Biobased

Surfactants and Detergents

Synthesis, Properties,
and Applications
Editors

Douglas G,Hayes
Dai Kitamoto
Daniel K.Y, Solaiman
Richard D, Ashby

Urbana, Illinois

AOCS Mission Statement

To be a global forum to promote the exchange of ideas, information, and experience, to enhance personal
excellence, and to provide high standards ofquality among those with a professional interest in the science
and technology of fats, oils, surfactants, and related materials.
AOCS Books and Special Publications Committee
M. Mossoba, Chairperson, U.S. Food and Drug Administration, College Park, Maryland
R. Adlof, USDA, ARS, NCAUR-Retired, Peoria, Illinois
M.L. Besemer, Besemer Consulting, Rancho Santa, Margarita, California
I? Dutta, Swedish University of Agricultural Sciences, Uppsala, Sweden
T. Foglia, ARS, USDA, ERRC, Wyndmoor, Pennsylvania
V. Huang, Yuanpei University of Science and Technology, Taiwan
L. Johnson, Iowa State University, Ames, Iowa
H. Knapp, DBC Research Center, Billings, Montana
D. Kodali, Global Agritech Inc., Minneapolis, Minnesota
G.R. List, USDA, NCAUR-Retired, Consulting, Peoria, Illinois
J.V. MAowski, Windsor Laboratories, Mechanicsburg, Pennsylvania
T. McKeon, USDA, ARS, WRRC, Albany, California
R. Moreau, USDA, ARS, ERRC, Wyndoor, Pennsylvania
A. Sinclair, RMIT University, Melbourne, Victoria, Australia
I? White, Iowa State University, Ames, Iowa
R. Wilson, USDA, REE, ARS, NPS, CPPVS-Retired, Beltsville, Maryland
AOCS Press, Urbana, IL 61802
02009 by AOCS Press. All rights reserved. No part of this book may be reproduced or transmitted in any
form or by any means without written permission of the publisher.
Library of Congress Cataloging-in-Publication Data
Biobased surfactants and detergents : synthesis, properties, and applications I editors, Douglas G. Hayes
... [et al.].
p. cm.
Includes index.
ISBN 978-1-893997-67-7 (alk. paper)
1. Biosurfactants. I. Hayes, Douglas G.
TP248.B57B557 2009
668.1--dc22
2008055036
Printed in the United States of America.
13 12 11 10 09 6 5 4 3 1
The paper used in this book is acid-free and falls within the guidelines established to ensure permanence
and durability.

Contents
Introduction

.............................................................................................................

.vii

Part 1 0 Introduction, Importance, and Relevance

1 Biobased Surfactants Overview and Industrial State-of-the-Art
Douglas G.Hayes ......................................................................................................

Part 2 0 Biosynthesis of Rhamnolipids and Sophorolipids

2 Production and Modification of Sophorolipids from Agricultural Feedstocks
Richard D. Ashby, Daniel K Y Solaiman, and Jonathan A. Zerkowski ....................... 29
3 Mannosylerythritol Lipids: Production and Downstream Processing
Udo Rau and Dai Kitamoto .................................................................................... .5 1
4 Advances in Bioprocess Development of Rhamnolipid and
Sophorolipid Production
Neissa M. Pinwn, Qin Zhang, Srujana Koganti, and Lu-Kwang Ju .......................... 77
5 Microemulsions of Rhamnolipid and Sophorolipid Biosurfactants
lhu ;$ Npyen and David A. Sabatini ...................................................................
107
6 Lipopeptide Biosurfactants and Their Use in Oil Recovery
Michael/. McInernq, Noha YOwseJf David I? Nagle ............................................... 129

Part 3 0 Employment of Phospholipids and Their Mimics

in Biomedical Applications
7 Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant
Hiromichi Nakahara, Sannamu Lee, and Osamu Shibata ...................................... 157
8 Lung Surfactants: Formulation, Evaluation, and Polymeric Additives
Edgar J. Acosta, Sameh M.I. Saad, Ningxi Kang, Zdenka Policova, Michael L. Hair, and
191
A. Wilhelm Neumann ...........................................................................................
9 Self-assembling Properties of Glycolipid Biosurfactants
and Their Functional Developments
Dai Kitamoto, Tomotake Morita, Tokuma Fukuoka, and Tomohiro Imura ................231

Part 4 0 Sugar-, Polyol-, and Amino-based Lipids: Biodegradable and

Biocompatible Surfactants for Foods, Health Care Products, and
Pharmaceuticals
10 Basic Properties of Sucrose Fatty Acid Esters and Their Applications
Naoya Otomo ....................................................................................................... 275
1 1 Selective Enzymatic Synthesis of N-Acylated Ahnolamine Emulsifiers
Cristina Otero ..................................................................................................... 299
V

12 Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase

Sang-Hyun Pyo and Douglas G. Hayes ..................................................................
323
13 Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived

from Arginine
Ma Rosa Infante, Lourder Pirez, Carmen Mordn, Ramon Pons,
andAurora Pinazo ............................................................................................ ..35 1
14 Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased
Surfactants
Fu Zhao, Kim Hayes, Steven J. Skerlos ................................................................... 389
15 Polyol and Amino Acid-based Biosurkctants, Builders, and Hydrogels
Kenneth M. Doll and Sevim Z. Erhan .................................................................. 425
16 Interfacial Properties of Sugar-basedSurfactants
OrlandoJ. Rojas, Cosima Stubenrauch, Lucian A. Lucia, and YOussefHabibi..........449
Index ........................................................................................................................

vi

473

For the majority of the 2OChcentury, petroleum was utilized as the main feedstock for
transportation and other fuels, chemical intermediates, and many co-products. The
Surfactants and Detergents industrial sector, like many others, relied heavily upon
petroleum as its main feedstock.
It is now clear that the relative abundance of petroleum is decreasing, leading to
increased prices, as experienced during 2008. In addition, it is generally accepted that
the heavy utilization of petroleum results in an increased emission of carbon dioxide
into earths atmosphere, which may lead to severe climate change throughout our
planet. More sustainable feedstocks for fuels and chemical intermediates are necessary.
The consumer is becoming aware of the potential danger of climate change and has
interest in products derived from more sustainable starting materials and processing
methods.
Due to the reasons discussed above, Surfactants and Detergents manufacturers have
increased interest in the utilization of biobased feedstocks. The primary motivation
is the rising costs of petroleum relative to oleochemical feedstocks, with the gap in
the price of the two anticipated to further narrow in the upcoming years. Of note,
the manufacturing costs associated with petrochemical and biobased feedstocks are
generally equal.
This bookwas prepared to highlight for the community ofsurfactant and Detergent
developers, formulators, and designers in industry, academia, and governmentsponsored laboratories throughout the world on the current industrial status of
biobased surfactants, new biobased surfactants being developed, the potential for
the green manufacturing of biobased surfactants via a biocatalytic route, and novel
applications for biobased surfactants. We hope this book will further the employment
of biobased surfactants in commercial products and inspire scientists worldwide to
develop new biobased surfactants and applications for their use.
The authors who prepared the chapters of this book are experts in their respective
fields, and thus are supplying the reader with a full, current, and critical evaluation of
biobased surfactant preparation and applications. We recognize that several research
groups are active in the research and development of biobased surfactants in addition
to those represented by this books authorship.
The book has been divided into four sections. The first section, Introduction,
Importance, and Relevance of Biobased Surfactants, consisting of Chapter 1,
reports on the motivation and current use for biobased surfactants, particularly in
the industrial sector, and to serve as an introductory chapter to the remainder of the
book, moreover, to highlight the other chapters in relation to biobased surfactant
research and development. The second section, Biosurfactants: Biosynthesis and
vii

0 D.G.Hayes et al.

Applications, consisting of Chapters 2-6 focuses upon biosurfactants, biobased

surfactants produced by living cells, particularly their fermentative production
and novel applications. The third section Biobased Surfactants in Biomedical
Applications, Chapters 7-9 describes the development of biobased lung surfactant
formulations, important for the treatment of several pulmonary diseases such as acute
respiratory distress syndrome, and the applications of biosurfactants in medicine. Of
relevance to Section 2, Chapter 9 also contains a thorough review of biosurfactant
preparation and utilization. The fourth section (Sugar-, Polyol-, and Amino-Based
Lipids: Biodegradable and Biocompatible Surfactants, Chapters 10-1 6) focuses upon
the preparation of sugar- and amino acid-based surfactants, with several chapters
describing their preparation by enzymatic synthesis, and their applications.
The genesis for this book was Biobased Surfactants and Oleochemicals, a
symposium chaired and organized by the Editors at the 9FhAmerican Oil Chemists
Society (AOCS) Annual Meeting and Exposition, May 13-16,2007, in Qutbec City,
Canada. The meeting was co-sponsored by the Japan Oil Chemists Society (JOCS).
Many of this books authors participated in the symposium, or in a symposium of the
same title at the 99chAOCS Annual Meeting and Exposition, May 18-21, 2008, in
Seattle, USA.
Therefore, we are grateful to both organizations, the AOCS (and the Biotechnology
and Surfactants and Detergents Divisions in particular) and the JOCS, for their
support in enabling the above-mentioned symposia. We also are very appreciative
of the staff of AOCS Press for their kind assistance and leadership in the preparation
and printing of this book. Last but not least, we thank the authors of this books
chapters for their willingness to contribute, and their time and diligence in preparing
the chapters. It is their expertise, shared in the pages of this book, which we hope will
make it a valuable resource for the reader.
Douglas G. Hayes
Dai Kitamoto
Daniel K. Y. Solaiman
Richard D. Ashby
February 13,2009

viii

0.

PART1

Introduction, Importance, and

Relevance

Biobased Surfactants: Overview and

Industrial State-of- the-Art
Douglas G. Hayes
Deportment of Biosystems Engineering andSoilScience;University of Tennessee,2506 E.J. Chapman Drive;
Knoxville, TN 37996-453 1

What Is a Biobased Surfactant?

Biobased products are defined as commercial or industrial products (other than
food or feed) that are composed in whole or in significant part of biological products
or renewable domestic agricultural materials (including plant, animal, and marine
materials) or forestry materials ( U S Senate Committee on Agriculture, Nutrition,
and Forestry, 2006). Biobased products do not contain components which are toxic or
endanger the environment; their use results in the following benefits: increased safety
for users, a low impact on the environment, a reduction of petroleum usage, and an
enhancement of the economy on both a domestic and local level, particularly for the
rural regional sector (Biobased US, 2007). Of the three broad categories attributed
to biobased products: biochemicals, biofuels, and biomaterials, biobased surfactants
best belong to biochemicals and, more specifically, to the following subcategories:
cleaning agents, paints and coatings, foods, personal-hygiene and cosmetics
products, lubricants, and pharmaceutical agents (Biobased US, 2007). An additional
category, a niche market, would be surfactants for environmental remediation or
environmentally-sensitive applications, such as tertiary oil recovery, particularly
well-suited for biosurfactants (Lu & Somasundaran, 2007; Quadri et al., 2008),
where the latter term refers to surface-active substances synthesized by living cells
(Rahman & Gakpe, 2008), and exists as a subset of biobased surfactants. (See also
McInerney et al., Lipopeptide Surfactants and Their Use in Oil Recovery, within
this book.) Metal-working fluids are a subcategory of lubricants where mixtures of
anionic and nonionic biobased surfactants are desired to replace synthetic surfactants
to greatly enhance their environmental profile. (See Zhao et al., Design ofvegetable
Oil Metalworking Fluid Microemulsions Using Biobased Surfactants in this book.)
Studies in 2003 and 2005 reported that the global surfactant industry produced
8.6 MMT and 12.5 MMT ofsurfactant, respectively, and was worth about $14 billion
and $28 billion, respectively, with annual growth being near 0.5 MMT (Anonymous,
2006; Chemistry in Britain, 2003; Patel, 2004). In the United States, ofthe 3.5 MMT

4 0 D.G. Hayes
of surfactants produced, roughly 35% were biobased (U.S. Department of Energy,
1999). The value of the laundry, personal-care, and dishwashing-detergents industries,
the major users of surfactants and detergents, was $70 billion globally in 2005, with
the projected value in 2010 expected to be $78 billion (Phillips et a]., 2007). The total
market value of the soaps and detergents industry within the European Union (EU)
countries in 2007 was 40.8 billion Euros ($64 billion) [International Association of
Soaps, Detergents, and Maintenance Products (AISE), 20071, with the E U consisting
roughly of one-third of the surfactant market (Chemistry in Britain, 2003).
Biobased feedstocks employed for surfactants are mainly used to form the
hydrophobe. The most common feedstock is seed oil, particularly from a source rich
in acyl groups of chain lengths of 12-16, such as palm, palm-kernel, particularly
palm stearine, a palmitic acyl-rich by-product from the purification and fractionation
of palm-kernel oil (Santosa, 2008), and coconut oils (Edser, 2004). Animal wastes,
such as lard or tallow, may be an additional source (e.g., cationic surfactants for
fabric softeners produced by Evonik, Essen, Germany) (McCoy, 2008b). Cuphea, a
potentially valuable new U.S. crop with an oil enriched in C,,-C,, fatty acyl groups,
may serve as a future feedstock (Pandey et al., 2000). Other potentially valuable
sources of fatty acyl groups for surfactants and detergents include algal oils and oils
from jatropha and soapnut (Supindus mukorossi) (Chetri et al., 2008; Chisti, 2007;
Scheibel, 2007). Many surfactants and emulsifiers require hydroxy fatty acids,
particularly ricinoleic acid from castor oil, and perhaps also, in the future, lesquerolic
and dimorphecolic acids from the seed oils of lesquerella and dimorphotheca, potential
new crops in the United States and the EU, respectively (Hayes, 2004). Epoxidized
seed oils (or their methyl esters), readily prepared from common oils such as soybean,
are potentially useful surfactant feedstocks (see Doll and Erhan, Polyol and Amino
Acid-Based Biosurfactants, Builders, and Hydrogels, in this book). In addition to
the oils triglyceride species, methyl ester derivatives, common intermediates in an
oleochemical biorefinery due to their use as a biodiesel feedstock (further discussed
below), would likely be employed (Ahmad et al., 2007). Sterols, minor components
of seed oils, also serve as effective surfactant hydrophobes (Svensson & Brinck, 2003).
Furthermore, lignin, a co-product from the conversion of lignocellulosic biomass to
biofuels, may be an additional feedstock for a surfactants hydrophobe (Holladay et
al., 2007; Johansson & Svensson, 2001).
Biobased feedstocks may also become more commonly employed as the sources
of a surfactants hydrophilic moiety, particularly for nonionic surfactants. Currently,
mono- and disaccharides (and their derivatives: sugar alcohols such as sorbitol and
xylitol and oxidation products such as furfuryl and levoglucosanyl compounds),
and glycols (produced via fermentation), derived from starches, cellulose, and other
naturally-derived polysaccharides, are becoming increasingly popular as head group
moieties of alkyl glycosides and esters (Werpy & Petersen, 2004). An additional
feedstock is glycerine, an inexpensive co-product from the manufacture of biodiesel, a
hydrophilic building block of mono- and diacylglycerols, common surfactants in food

Biobased Surfactants: Overview and Industrial State-of-the-Art 0 5

and cosmetics products (discussed below). In addition, glycerine can be converted into
other head group moieties: 1,2-propanediol or propylene glycol (1,3-propanediol),
the former via a catalytic process-a technological approach under development by
Dow, Huntsman, and others-and the latter, 1,3-propanediol, through a fermentation
process developed by DuPont Tate and Lyle LLC (Loudon, TN) (Shelley, 2007); and,
polyglycerol, through homogeneous or heterogeneous catalysis (Barrault et al., 2004).
Amino acids also serve as a feedstock for the hydrophilic moiety (Xia et al., 2001). In
addition, amino acids can be converted into ethanolamine and isopropylamine (from
serine and threonine, respectively), important intermediates for cationic surfactants
(Scott et al., 2007). (See Rosa Infante et al., Synthesis, Aggregation Properties
and Applications of Biosurfactants Derived From Arginine, and Otero, Selective
Enzymatic Synthesis of Alcoholamine Emulsifiers, in this book.) Surfactants with
DNA hydrophiles were recently reported (Bilalov et al., 2004; Leal et al., 2006; Xu et
al., 200.5). Alternatively, renewable resources serve as surfactant feedstocks indirectly
by serving as the carbon-energy source of biosurfactant synthesis from microorganisms
(Lang, 2003; Lu et al., 2007). (See Production and Modification of Sophorolipids
from Agricultural Feedstocks, Ashby et al., in this book.) Biosurfactants include
trehalose esters, rhamnolipids and sophorolipids (see chapters written by Ashby et
al. and Pinzon et al. in this book), mannosylerythritol lipids (chapter by Rau and
Kitamoto, Mannosylerythritol Lipids: Production and Downstream Processing),
other glycolipids, phospholipids, and lipopeptides (chapter by McInerney et al.)
(Lang, 2003; Lu et al., 2007). For synthetic nonionic surfactants employed today,
the major source of hydrophile is polyethylene glycol (or equivalently, polyethylene
oxide, or ethoxylate),which is derived from petroleum; however, one can potentially
derive it from bioethanol through the synthesis of ethylene as an intermediate (Gielen
et al., 2008). Ethylene oxide, the starting material, is carcinogenic, mutagenic, highly
flammable, volatile, and reactive. Therefore, for the reasons given above, biobased
hydrophiles for surfactants and detergents will become more commonly employed.
Perhaps polymerization products of bioderived 1,2- or 1,3-propanediol or polyglycerol
may become future substitutes.

Motivation for Biobased Surfactants

Since the press time of this book, petroleum crude oil prices have reached their highest
levels in recorded history ($140/bbl), and an increased replacement by oleochemical
or other biobased feedstocks is being pursued by the surfactants and detergents
industry (Phillips et al., 2008). The use ofoleochemical feedstocks makes further sense
since the production cost per mass for biobased materials is roughly equivalent to
petroleum-based materials (US.Department of Energy, 1999). The projected growth
has already begun; for example, the [g]lobal crude palm oil output has increased
from less than 3 million tonnes in 1974V.5 to almost 40 million tonnes in 2006/07,
[representing] an average annual growth rate of more than 8% (Carter et al., 2007).
However, before their more extensive utilization can occur, oleochemical feedstocks

6 0 D.G. Hayes
must become lower in cost, at least equal to that of petroleum-based alternatives,
with a lower degree of price fluctuations; and their supply must be more reliable
and readily available (Patel, 2004; Schalitz, 2007). In fact, oleochemical prices have
soared in recent months. For example, palm, soybean, and corn oils have achieved
record-high prices, due in part to the increased demand for biohels and increased
transportation costs (Brasher, 2008; Lewis, 2008; Phillips et al., 2008). But, genetic
engineering approaches are anticipated to enhance the supply-related issues in the
intermediate-to-long term by leading to an increased yield of oil and the ability to
tailor the chemical properties of the oil, such as acyl chain length, to meet product
needs (Gressel, 2008; Jambhulkar, 2007).
Another motivating factor is the desire of consumers to purchase eco-friendly
products, moreover, commodities that utilize renewable resources, employ sustainable
technologies for their manufacture, and result in a low environmental impact (McCoy,
2007a,b; Watkins, 2007). This area currently exists as a niche market that targets
environmentally-conscious consumers who are willing to pay for higher-priced items
that provide more ecological benefits. However, due to increased public concern [and
acceptance (Baum, 2008)] of the linkage of increased CO, and other greenhouse gas
production with climate change and the socioeconomic issues relating to petroleum
usage, which in turn will probably lead to increased government-imposed regulations,
this market will continue to grow. Many manufacturersare now seeking certification for
their products to be labeled as biobased or green or eco-friendlyor sustainable,
reflecting a trend of increased cooperation between environmental organizations
and manufacturers (McCoy, 2008a). A good description of sustainability, provided
by The Natural Step, an environmental nonprofit organization with the mission of
integrating sustainability practices into businesses, is: [nlature should not be subject
to increasing concentrations of substances extracted from the Earths crust, to high
concentrations of substances produced by society, or to physical degradation.. .
(McCoy, 2008a).
No universal set of standards or definitions exists that one can employ to achieve
certification on either a national or international level; moreover, several organizations
certih (Table 1.1). However, criteria for certification, even though (in many cases)
based partially on International Organization for Standardization protocols I S 0 14024
(environmental labels and declarations) and 14001 (environmental management
system), differ significantly from each other and change regularly (McCoy, 2008b;
Schalitz, 2007). This makes difficult both the job of producers to tailor their products
and to decide between certification labels, and comparison shopping for consumers
(Coons & Phillips, 2008). In addition, many manufacturers believe insufficient
evidence is available to support the claim that biobased surfactants are more degradable
than petroleum-based alternatives, particularly if the feedstocks undergo a significant
amount of chemical modification (McCoy, 2007a). Furthermore, the increased use
of biobased feedstocks can have detrimental environmental effects resulting from the
increased acreage of oleochemical-yielding crops-such as deforestation and animal

Biobased Surfactants: Overview and Industrial State-of-the-Art0 7

habitat loss, which have already occurred in Malaysia and Indonesia as a result of
palm-oil production-and a loss of biodiversity for oilseed-bearing plants ( McCoy,
2007a,b; Patel, 2004). However, sustainability certification is also being addressed for
feedstocks to prevent the occurrence of such events. The Roundtable for Sustainable
Palm Oil serves as an example (Roundtable for Sustainable Palm Oil-RSPO,
2008).
An important aspect of eco-friendliness is reduced environmental impact. Several
environmentally-related issues are facing the surfactants and detergents industry that
may be effectively addressed by using biobased surfactants (Table 1.2). The need
to replace alkylphenyl ethoxylates (APEs), nonionic surfactants commonly used in
cleaning products, was accelerated by the recent decision of Walmart to place APEs
on a list of "chemicals of concern" that its laundry product suppliers must replace
before 2010 (McCoy, 2007a) and by the request of the Sierra Club (Washington,
DC) to the U.S. EPA (Environmental Protection Agency) to ban the use of APES
(Sissell, 2007). Linear and branched alkyl ethoxylates, which one can manufacture
from petroleum and/or biobased feedstocks, are the leading substitution candidates
(McCoy, 2007a). The increased demand for cold water laundry detergents will provide
an opportunity for the use of methyl ester sulfonates (MESS), which are derived
from fatty acid methyl esters (FAMEs) (described below) (Roberts et al., 2008). For
Table 1.1. Certification Labelsfor BiobasedSurfactants
Name of certification
BioPreferred (US. Dept. Agriculture, Ames, IA)
CleanGredients (GreenBlue, Charlottesville,VA) a

Product type
Household, industrial cleaning, and
laundry products
Surfactants for household cleaning and
laundry products
Household and industrial cleaning
products

Design for the Environment (DR; US.

Environmental Protection Agency, Washington,
DC)
EcoCert (L'lsle Jourdain, France)
Detergents, cosmetics
EcoLogo (EnvironmentalChoice Program,
Industrial and household cleaning
Ottawa, Canada)
products
EU Flower/Eco-Label (European Union, Ivry-sur
Household and industrial cleaning
Seine, France)
products
Green Seal (Washington,DC)
Cleaning products and paintskoatings
GREENGUARD (Greenguard Environmental
Indoor air quality
Institute. Marietta, GA)
Leadership in Energy and Environmental Design
Industrial cleaners
(LEED US. Green Building Council, Washington,
DC)
Nordic Swan (Nordic EcolabellingBoard,
Household and industrial cleaning
Stockholm. Sweden)
oroducts
Does not certify, but provides a reference database 0f"green"and biobased ingredients.
availablefor formulators.

8 0 D.G. Hayes

example, the Japanese oleochemical manufacturer, Lion, is manufacturing MES to

include in the formulation of the Care Coldwash product by Danlind (Holstebro,
Denmark) (McCoy, 2008b). The growing demand for more concentrated liquid
laundry detergents, also driven by Wdmarts specifications, may lead to the increased
use of short-chain alkyl polyglucosides (described below) to provide cleaning activity
while enhancing the solubilization of other ingredients in the concentrated solution
(Watkins, 2007). Furthermore, a recent report states that [iln the last few years,
oleochemical surfactants have contributed more to the development of compact
detergents and the reduction of washing temperatures than petrochemical surfactants
have (Patel, 2004). Phosphates, used as builders for laundry detergents, help
inactivate hard water cations, provide pH buffering, and prevent redeposition of dirt;
however, their bioaccumulation in waterways leads to a temporary surge of algal and
bacterial growth, followed by oxygen depletion due to the thick slime layer that
forms at the waterair interface which blocks oxygen and light transport from the
air. This results in the loss of life for fish and aquatic wildlife and a source of drinking
water. (Further information is given in Doll and Erhan, Polyol and Amino Acid-Based
Biosurfactants, Builders, and Hydrogels in this book.) Recent industrial practice has
involved replacing phosphates with a blend of nonionic ethoxylated/propoxylated
fatty alcohol surfactants (Plurafac; BASF, Ludwigshafen, Germany), chelation agents,
and hydrophilic polymers (McCoy, 2007a). Biobased amino acid derivatives may be
important detergent builders as well (see chapter by Doll and Erhan). Regarding the
need to replace solvents in cleaning products, several companies are developing blends
of nonionic and cationic surfactants to replace ethylene glycol butyl ether, a volatile
organic compound commonly used in spray cleaners.
Table 1.2. Environmental Goals for the Surfactants and Detergents Industry
Environmental goal
Reduce carbon-dioxide emissions
Replacement of alkylphenyl ethoxylates (APES)
Use of cold-water-compatiblelaundry detergents
Increased concentration of liquid laundry
detergents
Replacement of phosphatesand zeolites as
builders for dishwasher detergent
Replacement of formaldehyde-releasing biocides
Replacement of solvents (e.g., ethylene glycol
butvl ether in mrav cleaners)
Replacement of quaternary ammonium
compounds
McCoy, 2007a,b; Watkins, 2007.

Rationale
Reduce climate change
Produce toxic biodegradation products;
may promote reproductive problems
Reduce household-energy consumption
Reduce packaging and shipping costs
Reduce phosphate eutrophication
in lakes and rivers; zeolites are
nondearadablesolids
Nasopharyngealcancer agent
Decreasevolatile organic compounds;
oossible asthma aaent
Possible asthma agent

Biobased Surfactants: Overview and Industrial State-of-the-Art0 9

A major advantage of biobased feedstocks for surfactants is that they require

less energy to manufacture than petroleum-based feedstocks, resulting in a lower
production of CO, and hence a more sustainable manufacturing approach (McCoy,
2007b). For example, a recent economic study states that if the EU would increase its
usage of oleochemical feedstocks in surfactant manufacture by 24%, from 880-1 100
kilotons, it would reduce CO, emissions by at least 8% (Patel, 2004). The cited article
employs a procedure that evaluates a manufacturing processs sustainability from
cradle to grave known as the Life Cycle Assessment (LCA).This procedure provides
quantitative measures of the environmental footprint of a manufacturing process
for all of its stages: from the growth, harvesting, and storage of the feedstocks to
pre-purification steps, to the manufacturing process steps, to the transportation and
delivery of the main products, to the treatment of by-products and wastes, and to the
ultimate fate of all products and by-products (Cowan et al., 2008; Hatti-Kaul et al.,
2007; Kloepffer, 2005; Patel, 2004 ). The footprint measured typically consists of
the production of CO, and other greenhouse gases that may contribute to global
warming and climate change; components that may acidify or degrade the quality
of the earths water resources (e.g., SOx,NOx,and phosphates); ingredients that will
not hlly degrade in the ecosphere; substances that contribute to smog and lower air
quality in urban areas, which produce respiratory illness, among other diseases; and
the overuse of land, water, and energy resources. In conclusion, LCA analyses or their
equivalent will become increasingly important as tools to evaluate the manufacturing of
surfactants; and, not surprisingly, the use of biobased feedstocks will receive favorable
LCA evaluations due to the reduction of energy use and, hence, CO, production.
An additional driving force for the hrther increase of biobased surfactants is
the increased employment of nonionic surfactants at the expense of cationics and
anionics (Patel, 2004). Biobased fatty acid and alcohol feedstocks are very suitable for
nonionics.

Examples and Applications of Biobased Surfactants

The chemical structures of several biobased surfactants are given in Fig. 1.1. Alkyl
glycosides (AGs, e.g., P-dodecyl maltoside of Fig. 1.l),saccharides, and fatty alcohols
linked together via an acetal bond-which occur in nature (in cyanobacteria) but are
more commonly synthesized by mixing saccharide and fatty alcohol at an elevated
temperature in the presence of an acidic catalyst (typically forming an acetal linkage
at the saccharides a-carbon)-are highly biodegradable biobased surfactants that
are commonly employed in personal care products, hard-surface cleaners, laundry
detergents, textiles, oil recovery, and foods (Balzer & Luders, 2000; Von Rybinski
& Hill, 2003). (See Rojas et al., Interfacial Properties of Sugar-based Surfactants
in this book.) The significant difference in manufacturing costs for their preparation
relative to the more commonly employed alkyl ethoxylate nonionic surfactants is
decreasing due to the rapid surge of petroleum prices, since ethylene oxide is primarily
derived from petroleum. Saccharide-fatty-acid esters (e.g., sucrose 6-monooleate of

10 0 D.G. Hayes

Fig. 1.1) have many uses as surfactants in foods, cosmetics, and pharmaceuticals, as
described in Table 1.3 (Nelen & Cooper, 2004). (See also Basic Properties of Sucrose
Fatty Acid Esters and their Application, Otomo, and Synthesis of Saccharide-Fatty
Acid Ester Biosurfactants via Lipase, Pyo and Hayes in this book.) Rhamnolipids,
sophorolipids, trehalose lipids, mannosylerythritol lipids (not depicted in Fig. 1. I),
and other polysaccharide-based biosurfactants derived from fermentation have many
potential applications in the petroleum, pharmaceutical, food-processing, pesticide,
and environmental-remediation industries. (See chapters written by Ashby et al.., Rau
and Kitamoto, Nguyen and Sabatini, Kitamoto et al., and Pinzon et al. in this book.)
Amino acid- and peptide- or protein-based surfactants, encountered in nature (e.g.,
lipoproteins; see McInerney et al., Lipopeptide Surfactants and Their Use in Oil
Recovery in this book) and synthesized chemically or chemoenzymatically (see Rosa
Infante et al., Synthesis, Aggregation Properties and Applications of Biosurfactants
Derived From Arginine, and Otero, Selective Enzymatic Synthesis of Alcoholamine
Emulsifiers in this book), have several current and potential applications in cosmetics
(e.g., thickeners, emollients, and conditioners), foods, and cleaners (Nnanna et al.,
2001; Sakamoto, 2001; Xia et al., 2001); and one can employ them in oil recovery
(McInerney et al.$ chapter). In addition, amphiphilic polypeptides (e.g., copolymers
of leucine, tryptophan, and lysine) are useful lung surfactants. (See Nakahara et al.s
Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant in
this book.)
MES, produced from FAMEs, derived primarily from coconut and palm oils
[with a C,, FAME-rich fraction, a possible by-product stream from a palm oilbased biorefinery produced by thermal fractionation due to its undesirability in
biodiesel formulations (Foster, 2006d), being of particular interest as an economic
feedstock], is a common and inexpensive anionic surfactant in powder laundry
detergent products (e.g., Nl/Surf powder detergent from Unilever). The process
for synthesizing MES, developed in the 1950s from USDA research, involves the
reaction of FAME with a slight stoichiometric excess of SO, in the presence of NaOH
(Ahmad et al., 2007; Edser, 2006; Foster, 2006a; Maurad et al., 2006; Roberts et
al., 2008). MES is produced commercially in Japan by Lion (Johor, Malaysia; 40
ktodyear) and in the United States by Stepan (Northfield, IL; 50 ktondyear) and
Huish Detergents (Salt Lake City, UT; 80 ktondyear) (Bognolo, 2008). Currently,
linear alkylbenzene sulfonates (LASs) are the most commonly employed anionic
surfactants for detergents and other cleaners, and have been so for the past 30 years
(Ahmad et al., 2007). Until recently, the manufacture of MESS was more expensive
than for LASs due to the slower reaction rate of sulfonation, the need to control
hydrolysis during the reaction, and the need to perform bleaching on the reaction
product (Edser, 2006). However, the higher rate of the price increase for petroleum
relative to palm and coconut oils is making MES production more cost-competitive
with that of LAS (Edser, 2006), and the product quality has improved as well
(Bognolo, 2008; Foster, 2006~).Moreover, the development of biodiesel facilities

Biobased Surfactants: Overview and Industrial State-of-the-Art 0 11

Sophorolipid

n
,

HO

OH

Amino Acid Surfactant (Lauryl Glutamate)

Fig. 1.1. Chemical structure of biobased surfactants. A and 6.

in Malaysia and Southeast Asia has increased the availability of FAME feedstock.
An additional economic advantage for MES versus fatty alcohol-based surfactants
such as alcohol (and alkyl ester) sulfates and alkyl ethoxylates is that the former does
not require high-temperature and high-pressure hydrogenation for the conversion
of FAME into fatty alcohol (Ahmad et al., 2007). MESs perform well compared
to other anionic surfactants for both hot- and cold- water formulations due to its
stability in hard water and its inertness to enzymes contained in laundry formulations
(Cohen et al., 2008; Foster, 2006c; Maurad et al., 2006; Roberts et al., 2008). They
rapidly undergo biodegradation aerobically (a-oxidation) to produce monomethyl
a-sulfosuccinate, which then undergoes desulfonation to produce succinic acid and
inorganic sulfate as its ultimate products, with its biodegradation reported to be at
least equal to, and in some reports, faster than that of LAS (Ahmad et al., 2007;
Edser, 2006; Masuda et al., 1993; Roberts et al., 2008). The latter's biodegradability
was investigated thoroughly, and performed well, yielding degradation products that
benignly undergo mineralization (HEM, 2008). However, MES does not undergo
mineralization effectively under anaerobic conditions (Garcia et al., 2008). Although
MESs present unique advantages, deficiencies in their performance were noted. They
possess low foam characteristics, limiting their utility in hand-washing and other
foam-based products, and are unstable in high-pH liquid formulations (Bognolo,
2008; Foster, 2006b). However, one can address these deficiencies by proper product
formulation (Foster, 20OGb,c). But as an overall assessment, MES surfactants are wellsuited to further increase their market share of anionic detergents at the expense of
petroleum-derived MS. Ahmad et al. (2007) suggest a goal for MES-replace 30%
of the 3.4 MMT LAS market by 2010, resulting in a 1 MMT production capability.
Esterquats are cationic, quaternary ammonium surfactants derived from animal
fats or vegetable oils (in the form of FAMEs) that are important rinse-added fabric
softeners and ingredients of hair conditioners, dyes, and bacterial and sanitizer
products, including swimming-pool biocides (Ahmad et al., 2007; Mishra & Tyagi,
2007; Overkempe et al., 2003). Gemini esterquats were developed that are useful
flotation agents, for example, in the mining of calcite (Mishra & Tyagi, 2007).
O n a molecular level, they adsorb via an electrostatic attractive driving force to the
negatively charged surfaces of cotton, collagen, and other fibrous biopolymers. The
hydrophobic moieties of the surfactants extend outward and prevent friction between
adjacent fibers. They typically are formed by first esterifying methyldiethanolamine
(or other amine derivative) with two mole equivalents of fatty acyl groups at ~ 2 5 0 C
for a few hours under vacuum to remove the product water and then quaternizing
the molecule (e.g., with methylethylsulfate at c1Oo"C for a few hours) to form the
derivative depicted in Fig. 1.1 (Mishra &Tyagi, 2007; Overkempe et al., 2003). They
are readily biodegradable and possess an excellent environmental profile. Although one
can consider the traditionally used fatty acid feedstock, tallow, biobased, the mediumchain fatty acyl groups derived from palm oil, for example, are becoming increasingly
popular and provide improved performance (Ahmad et al., 2007; Overkempe et al.,
2003).

Biobased Surfactants:Overview and Industrial State-of-the-Art 0

13

Fatty acid-polyol esters and their ethoxylates-monoglycerides, glycol esters

(ethylene and propylene), sorbitan derivatives, and polyglycerolesters-have numerous
applications as biodegradable and biocompatible surfactants in foods, cosmetics, and
pharmaceuticals as listed in Table 1.3. One can synthesize these derivatives chemically
(Honydonckx et al., 2004) or enzymatically (the latter discussed below). Ethoxylated
sorbitan esters, often referred to as polysorbates, are well-known to the field by
the commercial trade names Span and Tween. Solvay Chemicals, a major producer
of polyglycerol, states that polyglycerol esters yield surfactant properties similar to
polysorbates, making them very useful biobased emulsifiers for personal care products
[Solvay Chemicals (Deer Park, TX), 20041. Also, polyglycerol esters have unique
properties as antifogging agents for use in food packaging and greenhouses, and have
other industrial applications as well [Plasman et al., 2005; Solvay Chemicals (Deer
Park, TX), 20051. In addition to the esters of the polyols listed in Table 1.3, other
glycol esters (e.g., 1,3-propanediol) and sugar-alcohol esters (e.g., erythritol and
xylitol) can serve as effective biobased surfactants (Hayes, 2004). As discussed above,
most of the polyol groups described in this paragraph are also biobased.
Several other biobased surfactants are noteworthy but are not depicted in Fig. 1.1.
Phospholipids have numerous applications in pharmaceuticals (e.g., liposomal drug
delivery), foods, and cosmetics (Hayes, 2004). (For applications of phospholipids
as components of lung surfactant products for the treatment of Neonatal and Adult
Respiratory Distress syndrome, see Nakahara et al.; Lung Surfactants: Formulation,
Evaluation, and Polymeric Additives, Acosta et al.; and Influence of Pulmonary
Surfactant Protein Mimics on Model Lung Surfactant, in this book.) In addition to
their cosmetics- and pharmaceutical-related applications, fatty acid ethoxylates are
employed in laundry, dishwashing, floor- and wall-cleaning, metal-cleaning, and rugcleaning products, and are rewetting agents to improve the absorbency ofwet-strength
paper used in the manufacture of paper towels [Gujarat Chemicals (Nanpura, India),
20081. Compared to alkyl ethoxylates (ethoxylated fatty alcohols), ethoxylated fatty
acids possess some advantages: improved skin compatibility and lower susceptibility
toward foaming; however, the fatty acid ethoxylates undergo hydrolysis in alkaline
media and possess lower cloud point temperatures (Johansson & Svensson, 200 1).
Ethoxylated fatty amines are (or are components of) emulsifiers; wetting agents;
dispersants; stabilizers, sanitizers, and defoaming agents for agrochemical emulsifiers;
industrial cleaners; metal cleaners; textiles; paper de-inking; and drilling products
and detergents, and are intermediates for the synthesis of anionic surfactants [Gujarat
Chemicals (Nanpura India), 20081. Sterols, minor components of seed oils, upon
ethoxylation or glycosylation, are useful biobased surfactants for cosmetics and
pharmaceuticals (Svensson & Brinck, 2003). Saponins are steroid glucosides that occur
naturally in several plant species that may be useful biobased surfactants (Svensson
& Brinck, 2003). Lignin sulfonates were employed as dispersing agents for dyes and
other chemicals in water for employment in the textile industry; however, their use
is limited by their dark color and their poor surface tension-lowering capabilities
(Perkins, 1998).

Table 1.3. Food, Cosmetics, and Pharmaceutical Applications of Fatty-Acid-Polyol Ester and Ethoxylate-Fatty-Acid Ester Biobased
Surfactants
Polyolester

Food applications

Cosmetics applications

Pharmaceutical applications

Monoglycerides

Margarines, ice cream, bread,

peanut butter, chewing gum,
and cakes (Cottrell81van Peij,
2004;Hayes, 2004; Sparso 81
Krog, 2004)

Emulsifier, emollient, moisturizer,

and viscosity builder in creams
and lotions [cosmeticinfo.
org, 2008 Flora Health
Manufacturing& Distributing
Ltd. (Burnaby Canada), 20081
Skin-care products, moisturizers,
lipstick, and other makeup
products (cosmeticinfo.org,
2008)

Drug delivery vehicles (Sagalowiaet al.,

2006), emulsifier (Marti-Mestres& Nielloud,
2000)

Shampoo; skin care; creams;

bubble baths; makeup; hair,
skin, and nail care products; and
lotions (cosmeticinfo.org, 2008)
Moisturizers, cleansing products,
fragrance products, and makeup
products such as foundations
and lipsticks (cosmeticinfo.org,
2008)
Bath products, cleansing
products, makeup products,
hand and body preparations,
suntan products, and shampoos
(cosmeticinfo.org, 2008)

Emulsifiers (Marti-Mestres& Nielloud, 2000)

Acetate or
lactate esters of
monoglycerides

Beveragewhiteners, chewing
gum, meat products, sauces,
canned coffee, solubilization
agents for coloring agents and
antioxidants, and baked items
(Gaupp 81Adams, 2004)

Ethylene-glycolesters

Propylene-glycol
(12-propanediol)

Aerated cakes and soft ice

cream (Sparso & Krog, 2004)

esters

Sucrose and
saccharide esters

Sauces, mayonnaise, salad

dressings, caramel, chocolate,
icing, and bakery items (Nelen
&Cooper, 2004)

Plasticizer for coatings of pills [Morflex Inc.

(Greensboro, NC, 20081

Potentially valuable agents for

microemulsion-baseddrug delivery and
transdermal drug delivery (Marti-Mestres
81Nielloud, 2000) (Otomo, "Basic Properties
of Sucrose Fatty-AcidEstersand Their
Application,minthis book)

c
.
~

P
nI
l
Y

In

Table 1.3, cont. Food, Cosmetics, and Pharmaceutical Applications of Fatty-Acid-Polyol Ester and Ethoxylate-Fatty-Acid Ester
Biobased Surfactants
Polyolester
Sorbitanesters

Food applications
Cakes, baked goods, and bread
(Cottrell& van Peij, 200s)

Sorbitanesters,
ethoxylated (Spam,
Tween; polysorbates)

Cakes, baked goods, bread,

and dairy products (Cottrell&
van Peij, 2004)

Polyglycerolesters

Sponge cakes and to control

viscosity and prevent
"blooming" in chocolate
(Norn, 2004); bread making
(Miyamoto et al., 2005)
Monoglycerides: Dough
conditionerkrengthener,
emulsifier for cake batter,
coffee whitener, and
pan-release agent (Smith &
Hona. 20031

Ethoxylated
fatty acids or
monoglycerides

Cosmetics applications
Skin-care products, skincleansing products, moisturizers,
eye makeup, and other makeup
(cosmeticinfo.org, 2008)
Shampoos, hair conditioners,
hair dyes and colors, bath
products, skin cleansers, skin
fresheners, makeup bases and
foundations, and other hair- and
skin-care products (cosmeticinfo.
org, 2008)
Emulsifier [Allef et al., 2006;
cosmeticinfo.org, 200s; Plasman
et al., 2004; Solvay Chemicals
(Deer Park, TX), 20041
Emulsifiers (cosmeticinfo.org,
2008)

Pharmaceutical applications
Water-in-oil emulsifiersfor oral and
injectable preparations (Marti-Mestres &
Nielloud, 2000)
Water-in-oil emulsifiersfor oral and
injectable preparations, co-solubilization of
hydrophilic and hydrophobicvitamins, and
suspensions (Marti-Mestres& Nielloud, 2000)

Oral drug delivery (Andrysek, 2006) and

emulsifier (Allef et al., 2006)

Fatty acids: Emulsifier for self-emulsification

drug-delivery preparations (Trouve, 1996)

16 0 D.G. Hayes

Many of the products described above are synthesized directly from FAMEs;
e.g., saccharide- and polyol-fatty-acid esters, ethoxylated fatty acids, amino acid
surfactants, esterquats, and MESS. The majority of the other biobased surfactants
are indirectly manufactured from FAMEs. AGs and biobased alkyl ethoxylates are
synthesized from fatty alcohols, which are derived from FAMEs. Fatty amines are also
derived from FAMEs. Since FAMEs are primary targets of oleochemical processing
for the manufacture of biodiesel, biobased surfactant production should integrate
well into an oleochemical biorefinery. The latter term refers to the utilization
of biobased feedstock by a chemical plant, which through various processing and
separation schemes produces electrical power and yields an array of hels and
products/co-products/intermediateproducts, similar to a petrochemical refinery,
where petroleum is fractionated to produce home and transportation fuels, chemical
intermediates, lubricants, asphalt, etcetera. Moreover, FAME will serve as a key
intermediate oleochemical biorefinery product stream that can be utilized as biodiesel
he1 or as a chemical intermediate for the production of biobased products, including
biobased surfactants. Glycerine, a key by-product stream, will be readily utilized as
an intermediate for biobased surfactants as well, as described above. Other biobased
surfactant intermediates such as saccharides, proteins, and lactic acid (employed for
the synthesis of monoglyceride esters) would occur as biorefinery by-product streams
that utilize certain seed oils, starch, or lignocellulosic biomass as feedstock. A further
discussion of the oleochemical biorefinery concept is given elsewhere (Hatti-Kaul et
al., 2007; Hill, 2007; Johansson & Svensson, 2001).
Table 1.4 lists several biobased surfactants that are available commercially and
their manufacturers. This list also includes commercial products which incorporate
biobased surfactants, many of which have achieved one of the eco-friendly or biobased
certifications listed in Table 1.1. This list truly reflects the global nature of biobased
surfactants, with manufacturers based in several different countries represented.

Utilization of Enzymes for the Manufacture of

Biobased Surfactants
As demonstrated recently through LCAs, the replacement of chemical processes by
enzymic ones for oleochemical products has several advantages: lower energy use (due
to the use of lower temperatures and ambient pressure, resulting in reduced CO,
production), lower amounts of waste products and by-products, and the absence of
toxic metal catalysts or aciddbases (Cowan et al., 2008). Although not employed in
commercial processes to date due to their prohibitively high cost, enzymes may be
potentially useful for the synthesis ofseveral biobased surfactants ( Ducret et al., 2006;
Hayes, 2004; Karmee, 2008; van Rantwijk et al., 1999) (Table 1.5). As indicated in
this table, this book provides significant detail on the current state-of-the-art relating
to enzyme use. This list does not include the use of microorganisms to synthesize
biosurfactants, which is also described in this book.

Biobased Surfactants: Overview and Industrial State-of-the-Art 0

17

Table 1.4. Biobased-Surfactant Commercial Products

Industrial manufacturer
Abbott Laboratories(Abbott Park, IL)

Air Products (Allentown, PA)

American IngredientsCompany (Kansas City,

MO)
Bio-Kleen Products, Inc. (Kalamazoo, MI)
BLES Biochemicals Inc. (London, Ontario,
Canada)
Boehringer lngelheim (Ingelheim, Germany)

Centro China, LTD

Chiesi Farmaceutici (Parma, Italy)

I .

Chlorox (Oakland, CA)

Cognis (Cincinnati, OH)
Cognis Oleochemical (Selangor, Malaysia)
Croda, Inc. (Edison, NJ)
Danisco NS (Copenhagen, Denmark)

Deyer Chemicals Co., LTD (Zhejiang,China)

Dow Chemical (Midland, MI)
Dustbane (Ottawa, Canada)
Earth Friendly Products (Winnetka, IL)
Eco-Products, Inc. (Boulder, CO)
Esterchem (Leekbrook, UK)
eZall Technologies (Grove City, OH )

Products
Survanta lung surfactant (obtained from
processed, supplemented bovine-lung
tissue)
Tomadol surfactants (biobased alcohol
ethoxylates, household and industrial
cleaners)
Monoglyceridesand ethoxylated
monoglyceridesfor the food industry
Various household and industrial cleaning
supplies
BLES lung surfactant (obtained from the
lung lavages of benefited cows, >95% of
phosphoipids)
Alveofact lung surfactant (derivedfrom
bronchoalveolar lavage of bovine-lung
surfactant. >99% wt% of DhosDhalioids)
Ethylene-glycol esters
Curosurf luns surfactant (obtained from
minced porcine-lung tissue; contains 93% of
DhosDholiDids)
Green Works household cleaners (contain
alkylpolyglucosides)
Alkylpolyglucosides (APGs)
Methyl ester sulfonates (MESS)derived from
palm oil
Supplies fatty acids and fatty esters to the
soam and deteraents industrv
Dimodan distilled monoglycerides,Grindsted
ACETEM (aceticacid) and CITREM (lactic
acid) esters of monoglycerides,Grindsted
PGE (fatty acid) and PGPR (polyricinoleic
acid) esters of polyglycerol, Grindsted PGMS
(propylene- glycol esters of fatty acids), and
Grindsted SMS sorbitan monostearate
Polyglycerolesters, sorbitan monostearate,
sucrose esters
Ecosurf surfactants (alkanolsderived from
palm-kerneloil as feedstock)
Various industrial cleansers
Laundry detergent; household cleaners
Laundry detergent; household cleaners
Polyglycerolesters
Total Body Wash (biobased alcohol
ethoxvlatesfor animal care]

18 0 D.G. Hayes

Gemtek (Phoenix. AZ)

Greenfeet (Chico, CA)
Huish Detergents, Inc. (Salt Lake City, UT)

Hychem Corp. (Belmar, NJ)

Jeneil Biosurfactant Co., LLC (Saukville,WI)
JohnsonDiversey (Sturtevant,WI 1
Kawaken (Tokyo, Japan)

Lambent Technologies (Gurnee, IL)

Safe Care (industrial cleaner)

Laundry detergent; household cleaners
Methyl ester sulfonates (used in Costco
Kirkland Brand Select Ultra and Safeway
Select Ultra laundrv-deteraent Droducts)
ECOTERIColeo-derived nonionic surfactants,
including sorbitan esters and polysorbates,
alkyl polysaccharides, ethoxylates of natural
alcohols and polyglycol fatty acid esters; and
ALKADET alkyl polyglucosides
Monoglycerides,polyglycerol esters, and
propylene glycol esters
Rhamnolipid biosurfactants
Household cleaners
Nonionic and peptide-basedsurfactants for
cosmetics and personal-care products
Myverol monoglycerides,acetylated and
succinylated monoglycerides,propyleneglycol esters, Admul polyglycerol esters,
sorbitan esters, and polysorbates (for foods
and pharmaceuticals)
Sorbitan esters, polysorbates, ethoxylated
fatty acids, alkyl ethoxylates, monoglycerides
(for foods and pharmaceuticals)
des, sorbitan esters, and
^-^-

Japan)
MitsubishiTanabe Pharma Corporation
(Tokyo, Japan)

Nippon Fine Chemicals (Tokyo, Japan)

ONY Inc. (Amherst, NY)

^-

SURFACTEN lung surfactant (obtained from

finely ground bovine-lung tissue; contains
84% of DhosDholiDidsl
Polyoxyethylene-sorbitanesters, ethyleneglycol and propylene-glycol esters,
monoglycerides,and sorbitan esters
Sucraph AC-8 (caprylylglucoside)
lnfasurf lung surfactant (obtained from calf-

Biobased Surfactants: Overview and Industrial State-of-the-Art

19

Table 1.4., cont. Biobased-SurfactantCommercial Products

-~
___

Industrial manufacturer
Optionsforlife (New York, NY)
Orafti Group (Tienen, Belgium)
OurHouse (Niwot, CO)
Rhamnolipid,Inc. (St. Petersburg, FL)
Rhodia (Paris, France)
Riken Vitamin (Tokyo, Japan)
Seventh Generation (Burlington, VT)
ShreeVallabhChemicals (Gujarat, India)
Solutia Europe (Louvain-la-Neuve, Belgium)
Spartan Chemical (Maumee, OH)
Stepan (Northfield, IL)
Undesa (Barcelona, Spain)
Wtek (Milton, WI)
Win Chemicals (Burlington, Ontario, Canada)

Products
Window and glass cleaner
INUTECSPl (personal care products and
paints and coatings)
Shiny Surface Cleaner,Heavy Duty Cleaner
(householdcleaners)
Rhamnolipid biosurfactants
Rhodoclean (uses pinene from pine oil as
feedstock; alkyl phenyl ethoxylate replacer)
PolygI ycerol esters
Laundry detergents and household cleaners
(contains biobased alkyl ethoxylates)
Ethoxylatedfatty acids
Dequest PB (carboxymethylinulinderivative;
for oil recovervfrom soent wells)
All-purpose bathroom, glass, hand, and
industrial cleaners
Methyl ester sulfonates (MESS)
Kemifluid esterquats, Monestriol ethyleneglycol and propylene-glycol esters
Industrial cleaners
WinSurf Surfactant Systems (industrial
cleaners)

Conclusions
The ideal surfactant was recently described as possessing low surface and interficial
tension, low critical micelle concentration, low Krafi-point temperature, high
solubility in cold and hot water (or in alkanes, squalene, aliphatic esters, or other
lipophilic solvents), fast kinetics for their self-assembly, high biodegradability and
biocompatibility, an excellent environmental profile, and a low cost-to-performance
ratio (Scheibel, 2007). Up to recent years, biobased surfactants competed favorably
with, and in some cases outperformed, synthetic, petroleum-derived surfactants (e.g.,
for biodegradability and biocompatibility), except in terms of the cost-to-performance
ratio. However, with the recent surge of petroleum prices, biobased surfactants are
becoming a more economically viable choice. In addition, the increased consumer
desire to purchase eco-friendly and biobased products favors biobased surfactants.
Furthermore, the development of oleochemical biorefineries based on FAMEs for
biodiesel will lead to a more abundant and consistent feedstock supply for surfactant
synthesis. Therefore, biobased feedstocks will become increasingly popular for
surfactants and detergents manufacture in the near-term. The utilization of biobased
feedstock and sustainable manufacturing by the surfactants and detergents industry

20 0 D.G.Hayes
Table 1.5. Enzymes Employed for the Preparation of Biobared Surfactants
Enzyme type
Lipases

Glucosidase

Biobased surfactant
Monoglycerides, saccharideand polyol-fatty-acid esters,
amino-acid surfactants
(via N-acylation), and
lysophosphatidylcholine
Alkyl glucosides

Phospholipases
Papain

Phospholipids
Amino-acid surfactants

References
Hayes, 2004 and Karmee,
2008 this chapter and other
chapters in this book by
Rosa Infante et al., Otero, and
Pyo and Hayes
Ducret et al., 2006 and van
Rantwijk et al., 1999 in this
chaoter
Hayes, 2004 in this chapter
See chapter by Rosa Infante
in this book.

would be greatly assisted by an international effort to create a consistent, uniform,

and worldwide certification program for eco-friendlyand/or biobased. Although
not currently economically viable, the use of biocatalysts (enzyme or whole cells) is a
sustainable technological approach worthy of further study due to the reduced energy
costs, the formation of less toxic by-products, and higher yields.

Acknowledgments
?he author thanks the U.S. Department of Agriculture, Grant 2006-35504-17262,
and the National Science Foundation, Grant BES-0437507, for supporting his
research in the biobased surfactant and microemulsions areas; Dr. Warren Schmidt,
Shell Global Solutions (U.S.) Inc., Houston,TX, and Mr. William J. Schalitz, Spartan
Chemical, Maumee, OH, who provided technical assistance.

References
Ahmad, S.; I? Siwayanan; Z. Abd Murad; H. Abd Aziz; H. Seng Soi. Beyond biodiesel: Methyl
Esters as the Route for the Production of Surfactants Feedstock. INFORM2007, 18,216-220.
Allef, I?; W. Berkels; H. Foetsch; J. Meyer. Partial Esters of Polyglycerol with PolyricinoleicAcid
and Multifunctional Carboxylic Acids for Use in the Preparation of Emulsions and Dispersions.
European Patent EP 2006-243 1683781 (2006).
Andrysek, T. Excipients and their role in absorption: influencing bioavailability of cyclosporine by
triglycerides and polyglycerol esters. Biomed. Pap. 2006, 150, 227-233.
Anonymous. Surfactants and detergents: Cleaning the green way. Fonu on Su$2006,2,6.
Balzer, D.; H. Luders (Eds.) Nonionic Su$mtants: Alkyl Polyglucosides. Surfactant Science Series.
Marcel Dekker: New York, 2000; Vol. 91.
Barrault, J.; J.M. Clacens; Y. Pouillow. Selective Oligomerization of Glycerol Over Mesoporous
Catalysts. Top. Catal. 2004,27, 137-142.
Baum, R.M. Sustainability: Learning to Live Off the Sun. Chem. Eng. News 2008,86(33), 42-46.

Biobased Surfactants:Overview and Industrial State-of-the-Art0 2 1

Bilalov, A.; C. Leal; B. Lindman. Mixing Oil and Water by a DNA-based Surfactant. J. Pbys.
Chem. B 2004,108,15408-15414.
Biobased US. What Are Bio-Based Products?2007; http://www.biobased,us/biobased.html.
Bognolo, G. Surfactants Market Trends in Europe. CESIO 7th World Surfactants Conzrence, European Committee of Organic Surfactants and their Internaliates, Brossels, Paris, 2008.
Brasher, I? Biodiesel Comes Under Fire as Next Target of Fuel Costs. 7he Des Moines (L4) Regizter
2008, July 13, 2008, 1 .
Carter, C.; W. Finley; J. Fry; D. Jackson; L. Willis. Palm Oil Markets and Future Supply. Eur. /.
Lipid Sci. Technol. 2007, 109, 307-314.
Chemistry in Britain. Surfactants: The Ubiquitous Arnphiphiles. Chem. World 2003, (Iuuly), 1-7.
Chetri, A.B.; M.S. Tango; S.M. Budge; K.C. Watts; M.R. Islam. Non-edible Plant Oils as New
Sources for Biodiesel Production. Int.J. Mol. Sci. 2008,9, 169-180.
Chisti, Y. Biodiesel From Microalgae. Biotechnol Adv. 2007,25,294-306.
Cohen, L.; F. Soto; A. Melgarejo; D.W Roberts. Performance of @-sulfa Fatty Methyl Ester
Sulfonate versus Linear Alkylbenzene Sulfonate, SecondaryAlkane Sulfonate and a-sulfo Fatty
Methyl Ester Sulfonate.J. Su$ Deterg. 2008, 11, 18 1-1 86.
Coons, R.; K. Phillips. Soapers Awash in Green Issues and Feedstock Concerns. Chem. Week
2008, 170(5), 22.

cosmeticinfo.org. Cosmetic Ingredients Search Database, 2008; http://www.cosmeticsinfo.org/

ingredient-details.php?ingredient-id= 1 186.
Cottrell, T.; J. van Peij. Sorbitan Esters and Polysorbates.Emuls$ers in Food Zchnolau; R.J.
Whitehurst, Ed.; Blackwell Publishing: Oxford, UK, 2004; 162-185.
Cowan, D.; K.M. Oxenbell; H.C. Holm. Enzymatic Bioprocessing of Oils and Fats. INFORM
2008,19,210-212.

Ducret, A.; M. Trani; R. Lortie. Comparison Berween Various Commercial Sources of Almond
P-glucosidase for the Production of Alkyl Glucosides.J. Mol. Catal B: Enzym. 2006,38, 9 1-94.
Edser, C. Growing Use of Oleochemicals. F o w on Su$2004, (Nov), 1-2.
Edser, C. Is the Time Ripe for Methyl Ester Sulfonates?Focus on Su62006, (Sept.), 1-2.
Flora Health Manufacturing & Distributing Ltd. (Burnaby, Canada). Herb & Supplement Encyclopedia: Glycerol Stearate Supplement; 2008, http://www.florahealth.com/flordhome/Canada/
HealthInformation/Encyclopedias/GlycerolStearate.htm.
Foster, N. Commercial Applications of the Chemithon MES Process. INFORM2006a, (Methyl
Ester Sulfonates-A Guide to Biorenewabk Resources: A Special Supplement, Biorenewable Resources #3), 7-8.
Foster, N. Gimme a Break! (Price, that is). INFORM2006b, (Metbylfiter Sulfonates-A Guide to
Biorenewable Resources: A Special Supplement, Biorenewable Resources #3), 3.
Foster, N. Methyl Ester Sulfonates Industry Poll. INFORM 2 0 0 6 ~(Methyl Bter Su&nates-A
Guide to Biorenewable Resources: A Special Supplement, Biorenewabk Resources #3), 4-6.
Foster, N. Questions About MES. INFORM 20064 (Methyl firer Sulfonates-A Guide to Biorenewable Resources: A Special Supplement, Biorenewable Resources #3), 9-1 0 .

22 0 D.G.Hayes
Garcia, M.T.; E. Campos; A. Marsal; I. Ribosa. Biodegradability and Toxicity of Sulphonate-based
Surfactants in Aerobic and Anaerobic Aquatic Environments. Water Re5. 2008 (in press).
Gaupp, R.; W. Adams. Di-acetyltartaric Esters of Monoglycerides (DATEM) and Associated
Emulsifiers in Breadmaking. Emuhifirs in Food Technolog; R.J. Whitehurst, Ed.; Blackwell
Publishing: Oxford, UK, 2004;pp. 86-109.
Gielen, D.; J. Newman; M.K. Patel. Reducing Industrial Energy Use and CO, Emissions: The
Role of Materials Science. MRS Bull 2008,33,471-477.
Gressel, J. Transgenics are Imperative for Biofuel Crops. Plant Sci. 2008,174,246-263.
Gujarat Chemicals (Nanpura, India). Ethylene Oxide Derivatives, 2008;http://www.emulsifiers.
inlethylene-oxide-derivatives. htm.
Hatti-Kaul, R.; U. Toernvall; L. Gustafsson; P Boerjesson. Industrial Biotechnology for the Production of Bio-based Chemicals-A Cradle-to-grave Perspective. Trend Biotechnol. 2007,25,
119-124.
Hayes, D.G. Enzyme-Catalyzed Modification of Oilseed Materials to Produce Eco-Friendly Products. J. Am. Oil Cbem. SOC.2004,81,1077-1 103.
H E M . Human and Environmental Risk Assessment on Ingredients of Household Cleaning Products (HEM); Risk Assessments: Linear Alkylbenzene Sulphonate, 2008.
Hill, K. Industrial Development and Application of Biobased Oleochemicals. Pure Appl. Chem.
2007,79,1999-201 1.
Holladay, J.E.; J.J. Bozell; J.E White; D. Johnson. Top Value-Added Chemicals from Biomass.
Volume 11. Results of Screening for Potential Candidates from Biorefinery Lignin, l?N.N.L. US.
Dept. of Energy, 2007.
Honydonckx, H.E.; D.E. De Vos; S.A. Chavan; PA. Jacobs. Esterification and Transesterification
of Renewable Chemicals. Top. Catal. 2004,27,83-96.
International Association of Soaps, Detergents, and Maintenance Products (AISE). AISE Annual
Report 2007,2007;http://www.aise-net.org/downloads/ar~2007.pdf.
Jambhulkar, S.J. Mutagenesis: Generation and Evaluation of Induced Mutations. Ada Bot. Res.
2007,45417-434,
Johansson, I.; M. Svensson. Surfactants Based on Fatty Acids and Other Natural Hydrophobes.
Cum Opin. ColloidIntefae Sci. 2001,6 178-188.
Karmee, S.K. Lipase Catalyzed Synthesis of Ester-Based Surfactants from Biomass Derivatives.
Bio&eh Bioprod. Biorefin. 2008,2,144-154.
Kloepffer, W. Life Cycle Assessment as Part of SustainabilityAssessment for Chemicals. Environ.
Sci. Pollut. Res. Int. 2005,12,173-177.
Lang, S. Surfactants Produced by Microorganisms. Novel Surfartants: Preparation, Applications, and
Biodepahbilig 2nd ed.;. K. Holmberg, Ed.; Surfactant Science Series, Vol. 114.Marcel Dekker:
New York, 2003;pp. 279-315.
Leal, C.; A. Bilalov; B. Lindman. Phase Behavior of a DNA-based Surfactant Mixed with Water
and n-alcohols./. P l y . Chem. B 2006,110, 17221-17229.
Lewis, L. Palm Oil Crisis Driven by Ethanol Fuel Push. ZbeAustralian 2008,July 10,2008(Bwiness Section), 1.

Biobased Surfactants:Overview and Industrial State-of-the-Art0 23

Lu, S.; l? Somasundaran. Intermolecular Packing of Sugar-based Surfactant and Phenol in a Micellar Phase. Langmuir 2007,23,9960-9966.
Lu, J.R.; X.B. Zhao; M. Yaseen. Biomimetic Amphiphiles: Biosurfactants. Curr. Opin. Colloid
Interface Sci. 2007, 12, 60-67.
Marti-Mestres, G.; F. Nielloud. Main Surfactants Used in the Pharmaceutical Field. Drugs Pharm.
Sci. 2000, 105, 1-18.
Masuda, M.; H. Odake; K. Miura; K. Ito; K. Yamada; K. Oba. Biodegradation of 2-sulfonato
Fatty Acid Methyl Ester (a-SFMe). 11. Biodegradation pathways of a-SFMe. J. Jpn. Oil Chem.
SOC.1993,42, 905-909.
Maurad, Z.A.; R. Ghazali; I? Siwayanan; Z. Ismail; S. Ahmad. Alpha-sulfonated Methyl Ester as
an Active Ingredient in Palm-based Powder Detergents.J. Surf: Deterg. 2006,9, 161-167.
McCoy, M. Going Green. Chem. Eng. News 2007a, 85(5), 13-19.
McCoy, M. A New Kind of Clean. Chem. Eng. News 2007b, 85 (12), 29-3 1.
McCoy, M. Converging Pathways: Chemical Companies and Environmentalists Edge Closer
Together in the Pursuit of Sustainability. Chem. Eng. News 2008a, 86(33), 47-46.
McCoy, M. Greener Cleaners. Chem. Eng. News 2008b, 86(3), 15-23.
Mishra, S.; V.K. Tyagi. Ester Quats: The Novel Class of Cationic Fabric Softeners. J. Oleo Sci.
2007,56269-276,

Miyamoto, Y.; M. Sakamoto; T. Maeda; N. Morita. Application of Polyglycerol Mono-fatty Acid

Esters to Improve Breadmaking. FoodSci. Technol. Res. 2005, 11, 19-25.
Morflex Inc. (Greensboro, NC). Pharmaceutical Coatings Bulletin 102-3: Influence of Plasticizers
on the Dissolution and Physical Properties of Ethyl Cellulose Films and Coated Beads, 2008;
http://www.morflex.com/pdf/bul102.3.pdf.
Nelen, B.A.P.; J.M. Cooper. Sucrose Esters. Emuh$ers in Food Technology; R.J. Whitehurst, Ed.;
Plackwell Publishing: Oxford, UK, 2004; pp. 131-161.
Nnanna, LA.; G.Y. Cheng; J. Xia. Potential Applications of Protein-based Surfactants. ProteinBased Sut.factants:Synthesis, PbysicochemicalProperties, and Applications; I.A. Nnanna; J. Xia,
Eds.; Surfactant Science Series; Marcel Dekker: New York, 2001; Vol. 101, pp. 227-260.
Norn, V. Polyglycerol Esters. Emuh$en in Food Technology; R.J. Whitehurst, Ed.; Blackwell Publishing: Oxford, UK, 2004; pp. 110-130.
Overkempe, C.; A. Annerling; C.G. Van Ginkel; l?C. Thomas; D. Boltersdorf; J. Speelman.
Esterquats. Novel Surfactants: Preparation, Applications, and Biodegradability: K. Holmberg, Ed.;
Surfactant Science Series; Marcel Dekker: New York, 2003; Vol. 114, pp. 347-384.
Pandey, V.; R. Banerji; B.S. Dixit; M. Singh; S. Shukla; S.l?Singh. Cuphea, a Rich Source of
Medium-chain Triglycerides: Fatty Acid Composition and Oil Diversity in Cupheaprocumbens.
Eur. J. Lipid Sci. Technol. 2000, 102, 463-466.
Patel, M. Surfactants Based On Renewable Raw Materials: Carbon Dioxide Reduction Potential
and Policies and Measure for the European Union. J. Znd. Ecol. 2004, 7,47-62.
Perkins, W.S. Surfactants-A Primer. Am. Tmtil. Znt. 1998, Mug), 5 1-53.
Phillips, K.; R. Coons; A. Scott; K. Walsh. Soaps and Detergents: Producers Race to Lower Costs
and Go Green. Chem. Week 2008, 170(3),23-25.

24 0 D.G. Hayes

Phillips, K.; E. DAmico; K. Walsh. Soaps and Detergents: Measuring Out Costs. Chem. Week

2007,167(3),23-28.
Plasman, V.; T. Caulier; N. Boulos. Polyglycerols. Versatile Ingredients for Personal Care. Household Pers. Prod. Ind. 2004,41,94-97.
Plasman, V.;T. Caulier; N. Boulos. Polyglycerol Esters Demonstrate Superior Antifogging Properties for Films. P h t . Addit. Compd. 2005,7, 30-33.
Quadri, G.; X. Chen; J.W. Jawia; F. Tambone; I? Genevini; F. Faoro; F. Adani. Biobased Surfactant-Like Molecules From Organic Wastes: The Effect of Waste Composition and Composting
Process on Surfactant Properties and on the Ability to Solubilize Tetrachloroethene (PCE).
Environ. Sci. Zchnol. 2008,42,261 8-2623.
Rahman, EK.S.M.; E. Gakpe. Production, Characterization and Applications of BiosurfactantsReview. Biotechnohgy (Faisalabad, Pakistan) 2008,7, 360-370.
Roberts, D.W.; L. Giusti; A. Forcella. Chemistry of Methyl Ester Sulfonates. INFORM2008,
(Surfactantsj?om Biorenewable Sources: A Special Supplement, Biorenewable Resources #5), 2-9.
Roundtable for Sustainable Palm Oil (RSPO). Promoting the Growth and Use of Sustainable
Palm Oil, 2008;http://www.rspo.org/.
Sagalowicz, L.; M.E. Leser; H.J. Watzke; M. Michel. Monoglyceride Self-assembly Structures as
Delivery Vehicles. Trends Food Sci. Technol. 2006,17, 204-214.
Sakamoto, K. Current Market Developments and Trends in Amino Acid- and Protein-Based
Surfactants. Protein-Baed Surfactants: Synthesis, Physicochemical Propedes, and Applications; LA.
Nnanna; J. Xia, Eds.; Surfactant Science Series; Marcel Dekker: New York, 2001;Vol. 101,pp.

261-280.
Santosa, S.J. Palm Oil Boom in Indonesia: From Plantation to Downstream Products and Biodiesel. Clean: Soii Air, Water 2008,36453-465.
Schalitz, W.J. The Role Awaiting Bio Based Surfactants in the Industrial Cleaning Market. 78th
American Oil Chemists Society Annual Meeting, Quebec City, Canada, 2007.
Scheibel, J. The Impact of Feedstocks on Future Innovation in Surfactant Technology for the
Detergent Market. 78th American Oil Chemists Society Annual Meeting, Quebec City, Canada,

2007.
Scott, E.; F. Peter; J. Sanders. Biomass in the Manufacture of Industrid Products-The Use of
Proteins and Amino Acids. Appl. Microbiol. Biotechnol. 2007,75,751-762.
Shelley, S. A Renewable Route to Propylene Glycol. Chem. Eng. Prog. 2007,103,G-9.
Sissell, K. Sierra Club Asks EPA to Ban NPEs in Commercial and Household Detergents. Chem.
Week 2007,167(20), 45.
Smith, J.; L. Hong. FoodAdditives Databook; Blackwell Publishing: Oxford, UK, 2003.
Solvay Chemicals (Deer Park, TX). Polyglycerol in Personal Care (Technical Bulletin
PGL-03-00 l), 2004;http://www.solvaychemicals.us/static/wma/pdf/6/6/9/2/PGL-03-001
.p&.
Solvay Chemicals (Deer Park, TX). Polyglycerols in Industrial Applications (Technical Bulletin
TDS-PGL-05-00 1), 2005;http://www.solvaypolyglycerol.com/docroot/glycerol/static-files/attachments/polyglycerol-in_industrial-applications-li~t.pd~
Sparso, F.V.; N. Krog. Propylene Glycol Fatty Acid Esters. Emulsifiers in Food Technology; R.J.

Biobased Surfactants:Overview and IndustrialState-of-the-Art 0 25

Whitehurst, Ed.; Blackwell Publishing: Oxford, UK, 2004; pp. 186-205.

Svensson, M.; J. Brinck. Surfactants Based on Sterols and Other Alicyclic Compounds. Novel
Surfactants: Preparation,Applications, and Biodegrahbilig 2nd ed; K. Holmberg, Ed.; Surfactant
Science Series; Marcel Dekker: New York, 2003; Vol. 114, pp. 217-239.
Trouve, G. Utilization of Ethoxylated Fatty Acid Esters as Self-EmulsifiableCompounds, WO
Patent 19960118 (1996).
U.S. Department of Energy. The Technology Roadmap for PlantlCrop-Based Renewable Resources 2020, 1999.
US. Senate Committee on Agriculture Nutrition and Forestry. Energy Title (Title K) of the Farm
Security and Rural Investment Act of 2002, U.S. Senate, 2006.
van Rannvijk, F.; M. Woudenberg-van Oosterom; RA. Sheldon. Glycosidase-catalyzedSynthesis
ofAlkyl Glycosides.J. MoL Catal. B: Enzym. 1999, 6, 51 1-532.
Von Rybinski, W.; K. Hill. Alkyl Polyglycosides. Novel Surfactants: Preparation,Applications, and
Biodegradzbilig 2nd ed.;. K. Holmberg, Ed.; Surfactant Science Series; Marcel Dekker: New
York, 2003; Vol. 114, pp. 35-93.
Watkins, C. Detergents Go Green. INFORM2007, 18, 780-783.
Werpy, T.; G. Petersen. Top Value-Added Chemicals from Biomass, Volume I, Results of Screening for Potential Candidates from Sugars and Synthesis Gas, US.Department of Energy, 2004.
Xia, J.; I.A. Nnanna; K. Sakarnoto. Amino Acid Surfactants: Chemistry, Synthesis, and Properties.
Protein-BasedSurfactants: Synthesis, Physicochemical Properties, and Applications; LA. Nnanna; J.
Xia, Eds.; Surfactant Science Series; Marcel Dekker: New York, 2001; Vol. 101, pp. 75-122.
Xu, C.; l? Taylor; ED.1. Fletcher; V.N. Paunov. Adsorption and Hybridization of DNA-surfactants
at Fluid Surfaces and Lipid Bilayers. J. Mater. Chern. 2005, 15,394-402.

.*

PART2

Biosynthesis of Rhamnolipids and

Sophorolipids

Production and Modification of

Sophorolipids from Agricultural
Feedstocks
Richard D. Ashby, Daniel K.Y. Solaiman, and Jonathan A. Zerkowski
U.5. Department ofAgriculture,Agricultural Research Service, Eastern Regional Research Center,600 E.
Mermaid Lane, Wyndmoor, Pennsylvania 19038

Introduction
Surfactants or Surhce Acting Agents are molecules that are made up of hydrophobic
and hydrophilic components allowing them to either concentrate at interfaces such
as aidwater or oil/water and act to reduce surface and/or interfacial tension or selfassociate to form micelles or other nanostructured aggregates. Generally, surfactants
are classified as either ionic (e.g., alkyl sulfonates, anionic, phospholipids, zwitterionic;
or quaternary ammonium compounds, cationic) or nonionic (e.g., alkyl glucosides or
alkyl ethoxylates) based on their chemical structure. Regardless of their classification,
surfactant effectiveness tends to be concentration-dependent; increasing surfactant
concentrations beyond a critical value, unique to each surfactant, known as the critical
micelle concentration (CMC), causes micellar aggregation to occur. Thus above the
CMC many surfactants do not exhibit additional changes in surface and/or interfacial
tension.
Many microorganisms have the capability to naturally synthesize surfactanttype molecules. Many of these so-called biosurfactants express a broad range of
functional properties including emulsification, phase partitioning, wetting, foaming,
surface activity, and in some cases potential clinical applications and as such (along
with the mounting concern for the environment and the persistent rise in petroleum
prices) are drawing increased industrial interest for various applications traditionally
occupied by petroleum-based surfactants. Many different biosurfactants have been
identified. Some, such as glycolipids, neutral lipids, and lipopeptides, are of relatively
small molar mass while others may form polymeric structures or be created through
complex chemical interactions that result in larger molar mass materials. Examples of
such large molar mass biosurfactant materials include lipoproteins, lipopolysaccharide-

29

30 0 R.D. Ashby et at.

protein complexes, and polysaccharide-protein-fatty acid complexes (Muthusamy et

al., 2008).
Glycolipids are one class of biosurfactants that are receiving industrial
consideration. Many of the common glycolipids possess a chemical structure whose
hydrophobic region is generally composed of a long-chain fatty acid or hydroxy
fatty acid, and whose hydrophilic region is often a disaccharide. Three of the more
common natural glycolipid biosurfactants are rhamnolipids (whose components
generally include rhamnose sugar and P-hydroxyalkanoic acids), mannosylerythritol
lipids (4-O-(mono or di-O-acetyl-di-O-alkanoyl-D-mannopyranosyl)-erythritol),
and sophorolipids (most commonly composed of sophorose sugar and fatty acids)
each of which display surface-active properties, much like synthetic surfactants, but
which also afford inherent advantages over synthetic surfactants in that they are
synthesized by fermentation from renewable resources, are generally nontoxic and are
readily biodegradable in nature.
Sophorolipids are synthesized by yeasts principally belonging to the genus Candidz
(formerly Torulopsis). First described in 1961 (Gorin et al., 1961), sophorolipids
consist of 2 glucose units linked [P]-1,2 (sophorose; 2-0-[~]-D-glucopyranosyl[PI-D glucopyranose) and a fatty acid region attached at the reducing end of the
sugar through a glycosidic linkage. The fatty acid is generally linked to the sophorose
in an [ a ] or [w]-1 position and the primary hydroxyl groups at the 6 and 6 positions
on the sophorose sugar may be acetylated (Otto et al., 1999). The fatty acid chain
length varies between 16 and 18 carbons (Rau et al., 1996) (one exception is the
sophorolipids synthesized by R. bogoriensis,which include fatty acid side-chains of 22
and 24 carbons; Nuhez et al., 2004), and may be saturated or unsaturated (Davila
et al., 1994). They occur as mixtures of lactone (generally most prevalent; with the
carboxyl terminus of the fatty acid esterified to the 4-position of the sophorose sugar)
and open-chain forms (Nuhez et al., 2001) (Fig. 2.1). The most-studied sophorolipid
production system belongs to Candida bombicola ATCC 22214 although other species,
such as C.apicokz (Stiiwer et al., 1987; Hommel & Huse, 1993)) Tomlopsis magnoliae
(Gorin et al., 1961) and Rhodotomla bogoriensis (Tulloch et al., 1968a; Spencer et al.,
2002; Nuhez et al., 2004) have also been shown to produce sophorolipids.

Sophorolipid Biosynthesis
Introduction and Applications of Sophorolipid Biosurfactants
One of the proposed hnctions of sophorolipids in the producing organisms is to
assist in accessing lipophilic substrates (It0 & Inoue, 1982), but their large production
capacity and amphiphilic character have increased awareness for promising industrial
applications (Solaiman et al., 2004a). The industrial potential of these molecules is
dependent on their surface-active properties, which are dictated by the molecular
distribution of the open-chain and lactone structures. While a large capability
application has yet to be discovered, sophorolipids do have the capacity to lower
surface and interfacial tension (Cooper & Paddock, 1983) thus providing a potential

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks0 3 1

HO
0
A

Fig. 2.1. Structures of 17-L-[(2'-0-~-glucopyranosyl-~-D-glucopyranosyl)-oxyl-9-octadecenoic

acid
6:6"-diacetate sophorolipidsin the 1; <-lactone form (A) and the free acid form (B).

mechanism for use in many diverse applications (Banat et al., 2002). In fact,
sophorolipids have been reported to be active as primary surfactants for shampoos,
body washes, and detergents (Hall et al., 1995; Inoue et al., 1980) and as emulsifiers
for skin care products (Mager et al., 1987). In addition, they have been reported to
have applications as food encapsulants (Allingham, 1971), as degreasing agents (Hall et
al., 1995), and to enhance soil bioremediation and waste water treatment (Makkar &
Cameotra, 2002; Mulligan et al., 2001). Some reports have suggested that the lactone
form ofsophorolipids can be utilized as a bacteriostatic agent (Mager et al., 1987), has
anticancer properties (Scholz et al., 1998; Isoda et al., 1997), can be used to stimulate
skin fibroblast metabolism (Boneix, 2003), and treat skin diseases (Maingault, 1997).
In contrast, the acidic form of sophorolipids has been shown to be therapeutically
active for skin treatment, particularly as agents for fibrinolysis (promoting healing),
desquamation, depigmenting, macrophage activation (Maingault, 1999), and
as moisturizing agents (Abe et al., 1981; Tsutsumi et al., 1981). In addition, the
unique sophorolipid structure has increased interest in their use as a precursor for
the production of specialty chemicals such as sophorose, a known inducer of hngal
cellulase enzymes (Sternberg & Mandels, 1980), monohydroxy fatty acids (Rau et
al. 2OO1), and other derivatives (Zerkowski & Solaiman, 2006, 2007; Zerkowski et
al. 2008). To a limited extent, structural variation (and hence control over physical
properties) can be achieved by changing the lipidic carbon source, which alters the
sophorolipid fatty acid content (see below).

Saccharide/Lipidic Feedstocks
Since the 1980s research has focused on the use ofsimple monosaccharides (primarily
glucose) and lipids as substrates for sophorolipid synthesis. In order to minimize
production costs, low cost substrates, such as triacylglycerols and free fatty acids, were

32 0 R.D. Ashby et al.

thought to be good candidates for feedstock utilization thus providing an additional

outlet for these commodity materials. Since then, C bombicokz has been tested
extensively to ascertain its ability to synthesize sophorolipids from these materials in
yields that would allow industrial application. By varying the fermentation conditions,
many different research groups have successfully synthesized sophorolipids from a
number of different lipidic materials. Initially, Ito and Inoue reported sophorolipid
yields of 32 g/L when grown in shake flask cultures on a glucose/safflower oil medium
(It0 & Inoue, 1982). These production numbers were improved to 140 g/L by Zhou et
al. who studied the influence of the media composition on sophorolipid biosynthesis
from glucose and safflower oil (Zhou et al., 1992). Later, in the mid to late 1990s
sophorolipid yields were improved to over 150 g/L (Table 2.1) thus improving the
economics of sophorolipid production and making potential commercial use more
feasible. In fact, Daniel et al. reported production capacities as high as 422 g/L
(Daniel et al., 1998) from a two-stage fed batch process where deproteinized whey (a
waste product from the cheese industry) was initially used as a substrate to cultivate
the yeast Cyptococcus curvatus ATCC 20509 and use the lactose to produce a single
cell oil which was then used along with rapeseed oil as the medium for growth and
sophorolipid production by C. bombicoh. In most instances the medium used to
produce sophorolipids consists of glucose and a lipid source. The most common lipid
sources include triacylglycerols and free fatty acids however, fatty acid esters (methyl
and ethyl), n-alkanes (e.g., hexadecane), and 2-alkanols (e.g., 2-dodecanol) have also
been reported to support sophorolipid biosynthesis (Table 2.1).
A thorough examination of the sophorolipid structural variants produced from
C. bombicokz when grown on various free fatty acids, including pdmitic acid, stearic
acid, oleic acid, and linoleic acid, by Liquid Chromatography/Atmospheric Pressure
Chemical Ionization - Mass Spectroscopy (LCIAPCI-MS; for chromatograms see
Fig. 2.2) (Nufiez et al., 2001) revealed that under the growth conditions used all were
synthesized in greater than 92% lactone form with the predominant fatty acid present
reflective of the fatty acid feedstock. This allowed some control over the surfactant
properties (especially the CMC values) exhibited by each product and an opportunity
to use blending as a means to fine-tune those properties (Ashby et al., 2008).

Co-product Feedstocks
In addition to the lipid sources described above, inexpensive co-product streams such
as crude glycerine from biodiesel manufacturing (Ashby et al., 2005), soy molasses
from soybean processing (Solaiman et al., 2004b), and oil refinery waste including
soapstock and post-refinery fatty acids (Bednarski et al., 2004) have also been reported
as co-substrates in the production of sophorolipids. Biodiesel is synthesized by the
chemical transesterification of triacylglycerols with short chain alcohols, as shown in
Fig. 2.3.
Over the past few years the biodiesel industry has advanced rapidly throughout
the world because it is an attractive alternative to petroleum diesel as it burns cleaner

Productionand Modificationof Sophorolipidsfrom Agricultural Feedstocks 0 33

Table 2.1. Maximum Reported Sophorolipid Yields Produced by C. bombicola from Various
Lipidic Substrates (Unless Otherwise Noted the Carbohydrate Source Was Glucose)
~

Lipid Source

Yield (g/L)

Reference

Triacvlalvcerols:
Sunflower Oil

172

Davila et al., 1994

Safflower Oil

140

Zhou et al., 1992

Canola Oil

160

Zhou & Kosaric, 1995

Soybean Oil

33

Hu & Ju 2001; Asmer et al., 1988

Rapeseed Oil

255

Davila et al., 1994

Palm Oil

82

Davila et at., 1994

Fish Oil

51

Davila et at.. 1994

Rapeseed Oil'

422

Daniel et al., 1998

Corn Oilb

400

Pekin et at.. 2005

Animal Fat

120

Deshpande& Daniels, 1995

Waste Frying Oil

50

Fleurackers, 2006

Free Fatty Acids:

Palmitic Acid

42

Ashby et al., 2008

Stearic Acid

77

Ashbv et al.. 2008

Oleic Acid

180

Rau et al., 1996

Linoleic Acid

40

Ashby et at., 2008

340

Davila et al., 1994

Esters:
Rapeseed Oil EE
Sunflower Oil ME

235

Davila et at.. 1994

Palm Oil ME

240

Davila et at., 1994

Linseed Oil ME

122

Davila et at., 1994

Stearic Acid ME

32

Asmer et al., 1988

55

Hu & Ju, 2001

22

Brakemeier et al., 1998

Alkanes:
Hexadecane

2-Al kanols:
2-Dodecanol

'Carbohydrate source was deproteinized whey.

bCarbohydratesource was a mixture of glucose and honey.

34 0 R.D. Ashby et al.

SL-LA SL-PA
Lactone Lactone

SL-OA
Lactone

SL-SA
Lactone

Min. ST = 35 mN/m
CMC = 35 mg/L

o i ,

, , , ,

,) , , ) i , ~ , ? l ? * ~ , l & , t

Min. ST = 36 mN/m
CMC = 250 mg/L

Yo :
0

i.L
I

* . 9

Time (min)
Fig. 2.2. Liquid chromatograms of the region corresponding to the diacetylated lactonic forms of the
sophorolipids derived from glucose and palmitic acid (SL-PA, 92% lactone; A), stearic acid (SL-SA, 99%
lactone; B), oleic acid (SL-OA, 100% lactone; C), and linoleic acid (SL-LA, 98% lactone; D). Minimum
surface tensions (Min. ST) and critical micelle concentrations (CMC)were measured in a waterhapeseed
oil system (Ashby et al., 2008).

Triacylglycerol

Fatty Acid Esters

(Biodiesel)

Glycerol

Fig. 2.3. Chemical transesterificationreaction in the formation of fatty acid esters (biodiesel) with the
simultaneous formation of glycerol ( R l , R2, and R3 correspond to individual fatty acids).

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks 0 35

(with reduced emissions) and is sustainable through agricultural efforts. According

to the National Biodiesel Board, biodiesel production in 2007 in the United States
alone was estimated at 450 million gallons and is expected to continue to climb over
the next few years (National Biodiesel Board, 2007). This expected increase however,
will depend largely upon feedstock limitations including the price and availability
of vegetable oil substrates (primarily soybean oil in the U.S.); petroleum diesel and
methanol and may also be affected by the food vs. fuel debate.
For every 100 pounds of biodiesel produced, I0 pounds of crude glycerine is
generated. This translates to a crude glycerine yield of 45 million gallons in the United
States alone in 2007, exceeding its application capacity. This glut has driven the price
of glycerine down to historic lows of between $0.35 and $0.50 per pound for refined
glycerine and as low as $0.02 per pound for crude glycerine (Kotrba, 2007). At the
same time the new demands for biodiesel have caused the prices for soybean oil, the
primary biodiesel feedstock, to climb to historic highs of $0.58 per pound as of the
end of July 2008 (TFC Commodity Charts, 2008).
One of the economic strategies for the production of any new chemical includes
the use of low-cost raw materials in order to improve profit margins. Currently,
surfactant producers are finding it more and more difficult to generate revenue due
to the high cost of petroleum but at the same time biosurfactants are becoming more
expensive due to the increasing price of triacylglycerols and free fatty acids, as they
are increasingly shifted towards biodiesel production. This situation has increased
the opportunity to develop green surfactants by using cheap co-products as
starting materials. By using glycerine, surfactant producers can offer a supplementary
application to utilize glycerine (thus helping to maintain its value), and create an
environmentally friendly product without the hindrance of increasing petroleum
prices.
The chemical composition of the crude glycerine streams vary from reaction to
reaction but essentially are composed of glycerine and water with small concentrations
of alcohol (usually methanol), free fatty acids, fatty acid alkyl esters, and tri-, di- and
monoglycerides. The compositional differences that do occur in the crude glycerine
stream are by and large the result of the source of the lipid feedstock, the point of the
biodiesel synthetic process from which the crude glycerine was obtained (separation
stage vs. evaporation stage), the efficiency of the transesterification process, and
recovery effectiveness. The substitution of crude glycerine (with its accompanying fatty
acid component) in place of the more expensive glucose and pure free fatty acid as the
substrate material may result in reduced production costs for sophorolipids (provided
yields can be maintained) and offer a supplementary outlet for crude glycerine, thus
minimizing isolation and refinement costs for high purity applications. We showed
that C. bombicola does indeed grow and produce sophorolipids from pure glycerine
and crude glycerine at 10 g/L and 60 g/L, respectively (Ashby et al., 2005). In that
study, we used a crude glycerine sample that was derived from the separation stage of
a soybean oil-based biodiesel production system that was composed of 40% glycerine,

36 0 R.D. Ashby et al.

32% free fatty acid/fatty acid methyl ester, 2% partial glycerides, and 26% water. The
difference in sophorolipid productivity between pure glycerine and the crude material
was at least partially due to the presence of a relatively large exogenous fatty acid
source and the decreased osmotic stress put on the organism when crude glycerine was
used. In addition, by using crude glycerine that contained an appreciable amount of
fatty acid methyl esters, the conformational equilibrium was shifted from the lactonic
form towards the open-chain form. This was brought about by the presence of the
methyl group on the carboxyl end of the fatty acid which inhibited lactonization,
resulting in a larger concentration of open-chain sophorolipids, thereby facilitating
property control and permitting post-synthetic modification of the fatty acid sidechain without the need to chemically break the lactone ring. The chemical structures
of the sophorolipid products derived from the glycerine samples were determined
by LC/APCI-MS (Ashby et al., 2005). The results proved that the structures of the
sophorolipids derived from pure and crude glycerine were indeed different (Fig. 2.4).
When the yeast was grown with pure glycerine, LC/APCI-MS analysis revealed the
formation of the sophorolipid lactone forms with the majority of the fatty acids being
either palmitic, stearic, or oleic as the main products (99%) with negligible amounts
of open-chain form. In contrast, the use of crude glycerine resulted in sophorolipids
that favored the open-chain form, accounting for GO% (24% free acid and 36%
methyl esters) of the total sophorolipid content with oleic acid and linoleic acid
predominating.
Soy molasses is aco-product stream from soybean processing (Fig. 2.5).It is derived
from the whole bean after flaking and removal of the crude soy oil through solvent

aL

Time (min.)
Fig. 2.4.The total ion current (TIC) chromatogramsof the sophorolipids derived from pure glycerine (A)
and crude glycerine (co-product of biodieselsynthesis), B).The peaks eluting above 17 min. correspond
to lactonic sophorolipids while the peaks eluting between 2 min. and 17 min. correspond to open-chain
sophorolipids (specifically,peaks eluting prior to 10 min. are free acid, open-chain moleculeswhile those
eluting between 10 and 17 min. correspond to the methyl esterified, open-chain forms) (Ashby, 2005).

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks0 37

extraction. Then, after a grinding and sizing step to produce soy flour and soy protein/
concentrate, the remaining material is soy molasses. Soy molasses is high in potentially
fermentable carbohydrates (accounting for approximately 30% of the total solids) and
therefore has the potential to be a valuable feedstock in microbial fermentations. In
addition, the cost of this material can be as low as 20% of the commonly used glucose
substrate. Since feedstock costs are generally the driving force behind the economics
of biobased production systems, the inexpensive soy molasses could potentially result
in a large cost-saving measure for sophorolipid biosynthesis. Soy molasses is primarily
composed of sucrose, raffinose, and stachyose as soluble carbohydrates. When using
soy molasses in place of glucose in sophorolipid fermentations, we found that soy
molasses could indeed function as a substitute for glucose (Solaiman et al., 2004b)
Whole
Soybeans

Cleaning
Cracking
Dehulling
Flaking
Soy Flakes
I

Solvent Extraction
Crude
Soy Oil

Spent
Flakes

Soy Flour

Grinding
Sizing

Refining
Soy Protein
Concentrate
Isolate

Soy Lecithin

Coproduct
Soy Molasses

Fig. 2.5. Soybean processing and the generation and carbohydrate content of soy molasses.

Refined
Soy Oil

38 0 R.D. Ashby et at.

but only the sucrose fraction of the soluble carbohydrates was actually utilized by the
organism (Solaiman et al., 2007). The raffinose and stachyose remained unutilized
throughout the fermentations. The inability of the organism to utilize all the available
soluble carbohydrates resulted in diminished yields (Table 2.2) which were further
diminished when yeast extract and urea were completely omitted from the growth
media. A solution to this dilemma is currently being explored by our efforts to express
a-galactosidase activity in C. bombicokz, thus providing the organism a means to liberate
the a-D-galactose monosaccharides from the stachyose and raffinose molecules, thus
making more of the soluble carbohydrates utilizable to the organism. While yields
were affected by the change from glucose to soy molasses, sophorolipid structural
composition remained relatively constant. By using soy molasses in conjunction
with oleic acid as the lipid co-substrate, 97% of the sophorolipids were synthesized
in the lactone form. In comparison, glucose/oleic acid fermentations resulted in
sophorolipids that were greater than 99% lactonic. In addition, the distribution of
sophorolipid lactones between the two systems containing yeast extract and urea was
also comparable. As expected, C18:1 lactones predominated in both the glucose/oleic
acid and soy molasses/oleic acid fermentations with relative abundances of 80 and
8 I%, respectively. However, when yeast extract and urea were omitted from the growth
media the lactone distribution did show variability. From this data, we concluded that
soy molasses can be an effective substitute for glucose for sophorolipid fermentations
without sacrificing structural continuity. Also,when using soy molasses, it is possible
to omit the yeast extract and urea, thus reducing media cost, and still obtaining the
desired sophorolipid products. Success in the utilization of these abundant, low cost
co-product streams provides a new application for these materials and helps reduce
the cost of sophorolipid synthesis and co-product disposal.

Table 2.2. Maximum Volumetric Yields and Lactone Distribution for Sophorolipids
Produced by C. bombicolu from the Fermentation of Glucose or Soy Molasses with Oleic
Acid as the Lipid Source According to Hydroxy Fatty Acid Content
Substrates
Hydroxy Fatty Acid Content (96)'
Carbohydrate
Lipid
Yield (Q/L)~ C16:O
C18:O C181
C182 Others'
Glucose
Oleic acid
100
8(3)d
2(2)
80(71) 3
1
Soy molasses
Oleicacid
75
4(3)
6(6)
81(68) 5
4
Soy molasses*
Oleic acid
53
17(17) 12(11) 57(38) 7
7
'Percentages were determined based on peak integration of the total ion current (TIC) of LC/
APCI-MS ( Nuiiez et al., 2001).
bExpressedas g purified sophorolipid/L (starting volume) of culture broth. Values are the
averages of two determinations with standard deviations of ~ 1 0 % .
'These include the C170, C18l-monoacetylated, and C18:3 sophorolipid species.
dAll parentheticalvalues under"Hydroxy fatty acid content" correspond to the % of that
sophorolipid that was in the w-1 conformation. Those entries without parentheticalvalues
were unresolvableunder the LC/APCI-MS method used.
'Yeast extract and urea were completely omitted from the growth media.

Production and Modificationof Sophorolipidsfrom Agricultural Feedstocks 0 39

Biochemistry/Enzymology
Research on sophorolipids in the past decades has largely focused on the development
of fermentation processes to generate large amounts of these ecofriendly materials
and bioengineering methods to achieve an overall lowering of the production costs.
Comparatively, fundamental research on the biochemical pathways for the biosynthesis
of sophorolipids from C. bombicokz is limited even though production from numerous
feedstocks has been reported. Earlier biochemical studies on the enzymes putatively
involved in the biosynthesis of sophorolipid in Rbodotorukz bogoriensis (formerly
Cundidu bogoriensis) have nevertheless provided valuable information important to
the understanding of the biosynthesis pathway for this glycolipid in related yeast
strains. The production of an extracellular sophorolipid by R. bogoriensis was first
reported by Tulloch et al. (1968a). These investigators determined from chemical and
NMRstudies that the structure of the primary sophorolipid produced by R. bogoriensis
was 13-[(2-0-[~]-D-glucopyranosyl-[~]-D-glucopyranosyl)-oxy]docosanoic
acid
in its 6-mono- and 6,6-diacetylated forms. Ruinen (1963) and Ruinen and
Deinema (1 964) in fact had earlier described the production of extracellular lipids
in R. bogoriensis,but the definitive chemical structures of those compounds were not
elucidated.
Very recently, Nuhez et al. (2004) employed a previously developed LC/APCIMS method (Nuhez et al., 2001) to confirm the finding of Tulloch et al. (1968a)
that sophorolipids produced by R. bogoriensis contained a 22-carbon hydroxy fatty
acid (C22). In that same study, Nuhez et al. further reported that a small quantity
(< 10%) of the sophorolipids isolated from R. bogoriensh actually contained a hydroxy
fatty acid with 24 carbons (Nuiiez et al., 2004). Esders and Light (1972a) reported
that the sophorolipids of R. bogoriensis exist in di-, mono-, and un-acetylated forms,
in which the primary alcohol functions of the sophorose moiety were variously
esterified to an acetyl group. These investigators then pioneered the biochemistry
of sophorolipid biosynthesis in R. bogoriensis. The first enzymes they characterized
were the glycosyl- and acetyl-transferase activities of this yeast (Esders & Light,
1972b). With a combination of various column chromatographic techniques
(i.e., DFAE-Sephadex anion exchange, hydroxyapetite, and gel filtration) and a
disc electrophoretic separation step, these authors described the identification of
glucosyltransferases 1 and 2, which catalyze the addition of a glucose molecule in its
activated form (i.e., UDP-glucose) to the 13-hydroxydocosanoic acid (13-HDA) and
methyl 13-glucopyranosyloxydocosanoate,respectively, to form the mono- and diglucopyranosyl-13-HDA (i.e., sophorolipid). The two glucosyltransferases, however,
were co-purified throughout all the procedures and could not be individually separated
and/or isolated. The same study also described the presence of acetyltransferaseactivity
in the organism which was capable of catalyzing the transfer of an acetyl group from
acetyl coenzyme A (acetyl-CoA) to a mono- or un-acetylated sophorolipid. The
mono-glucopyranosyl- 13-hydroxydocosanote, however, was a poor substrate for this
acetyltransferase activity. The temporal appearance of these enzyme activities during

40 0 R.D. Ashby et al.

fermentation was also monitored. The glucosyltransferase activities were found to
reach a maximal value at 1.5 days of fermentation, while the level of acetyltransferase
reached the highest point on the third day of culture growth. Taken together, the
results of this study indicated that the acetylation of sophorolipids typically occurred
after the di-glucopyranosyl- 13-hydroxydocosanoate had been fully synthesized.
In the course of their study on the enzymes of sophorolipid biosynthesis, Esders
and Light (1972c) also identified in R. bogoriensis a glycosyltransferase activity, named
uridine diphosphate (UDP) g1ucose:sterol glucosyltransferase, which catalyzed the
transfer of activated glucosyl groups to a sterol, such as ergosterol (the ingenious
substrate) and cholesterol. The relationship between this UDP-g1ucose:sterol
glucosyltransferase and the previously reported glucosyltransferases 1 and 2 was
not apparent because data on the reactive nature of these enzymes toward the lipid
acceptors of the other transferases were not presented.
In a series of two papers, Bucholtz and Light (1976a, 1976b) hrther examined
the enzymatic acetylation of R. bogoriensis sophorolipids. Also, the effects of glucose
substrate and the yeast extract nitrogen source on the levels of acetyl- and glucosyltransferases were investigated (Cutler & Light, 1979). In that study, the investigators
found that either a low glucose or a high yeast extract concentration led to decreased
total yields of sophorolipid as well as slower rates of biosynthesis. Enzymatic assays
further showed that a low glucose concentration in the medium depressed the level
of glucosyltransferase, but the activity of acetyltransferase remained unaffected. The
study provided valuable information regarding the role of the carbohydrate substrate
in regulating the biosynthesis pathway of sophorolipid in R. bogoriensis.
Cutler and Light (1982) further observed that the biosynthesis of long chain fatty
acids, especially those with C22 and C24 chain-lengths found in the sophorolipids of
R. bogoriensis, were induced only in the presence of high glucose concentrations. The
glucosyltransferases 1 and 2 of R. bogoriensis,which had been proposed to catalyze the
first and second addition of glucose to the hydroxy fatty acid to form sophorolipids,
resisted all attempts to be individually purified. Breithaupt and Light (1 982) employed
an affinity chromatography technique to attempt the separation of the two enzymes.
The affinity groups included the 13-hydroxydocosanoic acid and its unacetylated and
diacetylated sophorolipids. Their results showed that both enzymes exhibited near
identical binding properties to the affinity gels, frustrating further efforts to purify
each activity separately.
The involvement of any of these enzymes in the biosynthesis pathway of
sophorolipid in R. bogoriensis still awaits definitive confirmation. For example, the
UDP-g1ucose:sterol glucosyltransferase activity described by Esders and Light (1 972c)
might not have relevance to sophorolipid biosynthesis. In fact, Kastelic-Suhadolc
, (1980) reported the extraction of a P-cholesteryl glucoside from the cell wall of R.
bogoriensis, and further co-localized this glucoside and sterol glucosyl transferase
activity to a single fraction of a cell-wall membrane preparation, suggesting that the
glucosyltransferase may be linked to the biosynthesis of P-cholesteryl glucoside.

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks 0 41

Another important metabolic pathway in the biosynthesis of sophorolipid is the

hydroxylation reaction of the fatty acid prior to its incorporation into the glycolipid
molecule. It is generally believed that cytochrome P450 enzyme(s), which are prevalent
in yeasts and fungi (Gppeli, 1986; Van den Brink et al., 1998), are featured in the
synthesis of the hydroxy fatty-acid moiety of sophorolipid. Because of the presence
of many cytochrome P450s in any particular yeast or fungal organism, however,
the mere identification of the enzyme activity does not represent its unequivocal
involvement in the sophorolipid biosynthesis pathway. In this context, Heinz et al.
(1969, 1970) reported the hydroxylation of fatty acids using the cell-free extracts
prepared from a strain of sophorolipid-producing Torulopsis.Years later, Hommel et
al. (1994) reported the kinetics of cytochrome P450 synthesis in Candida apicokz,
a sophorolipid-producing organism. The results showed that this enzyme activity
increased during the stationary phase of cell growth, which coincided with the onset
of sophorolipid synthesis. Lottermoser et al. (1996) first described the cloning of two
cytochrome P450 genes from Candidz apicokz. Based on sequence analysis of the
newly cloned genes with respect to the information found in the cytochrome P450
database, these investigators assigned the designates CYP52E1 and CYP52E2 to the
isolated genes. The hnctional characterization of the gene products, CYP52E1 and
CYP52E2, however, was not reported. Consequently, their roles in the biosynthesis
pathway of sophorolipid could not be confirmed.
The enzymatic and genetic studies over the years, while not extensive, have laid
important groundwork for the elucidation of the metabolic pathway of sophorolipid
biosynthesis. Additional research is needed, through concerted efforts in genetic
and biochemical methodologies, to conclusively map the pathway of sophorolipid
biosyn thesis.

Post-synthetic EnzyrnaticKhemical
Modifications of Sophorolipids
In an effort to expand the applications of sophorolipids, many chemical and enzymatic
processes have been reported that seek to modify the sugar or fatty acid moieties of
sophorolipids. Enzymes have been used to alter the structure of sophorolipid in two
principal ways, either to modify the carbohydrate moieties or to modify the terminal
carboxyl group on the fatty acid moieties. The former approach was the first to be
investigated in a degradative sense. The earliest report used acetylesterase (EC 3.1 .I .6)
to remove both the 6 and 6 acetyl groups while maintaining the macrolactone ring
(Asmer et al., 1988). The analytical yield was reported to be only 30%, but the product
was not isolated. The authors suggested that hrther degradation andlor concurrent
opening of the lactone ring occurred, limiting yield. By contrast, later reports used
a hngal cutinase expressed in E. coli (de Koster et al., 1995) or immobilized C
rugosa lipase (Otto et al., 1999) and found that the 6 acetyl group was removed
preferentially, again keeping the lactone intact. Molecular modeling concurred with
this observation: a structural model derived from NMR data (de Koster et al., 1995)

42 0 R.D. Ashby et al.

showed that the 6 acetyl group was hidden between the lipid chain and the sophorose
unit and so should be less accessible to enzymatic cleavage.
Proceeding in the synthetic direction, Bisht et al. (1999) showed that Novozyme
435 (immobilized lipase from Candida antarctica) could be used to add additional
acyl groups, either acetyl, acryloyl, or succinoyl, to the carbohydrates of acyclic
sophorolipid esters. With an excess of acylating reagent, both the 6 and 6 hydroxyl
groups were modified to give the diacyl compounds. In the absence of an external
acylating group, or with only one equivalent, the terminal fatty acid carboxyl group
reacted intramolecularly to form a macrolactone different from the native one. NMR
data showed that the new lactone had formed at the 6 position (Fig. 2.6a). This
compound was then further reacted with Novozyme 435 and an excess of vinyl
acrylate to give the monoacryloyl derivative at 6. The same non-native macrolactone
was observed by Nuhez, although in that case both the 6 acetylated and non-acetylated
versions were formed (Nuhez et al., 2003). These by-products were isolated from a
reaction where the 6, 6 diacetyl sophorolipid methyl ester was reacted with galactose
diacetonide using Novozyme 435. The intended product was the transesterified
galactose ester, linking the sophorolipid carboxylate to the primary 6-hydroxyl group
of galactose, which was obtained, although deacetylation at the sophorolipid 6 and
6 groups had occurred to varying extents. This one-pot reaction with immobilized
lipase therefore produced sophorolipid that was modified at both the carbohydrate
head and the carboxylic tail. A similar macrolactonization, again using Novozyme
435, was performed on a similar sophorolipid, namely one with a C22 lipid tail
(NuAa et al., 2004). These results were significant since the C22 sophorolipid is not
originally isolated in macrolactone form, but only as the acyclic free acid. It could not
be determined from mass spectral data whether lactonization had occurred at the 6 or
6 site, although the latter was proposed by analogy with results in Bisht et al. (1999)
and Nuhez et al. (2003).
The next advance came in determining that monoacylation could be performed
regioselectively (Singh et al., 2003). Novozyme 435, which, as described above,
gave diacetylated products, could be used to prepare 6 monoacetyl derivatives of
sophorolipid ester when shorter reaction times and less vinyl acetate were used.
Alternatively, Lipase PS (isolated from Burkholderia cepacia; Amano Enzyme USA,
Elgin, IL) on a ceramic support gave completely opposite regioselectivity, producing
the 6 monoacetylated products. Amides of sophorolipid were also suitable substrates
for these reactions. The amides themselves (five analogs from para-substituted
phenethylamines) were prepared by Novozyme 435-catalyzed amidation of the
desacetyl sophorolipid ethyl ester. A different set of substrates was used by Carr and
Bisht (2003) who employed per-acylated sophorolipid esters with four different lipases.
Here, the reaction was entirely at the carboxyl group to give either the n-butyl or isobutyl
ester; no de-acylation was observed. It was proposed that only when the macrolactone
was present could binding in the enzyme active site occur so that deacylation could
occur. With acyclic substrates, the carboxyl terminus was preferentially attacked. A

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks 0 43

possible item for hture study would be to use the hexa-acetylated macrolactone.
The deacylation reaction was put to use for in situ polymerization by Hu and Ju
(2003). Any of four lipases could be used to first generate the 6-hydroxy--acetyl
sophorolipid macrolactone, which then underwent self-oligomerization by enzymatic
ring-opening, forming esters between the 6 hydroxyl group and the C18 carboxyl
group. Unfortunately, the amount of oligomer extractable into solvents such as ethyl
acetate or tetrahydrofuran was small, reaction times were lengthy (10 or 20 days),
and the insoluble material could only tentatively be identified as polymer (a peak
corresponding to the nonamer was found by mass spectrometry). Finally, a more
extensive modification to the sophorolipid structure was effected by hesperidinase
(EC 3.2.1.40) or several other glycosidases (Rau et al., 1999, 2001). These enzymes
operated on the desacetyl acyclic free acid sophorolipid to remove the outer glucose,
leaving a monosaccharide glycolipid in 80% yield.
To the best of our knowledge, the only acyl groups that have been appended
enzymatically to the sophorolipid carbohydrates are acetyl, acryloyl, methacryloyl,
and succinoyl. It would be interesting to test how structurally variable is the set of
acyl donors tolerated by the lipases in conjunction with the sophorose headgroup
(more general studies on lipase utility have of course been performed). Branched
and aromatic acyl groups could afford novel and usehl sophorolipids. While the
methacryloyl group is a-branched, its vinyl unit is small, so it may be an exception.
Some of the earliest work on chemical modification to the sophorolipid skeleton
was performed in the course of initially determining the structure of the macrolactone
by Tulloch et al. (1968b).The techniques involved are essentially classical degradations
used in carbohydrate chemistry, such as peracetylation with HBr/acetic acid. Although
these methods apparently were applied afier these pioneering studies, they may bear
hrther investigation, since some interesting structural changes can be accomplished.
One intriguing derivative, for example, consists of a ring-opened form where the
C18 fatty acid remains esterified at the 4 hydroxy site, but has been cleaved at the
glycosidic 1 site.
The more common method of ring opening, which has been employed in a
number of studies, is alkaline hydrolysis or alcoholysis. Sodium alkoxide solutions
remove the acetyl groups and concurrently form the acyclic alkyl ester of the fatty
acid chain. Methyl through hexyl (although not, apparently, pentyl) esters have been
prepared in this way (Bisht et al., 1999; Nuhez et al., 2003; Singh et al., 2003; Carr
& Bisht, 2003; Zhang et al., 2004). Sodium or potassium hydroxide generates the
sophorolipid free acid, after acidification. A related method uses primary amines to
form the sophorolipid amides (Rau et al., 2001), although ammonium chloride was
apparently necessary as a catalyst, since other workers reported that refluxing the
sophorolipid lactone with amines, presumably without ammonium chloride, did not
work (Singh et al., 2003). More hindered amines, such as diethylamine, required
prior deprotonation with n-butyl lithium (Rau et al., 2001). A milder method of
converting the carboxylic acid tail to an amide employed dicyclohexylcarbodiimideas

44 0 R.D. Ashby et al.

a coupling reagent with ethyl or benzyl esters ofa-amino acids (Azim et al., 2006). The
amino acids used included nonpolar (leucine and phenylalanine) and polar (aspartic
and glutamic acids) types. The sophorolipid variants were assayed for antimicrobial
activity against seven species, with mixed results: minimal inhibitory concentrations
ranged from 5 mglmL to 10-4 mglmL. Anti-HIV and antispermicidal effects were
also investigated.
Chemical methods have also been used to add groups to the carbohydrate units.
Peracetylation is readily accomplishedwith acyl anhydrides anddimethylaminopyridine
catalysis (Carr & Bisht, 2003). Selective acylation through non-enzymatic means has
not yet been demonstrated. Our own work used carbodiimide-mediated esterification
to affix amino acids to the carbohydrate hydroxyl groups of sophorolipid, but
a mixture of two major and one minor isomer was obtained (Zerkowski et al.,
2006). We hypothesized that the two major isomers are the 6 and 6 derivatives.
An aminobenzoate linker unit was used (Fig. 2.6b) at the point of attachment to
the hydroxyl groups, since we found that aliphatic acyl groups (e.g., N-protected
glycine) were readily cleaved in the presence of weak bases and protic solvents, while
benzoates resisted such decomposition. The amino acids were primarily polar ones
such as hydroxyproline, glutamic acid, and lysine, which were chosen to increase
water solubility of the sophorolipid. We also used carbodiimides to prepare esters of
the fatty acid chain from the sophorolipid free acid and an alcohol, particularly benzyl
alcohol. The sophorolipid skeleton was not decomposed by treatment with 100%
trifluoroacetic acid or by catalytic hydrogenation over palladium on carbon (Azim et
al., 2006, Zerkowski et al., 2006).
Polymers of sophorolipid have been prepared through chemical means by two
routes. First was free-radical polymerization of the 6 monoacryloyl 6 macrolactone
(Bisht et al., 2000). Co-polymers with acrylic acid and acrylamide were also prepared.
Molecular weights were on the order of lo4 glmol, and the material was soluble in
polar aprotic solvents dimethylformamide and dimethyl sulfoxide but not in water.

Fig. 2.6. The non-natural macrolactone linking the carboxyl group to the 6 hydroxy group that is
produced enzymatically (Bisht et al., 1999) when a substoichiometric amount of external acylating
reagent is used (A). An example of a head-group-modified sophorolipid where a polar unit (here,
hydroxyproline) is attached via a linker unit (para-aminobenzoic acid) to one of the sophorose hydroxy
groups (6).

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks

45

The second route to sophorolipid polymers involved ring-opening metathesis

polymerization via ruthenium catalysts (Gao et al., 2007). This method made use
of the native olefin unit of the C18 chain-in
other words, there was no need
for derivatization of the sophorolipid as obtained from fermentation; the diacetyl
macrolactone was used after purification on silica gel. Molecular weights of up to
105 g/mol were attained. The authors proposed that robust enthalpies of interchain
aggregation of the linear polymer may be the driving force for opening the unstrained
macrolactone ring. The polymers were soluble in tetrahydohran, in contrast to those
reported in (Hu & Ju, 2003), and reactions were complete in an hour.

Concl usions
Many of the glycolipid biosurfactants, including sophorolipids, are making inroads
for potential industrial application simply because they can be synthesized biologically
in relatively large concentrations and they also exhibit properties that lend themselves
favorably as substitutes for or additives to currently used petroleum-based surfactants.
Reports suggest that sophorolipid concentrations on the order of hundreds of grams
per liter can be attained under suitable fermentation conditions and that some of
those sophorolipids (depending on their predominant fatty acid content and structural
form) possess surface active properties that approximate those shown by commonly
used surfactants. As further work continues to elucidate the biochemical pathways
involved in sophorolipid synthesis and to synthesize new structural variants through
post-synthetic chemical/enzymatic modifications, new products with improved
properties will be produced with diminished costs. This will make these molecules
even more favorable for industrial application and may also help stimulate the search
for new methods to utilize abundant, low value co-product materials such as crude
glycerine and soy molasses.
Mention of trade names or commercial products in this artick is soklyfor the purpose of
providing specifc information and does not imply recommendation or endorsement by the
US.Department OfAgricultUre.

References
Abe, Y.; S. Inoue; A. Ishida. Cosmetic Composition for Skin and Hair Treatment. US.Patent
M 2 9 Z 3 4 0 . 1 9 8 1.
Allingham, R. Sophoroside Esters in Prepared Food Products. US.Patent #3,622,344. 1971.
Ashby, R.; A. Nufiez; D. Solaiman;T. Foglia. Sophorolipid Biosynthesis from a Biodiesel CoProduct Stream.J Am. Oil Cbem. Soc., 2005, 82,625-630.
Ashby, R.; D. Solaiman;T. Foglia. Property Control of Sophorolipids: Influence of Fatty Acid
Substrate and Blending. Biotecbnol. Lett., 2008,30, 1093-1 100.
Asmer, H-J.; S. Lang; F, Wagner; V. Wray. Microbial Production, Structure Elucidation and Bioconversion of Sophorose Lipids.J. Am. Oil Cbem. Soc., 1988, 65, 1460-1466.
h i m , A.; V. Shah; G. Doncel; N.Peterson; W. Gao; R. Gross. Amino Acid Conjugated Sophoro-

46 0 R.D. Ashby et al.

lipids: A New Family of Biologically Active Functionalized Glycolipids. Bioconj. Chem., 2006,
17, 1523-1529.
Banat, I.; R Makkar; S. Cameotra. Potential Commercial Applications of Microbial Surfactants.
Appl. Microbiol. Biotechnof., 2002, 53, 495-508.
Bednarski, W.; M. Adamczak; J. Tomasik; M. Plaszczyk. Application of Oil Refinery Waste in the
Biosynthesis of Glycolipids by Yeast. Biores. Technof.,2004,95, 15-18.
Bisht, K.; R. Gross; D. Kaplan. Enzyme-Mediated Regioselective Acylations of Sophorolipids.
J. Org. Chem., 1999, G4, 780-789.
Bisht, K.; W. Gao; R. Gross. Glycolipids from Candid? bombicola: Polymerization of a 6-0-Acryloyl Sophorolipid Derivative. Macromolecules, 2000,33,6208-6210.
Borzeix, C. Use of Sophorolipids Comprising Diacetyl Lactone as Agent for Stimulating Skin
Fibroblast Metabolism. US.Patent #6,596,21;5. 2003.
Brakemeier, A.; D. Wullbrandt; S. Lang. Candida bombicola: Production of Novel AIkyl Glucosides Based on Glucosel2-Dodecanol. Appf. Microbiol. Biotechnol., 1998,50, 161-166.
Breithaupt, T.; R. Light. Atfinity Chromatography and Further Characterization of the Glucosyltransferases Involved in Hydroxydocosanoic Acid Sophoroside Production in Candida bogoriensis. J. Biol. Chem., 1982,257, 9622-9628.
Bucholtz, M.; R. Light. Acetylation of 13 Sophorosyloxydocosanoic Acid by an Acetyltransferase
Purified from Candid? bogoriensis. J. Biol. Chem., 1976a, 251,424-430.
Bucholtz, M.; R. Light. Hydrolysis of 13 Sophorosyloxydocosanoic Acid Esters by Acetyl and
Carboxyesterases Isolated from Candida bogoriensis. J. Biol. Chem., 1976b, 251,431-437.
Carr, J.; K. Bisht. Enzyme Catalyzed Regioselective Transesterification of Peracetylated Sophorolipids. Tetrahedron, 2003,59, 7713-7724.
Cooper, D.; D. Paddock. Torubpsis petrophilum and Surface Activity. Appl. Environ. Microbiol.,
1983,4G, 1426-1429.
Cutler, A.; R. Light. Regulation of Hydroxydocosanoic Acid Sophoroside Production in Candiuk
bogoriensis by the Levels of Glucose and Yeast Extract in the Growth Medium. J. Biol. Chem.,
1979,254, 1944-1950.
Cutler, A.; R. Light. Effect of Glucose Concentration in the Growth Medium on the Synthesis of
Fatty Acids by the Yeast Candida bogoriensis. Can. J. Microbiol., 1982, 28, 223-230.
Daniel, H-J.; M. Reuss; C. Syldatk. Production of Sophorolipids in High Concentration from
Deproteinized Whey and Rapeseed Oil in a Two Stage Fed Batch Process Using Candid?
bombicoh ATCC 222 14 and Ctyptococcus curvatus ATCC 20509. Biotecbnol. Lett., 1998,20,
1153-1 156.
Davila, A-M.; R. Marchal; J-I? Vandercasteele. Sophorose Lipid Production from Lipidic Precursors: Predictive Evaluation of Industrial Substrates. J. Ind. Mirrobiol., 1994, 13, 249-257.
de Koster, C.; W. Heerma; H. Pepermans; A. Groenewegen; H. Peters; J. Haverkamp. Tandem
Mass Spectrometry and Nuclear Magnetic Resonance Spectroscopy Studies of Candida bombicola Sophorolipids and Product Formed on Hydrolysis by Cutinase. Anal. Biochem., 1995,230,
135-148.
Deshpande, M.; L. Daniels. Evaluation of Sophorolipid Biosurfactant Production by Candida

Productionand Modification of Sophorolipids from Agricultural Feedstocks 0 47

bombicola Using Animal Fat. Biores. Technol., 1995,54, 143-1 50.

Esders, T.; R. Light. Characterization and in vivo Production of Three Glycolipids from Cundida
bogoriensis: 13 GlucopyranosylglucopyranosyloxydocosanoicAcid and Its Mono and Diacetylated Derivatives.]. Lip. Res., 1972a, 13,663-671.
Esders, T.; R Light. Glucosyl- and AcetyltransferasesInvolved in the Biosynthesis of Glycolipids
from Candida bogoriensis.]. Biol. Chem., 1972b, 247, 1375-1386.
Esders, T.; R. Light. Occurrence of a Uridine Diphosphate G1ucose:Sterol Glucosyltransferase in
Cundida bogoriensis. ]. Biol. Chem., 1972c, 247,7494-7497.
Fleurackers, S . On the Use of Waste Frying Oil in the Synthesis of Sophorolipids. Eur. ]. Lipid Sci.
Technol., 2006, 108, 5-12.
Gao, W.; R. Hagver; V. Shah; W. Xie; R. Gross; M. Ilker; C. Bell; K. Burke; E. Coughlin. Glycolipid Polymer Synthesized from Natural Lactonic Sophorolipids by Ring-Opening Metathesis
Polymerization. Macrornolenrles, 2007, 40, 145-147.
Gorin, I?; J. Spencer; A. Tulloch. Hydroxy Fatty Acid and Glycolipids of Sophorose from Torulopsis magnoliae. Can./. Chem., 1961,39,846-895.
Hall, I?; J. Haverkamp; C. van Kralingen; M. Schmidt. Synergistic Dual Surfactant Detergent
Composition Containing Sophoroselipid. US. Patent #5,417,879. 1995.
Heinz, E.; A. Tulloch; J. Spencer. StereospecificHydroxylation of Long Chain Compounds by a
Species of Torulopsis.]. Biol. Chem., 1969,244, 882-888.
Heinz, E.; A. Tulloch; J. Spencer. Hydroxylation of Oleic Acid by Cell-Free Extracts of a Species of
Torulopsis. Biochim. Biophys. Acta., 1970,202,49-55.
Hommel, R.; K. Huse. Regulation of Sophorose Lipid Production by Cundida apicokz. Biotechnol.
Lett., 1993,33, 853-858.
Hommel, R.; S . Stegner; K. Huse; H-I? Kleber. Cytochrome P-450 in the Sophorose-Lipid-Producing Yeast Candida (Torulopsis) apicola. Appl. Microbiol. Biotechnol., 1994, 40, 724-728.
Hu, Y.;L-K. Ju. Sophorolipid Production from Different Lipid Precursors Observed with LC-MS.
Enz Microb. Technol., 2001,29, 593-601.
Hu, Y.; L-K Ju. Lipase-Mediated Deacetylation and Oligomerization of Lactonic Sophorolipids.
Biotechnol. hog., 2003, 19,303-3 1 1 .
Inoue, S.; Y. Kimura; M. Kinta. Process for Producing a Glycolipid Ester. US. Patent #4,215,213.
1980.
Isoda, H.; D. Kitamoto; H. Shinmoto; M. Matsumura; T. Nakahara. Microbial Extracellular Glycolipid Induction of Differentiation and Inhibition of the Protein Kinase C Activity of Human
Promyelocytic Leukemia Cell Line HL6O. Biosci. Biotech. Biochem., 1997, GI, 609-614.
Ito, S.; S . Inoue. Sophorolipids from Torulopsis bombicola: Possible Relation to Alkane Uptake.
Appl. Environ. Microbiol., 1982, 43, 1278-1283.
Kappeli, 0. Cytochromes P-450 of Yeasts. Microbiol. Rev., 1986, 50, 244-258.
Kastelic-Suhadolc,T. Cholesteryl Glucoside in Candida bogoriensis. Biochim. Biophys. Actu., 1980,
620,322-325.
Kotrba, R. The Glycerine Spread. BiodieselMag., September 2007, pp. 50-55.
Lottermoser, K.; W Schunck; 0.Asperger. Cytochromes P450 of the Sophorose Lipid-Producing

48 0 R.D.Ashby et al.
Yeast Candida apicokz: Heterogeneity and Polymerase Chain Reaction-Mediated Cloning of Two
Genes. Yeart, 1996, 12, 565-575.
Mager, H.; R. Rothlisberger; E Wagner. Cosmetic Composition Containing One Sophorose
Lipid-Lactone and Its Use. European Patent EP0207783. 1987.
Maingault, M. Pharmaceutical and Cosmetic Compositions Containing Sophorolipids. Canadian
Patent cAN126242874.1997.
Maingault, M. Utilization of Sophorolipids as TherapeuticallyActive Substances or Cosmetic
Products, in Particular for the Treatment of the Skin. US.Patent #5,981,477,1999.
Makkar, R.; R. Cameotra. An Update on the Use of Unconventional Substrates for Biosurfactant
Production and Their Applications. Appl. Microbiol. Biotecbnol., 2002, 58, 428-434.
Mulligan, C.; R. Yong; B. Gibbs. Heavy Metal Removal from Sediments by Biosurfactants.
J Hamrd. Mater., 2001, 85, 1 1 1-125.
Muthusamy, K.; S . Gopalakrishnan;T. Ravi; I? Sivachidambaram. Biosurfactants: Properties,
Commercial Production and Application. Cum Sci., 2008,74,736-747.
National Biodiesel Board. Ektimated US. Biodiesel Production by Fiscal Year, http://www.biodiesel.org/pdf_files/fuelfactsheets/production-graph-slide.pdf, , 2007.
Nufiez, A,; R. Ashby; T. Foglia; D. Solaiman. Analysis and Characterization of Sophorolipids by
Liquid Chromatography with Atmospheric Pressure Chemical Ionization. Chromatographia,
2001,53,673-677.
Nufiez, A,; T. Foglia; R. Ashby. Enzymatic Synthesis of a GalactopyranoseSophorolipid Fatty
Acid-Ester. Biotechnol. Lett., 2003, 25, 1291-1297.
Nufiez, A,; R Ashby; T. Foglia. LC/MS Analysis and Lipase Modification of the Sophorolipids
Produced by Rbodotorukz bogoriensis. Biotechnol. Lett., 2004,26, 1087-1093.
Otto, R.; H. Daniel; G. Pekin; K. Miiller-Decker; G. Fiirstenberger; M. Reuss; C. Syldatk.
Production of Sophorolipids from Whey. 11. Product Composition, Surface Active Properties,
Cytotoxicity and Stability Against Hydrolases by Enzymatic Treatment. Appl. Microbiol. Biotecbnol., 1999,52495-501.
Pekin, G.; F. Vardar-Sukan; N. Kosaric. Production of Sophorolipids from Candida bombicokz
ATCC 22214 Using Turkish Corn Oil and Honey. En5 Life Sci., 20055,357-362.
Rau, U.; C. Manzke; F. Wagner. Influence of Substrate Supply on the Production of Sophorose
Lipids by Candida bombicokz ATCC 22214. Biotecbnol. Lett., 1996, 18, 149-154.
Rau, U.; R. Heckmann; V. Wray; S. Lang. Enzymatic Conversion of a Sophorolipid into a Glucose Lipid. Biotecbnol. Lett., 1999,21,973-977.
Rau, U.; S. Hammen; R. Heckmann; V. Wray; S. Lang. Sophorolipids:A Source for Novel Compounds. Ind. Crop Prod., 2001, 13, 85-92.
Ruinen, J. The Phyllosphere 11. Yeasts from the Phyllosphere of Tropical Foliage. Antonie van
Leeuwenhoek, 1963,27,425-438.
Ruinen, J.; M. Deinema. Composition and Properties of the Extracellular Lipids of Yeast Species
from the Phyllosphere. Antonie van Leeuwenhoek, 1964,30, 377-384.
Scholz, C.; S. Mehta; K. Bisht; V. Guilmanov; D. Kaplan; R. Nicolosi; R. Gross. Bioacrivity of
Extracellular Glycolipids-Investigation of Potential Anti-Cancer Activity of Sophorolipids and

Production and Modification of Sophorolipidsfrom Agricultural Feedstocks 0 49

Sophorolipid-Derivatives. Polym. Prep., 1998,39, 168-169.

Singh, S.; A. Felse; A. NuAa; T. Foglia; R. Gross. Regioselective Enzyme-Catalyzed Synthesis
of Sophorolipid Esters, h i d e s , and Multifunctional Monomers. 1.Or5 Chem., 2003, 68,
5466-5477.
Solaiman, D.; R. Ashby; T. Foglia; A. NuAa; W. Marrner. Sophorolipids-Emerging Microbial
Biosurfactants. Inform, 2004a, 15,270-272.
Solaiman, D.; R. Ashby; A. NuAa; T. Foglia. Production of Sophorolipids by Candida bombicokz
Grown on Soy Molasses as Substrate. Biotechnol. Lett., 2004b, 26, 1241-1245.

Solaiman, D.; R. Ashby; J. Zerkowski; T. Foglia. Simplified Soy Molasses-Based Medium for
Reduced-Cost Production of Sophorolipids by Candida bombicokz. Biotechnol. Lett., 2007,29,
1341-1347.
Spencer, J.; A. Ragout de Spencer; C. Lduce. Non-Conventional Yeasts. Appl. Mirrobiol. Biotecbno[., 2002,58, 147-156.
Sternberg, D.; G. Mandels. Regulation of the Cellulolytic System in Trichoderma reesei by
Sophorose: Induction of Cellulose and Repression of Beta-Glucosidase.J. Barteriol., 1980, 144,
1197-1199.
Stuwer, 0.;R. Hommel; D. Haferburg; H. Kleber. Production of Crystalline Surface-Active Glycolipids by a Strain of Torulopsis apicokz.J. Biotecbnol., 1987, 6, 259-269.
TFC Commodity Charts, Soybean Oil (BO, CBOT) Monthly Price Chart, http://futures.tradingcharts.com/chart/SO/M, 2008.
Tsutsumi, H.; J. Kawano; S. Inoue. Cosmetic Composition. US.Patent #4,305,961. 1981.
Tulloch, A.; J. Spencer; M. Deinema. A New Hydroxyl Fatty Acid Sophoroside from Candida
bogoriensis. Can./. Cbem., 1968a, 46, 345-348.
Tulloch, A.; A. Hill; J. Spencer. Structure and Reactions of Lactonic and Acidic Sophorosidesof
17-HydroxyoctadecanoicAcid. Can. J. Chem., 1968b, 46, 3337-335 1.
Van den Brink, H.; R. van Gorcum; C. van den Hondel; I? Punt. Cytochrome P450 Enzyme
Systems in Fungi. Fungal Gen. Biol., 1998,23, 1-17.
Zerkowski, J.; D. Solaiman; R. Ashby; T Foglia. Head Group-Modified Sophorolipids: Synthesis
of New Cationic, Zwitterionic, and Anionic Surfactants.J. Su$ Deterg., 2006, 9, 57-62.
Zerkowski, J.; D. Solaiman. Synthesis of Polyfunctional Fatty Amines from Sophorolipid-Derived
17-Hydroxy Oleic Acid. J. Am. Oil Chem. Soc., 2006,83, 621-628.
Zerkowski, J.; D. Solaiman. Polyhydroxy Fatty Acids Derived from Sophorolipids.j . Am. Oil
Chem. Soc., 2007,84, 463-471.
Zerkowski, J.; A. Nufia; D. Solaiman. Structured Estolides: Control of Length and Sequence.J
Am. Oil Chem. Soc., 2008,85, 277-284.
Zhang, L.; l? Somasundaran; S. Singh; A. Felse; R Gross. Synthesis and Interfacial Properties of
Sophorolipid Derivatives. Cod Su$ A., 2004,240,75-82.
Zhou, Q.; V. Kleckner; N. Kosaric. Production of Sophorose Lipids by Torulopsis bombicokz from
Safflower Oil and G1ucose.j. Am. Oil Chem. SOL, 1992, 69, 89-91.
Zhou, Q.; N. Kosaric. Utilization of Canola Oil and Lactose to Produce Biosurfactant with Candida bom6icokz.j. Am. Oil Chem. Sot., 1995, 72,67-71.

Mannosylerythritol Lipids:
Production and Downstream
Processing
Udo Rau' and Dai Kitamoto2
'TechnicalUniversityBraunschweig, lnstitute of Biochemistry and Biotechnology,Spielmannstr. 7,D-38 106
Braunschweig, Germany;'Researchlnstitute far lnnovation in SustainableChemistry,Nationallnstitute of
Advanced Industrial Science and Technology,AiST Tsukuba Central 5-2, 1 -1- 1, Higashi, Tsukuba, lbaraki
305-8565,Japan

Introduction
Biosurfactants are synthesized by several microorganisms and/or by enzymatic
treatment (Banat et al., 2000) and are assigned to three natural roles in the microbial
environment: (a) increasing the surface area of hydrophobic water-insoluble growth
substrates; (b) increasing the bioavailability of hydrophobic substrates by enhancing
their apparent solubility or desorbing them from surfaces; (c) regulating the attachment
and detachment of microorganisms to and from surfaces.
Microbial surfactants are used in a broad range of industrial applications, including
enhanced oil recovery, crude oil drilling, lubrication, surfactant-aided bioremediation,
health care, cosmetics, and food (Fiechter, 1992; Lin, 1996; Dossat et al., 2002;
Eliora & Rosenberg, 2002; Kitamoto et al., 2002; Lang & Trowitzsch-Kienast,
2002; Sekhon, 2006). The high-molecular-mass bioemulsifiers are amphipathic
polysaccharides, proteins, lipopolysaccharides, lipoproteins, or complex mixtures of
these biopolymers, which are effective at stabilizing oil-in-water emulsions. The lowmolecular-mass microbial surfactants are generally lipopeptides, such as surfactin,
gramicidin S, and polymyxin or glycolipids, such as mannosylerythritol lipids,
sophorolipids, and rhamnolipids, which primarily reduce surface and interfacial
tensions (Rosenberg & Ron, 1999). Biosurfactants possess important advantages
compared to chemical surfactants. These include mild production conditions, lower
toxicity, higher biodegradability, and environmental compatibility. Recently, common
reviews about biosurfactants were published by Van Hamme et al. (2006), Lu et al.
(2007) and Singh et al. (2007).
Among these amphiphilic molecules, the glycolipids represent a potent group
of biosurfactants. They are low-molecular-weight products ( ~ 1 , 5 0 0g mol-I). The
constituent mono-, di-, or oligosaccharides include glucose, mannose, galactose,
-51

52 0 U. Rau and D. Kitamoto

glucuronic acid, rhamnose, and galactose sulphate. The lipophilic moieties contain
saturated, unsaturated, and/or hydroxylated fatty acids or fatty alcohols (Karanth et
al., 1999; Lang, 2002). Mannosylerythritol lipid (MEL) is one of the most promising
glycolipids for widespread applications in the future. Beside mannose and fatty acids/
alcohols of different chain lengths, it also contains the sugar alcohol erythritol.
One of the reasons for increase interest in MEL is due to their pharmaceutical
applications (Shibahara et al., 2000; Zhao et al., 2000). For example, Aventis
Pharma Deutschland GmbH published a patent about a MEL, named Ustilipid,
that is produced by Ustilago maydis DSM 11494 and can be used in the treatment
of schizophrenia or diseases caused by dopamine metabolic dysfunction (Vertesy et
al., 2002). However, other applications, such as chemical tools for purification of
proteins (Im et al., 2003; Kitamoto et al., 2000) or anti-agglomeration agents of iceslurry (Kitamoto et al., 2OO1b) are also known. Kitamoto et al. (2002), reviewed the
functions and potential applications of MEL.
Despite the feasible use of MEL in such manifold areas, a real breakthrough with
this component has not yet occurred. One reason could be that the development and
scale-up ofa biotechnical production plant including thecost-intensive downstreamwill
require more money and time effort compared to a chemical surfactant manufacturing
process if there is no outstanding characteristic which justifies the higher investment.
The different applications for MEL are covered in another chapter of this book and
should stimulate commercial interests when combined with this chapter concerning
recent findings of bioreactor production and downstream processing of MEL.

Structure and Microbial Origin

Mannosylerythritol lipid, 2,3-di-O-alka(e)noyl-l3-D-mannopyranosyl-(
1+4)-0meso-erythritol, partially acetylated at C-4 and/or C-6, contains mannose and the
sugar alcohol erythritol as hydrophilic moiety and acetyl groups as well as fatty acids
as the hydrophobic moiety (Fig. 3.1).
The natural emerging MEL can be further differentiated into MEL A-D with
respect to the degree of acetylation at C-4 and C-6 (Kitamoto et al., 1990a). MEL-A
represents the diacetylated compound whereas MEL-B and MEL-C are monacetylated
at C-6 and C-4, respectively. The completely deacetylated structure is attributed to
MEL-D. A mixture of MEL A-D is always present whereas most microorganisms
secrete the hydrophobic MEL-A in highest concentration >70% (w/w) compared
to the other homologs. The mannose is further acylated with C4-Cl8 saturated
and unsaturated fatty acids depending on the microorganism and substrate used
(Table 3.1). The higher the grade of acylation, the lower is the hydrophilicity of the
component. Recently, the configuration of the sugar alcohol was confirmed by total
synthesis of the single diastereomer D-erythritol (Crich et al., 2002).
The MEL are secreted by several microorganisms and were first noted as oily
compounds in the culture suspension of the smut fungus Ustilago maydis, and were
termed Ustilipids (Haskins et al., 1955). Spoeckner et al. (1999) used Ustilago maydis

Mannosylerythritol Lipids: Production and Downstream Processing 0 53

Fig. 3.1. Molecular structure of mannosylerythritol lipid. The length and saturation of the fatty acid
residues (R2,R3)depend on the substrate and microorganismused. R4,R6 = acetyl or H. Ball-stick-model:
Diacetylated MEL-A with two C100 fatty acid esters at position C-2and C-3:

DSM 4500 in a shake flask with sunflower oil fatty acids and glucose as substrates.
However, the production of an additional glycolipid (cellobiose lipid or ustilagic acid)
could only be lowered by using a suitable ratio of these substrates but not hlly avoided.
Geotrichum candidum ST 0025 15 also secretes a mixture of both cellobiose lipid and
MEL (Kurz et al., 2003). Deml et al. (1980) used Schizonella melanogramma for the
production of MEL that was termed Schizonellin. Candida sp. B-7 (Kawashima et
al., 1983) and Kurtzmanomycessp. 1-1 1 (Kakugawa et al., 2002) are also known to be
producers of MEL (Table 3.1).
Pseudozyma (formerly Candidz) is the most potent genus for the secretion of
MEL. Primarily in the 1990s, Kitamoto and co-workers thoroughly investigated
the production characteristics and substrate influence on MEL structure by
using Pseudozyma (Candidz) antarctica T-34 (Table 3.2). Pseudozyma strains are
usitilaginomycetous anamorphic yeasts, which are classified into Ustilagininales and
show a close relationship (97% identity) from the sequence of ITS1, 5,8S rRNA
gene, and ITS2 (Morita et al., 2006a). Based on this sequence homology, analysis of
ribosomal DNA on yeast strains of the genus Pseudozyma was undertaken and yielded
I! ruplosa NBRC 10877 as a novel MEL producer (Morita et al., 2006b).
The results of Tables 3.1 and 3.2 obviously show that a prediction of the fatty
acids introduced into the MEL is not possible when employing the same substrate
(soybean oil) since it depends on the strain used. Related to the microorganisms listed
in Table 3.1, a broader spectrum of fatty acids was detected compared to the genus
Pseudozyma. However, all substrates were degraded by the release of one C2-unit at
minimum. In the past, this partial &oxidation was only known in mammalian cells
and was first described for the yeast I! antarctica by Kitamoto et al. (1998) as a chainshortening pathway.

Table 3.1.Variation of the Fatty Acid Moiety in MEL Depending on Microorganism and Substrate

c)

rn
3

c)

2
0

m
=I

cr:

5
0
*
In

Ustilago maydis
Candidasp. 6-7
Candida sp. SY 16 DSM 4500
Microorganism Candida sp. 6-7
Substrate
Mixture of n-alkanes Soybean oil
Soybean oil
Sunflower fatty acidsa
Reference
Kawashima et al. 1983 Kawashima et al. 1983 Kim et al. 1999 Spoeckner et al. 1999
MELb
A
Fatty acid
(%, WIW)
c40
9.1
c5:o
1.5
C69
49
19.6
c79
4.2
C80
18
19.2
0.7
C81
c99
21.6
c10 0
1.9
9.7
0.7
C10l
c102
c1 l:o
13.7
c120
24
25
23
3.0
c130
142
c140
2.4
45.7
19
3.8
c141
9
42.9
C160
1.9
C161
12.6
C189
C181
1.7
"Contained50% oleic acid
bA mixture of MEL was analysed if not indicated otherwise

Schizonella
melanogramma
GD 325
Glucose
Deml et al. 1980
SchizonellinA/B

Kurtzmanomyces
sp.1-1 1
Soybean oil
Kakuqawa et al. 2002

1.4
36.4
1.1
1.2
0.5
2.0
~

6.2
44.1
19
7.6
21

Table 3.2. Influence of Different Substrates on Variation of Fatty Acid Moiety in MEL Using Pseudozyma (Candido)antarcticaT-34
Substrate

Methyl ester of specified fatty acid

n-alkanes
Alkanols and Alkanones
Soybeanoil
C12 C13 C15 C16 C181 C182 C12 C13 C14 C15 C16 C17 C18 CIZb C12' C13d C13' C14'
Kitamotoet at. Kitamoto Kitamoto Kitamoto et
Kitamoto et al. 2001a
Kitamotoet al. 1999
1990a
et al. 1993 et al. 1995 al. 1993
gA
gB
A

Reference
MEL"
Fatty acid (96,w/w)
c7
C8
17.5
C81
c9
c91

26.6

22

1 0 8
2

14

30.7

--

50

45

1
-

67

c10
c101
c102

71.3

58.6

72

c11
C11:l
c12

10.1

13.3

40

40

32.1

4
22

11
4

1.6
-

55.7
6.2

14.8
46.4

1.2

18

1.9

c131
C14
3
0.2
c141
c142
"A mixture of MEL was analysed if not indicatedotherwise
b2-dodecanol, 2-dodecanone yielded only slightly different data
'3-dodecanone
d2-tridecanol,2-tridecanone yielded only slightly different data
'3-tridecanone
'2-tetradecanol, 2-tetradecanone yielded only slightly different data
gfatty acid methylester were analysed by GC after hydrogenation

2
20

5
6

3
23

6
8

3
16

31

33

28

11

59

57

12

0.2

45

<1

42

2.0

15

14

0.8

_
-

53

16 50

40

11

38.5

61.5

_ _*

7.9 5.2
12.4 9.5

2.8
27.3

47.7 37.9

11.7 3.7

32

22.7

51.3 51.9

23.1

4.4

0.7

1.6

2.5

2.7

1.3
39.7

23
3

w
u_
~

3
1.
o_

r:
~

z0

2
1
3
-u

a5.
3

0
Ln

Ln

56 0 U. Rau and D. Kitamoto

The fatty acid distribution in MEL derived from the same genus but a different
strain Pseudozyma aphidis DSM 70725 is shown in Table 3.3. Odd fatty acid esters
were primarily shortened to C7, C9, and C11 acids. The amount of the C7 acid
increased with the chain length of the substrate, but generally only small quantities
of the C7 acid were present. Even fatty acid esters were degraded to C8, C10, and
C12 with a preference for CIO, particularly for the unsaturated ones. The double
bond in the C18:l ester completely disappeared. The C18:2 ester was transformed
predominately to C1O:l acid. Soybean oil contains about 50% (w/w) C18:2 and
25% (wlw) C18:l fatty acids. The degradation pattern of the soybean oil fatty acids
was similar to that of the even fatty acid methyl esters, although the quantities were
different. The sum of C1O:O and C1O:l made up approximately 70% (w/w) of all
esters with the exception of stearic acid methyl ester as substrate, which yielded lower
amounts. Fatty acids larger than C14 or smaller than C7 were detected only in trace
amounts. Likewise C6 fatty acids were also not detected. These findings are in contrast
to the composition of MEL secreted by Candida sp. B-7 (Kawashima et al., 1983) and
Candida sp. SYl6 (Candida antarctica KCTC 7804) (Kim et al., 1999), respectively.
Both microorganisms were also cultivated using soybean oil as substrate (Table 3.1).
MEL B-7 and MEL SYl6 contained 45.7 and 19.0% (w/w), respectively, C14:O fatty
acids. The last MEL also contained 9.0% (wlw) C14:l and notable amounts of C6:O
(49.0% (w/w)). The MEL of Kurtzmanomyces sp. 1-1 1 (Kakugawa et al., 2002) even
contained 25.9% (w/w) C14:2. However, all these other MEL showed a relative low
content ~ 1 0 %
(w/w) of C10 fatty acids. MEL-T34 (Table 3.2) was similar to the
MEL of I? aphidis but differed in the fatty-acid content by an average of 5 1 0 % (w/w)
depending on the substrate used (Kitamoto et al., 1995,2001a). However, some clear
distinctions exist. For example, when using oleic acid methyl ester as a substrate the
resulting MEL differed by 10-20% (w/w) with respect to the concentrations of C8:O,
CIO:O, and C1O:l fatty acids (Kitamoto et al., 1993). Additionally, when soybean oil
was used Kitamoto et al. (1990a) did not find C10:l fatty acids, but the fatty acids
were subsequently detected as methyl esters by GC after a hydrogenation step.
Among the high-level MEL producers, such as I? antarctica and I? apbidis, MEL-A
is formed in largest amount. In contrast, Pseudozyma tsukubaensisJCM 10324 (Morita
et al., 2007c) and Pseudozyma sp. KM-160 (Konishi et al., 2007) selectively produce
only a single glycolipid product, which corresponds to MEL-B, in high yield (more
than 25 g L from 4% (w/w) of soybean oil). A detailed inspection of the H and
I3C-NMR data derived from the I? tsukubaensis product showed that beside the known
MEL-B one diastereomer having a different chirality is also formed. The diastereomer
was identified as 1-0-(6-0-acetyl-2,3-di-O-alka(e)noyl-i3-D-mannopyranosyl)-Derythritol, indicating that erythritol is bound to mannose in a manner different from
all known MEL (Fukuoka et al., 2008).
Among the other MEL homologs, the more hydrophilic MEL C can be produced
in increased amounts of 65% (w/w) and 85% (w/w) by Pseudozyma hubeiensis KM-59
(Konishi et al., 2008) and Pseudozymagraminicokz CBS 10092 (Morita et al., 2008).

Mannosylerythritol Lipids: Productionand Downstream Processing 0 57

Table 3.3. Influence of the Substrate (Soybean Oil or Fatty Acid Methyl Ester with Different
Chain Length) on the Fatty Acid Compositionof the MEL Secreted by Pseudozymaaphidis
DSM 70725"
Fatty acid content (96,w/w) of MELs and MEL A after cultivation on
Fatty acid methyl ester

Soybean oil
Fatty acid type

C7:O
C80
C81
c90
c9: 1
c100
c101
c102
c1 l:o
C11:l
c12:o
C12:l
C13:O
C13:l
C14:O
c141

O.lZb
18.54b
2.65b
O.lOb

0.04
12.96
2.38
0.20

36.18b
31.46b
0.51b
0.23b

55.83
16.81
0.10
0.39

2.53b 2.78
0.94b 0.92

C15:O

C160

C17:O

C180

C181

2.69
0.20

15.52

6.84
0.27

0.44
15-96

0.04
20.66

C182
0.06
12.65

44.63
0.08
1.18
0.31

0.22

41.18
0.12
1.48

1.68

0.53

57.10
1.86

74.81
0.99

17.49
56.69
0.48
0.19

71.55
0.26

48.50
0.94
0.16

0.52

45.36
0.19
0.15

3.41

0.16

21.76
0.73
0.48

1.74
0.05

1.24
4.67

0.73
0.09

10.42
0.07

0.35

0.05b 0.06

0.02

0.16

0.15

0.07

1.12

0.36

c142
0.25
2.45
'Related to soybean oil, isolated me1 a (left column) was additionally analyzed beside the
mixture of mels.
bpureMEL A
Rau et al., 2005c.

MEL, mono-acylated at R3 (Fig. 3.1) without any acetyl groups at the mannose
moiety, can be produced by using I? antarctica T-34 and JCM 10317 as well as Z?
parantarctica JCM 11752 and glucose as sole substrate. Due to its increased hydrophilic
character and higher water solubility, the CMC (critical rnicelle concentration) value
o f 3 . 6 ~ 1 0M
. ~ is 100-fold higher compared to the MEL hitherto reported and therefore
Edvors the use as oil-in-water type emulsifiers and washing detergents. However, the
yields are low (1.1-1.3 g 1-') and the di-acylated MEL are formed likewise (Fukuoka
et al., 2007b).
Recently, a more hydrophobic tri-acylated MEL was produced from cultures
using I? antarcticaT-34 and I? ruplosa NBRC 10877. Soybean oil served as substrate.
R2 and R3 carry the common medium-chain fatty acids (C8-Cl4). The third, longchain fatty acid (Cl6 and C18) is linked to the primary hydroxyl group of erythritol.
Morita et al. (2007b) assumed that this component could be directly introduced
into the MEL from soybean oil by mediation of extracellular lipase and/or esterase,

58 0 U. Rau and D. Kitamoto

which are present during growth on triglycerides. 'The assumption was confirmed by
successful esterification of purified di-acylated MEL-A with methyl oleate by using
the culture supernatant of I! mgulosa and Novozyme 435 (lipase).
Also a new MEL type, with shorter hydrocarbon chain (C2 or C4) at R2 but again
di-acylated, was found (Fukuoka et al., 2007a) by cultivation of the basidiomycetous
yeast Pseudozyma shanxiensir CBS 10075 (Wang et al., 2006) on soybean oil. Due to
shorter ester, an enhanced hydrophilicity of the molecule is assumed.

A nal yses
In order to quantitatively detect and separate MEL A-D, triglycerides and fatty acids,
a 3 mL culture suspension was acidified with two drops of 5 N HCI to p H 2 and was
subsequently extracted three times with 3 mL methyl tertiary butyl ether (MTBE). The
mixture was vortexed for 1 min and centrifuged for 5 min at 6000 rpm. The organic
phases were combined and analyzed either by TLC or HPLC. The distinct separation
of single spots for qualitative TLC analysis can be carried out by the development of
Silica gel 60 F254 (Merck, Germany) plates with the ternary solvent system CHCI,/
MeOH/H,O (65:152). HPLC is performable on a silica gel column (Nucleosil 100-5;
CS-Chromatographie Service GmbH, Langenvehe, Germany) with an evaporative
light scattering detector (ACS Mass Detector model 750/14; Houston, TX, USA)
using a gradient solvent program consisting of various proportions of CHCl, and
CH,OH (from 99:l to 0:lOO (v/v)) at a flow rate of 1 mL min-' (Fig. 3.2).
A qualitative and rapid method to determine MEL or other biosurfactants formed
throughout a cultivation can be attained by the drop collapse method (Kuiper et al.,
2004; Hewald et al., 2005; Morita et al., 2007b), which estimates the surface tension
of the culture medium (Fig. 3.3).
The size of a constant volume drop which spreads on a surface is a measure for the
content of MEL.

Biosynthesis
The biosynthetic route of MEL formation was thoroughly examined with Pseudozyma
antarctica T-34 (Kitamoto et al., 1993., 1998) and Ustilago maydis (Hewald et al.,
2006) and is depicted in Fig. 3.4.
Both strains are able to utilize triglycerides as soybean oil. Therefore, extracellular
action of a lipase/esterase is necessary for the cleavage into free fatty acid and glycerol.
Lipase activity from l? antarctica was also noted when the strain was grown on glucose,
and the activities reached levels corresponding to approximately 50% of that obtained
with soybean oil (Morita et al., 2007b). After ester cleavage, glycerol is consumed by
the cells for the production of energy and growing but not for MEL formation (Rau
et al., 2005b). Likewise the released fatty acid is transported into the cell and coupled
to CoA. One part of the activated fatty acid must be &oxidized to acetyl-CoA to
supply energy and precursors for anabolic pathways. Subsequent, mannose is de novo
synthesized by gluconeogenesis. Erythritol is formed through the gluconeogenesis and

MannosylerythritolLipids: Production and Downstream Processing 0 59

Fig. 3.2. HPLC and TLC of MEL containing culture suspension extracted with MTBE. FA=Fatty acid; SO=
Soybean oil.
Pseucbzyme antatctica T-34
Cultured medium
Medium

5 .O

+Glucose

9.0

+Glycerol

5.5

Soybean oil

9.5

Diameter (rnrn)

Fig. 3.3. Secretion of MELs reduce the surface tension of culture supernatant (Morita et al. 2007b).
The effects of MELs on the surface tension of the culture medium were visualized by placing 50 pI of
culture supernatant of fseudozyma antarctica T-34 grown with glucose, glycerol, or soybean oil (40 g I.
respectively), as the sole carbon source on the hydrophobic surface of Parafilm-M. The pictures of the
droplets formed were taken from side (upper panel) and top view (lower panel).

phosphogluconate pathways yielding erythrose which can be converted into the sugar
alcohol by action of erythrose reductase which was verified in Candida magnoliae
(Lee et al., 2003). However, this enzyme has not yet been found in I! antarctica or
1/. maydis. Enzymes indicated with asterisks in Fig. 3.4 were verified in 1/. maydis
(Hewald et al., 2005, 2006) and should be also present in I! antarctica. The activated
GDP-mannose is linked to erythritol by a glucosyltransferase. The transcription of

60 0 U. Rau and D. Kitamoto

Soybean oil
Lipase
+H,O
Esterase
Fatty acids, Glycerol
(C16, C18)

Acyl-CoA
Chain shortenin
pathway

&-Oxidation

C8-, CIO-, C12-CoA (P.anfarcfica T-34))

C4-,C6- and C14-,C16-CoA (Umaydis)

Acetyl-CoA
de novo
3ynthesis

Mannose Erythritol
GDP-Mannose

*GIUCQS~Itransferase

GDP

Mannosylerythritol
CoA
'Acyltransferase

C2-,CS-acylated MEL
2 Acetyl-COA
"Acetyltransferase

MEL A
C4-.Cf%acetvlated
*Major facilitator
Fig. 3.4. Presumptive biosynthesis of MEL-A in U. rnuydis and f! unturcticu based on soybean oil as
substrate. Enzymes indicated with an asterisk were verified so far in U. rnuydis only. The scheme was
adopted from Kitamoto et al. 1998 and Hewald et al. 2006.

Mannosylerythritol Lipids: Production and Downstream Processing 0 61

the related gene is strongly induced under conditions of nitrogen starvation. The
intermediate mannosylerythritol has been isolated in significant amounts from MELproducing cells (Boothroyd et al., 1956). The shortening of acyl-CoA was in the past
only found in mammalian peroxisomes until Kitamoto et al. (1998) could verify this
pathway in I? antarctica.
Even fatty acids, as contained in soybean oil, are primary degraded down to
C8-, C10-, and C12-acids. Odd chain substrates yield predominantly C7-, C9-,
and C1 1-acids. The regulation of this partial &oxidation is still unknown. There is
no preference for fatty acids connected at position C-2 or C-3 in I? antarctica. In
contrast, U. maydis links shorter fatty acids (C2-C8) at position C-2 and medium or
longer fatty acids at C-3 (Spoeckner et al., 1999; Kun. et al., 2003). The final step
of biosynthesis of MEL A proceeds with an acetyl-CoA dependent acetyltransferase
which displays relaxed regioselectivity and is able to transfer acetyl groups to both
positions C-4 and C-6 of the mannosyl moiety. Wild-type cells of U. maydis secrete
large amounts of MEL which are acetylated at only one position (MEL-B/ MEL-C).
This could indicate that the second acetylation reaction is slower than the first one
since partially acetylated glycolipids are secreted before the second acetylation occurs.
Secretion of MEL in U. maydis is supposed to be catalyzed by the major facilitator
Mmfl because the suppression mutant for mmfr is unable to produce extracellular
MEL. The exporter Mmfl appears to display only limited specificity for its substrates,
since deacetylated MEL are secreted as efficiently as the acetylated ones. Furthermore,
the observed spectrum of MEL carrying acyl groups of different lengths suggests that
the putative exporter does not distinguish between these derivatives.

Production
Substratesand Shake Flask Cultivation
Depending on the microorganism used (Table 3. l), manifold substrates (Tables 3.2
and 3.3) can be applied for MEL production as pentoses, hexoses, sugar alcohols,
soluble starch, alkanes, alkanols, alkanones, fatty acids, and their methyl or ethyl esters
as well as triglycerides (Kawashima et al., 1983; Kitamoto et al., 2001a; Kim et al.,
2002b; Sugita et al., 2003). In principle, glucose can be used for MEL formation but
always with reduced yields compared to hydrophobic substrates (Kim et al., 2002a;
Morita et al., 2007b). Therefore, vegetable oils (in particular soybean oil), fatty acids,
and their esters are the most suitable carbon sources for MEL production although a
further increase of yield can be attained by addition of glucose as co-substrate (Rau
et al., 2005b; Konishi et al., 2008). A variety of common triglycerides and fatty acids
used for MEL production by I? aphidis shows Fig. 3.5.
Beside soybean oil, oleic acid (containing 70% (vh) pure C18:l fatty acids),
and mixtures of different fatty acids derived from sunflower oil are also suitable
carbon sources. Even biodiesel (rapeseed oil methyl ester) from a petrol station was
convertible. O n an average, 250% (vh) of substrate is converted into the product
except for glycerol, which is not a proper carbon source for MEL formation using

62 0 U. Rau and D. Kitamoto

Soybesn Sunflower Rapeseed Corn oil

oil
oil
oil

Biodiesel Oleic scld Sunflower Glycerol

fatty acids

Fig. 3.5. MEL formation of R aphidis depending on the carbon source used (medium A). The cultivations
were terminated after 10 days (Rau et al. 2005~).Biodiesel=rapeseed oil methyl ester. Medium A (per liter
deionised water): 80 ml carbon source, 2 g NaNO,, 0.2 g KH,PO, 0.2 g MgS0,x7H20, 1 g yeast extract, pH
6.2 not adjusted, 27C.

I? aphidis. However, in a screening of 10 different strains of the genus Pseudozyrna,

only l? antarctica JCM 10317 secreted abundant amounts of MEL with glycerol
as substrate (Morita et al., 2007a). Using fed-batch cultivation in shake flasks and
mannose as co-substrate, MEL yield was increased to 16.3 g L-I. Likewise, C a n d i h
sp. SYlG converted in a two-stage culture glycerol and oleic acid as initial and feeding
carbon source, respectively, into 41 g L- MEL after 8 days (Kim et al., 2002a). These
promising results should favor the bioconversion of the by-product glycerol from
biodiesel production into highly functional bio-based material.
Mannose and erythritol are the invariant structural elements of MEL and have
to be synthesized by the microorganism. If these components are available during
cultivation the yield of MEL increases. Figure 3.6 shows the response of l? aphidis to
the addition of mannose or erythritol at different times of cultivation. It is important
that the second carbon source is added after growth has ceased due to nitrogen
limitation and MEL formation has been started. With this procedure, the yield can
be increased by 10-20% (w/w).
A considerably higher increase to 70-90% (w/w) yield was achieved with l? rugulosa
NBRC 10877 by intermittent feeding of soybean oil and erythritol. Under optimal
conditions in shake culture, maximum yield, productivity, and yield coefficient (on
a weight basis to soybean oil supplied) of 143 g Ll, 5.0 g L- dl, and 0.5 g g- were

MannosylerythritolLipids: Production and Downstream Processing 0 63

M6
Fig. 3.6. Ten days shake flask cultivation (medium see Fig. 3.5) of R aphidis using soybean oil (80 ml I-)
and erythritol (E, 40 g I-) or mannose (M, 40 g k) as second carbon source added at day 0,2,4 or 6 (Rau
et al. 200%). Ref.: 10 days reference cultivation only with soybean oil.

obtained (Morita et al., 2006b). This high concentration of MEL resulted in increased
viscosity of the culture suspension with paste-like characteristics which prevents
hrther shaking and sufficient oxygen supply and, therefore, probably constitutes the
limit of MEL yield in shake flasks. Likewise, 140 g L- MEL were attained using
resting cells of I! unturcticu T34 during a 28 day 30 mL fed-batch shake cultivation
with a total of 156.6 g L n-octadecane as the carbon source. The cultivation started
with 46.2 g L n-octadecane and afier every 7 days the same amount of substrate was
added (Kitamoto et al., 2001a).
As a secondary metabolite, most of the MEL is secreted during the stationary
growth phase under nitrogen-limited conditions. However, the nitrogen source is
essential for prepended growing and subsequent maintenance purposes; it also
influences the MEL formation as well as the grade of acetylation (Fig. 3.7).
Sodium nitrate was the best nitrogen source followed by ammonium nitrate when
pH was not controlled. The drastic decrease to pH 2 was clearly the reason for the
decreased MEL formation observed when ammonium salts were used as a nitrogen
source. The consumption of ammonium led to acidification of the medium. When
only ammonium nitrate was used, the ammonium was initially consumed. Afier 4

64 0 U. Rau and D. Kitamoto

=
=

30 -

20

10

MELA
MEL B
MELC

n
"-

day

7 11 21
NaNO,

7 11 21
NH,CI

7 11 21
NH,NO,

7 11 21
(NH,),SO,

Fig. 3.7. MEL formation by k! aphidis depending on the nitrogen source used. Each flask contained
medium A (Fig. 3.5) with 23.5 mM nitrogen. MEL was quantified after 7, 11 and 21 days (Rau et al.
2005c).

days the total ammonium was assimilated and the nitrate was consumed, accompanied
by simultaneous increase of pH. The concentration of MEL C was highest at the
beginning and but decreased in favor of MEL B at later stages. A deacetylation at C-6'
with subsequent acetylation at C-4' took place. A preference for sodium nitrate was
also reported in studies of Candida sp. S-10 (Kawashima et al., 1983), I? antarctica
T34 (Kitamoto et al., 1990b), and I? rugulosa NBRC 10877 (Morita et a]., 2006b). In
contrast, Candida sp. SYl6 produced higher amounts of MEL using NH,NO, than
NaNO, (Kim eta]., 2002a).

Bioreactor
Only limited information about bioreactor production of MEL is available. Kim et
al. (1999) used Candida sp. SYl6 with soybean oil as substrate in a 5-L bioreactor for
the formation of 100 g L-'crude MEL. However, the crude product contained only
4% (w/w) pure MEL. One year later, Adamczak and Bednarski (2000) succeeded in
the production of 46 g L-' MEL equivalent to a productivity of 7.4 g L-' d' achieved
by using I? antarctica ATCC 20509 in a 5-L bioreactor with 80 g L' soybean oil and
p02-control at 50%. Two-stage culturing avoided foaming but decreased MEL yield
to 28 g L'.
The MEL are metabolites that are primarily secreted if the nitrogen source is
depleted and cells have entered the stationary phase. One method to increase the yield

MannosylerythritolLipids: Productionand Downstream Processing0 65

for resting cells is to raise the amount of product-secreting biomass. Results from l?
aphidis in shake-flask cultivations (Rau et al., 2005b) showed that glucose is a better
substrate for growth compared with soybean oil and was therefore added at the start.
The amount of sodium nitrate was increased to 3 g L ' . The initial concentration
of soybean oil was essentially reduced from 80 mL L' to 20 mL L ' but could not
be fully avoided due to foaming. Foam formation is a serious problem during the
production of MEL which can lead to essential discharge of culture suspension
accompanied with loss of product and risk of infection. Conventional anti-foam
agents, such as polypropylene, polyethylene glycol, or silicone oil, have to be avoided
because the microorganisms do not consume them and, therefore the agents pollute
the hydrophobic MEL fraction during the downstream process. However, soybean
oil can simultaneously act as substrate and anti-foam agent. As a result, a medium
with two carbon sources and intermittent as well as controlled feeding was used for
cultivation (Fig. 3.8)
The cells first consumed nitrate and attained a short period of resting conditions
after about 1 day. At this time, glucose was still not completely consumed. In
order to initiate regrowth, a concentrated solution of glucose (285 g L'), sodium
nitrate (16 g L'), and yeast extract (14 g L') was added at a feed rate of 125 mL
h' after 1.75 days. Substrate feeding was stopped after 2.8 days. The feed rate and
substrate concentrations resulted from a detailed analysis of the preculture maximum
consumption rates of glucose (1.25 g L' h'), sodium nitrate (0.07 g L-' h'), and yeast
extract (0.06g L-'hl). Cell protein and bio dry mass were qualitatively determined
to be in accordance with the period of substrate and soybean oil supply. However,
during the first growth phase and late stages of cultivation essential differences are
apparent due to the accumulation of storage material, probably lipids, inside the cells
(Fig. 3.9). Therefore, cell protein was chosen as an indirect measure of biomass.
A conductivity sensor triggered the addition of soybean oil, which was primarily
emulsified in the culture suspension while glucose was present. Between 0.75 and
3.75 days, depending on the actual foam formation, a total volume of 6.1 L soybean
oil was added. Most of the consumption of soybean oil took place after glucose was
consumed. At the end of cultivation, soybean oil and released fatty acids were totally
converted by the cells. If the addition of soybean oil was stopped at the proper time,
the microorganisms were able to convert the total substrate which would facilitate the
subsequent downstream process.
MEL was secreted during the substrate feed and after the added glucose was
consumed, in spite of excess nitrate being present up to a cultivation time of 6.8
days. This observation is in contrast to the accumulation process in oleaginous yeast
(Ratledge, 1997) where nitrogen limitation is a prerequisite for lipid biosynthesis.
Potentially, the following explanations could be argued. l? aphidis is not a mere
oleaginous yeast, because the MEL is secreted in addition to the accumulation of
lipids inside the cells (Fig.3. 9). Exhaustion of nutrients other than nitrogen can also
lead to the onset of lipid formation (Granger et al., 1993). The biosynthesis of lipids

66 0 U. Rau and D. Kitamoto

P)

UI

I
E

U
0

ijj

Time (days)
Fig. 3.8. Fed-batch cultivation of i? aphidis DSM 14930 in 30-L medium B with soybean oil and glucose
as carbon sources. Three Rushton turbine impellers, 300 rpm, aeration rate: 540 L h-1. After 1.75 days,
a concentrated substrate solution 5 (glucose 285 g I-', sodium nitrate 16 g I-],yeast extract 14 gl-') was
Fed with 125 ml h.l. Due to Foam control, 6.14 soybean oil (SBO) was added between 0.75 and 3.75 days.
Insert: Formation of MEL-beads. Medium B (per liter deionised water): 30 g glucose, 20 ml soybean oil,
3 g NaNO,, 0.2 g KH,PO, 0.2 g MgS0,x7H20, 1 g yeast extract, pH=6.2 not controlled, T=27"C (Rau et al.
2005b).

could also be initiated by the first nitrate limitation after 1 day; and is not shut down
in the hrther course of cultivation.
After 5.2 days, high concentrated MEL beads appeared at a concentration of
about 50 g L' MEL. HPLC analysis of these beads revealed MEL content greater
than 80% (w/w) beside soybean oil and fatty acids. Occasionally, these beads were
also noticed during shake-flask cultivation. Due to the fact that these beads only
occurred at concentrations greater than 40 g L-l MEL, they could be regarded as
indicators for enhanced MEL production. After 11.8 days, 3.8 g L-I cell protein and

Mannosylerythritol Lipids: Productionand Downstream Processing 0 67

Fig. 3.9. Storage substances in f! aphidis observable as refractive globules during the assimilation of
soybean oil (Rau et al. 2005b).

110 g L-I MEL resulted from the last sample. After the cultivation was finished, a
huge amount of sticky MEL adhered to the cap and internal wall of the vessel. This
MEL was collected, resuspended into the culture fluid; the final result was 165 g Ll.
Kim et al. (2006)used NH,NO, as the N-source, which led to drastic variation
of pH due to consumption of first ammonia (pH decrease) and subsequent nitrate
(pH increase). Therefore, batch cultivation of Cundidu sp. SYl6 under pH control
at 4.0 yielded 48 g L ' MEL after 200 h. A two-stage culture system, similar to Fig.
3.8,under suitable growth-limiting conditions and subsequent fed-batch technique
during the stationary production phase, raised the MEL yield to 95 g L' after 200 h.
The cultivation was started with a combination of glucose and soybean oil (each 15
g L')as the initial carbon sources during the cell growth phase. The first and second
addition of 70 g L'and 100 g L' soybean oil, during the N-limiting MEL production
phase, occurred after 44 h and 104 h. The self-decreasing pH from 7.5 was controlled
at 4.0 after 24 h. With this microorganism, the foam-stat strategy, using soybean oil
as both an antifoam agent and carbon source, was not successhl and did not lead to
increased yield. The authors suppose that a higher substrate pressure is necessary for
enhanced MEL formation, which was not present due to antifoam sensor controlled
addition of soybean oil. As a result, Table 3.4 summarizes maximum MEL yields
hitherto attained in shake-flask and bioreactor cultivation.

Downstream Processing
To date, in spite of interesting applications, a real breakthrough of MEL is hampered
due to their high production cost. Beside a high yield and productivity of the bioreactor
process, the subsequent downstream handling of the product is economically crucial.
Therefore, an easy to handle procedure is required for economic isolation and
purification of MEL.

68 0 U. Rau and D.Kitamoto

Table 3.4. Maximum MELYields Achieved by Various Strains and Culture Systems
Culture Yield
Microorqanism system (q L-1
143
RrugulosaNBRC SF
10877
R antarcticaT34 SF
140

Productivity .Y,?,
(g L- d)
(q q ) Substrates
Reference
5
0.5
Soybean oil/
Morita et al. 2006b
Erythritol
5
0.89
n-octadecane Kitamoto et al. 2001a

Rantarctica ATCC BR
20509
Candidasp.SY16 BR

46

7.4

0.57

Soybean oil

95

11.4

0.45

Soybean oil/
Glucose
Soybean oil/
Glucose

RaphidisDSM
BR
165 13.9
0.92
14930
SF = Shake flask
BR = Bioreactor
.Yield coefficient = g MEL produced/ g substrate consumed

Adamuak & Bednarski

2000
Kim et al. 2006
Rau et al. 2005b

The choice of method for the isolation and purification of a particular

biosurfactant depends on its ionic charge, its solubility in water, and on whether
the product is cell bound or extracellular. The known methods used include solvent
extraction, adsorption followed by solvent extraction, precipitation, crystallization,
centrifugation, and foam fractionation. Desai and Desai (1993) as well as Syldatk and
Wagner (1987) gave an overview of these procedures.
During the MEL production process, green to yellow MEL beads separated at
the bottom at a concentration greater than 40 g L, which contain high quantities of
MEL >60% (w/w) and relatively small amounts of soybean oil ~ 2 0 %(w/w) as well
as fatty acids ~ 1 0 %
(w/w). The MEL beads can be taken as indicators for enhanced
product formation. Their consistency is similar to highly viscous oil drops, and
they could not simply be separated by filtration. Therefore, stepwise conventional
extraction techniques starting with the total culture suspension and using different
solvents is applicable for isolation and purification of MEL (Fig. 3.10).
The extraction step with MTBE yielded on average 75% (w/w) MEL, 15% (w/w)
soybean oil, and 10% (w/w) fatty acids after drying. Repeated extraction is necessary
for exhaustive transfer of MEL into the organic phase. The further enrichment to
91% (wlw) MEL and decrease to 5% (w/w) soybean oil and 4% (w/w) fatty acids is
achieved by subsequent repeated extraction using cyclohexane. The resulting purified
MEL fraction is a transparent, brown-colored, highly viscous fluid at ambient
temperature. During this procedure, about 20% (w/w) MEL is lost compared to
that contained in the culture suspension. A complete separation of residual soybean
oil and fatty acids is achievable by using n-hexane, methanol, and water as a solvent
mixture with subsequent repeated extraction by n-hexane. The purification of MEL
by TLC is documented (Fig. 3.10). Unfortunately, the process to yield pure MEL is
is inefficient and results in drop of recovery down to 8% (w/w) (Rau et al., 2005a).
Another disadvantage of this process is the production of huge amounts of waste
solvents that have to be recycled. Beside the costs for bioreactor production, recycling

Mannosylerythritol Lipids: Productionand Downstream Processing0 69

IOO%(w/w) MEL
Recovery 8% (w/w)

Fig. 3.10. Scheme of the stepwise extraction procedure by using different solvents for isolation and
purification of MEL. The TLC represents the repeated extraction steps by n-hexane. Lanes 1-6, organic
phase; lanes 7-12, aqueous phase; lanes 1,2 + 7,8, first extraction; lanes 3,4 + 9,10, second extraction;
lanes 5,6 + 11,12, third extraction. MTBE, methyl tertiary butyl ether; FA, fatty acid; SO, soybean oil (Rau
et al. 2005a).

70 0 U. Rau and D. Kitamoto

increases the manufacturing costs so that an acceptance of this bioprocess for industry
is further inhibited.
The MEL can also be isolated by preparative HPLC equipped with silica gel
columns (Kim et al., 1999; Kitamoto et a]., 2000; Rau et al., 2005b). 'This is a superior
method to produce a very pure MEL mixture or even to separate the individual MEL
A-D. However, the loss ofproduct is substantial, and so this is not a beneficial solution
in order to decrease the costs for the downstream process.
When the bioreactor production process is finished and the MEL-containing
culture suspension is transferred into a glass bottle, the separation of aggregated MEL
beads can be observed at the bottom as highly viscous fluid (Fig.3.11, left picture).
'This viscous MEL phase and the MTBE extract (Fig. 3.10) of the whole culture
suspension possessed a similar composition (Fig. 3.1 1). After sterilization of the
MEL-containing culture suspension at 121C for 20 min, two MEL-containing
phases, a solid sticky and an aqueous one, were formed, both fatty acid enriched as
well as soybean oil depleted (Fig. 3.1 I, right picture). A small volume of a primary

Before heating

After heating

v)

9
U
m

Viscous MEL-phase

Solid phase

Aqueous phase

Fig. 3.1 1. Composition of different MEL phases from a culture suspension before (left) and after (right)
heat treatment at 121C for 20 min (Rau et al. 2005a).

Mannosylerythritol Lipids:Production and Downstream Processing 0 71

soybean oil-containing top phase was also observed. After heating, the MEL extracted
by MTBE were distributed into the solid and aqueous phases at 89% and 11%
(w/v), respectively. This solid phase was easy to separate by pouring off the cell
debris-containing supernatant. About 11% (v/v) of MEL remained suspended in the
aqueous cell debris phase and could also be recovered by extraction with ethanol,
centrifugation, rotary evaporation of the solvent, and vacuum drying.
A comparison of known MEL downstream procedures is given in Table 3.5.
Unfortunately, to the best knowledge of the authors, only two references are available
with a quantitative description of the different methods. The heat treatment attained
the highest yield and is the fastest method by far, but only on average 87% (w/w)
MEL is contained in the precipitated fraction. However, this solid-enriched MEL
phase should be pure enough for most industrial applications (Kitamoto et al., 2001 b;
Hua et al., 2003, 2004; Im et al., 2003). For example, pure MEL-A, purified 100%
(w/w) MEL A-D, and a mixture of 88.3% MEL, 6.6% soybean oil and 5.1% (w/w)
fatty acids reduced the surface tension of water/air to similar data of 34.7, 26.7, and
31mN m-', respectively (Rau et al., 2005b).
Table 3.5. Different Quantitatively Described Methods for the
Downstream Processing of MEL
Method

bYield
% (w/w)

Ethyl acetate extraction + preparative aHPLC 79

Ethyl acetate extraction + preparativeaHPLC 4
8
Stepwise extraction with different solvents
Heat treatment
93
aPreparativeHPLC was performed with silica gel column.
gMEL recovered
x 100
byield= g MEL before downstream

Purity
% (w/w)

Reference

1 00

Kitamoto et al. 2000

Kim et al. 1999
Rau et al. 2OOSa
Rau et al. 2005a

100
100
a7

Conclusions
The recent results of MEL bioreactor production and downstream research show that
yields far over 100 g L' are possible by the use of the proper microorganism and
cultivation technique. In combination with facilitated heat treatment as a downstream
process, these findings should stimulate the industrial production of MEL.

References
Adamczak, M.; W. Bednarski. Influence of Medium Composition and Aeration on the Synthesis
of Biosurfactant Produced by Cundida unturcticu. Biotechnol. Lett. 2000,22, 3 13-3 16.
Banat, I.M.; RS. Makkar; S.S. Cameotra. Potential Commercial Applications of Microbial Surfacrants. Appl Microbiol Biotechnol2000,53,495-508.
Boothroyd, B.; J.A. Thorn; R.H.Haskins. Biochemistry of the Ustilaginales.XI. Characterization
of Extracellular Glycolipids Produced by Ustilago sp. Can. /. Biochem. Physiol. 1956,34, 10-14.
Crich, D.; M.A. Mora; R. Cruz. Synthesis of the Mannosyl Erythritol Lipid MEL A; Confirma-

72 0 U. Rau and D. Kitamoto

tion of the Configuration of the Meso-Erythritol Moiety. Tetrahedron 2002,58, 35-44.

Deml, G.; T. Anke; F. Obenvinkler; B.M. Gianetti; W. Steglich. Schizonellin A and B, New Glycolipids from Schimnella melanogramma. Phytochem 1980, 19, 83-87.
Desai, J.D.; A.J. Desai. Production of Biosurfactants. Biosurfactants:Production, Properties, Applications. Kosaric, N., Ed; Marcel Dekker, Inc.: New York, 1993; Vol48, 65-97.
Dossat, V.; D. Combes; A. Marty. Efficient Lipase Catalysed Production of a Lubricant and
Surfactant Formulation Using a Continuous Solvent-Free Process. J Biotechnol. 2002,97,
117-124.
Eliora, Z.R; E. Rosenberg Biosurfactants and Oil Bioremediation. Curr. Opin. BiotechnoL 2002,
12,249-253.
Fiechter, A. Biosurfactants: Moving Towards Industrial Application. Trend BiotechnoL 1992, 10,
208-2 17.
Fukuoka, T.; T. Morita; M. Konishi; T. Imura; D. Kitamoto. Characterization of New Types of
Mannosylerythritol Lipids as Biosurfactants Produced from Soybean Oil by a Basidiomycetous
Yeast, Pseudozyma shunxiensis. Jf oleo Sci. 2007% 56,435-442.
Fukuoka, T.; T. Morita; M. Konishi; T. Imura; D. Kitamoto. A BasidiomycetousYeast, Pseudozymu
tsukubuenris, Efficiently Produces a Novel Glycolipid Biosurfactant. The Identification of a New
Diastereomer of Mannosylerythritol Lipid-B. Curbo. Res. 2008,343, 555-560.
Fukuoka, T.; T. Morita; M. Konishi; T. Imura; H. Sakai; D. Kitamoto. Structural Characterization
and Surface-Active Properties of a New Glycolipid Biosurfactant, Mono-Acylated Mannosylerythritol Lipid, Produced from Glucose by Pseudozyrnu unturcticu. Appl. Microbio. Biotechn.
2007b, 76,801-810.
Granger, L.M.; I? Perlot; G. Goma; A. Pareilleux. Effect of Various Nutrient Limitations on Fatty
Acid Production by Rhodotorula glutinis. Appl. Microbiol. Biotechnol. 1993,38, 784-789.
Haskins, R.H.; J.A. Thorn; B. Boothroyd. Biochemistry of the Ustilagenales,XI. Metabolic Products of Ustilago zeue in Submerged Culture. Cun./. MicrobioL 1955, I , 749-756.
Hewald, S.; K. Josephs; M. Bolker. Genetic Analysis of Biosurfactant Production in Ustihgo muydis. Appl. Environ. Microbiol. 2005, 71,3033-3040.
Hewald, S.; U. Linne; M. Scherer; M.A. Marahiel; J. Kamper; M. Bolker. Identification of a Gene
Cluster for Biosynthesis of Mannosylerythritol Lipids in the Basidiomycetous Fungus Ustilago
muydis. Appl. Env. Microbiol. 2006, 72, 5469-5477.
Hua, Z.; Y. Chen; G. Du; J. Chen. Effects of Biosurfactants Produced by Candidd untazcticn on
the Biodegradation of Petroleum Compounds. WorkiJ MicrobioL BiotechnoL 2004,20,25-29.
Hua, Z.; J. Chena; S. Luna; X. Wang. Influence of Biosurfactants Produced by Cundidn antarcticu
on Surface Properties of Microorganism and Biodegradation of n-Alkanes. Water Res. 2003,37,
4 143-4 150.
Im, J.H.; H. Yanagishita; T. Ikegami; Y.Takeyama;Y. Idemoto; N. Koura; D. Kitamoto. Mannosylerythritol Lipids, Yeast Glycolipid Biosurfactants, Are Potential f i n i t y Ligand Materials for
Human Immunoglobulin G.]. Biomed. Muter. Res. 2003,65A, 379-385.
Kakugawa, K.; M. Tamai; K. Imamura; K. Miyamoto; S . Miyoshi; Y. Morinaga; 0.Suzuki; T.
Miyakawa. Isolation of Yeast Kurtzmunomyces sp. I- 1 1 , Novel Producer for Mannosylerythritol
Lipid. Biosci. Biotech. Biochem. 2002, 62, 188-1 9 1 .

Mannosyierythritol Lipids: Production and Downstream Processing 0 73

Karanth, N.G.K.; EG. Deo; N.K. Veenanadig. Microbial Production of Biosurfactants and Their
Importance. Cum Sci. 1999, 77, 116126.
Kawashima, H.; T. Nakahara; M. Oogaki; T. Tabuchi. Extracellular Production of a Mannosylerythritollipid by a Mutant of Candida sp. from n-Alkanes and Triacylglycerols. J. Ferment.
Zchnol. 1983, GI,143-149.
Kim, H.-S.; J.-W. Jeon; B.-H. Kim; C.-Y. Ahn; H.-M. Oh; B.-D. Yoon. Extracellular Production
of a Glycolipid Biosurfactant, Mannosylerythritol Lipid, by Candidz sp. SYl 6 Using Fed-Batch
Fermentation. App.! Microbiol. Biotechn. 2006, 70,39 1-396.
Kim, H.-S.; B.D. Yoon; D.H. Choung; H.-M. Oh; T. Katsuragi; Y. Tani. Characterization of a
Biosurfactant, MEL, Produced from Candida sp. SYl 6. Appl. Microbiol. Biotechnol. 1999,52,
71 3-721.
Kim, H.S.; J.W. Jeon; H.W. Lee; Y.I. Park; W.T. Seo; H.M. Oh; T. Katsuragi; Y.Tani; B.D. Yoon.
Extracellular Production of a Glycolipid Biosurfactant, Mannosylerythritol Lipid, from Candida
antarctica. Biotechn. Lett. 2002a, 24, 225-229.
Kim, S.W.; H.J. Hwang; J.P Park; Y.J. Cho; C.H. Song; J.W Yun. Mycelial Growth and ExoBiopolymer Production by Submerged Culture of Various Edible Mushrooms Under Different
Media. Lett. Appl. Microbiol. 2002b, 34, 5G-61.
Kitamoto, D.; S. Akiba; C. Hioki; T. Tabuchi. Extracellular Accumulation of Mannosylerythritol
Lipids by a Strain of Candida antarctica.Agric. BioL Chem. 199Oa, 54, 31-36.
Kitamoto, D.; K. Haneishi; T Nakahara; T. Tabuchi. Production of Mannosylerythritol Lipids by
Candida antarctica from Vegetable Oils. Agri. Bio. Chem. 1990b, 54, 37-40.
Kitamoto, D.; T. Yokoshima; H. Yanagishita; K. Haraya; H.K. Kitamoto Formation of glycolipid
biosurfactant, mannosylerythritol lipid, by Candida antarctica from aliphatic hydrocarbons via
subterminal oxidation pathway. JJpn Oil Chem SOC1999,48, 1377-1384.
Kitamoto, D.; S . Ghosh; 0. G.; Y. Nakatani. Formation of Giant Vesicle from Diacylmannosylerythritols and Their Binding to ConcanavalinA. Chem. Comm. 2000, 10, 861-862.
Kitamoto, D.; T. Ikegami; G. Suzuki; A. Sasaki; Y. Takeyama; Y. Idemoto; N. Koura; H. Yanagishita. Microbial Conversion of n-Alkanes into Glycolipid Biosurfactants, Mannosylerythritol
Lipids, by Pseudozyma (Candida) antarctica. Biotechnol. Lett. 2001a, 23, 1709-1714.
Kitamoto, D.; H. Isoda; T. Nakahara. Functions and Potential Applications of Glycolipid Biosurfactants - from Energy-Saving Materials to Gene Delivery Carriers./. Biosci. Bioeng, 2002, 94,
187-20 1.
Kitamoto, D.; T. Nemoto; H. Yagishita; T. Nakane; H. Kitamoto; T. Nakahara. Fatty Acid
Metabolism of Mannosylerythritol Lipids as Biosurfactants Produced by Candida antarctica
-Incorporation of Medium-Chain Fatty Acids as Beta-Oxidation Intermediates into Glycolipids.
J. Jap. Oil Chem. Soc. 1993, 42, 346358.
Kitamoto, D.; H. Yanagashita; K. Haraya; H.K. Kitamoto. Effect of Cerulenin on the Production
of Mannosyl-Erythritol Lipids as Biosurfactants by Candida antarctica. Biotechnol. Lett. 1995,
17,25-30.
Kitamoto, D.; H. Yanagishita; A. Endo; M. Nakaiwa; M. Nakane; T.Akiya. Remarkable Antiagglomeration Effect of a Yeast Biosurfactant, Diacylmannosylerythritol, on Ice-Water Slurry for
Cold Thermal Storage. BiotechnoL Progress 2001b, 17, 362-365.

74 0 U. Rau and D. Kitamoto

Kitamoto, D.; H. Yanagishita; K. Haraya; H.K. Kitamoto. Contribution of aCchain-Shortening
Pathway to the Biosynthesis of the Fatty Acids of MEL in the Yeast Candida antartica. Biotechnol. Lett. 1998,20, 813-818.
Konishi, M.; T. Morita; T. Fukuoka; T. Imura; K. Kakugawa; D. Kitamoto. Production of Different Types of Mannosylerythritol Lipids as Biosurfactants by the Newly Isolated Yeast Strains
Belonging to the Genus Pseudozyma. Appl. Microbiol. Biotechn. 2007, 75, 521-531.
Konishi, M.; T. Morita; T. Fukuoka; T. Imura; K. Kakugawa; D. Kitamoto. Efficient Production
of Nannosylerythritol Lipids with High Hydrophilicity by Pseudozyma hubeiensis KM-59. Appl.
Microbiol. Biotechn. 2008, 78, 37-46.
Kuiper, I.; E.L. Lagendijk; R. Pickford; J.P Derrick; G.E.M. Lamers; J.E. Thomas-Oates; B.J.J.
Lugtenberg; G.V. Bloemberg. Characterization of Two Pseudomonasputida Lipopeptide Biosurfactants, Putisolvin I and 11, Which Inhibit Biofilm Formation and Break Down Existing
Biofilms. Mol. Microbiol. 2004,51,97-1 13.
Kurz, M.; C. Eder; D. Isert; Z. Li; E.F. Paulus; M. Schiell; L. Toti; L. Vertesy; J. Wink; G. Seibert.
Ustilipids, Acylated P-D-Mannopyranosyl D-Erythritols from Ustilago maydis and Geotrichum
candidurn./. Antibiot. 2003,5G, 91-101.
Lang, S. Biological Amphiphiles (Microbial Biosurfactants). Cum Opin. Colloid Interjace Sci.
2002,7,12-20.
Lang, S.; W. Trowitzsch-Kienast. Biotenside. B. G. Teubner: Stuttgart, 2002, pp. 156170.
Lee, J.K.; S.Y. Kim; Y.W. Ryu; J.H. Seo; J.H. Kim. Purification and Characterization of a Novel
Erythrose Reductase from Candida magnoliae. Appl. Environ. Microbiol. 2003, G7,3710-3718.
Lin, S.-C. Biosurfactants: Recent Advances. J Chem. Technol. Biotechnol. 1996, GG,109-120.
Lu, J.R.; X.B. Zhao; M. Yaseen. Biomimetic Amphiphiles: Biosurfactants. Curt. Opin. Colloid
Inteface Sci. 2007, 12, 60-67.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; D. Kitamoto. Analysis of Expressed Sequence Tags
from the Anamorphic Basidiomycetous Yeast, Pseudozyma antarctica, Which Produces Glycolipid Biosurfactants, Mannosylerythritol Lipids. East 2006%23,66 1-67 l .
Morita, T; M. Konishi; T. Fukuoka; T. Imura; D. Kitamoto. Discovery of Pseudozyma rugulosa
NBRC 10877 as a Novel Producer of the Glycolipid Biosurfactants, Mannosylerythritol Lipids,
Based on rDNA Sequence. AppL Microbiol. Biotech. 2006b, 73, 305-313.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; D. Kitamoto. Microbial Conversion of Glycerol
into Glycolipid Biosurfactants, Mannosylerythritol Lipids, by a Basidiomycete Yeast, Pseudozyma
antarctica JCM 103 17T.J Biosci. Bioeng. 2007a, I04 78-8 1.
Morita, T.; M. Konishi; T. Fukuoka; T Imura; D. Kitamoto. Physiological Differences in the
Formation of the Glycolipid Biosurfactants, Mannosylerythritol Lipids, Between Pseudozyma
antarctica and Pseudozyma aphidis. Appl. Microbiol. Biotech 2007b, 74, 307-3 15.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; H.K. Kitamoto; D. Kitamoto. Characterization
of the Genus Pseudozyma by the Formation of Glycolipid Biosurfactants, Mannosylerythritol
Lipids. FEMS East Res. 2007c, 7,286292.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; S. Yamamoto; M. Kitagawa; A. Sogabe; D.
Kitamoto. Identification of Pseudozyma graminicokz CBS 10092 as a Producer of Glycolipid
Biosurfactants, Mannosylerythritol Lipids. J Oleo Sci. 2008,57, 123-13 1.

Mannosylerythritol Lipids: Production and Downstream Processing 0 75

Ratledge, C. Microbial Lipids. Biotechnology. Rehm, H.J., G. Reed, Eds.; VCH Verlagsgesellschaft:
Weinheim, 1997; Vol7, 133-197.
Rau, U.; L.A. Nguyen; H. Roeper; H. Koch; S. Lang. Downstream Processing of Mannosylerythritol Lipids Produced by Pseudozyrna aphidis. Eur. /. Lipid Sci. Technol. 2005a, 107, 373-380.
Rau, U.; L.A. Nguyen; S. Schulz; V. Wray; M. Nimtz; H. Roeper; H. Koch; S. Lang. Formation
and Analysis of Mannosylerythritol Lipids Secreted by Pseuhzyma aphidis. Appl. Microbiol.
Biotechnol. 2005b, 66, 551-559.
Rau, U.; L.A. Nguyen; S. Schulz; V. Wray; M. Nimtz; H. Roeper; H. Koch; S. Lang Formation
and analysis of mannosylerythritol lipids secreted by Pseudozyma aphidis. Appl Microbiol Biotechno1 2005c, 66, 551-559.
Rosenberg, E.; E.Z. Ron. High- and Low-Molecular-MassMicrobial Surfactants.Appl. Microbiol.
Biotechn. 1999,52, 154-162.
Sekhon, B.S. Biosurfactants: An Overview. Nut. Acad. Sci. Lett. 2006,29, 317-332.
Shibahara, M.; X. Zhao; Y. Wakamatsu; N. Nomura; T. Nakahara; C. Jin; H. Nagaso; T, Murata;
K.K. Yokoyama. Mannosylerythritol Lipid Increases Levels of Galactoceramide in and Neurite
Outgrowth from PC12 Pheochromocytoma Cells. Cytotechnol. 2OOO,33,247-25 1.
Singh, A.; J.D. Van Hamme; 0.P.Ward. Surfactants in Microbiology and Biotechnology: Part 2.
Application Aspects. Biotechn. Adv. 2007,25, 99-121.
Spoeckner, S.; V. Wray; M. Nimtz; S. Lang. Glycolipids of the Smut Fungus Ustikzgo maydb from
Cultivation on Renewable Resources. Appl. Microbiol. Biotechnol. 1999,51, 33-39.
Sugita, T.; M. Takashima; N. Poonwan; N. Mekha; K. Malaithao; B. Thungmuthasawat; S.
Prasarn; l? Luangsook; T. Kudo. The First Isolation of Ustilaginornycetous Anamorphic Yeasts,
Pseudozyma Species, from Patients Blood and a Description of Two New Species: I!parantarctica
and I! thailandica. Microbiol. Immunol. 2003, 47, 183-1 190.
Syldatk, C.; F. Wagner. Production of Biosurfactants. Biosu8actants and Biotechnology. Kosaric, N.,
W.L. Cairns, N.C.C. Gray, Eds; Marcell Dekker, Inc.: New York, 1987; Vol25, 89-120.
Van Hamme, J.D.; A. Singh; 0.P Ward. Physiological Aspects: Part 1 in a Series of Papers Devoted to Surfactants in Microbiology and Biotechnology. Biotechn. Adv. 2006,24, 604-620.
Vertesy, L.; M. Kurz; J. Wink; G. Noelken. Ustilipides, Method for the Production and the Use
Thereof, US 6,472,158, 2002.
Wang, Q.-M.; J.-H. Jia; F.-Y. Bai. Pseudozyma hubeiensis sp. nov. and Psetufozyma shanxiensis sp.
nov., Novel Ustilaginomycetous Anamorphic Yeast Species from Plant Leaves. Int. /. Syst. Evol.
Microbiol. 2006,56,289-293.
Zhao, X.; X. Geltinger; S. Kishikawa; S. Ohshima; S. Murata; S. Nomura; S. Nakahara; K.K.
Yokoyarna. Treatment of Mouse Melanoma Cells with Phorbol 12-Myristate 13-Acetate Counteracts Mannosylerythritollipid-Induced Growth Arrest and Apoptosis. Cytotechnol. 2000,33,
123-130.

Advances in Bioprocess
Development of Rhamnolipid and
Sophorolipid Production
Neissa M. Pinzon, Qin Zhang, Srujana Koganti, and Lu-Kwang Ju
Department of Chemical and BiomolecularEngineering,The Universityof Akron, Akron, OH 44325-3906

Introduction
Rhamnolipids and sophorolipids are two major glycolipid biosurfactants that have
attracted significant attention because of their potential medical, cosmetic, hygienic,
environmental, and other industrial applications. The focus of this chapter is on the
microbial-fermentation processes for the industrial production of these glycolipids.
Some brief descriptions about their structures, properties, and applications are
included, particularly those affecting the designs and operations of the fermentation
processes.

Rhamnolipids and Sophorolipids

Structures
Rhamnolipids consist of one or two molecules of rhamnose linked to a
P-hydroxyalcanoic acid or a chain of two P-hydroxyalcanoic acids joined by an ester
bond. Figure 4.1 shows the structures of rhamnolipids most commonly found in the
producing-fermentation processes.
The two most abundant rhamnolipids found in fermentation broths are rhamnosylP-hydroxydecanoyl-P-hydroxydecanoate (Rha-C10-CIO), a mono-rhamnolipid;
and rhamnosyl-rhamnosyl-P-hydroxydecanoyl-P-hydroxydecanoate
(Rha-Rha-C10CIO), a di-rhamnolipid (Soberon-Chavez et al., 2005). However, several hornolog
molecules with other fatty-acid chains, differing in chain length and/or the extent of
saturation, were identified, for instance, using high-performance liquid chromatography coupled with mass spectrometry (HPLC-MS) (Chayabutra & Ju, 2001; Deziel et
al., 2000; Haba et al., 2003). Lepine et al. (2002) further showed that, for the isomeric
rhamnolipids with an ester chain of two P-hydroxyalcanoic acids of different lengths
(for example, Rha-ClO-C8 and Rha-C8-C10) found in Pseudumonas aeruginusu culture broths, the compound with the shorter P-hydroxyalcanoate linked to the sugar
(for example, Rha-C8-C10) was at least three times more abundant than the other

77

78 0 N.M. Pinzon et al.

compound (for example, Rha-ClO-C8). In addition, if the longer P-hydroxyalcanoate

was unsaturated, the rhamnolipid with the shorter P-hydroxyalcanoate adjacent to
the sugar was more than 20 times more abundant than its isomeric counterpart (Lepine et al., 2002).
Sophorolipids consist of a sophorose (a dimer of glucose) linked with a glycosidic
bond to a terminal (a)or subterminal (0-1) hydroxyl fatty acid. Sophorolipids can
have open (acidic) and closed (lactonic) structures as shown in Fig. 4.2. The hydroxyl
fatty-acid moiety of the lactonic sophorolipids formed a macrocyclic lactone ring with
the 4"-OH group of the sophorose, while that of the acidic ones remained a free acid
(Casas & Garcia-Ochoa, 1999; Hu & Ju, 2001b). The sophorolipids can also vary
in their extents of acetylation at the 6'- and @'-OHgroups (typically diacetylated) as
well as the chain lengths of their fatty acids (usually C,6-C,8)(Hu & Ju, 2001b).
According to the above description, the two glycolipids have the following main
structural differences:
Sophorolipids always contain two 6-C sugar (glucose) molecules, whereas
rhamnolipids can have either one or two 6-C sugar (rhamnose) molecules.
The fatty-acid moiety of rhamnolipids has P-hydroxyl group(s), whereas that of
sophorolipids has the hydroxyl group on the 0-1 (next to the terminal) or, more
rarely, o (terminal) carbon. As a result, the rhamnolipids have 1 or 2 hydrophobic
(hydrocarbon) tails (Fig. 4. l), while the hydrophobic portion of sophorolipids is
sandwiched between the hydrophilic sophorose and the carboxylic acid or ester
(Fig. 4.2). The differently configured separation between the hydrophilic and

d
Fig. 4.1. Common types of rhamnolipids found in the Pseudornonas species: (a) Rha-C10-C10, (b)
Rha-C10, (c) Rha-Rha-C10-C10,and (d) Rha-Rha-00.

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 79

COOH
Lactonic SL

Acidic SL

Fig. 4.2. Structures of lactonic and acidic sophorolipids(SLs) (Hu & Ju, 2001a).

hydrophobic moieties is expected to affect the partition effectiveness and patterns

of the two glycolipids at the oil-water and gas-water interfaces.
Rhamnolipids always have a free carboxylic acid; consequently, the surface
activities, solubilities and other properties, and the applications of rhamnolipids
are sensitive to pH. Although the same is true for acidic sophorolipids, the
lactonic sophorolipids are not sensitive to pH change.
These structural differences not only lead to different properties and applications
of the two glycolipids, but also significantly affect the behaviors and designs of
their fermentation and downstream product separation and purification processes,
as described in the later sections of this chapter.

Properties
As their synthetic-surfactant counterparts, rhamnolipids and sophorolipids can reduce
the surface and interfacial tensions (IFTs), resulting in excellent detergent, emulsifying,
foaming, and/or dispersing properties. For instance, rhamnolipids reduce the IFT of
watedkerosene systems from 43 to <1 mN/m (Parra et al., 1989), and the surface
tension of water from 72 to <30 mN/m (Nitschke et al., 2005). Low critical micelle
concentration (CMC) properties were reported for different rhamnolipid molecules:
5 mg/L for the di-rhamnolipid Rha-Rha-C10-C10, and 40 mg/L for the monorhamnolipid Rha-C1 0-C10 (Nitschke et al., 2005). Rhamnolipids with shorter fattyacid chains, such as Rha-C10 and Rha-Rha-C10 (Fig. 4,1), have larger CMC values
of about 200 mg/L. The pKa of a mono-rhamnolipid mixture in water measured
by potentiometric titration was determined to be 4.28 f 0.16 and 5.50 f 0.06 for
concentrations below and above the CMC of this rhamnolipid, respectively (LebronPaler et al., 2006). In this same study, 'H nuclear magnetic resonance (NMR) titration
and spectrochemical titration using attenuated total reflectance Fourier transform-

80 0 N.M. Pinzon et al.

infrared (FT-IR) were also used to measure the pKa of the mono-rhamnolipid.
These analyses resulted in a value of 4.39 f 0.06 and 4.84 f 0.05, respectively. The
first value was measured at a concentration close to the CMC, and the latter at a
concentration higher than the CMC. The hydrophilicllipophilic balance (HLB) of a
rhamnolipid produced by l? aeiuginosa UG2 was estimated to be 24.1 using group
contributions and 17.0 using a correlation of HLB with CMC for sodium carboxylic
acids (Noordman et al., 2002). Another study using l? aeruginosa AT10 reported an
HLB value of 10.07 (Abalos et al., 2004).
Sophorolipids have diverse physicochemical properties because of their different
molecular structures, particularly the presence of both lactonic and acidic forms.
Properties such as lowering surface tension, foam-formation capacity, solubility in
organic solvents and water, and antimicrobial activity are affected by the degree of
lactone formation. In general, lactonic sophorolipids are more effective in lowering
the surface tension of water, and have stronger antimicrobial activity, whereas the
acidic sophorolipids at pH >pKa are more foaming and more soluble in aqueous
solutions. The presence of acetyl groups also affects the properties of sophorolipids:
for example, acetyl groups lower the hydrophilicity and, reportedly, enhance the
antiviral- and cytokine-stimulating effects of the glycolipids (Shah et al., 2005).
Accordingly, the diacetylated lactonic sophorolipids have oily property, and are
not readily soluble in water, whereas the nonacetylated acidic sophorolipids are readily
soluble in water. Purified diacetylated lactonic sophorolipids have solubility no more
than 70 mg/L (Otto et al., 1999). The properties of acidic sophorolipids are expected
to vary with pH. For example, the solubility of acidic sophorolipids in water (pH
uncontrolled) was reported to be 0.4 g/L (Rau et al., 1999), but was found to be in
the range of 12-15 g/L in phosphate buffers at a pH of 6-8 (Hu & Ju, 2001a). The
reported values of lowered surface and IFTs as well as the CMC in aqueous solutions
also varied substantially, presumably due to the different composition and purity of
the samples and the measurement methods employed. Davila et al. (1993) reported
that sophorolipids could lower the surface tension ofwater from 72.8 mN/m to 40-30
mN/m, with a C M C of 40 to 100 mg/L. A mixture of purified lactonic sophorolipids
lowered the surface tension of water to 36 mN/m, with a CMC of 10 mg/L (Otto
et al., 1999). Purified lactonic sophorolipids (10 mg/L) reduced the IFT between
n-hexadecane and water from 40 mN/m to about 5 mN/m, relatively independent
of a pH (in the range of 6-9) and a temperature (in the range of 20-90C). O n the
other hand, a mixture of acidic sophorolipids could reduce the IFT of hexadecane or
several vegetable oils to 1-2 mN/m (Rosenberg & Ron, 1999). Most recently, Ashby
et al. (2008) reported the following properties for four sophorolipid mixtures (with
292% being lactonic): minimal surface tension-35-36
mN/m, CMC-35 to >200
mg/L, and canola (rapeseed) oil IFT-3-7
mN/m.
Similarly, a wide range of HLB values was reported: lower for acetylated lactonic
sophorolipids and higher for acidic sophorolipids (pH-dependent). For example, the
HLB was given as 3-8 for a commercially available mixture of acidic and lactonic

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 8 1

sophorolipids at a pH of 4-6 (trade name Sopolin, from MG Intobio Co., Ltd.,

South Korea). The HLB of acidic sophorolipids varied from 8 to 15 when used in the
acid form or as the monovalent (sodium or potassium) salts, and was lower at 4-7
when used in the form of divalent (calcium) salts (Hillion et al., 1998). In addition,
diacetylated acidic sophorolipids acted as excellent emulsifiers; monoacetylated acidic
ones have both foaming and cleansing abilities; and nonacetylated acidic ones have
equivalent HLB values of 30 to 40, showing good cleansing ability (Inoue et al.,
1980).
For easy comparison, the surface tension and CMC values for the two glycolipids
are summarized in Table 4.1. Rhamnolipids appear to be more effective than
sophorolipids. This higher surface-activity of rhamnolipids is presumably due to
the clearer separation of hydrophobic and hydrophilic moieties in rhamnolipids, as
described earlier in the Structures section. The properties of two common synthetic
surfactants, alkylate dodecylbenzene and sodium dodecyl sulfate, are also included in
Table 4.1 for comparison. Both of the biosurfactants are significantly more surfaceactive than the two synthetic surfactants.

Applications
Both rhamnolipids and sophorolipids were reported as biosurfactants that are
biodegradable, nontoxic to humans, and able to be produced from renewable
resources (Desai & Banat, 1997; Furuta et al., 2003). These properties make these
glycolipids attractive for more environment-friendly usages of surfactants. Many
applications were also proposed and/or demonstrated in the food, cosmetic, and
pharmaceutical industries. Excellent reviews on the applications of these glycolipids
are available elsewhere (Maier & Soberon-Chavez, 2000; Mulligan, 2004; Nitschke et
al., 2005; Singh et al., 2007; Van Bogaert et al., 2007). Only some of them are briefly
mentioned here.
Due to their excellent effectiveness in solubilizing and emulsifying hydrocarbons,
rhamnolipids are proven to be one of the most effective surfactants for bioremediation
and enhanced oil recovery (Nguyen et al., 2008; Wang et al., 2007). They also have
antibacterial, antifungal, mycoplasmacidal, and antiviral activities that promote
their potential applications in the pesticide industry (Nitschke & Costa, 2007).
In addition, one can obtain rhamnose from rhamnolipids. Rhamnose is currently
Table 4.1. Comparison of Some Propertiesof Glycolipid Biosurfactants and
Synthetic Surfactants
Surfactant
Min. SurfaceTension (mN/m)
Sophorolipids
33-37
Rhamnolipids
26-29
Detergent alkylate dodecylbenzene 47
Sodium dodecyl sulfate
37
Abbreviation: CMC, critical micelle concentration.

CMC (mg/L)
10->200
5-200
590
2000-3000

82 0 N.M. Pinzon et al.

produced by extracting quercitrin from oak bark, naringin from citrus peels, or rutin
from oak bark or other plants. This process generates considerable amounts of toxic
wastes, and involves corrosive materials (Chayabutra et al., 2001). O n the other hand,
rhamnolipids can be easily hydrolyzed to produce rhamnose.
Sophorolipids are used as a cleaning agent in dishwashers (Furuta et al., 2003).
They also were proposed to be used in the petroleum industry because their emulsifying
properties and surface and IFT-lowering properties are not significantly affected even
at high salt concentrations. They are useful in washing the drill-cutting polluted by
hydrocarbons and in regenerating hydrocarbons from mud (Baviere et al., 1994).
Also,sophorolipids have properties such as inhibiting free radical and elastase activity,
stimulating the dermal fibroblast metabolism and collagen neosynthesis, and the
fibrinolytic property (Hillion et al., 1998; Maingault, 1999). These properties make
sophorolipids attractive for cosmetic formulations. One can also use the antibacterial
properties of sophorolipids to fight against infectious diseases (Napolitano Lena,
2006). In addition, purified sophorolipids showed cytotoxic effects on cancer cells,
and therefore, had potential applications in liver-cancer treatment. The anticancer
activity was effected by triggering apoptosis in the H7402 human liver-cancer cell
(Chen et al., 2006a).

Rhamnolipid Production
Producing Microorganisms and Characteristics
The production of rhamnolipids is a unique characteristic of the opportunistic
pathogen I? aeruginosa. While some isolates of the nonpathogenic I? rlororaphis
(Gunther et al., 2005) and the pathogenic Burkholderiapseudomallei (Haussler et al.,
2003) also produce rhamnolipids, the product concentrations (or yields) obtained in
the fermentations using I? aeruginosa are, so far, significantly higher than those from
the other species. Various I? aeruginosa strains-for example, DSM 2874 (Trummler
et al., 2003), ATCC 9027 (Chen et al., 2004), ATCC 10145 (Chayabutra & Ju,
2001), and UG2 (Mata-Sandoval et al., 2001)-are reportedly good producers of
rhamnolipids.
Rhamnolipids can be obtained with batch fermentation (Chayabutra et al.,
2001; Chen et al., 2007) and continuous culture (Chen et al., 2003; Gruber et al.,
1993), and in the processes using immobilized cells (Wagner et al., 1984; Wilson
& Bradley, 1996) and/or resting cells (Wagner et al., 1984). Since rhamnolipids are
predominantly reported as the secondary metabolites overproduced by cultures under
certain nutrient limitation, the fermentation strategies for rhamnolipid production
commonly involve designs of growth-limiting conditions or employ resting cells. In
these cases, rhamnolipid levels sharply increase as a consequence of the limitation of
one or more components in the medium.
We demonstrated the production of rhamnolipids when the culture reaches the
stationaryphase ofgrowth due to thelimitation ofnitrogen or another essential nutrient
(Chayabutra et al., 2001). Optimal p H and temperature for rhamnolipid production

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 83

under aerobic conditions and in continuous culture were 6.2-6.4 and 32"C-34"C,
respectively (Guerra-Santos et al., 1986). We confirmed that rhamnolipid production
is sensitive to pH. At a lower pH of 6.0, the specific rhamnolipid production rate
was approximately 4 mg of rhamnose per gram of cells (dry weight) per hour [mg
of rhamnose/(g CDW-h)], while at a higher pH of 6.5, it increased to 12.5 mg of
rhamnose/(g CDW-h).
Rhamnolipid production is not only affected by the bacterial strain selected, but
also by the carbon source used; the nitrogen and other nutrient concentrations in
the medium; the culture conditions such as pH, temperature, and dissolved oxygen
concentration; and the design or approach of the fermentation process. We can use
various carbon sources for rhamnolipid production, for example, glycerol, ethanol,
glucose, and corn oil (Nitschke et al., 2005). The Pseudomonus species can also use
hydrocarbons for growth in aerobic conditions. The rhamnolipid concentrations
obtained are usually higher when n-alkanes and vegetable oils are used as the carbon
sources as compared to when water-soluble carbon substrates are used (Nitschke et al.,
2005). However, in the absence of oxygen, hydrocarbons cannot be degraded, at least
not effectively, for industrial-production purposes, by these microorganisms. Recent
efforts to improve the yield of rhamnolipids were also directed toward genetically
engineered strains and the use of low-cost feedstock or agricultural by-products, such as
vegetable oils and the residue from the vegetable-oil refinery, as substrates (Benincasa,
2007; Mercade & Manresa, 1994). For example, the rhamnolipid production from
different carbon sources varies significantly: ~ 1 . 2g/L using corn oil (Anna et al.,
2002), 1.5 g/L using glucose (Guerra-Santos et al., 1984), 6.9 g/L using glycerol
(Anna et al., 2002), and 45 g/L using rapeseed oil (Trummler et al., 2003).
Currently, the industrial production of rhamnolipids faces significant obstacles
in high production costs and low production rates as compared to the production of
synthetic surfactants and certain other microbial products, including sophorolipids.
One particular process challenge is caused by the fast and stable foam generated
in aerobic I! ueruginosu fermentations (Wu & Ju, 1998). Rhamnolipids, with their
free carboxylic-acid group (Fig. 4.1), and other lipidic metabolites produced by Iz
ueruginosu cause extreme foaming in the fermentation broth at the near neutral pH
optimal for the bacterial fermentation. (The foaming was much more subdued at a
pH lower than 5.5.)
Foaming is common in aerobic fermentations. Mechanical foam breakers
and chemical antifoaming agents were developed for controlling the foaming in
many fermentation processes. Unfortunately, the foaming in aerobic rhamnolipid
fermentation appears extremely fast, and is too stable to be handled by these methods
(Siemann &Wagner, 1993). In addition, chemical antifoams can have adverse effects
on the downstream recovery and purification processes besides the considerable
extra costs associated with the large amounts of antifoams required to match the
rhamnolipids produced along the fermentation process. Different strategies were
attempted to break the foams. For example, we used an external vessel to hold and

84 0 N.M. Pinzon et al.

collapse the overflowed foams, with an automatic acid addition if necessary (Wu
& Ju, 1998). The obtained broth was pumped back into the fermentor after the
foam was broken down. Siemann and Wagner (1993) used a similar setup for their
process employing immobilized cells. However, even with this setup, the foaming
was still difficult to control, and the aeration rate had to be significantly lowered.
This caused oxygen limitation, and compromised cell viability as well as rhamnolipid
productivity.
We proposed an alternative to the conventional aerobic fermentation approach
by taking advantage of the capability of l? aeruginosa to perform denitrification as
a complementary or alternative respiration route. The denitrification approach can
completely eliminate the problems associated with the severe foaming nature of
rhamnolipid broth (Chayabutra et al., 2001; Chen et al., 2003). We describe the
progress in the development of the denitrification-based fermentation technology in
more detail in a later section.

Biosynthesis Pathway and Regulations

The synthesis of the hydroxyalkanoate used specifically in rhamnolipid synthesis is
controlled by the enzyme RhlG, whose encoding gene rhlG is homologous to the
fabG gene in the general fatty- acid synthesis (Campos-Garcia et al., 1998). The genes
rbk4B and rblCare known to encode for rhamnosyltransferases,which are the enzymes
that catalyze the transfer of one (or two) rhamnose from thymidine diphosphate-lrhamnose to P-hydroxyalkanoate to form rhamnolipids. This process is regulated at
the transcriptional level by at least two interactive quorum-sensing systems b and rhl
(Fig. 4.3).
The quorum-sensing systems involve the production of particular N-acyl
homoserine lactones known as autoinducers (AIs: PAI-1 and PAI-2 in Fig. 4.3).
As the bacterial concentration increases, AI concentrations build up in the growth

PA12
Fig. 4.3. Two known quorum-sensing systems (/as and rho in f! aeruginosa involved in rhamnolipid
synthesis (Chen et al., 2004).

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 85

medium. Each A1 binds to its specific recipient protein (LasR or RhlR), forming
an AI-protein complex that activates and/or regulates various genes, including the
rhlAB and rhlG genes for rhamnolipid synthesis (Chen et al., 2005). Note that the
AI-protein complexes also activate the rhll and lad genes, thus auto-inducing the
A1 syntheses themselves.
In the kzs system, PA11 [N-(3-oxododecanoyl) homoserine lactone] is first
synthesized by Lad (an AI synthase). Subsequently, PA1 1 binds with the hR-encoded
transcriptional activator, LasR (Chen et al., 2004). In the same way, the rhl system
controls the synthesis of the second AI, PA12 (N-butyryl homoserine lactone).
?he enzyme that catalyzes this reaction is the AI synthase RhlI. PA12 binds to the
transcriptional activator protein RhlR, forming the RhlR-PAI2 complex, which
controls the expression of rhlAB and rhlG. ?he two quorum-sensing systems are not
independent (Fig. 4.3). For instance, the production of PAI2 is not only promoted by
the RhlR-PAI2 complex but also by the LasR-PAI1 complex from the h quorumsensing system (Chen et al., 2005).
We studied the kinetics of PA12 synthesis and degradation in the batch
fermentation of I! aeruginosa using the wild-type PA01 and its rhll(-) and rhlR(-)
mutants (Chen et al., 2005). The rhll(-) and rhlR(-) mutants were first found to
produce insignificant amounts of rhamnolipids as compared to the wild-type I?
aeruginosa PAOl. Then, to study the PA12 degradation kinetics, the AI-bearing
supernatant was collected from a batch culture of PAO1, and added to two systems
for comparison: one system contained the washed cells of rhlZ(-) mutant collected
from a continuous culture maintained at the dilution rate of 0.025 hour-; the other
system was cell-free. The use of rhll(-) cells was to ensure that no PAI2 synthesis
occurred during the experiment so that the degradation could be examined apart
from biosynthesis. The results are shown in Fig. 4.4. They indicated that the PA12
degradation was predominantly cell-associated. While the A1 concentration decreased
insignificantly in the cell-free systems (especially when considering the range of data
fluctuation caused by the sensitive bioassay), clear degradation was observed in the
systems containing rhll(-) cells.
PA12 concentrations were next followed along the batch fermentation of the
wild-type PAOl (Fig. 4.5) and the rhlR(-) mutant (Fig. 4.6). Both AI synthesis and
degradation were expected to take place in the two systems, but the synthesis kinetics
might differ: the synthesis would be autoinduced by the RhlR-PA12 complex in the
wild-type culture but not in the rhlR(-) mutant (without RhlR production in the
latter). The stationary-phase decay of PAI2 in the wild type was slower than that in
the rhll(-) mutant (Fig. 4.5). Assuming that the mutant had the same degradation
kinetics as the wild type, the observed difference in stationary-phase decay rates could
be attributed to the PA12 synthesis occurring with the wild type [but, as expected,
not with the rhll(-) mutant]. Nonetheless, the synthesis rate was slower than the
degradation rate, resulting in the net decay profile seen in Fig. 4.5. Furthermore,
we observed different PAI2- concentration profiles during the growth phase of the

86 0 N.M. Pinzon et al.

120

110

8 100

50
40

30

50

100

150

250

200

300

350

Time (Min)
Fig. 4.4. Degradation profiles of Al PA12 in media with ( 0 )and without (A) rh//(-) mutant cells. The
smooth lines correspond to exponential degradation with the labeled best-fit values of the degradation
constant, k, (Chen et al., 2005).

p 200

10

2
C

'3 160

-m
0

120

.-

I
C

80

8C

40

p!

0 .
0

0
2

10

12

14

16

18

Time (Day)
Fig. 4.5. Profiles of autoinducer PA12 concentration in a batch cultivation of F! aeruginosa PAO1. The
empirically exponential decay profile, with the best-fit "net" decay constant (k,-k) taking into account
both the degradation and synthesis of PA12 in the culture, was shown for the stationary phase (Chen et
al., 2005).

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 87

wild type (Fig. 4.5) and the rhlR(-) mutant (Fig. 4.6). 'The mutant's growth-phase
AI production was slower, yielding a maximal PAI2 concentration of about one-third
of those from the wild-type PAO1. As the two cultures had similar profiles of cell
growth, the results implied that the autoinduction mechanism enabled a threefold
higher rate of AI synthesis.
We extended this work to evaluate the effect of PAI2 on specific rhamnolipid
productivity, using the rhll(-) mutant cultures added with various A1 concentrations
(Chen et al., 2004). We developed a model accordingly to link the essential features
of the rhl quorum-sensing system to the observed rhamnolipid-production profiles.
As expected, higher added PAI2 concentrations gave not only higher initial synthesis
rates but also longer induced synthesis (Fig. 4.7). Interested readers are referred to
the reference (Chen et al., 2004) for more information on the mathematical model
developed, the best-fit empirical constants obtained, and the potential physical
significance of the quorum-sensing systems on rhamnolipid synthesis.

Advances in Rhamnolipid Production by Denitrifying

Fermentation Technology
Aeration andlor agitation are the mechanisms typically used in microbial fermentations
to provide the interfacial transfer of 0, from gas bubbles to the aqueous media. Oxygen
supply by surface aeration alone is generally not sufficient to provide the oxygen
requirement for processes employing high cell concentrations for high volumetric
productivity. Submerged aeration is used instead, providing air from the bottom

Time (Day)
Fig. 4.6. Profiles of autoinducer PA12 concentration in a batch cultivation of the rh/R(-) mutant of P.
aeruginosa PA01 (Chen et al., 2005).

88 0 N.M. Pinzon et al.

Fig. 4.7. Experimental results and model profiles for the rhamnolipid production by a rh//(-) mutant of
PAO1, effected by the addition of autoinducers at different initial concentrations (Chen et al., 2004).

of the fermentor that bubbles up through the fermentation broth. By breaking the
large bubbles into fine ones, the agitation also increases the gas-liquid interfacial area
available for oxygen transfer to take place.
Unfortunately, these methods cannot be readily employed for rhamnolipid
production because of the highly foaming nature of the broth. To solve the
problems associated with severe foaming, we are developing the denitrification-based
fermentation technology for rhamnolipid production. I? aeruginosa is a facultative
aerobe, capable of performing aerobic and anaerobic respiration, for the latter, in
the presence of nitrate and/or nitrite as alternative electron acceptors. The anaerobic
nitrate-based respiration is known as denitrification. Nitrate is highly water-soluble.
It can be easily supplied to the growth media, thus, eliminating the vigorous, highshear agitation required in conventional fermentors for generating fine bubbles. Since
aeration is also not essential in denitrification-based fermentation, the aeration rate
can be adjusted to avoid uncontrollable foaming problems.
In the denitrification process, nitrate is first converted to nitrite, which is then
reduced to nitrogen (N,). This process is energetically favorable, and provides N,
gas as the principal end product (Ferguson, 1994; Hernandez et al., 1991). Four
reduction steps are involved in the denitrification mechanism:

NO,-

+ NO,- + NO + N,O + N,

Although not explored in industrial fermentation processes prior to our work,

denitrification was used in biological wastewater treatment for many years (Ju et al.,
1995).
The enzyme involved in the first step of nitrate reduction, from nitrate to nitrite,
is the nitrate reductase, which is a molybdenum-containing enzyme (Zumft, 1997).

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 89

The catalytic site of nitrate reductase is oriented toward the cytoplasm, and generates
nitrite at the inner face of the membrane (Zumfi, 1997). Once nitrite is produced,
another enzyme called nitrite reductase is responsible for the next reducing step to
nitric oxide. Unlike nitrate reductases, nitrite reductases are periplasmic enzymes.
After nitric oxide is produced, the nitric reductase and nitrous-oxide reductase further
reduce the nitricoxide and nitrous oxide to nitrous and nitrogen (NJ, respectively. The
enzymes involved in the dissimilative nitrate reduction process are generally repressed
by 0, and are synthesized only under anaerobic or anoxic conditions. Nonetheless,
denitrification under aerobic conditions was reported (Ka et al., 1997; Zhao et al.,
1999). We observed this with I? aeruginosa ATCC 9027 culture, which performed
denitrification even at dissolved oxygen concentrations higher than 1 mg/L when
grown in continuous culture using glucose as the carbon source (Chen et al., 2003).
Besides the effect of oxygen on the mechanism of denitrification, Thomas et al.
(1994) studied the effect of pH on denitrification using Pseudomonus species. A p H
range from 4.0-8.8 was examined, and the maximal denitrification rate took place at
a p H around 7.0-7.5. At low p H values, such as 4.0, the nitrogen-oxide reductases,
especially the N,O reductases, were progressively inhibited, resulting in a decrease of
the overall denitrification rate.
We examined the effect of different carbon sources (i.e., palmitic acid, stearic
acid, oleic acid, linoleic acid, glycerol, vegetable oil, and glucose) on rhamnolipid
production (Chayabutra et al., 2001). All of these substrates were able to support the
growth of I? aeruginosa ATCC 10145 under both aerobic and denitrifying (anaerobic)
conditions (Fig. 4.8). However, under magnesium-limited conditions, rhamnolipid
production was only detected when hexadecane, palmitic acid, or stearic acid was
used as the carbon substrate. O n palmitic acid and stearic acid, the growth of the
bacterium was slower probably because these were solid substrates that were not
readily available to the cells.
Rhamnolipid production in aerobic fermentationswas typically done in nitrogensource-limited media. Since I? aeruginosa can use nitrate as the N-source for growth,
the media may need to be designed with other limiting nutrients. We investigated the
rhamnolipid production in systems limited by nitrogen (N), iron (Fe), magnesium
(Mg), phosphorus (P), sulfur (S), and calcium (Ca) sources, respectively (Chayabutra
et al., 2001). P limitation gave four to five times higher specific rhamnolipid
productivity than N limitation. S limitation was comparable to N limitation in
supporting rhamnolipid production, while Mg limitation gave much poorer results.
Ca and Fe were not effective for limiting cell growth.
Thevalues ofspecific rhamnolipid productivity obtainedwith different C substrates
(hexadecane versus palmitic acid) and limiting nutrients (N versus P) under aerobic
versus denitrihing conditions are summarized in Fig. 4.9. The productivity with
palmitic acid as substrate was about three times higher under the aerobic conditions
than under the denitrifying conditions. As for the effect of C substrates, the aerobic
productivity from palmitic acid was lower than that from hexadecane. The former was
about 72% of the latter under N limitation and 82% under P limitation. Furthermore,

90 0 N.M. Pinzon et al.

0.4
h

.-5

0.3

c,

c,

8
C

0.2

-aJ
0.0

10

20

30

40

50

60 70 80 90 100

Time (h)
Fig. 4.8. Growth of F! aeruginosa under denitrificationwith different carbon substrates:(A)
palmitic acid;
(0)
stearic acid; (V)oleic acid; (0)
linoleic acid; (m) glycerol; and (A)glucose (Chayabutraet al., 2001).

P limitation was about four to five times more effective than N limitation, with either
hexadecane or palmitic acid as the substrate.
We also demonstrated the feasibility of rhamnolipid production under denitrifying
conditions by using a phosphorus-limited medium with palmitic acid as the carbon
substrate (Fig. 4.10) (Chayabutra et al., 2001). Although the specific rhamnolipid
productivity under denitrification was about one-third of that under aerobic
conditions, the process did not encounter the problems of foaming and respiration
limitation.

Ongoing and Future Work

Despite the demonstrated feasibility, obstacles are associated with rhamnolipid
production by denitrification-based fermentation processes. The most serious
challenges come from the need of the controlled addition of nitrate. Although
nitrate is highly water soluble and is largely not harmfd to the bacterium under
aerobic conditions (Chayabutra & Ju, 2000), it cannot be added in batches of high
concentrations for denitrification-based fermentations. Under anaerobic, denitrifying
conditions, nitrate and particularly nitrite, which is an intermediate formed in the
denitrification process, are inhibitory and/or toxic to I! aeruginosa depending on the
concentrations added (Fig. 4.1 1). Nitrite could also accumulate rapidly to toxic levels
in starved or stressed denitrifying cultures (Chayabutra & Ju, 2000). In our earlier
studies, we demonstrated the feasibility of using online fluorometers for intracellular
NAD(P)H (the reduced form of coenzymes NADH and NADPH) to monitor the

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 9 1

Time (h)

Fig. 4.9. Comparison of specific rhamnolipid productivity obtained with different carbon substrates
(hexadecane versus palmitic acid) and limiting nutrients(N versus P) under aerobic versus denitrifying
conditions (Chayabutra et al., 2001).

Respiration:

Substrate:
Limiting Nutrient:

NO3-

02

Palmitic Palmitic
Acid
Acid

02

02

02

Hexadecane

Hexadecane

Palmitic
Acid

Fig. 4.10. Experimentalresults of P-limitedfermentation with palmitic acid as C substrate, made under
denitrifying and then aerobic conditions (Chayabutra et at., 2001).

92 0 N.M. Pinzon et at.

Time (h)
Fig. 4.1 1. Effect of nitrate and/or nitrite concentrations on the growth oft? aeruginosa under anaerobic
denitrifying conditions using glucose as thecarbon source (OD,,,:optical
density at 460 nm) (Chayabutra
& Ju, 2000).

change of the cellular electron-accepting state, including, for example, the start and
end points of denitrification in I! aeruginosa cultures (Chen et al., 2003; Ju et al.,
2005). We are working on developing a robust control strategy, based on the online
NAD(P)H fluorescence monitoring, for industrial rhamnolipid production using the
denitrification-based fermentation technology.

Sophorolipid Production
Producing Microorganisms and Characteristics
Sophorolipids were first identified as the extracellular glycolipids produced by the yeast
Tordopsis magnoliae, originally isolated from sow-thistle petals (Gorin et al., 1961).
The species was later identified as 7:apicola, currently known as Candida bombicokz.
The novel yeast species Starmella bombicola (isolated from flowers Calystegiasepium and
the associated sap beetles of genus Conotelus) identified in 1998 was the teleomorph of
C. bombicokz because of their high ability to mate with each other to form ascospores
(Rosa & Lachance, 1998). More recently, a new strain of Wickerhamiella domericgiae
synthesized sophorolipids. This strain was isolated from oil-containing wastewater
samples by using the enrichment techniques, with the measurements of the oil filmcollapsing activity and surface tension (Chen et al., 2006b).

Biosyn thesis Pathway

Overall, sophorolipid synthesis includes the synthesis of both hydrophobic (fattyacid) and hydrophilic (sophorose) moieties and the subsequent linkage of the two

Advances in Bioprocess Development of Rharnnolipidand Sophorolipid Production 0 93

moieties. C. bombicokz can synthesize sophorolipids by using various carbon substrates:

hydrophilic substrate alone (such as carbohydrate and glycerol), hydrophobic/lipidic
substrate alone (such as fatty acids, hydrocarbons, fatty-acid esters, and glycerides), and
both hydrophilic and hydrophobic substrates (Fiehler et d., 1997; Hu & Ju, 2OO1b;
Kim et al., 1997; Rau et al., 1996). The possible sophorolipid synthesis pathway
from carbohydrate is shown in Fig. 4.12 (Lang &Wagner, 1993). The sophorose in
sophorolipids is generated by the modification and transformation of the original
substrate, and the fatty-acid part is synthesized de novo via acetyl-CoA. 'These two
moieties are connected to form the sophorolipids.
Another example is shown in Fig. 4.13 for the possible pathway for sophorolipid
synthesis from triglycerides (Lang & Wagner, 1993). The triglycerides are broken
down to glycerol and fatty acids, which are incorporated into the sophorose synthesis
and fatty-acid synthesis or modification. The sophorose and fatty acid are then
combined to form the sophorolipid.

Effects of Culture Conditions on Sophorolipid Formation

C. bombicokz produces sophorolipids as the secondary metabolites. 'The synthesis is
not associated with cell growth, and is triggered by nutrient (particularly the nitrogen
source) limitation. Sophorolipids are excreted into the aqueous media during the
fermentation. 'The sophorolipid synthesis is affected by: the carbon sources, which
can function as carbon and energy substrate and/or precursors; dissolved oxygen

Sophorolipid
Fig. 4.1 2.The possible biosynthetic pathway of sophorolipidsby Candida sp. when using carbohydrates
as the substrate (Lang & Wagner, 1993).

94

N.M. Pinzon et aI.

Sophoroiipid

Fig. 4.13. The possible biosynthetic pathway of sophorolipids by Candido sp. when using triglycerides
as the substrate (Lang &Wagner, 1993).

concentration; temperature; and pH and medium composition, such as the sodiumcitrate concentration. The regulation by each of these factors is described in the
following:
A wide range ofcarbon sources can be used by C. bombicokz to grow and synthesize
sophorolipids. Nonetheless, to achieve high sophorolipid yields, the addition of both
carbohydrate (such as glucose) and a lipid precursor is essential in the production
phase (Rau et al., 1996). The type of precursor (lipidic substrate) has a significant
effect on the yield and structure of sophorolipids. For example, the precursors having
chain lengths of C,,-C,, can be incorporated directly into the hydrophobic fatty
acids in the sophorolipids without being alternated in the length or structure of the
hydrocarbon chain. For the precursors with chain lengths longer than C,, or shorter
than C,,, their carbon chains are shortened or lengthened by the cells before being
incorporated into the sophorolipids (Tulloch et al., 1962). High yields of sophorolipids
were achieved when precursors with chain lengths of C,, to C,, were used, whereas
only a low conversion of the precursors with shorter chain lengths of C,, to C,, was
typically obtained (Davila et al., 1994; Tulloch et al., 1962).
The fed-batch mode was often employed to achieve higher sophorolipid
production (Davila et al., 1992). The yield could be further enhanced by the stepwise or continuous feeding of the lipid precursor to prevent the inhibition caused by
the fatty-acid accumulation (Fiehler et al., 1997; Kim et al., 1997). The accumulation
of fatty acids in the cells represses NADPH production through inhibiting glucose-6phosphate dehydrogenase, which is the first enzyme in the pentose phosphate pathway
(Kim et al., 1997). NADPH is the cofactor for the hydroxylase that catalyzes direct

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 95

hydroxylation of fatty acids to form the hydroxyl fatty acids required for sophorolipid
synthesis. Consequently, fatty-acid accumulation would inhibit sophorolipid
production.
The effect of dissolved oxygen concentration on the sophorolipid synthesis by C
bombicokz was fully investigated. Sufficiently high dissolved oxygen concentrations,
25-1 5% ofair saturation, are necessary to achieve high sophorolipid yields (Guilmanov
et al., 2002; Lang et al., 2003; Spencer et al., 1979). Garcia-Ochoa and Casas (1999)
used an unstructured kinetic model to demonstrate the effect of the dissolved oxygen
concentration. The results also confirmed the beneficial effects of higher dissolved
oxygen concentrations on sophorolipid production.
Fermentation temperature affects cell growth and product formation. For
sophorolipid production by C. bombicokz, the optimal temperature is around 2 1C
(Goebbert et al., 1984). C bombicokz grew well in the temperature range from
22-37"C, but the sophorolipid formation dropped significantly at temperatures
above 30C (Spencer et al., 1979).
Stuewer and co-workers (1987) have thoroughly studied the effect of pH on the
sophorolipid formation by C bombicokz.Different types ofsophorolipids, such as watersoluble (acidic) or crude crystalline (lactonic), are formed at different cultivation pH
values. Water-soluble sophorolipids are formed at a pH below 2, while the crystalline
sophorolipids form at a pH between 3 and 5. Sophorolipids are not significantly
produced at a pH higher than 5 (Spencer et al., 1979; Stuewer et al., 1987). Similar
results were obtained by Davila et al. (1997) and Goebbert et al. (1984). The optimal
pH for sophorolipid formation with resting cells was 3.5, while that with the batch
fermentation was in the range of 3.5-4.5. The higher yield at lower pH was attributed
to the heightened end-product inhibition resulting from the increased solubility of
sophorolipids in the aqueous medium when pH was raised from 3.5 to 5 (Davila et
al., 1997). Stuewer et al. (1987) further observed that C. bombicokz could convert
the water-soluble sophorolipids into the crystalline sophorolipids by adjusting the
pH but not vice versa. O n the other hand, Goebbert et al. (1984) and Spencer et al.
(1979) did not observe the change in product nature (water-soluble versus crystalline)
with the changed cultivation pH, although the observations on the effect of the pH
on product yield were largely consistent among these researchers.
Other factors also affect the sophorolipid formation. For example, the inclusion
of sodium citrate in the culture medium, in the concentration range of 2.5-10 g/L,
increased the biomass yield slightly, but the sophorolipid yield significantly (Stuewer
et al., 1987). The optimal concentration of sodium citrate was 5 g/L, and some other
organic acids, such as succinate, tartrate, or malonate, can be used as alternatives.
With the addition of these organic acids, the exolipid pattern was also altered,
and more hydrophobic (lactonic) sophorolipids were formed. Evans and Ratledge
(1985) reported that citric acid was one of the factors involved in the regulation of
lipid accumulation with oleaginous yeast. Stuewer et al. (1987), on the other hand,
proposed that the main effect of the organic acids was caused by their influence on the
cultivation pH.

96 0 N.M. Pinzon et al.

Sophorolipid Production and Purification

We compared the yields and structures of sophorolipids production by C.bombicoh
using glucose as the growth C-source with or without hexadecane or soybean oil as the
second C-source in the stationary phase (Hu & Ju, 2001a). The structure identification
and quantification of all the major components of sophorolipids formed were done
using an HPLC-MS" system (HP 1100 series with W-visible diode-array detector +
Bruker's Esquire electrospray MS" analyzer), which employed electrospray-ionization
and ion-trap techniques. We found that the yields of crude sophorolipids were 0.84,
0.20, and 0.03 grams per gram of the hexadecane, soybean oil, and glucose consumed,
respectively. The types of sophorolipids produced also varied considerably with the
different C-sources used. The sophorolipid structures (and the MS spectra) obtained
are shown in Fig. 4.14. When the cells were grown on glucose alone, they produced
mainly acidic sophorolipids with the fatty-acid moiety being C18:l (chain length of
C,, with 1 double bond, m/z = 704.8) and Cl6:O (hexadecanoic, no unsaturation,
m/z = 679.3). When soybean oil was used as the second C-source, both lactonic
and acidic sophorolipids with C,, and C,, fatty-acid moieties of different extents of
unsaturation were produced. In the system with hexadecane as the second C-source,
the glycolipids produced were predominately the diacetylated lactonic sophorolipids
with palmitate as the fatty-acid moiety (Hu & Ju, 2001a).
?he changes of sophorolipid composition along the fermentation with the
addition of hexadecane or soybean oil as the second C-source are summarized in
Fig. 4.15. In the hexadecane system, the two major types of sophorolipids, with the
fatty-acid moiety being diacetylated (2-Ac) C16:O with -OH at o (16) or O- 1 (15 )
position, were produced parallel at a relatively constant rate afier about 60 hours. Very
little acidic sophorolipids were produced. In the system with soybean oil as the second
C-source, sophorolipids were produced in more complex mixtures. While acidic
sophorolipids were still formed at low concentrations, the dominant components
were a complex mixture of diacetylated (2-Ac) lactonic sophorolipids with the fattyacid chain length of C,, or C,, and with various extents of unsaturation. The results
indicated a close correspondence between the structures of the sophorolipids' lipid
moiety and the lipid precursor used. The correspondence suggested a possibility of
synthesizing sophorolipids with various desired structures.
We also studied the purification of lactonic sophorolipids from crude fermentation
products by crystallization (Hu & Ju, 2001a). As shown in Fig. 4.16a, the commonly
used crystallization solvent, ethanol, was found to have a much higher solubility of
the lactonic sophorolipids than the acid ones. Consequently, for the purification of
lactonic sophorolipids, ethanol not only lacked the selectivity in removing acidic
sophorolipids, but also resulted in a significant loss of the desired products.
The use of aqueous buffers as the solvent for crystallization proved to be much
more beneficial. The lactonic sophorolipids were considerably less soluble in aqueous
buffers than in ethanol (Fig. 4.16). Furthermore, the lactonic sophorolipids were
less soluble than their acidic counterparts (Fig. 4.16b). While both phthalate and

Advances in Bioprocess Development of Rhamnolipid and Sophorolipid Production 0 97

Fig. 4.14. Structures and mass spectra of sophorolipids produced by C. bombicolo in media containing
the following C-source(s):(a) glucose only, (b) glucose plus hexadecane as the second C-source, and (c)
glucose plus soybean oil as the second C-source.

98 0 N.M. Pinzon et al.

phosphate bufferswere more effective for purifying lactonic sophorolipids than ethanol,
phthalate buffers gave rise to a slightly more loss of the lactonic sophorolipids than
phosphate buffers. The phosphate buffers with higher pH were particularly promising
because they would keep more acidic sophorolipids soluble while precipitating out
the very poorly soluble lactonic sophorolipids. As a result, the loss of the lactonic
sophorolipids in the phosphate buffer (2.4-3.1 % of the lactonic sophorolipids
originally present in the crude product, shown in Fig. 4.17a) was much less than the
loss in ethanol (47-77%), and the purity of lactonic sophorolipids in the crystallites
improved. For example, the purity reached up to 99.6% (Fig. 4.17b) in the phosphate
buffer of pH 7.5. ?he FTIR spectroscopy confirmed that indeed significantly less
acidic sophorolipids were in the purified product than in the crude product (spectra
not shown).
According to the study results, we designed a two-step process for purifying the
lactonic sophorolipids. The first step was a wash by phosphate buffer (pH 26.5) at
a low (room) temperature to remove the more soluble acidic sophorolipids while

a
f?

i3ms (h)

Fig. 4.15.The changes of sophorolipid composition when using hexadecane (a and b) or soybean oil (c
and d) as the second C-source during the fermentation (Hu& Ju,2001b).

Advances in Bioprocess Development of Rhamnolipidand Sophorolipid Production 0 99

10

20

40

30

4.0 4.5

50

Temperature (C)

5.0 5.5 6.0 6.5 7.0 7.5 8.0

PH

Fig. 4.16. (a) Solubility of different sophorolipids in ethanol at various temperatures. (b) Solubility of
different sophorolipids in aqueous buffers at various pHs at 25C (Hu & Ju, 2001a).
3.5000

100.00

3.0000
2.5000

g 2.0000

,g

1.5000
23i .I1.0000
2:s 05000

$ 0.0000

Purity

99.00

Lost Lactonic

99.50

SLS

98.50
98.00

6.5

7.5

PH

6.5

7.5

PH

Fig.4.17. (a) Lossof lactonicsophorolipids in the phosphatebuffer at adifferent pH during crystallization;

(b) Purity of lactonic sophorolipids in the crystallites (Hu & Ju, 2001a).

minimizing the dissolution of lactonic sophorolipids. After decanting the liquid

phase, a fresh batch of the phosphate buffer was added to the remaining solids. The
second step was to first completely dissolve the sophorolipids by heating the buffer to
a higher temperature (65C)
and then to precipitate out the sophorolipid crystals by
cooling the buffer to room temperature. The process could be repeated for improved
purity if desirable.

Conclusions
Rhamnolipids and sophorolipids are two major groups of glycolipid biosurfactants that
have attracted much attention for various potential applications. The two groups have
different properties primarily because of their different hydroxyl fatty-acid moieties:
rhamnolipids lipid moiety is P-hydroxylated and has a free carboxylic- acid end group;
sophorolipids lipid moiety is primarily hydroxylated at the (O-1) position, and the
carboxylic acid group is either free (acidic sophorolipids) or intraesterified with the
sugar (lactonic sophorolipids). The above differences affect not only their properties

100 0 N.M. Pinzon et al.

(HLB values, pH dependency, water solubility, etc.) and, accordingly, applications

but also the behaviors of the fermentation processes used for their production.
Rhamnolipids are primarily produced by the bacterium I! aeruginosa at close to
neutral pH. At this pH range, rhamnolipids are negatively charged (due to proton
dissociation) and are excellent foaming agents. This property caused severe operational
difficulties and poor productivity in the conventional aerated fermentation processes.
The severe foaming prevented the use of high aeration rates and, consequently,
limited the maximal cell concentrations employable for stationary-phase rhamnolipid
production in the bioreactors. I! aeruginosa was recently identified to possess
some fermentative capability, which is, however, too weak for supporting active
industrial-production processes. O n the other hand, I! aeruginosa can perform active
denitrification as alternative anaerobic respiration. Rhamnolipid production by the
denitrifying fermentation process is currently under development.
O n the other hand, the process technology for industrial sophorolipid production
using the yeast C. bombicokz is largely mature. The optimal pH for sophorolipid
production is in the range of 3.54.5. Under this pH condition, even the acidic
sophorolipids are mostly not charged (protonated). Foaming, therefore, is not
a problem in the yeast fermentation for sophorolipid production. Furthermore,
while optimal sophorolipid production prefers a relatively highly dissolved oxygen
concentration, the yeast can tolerate anaerobic conditions without permanent
damages to the culture (unlike I! aeruginosa). Industrial- fermentation processes for
sophorolipid production are therefore easier to design and more robust to operate
than the processes for rhamnolipid production. The most active sophorolipid
production is typically achieved by using a carbohydrate (preferably glucose) as the
primary substrate and a lipid source (e.g., triglyceride oil) as the second substrate. The
lipid substrate is added continuously or in frequent batches to avoid the inhibition
of accumulated fatty acids. The structures of sophorolipids' lipid moiety correlated
with the fatty-acid structures of the lipid precursors used, particularly in the chain
lengths of C.,'-C,,, thus offering the possibility of designing the desired sophorolipid
structures. An effective method for the purification of lactonic sophorolipids by
crystallization was also developed. 'The hture research and development efforts for
sophorolipids should be focused more on their applications, and chemicaUenzymatic
modifications if necessary, in addition to the process modifications and specieslstrain
selection/engineering for using inexpensive by-products or wastes as the substrates.

Acknowledgments
The original work from the group of Lu-Kwang Ju reported in this chapter was
supported by National Science Foundation (BES-9900694 and 01 04122), U.S.
Department ofTransportation (Office of the Secretary, DTOS59-07-G-00050), Ohio
Board of Regents (for Ohio Bioprocessing Research Consortium), and The University
of Akron (Faculty Research Grant).

Advances in Bioprocess Developmentof Rhamnolipid and Sophorolipid Production 0 101

References
Abalos, A.; M. Vinas; J. Sabate; M.A. Manresa; A.M. Solanas. Enhanced biodegradation of Casablanca crude oil by a microbial consortium in presence of a rharnnolipid produced by Pseudomonas aeruginosa AT10. Biodegradation 2004, 15(4), 249-260.
Anna, L.M.S.; G.V. Sebastian; E.P. Menezes; T.L.M. Alves; AS. Santos; N. Pereira, Jr.; D.M.G.
Freire. Production of biosurfactants from Psdomonas aeruginosa PA1 isolated in oil environments. Brazilian/. Chem. Eng. 2002, 17(2), 159-166.
Ashby, R.D.; D.K.Y. Solaiman; T.A. Foglia. Property control of sophorolipids: influence of fatty
acid substrate and blending. Biotechnol. Lett. 2008,30(6), 1093-1 100.
Baviere, M.; D. Degouy; J. Lecourtier. Process for Washing Solid Particles Comprising a Sophoroside Solution. U.S.Patent 5,326,407 (1994).
Benincasa, M. Rhamnolipid produced from agroindustrial wastes enhances hydrocarbon biodegradation in contaminated soil. Curr. Microbiol. 2007,54(@, 445-449.
Campos-Garcia, J.; A.D. Caro; R. Najera; R.M. Miller-Maier; R.A. Al-Tahhan; G. SoberonChavez. The Pseudomonas aeruginosa rhlG gene encodes an NADPH-dependent j3-ketoacyl
reductase which is specifically involved in rhamnolipid synthesis.J. Bacteriol. 1998, 180(17),
4442-4451.
Casas, J.A.; F. Garcia-Ochoa. Sophorolipid production by Candidz bombicola: Medium composition and culture methods. J. Biosci. Bioeng. 1999, 88(5), 488-494.
Chayabutra, C.; L.-K. Ju. Degradation of n-hexadecane and its metabolites by Pseudomonas
aeruginosa under microaerobic and anaerobic denitrifying conditions. Appl. Environ. Microbiol.
2000, GG(2), 493-498.
Chayabutra, C.; L.-K. Ju. Polyhydroxyalkanoicacids and rharnnolipids are synthesized sequentially in hexadecane fermentation by Pseudomonas aeruginosa ATCC 10145. Biotechnol. Prog.
2001, 17(3), 419-423.
Chayabutra, C.; J. Wu; L.-K. Ju. Rharnnolipid production by Pseudomonas aeruginosa under denitrification: Effects of limiting nutrients and carbon substrates. Biotechnol.Bioeng. 2001, 72(1),
25-33.
Chen, F.; C.-C. Chen; L. Riadi; L.-K. Ju. Modeling rhl quorum-sensing regulation on rharnnolipid production by Pseudomonas aeruginosa. Biotechnol. Prog. 2004,20(5), 1325-1331.
Chen, C.-C.; L. Riadi; S.-J.Suh; D.E. Ohman; L.-K. Ju. Degradation and synthesis kinetics of
quorum-sensing autoinducer in Pseudomonas aeruginosa cultivation. JBiotechnol. 2005, 11 7(1),
1-10.
Chen, J.; X. Song; H. Zhang; Y.-B. Qu; J.-Y Miao. Sophorolipid produced from the new yeast
strain Wickerhamieh domercqiae induces apoptosis in H7402 human liver cancer cells. Appl.
MicrobioL Biotechnol. 2006a, 72(1), 52-59.
Chen, J.; X. Song; H. Zhang; Y Qu. Production, structure elucidation and anticancer properties of sophorolipid from Wickerhamiella domercqiae. Enzyme Microb. Technol. 2006b, 37(3),
50 1-506.
Chen, S.-Y.; Y-H. Wei; J.-S. Chang. Repeated pH-stat fed-batch fermentation for rhamnolipid
production with indigenous Pseudomonas aeruginosa S2. Appl. Microbiol. Biotechnol. 2007,
7G(l), 67-74.

102 0 N.M. Pinzon et al.

Chen, F.; Q. Xia; L.-K. Ju. Aerobic denitrification of Pseudomonas aeruginosa monitored by online
NAD(P)H fluorescence.Appl. Environ. Microbiol. 2003, 69(11), 671 5-6722.
Davila, A.M.; R. Marchal; N. Monin; J.l? Vandecasteele. Identification and determination of individual sophorolipids in fermentation products by gradient elution high-performance liquid chromatography with evaporative light-scattering detection. J. Chromatogx 1993, 648(1), 137-147.
Davila, A.-M.; R. Marchal; J.-I? Vandescasteele. Sophorose lipid production from lipidic precursors:Predictive evaluation of industrial substrates.J. Indus. Microbiol. 1994, 13(4), 247-257.
Davila, A.M.; R. Marchal; J.I? Vandecasteele. Kinetics and balance of a fermentation free from
product inhibition: Sophorose lipid production by Candida bombicola. Appl. Microbiol. Biotechno!. 1992,38(1), 6-1 1.
Davila, A.M.; R. Marchal; J.l? Vandecasteele. Sophorose lipid fermentation with differentiated
substrate supply for growth and production phases. Appl. Microbiol. BiotechnoL 1997, 47(5),
496-501.
Desai, J.D.; I.M. Banat. Microbial production of surfactants and their commercial potential.
Microbiol. Mol. BioLReu. 1997, 61 (1). 47-64.
Daiel, E.; F. Lepine; S. Milot; R. Villemur. Mass spectrometry monitoring of rhamnolipids from
a growing culture of Pseudomonas aeruginosa strain 57W. Biochim. BiophysActa Mol. Cell Biol.
Lipids 2000,1485(2-3), 145-152.
Evans, C.T.; C. Ratledge. The physiological significance of citric acid in the control of metabolism
in lipid-accumulating yeasts. Biotechnol. Genetic Engin. Rev. 1985,3. 347-375.
Ferguson, S.J. Denitrification and its control. Antonie van Leeuwenhoek 1994, 66(1-3), 87-1 10.
Fiehler, K.; A. Albrecht; D. Rasch; U. Rau. Continuous production of sophorolipids by Cundida
bombicola ATCC 22214. Fett/Lipid 1997,99(1), 19-24.
Furuta, T.; K. Igarashi; Y. Hirata. Biodegradable Low-Foaming Detergent Compositions Containing Sophorolipid. International Patent 2002-JP6457 (2003).
Garcia-Ochoa, F.; J.A. Casas. Unstructured kinetic model for sophorolipid production by Candida
bombicola. Enqme Microb. Echnol. 1 9 9 9 , 2 5 ( 3 , 613-621.
Goebbert, U.; S. Lang; F. Wagner. Sophorose lipid formation by resting cells of Tordopsis bombicola. Biotechnol. Lett. 1984, 6(4), 225-230.
Gorin, P.A.J.; J.ET Spencer; A.R Tulloch. Hydroxy fatty acid glycosides of sophorose from Torulopsis magnoliae. Can. J. Chem. 1961,39(4), 846-855.
Gruber, T.; H. Chmiel; 0. Kaeppeli; I? Sticher; A. Fiechter. Integrated process for continuous
rhamnolipid biosynthesis. Su$ Sci. Series 1993,48(Biosu~actants),157-173.
Guerra-Santos, L.; 0. Kaeppeli;A. Fiechter. Pseudomonas aeruginosu biosurfactant production
in continuous culture with glucose as carbon source. Appl. Environ. Microbiol. 1984, 48(2),
301-305.
Guerra-Santos, L.H.; 0. Kaeppeli; A. Fiechter. Dependence of Pseudomonds aeruginosu continuous culture biosurfactant production on nutritional and environmental factors. Appl. Microbiol.
Biotechnol. 1986, 24(6), 443-448.
Guilmanov, V.; A. Ballistreri; G. Impallomeni; R.A. Gross. Oxygen transfer rate and sophorose
487-474.
lipid production by Candida bombicola. Biotechnol. Bioens 2002, 77(5),,:

Advances in Bioprocess Developmentof Rhamnolipid and Sophorolipid Production 0 103

Gunther, N.W.I.V.; A. Nunez; W. Fett; D.K.Y. Solaiman. Production of rhamnolipids by

Pseudomonas chlororaphis, a nonpathogenic bacterium. Appl. Environ. Microbiol. 2005, 71(5),

2288-2293.
Haba, E.; A. Abalos; 0.Jauregui; M.J. Espuny; A. Manresa. Use of liquid chromatography-mass
spectroscopy for studying the composition and properties of rhamnolipids produced by different
strains of Pseudomonas ueruginosa.J. Sutf Deteq. 2003, 6(2), 155-161.
Haussler, S.; M. Rohde; N. Von NeuhofX M. Nimtz; I. Steinmetz. Structural and functional cellular changes induced by BurkhoMeriapseudomallei rhamnolipid. In&. Immun. 2003, 71(5),

2970-2975.
Hernandez, D.; EM. Dias; J.J. Rowe. Nitrate transport and its regulation by 0, in Pseudomonas
aeruginosu.Arch.Biochem. Biopbys. 1991,28G(I), 159-163.
Hillion, G.; R. Marchal; C. Stoltz; F. Borzeix. Use of a Sophorolipid to Provide Free Radical Formation Inhibiting Activity or Elastase Inhibiting Activity. U.S. Patent 5,756,471(1998).
Hu, Y.; L.-K. Ju. Purification of lactonic sophorolipids by crystallization.J. Biotechnol. 2001a, 87,

263-272.
Hu, Y.; L.-K. Ju. Sophorolipid production from different lipid precursors observed with LC-MS.
Enzyme Microb. Echnol. 2001b, 29(10), 593-601.
Inoue, S.; M.Kinta; Y. Kimura. Dehydrating Purification Process for a Fermentation Product.
U.S. Patent 4,197,166(1980).
Ju, L.-K.; F. Chen; Q. Xia. Monitoring microaerobic denitrification of Pseudomonas ueruginosa by
online NAD(P)H fluorescence.j.Indus. Microbiol. Biotechnol. 2005,32(11-12), 622-628.
Ju, L.-K.; X.Yang; J.F. Lee; W.B. Armiger. Monitoring of the biologid nutrient removal process
by an online NAD(P)H fluorometer. Biotechnol. Prog. 1995, 11(5), 545-551.
Ka, J.-0.; J. Urbance; R.W. Ye; T.-Y. Ahn; J.M. Tiedje. Diversity of oxygen and N-oxide regulation of nitrite reductases in denitrifying bacteria. FEMS Microbiol. Lett. 1997, I56(1), 5560.
Kim, S.Y.; D.K. Oh; K.H. Lee; J.H. Kim. Effect of soybean oil and glucose on sophorose lipid
fermentation by Torulopsis bombicola in continuous culture. Appl. Microbiol. Biotechnol. 1997,
48(1), 2326.

Lang, S.; A. Brakemeier; D. Wullbrandt; A. Seiffert-Storicko. Sophoroselipids,Method for Their

Production and Use. U.S. Patent 6,433,152(2003).
Lang, S.; F. Wagner. Bioconversion of oils and sugars to glycolipids. Sutf Sci. Series 1993, 48(Biosurf ctants), 205-227.
Lebron-Paler, A.; J.E. Pemberton; B.A. Becker; W.H. Otto; C.K. Larive; R.M. Maier. Determination of the acid dissociation constant of the biosurfactant monorhamnolipid in aqueous solution
by potentiometric and spectroscopic methods. Anal. Chem. 2006, 78(22), 7649-7658.
Lepine, F.; E. Deziel; S. Milot; R. Villemur. Liquid chromatographiclmass spectrometric detection
of the 3-(3-hydroxyalkanoyloxy)alkanoicacid precursors of rhamnolipids in Pseudomonas ueruginosa cultures. J. Mass Spectrom. 2002,37(1), 41-46.
Maier, RM.; G. Soberon-Chava. Pseudomonas aeruginosa rhamnolipids: biosynthesis and potential applications.Appl. Microbiol. Biotechnol. 2000, 54(5), 625-633.
Maingault, M. Utilization of Sophorolipids as Therapeutically Active Substances or Cosmetic

104 0 N.M. Pinzon et al.

Products, in Particular for the Treatment of the Skin. U.S. Patent 5,981,497 (1999).
Mata-Sandoval, J.C.; J. Karns; A. Torrents. Effect of nutritional and environmental conditions on
the production and composition of rhamnolipids by I? aeruginosa UG2. Microbiol. Res. 2001,
155(4), 249-256.
Mercade, M.E.; M.A. Manresa. The use of agroindustrid byproducts for biosurfactant production.
J. Am. Oil Chem. SOC.1994,71(1), 61-64.
Mulligan, C.N. Environmentd applications for biosurfactants. Environ. Pollut. (Amsterdam, Netherlandt) 2004,133(2), 183-198.
Napolitano Lena, M. Sophorolipids in sepsis: Antiinflammatory or antibacterial? Crit. Care Med.
2006,34(1), 258-259.
Nguyen, T.T.; N.H. Youssef; M.J. Mcinerney; D.A. Sabatini. Rhamnolipid biosurfactant mixtures
for environmental remediation. Water Res. 2008, 42(G-7), 1735-1743.
Nitschke, M.; S.G.V.A.O. Costa. Biosurfactants in food industry. Tzendt Food Sci. Technol. 2007,
I8(5), 252-259.
Nitschke, M.; S.G.V.A.O. Costa; J. Contiero. Rhamnolipid surfactants: an update on the general
aspects of these remarkable biomolecules. Biotechnol. Prog. 2005, .?I(@,
1 593-1600.
Noordman, W.H.; J.H.J. Wachter; G.J. De Boer; D.B. Janssen. The enhancement by surfactants
of hexadecane degradation by Pseudomonas acruginosa varies with substrate availability.J. Biotechnol. 2002, 94(2), 195-212.
Otto, R.T.; H.J. Daniel; G. Pekin; K. Muller-Decker; G. Furstenberger; M. Reuss; C. Syldatk.
Production of sophorolipids from whey. 11. Product composition, surface active properties,
cytotoxicity and stability against hydrolases by enzymatic treatment. Appl. Microbiol. Biotechnol.
1999,52(4), 495-501.
Parra, J.L.; J. Guinea; M A . Manresa; M. Robert; M.E. Mercade; E Comelles; M.P Bosch.
Chemicd characterization and physicochemical behavior of biosurfactants.].Am. Oil Chem. SOC.
1989, &(I),
141-145.
Rau, U.; R. Heckmann; V. Wray; S. Lang. Enzymatic conversion of a sophorolipid into a glucose
lipid. Biotechnol. Lett. 1 9 9 9 , 2 1(I I), 973-977.
Rau, U.; H. Manzke; E Wagner. Influence of substrate supply on the production of sophorose
lipids by Candida 6ombicolaATCC 22214. Biotechnol. Lett. 1996, 18(2), 149-154.
Rosa, C.A.; M.A. Lachance. The yeast genus Starmerella gen. nov. and Starmerelh bombicola sp.
nov., the teleomorph of Candida bombicola (Spencer, Gorin & Tullock) Meyer & Yarrow. Znt. J.
Syst. Evol. Microbiol. 1998,48(4), 1413-1417.
Rosenberg, E.; E.Z. Ron. High- and low-molecular-mass microbid surfactants.Appl. Microbiol.
Biotechnol. 1999, 52(2), 154-162.
Shah, V.;G.F. Doncel; T. Seyoum; K.M. Eaton; I. Zalenskaya; R. Hagver; A. h i m ; R. Gross.
Sophorolipids, microbial glycolipids with anti-human immunodeficiency virus and spermimmobilizing activities. Antimicrob. Agents Chemother. 2005, 49(10), 4093-4 100.
Siemann, M.; F. Wagner. Prospects and limits for the production of biosurfactants using immobilized biocatdysts. Su$ Sci. Series. 1993, 48(Biosufactants), 99-133.
Singh, A,; J.D. Van Hamme; 0.P Ward. Surfactants in microbiology and biotechnology: part 2.

Advances in Bioprocess Developmentof Rhamnolipid and Sophorolipid Production 0 105

Application aspects. Biotecbnol. Adv. 2007,25(1), 99-121.

Soberon-Chava, G.; F. Lepine; E. Daiel. Production of rhamnolipids by Pseudomonas aeruginosa.
Appl. Microbiol. Biotecbnol. 2005, G8(@, 718-725.
Spencer, J.F.T.; D.M. Spencer; A.P. Tulloch. Extracellular glycolipids of yeasts. Economic Microbiol
1979,3(Second. Prod. Metab.), 523-540.
Stuewer, 0.;R Hommel; D. Haferburg; H.P. Kleber. Production of crystalline surface-active
glycolipids by a strain of Torulopsis apicola. J. Biotechnol. 1987, G(4), 259-269.
Thomas, K.L.; D. Lloyd; L. Boddy. Effects of oxygen, pH and nitrate concentration on denitrification by Pseudomonas species. FEMS Microbiol. Lett. 1994, 118(1-2), 181-186.
Trummler, K.; F. Effenberger; C. Syldatk. An integrated microbiallenzymaticprocess for production of rhamnolipids and L-(+)-rhamnose from rapeseed oil with Pseudomonas spaDSM 2874.
Eur. J. LipidSci. Tecbnol. 2003, 105(10), 563-571.
Tulloch, A.P.; J.F.T. Spencer; P.A.J. Gorin. Fermentation of long-chain compounds by Torulopsis
magnoliae. I. Structures of the hydroxy fatty acids obtained by the fermentation of fatty acids
and hydrocarbons. Can. 1.Cbem. 1962,40, 1326-1338.
Van Bogaert, I.N.A.; K. Saerens; C. De Muynck; D. Develter; W. Soetaert; E.J. Vandamme.
Microbial production and application of sophorolipids.Appl. Microbiol.Biotecbnol. 2007, 7G(l),
23-34.
Wagner, F.; J.S. Kim; S. Lang; Z.Y. Li; G. Marwede; U. Matulovic; E. Ristau; C. Syldatk. Production of surface active anionic glycolipids by resting and immobilized microbial cells. Proc. 3rd
Eur. Congress ofBiotecbno1. 1984, I , 3-8.
Wang, Q.; X. Fang; B. Bai; X. Liang; P.J. Shuler; W.A. Goddard; Iii; Y. Tang. Engineering bacteria
for production of rhamnolipid as an agent for enhanced oil recovery. BiotecbnoLBioeng. 2007,
38(4), 842-853.
Wilson, N.G.; G . Bradley. The effect of immobilization on rhamnolipid production by Pseudomonasjhorescens. JAppl. Bacteriol. 1996, 81(5), 525-530.
Wu, J.; L.-K. Ju. Extracellular particles of polymeric material formed in n-hexadecane fermentation by Pseudomonas aeruginosa.J. Biotechnol. 1998,59(3), 193-202.
Zhao, H.W.; D.S.Mavinic; W.K. Oldham; F.A. Koch. Controlling factors for simultaneous nitrification and denitrification in a two-stage intermittent aeration process treating domestic sewage.
Water Res. 1999,33(4), 961-970.
Zumft, W.G. Cell biology and molecular basis of denitrification. Microbiol. Mol. Biol. Rev. 1997,
G1(4), 533-616.

Microemulsions of Rhamnolipid
and Sophorolipid Biosurfactants
Thu T. Nguyenand David A. Sabatini
Universityof Oklahoma,202 W Boyd Street, Rm 334, Norman, OK 73019

Introduction
Microemulsions have received significant attention during the last several decades
due to their thermodynamic stability, favorable phase behavior, ultralow interfacial
tension (IFT), and high solubilization capacity of both organic and aqueous phases
(Kumar & Mittal, 1999). These features of microemulsions make them valuable for
a wide range of applications.
Microemulsions produce ultralow IFT (less than 0.1 mN/m) between the
microemulsion phase and the excess oil and/or water phases, thus overcoming the
capillary forces that trap oil in porous media. In addition, microemulsions have
the ability to solubilize both oil and water. These two features promote the use of
microemulsions in enhanced oil recovery and detergency (Uchiyama et al., 2000;
Acosta et al., 2002b., 2003). The transparency of microemulsions makes them attractive
in cosmetic formulations. The ultralow IFT between oil and water also facilitates
the penetration of the formulations into human skin, making microemulsions of
interest in personal care products and drug delivery applications (Acosta et al., 2005;
Komesvarakul et al., 2006).
While only a limited number of studies have evaluated the use of biosurfactants
in microemulsion formulations, several researchers have shown that glycolipid
biosurfactants, the most common class of biosurfactants composed of carbohydrate
heads and fatty acid tails (e.g., rhamnolipid and sophorolipid), can be used in a wide
range of applications (Gautam & Tyagi, 2006). The rhamnolipid biosurfactant has
been evaluated for the remediation of petroleum-contaminated sites due to its ability
to solubilize, mobilize, and stimulate the biodegradation of petroleum hydrocarbons
(Georgiou et al., 1992; Zhang & Miller, 1992 , 1995; Herman et al., 1997; Maier &
Soberh-Chivez, 2000; Kitamoto et al., 2002; Whang et al., 2008). The rhamnolipid
biosurfactant has also been used in applications such as removal of heavy metals from
contaminated soil (Kitamoto et al., 2002; Maier & Sober&-Chivez, 2000; Mulligan
et al., 2001) and has antimicrobial and antiviral activities (Kitamota et al., 2002;
Cameotra & Makkar, 2004). The sophorolipid biosurfactant, on the other hand, has
107

108 0T.T. Nguyen and D.A. Sabatini

been evaluated for pharmaceutical application (Shah et al., 2005; Rodrigues et al.,
2006) and for cosmetic and personal care applications (Van Bogaert et al., 2007) due to
its good surface active properties and excellent skin compatibility. Replacing synthetic
(petroleum-based) surfactants with biosurfactants in product formulations is of great
interest in many applications due to their lower toxicity, higher biodegradability, and
biorenewable nature.
Although microemulsions using a single surfactant have been reported (Holmberg
& Osterberg, 1986; Sunwoo &Wade, 1992), consumer product formulations are most
often prepared with surfactant mixtures (Rosen, 1989). The purpose of combining
surfactants/co-surfactants is to obtain the proper hydrophilic/lipophilic balance for
the required oil and water phases under a specific condition, which is quite difficult
to achieve in a single-surfactant system. In most of the studies where mixtures of
surfactantdco-surfactantswere used, the surfactant mixtures show apparent advantages
and synergism over the use of a single surfactant (Huibers & Shah, 1997; Li et al.,
2005; Nguyen et al., 2008). This is especially true for biosurfactants which are not
available with the wide range of properties found in synthetic surfactants. Single
biosurfactant systems are thus less likely to provide a proper hydrophilic/lipophilic
balance for a required oil while mixtures of two or more biosurfactants with different
hydrophilicity/lipophilicity can. This chapter thus focuses on the use of biosurfactant
mixtures to achieve microemulsion formulations and improve system properties.

Microemulsions in General
Phase Behavior and interfacial Tension
Microemulsions are thermodynamically stable dispersions of two immiscible liquids
(oil and water) stabilized by surfactant films (Rosen, 1989). Microemulsions can be
distinguished from macroemulsions and miniemulsions in different ways. First of all,
microemulsions have much smaller droplets (less than 100 nm) than macroemulsions
and miniemulsions (100-400 nm or greater). Secondly, microemulsion formation
is spontaneous while macroemulsion formation requires intense agitation and
miniemulsion formation also requires prolonged mixing. Finally, and most
importantly, microemulsions are thermodynamically stable while macroemulsions
and miniemulsions are not.
The surfactant concentration at which surfactant monomers start to aggregate
to form micelles is defined as the critical micelle concentration (CMC). Above the
CMC, many system properties remain unchanged since additional surfactant forms
micelles rather than increasing the surfactant aqueous activity (Rosen, 1989). The
surfactant concentration at which a microemulsion first forms is defined as the critical
microemulsion concentration (CpC).
Winsor introduced four types of microemulsions, known as Winsor Type I, 11,
111, and IV microemulsions (Winsor, 1948; Rosen, 1989). Winsor Type I and Type
I1 microemulsions are two-phase systems. Winsor Type I (oil-in-water or Onxr)
microemulsions solubilize oil (the excess phase) in spherical normal micelles within

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants0

109

the water phase while Type I1 (water-in-oil or W/O) microemulsions solubilize water
(the excess phase) in reverse micelles within the oil phase. Type I11 microemulsions,
on the other hand, are three-phase systems which contain the excess oil and excess
water phases in equilibrium with the bilayer middle phase microemulsion. Type I11
microemulsions are formed when lamellar micelles exist. With increasing surfactant
concentration, the volume of the middle phase microemulsion increases and
eventually incorporates all the excess oil and excess water phase into a single phase
microemulsion, known as a Winsor Type IV microemulsion.
The phase change or microemulsion transition from Winsor Type I + I11 +
I1 is often produced by the change in a tuning parameter, such as temperature for
alkyl ethoxylate (AE) surfactants or electrolyte concentration for ionic surfactants.
For AE nonionic surfactants, increasing temperature causes the ethoxylate groups
to dehydrate, which increases the lipophilicity of the surfactant. As a result, the
surfactant can solubilize more oil and the micelles change from the normal spherical
shape (Type I microemulsion) to a rod-like shape and form a bilayer phase (Type I11
microemulsion) that solubilizes both oil and water in a thermodynamically stable
phase. As the temperature continues to increase, the surfactant becomes more and
more lipophilic until a point is reached where the micelles start to invert into reverse
micelles which solubilizes water into the excess oil phase (Type I1 microemulsion).
Therefore, the microemulsion transitions from Type I to Type I11 to Type I1 with
increasing temperature for AE surfactants. For ionic surfactants, adding electrolyte
reduces the repulsion between the ionic head groups, thus increasing the surfactant
lipophilicity. As the electrolyte concentration increases, the shape of the micelles
changes from normal spherical to rod-like to reverse micelles. Similar to the change in
temperature of AE surfactant systems, the microemulsion transitions from Type I to
111to I1 with increasing electrolyte concentration for ionic surfactant systems (Rosen,
1989). As seen in Fig. 5.1, with increasing lipophilicity of the surfactant, the micelles
not only change in shape, but also in size (Komesvarakul et al., 2006). R, is the radius
of the Type I micelles and R, is the radius of the Type I1 micelles. The curvature (H)
of the micelles is equal to 1/R, and 1/R, respectively. At low salinity, the surfactant is
hydrophilic and the micelle curvature is large and positive, andType I microemulsions
are formed. As the salinity increases, the net curvature decreases and approaches a net
value of zero with Type I11 microemulsion formation. When the surfactant becomes
more hydrophobic at higher salinity, the curvature decreases further and becomes
negative. At this point, Type I1 microemulsions are formed (Bourrel & Schechter,
1988; Rosen, 1989). Figure 5.1 also demonstrates that at a given temperature/salinity,
as the surfactant concentration increases the Winsor Type IV system is approached.
The IFT ofwatedoil systems y(o),
is closely related to microemulsion formation
(Bourrel & Schechter, 1988). Ultralow IFT (< 0.1 mN/m) is achieved in Type I11
middle phase microemulsions (Rosen, 1989). Kunieda and Shinoda (1982) plotted
the IFT of an AE surfactant/oil/water system as a function of temperature as shown
in Fig. 5.2. As the temperature increases, the surfactant becomes more hydrophobic.

1 10 OT.T. Nguyen and D.A. Sabatini

Type IW Single-phase microemulsion

Type ZI:
Reverse
miceller

GwMtUn(H)=l/&

Reducing Curvature
Fig. 5.1. Phase behavior and changes in curvature with surfactant concentration and hydrophobicity
as adjusted by a scanning variable (salinity for anionic surfactant (AIS); temperature for alkyl ethoxylate
surfactant (NISI) (Reprinted from Komesvarakul et al., 2006. Reprinted with permissionfrom Society of
Cosmetic Chemists).

As a result, the interaction between the surfactant and the water decreases while
the interaction between the surfactant and the oil increases. Consequently, the IFT
between surfactant-water increases and that between surfactant-oil decreases. This
occurs with the phase change from the Type I O W microemulsion to the Type I1
W/O microemulsions. 7he point when the IFT values yo-D and yD-wcross is the
optimum middle phase microemulsion where the IFT between the excess oil and
excess water (yo,)
is minimum (Bourrel & Schechter, 1988).

Solubilization and Winsor R-Ratio

Solubilization, defined as the surfactant-enhanced aqueous concentration of the solute,
is an important result of surfactant micelle formation. Solubilization is a reversible
interaction of the solute with surfictant micelles to form a thermodynamically stable
system (Rosen, 1989). Solubilization potential is represented by the volume of oil
(water) that dissolves within the micellar phase and distinguished by the solubilization
parameters (Sp3defined as

SP0 =-VO
VS

and

SPw =-VW
VS

Eq. 5.1.

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants 0

111

Fig. 5.2. Interfacial tension values between oil and microemulsion phase ,)y(,
microemulsion phase
) ,7(
as a function of temperature for an alkyl ethoxylate surfactant
and water (yBw) and oil and water
(Adapted from Kunieda and Shinoda, 1982).

where SPoand SP,are solubilization parameters for oil and water, respectively; V, V,
and V, are the volume of oil, water, and surfactant in the micellar phase, respectively.
These solubilization parameters are usually calculated excluding additives even
though they are present in the system (Bourrel & Schechter, 1988). As the salinity
concentration increases, the surfactant micelle size grows and thus can solubilize more
oil and less water. As a result, the SP, increases and the SP, decreases with the scan
of the tuning parameter (increasing temperature for AE surfactants and increasing
electrolyte concentration for ionic surfactants). Similar to the optimum formulation
obtained by evaluating IFT values, optimum solubilization can also be identified as
the point where SP, and SP, intercept. At this point, the volumes of oil and water
solubilized in the micellar phase are equal and we have the optimum solubilization
parameter (SP' = SP, = SP,; Graciaa et al., 1993). SP* can also be calculated by
equation 2 (Bourrel & Schechter, 1988):

sp* =-v*
m,
Eq. 5.2.

1 12 OT.T. Nguyen and D.A. Sabatini

where SP" is the solubilization at optimum formulation (mL oil/g surfactant), Vo is

the volume of oil solubilized in the middle phase microemulsion (mL oil) when V, =
V,(equal volumes of both in the middle phase) and m, is the amount of surhctant in
the middle phase (g surhctant). All of the surfactant is usually assumed to be present
in the middle phase.
Figure 5.3 shows a typical plot of phase behavior and solubilization parameters
as a function of electrolyte concentration (Acosta et al., 2002a). As can be seen, the
IFT of the water/oil/surfictant system and the solubilization of oil in the micellar
phase are inversely related. At the optimum point (where the IFT between the oil
and the microemulsion phase ( O h interface) and that between the water and the
microemulsion phase ( W / S interhce) are equal and minimum) the solubilization of
oil in the middle phase microemulsion is at its maximum. Therefore, at the optimum
formulation, the minimum oillwater IFT and maximum solubilization parameter are
realized. The inverse relationship is captured by the Chun-Huh relationship, which
indicates that the IFT (yoJ is inversely proportional to the square of SP by ChunHuh constant C, which is dependent on the microemulsion system, as shown in the
following equation (Huh, 1979):
fl

IFT = SP2
Eq. 5.3.

NaCI, wt%
Fig. 5.3. Phase behavior showing the transition for Winsor type I to 111 to II with corresponding interfacial
tensions and solubilization level as a function of salinity of an ionic surfactant (Adapted from Acosta et
al., 2002a).

Microemulsions of Rharnnolipid and Sophorolipid Biosurfactants0

113

The phase changes in a water/surfactant/oil system or the microemulsion

transition can also be interpreted by the Winsor's R-ratio as seen in Fig. 5.4 (Winsor,
1954). The Winsor R-ratio represents the interaction of water-surfactant-oil in water
phase and oil phase (Bourrel & Schechter, 1988):

R = Aco -Aoo - 4 - Aco

ACW
Acw - A,, - A""
Eq. 5.4.
whereAco is the interaction between the surfactant and the oil, Acw is the interaction
between the surfactant and the water phase; A,, andA,,are the interactions between
oil molecules and water molecules, respectively; A,, and A , are the interactions
between the lipophilic tails and hydrophilic heads, respectively, of the surfactant
molecules at the interface. In equation 4, the numerator and denominator represent
the net interactions of oil-surfactant and water-surfactant, respectively. As seen in Fig.
5.4, for R < 1, the water-surfactant interaction (Acw)is stronger than the oil-surfactant
interaction (AcJ and the surfactant resides in the water phase making normal micelles
(Type I) while the opposite is true for R > 1. As a result, when R c I , the characteristic
system is Type I and when R > 1, it is Type 11. At optimum formulation, R = 1, which
indicates that the oil-surfactant and water-surfactant interactions are balanced and
the surfactant prefers to be at the interface. As both the numerator and denominator
increase, in other words, as the oil-surfactant and water-surfactant interactions at the
optimum both increase while maintaining an Rvalue of 1, the optimum solubilization
(SPJis enhanced (Winsor, 1954; Bourrel & Schechter, 1988).

PLCO
Fig. 5.4. Microemulsiontransition with changes in Winsor's R-ratio (Adapted from Winsor, 1954).

1 14 0T.T. Nguyen and DA. Sabatini

The hydrophilic-lipophilic balance (HLB) method has been frequently used in

selecting the most suitable surfactant or combination of surfactants to formulate the
desired form of microemulsion (Griffin, 1949; Rosen, 1989). Lipophilic surfactants
are assigned a low HLB number while hydrophilic surfactants are assigned a high
HLB number with HLB values generally falling in the range of 0 to 40 (Griffin,
1949). Therefore, a surfactant system with a high HLB value tends to form
O N microemulsions while a low HLB surfactant system tends to form W / O
microemulsions. When the HLB of the surfactant system is approximately equal to
the HLB value required for the oil, a Type I11 microemulsion tends to form. However,
it is often difficult to obtain the required HLB for a given oil from a single surfactant.
Mixtures of surfactants tend to be more powerful in producing the required HLB
for the desired oil (Rosen, 1989). In the following sections, we will discuss the use of
surfactant mixtures in formulating microemulsions. For a mixture of two surfactants
(for example, surfactant A and surfactant B), Griffin (1949) proposed an equation that
correlates the optimum ratio between two surfactants that gives the best emulsification
and the required HLB for the oil:

WA HLBA + WBHLBB
= HLB,,
WA +WB
Eq. 5.5.
where W,and W, are the amounts (weight) of the first emulsifier (A) and the second
emulsifier (B) used, respectively; HLB, and HLB, are the assigned HLB values for
emulsifiers A and B; and HLB0,,
is the required HLB of the oil for the desired
type of emulsion. The one Caveat to this approach is that if the two surfactants have
drastically different HLB values, they may not act together but rather segregate into
separate phases; a third surfactant with intermediate HLB can prevent this segregation
(Tongcumpou et al., 2006). This equation can also be applied for a surfactant system
composed of more than two components with the expansion of the left side of the
equation.
In the following sections we will discuss the surfactant mixtures in formulating
microemulsions and the improved microemulsifying power of surfactant mixtures
over single surhctant systems. All the surfactant mixture results we present contain at
least one biosurfactant in the mixture.

Rhamnolipid-based Microemulsions
To date only limited studies have evaluated microemulsion formulations with
biosurhctants. Most of the microemulsion studies have evaluated rhamnolipid
biosurhctants with the addition of alcohol to enhance the microemulsion formation
(Xie et al., 2005). Xie et al. (2005) studied system of rhamnolipids + alcohol/nheptane/water. When n-propanol was used as the co-surfactant, only Winsor Type

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants 0

115

I microemulsions were formed, indicating the hydrophilicity of the system (R c 1).

In order to see a phase transition, the R-ratio had to be increased and the HLB had
to be decreased for this surfactant system and oil. Thus, n-butanol was used as the
co-surfactant and the phase transition was observed. The rhamnolipid biosurfactant
discussed in this section was purchased from Jeneil Biosurfictant Co. (Saukville,
Wisconsin) as a blend of 50 w/v% monorhamnolipid (CZbH4809,
MW = 504, CMC
= 104 M at neutral pH) and 50 w/v% dirhamnolipid (C32HS80,3,
MW = 650, CMC
= 1.5 x l o 4 M at neutral pH) (Helvaci et a]., 2004; Ozdemir & Malayoglu, 2004;
Ozdemir et al., 2004; Xie et al., 2005).

Rhamnolipid-aloneMicroemulsions
Rhamnolipid biosurfactants with two head groups-the anionic carboxylated head and
the bulky rhamnosyl head and two identical tails of C8 hydrocarbon (Fig. 5.5) are
known to have high HLB values of 22-24 (Xie et al., 2005). Based on the guideline
of formulating microemulsions from the Winsor R-ratio and the HLB method,
rhamnolipid is expected to be a good emulsifying agent for extremely hydrophilic
oils. Nguyen et al. (2008) studied the interfacial properties and phase behavior of
rhamnolipid biosurfactant with oils of different equivalent alkane carbon number
(EACN). The IFT between the oil phase and the aqueous phase was measured using
glass capillary tubes and a spinning drop tensiometer (Model 500 purchased from the
University ofTexas). The capillary tube is 2 mm in diameter and has a volume of 300
pL. The tube was filled with the aqueous phase (the denser phase), and then 1 pL of
the oil phase (the less dense phase) was injected into the aqueous solution to form
a droplet (Childs et al., 2004). The filled tube was then placed in the spinning drop
tensiometer and the oil droplet size was measured.

(b)
Fig. 5.5. Molecular structures of the rhamnolipids: (a) monorhamnolipid, (b) dirhamnolipid (Adapted
from Helvaci et al., 2004).

116 0T.T. Nguyen and D.A. Sabatini

EACN represents the hydrophilicity/lipophilicityof the oil; more lipophilic oils

have higher EACN oils. The four oils studied were toluene, hexane, decane, and
hexadecane with EACN values of 1, 6, 10, and 16, respectively. Rhamnolipid was
reported to produce ultralow IFT (less than 0.1 mN/m) for only toluene, which is
extremely hydrophilic as seen by its EACN value of 1 (see results in Fig. 5.6). The IFT
values for higher EACN (more lipophilic) oils were all higher than 0.1 mN/m.
Salager et al. (1979) proposed a relationship between optimal salinity (S) and different
variables in microemulsion formulations:

1n$)

= k(ACN)

+f(A) - CT + aTAT
Eq. 5.6.

where k is a constant (often between 0.1 to 0.17), ACN is the alkane carbon number
(for nonlinear hydrocarbons, the ACN becomes the EACN), AA) represents the
effect of alcohol, 0 is a function of surfactant type, aTis a constant, and AT is the
temperature difference between the studied temperature and the reference temperature
(25C). This equation indicates that as the ACN (EACN) value increases, 5" is also
expected to increase, all other variables being held constant. This is consistent with the
rhamnolipid microemulsion results shown in Fig. 5.6. As the electrolyte was added
to the ionic rhamnolipid solution, it reduced the electrical repulsion between the
ionic head groups, causing the system net curvature to decrease towards zero where

c
3

EA( "
Fig. 5.6. Relationship between optimal salinity (53 (0)
and optimal interfacial tension (/FT*) ( 0 )versus
oil EACN (toluene (1). hexane (6),decane (lo),and hexadecane(16)).The rhamnolipidconcentration was
0.1 wt% and IFT was measuredat 25 f 1C for all oils (Reprinted from Nguyen et al., 2008. Reprinted with
permission from Elsevier).

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants 0 1 17

the optimal middle-phase microemulsion is achieved (the lowest IFT and highest SP)
(Rosen, 1989). The S'increases with increasing EACN value, suggesting that more salt
was required to force the ionic surfactant into the oil/water interface. At neutral pH,
rhamnolipid is anionic in character due to the presence of the carboxylate head group.
The lowest IFT was observed for the lowest optimum salinity (i.e., toluene with the
lowest EACN value of I), indicating that rhamnolipid is best matched to toluene, and
thus requires the least salt addition to achieve the lowest IFT and generates the lowest
IFT observed for all the oils.
Nguyen et al. (2008) also developed a "fish" phase diagram of rhamnolipid
biosurfactant with toluene (Fig. 5.7), which is similar to the diagram illustrated in Fig.
5.1. The phase diagram was generated by varying the salt concentration for a series
of surfactant concentrations. The phase boundary is shown as a line that connects all
the points where a transition in microemulsion type was observed. The phase diagram
was reported at 23C (solid line) and 55C (dashed line). For a fixed rhamnolipid
concentration (e.g., 1 w/w%), the microemulsion transitions from Winsor Type I to
111 to I1 as the NaCl concentration increased. At a fixed electrolye concentration (e.g.,
12 w/w?/o), the volume of the middle phase increased with increasing rhamnolipid
concentration. As the concentration of rhamnolipid reached 10 w/w?h, the middle
phase incorporated all the excess oil and excess water into a single phase microemulsion.
Thus above 10 w/w?h rhamnolipid concentration, a single phase microemulsion Winsor
Type IV phase was observed. Figure 5.7 also demonstrates that temperature has only
a slight effect on the phase behavior of rhamnolipid biosurfactant; the system requires
slightly less salinity to form middle phase at higher temperature, indicating that the

Fig. 5.7. Phase diagram of rhamnolipid with toluene (CMC of 0.001 wt%). The boundary lines (solid for
23C; dashed for 55C) connect all the points where a transition for microemulsion type was observed.
The line labeled optimum salinity which cuts through the interior of the phase diagram corresponds to
the optimum salinity at each surfactant concentration (Reprinted from Nguyen et al., 2008. Reprinted
with permission from Elsevier).

118 0 T.T. Nguyen and D.A. Sabatini

sugar groups are less soluble at elevated temperature and thus requires lower NaCl
concentration to form Type 111 systems.
Corresponding to the fish diagram in Fig. 5.7, the IFT was plotted as a hnction
of rhamnolipid concentration at optimum formulations (Fig. 5.8). By the definition
of CMC and CpC, rhamnolipid was reported to have the CMC and CpC of 0.001
w/w% (or 1 . 9 ~ 1 0M)
~ and 0.01 w/wYo (or 1 . 9 ~ 1 0M),
~ respectively, at a salinity of
14.5 w/w%. This high optimum salinity indicates that rhamnolipid is very hydrophilic
even for the extremely hydrophilic toluene and thus is much too hydrophilic for
higher EACN oils (hexane, decane, and hexadecane).

Rhamnolipid-Co-surfactant Microemulsions
Due to its hydrophilic nature, rhamnolipid was able to form middle phase
microemulsions with only toluene, and even then the optimum salinity was very
high, ranging from 11 to 17 w/w %. In order to increase the hydrophobicity of a
system containing rhamnolipid (decrease its HLB), a surfactant with lower HLB was
mixed with rhamnolipid. When mixed with a lower HLB surfictant, the rhamnolipid
surfictant system should be able to form microemulsions with more hydrophobic oils
and requires less salinity to achieve optimum conditions.
Nguyen et al. (2008) studied the effect of mixing synthetic (petroleum-based)
surhctants with lower HLB values with rhamnolipid to formulate microemulsions
for different oils. Figure 5.9 shows the synergistic effect of using surfictant mixtures

Riiamnoiipid, wt%
Fig. 5.8. Optimum equilibrium IFT for toluene versus rhamnolipid concentration (following vertical line
in Fig. 5.7). IFTwas measuredat 25 f 1C. (Reprinted from Nguyen et al., 2008. Reprinted with permission
from Elsevier).

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants 0

1 19

2
E

Fig. 5.9. IFT of mixtures of rhamnolipid and Cl,,l,-8PO-S0,Na for toluene (01,hexane (01
decane
,
(V),
and hexadecane (A)at optimum salinity (S)for each system. Fractionis by mass and the total surfactant
concentration isO.l wt%.The IFTshown in this figure was the optimum IFTforeach formulation and was
measuredat 25 f 1C. (Reprinted from Nguyen et al., 2008. Reprinted with permission from Elsevier).

in reducing the IFT for toluene and hexane; the surfactant mixture was rhamnolipid
and a more hydrophobic co-surfactant alkyl propoxylated (PO) sulfate (C,2,,3-8POSO,Na, MW = 713), which was provided by Sasol Chemical Company. This cosurfactant has a hydrophobic characteristic due to the long hydrocarbon chain length
(CIz,13
versus C, for rhamnolipid) and the presence of the hydrophobic PO groups.
The two surfactants were mixed at different ratios that produce lower IFT as well as
required less salt at optimum formulations for toluene than rhamnolipid alone (within
the range of 3-8 w/w % of salt, compared to 1 1-1 7 w/w Yo for toluene microemulsion
when only rhamnolipid was used). We also see synergism for hexane, with IFT values of
the mixture dropping below 0.1 mN/m (middle phase formation), albeit with a higher
fraction of the hydrophobic C,z,,3-8PO-S0,Na in the mixture (higher value on the
x-axis or C,2,,3-8PO-S0,Naconcentration in the mixture-less rhamnolipid and more
CIz,,,-8PO-SO,Na in the optimum mixture than for toluene). The synthetic surfactant
alone (x value of 1 in Fig. 5.9) still generates the lowest IFT for decane and hexadecane,
demonstrating the more hydrophobic nature of this surfactant. The IFT for decane and
hexadecane increases with the increasing fraction of rhamnolipid in the mixture. This
indicates that this synthetic surhctant is not hydrophobic enough to balance the HLB
of the rhamnolipid surfactant system for the more hydrophobic decane and hexadecane.
Figure 5.9 thus not only illustrates the synergism of using surfactant mixtures, but also

120 0T.T. Nguyen and D.A. Sabatini

demonstrates the importance of having the right surfactant combination to match the
hydrophobicity of the target oil.
The results in Fig. 5.9 suggest that we should use a surfactant even more lipophilic
than C,, ,,-8PO-S04Na to achieve a low IFT for decane and hexadecane. We thus
evaluated the more lipophilic alkyl polypropylene oxide ether sulfates (C,6-18PO2EO-S04Na, MW = 1590.7) provided by Huntsman Chemical Company; its longer
alkyl group and greater number of lipophilic PO groups makes it more lipophilic than
the C,,,,3-8PO-S0,Na discussed above. The resulting data are shown in Fig. 5.10 for
an equal molar mixture of rhamnolipid and C,,-18PO-2EO-SO4Na. The optimum
salinity decreases for all oils to the range between 2 to 8 w/w% and the relationship
between optimum salinity and EACN follows the trend described in equation 6. In
contrast to the results in Fig. 5.9, the IFT for decane and hcxadecane shown in Fig.
5.10 are observed to be ultralow (< 0.1 mN/m) and about 10 times lower than the IFT
for toluene and hexane. Thus C,,-18PO-2EO-S04Na is lipophilic enough to counter
the hydrophilicity of rhamnolipid and provide an appropriate HLB to the surfactant
mixture to form middle phase microemulsions with decane and hexadecane. The
results also indicate that C,6-18PO-2EO-SO4Na is too hydrophobic for toluene and
hexane (the IFT values are larger for these oils).
Nguyen and Sabatini (2008) hrther studied the microemulsion formation of
rhamnolipid biosurfactant in mixtures with conventional synthetic surfactant sodium

EACN
Fig. 5.10. Optimum IFT ( 0 )and optimum salinity (5)(0)
of a mixture of rhamnolipid and C16-18PO-2E0
sulfate versus EACN (EACN values of 1 for toluene, 6 for hexane, 10 for decane, and 16 for hexadecane).
The total surfactant concentration was 0.1 w/w%. The ratio of rhamnolipid to Cl6-?8PO-2K)sulfate was
fixed at 3 7 by weight basis (or 1:l by molar basis). IFT was measured at 25 f 1C for all oils. (Reprinted
from Nguyen et al., 2008. Reprinted with permission from Elsevier).

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants 0

121

bis(2-ethyl) dihexyl sulfosuccinate (AOT; AOT has lower HLB, more soluble in oil) and
lipophilic linker oleyl alcohol for limonene and diesel oils. Limonene has a moderate
hydrophilicity (EACN value of =6 behaves similar to hexane from a microemulsion
perspective) and diesel is relatively hydrophobic (EACN value of 12 to 14); thus,
mixing rhamnolipid with the more hydrophobic surhctant AOT is required to form
microemulsions with these oils. Rhamnolipid is too hydrophilic while AOT is too
hydrophobic when used alone for both limonene and diesel. Therefore, neither of
these surfactants alone was able to form microemulsions with these oils. However,
mixing the two surfactants at an equal molar ratio was able to produce all three types
of microemulsions (Winsor Type I, 111, and 11) for both oils (Fig. 5.1 1). With the same
surfactant system, diesel microemulsion has higher optimum salinity (=8wt%) than
limonene ( ~ 2 . 5wt%) microemulsions, which agrees with the relationship between
optimum salinity and oil EACN proposed in equation 6. These results demonstrate
the advantage of using surfactant mixtures to promote microemulsion formation.
A "fish" phase diagram was also developed for the surfactant system containing soy
methyl ester ethoxylate (SMEE3EO), rhamnolipid biosurfactant UBR) and lipophilic
linker oleyl alcohol (OA) at a fixed ratio of SMEE3EO/JBR/OA = 4/1.75/2.5 by
weight percent with limonene oil (Nguyen & Sabatini, 2008). As can be seen in
Fig. 5.12, at 12.4 wt% total surfactant and linker concentration, the salinity of
middle phase microemulsions ranges from 7.5 to 10.5 wt% while at 2.1 wt% total
concentration, the salinity range is larger, from 11.5 to 17.5 wt%. Also, we can see

Ka('l,

?Vl%

Fig. 5.1 1. Microemulsionswith limonene and diesel oils using surfactant mixtures of AOThhamnolipid
= 0.05/0.05M. IFT was measured at 25 f 1C for both oils.Three types of microemulsions (WinsorType I,
111, and II) were observed for both oils.

122 0 T.T. Nguyen and D.A. Sabatini

e
z

NVuCl, w t v o

Fig. 5.12. Phase behavior of surfactact and linker system containing SMEEKO/JBWOA at ratio of
4/1.75/2.5wt96with limoneneoil.lnterfacialtensionsweremeasuredatlow(2.1
wt96)and high(12.4wt96)
total surfactant and linker concentrations.IFTwas measuredat 25 f 1"C.Three types of microemulsions
(WinsorType I, 111, and II)were observed at both total concentration.

that a lower salinity is required for a higher total surhctant/linker concentration

to form middle phase microemulsion, indicating that the surfactant/linker system
becomes more lipophilic at high concentration, thus requiring less salt to concentrate
the surfactant at the interface. The fish tail thus slants to the left at higher total
surfactant/linker concentrations, as illustrated in Fig. 5.1. When the hydrophilic/
lipophilic nature of the surfactant system and the oil are in balance with each other
and no relative partitioning occurs, the phase diagram is vertical; in other words, the
optimum salinity is the same at all surfactant concentrations (Bourrel & Schechter,

1988).
These results thus provide a guideline in formulating microemulsions with other
oils using the same surfactant system by manipulating the surFactant/linker ratio. For
example, to formulate microemulsion with diesel, which is more hydrophobic than
limonene, we needed to either increase the fraction of lipophilic linker oleyl alcohol
or decrease the fraction of hydrophilic biosurfactant rhamnolipid to form a more
lipophilic surfactant mixture.

Sophorolipid-based Microemulsions
Sophorolipid biosurfactant (Fig. 5.13), donated by the United State Department
of Agriculture (USDA) with high purity (~100%active), was a diacetylated open
ring with two heads-sucrose head attached to two acetyl groups and carboxylic
acid head. The hydrophilicity/lipophilicity of sophorolipid was assessed based on its

Microemulsionsof Rhamnolipid and Sophorolipid Biosurfactants 0

123

H
Fig. 5.1 3. Molecular structure of sophorolipid biosurfactant.

ability to shift the optimum salinity in microemulsion formulations. Before studying

sophorolipid, the surfactants AOT and sodium dihexyl sulfosuccinate (AMA) and the
hydrotrope sodium mono- and dimethyl naphthalene (SMDNS) were used to make
microemulsions with limonene. This system was able to form microemulsions with
limonene with an optimum salinity of 3 wt% (Fig. 5.14). By adding sophorolipid
into this surfactant system, the optimum salinity was reduced from 3.0 to 1.5 wt%
(Fig. 5.14). This result indicates that sophorolipid is more lipophilic than the AOT/
AMA/SMDNS mixture. As also seen in Fig. 5.14, adding sophorolipid increased the
optimum solubilization parameter from 3.4 to 5.5 mL oil/g surfactant, showing its
greater solubilization potential.
Microemulsions for limonene and diesel oils were also studied using surfactant
mixtures of biorenewable surfactant methyl ester ethoxylate (MEE), rhamnolipid
and sophorolipid. Biorenewable surfactants are defined as surfactants made from
renewable feed stocks, such as plants; thus, they are readily biodegradable, have low
toxicity and come from biorenewable sources. The biorenewable surfactant used here
is SMEE3EO made from soya and has three ethylene oxide groups. Two surfactant
mixtures of different ratios were found to form microemulsions with limonene. The
surfactant mixture that has SMEE3EO/sophorolipid/rhamnolipid ratio of 6/3/2
wt% produced an SPvalue of 1.8 mL oil/g surfactant with limonene. However, by

124 0 T.T. Nguyen and DA. Sabatini

K d l , Wt%l
Fig. 5.14. Interfacial tensions of limonene microemulsions with and without sophorolipid in the
formulation, measured at 25 f 1C.

adjusting this ratio to 5/5.25/5 wt%, the SP'value increased to 3.4 mL oil/surfactant.
Both systems have low optimum salinity, 1.2 and 3 wt%, respectively, which is
appropriate for use in cosmetics and pharmaceutical applications.
The results shown in Table 5.1 demonstrate the effectiveness of rhamnolipid
and sophorolipid biosurfactants in formulating microemulsions with limonene as
compared to conventional synthetic surfactants. Comparing the SP' of the first two
surfactant systems, we observe that at the same surfactant molar concentration, the
surfactant system consisting of rhamnolipid has an SP' value about two times higher
than that of the other surfactant system that does not contain rhamnolipid (8.1
versus 3.4 mL oil/g surfactant). Looking at the SP' values of the last two surfactant
systems, we observe that the formulation that has sophorolipid achieves the SP' value
of 5.2 mL oil/g surfactant at total surfactant concentration of 15.25 wt% while the
surfactant system that uses oleyl alcohol instead of sophorolipid has the SP' value of
5 mL oil/g surfactant at total surfactant concentration of 24.75 mL oil/g surfactant).
Thus, replacing the lipophilic linker oleyl alcohol with the hydrophobic biosurfactant
sophorolipid achieves similar SP' values but at lower total surfactant concentration.

Conclusions
Results presented here show that biosurfactants can be as effective as conventional
surfactants when used in mixtures that tailor the surfactant system hydrophilicity/
lipophilicity to the EACN of the oil. We found that rhamnolipid works best for
very hydrophilic oils (low EACN values) while sophorolipid works best for more
hydrophobic oils (high EACN oils). Mixtures of these and other surfactants produce

Microemulsionsof Rhamnolipidand Sophorolipid Biosurfactants0 125

Table 5.1. Optimum Solubilization Parameter (SP*) for Different Surfactant Formulations
with Limonene Microemulsions
Formulationa
Total Concentration
SP (mL Oil/g Surfactant)
AOT/JBR
(0.05/0.05 M or
0.1 M
8.1
2.2/2.0 wt%)
(4.2 wt%)
AOT/AMA/SMDNS
(0.025/0.025/0.05 M or
0.1 M
3.4
1.1/1.0/1.25 wt%)
(3.35 wt%)
SMEE3EO/SPL/JBR
(5/5.25/5 wt%)
15.25 wt%
5.2
SMEE3EO/OA/JBR
(12/7.5/5.25 wt%)
24.75 wt%
5.0
AOT, Sodium bis(2-ethyl)dihexyl sulfosuccinate; JBR, Rhamnolipid; AMA, Sodium dihexyl
sulfosuccinate; SMDNS, Sodium mono- and dimethyl naphthalene; SMEE3E0, Soy methyl ester
ethoxylate; SPL, Sophorolipid; OA, Oleyl alcohol.

high efficiency systems for a wide range of oils (from toluene (EACN of 1) to
hexadecane (EACN of 16)).Future research should develop additional biosurfactants
with a range of hydrophilicity/lipophilicityvalues to hrther expand our ability to
formulate with biosurfactant mixtures and replace petroleum-based surfactants in
pursuit of economical and sustainable products and processes.

References
Acosta, E.; H. Uchiyama; D.A. Sabatini; J.H. Hanvell. The Role of Hydrophilic Linkers.]. Sufact a m Deterg. 2002a, 5, 1 5 1- 157.
Acosta, E.; S. Tran; H. Uchiyama; D.A. Sabatini; J.H. Hanvell. Formulating Chlorinated Hydrocarbon Microemulsions Using Linker Molecules. Environ. Sci. Technol. 2002b, 36, 4618-4624.
Acosta, E.; PD. Mai; J.H. Harwell; D.A. Sabatini. Linker-Modified Microemulsions for a Variety
of Oils and Surfactants./. Sufactants Deterg. 2003, 6, 353-363.
Acosta, E.J.; T. Nguyen; A. Witthayapanyanon; J.H. Harwell; D.A. Sabatini. Linker-Based BioCompatible Microemulsions. Environ. Sci. Tecbnol. 2005,39, 1275-1282.
Bourrel, M.; and R.S. Schechter. Microemulsions and Related Systems: Formuhion, Solvenq and
Physical Properties. Surfactant Science Series 30; Marcel Dekker, 229-302, 1998.
Cameotra, S.S.; R.S. Makkar. Recent Applications of Biosurfactants as Biological and Immunological Molecules. Curr. Opin. MicrobioL 2004, 7, 262-266.
Childs, J.D.; E. Acosta; R. Knox; J.H. Hanvell; D.A. Sabatini. Improving the Extraction of Tetrachloroethylene from Soil Columns Using Surfactant Gradient Systems.]. Contam. Hydrol. 2004,
71,27-45.
Gautam, K.K.; V.K. Tyagi. Microbial Surfactants:A Review.]. Oleo Sci. 2006,55, 155-166.
Georgiou, G.; S.C.Lin; M.M. Sharma. Surface-Active Compounds from Microorganisms. Biol
Tecbnol. 1992,10, 60-65.

1260T.T. Nguyen and D.A. Sabatini

Graciaa A.; J. Lachaise; C. Cucuphat; M. Bourrel; J.L. Salager. Improving Solubilization in Microemulsions with Additives. 1. The Lipophilic Linker Role. Lungmuir 1993,9, 669-672.
Griffin, W.C. Classification of Surface-Active Agents by "HLB".J. SOC.Cosmct. Chemkts 1949, I,
31 1-326.
Helvaci, $$.; S. Pecker; G. m e m i r . Effect of Electrolytes on the Surface Behavior of Rhamnolipids R1 and R2. CoUoidr Surf B 2004,35, 225-233.
Herman, D.C.; Y. Zhang; R.M. Miller. Rhamnolipid (Biosurfactant) Effects on Cell Aggregation
and Biodegradation of Residual Hexadecane under Saturated Flow Conditions. AppL Environ.
MicrobioL 1997,63, 3622-3627.
Holmberg, K.; E. h e r b e r g . Structure of a Cosurfactant-Free Microemulsion.J Dkpersion Sci.
TecbnoL 1986, 7, 299-306.
Huh, C. Interfacial Tensions and SolubilizingAbility of a Microemulsion Phase That Coexists
with Oil and Brine. J. Colloid Interface Sci. 1979, 71,408-426.
Huibers, P.D.T.; D.O. Shah. Evidence for Synergism in Nonionic Surfactant Mixtures: Enhancement of Solubilization in Water-in-Oil Microemulsions. Langmuir 1997, 13, 5762-5765.
Kitamoto, D.; H. Isoda; T. Nakahara. Functions and Potential Applications of Glycolipid
Biosurfactants-from Energy-SavingMaterials to Gene Delivery Carries. J Biosci. Bioeng. 2002,
94, 187-201.
Komesvarakul, N.; M.D. Sander; E. Szekeres; E.J. Acosta; J.F. Faller; T.Mendik; L.B. Fisher; G.
Nicoll; DA. Sabatini; J.E Scamehorn. Microemulsions of Triglyceride-Based Oils: The Effect of
Co-Oil and Salinity on Phase Diagrams.J. Cormet. Sci. 2006,57,309-325.
Kumar, I?; K.L. Mittal. Handbook ofMicroemulsion Science and Tecbnohn. Marcel Dekker, Inc.:
New York, 1999, pp 13-44.
Kunieda, H.; K. Shinoda. Correlation between Critical Solution Phenomena and Ultralow Interfacial Tensions in a SurfactantIWaterlOil System. BufL Cbem. SOC.Jpn. 1982,55, 1777-1781.
Li, I?; A. Ghosh; RF. Wagner; S. Krill; Y.M. Joshi; A.T.M. Serajuddin. Effect of Combined Use of
Nonionic Surfactant on Formation of Oil-in-Water Microemulsions. Int. J. Pbarm. 2005,288,
27-34.
Maier, R.M.; G. Sober6n-Chiva. Prcudomanus ueruginosa Rhamnolipids: Biosynthesisand Potential Applications. Appf. MicrobioL Biotecbnol. 2000,54, 625-633.
Mulligan, C.N.; R N . Yong; B.F. Gibbs. Heavy Metal Removal from Sediments by Biosurfactants.
J. Hazard. Matez 2001,85, 111-125.
Nguyen, T.T.; N.H. Youssefi M.J. McInerney D.A. Sabatini. Rhamnolipid Biosurfactant Mixtures
for Environmental Remediation. Water Rcr. 2008,42, 1735-1743.
Nguyen, T.T.; DA. Sabatini. Formulating Alcohol-Free Microemulsions Using Rhamnolipid
Biosurfactant and Rhamnolipid Mixtures. J Sufm. Detrrg. 2008, DO1 10.1007/sl1743-0081098-y
Ozdemir, G.; U. Malayoglu. Wetting Characteristicsof Aqueous Rhamnolipids Solutions. Cofhidr
Surf B 2004,39, 1-7.
&emir. G; S. Peker; S.S. Helvaci. Effect of pH on the Surface and Interfacial Behavior of Rhamnolipids R1 and R2. CoUoidr Surf A 2004,234, 135-143.

Microemulsionsof Rharnnolipidand Sophorolipid Biosurfactants 0 127

Rodrigues, L.; I.M. Banat; J. Teixeira; R. Oliveira. Biosurfactants: Potentid Applications in Medicine. J. Antimicrob. Chemother. 2006,57, 609-61 8.
Rosen, M.J. Surfactants andIntefuciul Phenomena. 2nd ed. John Wiley & Sons, Inc., 1989. pp.
304-336393-420.
Sdager, J.L.; M. Borrel; R.S. Schechter; W.H. Wade. Mixing Rules for Optimum Phase-Behavior
Formulations of Surfactant/Oil/Water Systems. SOC.Petro. Eng. /. 1979, 19, 271-278.
Shah, V.; G.E Doncel; T Seyoum; K.M. Eaton; I. Zdenskaya; R. Hagver; A. h i m ; R. Gross.
Sophorolipids, Microbial Glycolipids with Anti-Human ImmunodeficiencyVirus and SpermImmobilizing Activities.Antimicrob. Agents Chemother. 2005, 49, 4093-4 100.
Sunwoo, C.K.; W.H. Wade. Optimal Surfactant Structures for Cosurfactant-Free Microemulsion
Systems I. C16 and C14 Guebert Alcohol Hydrophobes.J. Dispersion Sci. Technol. 1992, 13,
491-5 14.
Tongcumpou, C.; E.J. Acosta; L.B. Quencer; A.F. Joseph; J.F. Scamehorn; D.A. Sabatini; S.
Chavadej; N. Yanumet. Microemulsion Formation and Detergency with Oily Soils: I. Phase
Behavior and Interfacial Tension. J. Sufuctunts Deterg. 2006, 6 191-203.
Uchiyama, H.; E. Acosta; S. Tran; D.A. Sabatini; J.H. Hanvell. Supersolubilizationin Chlorinated
Hydrocarbon Microemulsions: Solubilization Enhancement by Lipophilic and Hydrophilic
Linkers. Ind. Eng. Chem. Res. 2000,39, 2704-2708.
Van Bogaen, I.N.A.; K. Saerens; C.D. Muynck; D. Develter; W. Soetaert; E. Vandamme. Microbial Production and Applications of Sophorolipids.Appl. Microbiol. BiotechnoL 2007, 76 23-34.
Whang, L.M.; P.W.G. Liu; C.C. Ma; S.S. Cheng. Application of Biosurfactants, Rhamnolipid,
and Surfactin, for Enhanced Biodegradation of Diesel-Contaminated Water and Soil. J. Huazrd.
Muter. 2008, 151, 155-163.
Winsor, I? Hydrotrophy, Solubilisation and Related Emulsification Processes. Trans. Furday SOC.
1948944, 376-389.
Winsor, I? Solvent Properties ofAmphiphilic Compounak Butterworth, London, 1954, pp. 1-10.
Xie, Y.; Y. Li; R. Ye. Effect of Alcohols on the Phase Behavior of Microemulsions Formed by a
Biosurfactant-Rhamnolipid. J. Dicpersion Sci. Technol. 2005,26 455-461.
Zhang, Y.; RM. Miller. Enhanced Octadecane Dispersion and Biodegradation by a Pseudomonar
Rhamnolipid Surfactant (Biosurfactant).Appl. Environ. Microbiol. 1992, 58, 3276-3282.
Zhang, Y.; R.M. Miller. Effect of Rhamnolipid (Biosurfactant) Structure on Solubilization and
Biodegradation of n-Alkanes. Appl. Environ. Microbiol. 1995, GI,2247-225 1.

Lipopeptide Biosurfactants and

Their Use in Oil Recovery
MichaelJ. Mclnerneyl, Noha YousseP, David P. Naglel
- Departmentof Botany and Microbiology, Universityof Oklahoma,770 Van Vleet Oval, Norman, OK73019;
Department of Microbiology and Molecular Genetics, Oklahoma State University, I 1 10s. Innovation Way,
Stillwater, OK 74074

Introduction
The standard of living that many nations enjoy is highly correlated to energy
consumption (Hall et al., 2003). The desire of individuals to improve their standard
of living coupled with the predicted increase in the worlds population will place large
demands on the worlds energy resources in the future. How will we meet the future
demand for more energy? Historically, we use fossil fuels as our main energy sources
with oil, coal, and natural gas supplying about 85% of worlds energy needs (Energy
Information Administration, 2006). The reliance on fossil fuels increased CO,
emissions and fostered global climate change. The use of more carbon-neutral energy
sources is desired, but difficult to achieve at the magnitude required. Even the most
optimistic projections suggest that renewable energy sources will meet less than 10%
of the worlds requirement for energy by 2030 (Energy Information Administration,
2006). The most difficult energy demand to meet is liquid transportation fuel. The
use of biofuels such as ethanol will increase substantially, but will only account for
~10%
of the demand by 2030 (Energy Information Administration, 2007). Most
likely, crude oil will continue to be the dominant source of transportation fuels until
adequate supplies of more carbon-friendly fuels can be developed.
Are sufficient crude-oil reserves available to meet future energy demands?
The answer to this question is controversial, and answers range from Global oil
production has peaked because the discovery of additional resources is unlikely to
Global oil production may increase because new resources may be discovered or
new technologies may be developed (Witze, 2007). Current technologies recover
only about one-third to one-half of the oil contained in the reservoirs. Globally, this
means that about 1 trillion barrels (0.16 Tm3) of oil were recovered to date and that
about 2 to 4 trillion barrels (0.3 - 0.6 Tm3) of oil remain entrapped in reservoirs
(Hall et al., 2003). In the United States, about 600 billion barrels (=96 Gm3) of oil
remain in U.S. reservoirs after conventional technologies reach their economic limit
(Energy Information Administration, 2007; Witze, 2007). A critical feature of U.S.
129

130 0 MJ. Mclnerney et al.

oil production is that about 27% of U.S.production is from low-production wells

where the daily oil production is cl.6 m3 of oil. About 17.5 million m3 of oil were
lost due to the plugging and abandonment of low-production wells between 1994
and 2003 (http://www.fossil.energy.gov/programs/oilgas/margina~wel~s/index.htm~).
New, cost-effective technologies must be developed if we are to preserve this critical
oil resource.
Oil production from low-production wells places economic constraints on any
technology to recover this oil. The treatment and capital costs must be low. Chemical
flooding technologies such as micellar or alkaline-surfactant-polymer flooding are
very effective in mobilizing entrapped oil, but they are marginally economical due
to the high cost of chemicals and their loss due to adsorption or other mechanisms
(Green & Willhite, 1998; Strand et al., 2003; Weihong et al., 2003). Microbial
processes do not consume large amounts of energy as do steam flooding or other
thermal recovery methods. Because microbial processes often use renewable resources
(carbohydrates) rather than petroleum to make recovery chemicals, the cost of
microbial processes should not depend on crude-oil prices as much as chemical
processes do. Microorganisms grow at exponential rates so the amount of microbial
products will increase exponentially as the cells grow. Thus, the possibility exists
to make large amounts of useful products quickly from inexpensive and renewable
resources. The critical question is whether microbial oil recovery is effective (Bryant
& Lockhart, 2002).
We discuss here the role that lipopeptide biosurfactants play in recovering the
entrapped oil. First, we discuss what these compounds are, how they are made, and
what regulates their production. We then address the technical feasibility of lipopeptide
oil recovery by analyzing laboratory and field data. Other aspects of petroleum
microbiology are covered in a recent book (Ollivier & Magot, 2005). Numerous
compilations of the results of microbial field trials are available (Bass & LappinScott, 1997; Hitzman, 1983, 1988; Lazar, 1991). Other reviews and summaries of
microbially enhanced oil recovery (MEOR) are available (Finnerty & Singer, 1983;
Islam & Gianetto, 1993; Jack, 1988; McInerney et al., 2005b, 2007; McInerney &
Sublette, 1997).

What Limits Oil Production?

First, let us briefly discuss how oil is produced. Initially, the natural pressure in the
oil reservoir pushes the oil and water into the well and up to the surface. Eventually,
the natural pressure in the reservoir dissipates, and pumps are used to bring the fluids
to the surface. This stage of oil production is called primary production (Planckaert,
2005). The next stage is secondary production, often called waterflooding, where
operators inject surfice water, seawater, or brine from a subterranean formation into
one well to push the oil to other wells called production wells. 'The primary and
secondary stages of oil production recover about one-third to one-half of the oil
initially present in the reservoir. As mentioned above, a large amount of oil remains
entrapped in the reservoir (one-half to two-thirds of the oil initially present), and is

Lipopeptide Biosurfactantsand Their Use in Oil Recovery 0 131

the target of tertiary (enhanced) oil- production processes.

Primary and secondary stages of oil production fail to recover all of the oil in a
reservoir because oil is trapped in tiny pores by capillary forces (Planckaert, 2005;
Witze, 2007). The forces that entrap oil are expressed as a ratio called the capillary
number (No) (Taber, 1969):

Na = (oil viscosity x fluid flux)/(oil-water interfacial tension).

To recover entrapped oil, we need a large increase in the capillary number (100to 1000-fold) (Reed & Healy, 1977). How can this be achieved? The fluid flux is
controlled by rock properties (permeability and porosity), and is not easily altered.
Large changes in oil viscosity are possible only for the recovery of heavy oil. Thus,
for most crude oils, the only practical solution is to reduce the interfacial tension
(IFT) between the oil and aqueous phases. We need IFT reductions of about 100- to
1OOO-fold, and we can achieve these reductions with properly designed surfactant
solutions. Lipopeptides are biologically-made surfactants (biosurfactants), some of
which reduce IFT by factors of 100-fold or more (Lin et al., 1994b; McInerney et al.,
1990). Lipopeptide biosurfactants have ideal properties for use in the oil reservoir, as
described in the following sections.

Lipopeptide Biosurfactants
Lipopeptide biosurfactants are surface-active agents produced mostly by members of
the Bacillus and Pseudomonas genera (Banat, 1995a; Bodour & Miller-Maier, 2002;
Van Dyke et al., 1991). Types of lipopeptides include surfactins, iturins, mycosubtilin,
bacillomycin, lichenysins, arthrofactin, putisolvin, and others. Besides their surfaceactive properties, lipopeptide biosurfactants have biomedical applications as
therapeutic agents. Surfactins and iturins produced by Bacillus subtilis have antifungal
and antimicrobial properties as well as some antiviral activity (Kleinkauf & H. von
Dohren, 1990; Rodrigues et al., 2006; Singh & Cameotra, 2004; von Dohren,
1995). Surfactin also inhibits fibrin-clot formation, making it a good candidate in
thrombolytic therapy (Arima et al., 1968).

Main Structural Features

Figure 6.1 shows the different types of lipopeptide biosurfactants produced by Bacillus
and Pseudomonas species. With the exception of putisolvin produced by Pseudomonas
p u t i h (where only 4 of the 12 amino-acid residues in the peptide head are cyclic),
lipopeptide biosurfactants consist of a short peptide polar head (ca. 5-12 residues)
and a hydrophobic P-hydroxy or p-amino fatty-acid tail (C13-Cl8 in length with
chains that are linear, iso-, or anteiso-branched). The amino terminus of the peptide is
amide-linked to the carboxyl group of the P-hydroxy fatty acid; the carboxyl terminus
is esterified to the P-hydroxyl (or @-amino)group of the acid, resulting in a lactonelinked cyclic peptide structure with a hydrophobic tail.

132 0 MJ. Mclnerney e t al.

Lipopeptldes produced by Bacillw subtilis

R-vH-CH,-CO-Lasn-DtDasn-Lgln-Lpro-Dasn-Ls~
Iturln A

R-~H-CH,-CO-Lglu-LIm-Dlm-Lval-LaspDleu
0
I
R-~H-CH,-CO-Lglu-L1eu-Dlcu-Lval-Lasp-Dlcu-Lval
surfactin
R=CC,o-C,,
n,iso, anteiso
R-~H-CH,-CO-Lglu-LIm-Dleu-Lval-Lasp-D1-L11~

R-~H-CH,-CO-Lasn-Dtyr-Dasn-tgln-Lgln-Lpro-Dser-~n
MycoinbtUln
Nu
1
R=C,*-C,'

Lipopeptlder produced by Bacillus licheniformis

R-~H-CH,-CO-Lglu-Lle-Dleu-Lval-La~-Dle-Li~e
LicbenYsinA
I
R'clo-cl4.
n,iso,
antelso
R-vH-CH,-CO-Lgln-Lleu-Dleu-Lval-LaspDleu-Lile Lichen~ia

R
~lo-cl,
n1so. antelso

R-~i-CH,-CO-Lglu-Lleu-Dleu-Lval-Lasn-Dle~-~~~l
S u r f s c t ~ 86
t
1
R=C I o-cy
0
n,iso, anteiso
R-~H-CH,-CO-Lgln-Lle-Dleu-Lval-Lasp-Dleu-Lval
Surfsctsnt *6
1
R=CI,-C,,
0
n,iso, anteno

Lipopeptides produced by Psendomonas rpp.

R-~H-CH,-CO-Dleu-D~-D~-Dleu-Dleu-Dser-Lleu-Ds~-Lile-Ljl~~p
Arthrofsctin
0
1
R=G

-7

R-CO-Lleu-Lglu-Llm-Lile-Lgln-Lser-Lval-Lile-La
-Llm Lval L r

R-CO-Llm-Lglu-Lleu-Lile-Lgln-Lscr-Lval-~Lleu
Lleu Ls

4 2

R-CO-Lleu-Lglu-Lleu-Lile-Lgln-Lser-Lval-Lib
-Lleu-Lile-L

'StisoMn I

R=C,

PuthM~
11

R=C,
PutLoMn 11 vsr
R=C,

Fig. 6.1 .Types of lipopeptide biosurfactants produced by Bacillus and Pseudomonas species.The lactone
ring is formed between the carboxyl group of t h e C-terminal amino acid and t h e p-hydroxyl o r p-amino
group of t h e fatty acid.

Lipopeptide Biosurfactants and Their Use in Oil Recovery 0

133

Surfactin is a cyclic lipoheptapeptide with a P-hydroxy fatty- acid tail (Arima et

al., 1968; Besson & Hourdou, 1987; Besson et al., 1992; Grangemard et al., 1999;
Jenny et al., 1991). Surfactin, which differs from lichenysin A in the amino acids
at positions 5 and 7, is a more effective antibiotic than lichenysin A (Yakimov et
al., 1995). Some surfactin-producing bacilli also synthesize the antibiotic Iturin, a
compound that is strongly membrane-active against target organisms (Besson &
Hourdou, 1987; Besson et al., 1992; Hourdou et al., 1989). Iturins fatty- acid moiety
is a C15 is0 p-amino acid; the peptide structure of Iturin A is L-Asn, D-Tyr, D-Asn,
L-Gln, L-Pro, D-Asn, L-Ser. The peptide structures of other iturins are shown in Fig.
6.1.
Two other lipopeptide biosurfactants are noteworthy. Serrawettin W2, produced
by Serratia marcescens, is implicated in the motility of this organism. The compound
comprises a pentapeptide (D-Leu, Ser, Thr, D-Phe, Ile) and a C10 hydroxy-fatty-acid
moiety (Matsuyama et al., 1992). Arthrofactin, produced by Pseudomonas sp. MIS38,
has an 11-residue cyclic peptide sequence (D-Leu, D-Asp, D-Thr, D-Leu, D-Leu,
D-Ser, Leu, D-Ser, Ile, Ile, Asp) and a C10 hydroxy fatty-acid derivative (Morikawa
et al., 1993).

Heterogeneity of Lipopeptide Structures

Crude lipopeptide preparations from a given culture often contain a mixture of
lipopeptides with different P-hydroxy fatty acids. One or more of the amino acids may
be in the D configuration. When isolated, surfactins and lichenysins are structurally
heterogeneous. The hydrophobic groups of lichenysin A are heterogeneous, being
derived from P-hydroxy acids of 12 to 15 carbons with straight, iso, or anteiso
terminal structures. The heterogeneities in surfactin mixtures were characterized by
nuclear magnetic resonance (NMR) and mass spectroscopy methods (Baumgart et al.,
1991; Oka et al., 1993; Peypoux et al., 1991, 1994). The fatty-acid moieties may be
linear or branched, and peptide chain variants exist which contain various amino-acid
substitutions in positions 2,4, and 7 ([Ile7], [Va17], [Ala4], [Leu4], [Ile4], [Ile2], and
[Val21 Surfactins) (Peypoux et al., 1991, 1994, 1999; Peypoux & Michel, 1992).

Biosynthesis
Nonribsomal biosynthesis is an important mechanism for the production of natural
products. These include a variety of compounds: glutathiones, antibiotics (p-lactams,
gramicidins, tyrocidins, surfactins, etc.), and lipopeptide biosurfactants (Desai &
Banat, 1997; Kleinkauf & H. von Dohren, 1990; von Dohren, 1995). A molecular
analysis of the genetic sequences of the nonribosomal peptide synthetases shows very
high levels of homology between organisms, even for peptide products of significantly
different properties (see von Dohren, 1995).
Nonribosomal peptide synthetases are modular proteins that catalyze the selective
binding, activation, and condensation of the amino acids in the peptide polar head
of lipopeptides via specific domains of each module according to the multiple-carrier

134 0 MJ. Mclnerney et al.

thiotemplate mechanism (Berti et al., 2007; Galli et al., 1994; Menkhaus et al., 1993;
Stein et al., 1996; Vater et al., 1997). The groups of Zuber, Marahiel, and Grandi
developed the understanding of this process, which was exploited to engineer strains
of B. subtilis that produced altered surfactins (Peypoux et al., 1999; Schneider et al.,
1998; Stachelhaus et al., 1995). Figure 6.2 represents the functional gene organization
of the surfactin synthetase operon. Each module consists of at least three domains:
(i) the adenylation or amino- acid activation domain where the amino-acid residue is
recognized and activated to its acyl adenylate derivative by a reaction with adenosine
triphosphate (ATP), (ii) the thiolation domain where the activated amino-acid ester is
covalently linked as its thioester to the enzyme-bound 4-phosphopantothiene group
(transferred to a conserved serine residue in the thiolation domain by the action of
4-phosphopantothienyl transferase, Sfp in case of surfactin synthetase), and (iii) the
condensation domain catalyzing the direct transfer of the amino-acid residue from
the 4-phosphopantothiene to an adjacent acylamino acid covalently linked to the
downstream module via peptide bond formation.
Additional domains within the module catalyze modifications to the structure
(e.g., epimerization of L-leu to D-leu). Thioesterase domains at the C-terminal end
of the module that add the C-terminal amino acid to the peptide chain (Schneider &
Marahiel, 1998) and distinct thioesterases encoded by genes present downstream of the
nonribosomal peptide synthetase operon catalyze the release of the peptide from the
enzyme complex and the lactonization or cyclization step (Bruner et al., 2002; Kohli
et al., 2001). SrfXD is an external thioesterase associated with surfactin synthetase
that stimulates the initiation of surfactin synthesis by transferring P-hydroxy fatty
acyl-CoA to glutamate to form P-hydroxymyristoyl-glutamate(Steller et al., 2004).

Regulation
Tratucriptional Regulation of Sur&tin Bioryntbesh
Biosurhctant biosynthesis has complex regulation. The genes encoding nonribosomal
peptide synthetases are organized in operons that are transcriptionally induced in

Condensation domain
Adenylation domain

Epimerization domain
ThiWSteraSe

Module

Gene

Thiolation (peptidy1 carrier protein) domain

Fig. 6.2. Surfactin synthetase gene organization. Each module consists of at least three domains:
adenylation, thiolation, and condensation. Modules 3 and 6 also have an epimerization domain for
D-leu formation. Module 7 has an additional thioesterasedomain at the 3end. Adenylation domains are
labeled by the amino acid activated by that domain.

Lipopeptide Biosurfactants and Their Use in Oil Recovery 0 135

response to nutritional stress. In the case of the B. subtilis antibiotic/biosurfactant

surfactin, the surfactin synthetase multienzyme complex is expressed most strongly
when cells are starved for carbon and nitrogen (Desai & Banat, 1997; Marahiel et al.,
1993). Such conditions are the same as those that induce sporulation. Transcription of
the s f l genes depends on a locus called c o d . The product of this locus interacts with
regulatory regions upstream of the s&4 promoter to stimulate transcription (Marahiel
et al., 1993). Another level of regulatory control of surfactin synthetase expression
occurs by the sporulation-specific locus sp00, which promotes transcription of the
surfactin synthetase operon when glucose is in excess (Marahiel et al., 1993). The s f l
is also controlled extracellularly via cell-density regulation pathways (quorum-sensing
peptides) (Schneider et al., 2002). While surfactin is induced by nitrogen limitation
(Davis et al., 1999) and repressed by glutamate (Marahiel et al., 1993), lichenysin A
formation by B. lichenijormis strain BAS-50 is stimulated several-fold when nitrogen
sources such as glutamate or asparagine are provided (Yakimov et al., 1996). The
Bacillus mojavenesis JF-2 biosurfactant is produced maximally by cells in logarithmic
phase, apparently under growth-rate control (Lin, 1996; Lin et al., 1994a). When
cells enter stationary phase, they stop biosurfactant synthesis and take the compound
back up (Lin et al., 1993).

Common TransrriptionRegulation with Other Starvation-InducedPhenomena

As mentioned, lipopeptide antibiotic biosynthesis is regulated by mechanisms shared
with sporulation, competence development, and other starvation-induced activities.
The regulation of surfactin biosynthesis and competence development (ability to
uptake external DNA) is closely related in B. subtilis (Marahiel et al., 1993). The comS
gene involved in competence development is located within the slfA gene complex
(DSouza et al., 1994). Both s f l and corns are under transcriptional control of the
two-component system comPA, and directly through the C o d peptide factor, both
ofwhich are involved in quorum sensing (Hamoen et al., 2003; Stein, 2005). Surfactin
is necessary but not sufficient for swarming motility in undomesticated B. subtilis
strains (Kearns et al., 2004; Kearns & Losick, 2003; Kinsinger et al., 2003). Surfactin
production and biofilm formation are interdependent in natural environments.
Surfactin enabled a natural B. subtilis isolate A1/3 to form biofilms (Hofemeister et
al., 2004). Surfactin also inhibited biofilm formation in other bacteria (Bais et al.,
2004).
Both surfactin synthesis and biofilm formation are regulated by common
transcription factors (e.g., SpoOA, AbrB) (Branda et al., 2001; Hamon & Lazazzera,
2001; Stein, 2005). Fruiting body formation (a highly ordered surface structure
intertwined tightly with sporulation) in B. subtilis involves two exopolysaccharide
production genes and s j j (the gene encoding for phosphopantethiene transferase
essential for the activity of the surfactin synthetase multienzyme complex) (Branda
et al., 2001; Stein, 2005). Analogous to the need for surface-active molecules for
hyphae development in fungi, fruiting body development in B. subtilis might require

136 0 MJ. Mclnerneyet al.

the activity of lipopeptide biosurfactants (Stein, 2005). B. subtilis sporulation is also

regulated by the spoOA gene (one of the srfA transcriptional hctors) (Stein, 2005).

Lipopeptide Surface and Interfacial Activities

As noted above, a large decrease in the IFT between the oil and aqueous phases is
needed to recover entrapped oil (McInerney et al., 2005b). Lipopeptides are ideal
oil-recovery agents because they partition at the oil-rock interface, and promote the
mobilization of oil from the rock. The most common biosurfactants used in MEOR
are lipopeptides produced by the Bacillus species and some Pseudomonas species,
glycolipids (rhamnolipids) produced by the Pseudomonas species, and trehalose lipids
produced by the Rhodococcw species (Banat, 1995a, b; Bodour & Miller-Maier, 2002;
Youssef et al., 2004). Lipopeptides and rhamnolipid biosurfactants lower IFT between
the hydrocarbon (crude oil or pure hydrocarbons) and aqueous phases to values of 0.1
mN/m or lower (Lin et al., 1994b; Maier & Soberon-Chava, 2000; McInerney et
al., 1990; Nguyen et al., 2008; Wang et al., 2007) (Table 6.1). These low interfacial
tension (IFT) values are sufficient to lower the capillary number and to mobilize
significant amounts of oil (see below).
The critical micelle concentrations (CMCs) of lipopeptides vary from 8-50
mg/L (Table 6.1), and are orders of magnitude lower than synthetic surfactants. This
indicates that lipopeptides are effective at much lower concentrations than synthetic
surfactants (Georgiou et a]., 1992; McInerney et al., 2005b; Youssef et al., 2004). The
lipopeptide made by the B. mojavemis strain JF-2 is similar in structure to surfactin
(Fig. 6.1). A partially-purified preparation of the JF-2 lipopeptide is a mixture of
compounds with linear or branched fatty-acid side chains (Lin et a]., 1994b; Youssef
et al., 2005), and have remarkable surhce-active properties. At concentrations of 10
mg/L, the JF-2 lipopeptide lowers the surhce tension of aqueous solutions to 27
mN/m and the IFT between water and decane to c 0.01 mN/m (Lin et al., 1994b;
McInerney et al., 1990). The CMC is 10 pM; the pH optimum for interfacial activity
is 6.0. Ten-fold higher IFT values are obtained one pH unit lower or higher than
the optimum. Activity is highest in solutions containing 50 g/L of NaCl (Lin et
al., 1994b). Under optimized conditions, the JF-2 lipopeptide is one of the most
effective biosurfactants known, decreasing the oil/water IFT by a factor of 500. The
B. licbeni$rmis strain BAS-50 biosurhctant has surhce-active properties very similar
to those of the JF-2 lipopeptide (surface tensions lowered to 28 mN/m; IFT values
not reported) (Yakimov et al., 1995).
Surfactin is not as effective as the JF-2 lipopeptide. The CMC is 2-fold higher
(24 pM), and the IFT values are at least ten-times higher than achieved by the JF-2
lipopeptide (Desai & Banat, 1997). Surhctin produced by B. nrbtilis was more effective
than sodium lauryl sulhte (an anionic chemical surhctant) in changing the rock
wettability from an oil-wet to a water-wet system (Johnson et al., 2007). Serrawettin
W2 lowers the surfice tension of water to ca. 34 mN/m (Matsuyama et al., 1992).
Arthrohctin lowers the surhce tension to 27 mN/m (Morikawa et a]., 1993).

Table 6.1. Critical Micelle Concentrationsand Interfacial Tensions of Various Lipopeptide Biosurfactants
Lipopeptide
Surfactins

Microorganism
Bacillus subtilis

Strain
OK6105

8. subtilis subsp. subtilis

T89-42
ROGG-2
T89-3
JF-2

8. subrilis subsp. spizzenii

8. mojavenesis

IFT(mN/m)
2 (dodecane)
1 (hexadecane)
0.63 (toluene)
2.1 7 (toluene)
0.3 (toluene)
0.001 (decane)

17.4 mg/L 0.66 (toluene)

12mgA
NA
10 mgA
0.36 (hexadecane)
B.subtilis
20 mgA
14.9 (dodecane)
lturins
Arthrofactin Pseudomonas sp.
MIS38
lOpM
NA
P. putida
267
50mg/L
NA
Putisolvin
a Word in the parentheses is the hydrocarbon used in the measurement.
Abbreviations: NA, not available; CMC, critical micelle concentration; In, interfacial tension.
Lichenysins

8. licheniformis

ROB-2
BAS50
BL86

CMC
10-20
mg/L
10 mg/L
10 mg/L
10 mg/L
7.8 mg/L

References
(Delwet al., 1999)
(Rosenberg& Ron, 1999)
(Youssef et al, 2007a)
(Youssef et al., 2007a)
(Youssefet al., 2007a)
(Knapp et al., 2002; Maudgalya
et al., 2004; Youssef et al., 2007a)
(Youssef et al., 2007a)
(Yakimov et al, 1995)
(Horowitz & Griffin, 1991)
(Deleu et al., 1999)
(Morikawaet al., 2000)
(Tran, 2007)

138 0 MJ. Mclnerneyet al.

Hydrophobic and Hydrophilic Regions and Surface Activity

Infante and Moses (1994) studied the surface activity of short linear peptides of
alternating amino acids-Leu and Gln. The peptides (two to five residues in length)
were highly surface-active, lowering the surface tension of aqueous solutions from
72 - e40 mN/m and the IFT between oil and water from 45.2 to ca. 20 mN/m, but
they did not form micelles, and did not stabilize oil emulsions. Hexadecanoyl esters
of the same peptides lowered surface tension values to e40 mN/m and decreased the
oil/water IFT to 1.5 - 4 mN/m. The hexadecanoyl ester peptides had CMCs of ca.
0.05 mM, and were highly effective at stabilizing oil/water emulsions. Thus, both
the hydrophobic and hydrophilic portions of lipopeptide biosurfactants contribute
to the surface activity. Nuclear magnetic resonance (NMR) studies showed that
linear lipopeptides interact strongly with phospholipid phases, which stabilizes the
flexibility of the peptide structure and is consistent with the role for the peptide in
surface activity (Macquaire et al., 1992).

Effect of Temperature,pH, and Salinity

Lipopeptides are active over a wide range of environmental conditions often present
in oil reservoirs, temperatures up to 100"C, pH from 6 1 0 , and salt concentration
up to 70 g/L (Cameotra & Makkar, 1998; Jenneman et al., 1983; Joshi et al., 2008b;
Makkar & Cameotra, 1997,1998; McInerney et al., 1990). The lipopeptide produced
by B. subtilis MTCC 1427 is active at a pH range of 3-1 1 and at temperatures up
to 100C. B. subtilis 20B, B. subtilis R1, B. lichentj3rmis K51, and B. strain HS3
produce lipopeptide biosurfactants from inexpensive carbon sources such as molasses
and whey (Joshi et al., 2008ab). These lipopeptides are active at temperatures up to
80C and with NaCl concentrations up to 50 g/L.

Improving Biosurfactant Activity

Improvements in biosurfactant activity can be obtained by altering cultivation
conditions or by genetic manipulation (for detailed review see Abu-Ruwaida et al.,
1991; Bordoloi & Konwar, 2007; Das & Mukherjee, 2005; Davis et al., 1999; Joshi
et al., 2008a; Makkar & Cameotra, 1998; Mukherjee et al., 2006; Schaller et al.,
2004) (Table 6.2).

Changes to the Amino-acid Composition

The three-dimensional structure of surfactin (Bonmatin et al., 1994) shows that the
carboxylic groups of glutamate and aspartate form a hydrophilic domain, and the
nonpolar residues in position 4 and, to a lesser extent, in positions 2 and 7 form
a hydrophobic domain with the lipid tail. The presence of these two domains is
important for surface activity (Bonmatin et al., 1995; Grangemard et al., 1999;
Peypoux et al., 1994, 1999; Schneider et al., 1998; Stachelhaus et al., 1995; Thimon
et al., 1994). The substitution of valine by isoleucine in position 4 decreases the
CMC by 2-fold and increases the surface activity, possibly due to the expansion of

Lipopeptide Biosurfactantsand Their Use in Oil Recovery 0 139

the hydrophobic domain (Bonmatin et al., 1995) (Table 6.2).Morikawa et al. (1993)
argued that the acidic residues- Asp and Glu-are most important for activity. But,
the monoanionic biosurfactant, lichenysin A, which has asparagine in position 5 , has
better surface activity than the dianionic biosurfactant, surfactin, which has aspartate
in position 5 (Grangemard et al., 1999; Yakimov et al., 1995).

Changes in Fatty-acid Composition

The fatty-acid moiety is also important for surfactant activity. The CMC depends
on the chain length of the fatty acid (Morikawa et al., 2000). Yakimov et al. (1996)
found that with lichenysin A, a monoanionic heptapeptide (Glu: Asn: Val: Leu: Ile;
1:1:1:3:1) (Fig. &I), surface activity decreased with an increase in the percentage
of branched-chain fatty acids, and surface activity increased with an increase in the
percentage of straight-chain 3-hydroxy-tetradecanoate (n-30H-Cl4). With the
dianionic heptapeptide, surfactin (Glu: Asp: Val: Leu: Ile; 1: 1:1:3: 1) (Fig. 6. l),
biosurfactant activity increased with an increase in the percentage mass of 3-hydroxy
is0 even-numbered fatty acids (iso-3-OH-Cl4) (Youssef et al., 2005).
Table 6.2. The Effect of Alterations t o the Lipopeptide Structure on
Biosurfactant Activity
Alteration
Replacement of leucine with
valine; [Val7]-surfactinvariant
Replacement of the valine
in position 4 with leucine
in position 7 or isoleucine:
[Leu4]- or [Ile4]-surfactin
variants
[lle4, 71-surfactin variant
[Ile2,4,7]-surfactin variant

Effect on activity
Hydrophobicity decreased and
CMC increased
Improved surfactant activity

Reference
(Peypoux & Michel, 1992)

Hydrophobicity increased and

CMC decreased (220 to 90
pM); affinity for calcium ions
increased
Hemolytic activity lowered:
[Va17]-, [Phe7]-, [Orn7]-, and
surface activitv not tested
ICvs71-surfactinvariants
Lichenysinwith variable fattyIncreasingthe proportion of
acid composition.
branched-chainfatty acids
decreased surface activity.
Normal-chain C14 fatty acid
imDortant for surface activitv.
Surfactin with variable fattyOil-displacement activity
acid composition.
against crude oil increased
with an increase in the ratio of
is0 to normal even-numbered
fatty acids
Abbreviation: CMC, critical micelle concentration.

(Grangemard et al., 1997)

(Bonmatinet al., 1995)

(Stachelhauset al., 1995)

(Yakimov et al., 1996)

(Youssefet al., 2005)

140 0 MJ. Mclnerney et al.

The percentage of branched-chain htty acids in the biosurhctant can be
manipulated by the addition of branched-chain amino acids (L-valine, L-isoleucine,
and L-leucine) to the growth medium (Akpa et al., 2001; Besson & Hourdou, 1987;
Besson et al., 1992; Desai & Banat, 1997; Hourdou et d., 1988, 1989; Kaneda,
1966). The addition of valine increased the ratio of iso to normal even-numbered
htty acids by 2.8-fold and surface activity by 3.2-fold compared to controls with no
amino-acid addition (Youssef et al., 2005). The addition of leucine decreased the ratio
of anteiso to ho odd-numbered fatty acids by 15-fold, and increased surfice activity by
2-fold. The addition of isoleucine increased this ratio by 2.7-fold, but did not affect
surface activity.

Formulating Biosut$actant M h m s
An optimal balance between the hydrophilic-hydrophobic interactions between the
displacing fluid and the crude oil is critical to obtain very low IFTs (<0.1 mN/m).
Table 6.3 shows several approaches to alter hydrophilic-hydrophobic interactions to
achieve low IFTs. One approach is to mix biosurfictants, either lipopeptides with
different fatty-acid compositions or lipopeptides with rhamnolipid. In both cases, the
mixtures became more hydrophilic, and had IFT values <0.1 mN/m against toluene
(Youssef et al., 2007a). Low IFT values against hydrophobic hydrocarbons (decane or
hexadecane) require the addition of a hydrophobic, synthetic surhctant.
Lastly, the salt concentration and the pH of the displacement fluid can be
manipulated to enhance interficial activity (Nguyen et al., 2008).

Oil Recovery by Lipopeptides

Biosurfactant Flooding Experiments
The lipopeptide-produced by the Bacillus strain JF-2, which was reclassified as a
strain of B. mojavensis (Folmsbee et al., 2006)-was partially purified and used in flow
experiments with sand-packed columns to test its effectivenessin mobilizing entrapped
crude oil. The injection of the JF-2 biosurfictant in 50 g/L of NaCl recovered little
entrapped oil even when very high lipopeptide concentrations (12.3 g/L) were used
(Knapp et al., 2002; Maudgalya et al., 2004,2005; McInerney et d.,2005a). An oil
bank formed in the sand-packed columns when the JF-2 biosurfictant solution was
added, but quickly dissipated before the oil bank reached the effluent end of the sand
pack. The addition ofa viscosifyingagent, 1 g/L of partially hydrolyzed polyacrylamide
(PHPA), and a co-surfactant (10 mM of 2,3 butanediol) to the biosurhctant solution
significantly increased oil recovery even with low biosurfactant concentrations (43
mg/L) (Table 6.4). Oil recovery was proportional to the concentration of the JF-2
biosurhctant in the presence ofPHPA and 2,3-butanediol. Up to 83% of the entrapped
oil in the sand pack was recovered when 920 mg/L of the JF-2 biosurfactant was used
(Knapp et al., 2002; Maudgalya et al., 2004,2005). Significant oil recoveries occurred
from Berea sandstone cores with as little as 40 mg/L of the JF-2 biosurfactant as long
as PHPA and butanediol were present (Maudgalya et al., 2005). Oil-recovery yield
was 2.2 mL of crude oil per mg of biosurhctant (Youssef et al., 2007b).

Lipopeptide Biosurfactants andTheir Use in Oil Recovery 0 141

Table 6.3. Formulating Effective Lipopeptidedurfactant Mixtures to Minimize Interfacial

Tension
~~

Mixture components
Lipopeptides with different, fatty-acid
comoositions

Lipopeptides and hydrophilic biosurfactant

(rhamnolipid)
Lipopeptides with hydrophobic surfactant

Effect on IFT
Low IFTagainst a hydrophilic hydrocarbon
(toluene)
Low IFT against toluene
Low IFT against hydrophobic hydrocarbons
(e.g., hexane, decane, or hexadecane)

Data from Youssef et al. (2007a).

Abbreviation: IFT, interfacial tension.

Others also found that that low lipopeptide biosurfactant concentrations recover
residual oil from model porous systems at elevated temperatures and salinities (Table
6.5). One pore volume of 1 mg/L solution of a lipopeptide purified from B. subtilis
MTCC 1427 cultures recovered 56% of the residual kerosene from sand-packed
columns (Makkar & Cameotra, 1998). The lipopeptide produced from molassesgrown Bacillus cells lowered surface tension to 29 mN/m, and recovered 34 to 39%
ofentrapped oil from sand-packed columns (Makkar & Cameotra, 1997) (Table 6.5).
Two thermophilic B. subtilis strains, DM-03 and DM-04, produced a lipopeptide
biosurfactant when grown with cheap nutrients (potato peel) that lowered surfice
tension to 32-34 mN/m and recovered 56-60% of entrapped oil from sand-packed
columns (Das & Mukherjee, 2007). Lipopeptide biosurfactants produced by B. subtilis
and B. licbeniformis strains from molasses- and whey-grown cultures also mobilized
entrapped oil (Table 6.5) (Joshi et al., 2008ab).
In situ growth of the B. mojavensis strain JF-2 recovered entrapped oil from sandpacked and crushed-limestone columns (Adkins et al., 1992; McInerney et al., 1985)
(Table 6.5). In situ lipopeptide production by other Bacillus strains also recovered
entrapped oil from model porous systems (Chang, 1987; Thomas et al., 1993;
Yakimov et al., 1997; Zekri et al., 1999) (Table 6.5).
Combining biosurfactant production with the production of other useful
products such as alcohols, gases, and acids is an effective strategy for oil recovery
(Bryant & Burchfield, 1989; Bryant et a]., 1988; Bryant & Douglas, 1988). A
consortium with a biosurfactant producer (B. licbeniformis), an acid, gas, and solventproducer (Clostridium sp.), and a facultatively anaerobic bacterium recovered 60% of
the entrapped oil from etched-glass micromodels and 28% of the entrapped oil from
Berea sandstone cores. A consortium with biosurfactant, solvent, and acid producers
isolated from oil-reservoir fluids recovered 10 to 60% of the entrapped oil from sandpacked columns (Sugihardjo & Pratomo, 1999).

Field Applications of Biosurfactant-mediated Oil Recovery

Well Treatments
A recent, well-controlled field experiment conclusively showed that large amounts of
a lipopeptide biosurfactant can be made in situ and that the inoculation of the wells

142 0 MJ. Mclnerney et al.

Table 6.4. Oil Recovery by Low Concentrations of Lipopeptide Biosurfactants in the

Presence of an Alcohol and a Viscosifying Agenr
Biosurfactant
concentration
(mg/L)
43
43
43

Residual-oil
saturation
(96)
12f1.6
192 1.6
19f3

Oil volume
recovered
(mL)
2 f 0.2
3 0.1
4 f 0.1

Percentage
of oil
recovery
13f2
12f2
17f3'

Additions to spent mediumb

None
2-3-Butanediol(lOmM)
Partially hydrolyzed
polyacrylamide(1 g/L)
22f1
5.5f1.4
22f0.1'
2,-3-Butandiol(lOmM)
43
and partially hydrolyzed
polyacrylamide(1 g/L)
a Data from (Knapp et al., 2002; Maudgalya et al., 2004).
Bacillus mojavensis strain JF-2 was grown aerobically, cells were removed, and 10 mM
of 2-3-butanediol and 1 g/L of partially hydrolyzed polyacrylamidewere added as
indicated. Sterile medium with 10 mM of 2-3-butanediol and 1g/L of partially hydrolyzed
polyacrylamiderecovered <0.2 mL of entrapped oil (<1.2% residual=oil recovery).
'Means with the same letter are significantly different from other data.

Table 6.5. Oil Recovery By Lipopeptides From Model Porous Systems

Microorganism
Bacillus subtilis

Type of experiment
Core flood

6.subtilis

Sand-packed
columns
with sodium
pyrophosphate
Sand-packed
columns with
kerosene

Percentage of residualhydrocarbon recovery

14

References
(Abhati et al.,
20031

6. subtilis strain
MTCC1427

6. subtilis strains DM03,

DM04 (thermophiles)
6. subtilis 20B
6. licheniformis K51
6. subtilis R1
Bacillus strain HS3
6. mojavensis strain JF-2

Sand-packed
columns with
crude oil
Sand-packed
columns
Sand-packed
columns

Sand-packed
columns and
Berea-sandstone
cores flooded
Crushed-limestone
columns

35

(Chang, 1987)

56 (with 100 mL
of a 1 mg/L of
crude- biosurfactant
preparation)
34-39

(Makkar &
Cameotra, 1998)

56-60

(Das & Mukherjee,

2007)
(Joshi et al.,
2008ab)

25-33

10-83 (depending
on biosurfactant
concentration)
27

(Makkar &
Cameotra, 1997)

(Maudgalyaet
al., 2004,2005;
Mclnerney et al.,
2005a)
(Adkinset al.,
1992)

Lipopeptide Biosurfactants and Their Use in Oil Recovery 0 143

with lipopeptide-producing strains is feasible (Simpson et al., 2007; Youssef et al.,

2007b) (Table 6.6). In situ biosurfactant production stimulated oil production from
one of the treated wells (Simpson et al., 2007; Youssef et al., 2007b). The amount
of additional oil produced from this well was close to that predicted by correlations
between biosurfactant concentration and oil recovery generated in laboratory
experiments.
Consistent with the laboratory findingsare that (i) the combination ofbiosurfactant
production with acid, gas, and solvent production is beneficial for oil recovery and (ii)
several studies showed that treating wells with mixtures of microorganisms-some of
which make biosurfactants, others which make acids, gas, and solvents- stimulates
oil production (Table 6.6). Oil production increased from 30 to 100% for up to 18
months (Table 6.6).

Wat&oding
Biosurfactant-enhanced water flooding differs from well treatments in that the goal
of biosurfactant-enhanced water flooding is to mobilize entrapped oil deep within
the reservoir rather than to increase the production of a single well. Nutrients with
or without a bacterial inoculum are injected into the reservoir to stimulate microbial
activity and biosurfactant production throughout the reservoir (Fig. 6.3). While
laboratory studies indicate that this approach is very effective, only two field tests of
biosurfactant-enhanced waterflooding were performed (Bryant & Burchfield, 1991;
Bryant et al., 1990, 1993, 1994). In both tests, the investigators used a mixed culture
with the B. mojavensis strain JF-2 and molasses as the nutrient. After inoculation,
the molasses was periodically added to each injection well into the first test. The oilproduction rate of the field increased by 14%, and the ratio of water-to-oil produced
from the field decreased after the microbial treatment (Bryant & Burchfield, 1991;
Bryant et al., 1990). The only evidence for in situ biosurfictant production was the
lowering of the surfice tension of produced fluids 6 weeks after inoculation. The
presence ofa lipopeptide biosurfactant was not confirmed by hrther chemical analyses.
In the second test, the inoculum was added, and then molasses was continuously
injected along with the injected brine (Bryant et al., 1993, 1994). The oil-production
rate increased in the second field by about 19% for 3 years. In each case, the change
in oil production and the total incremental oil recoveries, 88 and 400 m3in 2 years,
were low, which raised skepticism as to whether the microbial treatment improved oil
recovery above pretreatment levels (Bryant & Lockhart, 2002).

Conclusions
We must develop new, cost-effective technologies to recover the large amounts of
entrapped oil that exist in oil reservoirs to meet the growing demand for energy.
Laboratory studies show that lipopeptide biosurfactants are promising oil-recovery
agents. Lipopeptides have CMCs that are much lower than synthetic surfactants,
they generate low IFT between the hydrocarbon and aqueous phases (cO.1 mN/m),
and they remain active over a wide range of environmental conditions. Genetic and

144 0 MJ. Mclnerney et al.

Table 6.6. Stimulation of Oil Production by Biosurfactant Production and

Other Microbial Products
lnoculum
Surfactant, alcohol,
and polymer
producers

Nutrients and
shut-in time
Sugar, molasses, yeast
extract, PO; and NO;
and a 3-week shut-in
Deriod
4% molasses

Results
Oil production
increased from 2.4
to 6.3 m3per day

References
(Zaijic, 1987)

Oil production
increased by 7996

(Bryant et al., 1988;

Lazar, 1991)

Unspecified nutrients
and a 40-64 day shutin period

Oil production
increased by 2.3
to 3.4 mVday for
8 to 18 months in
the two treated
wells; 1140 m3of

(Zhang et al., 1993)

Two Bcrcillus
strains and one
pseudomonad

Waste fluids from a

fermentation industry
and a 7-day shut-in
period

Bocillus sp. and

clostridia

Unspecified nutrients
and a 7-day shut-in
period

B. licheniformis and
B. subtilis subsp.
subtilis spizizenii

Glucose-nitrate-trace
metals and a 4-day
shut-in period

Oil production
increased 60% (1.9
mVday) in one
well; no change in
another well
Incremental
oil production
increased by 56 to
137 m3in two wells;
no change in three
other wells
Oil production
increased 30 to
100%in two wells
for 100 days; 380 m3
of incrementaloil

Mixed culture of
acid, gas, solvent
and biosurfactant
Droducers
Pseudomonos
oeruginoso,
Xonthomonas
compestris,and
Bacillus licheniformis

incremental oil

(He et al., 2006)

(Buciak et al., 1994)

(Simpson et al.,
2007)

cultural approaches exist to improve lipopeptide surface activity. Flow-displacement

experiments show that low concentrations of lipopeptide biosurfactants mobilize
entrapped oil if a polymer and an alcohol are also present in the displacement
fluid. Nutrient injection stimulates the in situ growth of the lipopeptide-producing
microorganisms in sandstone cores, which mobilizes entrapped oil. However, oil
recoveries are low and variable in a range of 1040%.
While laboratory studies argue for the success of lipopeptide biosurfactants for oil
recovery, little information is available to know whether the success in the laboratory
can be translated to the field. Generating large amounts of lipopeptide biosurfactants
when nutrients and microorganisms (about 0.65 m3>are injected into individual wells

Lipopeptide Biosurfactants and Their Use in Oil Recovery 0

145

Biosurfactant
Acid
Solvent
Fig. 6.3. Biosurfactant-mediatedoil recovery. Nutrientsand/or biosurfactant-producing microorganisms
are injected into the oil reservoir through the injection well (I) to stimulate the production of
biosurfactants. Micelles with oil form an oil bank (grey-shaded region), which is pushed to production
wells (P).

is possible. Whether large amounts of lipopeptides can be made throughout an oil

reservoir is still an unanswered question and one that must be answered before a largescale implementation of biosurfactant-mediated oil recovery becomes a reality.

Acknowledgments
We thank the U.S.Department of Energy (contracts DE-FC26-02NT15321 and
DE-FC26-04NT15522) for support of our work.

References
Abhati, N.; R. Roostaazad; F. Ghadiri. Biosurfactant production in MEOR for improvement of
Irans oil reservoirs production experimental approach. SPE 84907. SPE International Improved
Oil Recovery Conference in Asia Pacgc; Kuala Lumpur, Malaysia, 2003; Society of Petroleum
Engineers, Richardson, TX.
Abu-Ruwaida, A.S.; I. M. Banat; S. Haditiro; A. Salem; M. Kadri. Isolation of biosurfactant-producing bacteria: Product characterization and evaluation. Acta Biotechnol. 1991, 11, 31 5-324.
Adkins, J.I?; RS. Tanner; E.O. Udegbunam; M.J. McInerney; R.M. Knapp. Microbially enhanced
oil recovery from unconsolidated limestone cores. Geomicrobiol.1.1992, 10, 77-86.
Akpa, E.J.P.; B. Wathelet; M. Paquot; R. Fuchs; H. Buzikiewicz; I? Thornart. Influence of culture
conditions on lipopeptide production by Bacillw subtilis. Appl. Microbiol. Biotechnol. 2001,
91-93, 551-561.

146 0 MJ. Mclnerney et al.

Arima, K.; A. Kakinuma; G. Tamura. Surfactin, a crystaline peptidelipid surfactant produced by
Bacillus subtilk Isolation, characterization, and its inhibition of fibrin clot formation. Biochem.
Biophys. Rcs. Commun. 1%8,31,488-497.
Bais, H.P.; R Fall; J.M. Vivanco. Biocontrol of Bacilllus zubrik against infection of Arabihpsis
mots by Psrudomonac syringa is facilitated by biofilm formation and surfactin production. Plant
PhysioL 2004,134,307-319.

Banat, I.M. Biosurfactants production and possible uses in microbial enhanced oil recovery and
oil pollution remediation: A review. Bioes. Tccbnol. 1995% 51, 1-12.
Banat, I.M. Characterization of biosurfaaants and their use in pollution removalState of the art
(Review). Acta Biotechnol. 1995b,15, 251-267.
Bass, C.J.; H.M. Lappin-Scott. The bad guys and the good guys in petroleum microbiology.
Oilfield Rev,,Spring 1997,17-25.
Baumgart, E; B. Kluge; C. Ullrich; J. Vater; D. Ziessow. Identification of amino acid substitutions in the lipopeptide surfactin using 2D NMR spectroscopy.Biocbem. Biophys. Rcs. Commun.

1991,177,998-1005.
Berti, A.D.; N.J. Greve; Q.H. Christensen; M.G. Thomas. Identification of a biosynthetic gene
duster and the six associated lipopeptides involved in swarming motility of Psnuiomonac syringae
pv. tomato DC3000. J. Bartrriol. 2007, 189,6312-6323.
Besson, F.; M. Hourdou. EBect of amino acids on the biosynthesis of beta amino acids, constituents of bacillomycin F.J. Antibiot. 1987,41,221-223.
Besson, E; I. Tenoux; M. Hourdou; G. Michel. Synthesis of P-hydroxy fatty acids and P-amino
acid fatty acids by strains of B a c i h subtifis producing iturinic antibiotics. Biochim. Biophys. Acta

1992,1123,51-58.
Bodour, AA.; R M . Miller-Maier. Biosurfactants: types, screening methods, and applications.
Encychpoedia of EnvironmentalMimbiohgy, G. Binon,Ed.; Wdey-Interscience: New York, NY,
2002; pp. 750-770.
Bonmatin, J.M.; M. Genest; H. Labbe; M. Ptak. Solution three-dimensional structure of surfactin: a cyclic lipopeptide studied by 'H-NMR, distance geometry, and molecular dynamics.
Biopobmen 1994,34975-986.
Bonmatin, J.M.; H. Labbe; I. Grangemard; E Peypoux; R. Maget-Dana; M. Pt&; G. Michel.
Production, isolation and characterization of [Led]- and [IlcQI-surfactinsfrom Bacillus subtilis.
Lm.Pepride Sci. 1995,2,41-47.
Bordoloi, N.K.; B.K. Konwar. Microbial surfactant-enhanced mineral oil recovery under laboratory conditions. Colloid Su$, B. Biointerfaces 2007,63,73-82.
Branda, S.S.;J.E. Gonzala-Pastor; S. Ben-Yehuda; R. Losick; R Kolter. Fruiting body formation
by Bacillus subtik. Proc. NatL h a d Sci. USA 2001,98,11621-1 1626.
Bruner, S.; T Weber; R Kohli; D. Schwaner; M. Marahiel; C. Walsh; M. Stubbs. Structural basis
for the cyclization of the lipopeptide antibiotic surfactin by the thioesterase domain SrtTE.
Smcrtlr~~2002,10,301-310.

Bryant, RS.; T.E. Burchfield. Review of microbial technology for improving oil recovery. Soc. Pet.
Eng. Reservoir Ens 1989,4,151-154.
Bryant, RS.; T.E. Burchfield. Microbial enhanced waterflooding: a pilot study. Drv. Pet. Sci.

Lipopeptide Biosurfactants and Their Use in Oil Recovery 0 147

1991,31,399-420.

Bryant, RS.; T.E. Burchfield; D.M. Dennis; D.O. Hinman. Microbial-enhanced waterflooding:
Mink Unit project. SOC.Pet. Eng. Reservoir Eng. 1990,5,9-13.
Bryant, RS.; T.E. Burchfield; M.D. Dennis; D.O. Himman. Microbial enhanced waterflooding
Mink Unit Project. SPE 17341. SPE/DOE Enhanced oilRecovmy Symposium 1988; Society of
Petroleum Engineers, Richardson, Tx
Bryant, RS.; J. Douglas. Evaluation of microbial systems in porous media for E O R SOC.Pet. Eng.
Reservoir Eng. 1988,3,489-495.
Bryant, S.L.; TI?Lockhart. Reservoir engineering analysis of microbial enhanced oil recovery. SOC.
Pet. Eng. Reservoir EvaL Eng. 2002,5, 365-374.
Bryant, RS.; A.K. Stepp; KM. Bertus; T.E. Burchfield; M. Dennis. Microbial-enhanced waterflooding field pilots. Dcv. Pet. Sci. 1993,39,289-306.
Bryant, RS.; A.K. Stepp; K.M. Bertus; T.E. Burchfield; M. Dennis. Microbial enhanced waterflooding field tests. Be 9th Symposium on improved Oil Recourry, Tulsa, OK, USA, 1994; Society
of Petroleum Engineers (SPE): Richardson, TX, USA, pp. 159-170.
Buciak, J.M.; A.E. Vazqua; RJ. Frydman; J.A. Mediavilla; RS. Bryant. Enhanced oil recovery
by means of microorganisms: pilot test. Proceedings of the 3tdLatin American and Caribbean Pet r o h m Engineering Confirenre,Buenos Ares, Argentina, 1994; Society of Petroleum Engineers
(SPE): Richardson, TX, USA, pp. 279-287.
Cameotra, S.S.; RS. Malckar. Synthesis of biosurfactants in extreme conditions. AppL Microbiol.
Biotecbnol. 1998,50, 520-529.
Chang, Y. Preliminary studies assessing sodium pyrophosphate effects on microbially mediated oil
recovery. Ann. N . Y A c d . Sci. 1987,506,296-307.
DSouza, C.; M.M. Nakano; I? Zuber. Identification of comS, a gene of the s@ operon that
regulates the establishment of genetic competence in Baiflw subtilis. Prof. NatL A c d . Sci. USA
1994,91,9397-940 1.

Das, K.; A.K. Mukherjee. Characterization of biochemical properties and biological activities of
biosurfactants produced by Pscudomonac amginosa mucoid and non-mucoid strains isolated
from hydrocarbon-contaminated soil samples. Appl. MicrobioL BiotecbnoL 2005,69, 192-1 99.
Das, K.; A.K. Mukherjee. Comparison of lipopeptide biosurfactants production by Bacillw subtilis
strains in submerged and solid state fermentation systems using a cheap carbon source: Some
industrial applications of biosurfactants. Proccss Biocbem. 2007,42, 1191-1 199.
Davis, D.A.; H.C. Lynch; J. Varley. The production of surfactin in batch culture by Baillw
subtilis ATCC 21332 is strongly influenced by the conditions of nitrogen metabolism. Enyme
Microb. Techno1 1999,25, 322-329.
Deleu, M.; H. Radindralambo; Y. Popineau; I? Jacques; I? Thonart; M. Paquot. Interfacial and
emulsiFying properties of lipopeptides from Bacillw subtilis. Colloidc Su$, A. Pbysicocbem. Eng.
Aspects 1999,152,3-10.
Desai, J.; I.M. Banat. Microbial production of surktants and their commercial potential. Microbiol. MoL BioL &. 1997,61,47-64.
Energy Information Administration. international Energy OutlooR, U.S.Department of Energy:
Washington, D.C., 2006.

148 0 M.J. Mclnerney et at.

Energy Information Administration. Annual Energy Outlook; U. S. Department of Energy: Washington, D.C., 2007.
Finnerty, W R ; M.E. Singer. Microbial enhancement of oil recovery. Bio/Technologr 1983, I,
47-54.
Folmsbee, M.; K.E. Duncan; S.-0. Han; D.P. Nagle; E. Jennings; M.J. McInerney. Re-identification of the halotolerant, biosurfactant-producing Bacillus licheniformis strain JF-2 as Bacillus
rnojavensis strain JF-2. Syst. Appl. Microbiol. 2006, 29, 6 4 5 4 4 9 .
Galli, G.; E Rodriguez; I? Cosmina; C. Pratesi; R. Nogarotto; F. de Ferra; G. Grandi. Characterization of the surfactin synthetase multi-enzyme complex. Biochim. Biophys. Acta 1994, 1205,
19-28.
Georgiou, G.; S.-C. Lin; M.M. Sharma. Surface-active compounds from microorganisms. Biol
Technology 1992,10, 6 0 4 5 .
Grangemard, I.; J.-M. Bonmatin; J. Bernillon; B.C. Das; F. Peypoux. Lichenysins G, a novel family of lipopeptide biosurfactants from Bacillus licheniformis IM 1307: production, isolation and
structural evaluation by NMR and mass spectrometry. J. Antibiot. 1999,52, 363-373.
Grangemard, I.; E Peypoux; J. Wallach; B.C. Das; H. Labbi; A. Caille; M. Genest; R. MagetDana; M. Ptak; J.M. Bonmatin. Lipopeptides with improved properties: structure by NMR,
purification by HPLC, and structure-activity relationships of new isoleucyl-rich surfactins. J.
Peptide Sci. 1997,3, 145-154.
Green, D.W.; G.I? Willhite. Enhanced oil recovery. Society of Petroleum Engineers: Richardson,
TX, 1998.
Hall, C.; I? Tharakan; J. Hallock; C. Cleveland; M. Jefferson. Hydrocarbons and the evolution of
human culture. Nature 2003, 426 318-322.
Hamoen, L.W.; G. Venema; 0 . P Kuipers. Controlling competence in Bacillus subtilix shared use
of regulators. Microbiologr 2003, 149, 9-17.
Hamon, M.A.; B.A. Lazazzera. The sporulation transcription factor SpoOA is required for biofilm
development in Bacillus subtilis. Mol. Microbiol. 2001, 42, 1199-1209.
He, Z.; Y. She; T. Xiang; F. Xue; B. Mei; Y. Li; B. Ju; H. Mei; T.F. Yen. MEOR pilot sees encouraging results in Chinese oil field. Oil GasJ. 2000,98, 4 6 5 1 .
Hitzman, D.O. Petroleum microbiology and the history of its role in enhanced oil recovery. ?he
1982 International Conference on Microbial Enhancement of Oil Recovery, Bartlesville, OK, 1983;
U. S. Department ofEnergy: Bartlesville, OK, pp. 162-218.
Hirzrnan, D.O. Review of microbial enhanced oil recovery field tests. ?he Symposium on Applications ofMicroorganisms to Petrokum Technologr, Springfield, VA, 1988; T.E. Burchfield, R.S.
Bryant, Eds.; National Technical Information Service: Springfield, VA, pp. VI 1-VI 41.
Hofemeister, 1.;B. Conrad; B. Adler; B. Hofemeister; J. Feesche; N. Kucheryava; G. Steinborn; I?
Franke; N. Grammel; A. Zwintscher, et al., Genetic analysis of the biosynthesis of non-ribosomal peptide- and polyketide-like antibiotics, iron uptake and biofilm formation by Bacillw subtilis
All3. Mol. Genet. Genomics 2004,272, 363-378.
Horowitz, S.; W.M. Griffin. Structural analysis of Bacillis lichenformis 86 surfactant. J. Ind. Microbiol. 1991, 7,45-52.
Hourdou, M.; E Besson; G. Michel. Studies on the biosynthesis of beta-amino acids, the lipid

Lipopeptide Biosurfactants and Their Use in Oil Recovery 0

149

moiety of iturins A, in Bacifus subtifis.J. Antibiot. 1988,41, 207-2 1 1.

Hourdou, M.L.; F, Besson; I. Tenoux; G. Michel. Fatty acid and beta-amino acid synthesis in
strains of Bacillus subtifis producing iturinic antibiotics. Lipid 1989,24,940-944.
Infante, M.R; V. Moses. Synthesis and activity properties of hydrophobidhydrophilic peptides.
Int. J. P e p t i & k . 1994,43, 173-179.
Islam, MA.; A. Gianetto. Mathematical modelling and scaling up of microbial enhanced oil
recovery. J. Can. Pet. Tecbnol 1993,32,30-36.
Jack, T.R. Microbially enhanced oil recovery. Bioncovery 1988, I, 59-73.
Jenneman, G.E.; M.J. McInerney; R M . Knapp; J. B. Clark; J.M. Feero; D.E. Revus; D.E. Menzie. A halotolerant, biosufactant-producing Bacillus species potentially useful for enhanced oil
recovery. Dn? Ind MirrobioL 1983,24,485-492.
Jenny, K.; 0. Kaeppeli; A. Fiechter. Biosurfactants from Bacillus licbenrjmnis: Structural analysis
and characterization.Appl. MirrobioL BiocecbnoL 1991,36, 5-13.
Johnson, S.J.; M. Salehi; ICE. Eisert; J.-T. Liang; G. A. Bala; S.L. Fox. Biosurfactants produced
from agricultural process waste streams to improve recovery in fractured carbonate reservoirs.
SPE 106078. SPE International Symposium on Oilfield Chemistry, Society of Petroleum Engineers, Houston, TX,2007.
Joshi, S.; C. Bharucha; A.J. Desai. Production of biosurfactant and antifungal compound by
fermented food isolate Bacillus subtifis 20B. Bions. EcbnoL 2008% 99, 4603-4608.
Joshi, S.; C. Bharucha; S. Jha; S.Yadav; A. Nerurkar; A.J. Desai. Biosurfactant production using
molasses and whey under thermophilic conditions. Biores. Tecbnol. 2008b, 99, 195-199.
Kaneda, T. Biosynthesis of branched chain fatty acids. IV-Factors affecting relative abundance of
fatty acids produced by Bacillussubtilis. Can. J. Mirrobiol. 1966,12, 501-514.
Kearns, D.B.; E Chu; R Rudner; R. Losick. Genes governing swarming in Bacillus subtilis and
evidence for a phase variation mechanism controlling surface motility. MoL MicrobioL 2004,52,
357-369.

Kearns, D.B.; R Losick. Swarming motility in undomesticated Bacillus subtilis. Mol. Mirrobiol
2003,49,581-590.

Kinsinger, R.F.; M.C. Shirk; R Fall. Rapid surface motility in Bacillus subtilis is dependent on
extracellular surfactin and potassium ion. J. Bactcriol. 2003, 185, 5627-5631.
Kleinkauf, H.; H. von Dohren. Biochemistry ofpepti& antibiotics: Recent advances in the biotecbnology ofp-hctamc and microbial bioactiveprptidcs. Walter de Gruyter: Berlin, 1990; pp. 1-522.
Knapp, RM.; M.J. McInerney; D.I? Nagle; M.M. Folmsbee; S. Maudgalya. Development of a
microbially enhanced oil recovery process for the Delaware Childers Field, Nowata County.
Okh. GeoL Survcy Circ. 2002, 108, 193-200.
Kohli, RM.; J.W. Trauger; D. Schwarzer; M.A. Marahiel; C.T. Walsh. Generality of peptide
cydization catalyzed by isolated thioesterase domains of nonribosomal peptide synthetases.
Biochemistry 2001,40,7099-7108.
Lazar, I. MEOR field trials carried out over the world during the last 35 years. Dw. Pet. Sci. 1991,
31,485-530.

Lin, S.-C. Biosurfactants: Recent advances./. Cbem. Tecbnol. Biotecbnol. 1996,66, 109-120.

150 0 M.J. Mclnerney et al.

Lin, S.-C.; K.S. Carswell; M.M. Sharma; G. Georgiou. Continuous production of the lipopeptide
biosurfactant of Bacillus licbeniformis JF-2. Appl. Microbiol. Biotecbnol. 1994a, 41,281-285.
Lin, S.-C.; M.A. Minton; M.M. Sharma; G. Georgiou. Structural and immunological characterization of a biosurfactant produced by Bucilh licbeniformis JF-2. AppL Environ. Microbiol.
1994b, GO, 31-38.
Lin, S.-C.; M.M. Sharrna; G. Georgiou. Production and deactivation of biosurfactant by Bacillus
licbenijirmis JF-2. Biotecbnol. Prog. 1993,9, 138-145.
Macquaire, F.; F. Baleux; E. Giaccobi; T. Huynh-Dinh; J.-M. Neumann; A. Sanson. Peptide secondary structure induced by a micellar phospholipidic interface: proton NMR conformational
study of a lipopeptide. Biochemistry 1992,31,25762582.
Maier, R.M.; G. Soberon-Chavez.Pseudomonac ueruginosu rhamnolipids: biosynthesis and potential apllications. Appl. Microbiol. Biotechnol. 2000,54, 625-633.
Makkar, R.S.; S.S. Cameotra. Utilization of molasses for biosurfactant production by two Bacillus
strains at therrnophilic conditions./. Am. Oil. Cbem. SOC.1997, 74, 887-889.
Makkar, R.S.; S.S. Cameotra. Production of biosurfactant at mesophilic and thermophilic conditions by a strain of'Bucillus subtilis. /. Ind. Microbiol. Biotechnol. 1998, 20, 48-52.
Marahiel, M.A.; M.M. Nakano; l? Zuber. Regulation of peptide antibiotic production in Bacillus.
Mol. Microbiol. 1993, 7, 631-636.
Matsuyama, T.; K. Kaneda; Y. Nakagawa; K. Isa; H. Hara-Hotta; I. Yano. A novel extracellular
cyclic lipopeptide which promotes flagellum-dependentand -independent spreading growth of
Serrutiu murcescens.J. Bucteriol. 1992, 174, 1769-1 776.
Maudgalya, S.; R.M. Knapp; M.J. McInerney; D.l? Nagle; M.J. Folsmbee. Development of a
bio-surfactant-based enhanced oil recovery procedure. SPE 89473. ?he SPEIDOE Improved Oil
Recovery Symposium, Tulsa, OK, 2004; Society of Petroleum Engineers, Richardson, TX.
Maudgalya, S.; M.J. McInerney; R.M. Knapp; D.l? Nagle; M.M. Folmsbee. Tertiary oil recovery
with microbial biosurfactant treatment of low-permeability Berea sandstone cores. SPE 94213.
Society of Petroleum Engineers Production and Operutions Symposium, Oklahoma City, OK, 2005;
Society of Petroleum Engineers: Richardson, TX.
McInerney, M.J.; K.E. Duncan; N. Youssef; T. Fincher; S.K. Maudgalya; M.J. Folrnsbee; R.
Knapp; D.R. Simpson; N. Ravi; D.l? Nagle. Development ofMicroorgunisms with Improved Transport and Biosurfactant Activityfor Enhunced Oil Recouep Department of Energy, ASM Press,
Inc., Washington, D.C.: 2005a.
McInerney, M.J.; M. Javaheri; D.l? Nagle. Properties of the biosurfactant produced by Bacillw
licbenzjirmis strain JF-2. /. Ind. Microbiol. 1990,5, 95-102.
McInerney, M.J.; G.E. Jenneman; R.M. Knapp; D. E. Menzie. Biosurfactant and enhanced oil
recovery. 1985.
McInerney, M.J.; D.l? Nagle; R.M. Knapp. Microbially enhanced oil recovery: Past, present,
and future. Petrokum Microbiology; B. Ollivier, M. Magot, Eds.; ASM Press: Washington, DC,
2005b; pp. 215-238.
McInerney, M.J.; K.L. Sublette. Petroleum microbiology: Biofouling, souring, and improved oil
recovery. Manual ofEnvironmentul Microbiohgy; C.J. Hurst, G.R. Knudsen, M.J. McInerney,
L.D. Stetzenbach, M.V. Water, Eds.; 1997; pp. 600-607.cPublisher and location?>

LipopeptideBiosurfactants and Their Use in Oil Recovery 0 15 1

McInerney, M.J.; G. Voordouw; G.E. Jenneman; K L. Sublette. Oil field microbiology. Manual of
Environmental Microbiology; C.J. Hurst, RL. Crawford, J.L. Garland, D.A. Lipson, A.L. Mills,
L.D. Stetzenbach, Eds.; ASM Press, Inc.: Washington, D.C, 2007; pp. 898-91 1.
Menkhaus, M.; C. Ullrich; B. Kluge; J. Vater; D. Vollenbroich; R.M. Kamp. Structural and functional organization of the surfactin synthetase multienzyme system./. BioL Chcm. 1993,2G8,
7678-7684.
Morikawa, M.; H. Daido; T. Takao; S. Murata; Y. Shimonishi; T. Imanaka. A new lipopeptide
biosurfactant produced by Arthrobacter sp. strain MIS38. J. Bacteriol 1993, 175,01594466.
Morikawa, M.; Y. Hirata; T. Imanaka. A study on the structure-function relationship of lipopeptide biosurfactants. Biocbim. Biopbys. Acta 2000, 1488,211-218.
Mukherjee, S.; I? Das; R. Sen. Towards commercial production of microbial surfactants. Trcndc
Biotcchnol. 2006,24, 509-5 15.
Nguyen, T.; N. Youssef; M.J. McInerney; D. Sabatini. Rhamnolipid biosurfactant mixtures for
environmental remediation. Water &s. 2008,42, 1735-1743.
Oka, K.; T. Hirano; M. Homma; H. Ishii; K. Murakami; S. Mogami; A. Motizuki; H. Morita; K.
Takeya; H. Itokawa. Satisfactory separation and MS-MS spectrometry of six surfactins isolated
from Bacillw subtilis natto. Chcm. Pharm. BuU. 1993,41, 1000-1002.
Ollivier, B.; M. Magot. Pctrohm Microbiology. ASM Press: Washington, D.C., 2005.
Peypoux, F.; J.M. Bonmatin; J. Wallach. Recent trends in the biochemistry of surfactin. Appl
MicrobioL BiotecbnoL 1999,5I, 553-563.
Peypoux, F.; J-M. Bonmatin; H. Labbe; I. Grangemard; B.C. Das; M. Ptak; M. Wallach; G. Michel. [Ala4]Surfactin, a novel isoform from Bacillw subtilis studied by mass and NMR spectrocopies. Eur, /. Biocbem. 1994,224,89-96.
Peypoux, F.; 1.-M. Bonmatin; H. Labbe; B.C. Das; M. Ptak; G. Michel. Isolation and characterization of a new variant of surfactin, the [Val7]-surfactin.Eur. /. Biocbem. 1991,202, 101-106.
Peypoux, F,; G. Michel. Controled biosynthesis of Vd7- and Leu7-surfactins. Appl. Microbiol.
Biotecbnol. 1992,36, 5 15-5 17.
Planckaert, M. Oil reservoirs and oil production. Pctrohm Microbiolou; 8. Ollivier, M. Magot,
Eds.; ASM Press: Washington, D.C., 2005; pp. 3-20.
Reed, RL.; R. Healy. Some physical aspects of microemulsion flooding: a review. Improved Oil
Rccovrry by Surfacklnt and Polymer Flooding; D.O. Shah, RS. Scheckter, Eds.; Academic Press,
Inc.: New York, 1977; pp. 383437.
Rodrigues, L.; I.M. Banat; J. Teixeira; R Oliveira. Biosurfactants: potential applications in medicine. /. Antimicrob. Chcmotber. 2006,57,609-618.
Rosenberg, E.; E.Z. Ron. High- and low-molecular-mass microbial surfactants.AppL Microbiol.
Biotecbnol. 1999,52, 154-162.
Schaller, K.D.; S.L. Fox; D.F. Bruhn; KS. Noah; GA. Bala. Characterization of surfactin from
Bacillw subtilis for application as an agent for enhanced oil recovery. AppL Biocbm. BiotecbnoL
2004,113-116,827-836.
Schneider,A.; M.A. Marahiel. Genetic evidence for a role of thioesterase domains, integrated in
or associated with peptide synthetases, in non-ribosomal peptide biosynthesis in Baciffzasubtilis.

152 0 M.J. Mclnerney et al.

Arch. Microbiol. 1998, 169, 404-410.

Schneider,A.; T. Stachelhaus; M.A. Marahiel. Targeted alteration of the substrate specificity of
peptide synthetases by rational module swapping. Mol. Gen. Genet. 1998,257, 308-318.
Schneider, K.B.; T.M. Palmer; A.D. Grossman. Characterization of comQ and c o d , two genes required for production of Codpheromone in Bacillus subtilis. /. Bacteriol. 2002, 184,410419.
Simpson, D.R.; R.M. Knapp; N. Youssef; K.E. Duncan; M.J. McInerney; C. Brackin. Microbial
stimulation treatment of high water-cut wells in the Viola formation, Pontotoc County, OK.
Tertiary Oil Recovery Conjrence, Wichita, KA, 2007.
Singh, I?; S.S. Carneotra. Potential applications of microbial surfactants in biomedical sciences.
Trend Biotechnol. 2004,22, 142-146.
Stachelhaus,T.; A. Schneider; M.A. Marahiel. Rational design of peptide antibiotics by targeted
replacement of bacterial and fungal domains. Science 1995,269, 5571-5574.
Stein, T. Bacillus subtilis antibiotics: structures, syntheses and specific functions. Mol. Microbiol.
2005,56, 845-857.
Stein, T.; J. Vater; V. Kruft; A. Otto; B. Wittmann-Liebold; I? Franke; M. Panico; R. McDowell;
H. R Morris. The multiple carrier model of nonribosomd peptide biosynthesis at modular
multienzymatic templates. /. Biol. Chem. 1996,271, 15428-1 5435.
Steller, S.; A. Sokoll; C. Wilde; F. Bernhard; I? Franke; J. Vater. Initiation of surfactin biosynthesis
and the role of the SrfD-thioesteraseprotein. Biochemistry 2004,43, 11331-1 1343.
Strand, S.; D.C. Standnes; T. Austad. Spontaneous imbibition of aqueous surfactant solutions into
neutral to oil-wet carbonate cores: Effects of brine salinity and composition. Energy Fuels 2003,
17, 1133-1144.
Sugihardjo, E.H.; S.W. Pratomo. Microbial core flooding experiments using indigenous microbes.
SPE 57306. SPEAsia Pacijic Improved Oil Recovery Conjrence, Kuala Lampur, Malaysia, 1999;
Society of Petroleum Engineers, Richardson, TX.
Taber, J.J. Dynamic and static forces required to remove a discontinuous oil phase from porous
media containing both oil and water. SOC.Pet. Ens /. 1969,9,3-12.
Thimon, L.; E Peypoux; B.C. Das; J. Wallach; G. Michel. Selective esterification of surfactin:
Preparation and properties of surfactin methyl esters. Biotechnol. Appl. Biochem. 1994,2U,
4 15-423.
Thomas, CJ?; G.A. Bala; M.L. Duvall. Surfactant-basedenhanced oil recovery mediated by naturally occurring microorganisms. SOC.Pet. Eng. Reservoir Eng. 1993, 11,285-291.
Tran, TX H. Interactions between biosurfactant-producing Pseudomonas and Phytophthom species.
Wagenungen University: The Netherlands, 2007.
Van Dyke, M.I.; H. Lee; J.T Trevors. Applications of microbial surfactants. Biotechnol. Adv. 1991,
9,241-252.
Vater, J.; T. Stein; D. Vollenbroich; V. Kruft; B. Wittmann-Liebold; I? Franke; L. Liu; I? Zuber.
The modular organization of multifunctional peptide synthetases./. Protein Chem. 1997, 16,
557-564.
von Dohren, H., Peptides. Genetics and Biochemistry ofAnti6iotic Production; L. C. Vining; C.
Stuttard, Eds.; Buttenvorth-Heineman: Boston, MA, 1995; pp. 121-171.

Lipopeptide Biosurfactants andTheir Use in Oil Recovery 0 153

Wang, Q.; X Fang; 8. Bai; X. Liang; RJ. Shuler; W. A. Goddard; Y. Tang. Engineering bacteria for
production of rhamnolipid as an agent for enhanced oil recovery. Biotecbnol. Bioeng. 2007,98,
842-853.

Weihong, Q.D.; D. Liangjun; 2. Zhongkui; Y. Jie; L. Huamin; L. Zongshi. Interfacial behavior of

pure surfactants for enhanced oil recovery-Pan 1: A study on the adsorption and distribution
of cetylbenzene sulfonate. Tensidc Surfactants DeteTents 2003, 40, 87-89.
Wine, A. That's oil, folks... Nature 2007, 445, 14-17.
Yakimov, M.M.; M.M. Amro; M. Bock; K. Boseker; H.L. Fredrickson; D.G. Kessel; K.N. Timmis. The potential of Bacillw Licbenz$tmis strains for in situ enhanced oil recovery. /. Pet. Sci.
Ens 1997,18, 147-160.
Yakimov, M.M.; H.L. Fredrickson; K.N. limmis. Effect of heterogeneity of hydrophobic moities
on surface activity of lichenysin A, a lipopeptide biosurfictant from Bacillw licbenz$rmis BASSO.
Biotecbnol. Appl Biocbem. 1996,23, 13-18.
Yakimov, M.M.; K.N. Timmis; V. Wray; H.L. Fredrickson. Characterization of a new lipopeptide
surfactant produced by thermotolerant and halotolerant subsurface Bacillw licbenz$rmis BASSO.
Appl Environ. Mirrobiol. 1995, 61, 1706- 17 13.
Youssef, N.; K. Duncan; M. McInerney. Importance of the 3-hydroxy fatty acid composition of
lipopeptides for biosurfactant activity. Appl. Environ. Microbiol. 2005, 71,7690-7695.
Youssef, N.; K. Duncan; D. Nagle; K. Savage; R Knapp; M. McInerney. Comparison of methods
to detect biosurfactant production by diverse microorganisms. /. Mimbiol Metboa? 2004,56,
339-347.

Youssef, N.; T. Nguyen; DA. Sabatini; M.J. McInerney. Basis for fbrmulating biosurfactant
mixtures to achieve ultra low interfacial tension values against hydrocarbons./. Znd Microbiol.
Biotecbnol. 2007% 34,497-507.
Youssef, N.; D.R. Simpson; K.E. Duncan; M.J. McInerney; M. Folmsbee; T Fincher; RM.
Knapp. In situ biosurfactant production by Baciflw Strains injected into a limestone petroleum
reservoir. Appf. Environ. Mirrobiol 2007b, 73, 1239-1247.
Zaijic, E.J. Scale up of microbes for single well injection. Proceedings of the First International
MEOR Workshop; 1987; J. King, J.; D. Stevens, Eds.; U.S.Department of Energy, Bartlesville
Project Office: Bartlesville, OK, 1987, pp. 241-246.
Zekri, A.Y.; R Almehaideb; 0. Chaalal. Project of increasing oil recovery from UAE reservoirs
using bacteria flooding. SPE 56827. SPEAnnual Technical Conjkence, Richardson, TX,1999;
Society of Petroleum Engineers.
Zhang, X.; R M . Knapp; M.J. McInerney. A mathematical model for microbially enhanced oil
recovery process. Dcy. Pet. Sci. 1993,39, 171-186.

Employment of Phospholipids
and their Mimics in Biomedical
Applications

Influence of Pulmonary Surfactant

Protein Mimics on Model Lung
Surfactant
Hiromichi Nakahara', Sannamu Leez, and Osamu Shibata'
'Departmentof Biophysical Chemistry,Facub of PharmaceuticalSciences, Nagasaki lnternational
University,2825-7Huis Ten Bosch, Sasebo, Nagasaki 859-3298,Japan, 'Department of Chemistry,Faculty of
Science,Fukuoka University,8- 19- 1 Nanakuma,Johnan-ku, Fukuoka814-0180, Japan,

Introductionto Pulmonary Surfactants

Componentsund Compositions
Surface-active materials lining up at the air-alveolar fluid interface in mammalian lungs
or pulmonary surfactants (PS)play an important role in vital physiological processes,
such as respiratory movement. First, at the interface, PS reduces surface tension
to minimize the work of breathing and to prevent alveolar collapse in expiratory
movements (Schurch et al., 1976). Second, PS rapidly adsorbs on and spreads to
the interfice (Bastacky et al., 1995). PS is synthesized by type I1 pneumocytes and
then stored in characteristic lamellar body organelles in the cytoplasm prior to
secretion into the alveolar hypophase. It exists in the alveolar hypophase in the form
of heterogeneous phospholipid-rich aggregates, including tubular myelin, which is
associated with rapid adsorption to the air-water interface (Magoon et al., 1983;
Notter et al., 1986; Putman et al., 1996). Then, it adsorbs from these aggregates
to form a film at the air-water interface. PS is a mixture of lipids (=90 wt%) and
proteins (= 10 wtYo). The material consists mainly of phosphatidylcholines (especially
dipalmitoylphosphatidylcholine,DPPC; disaturated phospholipid, =50 wt%) and
smaller but significant amounts of phosphatidylglycerol (PG), palmitic acid (PA),
and four proteins (surfactant proteins: SP-A, -B, -C, and -D) (Veldhuizen et al.,
1998; Postle et al., 2001; Kruger et al., 2002; Yu & Possmayer, 2003). DPPC is
not common in cell membranes and has been recognized as an important surfactant
component soon after the discovery that surface active agents are present in the
lungs (Brown, 1964). SP-A, -B, and -C have crucial biophysical significance in PS
function. SP-A comprises =5% of lavaged PS on a weight basis and can be isolated
as a material closely associated with surfactant lipids during centrifugation. SP-B and
SP-C together make up = 1.5% by weight of typical lavaged surfactant isolates and
157

158 0 H. Nakaharaet al.

also are tightly associated with surfactant lipids. SP-D is less abundant than SP-A and
is found largely in the lipid-depleted supernatant of centrifuged lavage rather than
sediments with the majority of surfactant lipids and the other proteins (Kuroki et al.,
1991). SP-D is absent from packaged intracellular surfactant in type I1 cell lamellar
bodies (Crouch et al., 1991; Voorhout et al., 1992).
DPPC is known to form tightly packed and solid surface films and has properties
that lower surface tension to near zero in the multilayer state during Z-A (surface
pressure-molecular area) cycling. Disaturated phospholipids in PS are thought to
become enriched in the surface films at the expense of fluid components during
compression to low interfacial area, or high surface concentration. However, the
resultant rigidity of the condensed films at physiological conditions causes it to
adsorb slowly from the alveolar fluid and to respread poorly from multiple materials
(Fleming & Keough, 1988). The unsaturated and anionic PG is thought to help
DPPC molecules adsorb and respread rapidly despite the fact that it does not
lower surface tension as effectively as DPPC. In addition, the potential for specific
molecular biophysical associations between anionic phospholipids and one or more
of the surfactant apoproteins has also been widely suggested (Baatz et al., 1990;,
Johansson et al., 1991; Chang et al., 1998). PA is present in a relatively low amount
in comparison with DPPC, but it is used as a very important additive for the proper
Lnctioning of both natural and synthetic PS replacement formulations. The addition
of 510 wt% PA to natural PS extracts obtained from animal sources has induced a
significant improvement in their properties both in vitro and in vivo (Cockshutt et
al., 1991; Gorree et al., 1991).
Besides lipids, the four proteins (SP-A, -B, -C, and -D) are associated with PS
functions in many cases (Hawgood & Shiffer, 1991; Johansson & Curstedt, 1997).
SP-A and SP-D, which contain hydrophilic proteins, play an important role in the
first-line defenses against inhaled pathogens (Miyamura et al., 1994) and in the
storage and transport of PS (Goerke, 1974; Voorhout et al., 1991). O n the contrary,
SP-B and SP-C are hydrophobic proteins that dominate the surface activity of PS.
They promote considerable adsorption of PS from the hypophase to the surface
at the air-water interface. Furthermore, they trigger a reversible exclusion of fluid
materials from the multicomponent monolayers when they are compressed beyond
their collapse pressures (Taneva & Keough, 1994a, 1994b, 1994c). SP-A is the most
abundant apoprotein in endogenous lung surfactants and has an amino acid sequence
that is highly conserved among animal species (Benson et al., 1985; White et al.,
1985; Floros et al., 1986; Boggaram et al., 1988). The N-terminal portion of the
molecule is collagen-like, while its C-terminal carbohydrate-binding sequence places
it in the family of mammalian C-type lectins and the collectin family. The human
SP-A (35 kDa) contains 228 amino acids with a mix of hydrophilic and hydrophobic
residues. This monomer has a short N-terminal domain of 7 amino acids followed
by a collagen-like region with 23 proline-rich Gly-X-Y repeats broken with a ProCys-Pro-Pro sequence between the lYh and 14th residues. SP-A also contains an

Influenceof Pulmonary Surfaaant Protein Mimics on Model Lung Surfactant0

159

intermediate amphipathic helical domain which is known to contribute to lipid

binding and a calcium-dependent C-terminal lectin-like carbohydrate-binding
regions. The major molecular biophysical action of SP-A is to increase the aggregation
and ordering of phospholipids in a calcium- and pH-dependent fishion (Cajal et
al., 1998; Ruano et al., 1998). SP-D (43 kDa) is distinct from the other surfactant
apoproteins in terms of biophysical function in PS. This hydrophilic calciumdependent C-type lectin has significant structural analogy to SP-A. The molecular
structure of monomeric SP-D contains an N-terminal collagenous domain, a linking
region, and a C-terminal carbohydrate-binding domain (Persson et al., 1990). The
N-terminus has a short initial noncollagen region, followed by a collagen-like domain
containing 59 Gly-X-Y repeats that are significantly longer than those in SP-A. This
is followed by a short linking sequence connected to the C-terminal carbohydrate
recognition domain containing invariant amino acids including four cysteine residues
to allow intramolecular disulfide bridging as in SP-A. SP-D is not fbund in either
lamellar bodies or in tubular myelin (Crouch et al., 1991; Voorhout et al., 1992).
SP-D, like SP-A, is a member of the collectin family of host defense proteins and may
also participate in surfactant metabolism and other aspects of pulmonary biology
(Crouch, 1998; Mason et al., 1998).
The human SP-B contains 79 amino acids and has a molecular weight of 8.7
kDa (Glasser et a]., 1987; Hawgood et al., 1987). SP-B also forms dimers and higher
oligomers through intermolecular cysteine-linked sulfhydryl bridges ( T h h a s h i
et a]., 1990; Johansson et al., 1991). SP-B contains not only significant residues
of hydrophobic amino acids but also multiple polar ones that can interact with
headgroup moieties of phospholipids. Although the exact start and stop boundaries
of its amphiphilic helices have not been determined yet, in the model proposed
by Andersson and co-workers (Andersson et al., 1995), four amphipathic helices,
bounded by residues 8-22,27-38,42-50, and 67-74, are aligned in an antiparrallel
left-handed hair-pin like motif. SP-B has been shown to increase phospholipid
aggregation as well as to disrupt and hse phospholipid bilayers and vesicles, and to
promote phospholipid insertion into surface films (Oosterlaken-Dijksterhuis et al.,
1991; Chang et al., 1998), leading to a relatively peripheral location in phospholipid
bilayers. SP-B, having a net charge of +7, affects phospholipids not only through
hydrophobic interactions but also through electrostatic interactions between its
polar residues and the headgroups of both zwitterionic and anionic phospholipids
in particular (Baatz et al., 1990). ?he human SP-C contains 35 amino acids and
has a molecular weight of 4.2 kDa including cysteine-linked palmitoyl groups
Uohansson et al., 1988; Curstedt et a]., 1990). Due to its extreme hydrophobicity,
SP-C forms a transmembrane a-helix that interacts primarily with phospholipid
fatty chains (Johansson, 1998). SP-C with a net charge of +2 primarily associates
with phospholipids through hydrophobic interaction, although its positively charged
residues can also interact with their headgroups. The biophysical molecular behavior
of SP-C largely resembles that of SP-B. However, SP-C appears to have less selectivity

160 0 H. Nakahara et al.

than SP-B for particular phospholipids (Pkrez-Gil et al., 1995). Both SP-B and SP-C
participate in the selective refining process during expiration. Fluid lipids with less
ability to sustain high surface pressures are preferentially ejected to enrich the surface
films in rigid phospholipids (mainly DPPC) (Nahmen et al., 1997; Lipp et al., 1998).
In general, this behavior has been known as the extraction phenomenon. It has been
suggested that the extracted materials form and occupy a surface-associated reservoir
at the adjacent interface (Schiirch et al., 1998; Takamoto et al., 2001; Ding et al.,
2003). This extraction is indispensable for PS function. However, its mechanism has
not been made clear yet.

Deficiency Symptoms in the Lungs

A deficiency of PS (mainly, newborn babies) has been shown to cause neonatal
respiratory distress syndrome (NRDS) in premature infants (Avery & Mead, 1959).
In a related distress for adults, an acquired inactivation of PS is likely to set up adult
acute respiratory distress syndrome (ARDS). So far, a common treatment of NRDS
is the surfactant replacement therapy in which exogenous surfactant preparations are
administered to NRDS infants. These medicines can be roughly divided into three
types. The first is the natural based source obtained from animals. This preparation is
mostly used in clinical cases. For example, Curosurf (Chiesi Pharmaceutici, Parma,
Italy), Survanta (Ross Laboratories, Columbus, O H , USA), and Surfacten (Surfactant
TA; Mitsubishi Pharma Corporation, Osaka, Japan) are analogs to native PS, and
they have been clinically used in several countries. Although the preparations above
are quite effective for the patients, they involve the risk of animal infections, such
as bovine spongiform encephalopathy (BSE), potential viral contamination, and
inherent immunity. Other drawbacks include a costly purification procedure and a
difficulty of producing batch-to-batch uniformity.
The second is asynthetic type made from synthetic surfactants without proteins. It
had previously been studied as a substitute for natural types, and then some medicines
(e.g., Exosurf and ALEC) were developed (Kirkness et al., 2003a, 2003b). Although
these medicines can be used at a lower cost, they do not have a sufficient beneficial
effect since they contain no proteins. Therefore, the preparations completely made of
synthetic surfactants with proteins (or peptides) is being developed for in the clinical
surhctant replacement therapy for NRDS. In addition, basic and clinical research on
an expanded application of these preparations to acute respiratory distress syndrome
(ARDS) has recently been carried out (Notter, 2000). Tanaka and co-workers (Tanaka
et al., 1986) have developed a most effective lipid replacement, DPPC/PG/PA (=
68:22:9 by weight) mixture. The composition of this mixture mimics that of the
lipids existing in the amnion liquid. It has been reported that the surface properties of
this mixture are enhanced by adding a small amount of lipid-binding protein (Tanaka
et al., 1986). Therefore, the DPPC/PG/PA mixture has been examined by many
researchers to develop new pulmonary preparations and to clarify pulmonary functions
(Gustafsson et al., 1996, 2000 Ma et al., 1998). Recently, Surfaxin, a synthetic PS

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

161

preparation with KL, peptide (consisting of novel 21-amino acid residues) has been
approved for clinical use (Revak et al., 1996; Ma et al., 1998; Cai et al., 2003).
Moreover, the preparation containing recombinant SP-C (rSP-C; 34 amino acid
residues) has been under active investigation for clinical efficacy (Amrein et al., 1997;
Krol et al., 2000; Grigoriev et al., 2003). Both preparations are composed of the
DPPC/PG/PA (= 68:22:9 by weight) mixture with the exception of minor differences
in the PG species. Although a few medicines containing new protein analogs have
been developed (Ma et al., 1998; Veldhuizen et al., 2000), no replacement surfactant
has been found comparable to the complete native surfactant with SP-B and -C.
Therefore, the development of such synthetic-type medicines is strongly desired.

Peptide Designing for Replacement Treatments

Recently, the authors synthesized a series of 18-mer amphiphilic a-helical peptides
made up of hydrophobic and hydrophilic residues of a ratio of 1 3 5 , 11:7,9:9, 7: 11,
and 5 1 3 (abbreviated as He1 13-5, He1 11-7, He1 9-9, He1 7-1 1, and He1 5-13,
respectively), which are the shorter peptides by 3 residues than KL, (Fig. 7.1) (Kiyota,
et al., 1996).
When they take a-helical structures, the hydrophobic part and the hydrophilic
part are completely separated in a-helical structure (Fig. 7.2). It has been suggested
that the most hydrophobic He1 13-5 of the five residues interacts with specific
phospholipid mixtures to adopt long nanotubular structures (Kitamura et al., 1999).
In addition, the interaction could induce the formation of membrane structures
resembling cellular organelles, such as the Golgi apparatus (Lee et al., 2001; Furuya
et al., 2003). The structure has a strong resemblance to the tubular myelin structure
made from secreted PS from type I1 pneumocytes. Therefore, it is expected to be
able to mimic the biophysical hnctions of SP-B and to be safe for RDS patients.
Furthermore, monolayer experiments using a modified Wilhelmy surface balance
showed that a mixture of He1 13-5 and phospholipids are spread and adsorbed
quickly, comparable with Surfacten, which is a modified natural bovine PS and is
used for RDS patients in Japan (Lee et al., 2004).
The authors have focused specifically on newly designing PS preparations and
elucidating the interfacial biophysical behavior and mechanism of SP-B during
respiration using the model peptide (He1 13-5). The interfacial behavior of spread
Number of
residue

1-10

1 1-20

He1 13-5

KLLKLLLKLW

LKLLKLLL

KL.

KLLLLKLLLL

KLLLLKLLLL

rSP-C

GIPFFPVHLK

RLLIWWW

LlVWlVGAL

21-30

31-34

LlGL

Fig. 7.1. Amino acid sequences of synthetic peptides (He1 13-5, KL,, and rSP-C) mimicking native
SP-B and SP-C. Abbreviations: K, lysine; L leucine; W, tryptophan; G, glycine; I, isoleucine; P, proline; F,
phenylalanine;V, valine; H, histidine; R, arginine; A, alanine.

162 0 H.Nakahara et al.

G
Fig. 7.2. Helical wheel representations of He1 13-5, KL,, and rSP-C. The abbreviations with open cycles

denote hydrophilic amino acids.

monolayers for the DPPC/Hel 13-5, DPPC/PG/Hel 13-5, DPPC/PA/Hel 13-5,

and DPPC/PG/PA/Hd 13-5 systems and Surfacten has been investigated employing
Langmuir (z-A and A V-A) isotherms, fluorescence microscopy (FM), and atomic
force microscopy (AFM).

Research and Development of PSs

Interfacial Properties of He1 73-5
Pepti&

Sh?lC&lreS

The amphiphilic a-helical peptide (He1 13-5) consists of 13 hydrophobic residues

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0 163

(12 Leu and 1 Trp) and 5 hydrophilic residues (5 Lys) (Fig. 7.1 and 7.2). The
hydrophobic-hydrophilic balance was estimated both theoretically from the calculated
hydrophobicity value (or the magnitude of hydrophobic faces) and experimentally
from the retention times in reverse phase high-performance liquid chromatography
(RP-HPLC). For this modulation (hydrophobicity of He1 13-5 is 0.07) (Kiyota et
al., 1996), it is supposed that He1 13-5 mimics SP-B of the amphipathic helical
structure rather than SP-C of the transmembrane helical structure at the alveolar
air-liquid interface. In addition, when it forms an ideal a-helical structure, it has the
amphiphilic structure with a 260" hydrophobic sector region as shown in the helical
wheel representation (Fig. 7.2). This means that the hydrophobic and the hydrophilic
parts are completely separated in the a-helical structure. KL, peptides have been used
as an element in synthetic-type RDS medicines (or Surfaxin) to mimic the native
SP-B (Gustahon et al., 1996; Ma et al., 1998; Cai et al., 2003), whereas rSP-C
peptides are in clinical trials (Phase 111) as Venticute and are based on recombinant
SP-C (Amrein et al., 1997; Nahmen et al., 1997; Krol et al., 2000; Grigoriev et al.,
2003). Although rSP-C forms a transmembrane a-helix across the air-water interface,
He1 13-5 and KL, take amphiphilic a-helical structures at the interface. Notice that
the a-helical representations for KL, and rSP-C reveal a disheveled configuration of
hydrophobic and hydrophilic amino acids (Fig. 7.2).

Adrotpion Isotberms
Surface pressure (z)-time (t) isotherms of pure DPPC and the representative DPPC/
He1 13-5 mixture (XHe,
,3-5 = 0.1) formed by adsorption from their vesicle solutions
are shown in Fig. 7.3. It indicated a slow adsorption of DPPC molecules from
0.15 M NaCl subphase as reported previously (Serrano et al., 2005). For X,,,3-5 =
0.1, on the contrary, the z-t isotherm had a shoulder at 5 15 min and reached the
equilibrium surface pressure of 537 mN/m. SP-B accelerated the adsorption rate of
vesicular phospholipids from the subphase to the air-water interface (OosterlakenDijksterhuis et al., 1991). Similarly to native SP-B,our mimicking peptide, He1 13-5,
facilitated the interfacial adsorption of DPPC at the air-water interface (Nakahara et
al., 2006b).

Tmperatum Depmdcnce in Langmuk Isotberms

Figure 7.4 shows surface pressure (z)-molecular area (A) and surface potential
(AW-A isotherms of He1 13-5 monolayers. For the z-A isotherms, He1 13-5 had
no sharp bend and was expanded, indicating that the monolayer of He1 13-5 was
in the expanded state over the all of the surface pressures and temperatures studied.
He1 13-5 formed the expanded (disordered) film up to 542, 539, and 539 mN/m at
298.2, 303.2, and 310.2 K,respectively. However, the extrapolated area of He1 13-5
was ~ 2 . 6nm2, irrespectively of temperature. This suggests that He1 13-5 forms stable
disordered films under biological temperature ranges and has low solubility in the
aqueous subphase. However, the low solubility is not related to the exclusion of He1

164 0 H. Nakahara et al.

70
DPPC
DPPC/Hel13-5(XHe1 13-5=o,1)

60

50

20

10

0
0

20

40
60
t l min

80

Fig. 7.3. n-t isotherms of pure DPPC and the representative DPPC/Hel 13-5 system (X,,,,
adsorbing from their vesicular solutions in 0.15 M NaCl at 298.2 f 1.0 K.

100
,3-5

= 0.1)

13-5 from surface monolayers into the subphase upon compression. The value of the
extrapolated area reflects the molecular conformation of peptides. The value of =2.6
nm2 supports a-helical structure rather than Fsheet, random, and other structures.
Further investigation of its conformation at the interface was referred to in a previous
paper (Nakahara et al., 20OGb). The surface potential (AV) of He1 13-5 always showed
a positive variation under compression (Fig. 7.4). The surface potential of He1 13-5
monotonically increased to =380 mV, =350 mV, and =310 mV on compression
at 298.2, 303.2, and 310.2 K, respectively. The reduction of AVvalues reflects an
aggravation of molecular orientation due to the increase in molecular motions.
Fluorescence Microscopy (FM)
Fluorescence images of the phase state in the He1 13-5 films at various surface pressures
are shown in Fig. 7.5. It is widely accepted that the fluorescent probe is selectively
dissolved in the disordered phase, and not dissolved in the ordered phase (Losche &
Mohwald, 1984). Therefore, ordered domains can be visualized as dark domains in
the disordered/ordered coexistence region. The pure He1 13-5 films are indicated by
the bright homogeneous images from the low surface pressure to the collapsed film. It

Influence of Pulmonary Surfactant Protein Mimicson Model Lung Surfactant 0 165

>
.
a
a
E

z
E

Area per molecule / nm2

Fig. 7.4. n-A and AV-A isothermsof pure He1 13-5 monolayer in a 0.02MTris buffer solution (pH 7.4) with
0.13MNaClat298.2,303.2,and310.2K.

provides the evidence that He1 13-5 forms the disordered film miscible with the FM
probe, independent of surface pressure. The results also correspond to those of the
PA isotherms for He1 13-5 (Fig. 7.4).

Atomic Forcc Microscopy

AFM for this study provided both topography and phase contrast images. The
topography image reflects the sample topography, whereas the phase contrast image,
which is originated from the energy loss of the oscillating AFM tip, shows the
chemical structures of heterogeneous samples. Also, on surfaces with local variations
of mechanical properties, such as biological samples, the AFM phase image provides
the best contrast of fine morphological and nanostructural features. Figure 7.6 shows
AFM images of a pure He1 13-5 monolayer transferred to mica substrates below
and above the collapse pressure. AFM images of He1 13-5 at below 35 mN/m were
homogeneous (not shown). At 35 mN/m, on the other hand, some bright domains
appeared in the topography image (Fig. 7.6a), which are slightly higher than the
surrounding regions by 10.2 nm. The small protrusions suggest the existence of
intermediate states toward monolayer collapse, because the height of the protrusions

166 0 H. Nakahara et al.

10

Fig. 7.5. Fluorescence microscopy (FM)images of pure He1 13-5 spread in 0.02 M Tris buffer with 0.13 M
NaCl at 298.2 K for various surface pressures.The monolayers contain 1 mol% fluorescent probe (R18).
The scale bar in the lower right represents 100 pm.

is much smaller than the diameter of a-helical He1 13-5 (=I nm) as estimated by a
computer simulation (CS ChemOffice Ultra 5.0). After the collapse of He1 13-5 at 45
mN/m, many protrusions are observed over the whole range as shown in Fig. 7.6~.
They
are located in a line, implying that the collapse of He1 13-5 promotes another collapse
beside it. Zasadzinski and co-workers (Ding et al., 2001) reported that above the
plateau pressure many protrusions appeared in the natural system of Survanta, which
is a commercial R D S medicine containing a natural bovine PS. The resultant patches
were also observed in the DPPC/POPG (palmitoyloleoyl-phosphatidylglycerol)films
containing low amounts of protein analogs (Diemel et al., 2002), indicating that
they were induced by the extraction of fluid compositions (one of PS functions). In
the topography image (Fig. 7.6c), the protrusions (bright) and monolayers (dark) of
single-species He1 13-5 molecules coexist. The height of these protrusions is -2.0-3.0
nm, suggesting that He1 13-5 are excluded from the monolayer after plateau regions
on z-A isotherms (Fig. 7.4) and then a three-dimensional folding made of two or
three He1 13-5 molecules is formed. Notice that two kinds of domains are observed:
the domain with a bright midpoint (indicated by an arrow) and a domina without
it (indicated by a dashed arrow). The former shows a central higher protrusion, as
shown in the phase contrast image of Fig. 7.6d. These protrusions are found to be
disk like in shape with a typical diameter of =33 nm, corresponding to aggregations
containing =340 molecules of He1 13-5.

Binary DPPC/Hel13-5 Preparations

Langmuir Isotherms
The z-A and AV-A isotherms of DPPC containing He1 13-5 were studied to assess
the effect of He1 13-5 as a model PS protein. In Fig. 7.7, z-A isotherms exhibited two
small plateau regions (or two clear sharp bends) at surface pressures of -12-19 and
-42 mN/m as indicated by the arrows for XHe,
,3-5 = 0.3. The first plateau indicates
the first-order disordered/ordered transition and the second one at higher surface
pressures indicates the film collapse of He1 13-5, respectively. As the mole fraction of

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

167

(a)

Fig. 7.6.Atomicforce microxopy(AFM)imagesoftheHe113-5 monolayerin a tapping mode: topography

(a) and corresponding phase contrast images (b) transferred onto mica at 35 mN/m at the scan area of
400 x 400 nm; topography (c) and corresponding phase contrast image (d) transferred onto mica at
45 mN/m at the scan area of 400 x 400 nm.The solid and dashed arrows indicate the domain with and
without the bright midpoint, respectively.

He1 13-5 increased, the first transition points of all the z-A isotherms increased from
= 12 to 19 mN/m and became more and more unclear. The second sharp bend in the
z-A isotherm appeared at 542 mN/m (or the collapse pressure of pure He1 13-5). It is
quite notable that the second sharp bend appeared at the same surface pressure (=42
mN/m), independent of their constituents. The DPPC monolayer formed a stable film
up to =55 mN/m, and the monolayer for different compositions of the DPPC/Hel
13-5 systems (0 < X,,
I
0.3) remained stable up to =55 mN/m despite differences
in the fraction of He1 13-5, indicating that He1 13-5 stabilizes DPPC films, much
the same way SP-B hnctions (Taneva & Keough, 1994c; Piknova et al., 2001). This
behavior also confirms that He1 13-5 weakly interacts with specific compositions of
DPPC at high surface pressures, and then only the He1 13-5 component is extracted
from the DPPC/Hel 13-5 system at =42 mN/m. The extraction increased with
hrther compression, until only the DPPC component is located at the air-water
interface at =55 mN/m. Beyond 42 mN/m, the AV-A isotherms of 0.05 I XHe,
,3-5
I
0.3 gradually increased, whereas the AVof the other mole fractions shows nearly

,~~

168 0 H. Nakahara et al.

constant value. Judging from the above phenomena, if the surface pressure goes
beyond the collapse pressure, the A V-A isotherms become almost parallel to the area
axis. Namely, the rising of A Vbeyond the second sharp bend means that the packing
state or orientation of the molecule is still developing. At least, these results support
the extraction phenomenon of He1 13-5 from the monolayer surface, However, the
value of AV became almost similar ( ~ 5 0 0mV) at =55 mN/m, indicating that He1
13-5 molecules are not extracted completely from the monolayer.

Area per maIecuIe I nmL

Fig. 7.7. rr-A and AV-A isotherms of the DPPC/Hel 13-5 mixtures on a 0.02 M Tris buffer solution (pH 7.4)
with 0.13 M NaCl at 298.2 K.The arrows indicate two clear sharp bends forXHc,,3-5
= 0.3.

Influenceof Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

169

FM Obsmations
A series of FM images for the DPPC/Hel 13-5 system at 15 and 20 mN/m is
presented in Fig. 7.8, which show the disordered/ordered coexistence states; that is,
the dark domains reflect the ordered phase of DPPC, whereas the bright regions
reflect the disordered phases of DPPC and He1 13-5. The ordered domains of all
molar fractions grow in size with increasing surface pressure from 15 to 20 mN/m.
When the films were compressed further, they formed dark homogeneous images
consisting of almost the ordered phase up to the collapse pressure except for X,,,3-5
= 0.025. Note that the addition of a small amount of He1 13-5 to DPPC induced
the moth-eaten aggregation (shown by an arrow) of LC domains of DPPC. This
moth-eaten aggregation occurred only when a small amount of He1 13-5 coexists
with DPPC. In addition, its aggregation enlarged and expanded the regions of each
LC domain. Therefore, the disordered phases of He1 13-5 penetrated into the DPPC
ordered domains and promoted their nucleation.

Fig. 7.8. FM images of the DPPC/Hel 13-5 mixture system at 298.2 K for surface pressures of 15 and 20
mN/m: (a) pure DPPC; (b)XMII3.+= 0.005; (c) X, , ~ =
+ 0.025. In the co-existencephase, percentagerefers
to the ordered domains in the micrcgraph.The monolayerscontain 1 mol%fluorexent probe (R18).The
arrow shows themoth-eatenaggregation of LC domains made of pure DPPC.The scale bar in the lower
right represents 100pm.

170 0 H. Nakahara et al.

M M Obsewations
AFM images (500 nm x 500 nm) ofa LB film ofDPPC films containingsmall amounts
13-5 = 0.1) transferred onto mica at 35,45, and 5 5 mN/m are shown
of He1 13-5 (XHe,
in Fig. 7.9. Pure DPPC monolayers provided a homogeneous AFM image (Nakahara
13-5 = 0.1) at 35
et al., 2005a). In contrast, the monolayer containing He1 13-5 (XHe,
mN/m shows two different phases as shown in Fig. 7.9a. Because diacyl chain lengths
in DPPC molecules are ~ 2 . 5nm, bright phases represent DPPC monolayer while
dark phases indicate He1 13-5 monolayer. The height difference of these phases is = 1.O
nm, demonstrating that experimental values agree with theoretical ones.

Fig. 7.9. AFM images of the DPPC monolayer containing moderately low amounts of He1 13-5 (at X,,
,3.5 = 0.1) in a tapping mode at the scan area of 500 x 500 nm: topography (a) and corresponding phase
contrast image (b) transferred onto mica at 35 mN/m; topography (c) and corresponding phase contrast
image (d)transferred onto mica at 45 mN/m; topography (e) and correspondingphase contrast image (f)
transferred onto mica at 55 mN/m.The ratio (percent)of occupied areas by DPPC monolayersto whole
images is shown in (a), (c), and (e). The arrow (c) shows the brightest lobes by the extracted particles.

Influence of Pulmonary Surfactant Protein Mimicson Model Lung Surfactant 0

171

The corresponding phase contrast image (Fig. 7.9b) shows the morphological
character more clearly due to the difference in the surface physicochemical property
between DPPC and He1 13-5. The behavior shown in these AFM images agrees with
the FM images that the way to disperse the DPPC-ordered domains is to add a small
amount of He1 13-5. O n the other hand, three different phases appear at 45 mN/m,
where He1 13-5 is extracted from binary DPPC/Hel 13-5 monolayers (Fig. 7.9~).The
brightest lobes (indicated by an arrow) by the extracted particles appear. Interestingly,
most of the He1 13-5 protrusions sit on DPPC monolayers. The height of protrusions
from the DPPC monolayer is sl.5 nm, which corresponds to that of one and one-half
a-helical He1 13-5 molecules (= 1 nm). The corresponding phase contrast image (Fig.
7.9d) clearly shows morphological changes between respective monolayers.
Upon further compressionto 55 mN/m, the topography image (Fig. 7.9e) indicates
that DPPC regions and the protrusions increase in number, while the He1 13-5 rich
regions decrease in comparison with those in Fig. 7.9a and 7.9~.This indicates that the
ratio of DPPC-occupied areas increases as the surface pressure increases from 35-55
mN/m. That is, the content of ordered domain is increased by compression. The
percentage of the ordered domain becomes 46,66, and 78%, respectively, for surface
pressures of 35, 45, and 55 mN/m. This result demonstrates that only He1 13-5 is
extracted from binary monolayers beyond the plateau regions on the rr-A isotherms
and that the extracted He1 13-5 molecules occupy 3-D surface-associated reservoirs.
As a result, the surface is refined to a DPPC monolayer. There is no difference in
the height and size of the protrusions between 45 and 55 mN/m. The diameter of
disk-like protrusions is found to be 4 . 6 nm, corresponding to 10 molecules of He1
13-5. Note that all of the protrusions are located on DPPC monolayers, whereas
the He1 13-5 molecules that are still not extracted exist as a monolayer regardless
of the high surface pressure ( 5 5 mN/m), indicating that all He1 13-5 molecules are
not completely extracted from the binary system. This supports the result from the
quantitative analysis based on rr-A isotherms. In the phase contrast image (Fig. 7.90,
the protrusions become indistinct. This reason and mechanism have been discussed
by others (Krol et al., 2000; Nakahara et al., 2006b).

Cyclic Compression and Eapansa'on Isonberms

Cyclic compression-expansion isotherm experiments (known as hysteresis curves) of
the DPPC/Hel 13-5 (XHd
,3-5 = 0.1) system are shown in Fig. 7.10. The films were
compressed up to a surface pressure of 55 mN/m and then returned to the starting
conditions. In general, surfactants would be continuously exhausted at the interface
during the compression and expansion cyclic processes. However, these cycles did
not lead to loss of materials from the surface. The first kink point became unclear
by the cyclic processes from the first to the fifth cycle in Fig. 7.10. The second sharp
bend, however, remained clear despite repeated cycling processes, indicating that He1
13-5 was desorbed from the interface. Although the first cycling z-A isotherm almost
coincided with the isotherm for fifth cycle at low surface pressures, the rr-A isotherms

172 0 H. Nakahara et al.

shifted to smaller areas in the repeated cycling processes at high surface pressures.
Figure 7.10 indicates that the easy re-spreading behavior displays the ability for
desorbed molecules to re-enter into the interface and works for a highly reproducible
hysteresis loop. 'These results demonstrate that He1 13-5 can accelerate the spreading
of DPPC and induce good re-spreading. Furthermore, these hysteresis curves resemble
those of the DPPC/SP-B and DPPC/SP-C mixtures that were reported previously
(Wustneck, eta]., 2001,2002).

Ternary DPPC/PG/Hell3-5 and DPPC/PA/Hell3-5 Systems

Langmuir Isotherms
Typical isotherms for the fixed composition of DPPC/PG (68/22, by weight) and
DPPC/PA ( 9 0 0 , by weight) systems with He1 13-5 peptides have been investigated
to assess each role of the PG and PA components in the ternary systems. For this
purpose, the z-A and AV-A isotherms of the three component monolayers were

70

60
7

50

E
z 40
E

-%

30

20

10
0

0.4

0.6
0.8
1
1.2
Area per molecule I mi2

Fig. 7.10. Cyclic compression and expansion isotherms (or hysteresis curves) of the DPPC/Hel 13-5 (X,,,,
13-5 = 0.1) mixture on a 0.02 MTris buffer solution (pH 7.4) with 0.13 M NaCl at 298.2 K.The compression
and expansion cycle was repeated five times at a compression rate of 0.1-0.2 nm2molecule-' m i d .

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

173

13-5) in Fig. 7.1 1 and 7.12. For the

measured at various He1 13-5 molar fractions (XHCI
DPPCIPGIHel 13-5 system, as shown in Fig. 7.1 1, the n-A isotherms exhibit two
plateau regions at =21 and =42 mN/m; they are indicated by arrows at X,, 13-5 = 0.1
(curve f). All of the n-A isotherms have the first plateau at the same surface pressure
(transition pressure, f l = 521 mN/m), and the sharp bend become less visible with
increasing X,,
The second sharp bend at X,, I E 5 = 0.1 appears at -42 mN/m
(or the collapse pressure of He1 13-5), and the second plateau range expands as the
amount of He1 13-5 increases. The DPPC/PG monolayer forms a stable film up to
=45 mN/m. O n the other hand, the ternary DPPClPGlHel 13-5 monolayers (0 <
XHC1
1E5s 0.1) can retain stable states up to 155 mN/m (see the inset in Fig. 7.1 1). This
is the first evidence that He1 13-5 interacts with the fluid component (PG), and the
two fluid components begin to be extracted from the ternary mixtures at 542 mN/m.
Then the degree of extraction is gradually increased upon Further compression, and
the interface is ultimately refined to DPPC-rich monolayers at =55 mN/m because
the collapse pressure of DPPC monolayers is =55 mN/m (Fig. 7.7). At =42 mN/m,
the AV-A isotherm at XHe,
13-5 = 0.1 shows a sharp bend corresponding to the second
plateau on its n-A isotherm, as indicated by an arrow, and then the AVvalue gradually
rises with Further compression, indicating that monolayer orientations change due to
the exclusion of fluid materials. Finally, it becomes constant above 55 mN/m. This
also supports the extraction phenomena of He1 13-5.
For the DPPC/PA/Hel 13-5 system shown in Fig. 7.12, n-A isotherms also have
two plateau regions at 910-14 mN/m and =42 mN/m. The n-A isotherm at X,, 13-5
= 0.005 has only the first plateau. On the other hand, two plateaus are observed in
the n-A isotherms of 0.01 c X,,,13-5 c 0.1. The transition pressures of the first plateau
change from = 11 to = 14 mN/m and become less visible with increasingXHd which
is the same as the DPPC/PG/Hel 13-5 system. The second sharp bend in 0.01 < X,, 13-5
cO.1 also appears at 242 rnN/m, and the second plateau range elongates as the amount
of He1 13-5 increases. This behavior is the same as that of the DPPC/PG/Hel 13-5
system. However, the DPPC/PA/Hel 13-5 system displays a jagged second plateau,
revealing that the He1 13-5 species is not easily extracted from the ternary monolayers
because of the absence of fluid components. In contrast to the collapse behavior of the
DPPC/PG/Hel 13-5 system, the DPPC/PA monolayer forms a stable film up to 4 0
mN/m, and all of the DPPC/PA/Hel 13-5 monolayers (0 < X,,
c 0.1) also remain
stable up to 560 mN/m, despite adding He1 13-5 to the DPPC/PA monolayer (see
the inset in Fig. 7.12). This also reveals that He1 13-5 interacts with rigid components
(DPPC and PA), even at high surface pressures, and only He1 13-5 is gradually
extracted from the ternary monolayers above =42 mN/m. That is, the extraction
slowly starts upon hrther compression, and the interface becomes rich in the DPPC/
PA monolayer at =60 mN/m. At more than 42 mN/m, the AV-A isotherms within
0.01 c X,, c 0.1 indicate the same behavior as that of the previously mentioned
DPPC/PG/Hel 13-5 system.

174 0 H. Nakahara et al.

z,

Area per molecule / nmL

Fig. 7.1 1. n-A and AV-A isotherms of the ternary DPPC/PG/Hel 13-5 (fixed DPPC/PG ratio) mixtures on
a 0.02 MTris buffer solution (pH 8.4) with 0.13 M NaCl at 298.2 K. (Inset) Enlarged n-A isotherms at high
surface pressures. Two sharp bends are indicated by arrows on the n-A isotherm of representative XHe,
,3.5 = 0.1. The sharp bend corresponding to the second plateau on its n-A isotherm is also shown by an
arrow on the correspondingAV-A isotherm.

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

175

FM ObservationsJbrthe DPPUPMHel13-5 System

An FM observation was carried out to understand the morphological effects of He1
13-5 on binary DPPC/PA monolayers. Figure 7.13 shows FM images of ternary
DPPCIPAIHel 13-5 monolayers at X,, ,3-5 = 0.01. Two-phase coexistence states
similar to the resultant images in the DPPC/PA monolayer system are also clearly
observed (Nakahara et al., 2006a). They indicate a homogeneous disordered phase
at 10 mN/m and disordered/ordered coexistence phases at 15-30 mN/m. However,
the image at 30 mNlm displays an almost ordered phase. With increasing surface
pressure from 15 to 20 mN/m, ordered domains grow larger: the ordered percentage
increases from 57 to 65%. The micrographs at 35 and 40 mN/m show the dark
homogeneous images. Once the micrographs became dark throughout, the dark
images remained, as is the case in the DPPC/PA system. Generally, aggregations of
fluorescent probes induces self-quenching, and the probe cannot fluoresce. From the
image at 45-55 mN/m, the micrographs no longer quench, and the contrast gradually
become clearer. This specific behavior reflects the extraction of He1 13-5 between
40 and 45 mN/m. In addition, the micrographs at 20 and 55 mN/m have the same
sized ordered domains. However, smaller disordered regions are observed in the FM
image at 55 mN/m; ordered domains in the image at 55 mN/m are more packed,
supporting the extraction of He1 13-5. More detailed mechanisms were discussed and
referred to in a previous paper (Nakahara et al., 2006a).
Hystmsir Curves
The repeated compression-expansion cycling z-A and A V-A isotherms of ternary
DPPC/PG/Hel 13-5 (A) and DPPC/PA/Hel 13-5 (B) systems at X,, ,3-5 = 0.1 are
shown for the first and fifth rounds in Fig. 7.14. These two systems were compressed
up to a surface pressure of =55 mN/m and then expanded to the respective starting
areas. In Fig. 7.14a, the second sharp bend clearly appears, despite the repeated
cycling processes, indicating that He1 13-5 is reversibly desorbed into the subphase
in the z-A isotherms. In Fig. 7.14b, for z-A isotherms, the second kink shoulder for
repeated hysteresis curves clearly appears without becoming unclear as it does in Fig.
7.14a The surhce pressure decreases sharply like a solid film in the expansion process.
Consequently, Fig. 7.14 indicates good re-spreading behavior from the first to fifth
rounds, showing the ability of the desorbed molecules to relocate to the interface.
These results demonstrate that small amounts of He1 13-5 peptides can accelerate the
good re-spreading of rigid components (DPPC and PA).

Multicomponent DPPC/PG/PA/Hel13-5 Systems and Surfactents

Langnuir Isotherms
The isotherms for the fixed composition of DPPC/PG/PA (68/22/9 by weight)
with He1 13-5 peptides were investigated to determine the optimal amount of
He1 13-5 to add to the DPPC/PG/PA mixture and to veritjr whether the synthetic
preparations have effective pulmonary functions by in vitro examinations. The z-

176 0 H. Nakahara et al.

Area per molecule / nmL

Fig. 7.12. n-A and AV-A isotherms of the ternary DPPUPNHel 13-5 (fixed DPPC/PA ratio) mixtures on
a 0.02 MTris buffer solution (pH 8.4) with 0.13 M NaCl at 298.2 K. (Inset) Enlarged n-A isothermsat high
surface pressures.

Influenceof Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

177

Flg.7.13.FM micrographsoftheDPPUPA/Hell3-5mixturesystematX,,,,,=O.Ol
frornn= 10-55 mN/m.
In the coexistence phase, the percentage refers to ordered domains in the micrograph.Themonolayers
contain 1 mol%of fluorescent probe (R18).Thescale bar in the lower right represents 100pm.

178 0 H. Nakahara et al.

Area uor molecule / nmz

Area per molecule / nni2

Fig. 7.14. Cyclic compression and expansion isotherms of (A) DPPC/PG/Hel 13-5 and (B) DPPC/PA/Hel
13-5 at&-, ,)-5 = 0.1 on a 0.02 M Tris buffer solution (pH 8.4) with 0.1 3 M NaCl at 298.2 K.The compression
and expansion cycle was performed five times at the compression rate of ~ 0 . 1nm2molecule-' min-'.

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant0 179

and A V-A isotherms of the multicomponent monolayers (DPPC/PG/PA/Hel 13-5)

were measured at various He1 13-5 mole fractions (XHe,
13-5) (Fig. 7.15). All of the
z-A isotherms have the transition pressures ( f i q ) , and one of them is indicated by a
straight-line arrow (XHe,= 0.1). The f l increases slightly with increasing X,,,13+.
The increase means that lipid monolayers are difficult to transform from LE to LC
states due to the addition of He1 13-5. This behavior also occurred in the DPPC/PA/
He1 13-5 system (Nakahara et al., 2006a). At 542 mN/m, where He1 13-5 starts to
change from monolayer states to other states, the z-A isotherms ofXHel13-5 = 0.025 to
0.1 exhibit plateau regions typically indicated by dashed arrows (curves e and f). These
plateaus on the z-A isotherms correspond to the extraction of He1 13-5 from surface
monolayers. That was previously supported by AFM measurements (Nakahara et al.,
2006b). The plateau regions in the present system are almost parallel to the abscissa
of area compared with those in the previous systems. In particular, the slope of the
plateau region at XHe,
117-5 = 0. I (curve f) is nearly zero on the lateral compression from
0.5 down to 0.3 nm The slope difference between the present mixtures and the
previous ternary systems shows that the present preparations generate a critical effect
on the extraction. The smaller slope implies that the extraction of He1 13-5 with PG
preferentially occurs at the expense of improving the orientation and packing of the
DPPC/PA surface monolayers in the plateau region during the compression process.
O n further compression, the surface pressures rapidly increase until monolayer
collapse, resulting in the production of close-packed monolayers (DPPC/PA) at the
interface. As for the monolayer collapse, the DPPC/PG/PA/Hel 13-5 and DPPC/
PAlHel 13-5 systems indicate similar f values at X,, 13-5 = 0.05 and 0.1. This also
supports that the mixed DPPC/PA monolayers dominate the aidliquid interface at
higher surface pressures in the plateau region (Nakahara et al., 2006a, 2008).
For the surface potentials, the AV-A isotherms change with X,, 13 at low surfice
pressures. However, the inclination of curves e and f a t the onset o t t h e extraction
starts to vary and then remains nearly zero (indicated by dashed arrows). These AV
results support the dominant exclusion of He1 13-5 with PG from surface monolayers
(at the plateau region) at the expense of improving the orientation of the binary
DPPC/PA monolayer. O n further compression, all of the AV-A isotherms converged
to 5460 mV, suggesting that the close-packed DPPC/PA monolayers are formed in
the high surface pressure region. These AV results also support the observations for
the corresponding z-A isotherms. These isotherms demonstrate that the preparations
in the present study take on better pulmonary functions (the extraction, high collapse
pressure, and so on) and require smaller amounts of He1 13-5 for wider plateau lengths
on the z-A isotherm than the previous ternary systems.

FM Observations
Shown in Fig. 7.16 are the selected FM micrographs of the DPPC/PG/PA/Hel
13-5 preparations. For XHe,
13-5 = 0.01 (Fig. 7.lba), where there is no plateau region
corresponding to the extraction of the P A isotherm, two-phase coexistence states

180 0 H.Nakahara et al.

z
E

ires per molecule / nm2

Fig. 7.15. n- and AV-A isotherms of the DPPC/PG/PA/Hel 13-5 preparations with a fixed DPPC/PG/PA
ratio on a 0.02 M Tris buffer solution (pH 7.4) with 0.1 3 M NaCl at 298.2 K. The transition pressure for X,,
13.5 = 0.1 is indicated by a straight-line arrow. The plateau regions displaying the extraction are shown by
dashed arrows on the n-A isotherms of X,, 13.5 = 0.05 and 0.1. (Inset) Enlarged x-A isotherms show the
plateau region.

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0 181

similar to the resultant images in the lipid mixture (Nakahara et al., 2008) are
observed. The FM image at 30 mN/m indicates a clear contrast between ordered
and disordered domains. With an increase in surface pressure, however, the images
become slightly unclear due to the quenching of FM probes. Moreover, the percentage
of ordered domains doess not vary in spite of an increase in surface pressure and show
the formation of phase-separated films at high surface pressures. For X,, ,s5 = 0.05
(Fig. 7.16b), at which the n-A isotherm has a plateau, FM images indicate a different
behavior from those of X,, ,3-5 = 0.01, Although a completely dark image appears
at 40 mN/m due to the quenching, the image at 50 mN/m no longer suffers from
quenching and shows better contrast. The recovery of FM contrast results from the
extraction of He1 13-5 with fluid components including PG and FM probes in the
plateau region, with the surface concentration of FM probes decreasing. This behavior
was previously observed in the DPPC/PA/Hel 13-5 systems as a refluorescence
phenomenon, and provides novel morphological evidence of He1 13-5extraction
events (Nakahara et al., 200Ga).

HysteresJrjcurves
Repeated cycling x- and AV-A isotherms of the selected DPPC/PG/PA/Hel 13-5
preparations for first and fifth rounds are shown in Fig. 7.17. These monolayers are
compressed to a surface pressure of =62 mN/m and then expanded to the respective

Fig. 7.16. FM micrographs of the DPPC/PG/PA/Hell3-5 preparations at X,,,,, = 0.01 (a)and 0.05 (b)on a
0.02 MTris buffer solution (pH 7.4) with 0.13 M NaCl at 298.2 K. In the coexistencephases,the percentage
(96)refers to ordered domains in the micrograph. The monolayers contain 1 mol %of fluorescent probe

(R18).The scale bar in the lower right represents 100gm.

182 0 H. Nakaharaet al.

starting molecular areas. Both of z-A isotherms have large hysteresis in area and
indicate good reproducibility during cycling. Afier reaching the maximum surface
pressure on compression, the surface pressure rapidly decreases by 120 mN/m by a
small expansion, suggesting that the extracted He1 13-5 instantly reenters the surface
and then disperses the close-packed monolayers (DPPC/PA) during expansion. The
ability of He1 13-5 to respread with PG in the present preparations was improved
in terms of the rapid reentry to the surface in comparison with that in the previous
systems (Nakahara et al., 2005b, 2006b). Furthermore, reproducible plateaus at 42
mN/m demonstrate that the exclusion and the successive inclusion of He1 13-5 are
a reversible process during cycling. Consequently, these actions during compressionexpansion cycling demonstrate that the preparations have sufficient pulmonary
functions in the basic in vitro level (Notter, 2000).
Comparison to Surj4acten (Surj4actant TA) in Hysteresis Measurements
The cyclic compression and expansion isotherms for Surfacten are shown in Fig.
7.18a. The z-A isotherms have large hysteresis area and good reproducibility during
the cyclic process. After reaching the maximum surface pressure on compression, the
surface pressure results in a rapid reduction of =15 mN/m by a small expansion. The
isotherms of the first compression cycle for X,, 13-5 = 0.05 and 0.1, and Surfacten
are superimposed in Fig. 7.18b. The hysteresis area on z-A isotherms for XHe,
13-5
= 0.1 is the largest of the three preparations by an integrated area; this implies that
the preparation (XHel
13-5 = 0.1) is more capable of respreading than Surfacten. In
addition, an analogous trend in area and profile between XHe,
,3-5 = 0.05 and Surfacten
is observed for z-A isotherms. These results demonstrate that when He1 13-5 is mixed
with the DPPC/PG/PA lipid mixture, it has similar capabilities to SP-B and SP-C,
such as rapid respreading and adsorption through the interface.

Concl usions
The present work is meant to investigate the interfacial behavior of a synthetic peptide
(He1 13-5), a mimic peptide of human surfactant protein B, in a phospholipid
and various lipid mixtures. He1 13-5 itself is an amphiphilic peptide and can form
an a-helical structure at the air-water interface. DPPC is a main phospholipid
component in PS and contributes to lowering surface tension during exhalation.
DPPC alone has some defects in pulmonary functions such as rapid respreading,
adsorption, and large hysteresis. However, addition of a small amount of He1 13-5
to DPPC led to improved adsorption of DPPC. Moreover, the addition resulted in
enlargement of reversible hysteresis loops of z-A curves for DPPC. These results
support the possibility of He1 13-5 as a substitute for SP-B (Nakahara et al., 2005b,
2006b). In addition, both PG and PA components prevented the DPPC monolayer
from collapsing after the extraction of He1 13-5. PG is thought to interact selectively
and electrostatically with cationic He1 13-5 during the extraction process; then 3-D
aggregates containing He1 13-5 and PG could be formed just below the surface

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0

183

'E

Area per molecule I nm2

2:

Area per molecule / nm'

Fig. 7.17. Cyclic compression and expansion isotherms of the DPWPG/PA/Hel 13-5 preparations at
X,, 13-5 = 0.05 (A) and 0.1 (B) on a 0.02 M Tris buffer solution (pH 7.4) with 0.13 M NaCl at 298.2 K. The
compression-expansioncycle was performed five times at the compression rate of 50.17 nml moleculelmin-l.

184 0 H. Nakahara et al.

5-r

Trough area I YO
Fig. 7.18. Cyclic compression and expansion isotherms of Surfacten (A) on a 0.02 M Tris buffer solution
(pH 7.4) with 0.1 3 M NaCl at 298.2 K. Hysteresis curves for the first round for the DPPC/PG/PA/Hel 13-5
preparations at X,,,,,, =0.05 and 0.1, and Surfacten (B).The compression-expansioncycle was performed
five times at the compression rate of 390 cm2min-. The area (abscissa axis) represents the relative area
(%) to the initial surface area prior to compression.

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0 185

monolayer (Diemel et al., 2002; Ding et al., 2003; Alonso et al., 2004; Nakahara et
al., 2006b). This formation contributes to the stability of DPPC monolayers. On the
other hand, PA components are not mainly extracted from surface monolayers, and
form a close packed monolayer with DPPC at the interface (Nakahara et al., 2006a).
Finally, various kinds of measurements for surface chemistry (Langmuir isotherms,
fluorescent images, temperature dependence, and hysteresis curves) have been made
for the comparison between the DPPCIPGIPAIHd 13-5 preparations and Surfacten,
which is commercially used for NRDS patients in Japan (Nakahara et al., 2008).
These results suggested that the preparations are comparable to Surfacten in terms
of surface activity and the interfacial behavior. The present work on PS substitutes
containing a synthetic peptide provides a deep insight into the clarification of
functional mechanisms of PS near the air/alveolar liquid interface. At the same time,
it will contribute to tailoring surfactant preparations to various states of surfactant
deficiency, insufficiency, and inactivation.

Acknowledgment
This work was supported by a Grant-in-Aid for Scientific Research 20500414 from
the Japan Society for the Promotion of Science (JSPS).

References
Alonso, C.; T. Alig J. Yoon; F, Bringezu; H. Warriner; J.A. Zasadzinski, More than a Monolayer:
Relating Lung Surfactant Structure and Mechanics to Composition. Biophys. J 2004,87,
4188-4202.
Amrein, M.; A. von Nahmen; M. Sieber, A Scanning Force- and Fluorescence Light Microscopy
Study of the Structure and Function of a Model Pulmonary Surfictant. Eur. Biophys. /. 1997,
26,349-357.
Anderson, M.; T.Curstedt; H. Jornvall; J. Johansson, Amphipathic Helical Motif Common to
Tumourolytic Polypeptide NK-Lysin and Pulmonary Surfactant Polypeptide SP-B. FEBS Len.
1995,362,328-332.
Avery, M.E.; J. Mead, Surface Properties in Relation to Atelectasis and Hyaline Membrane Disease. Am. J. Dis. Cbiki. 1959,97,517-526.
Baatz, J.E.; B. Elledge; J.A. Whitsett, Surfictant Protein SP-B Induces Ordering at the Surfice of
Model Membrane Bilayers. Biocbem. 1990,29,6714-6720.
Bastacky, J.; C.Y. Lee; J. Goerke; H. Koushafir; D. Yager; T.P. Speed; Y. Chen; J.A. Clements, Alveolar Lining Layer Is Thin and Continuous: Low-Temperature Scanning Electron Microscopy
of Rat LungJ AppL PbysioL 1995,79,1615-1628.
Benson, B.; S. Hawgood; J. Schilling; J. Clements; D. Damm; B. Cordell; RT. White, Structure
of Canine Pulmonary Surfactant Apoprotein: cDNA and Complete Amino Acid Sequence. Proc.
NatL Acad Sci. USA 1985,82,63794383.
Boggaram, V.;K. Qing; C.R Mendelson, The Major Apoprotein of Rabbit Pulmonary Surfactant. Elucidation of Primary Sequence and Cyclic AMP and Developmental Regulation. J Biol.
Cbem. 1988,263,2939-2947.

186 0 H. Nakaharaet al.

Brown, E.S., Isolation and Assay of Dipalmityl Lecithin in Lung Extracts. Am. J Pbysiol. 1964,
207,402-406.
Cai, I?; C.R. Flach; R. Mendelsohn, An Infrared Reflection-Absorption Spectroscopy Study of
the Secondary Structure in (KLJ4K, a Therapeutic Agent for Respiratory Distress Syndrome, in
Aqueous Monolayers with Phospholipids. Biochem. 2003,42,9446-9452.
Cajal, Y.; C. Dodia; A.B. Fisher; M.K. Jain, Calcium-Triggered Selective Intermembrane Exchange
of Phospholipids by the Lung Surfactant Protein SP-A. Biochem. 1998,37, 12178-12188.
Chang, R.; S. Nir; F.R. Poulain, Analysis of Binding and Membrane Destabilization of Phospholipid Membranes by Surfactant Apoprotein B. Biochim. Biopbys. Acta 1998, 1371,254-264.
Cockshutt, A.M.; D.R. Absolom; F. Possmayer, The Role of Palmitic Acid in Pulmonary Surfactant: Enhancement of Surface Activity and Prevention of Inhibition by Blood Proteins. Biochim.
Biopbys. Acta 1991, 1085, 248-256.
Crouch, E.; A. Persson; D. Chang; D. Parghi, Surfactant Protein D. Increased Accumulation in
Silica-Induced Pulmonary Lipoproteinosis. Am. /.Pathol. 1991, 139, 765-776.
Crouch, E.C., Collectins and Pulmonary Host Defense. Am. J Respir. CellMol. Biol. 1998, 19,
177-20 1.
Curstedt, T.; J. Johansson; I? Persson; A. Eklund; B. Robertson; B. Uwenadler; H. Jornvall,
Hydrophobic Surfactant-Associated Polypeptides: SP-C Is a Lipopeptide with Two Palrnitoylated
Cysteine Residues, Whereas SP-B Lacks Covalently Linked Fatty Acyl Groups. Proc. Natl. Acad
Sci. US4 1990,87, 2985-2989.
Diemel, R.V.; M.M.E. Snel; A.J. Waring; EJ. Walther; L.M.G. van Golde; G. Putz; H.I? Haagsman; J.J. Batenburg, Multilayer Formation upon Compression of Surfactant Monolayers
Depends on Protein Concentration as Well as Lipid Composition. An Atomic Force Microscopy
Study.j. Biol. Chem. 2002,277, 21 179-21 188.
Ding, J.; D.Y. Takamoto; A. Von Nahmen; M.M. Lipp; K.Y.C. Lee; A.J. Waring; J.A. Zasadzinski,
Effects of Lung Surfactant Proteins, SP-B and SP-C, and Palmitic Acid on Monolayer Stability.
Biopbys. J 2001,80,2262-2272.
Ding, J.; I. Doudevski; H.E. Warriner; T. Alig; J.A. Zasadzinski; A.J. Waring; M.A. Sherman,
Nanostructure Changes in Lung Surfactant Monolayers Induced by Interactions between Palmitoyloleoylphosphatidylglyceroland Surfactant Protein B. Langmuir 2003,19, 1539-1 550.
Fleming, B.; K.M.W. Keough, Surface Respreading after Collapse of Monolayers Containing
Major Lipids of Pulmonary Surfactant. Chem. Phys. Lipid 1988,49, 81-86.
Floros, J.; R. Steinbrink; K. Jacobs; etc., Isolation and Characterization of cDNA Clones for the
35-kDa Pulmonary Surfactant-Associated Protein. /. Biol. Chem. 1986,261, 9029-9033.
Furuya, T.; T. Kiyota; S. Lee; T. Inoue; G. Sugihara; A. Logvinova; I? Goldsmith; H.M. Ellerby,
Nanotubules Formed by Highly Hydrophobic Amphiphilic a-Helical Peptides and NaturaJ
Phospholipids. Biopbys. J 2003,84, 1950-1959.
Glasser, S.W.; T.R. Korfhagen; T Weaver; T. Pilot-Matias; J.L. Fox; J.A. Whitsett, cDNA and Deduced Amino Acid Sequence of Human Pulmonary Surfactant-Associated Proteolipid SPL(Phe).
Proc. Natl. Acad. Sci. USA 1987,84,4007-40 1 1 .
Goerke, J., Lung Surfactant. Biochim. Biophys. Acta 1974,344, 241-261.
Gorree, G.; J. Egberts; G. Bakker; A. Beintema; M. Top, Development of a Human Lung Surfac-

influenceof Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0 187

tant, Derived from Extracted Amniotic Fluid. Biochim. Biopbys. Acta 1991, 1086,209-216.
Grigoriev, D.O.; J. Kragel; A.V. Akentiev; B.A. Noskov; R. Miller; U. Pison, Relation Between
Rheological Properties and Structural Changes in Monolayers of Model Lung Surfactant under
Compression. Biophys. Chcm. 2003, 104, 633-642.
Gustafsson, M.; G. Vandenbussche; T. Curstedt; J.M. Ruysschaert;J. Johansson, The 21-Residue
Surfactant Peptide (LysLeu,),Lys(KL,) is a Transmembrane Alpha-helix with a Mixed Nonpolar/
Polar Surface. FEBS Letten 1996,384, 185-188.
Gustafsson, M.; M. Palmblad; T. Curstedt; J. Johansson; S. Schurch, Palmitoylation of a Pulmonary Surfactant Protein C Analogue Affects the Surface Associated Lipid Reservoir and Film
Stability. Biochim. Biophys. Acta 2000, 1466, 169-178.
Hawgood, S.; B.J. Benson; J. Schilling; D. Damm; J.A. Clements; R.T. White, Nucleotide and
Amino Acid Sequences of Pulmonary Surfactant Protein SP 18 and Evidence for Cooperation
between SP 18 and SP 28-36 in Surfactant Lipid Adsorption. Proc. Natl. Acad. Sci. USA 1987,
84, 66-70.
Hawgood, S.; K. Shiffer, Structures and Properties of the Surfactant-Associated Proteins. Ann. Rev.
Pbysioi. 1991,53,375-394.
Johansson, J.; H. Jornvall; A. Eklund; N. Christensen; B. Robertson; T. Curstedt, Hydrophobic
3.7 kDa Surfactant Polypeptide: Structural Characterization of the Human and Bovine Forms.
FEBS Len. 1988,232,6144,
Johansson, J.; T Curstedt; H. Jornvall, Surfactant Protein B: Disulfide Bridges, Structural Properties, and Kringle Similarities. Biochcm. 1991,30, 6917-6921.
Johansson, J.; T. Curstedt, Molecular Structures and Interactions of Pulmonary Surfactant Components. Eur. J. Biochcm. 1997,244,675-693.
Johansson, J., Structure and Properties of Surfactant Protein C. Biochim. Biopbys. Acta 1998,
1408, 161-172.
Kirkness, J.I?; I?R. Eastwood; I. Szollosi; RR Platt; J.R. Wheatley; T.C. Amis; D.R Hillman,
Effect of Surface Tension of Mucosal Lining Liquid on Upper Airway Mechanics in Anesthetized
Humans. J. AppL Pbysio.! 2003a, 95,357-363.
Kirkness, J.I?; M. Madronio; R. Stavrinou; J.R. Wheatley; T.C. Amis, Relationship Between Surface Tension of Upper Airway Lining Liquid and Upper Airway Collapsibilityduring Sleep in
Obstructive Sleep Apnea Hypopnea Syndrome.J. Appl. Pbysiot! 2003b, 95, 1761-1766.
Kitamura, A.; T. Kiyota; M. Tomohiro; A. Umeda; S. Lee; T.Inoue; G. Sugihara, Morphological
Behavior of Acidic and Neutral Liposomes Induced by Basic Amphiphilic a-helical Peptides
with SystematicallyVaried Hydrophobic-Hydrophilic Balance. Biopbys. J. 1999, 76, 1457-1468.
Kiyota, T.; S. Lee; G. Sugihara, Design and Synthesis of Amphiphilic a-Helical Model Peptides
with SystematicallyVaried Hydrophobic-Hydrophilic Balance and Their Interaction with Lipidand Bio-Membranes. Biochcm. 1996,35, 13196-13204.
Krol, S.; M. Ross; M. Sieber; S. Kunneke; H.-J. Galla; A. Janshoff, Formation of Three-Dimensional Protein-Lipid &regates in Monolayer Films Induced by Surfactant Protein B. Biophs. 1.
2O00,79,904-9 18.
Kriiger, I?; J.E. Baatz; RA. Dluhy; M. Usche, Effect of Hydrophobic Surfactant Protein SP-C
on Binary Phospholipid Monolayers. Molecular Machinery at the Airwater Interface. Biopbys.

188 0 H. Nakahara et al.

Chem. 2002,99,209-228.
Kuroki, Y.;M. Shiratori; Y. Ogasawara; A. Tsuzuki; T. &no, Characterization of Pulmonary Surfactant Protein D: Its Copurification with Lipids. Biochim. Biophys. Acta 1991, 1086, 185-190.
Lee, S.; T. Furuya; T. Kiyota; N. Takami; K. Murata; Y. Niidome; D.E. Bredesen; H.M. Ellerby;
G. Sugihara, De Novo-Designed Peptide Transforms Golgi-SpecificLipids into Golgi-Like
Nanotubu1es.J Biol. Chem. 2001,276,4122441228.
Lee, S.; G. Sugihara; 0. Shibata; H. Yukitake JP Patent Appl. P 2004-305006A,2004.

Lipp, M.M.; K.Y.C. Lee; D.Y. Takamoto; J.A. Zasadzinski; A.J. Waring, Coexistence of Buckled
and Flat Monolayers. Phys. Rev. Lett. 1998,81,1650-1653.
Losche, H.; H. Mohwald, Fluorescence Microscopy on Monomolecular Films at an Aidwater
Interface. Colbidc Surf: 1984, 10,217-224.
Ma, J.; S. Koppenol; H. Yu; G. Zografi, Effects of a Cationic and Hydrophobic Peptide, KL,, on
Model Lung Surfactant Lipid Monolayers. Biophys. J 1998, 74,1899-1907.
Magoon, M.W.; J.R. Wright; A. Baritussio; M.C. Williams; J. Goerke; B.J. Benson; R.L. Hamilton; J.A. Clements, Subfractionation of Lung Surfactant. Implications for Metabolism and
Surface Activity. Biochim. Biophys. Acta 1983, 750,18-3 1.
Mason, R.J.; K. Greene; D.R. Voelker, Surfactant Protein A and Surfactant Protein D in Health
and Disease. Am. J PhysioL 1998,275,Ll-Ll3.
Miyamura, K.;L.E. Leigh; J. Lu; J. Hopkin; A. Lopez Bernal; K.B. Reid, Surfactant Protein D
Binding to Alveolar Macrophages. Biochem. J 1994,300,237-242.
Nahmen, A.V.; M. Schenk; M. Sieber; M. Amrein, The Structure of a Model Pulmonary Surfacrant as Revealed by Scanning Force Microscopy. Biophys. J 1997, 72,463-469.
Nakahara, H.; S. Nakamura; H. Kawasaki; 0. Shibata, Properties ofTwo-Component Langmuir
Monolayer of Single Chain Perfluorinated Carboxylic Acids with Dipalmitoylphosphatidylcholine (DPPC). Colloidr Surf: B 2005% 41,285-298.
Nakahara, H.; S. Nakamura; S. Lee; G. Sugihara; 0. Shibata, Influence of a New Amphiphilic
Peptide with Phospholipid Monolayers at the Air-Water Interface. Colloidr Surf: A 2005b,
270-271,
52-60.
Nakahara, H.; S. Lee; G. Sugihara; 0. Shibata, Mode of Interaction of Hydrophobic Amphiphilic
a-Helical Peptide/Dipalmitoylphosphatidylcholinewith Phosphatidylglycerol or Palmitic Acid
at the Air-Water Interface. Langmuir 200Ga,22,5792-5803.
Nakahara, H.; S. Nakamura; T. Hiranita; H. Kawasaki; S. Lee; G. Sugihara; 0. Shibata, Mode
of Interaction of Amphiphilic a-Helical Peptide with Phosphatidylcholinesat the Air-Water
Interface. Langmuir 200Gb,22,1182-1 192.
Nakahara, H.; S. Lee; G. Sugihara; C.-H. Chang; 0. Shibata, Langmuir Monolayer of Artificial
Pulmonary Surfactant Mixtures with an Amphiphilic Peptide at the Aidwater Interface: Comparison of New Preparations with Surfacten (Surfactant TA). Langmuir 2008,24,3370-3379.
Notter, R.H.; D.P. Penney; J.N. Finkelstein; D.L. Shapiro, Adsorption of Natural Lung Surfactant and Phospholipid Extracts Related to Tubular Myelin Formation. Pediatr. Res. 198G,20,

97-101.
Notter, R.H., Lung Surfactants: Basic Science and ClinicalApplications. Marcel Dekker, Inc.: New

Influence of Pulmonary Surfactant Protein Mimics on Model Lung Surfactant 0 189

York, Basel, 2000;Vol. 147,pp. 1-444.

Oosterlaken-Dijksterhuis, M A ; H.P Haagsman; L.M.G. Van Golde; RA. Demel, Interaction of
Lipid Vesicles with Monomolecular Layers Containing Lung Surfactant Proteins SP-B or SP-C.
Biocbcm. 1991,30,827643281.
Pbrez-Gil, J.; C. Casals; D. Marsh, Interactions of Hydrophobic Lung Surfactant Proteins SP-B
and SP-C with Dipdmitoylphosphatidylcholine and Dipalmitoylphosphatidylglycerol Bilayers
Studied by Electron Spin Resonance Spectroscopy.Biocbcm. 1995,34,3764-3771.
Persson, A.; D. Chang; E. Crouch, Surfactant Protein D Is a Divdent Cation-Dependent
Carbohydrate-Binding Protein. /. Biof. Cbcm. 1990,265,5755-5760.
Piknova, B.; W.R. Schief; V. Vogel; B.M. Discher; S.B. Hall, Discrepancy Between Phase Behavior
of Lung Surfactant Phospholipids and the Classical Model of Surfactant Function. Biopbys. J

2001981,2172-2180.
Postle, AD.; E.L. Heeley; D.C. Wilton, A Comparison of the Molecular Species Compositions of
Mammalian Lung Surfactant Phospholipids. Comp. Biocbcm. PbysioL A 2001, 129,65-73.
Putman, E.; L.A. Creuwels; L.M. van Golde; H.P. Haagsman, Surface Properties, Morphology and
Protein Composition of Pulmonary Surfactant Subtypes. Biocbcm. /. 1996,320,577-605.
Revak, S.D.; T.A. Merritt; C.G. Cochrane; G.E Heldt; M.S. Alberts; D.W. Anderson; A. Kheiter,
Efficacy of Synthetic Peptide-Containing Surfactant in the Treatment of Respiratory Distress
Syndrome in Preterm Infant Rhesus Monkeys. Pediatric Rcs. 1996,39,715-724.
Ruano, M.L.; J. Perez-Gil; C. C a d s , Effect of Acidic pH on the Structure and Lipid Binding
Properties of Porcine Surfactant Protein A. Potential Role of Acidification Along Its E x q i c
Pathway.]. BioL Cbcm. 1998,273, 15183-15171.
Schurch, S.; J. Goerke; J.A.Clements, Direct Determination of Surface Tension in the Lung. Proc.
NatL Acaa! Sci. USA 1976,73,4678-4702.
Schurch, S.; F.H.Y. Green; H. Bachofen, Formation and Structure of Surface Films: Captive
Bubble Surfactometry. Biocbim. Biopbys. Acta 1998, 1408,180-202.
Serrano, A.G.; A. Cruz; K. Rodriguez-Capote; F. Possmayer; J. Perez-Gil, Intrinsic Structural and
Functiond Determinants within the Amino Acid Sequence of Mature Pulmonary Surfactant
Protein SP-B. Biocbcm. 2005,44,417-430.
Takahashi, A; A.J. Waring; J. Amirkhanian; B. Fan; H.W. Taeusch, Structure-Function Relationships of Bovine Pulmonary Surfactant Proteins: SP-B and SP-C. Biocbim. Biopbys. A m 1990,

1044,43-47.
T h o t o , D.Y.; M.M. Lipp; A. Von Nahmen; K.Y.C. Lee, A.J. Waring; J.A. Zasadzinski, Interaction of Lung Surfactant Proteins with Anionic Phospholipids. Biopbys. J 2001,81,153-167.
Tanaka, Y.; T. Takei; T. Aiba; K. Masuda; A. Kiuchi; T. Fujiwara, Development of Synthetic Lung
Surfactants.J LipidRes. 1986,27,475-485.
Taneva, S.; K.M. Keough, Pulmonary Surfactant Proteins SP-B and SP-C in Spread Monolayers at
the Air-Water Interface: 111. Proteins SP-B Plus SP-C with Phospholipids in Spread Monolayers.
Biopbys. J 1994% 66, 1 1 58-1 166.
Taneva, S.; K.M. Keough, Pulmonary Surfactant Proteins SP-B and SP-C in Spread Monolayers at
the Air-Water Interface: 11. Monolayers of Pulmonary Surfactant Protein SP-C and Phospholipids. Biopbys. J 1994b,66, 1147-1 157.

190 0 H. Nakahara et al.

Taneva, S.; K.M. Keough, Pulmonary Surfactant Proteins SP-B and SP-C in Spread Monolayers at
the Air-Water Interface: I. Monolayers of Pulmonary Surfactant Protein SP-B and Phospholipids. Biopbys. /. 1994c, GG, 1137-1 148.
Veldhuizen, E.J.A.; A.J. Waring; F.J. Walther; J.J. Batenburg; L.M.G. van Golde; H.P. Haagsman,
Dimeric N-Terminal Segment of Human Surfactant Protein B (dSP-B1-25) Has Enhanced
Surface Properties Compared to Monomeric SP-B1-25. Biopbys. 1.2000, 79,377-384.
Veldhuizen, R.; K. Nag; S. Orgeig; F. Possmayer, The Role of Lipids in Pulmonary Surfactant.
Biocbim. Bioplys. Acta 1998, 1408,90-108.
Voorhout, W.F.; T. Veenendaal; H.P. Haagsman; A.J. Verkleij; L.M.G. van Golde; H.J. Geuze,
Surfactant Protein A Is Localized at the Corners of the Pulmonary Tubular Myelin Lattice. 1.
Histocbem. Cytochem. 1991,39, 1331-1336.
Voorhout, WX; T. Veenendaal; Y. Kuroki; Y. Ogasawara; L.M. van Golde; H.J. Geuze, Immunocytochemical Localization of Surfactant Protein D (SP-D) in Type I1 Cells, Clara Cells, and
Alveolar Macrophages of Rat Lung. J. Histocbem. Cytocbem. 1992, 40,1589-1 597.
White, R.T.; D. Darnm; J. Miller; K. Spratt; J. Schilling; S. Hawgood; B. Benson; B. Cordell,
Isolation and Characterization of the Human Pulmonary Surfactant Apoprotein Gene. Nature
1985,317,361-363.
Wiistneck, N.; R. Wiistneck; V.B. Fainerman; R. Miller; U. Pison, Interfacial Behaviour and
Mechanical Properties of Spread Lung Surfactant Protein/Lipid Layers. Colloids Surj B 2001,
21, 191-205.
Wiistneck, R.; N. Wiistneck; B. Moser; U. Pison, Surface Dilatational Behavior of Pulmonary
Surfactant Components Spread on the Surface of a Pendant Drop. 2. Dipalmitoyl Phosphatidylcholine and Surfactant Protein B. Langmuir 2002, 18,1125-1 130.
Yu, S.-H.; F. Possmayer, Lipid Compositional Analysis of Pulmonary Surfactant Monolayers and
Monolayer-Associated Reservoirs.1.Lipid Res. 2003, 44,62 1-629.

Lung Surfactants : Formulation,

Evaluation, and Polymeric Additives
Edgar J. Acostalez, Sameh M.I. Saad2,Ningxi Kangl, Zdenka Policova',
Michael L. Hairz,and A. Wilhelm Neumann'
'Departmentof Chemical Engineeringand Applied Chemistv Universityof Toronto, 200 CollegeStreet,
Toronto,Ontario, Canada M5S 35; 'Deportment ofMechanical and lndustrial Engineering,Universityof
Toronto,5 King's CollegeRoad, Toronto,Ontario, Canada MSS 3G8

Introduction
This chapter opens with a brief historical review of the development of lung surfactant
therapy for the treatment of neonatal respiratory distress syndrome (nRDS), which
is now a standard therapy that has increased the survival of preterm neonates from
50% to over 80%. The most successhl formulations are extracts from lung lavages
of mammals, in particular from bovine and porcine sources. The current methods of
extraction and evaluation are also reviewed and discussed. The challenges for these
preparations include the variability of their performance, and their susceptibility to
inhibition by fluidizing lipids and proteins. To address these limitations, numerous
additives were proposed including neutral, anionic, and cationic polyelectrolytes.
Among these additives, cationic polyelectrolytes and amphiphilic cationic peptides
were among the most effective in resisting inhibitory factors. The mechanism of
action of these cationic additives on extract lung surfactants is discussed.

Historical Overview and Medical Needs for Exogenous

Lung Surfactants
Respiratory distress Syndrome and Lung Surfactants
Respiratory distress syndrome (RDS) is a clinical condition defined by the onset of
poor blood oxygenation due to lung injury or lack of lung surfactant (Petty, 2003).
RDS is typically classified into neonatal RDS (nRDS) and adult (sometimes referred
to as acute) RDS (ARDS) (Notter, 2000; Petty, 2003). About 50,000 cases of nRDS
and 150,000 ofARDS patients are diagnosed in the United States each year (McIntyre
et al., 2000; Notter, 2000; Rubenfield & Neff, 2003). Based on a recent U.S.survey,
the cost of caring for an ARDS patient averages U.S.$73,100 (Valta et al., 1999),
which represents a significant economical impact.
191

192 0 EJ. Acosta et al.

Before lung surfactant therapy became commonly used, nRDS was the leading
cause of death (>50% mortality) of premature babies with less than 29 weeks of
gestation (McDorman & Rosenberg, 1993). The development of therapeutical
surfactant formulas was a slow process. Avery and Mead, in 1959, were the first to show
that RDS is linked to an abnormal function of lung surfactant, and they proposed the
need for lung surfactant replacement therapy. The development of various formulas
is described in the next section. With the introduction of exogenous lung surfactant
formulations, the rate of the mortality of nRDS patients was reduced to approximately
20% (Hoekstra et al., 1991; Jacobs et al., 2000; McMillan et al., 1995; Petty, 2003).
Unfortunately, lung surfactant therapy was not effective in reducing the mortality of
ARDS patients (=40%), where surfactant inhibition (rather than lack of surfactant) is
associated with the pathology (Adhikari et al., 2004; Lewis et al., 2003; Notter, 2000;
Petty, 2003; Taeusch, 2000; Zasadzinski et al., 2001).

Lung Surfactants: Discovery,Physiology,and

Lung Surfactant Therapy
To clarify that a difference exists between the concept of surfactants within the
framework of lung physiology and the conventional definition of a surfactant is
important. Lung surfactants are water-insoluble, surface active agents, mainly
phospholipids (PLs), that are dispersed in water through vesicles or other similar
structures and that form adsorbed films at the air-water interface which reduce the
surface tension of the system.
The concept of lung surfactants was introduced by Pattle in 1955, who
recognized that these surfactants are a complex mixture of PLs and proteins (Pattle
& Thomas, 1961). According to Pattle and earlier studies by Von Neergaard (1929),
the surface tension (y) of the alveoli should be close to zero to avoid the collapse of
the smaller alveoli due to the difference in Laplace pressure (AP,+y/R) between the
small and large alveoli. Generally considered is that surface tensions near 1 mJ/m2 or
less are reached at the end of the expiration cycle in healthy individuals (Hall et al.,
1993; Schurch et al., 1976, 1989). Clements (1957) initiated detailed studies of the
surface properties of lung surfactants, and various reviews summarize the advances in
this field (Petty, 2003; Zasadzinski et al., 2001; Zuo et al., 2008). Figure 8.1 shows
a schematic of the lungs, including a section of the alveoli and a cross section of one
alveolus that includes a schematic of the adsorption of lung surfactant PLs in the
alveolar lining (not to scale). The thickness of the alveolar fluid (lining) ranges from
0.1 to 1 pm.
Lung surfactants are produced and recycled by Type I1 pneumocytes, and they are
composed of 85-90% of PLs, 7-10% of proteins, and 6 7 % of neutral lipids (NLs),
mostly cholesterol (Scott, 1992). ?he PLs are composed of phosphatidylcholines
(PCs, n80%), phosphatidylglycerols (PGs, =lo%), and minor proportions of
phosphatidylinositols (PIS), phosphatidylserines (PSs), phosphatidylethanolamines
(PEs), and sphingomyelins (SPHs) (Bernhard et al., 2000; Veldhuzien & Possmayer,

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0

193

Alveolar lining
layer

Lung rurfaclant
components

Fig. 8.1. Schematic of the alveoli and the distribution of lung surfactants in the alveolus (not to scale).

II

0 - IP - x

I
I

n -

0
Zwitterionic (net zero charae) DhospholiDids:
Phosphatidylcholine (PC) => X = -(CH 2)z(N+)(CH3)3
Phosphatidylethanolamine (PE) => X= -(CH2)z(N+)H3
Anionic phospholipids:
Phosphatidylglycerol (PG) => X= -CH2 -CH(OH) -CH2 -OH
Phosphatidylinositol (PI) => X= -C 6H1105
Phosphatidylserine (PSI => X= -CH2 -CH(NH2) -COOSaturated DhosDholiDids:
R1 ,R 2 :e.g. palmitic (P)(C16), steric (Cl8), etc.
Unsaturated phospholipids:
R 1,R2: e.g. oleyl (0)
(C18, one double bond), etc.
Fig. 8.2. Molecular structures of phospholipidscommonly found in lung surfactants.

194 0 EJ. Acosta et al.

2004). The general structure for the most common PLS found in lung surfactant is
presented in Fig. 8.2.
At neutral pH, PCs, PEs, and SPHs are zwitterionic, and have both negative
and positive charges within the molecule. O n the other hand, PGs, PSs, and PIS
are anionic. Dipalmitoyl-phosphatidylcholine (DPPC) accounts for nearly one-half
of PC components. When DPPC films are compressed, they form a closely packed
monolayer at the air-water interface, reducing the surface tension to values of 1 mJ1
m2 or less (Notter et al., 1980). The reduction in surface area per molecule and in
surbce tension is accompanied by a phase transition of the surfactant film from a
gaseous phase to a liquid-expanded (LE) phase, and finally to a liquid-condensed
(LC) phase (Nag et al., 1998; Taneva & Keough, 2000). In general, saturated PLs,
such as DPPC, form films capable of resisting larger compressions and reaching
lower surface tensions, and this is typically correlated with a higher melting point
of the PL (=41.5C for DPPC). According to one of the mechanisms that explain
lung- surfactant behavior, after a series of compression cycles, surfactant molecules
that are less surface active are squeezed out of the film, thus producing DPPCrich films that yield lower minimal surface tensions (Goerke, 1998; Notter et al.,
2000; Possmayer et al., 2001). Surfactant proteins have an important influence on the
physicochemical properties of the surfactant film. Four types of proteins are in lung
surfactants: SP-A, SP-B, SP-C, and SP-D (Persson et al., 1989; Possmayer, 1988).
SP-A and SP-D are hydrophilic, whereas SP-B and SP-C are hydrophobic. SP-A is the
most abundant of these proteins. It has a large molecular weight (25-38 KDa), and is
relatively acidic. SP-A contributes to rapid film formation, the formation of tubular
myelin, and it interacts mostly with the hydrophilic groups of the PLs (Creuwels
et al., 1997; McCormack, 1998; Morrow et al., 2003; Notter, 2000; Weaver &
Whitsett, 1991). SP-B has a small molecular weight (=9KDa), and it contributes
to the formation of tubular myelin and other large aggregates, contributes to film
refining, increases respreading, and interacts with the heads and tails of the PLs
(Creuwels et al., 1997; Morrow et al., 2004; Notter, 2000; Weaver &Whitsett, 1991;
Withsett et al., 1995). SP-C is the most hydrophobic of all surfactant proteins, and
contributes to softening lipid bilayers which is important for fast film formation
and film respreading. SP-D does not seem to have any impact on the physicochemical
properties of lung- surfactant films, but it plays an important role (along with SP-A)
in the immunological system (Creuwels et al., 1997; Notter et al., 2000; Weaver &
Whitsett, 1991). An important note to highlight is that SP-A and SP-D are absent
in exogenous lung surfactant formulations. The role of SP-A in exogenous lung
surfactant preparations was imitated, at least in principle, with nonionic and anionic
additives. SP-A has a collagenous domain in its tail and a carbohydrate-recognition
domain (CRD) at its head (McCormack, 2001). The CRD groups are known for
their binding capacity to PL monolayers through electrostatic interactions. This often
occurs through calcium ions (McCormack, 2001; Nag et al., 2004; Worthman et al.,
2000).

Lung Surfactants:Formulation, Evaluation, and Polymeric Additives 0 195

Four types of exogenous lung surfictant preparations are used in treating RDS:
synthetic PL mixtures (e.g., Exosurf), extracts from lung tissue (e.g., Survanta),
extracts from lung lavages (e.g., BLES, C U E ) , and whole surfactant from amniotic
fluid (Notter, 2000). The first lung surfactant formulations to be developed were
based on synthetic PL mixtures, in part to avoid an adverse immunological response to
surfactant proteins that are present in complete lung surfactants (e.g., from amniotic
fluid). In clinical trials, these formulations failed and were eventually phased out.
Assumably, lung surfactants obtained from a chloroform extraction of lung lavages
would be free of proteins; however, later it was found that, in the case of BLES they
are free of surfactant proteins SP-A and SP-D,
but they still contain SP-B and SP-C
(Smyth et al., 1983).Current therapeutical formulations are based on lung extract
from large mammals including bovine and porcine. The typical composition and
method of extraction of BLES and other surfactant formulations are discussed in the
next section.
While no concrete numbers exist on the market for lung surfictants, the estimate
from one of the manufacturers (Discovery Laboratories, Inc.) suggests that the
global market for these kinds of formulations is about U.S.$90 million for NRDS
(lung deficiency) and U.S. $95 million for meconium induced lung deficiency
(Capetole, 1998).The larger market, for which lung surfictant therapy is yet to be
proven effective, is for the treatment of ARDS, and is estimated to be U.S.$4 billion
worldwide (Capetole, 1998).

Commercial Exogenous Lung Surfactants

Several classification systems are used for commercial exogenous lung extracts
(Dargaville & Mills, 2005; Frerking et al., 2001; Lewis & Brackenbury, 2003;
Notter, 2000). Figure 8.3 divides lung surfactant formulations into two main
categories, according to the source of the preparation. The first category corresponds
to natural lung surfactants that contain endogenous lung extracts from mammals.
The second category corresponds to synthetic lung surfictants that contain a mixture
of synthetic PLs. Preclinid animal testing and clinical trials generally suggest that
the performance of natural animal-derived preparations is much better than that
of synthetic preparations (Been & Zimmermann, 2007; Halliday, 1995; Moya &
Maturana, 2007). However, batch-to-batch variations, the low content of proteins,
and the high cost of extraction and purification are among the limitations of animalderived preparations (Bernhard et al., 2000;Seeger et al., 1993).

Natural Surfactants' Extracts

One can also subdivide natural surfactants (Robertson & Halliday, 1998; Rudiger
et al., 2005; Seeger et al., 1993) into various categories. 'The first group consists
of exogenous surfactants extracted from animals (mammals) using a lung lavage
(Alveofact, BLES, Infisurf). The general extraction procedure of this category is
summarized in Fig. 8.4.The second group consists of exogenous surfactants extracted

196 0 EJ. Acosta et al.

t
LI-4
Lung Surfactant formations

Natural

Synthetic

Fig. 8.3. Classification of lung surfactant formulations and their commercial name.

1
Intact Calf Lungs

Intact Cow Lungs

cells removed by low-speed

centrifugation (150-2009)

pelleted by'high-speed
centrifugation (lOOOO-12000g)

pelleted by high-speed
centrifugation (10000-1 2000g)

Bronchoalveolar Lavage
(with 0.15 M NaCL)
I
cells removed by low-speed
centrifugation (1SO-200s)
pelleted bylhigh-speed
centrifugation (10000-1 20009)

extract with'ChloroformMethanol (remove SP-A)

extract with ChloroformMethanol (remove SP-A)

Bronchoalveolar Lavage
(with 0.15 M NaCL)

Bronchoalveolar Lavage
(with 0.15 M NaCL)
I
cells removed by low-speed
centrifugation (150-2009)

$.
Endogenous Lung
Surfactant

I
extract with'ChloroforrnMethanol (remove SP-A)

I*
extract with Acetone
(deplete Cholesterol and NL)

Alveofact

Fig. 8.4. Methods of extraction of natural lung surfactants from lung lavages.

lnfasurf

Lung Surfactants: Formulation, Evaluation, and Polymeric Additives 0

197

from animal (mammals) lung tissue (Curosurf). The third group is similar to the
second, except that the natural surfactants are supplemented with synthetic additives
(Surfacten, Survanta) to improve the surface activity of the formulation. The general
extraction procedure of the second and third groups (categories) is summarized in Fig.
8.5. A comparison of the composition of the six natural surfactants extracts is shown
in Fig. 8.6. More details about the extraction, composition, and surface activity of
natural lung surfactants are found in a recent review by Blanco and Pkrez-Gil (2007).
In the following paragraphs, we concentrate on providing the essential information
about these different surfactants.

Alveofact
Alveofact (bovactant) is a clinical exogenous lung surfactant obtained from the
bronchoalveolar lavage of bovine lung surfactant (Bartmann et al., 1992; Gortner
et al., 1990, 1992). The lavage is formed into a pellet by centrifugation, followed by
extraction with chloroform and methanol to remove hydrophilic surfactant protein
(Fig. 8.5). It contains a weight distribution of 99% of PLs and NLs (including 4% of
cholesterol), and 1% of hydrophobic surfactant proteins SP-B and SP-C as shown in
Fig. 8.6. It is produced as a sterile suspension in 0.15 M of NaCl at 45 mg/mL, and
is dosed at about 1.2 mL/kg. Alveofact is produced by Boehringer Ingelheim Co.,
Ingelheim, Germany.

BLES
BLES is a clinical exogenous lung surfactant obtained from the lung lavages of
benefited cows (Dunn et al., 1991; Enhorning et al., 1985; Smyth et al., 1983). The
lavage is pelleted by centrifugation, followed by extraction with chloroform methanol
to give BLSE (bovine lung surfactant extract). According to a recent composition
analysis of BLES by thin layer chromatography (TLC), this clinical surfactant
contains nearly 40% of disaturated PC (29.7% of DPPC), ~ 4 0 %of unsaturated
PCS, 13.6-14.5% of PGs, 0.9% of PIS, 2.8% of PEs, 0.3% of PSs, 1.4% of SPHs,
and 0.7% of lysophosphatidycholine (LPCs) (Nag et al., 2007). A range of 1-2% of
proteins SP-B and SP-C is contained in BLES, close to their natural amount NLs
(3%, mainly of cholesterol and diacylglycerol) are removed from BLES by acetone
washing, because this provides better surface activity in vitro. BLES is delivered as
a sterile suspension in 0.15 M of NaCl and 1.5 m M of CaCl at 25 mg/mL, and
its recommended dose for NRDS is 5 mL/kg (135 mg/kg), repeated every 8 hours.
BLES is produced by BLES Biochemicals Inc., London, Ontario, Canada.

Infasutf
Infasurf (CLSE or calfactant) is a clinical exogenous lung surfactant obtained from
calf lung lavage (CLSE) (Kendig et al., 1998; Kwong et al., 1985; Willson et al.,
1999). The lavage is pelleted by centrifugation at 10,000-12,000 g, followed by
extraction with 2:1 chloroform/methanol to give CLSE or calf lung surfactant extract

198 0 EJ. Acosta et al.

1
I

Bovine Lung Tissue

Porcine Lung Tissue

I I
Bovine LungTissue

homogenizeor mince
(in 0.1 5 M NaCL)

homogenizeor mince
(in 0.15 M NaCL)

homogenizeor mince
(in 0.15 M NaCL)

wash and ientrifuge

(1,000-3,OOO~)

differential cLntrifugation
and flotation steps

differential cLntrifugation
and flotation steps

extract with'ChloroformMethanol (21)

extract with ;thy1 acetate

(reduce neutral lipids)
I
extract with ChloroformMethanol (remove ethyl acetate)

extract with kthyl acetate

(reduce neutral lipids)
I
extract with ChloroformMethanol (remove ethyl acetate)

-r
i

liquid-gel affinity
chromatography

Lung Tissue Extract

LI
Curosurf

f
Lung Tissue Extract

supplement h t h synthetic
DPPC, Palmitic Acid, and
Tripalmitin

supplement with synthetic

DPPC, Palmitic Acid, and
Tripalmitin

Surfacten

Survanta

Fig. 8.5. Methods of extraction of natural lung surfactants from lung tissue.

Survanta

Fig. 8.6. Dry base composition (wt%) of lung surfactant extracts from lung lavages and lung tissue.
PL: phospholipids, NL: neutral lipids, P C phosphatidylcholines, P G phosphatidylglycerols, PI/PS
phosphaidylinositoVserine, PE: phosphatidyl ethanolamine, SPH: sphingolipids. SP-B/SP-C surfactant
proteins B/C.

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0

199

(Fig. 8.5). It contains a weight distribution of 93% of PLs (that include 83% of PC,
6% of PG, 4% of PI and PS, 3% of PE, and 2% of SPH), 5% of cholesterol and NLs,
and 1.5% of hydrophobic surfactant proteins SP-B and SP-C as shown in Fig. 8.6.
Disaturated PC is approximately 50% of the total PLs. I n h u r f is produced as a heatsterilized suspension in 0.15 M of NaCl at 35 mg/mL. The recommended dose for
NRDS is 3 mL/kg (105 mg/kg). Infasurf is produced by O N Y Inc., USA.

cumsu?f
Curosurf (poractant alpha) is a clinical exogenous lung surfactant obtained from
minced porcine lung tissue (Bevilacqua et al., 1996; Noack et al., 1987; Robertson,
1991). The process includes washing, centrihgation at 1,000-3,000 g, extraction
with 2: 1 chloroform/methanol, and liquid-gel affinity chromatography (see Fig. 8.5).
The average composition of Curosurf contains 93% of PLs (including 67-75% of
PC, 12-22% of PE, and SPH), and 1% of hydrophobic surfactant proteins SP-B
and SP-C, as shown in Fig. 8.6. Curosurf is produced as a suspension sterilized by
ultrafiltration (0.45- and 0.2-pm filters). Curosurf's recommended dose is 2.5 mL/kg
(200 mg/kg). Curosurf is produced by Chiesi Farmaceutici, Parma, Italy.

su?j2cten
Surfacten (Surfactant TA) is a clinical exogenous lung surfactant obtained after the
organic solvent extraction of finely ground bovine lung tissue. The final product is
supplemented with synthetic DPPC, palmitic acid, and tripalmitin (Fujiwara et al.,
1990; Konishi et al., 1992; Taeusch et al., 1986). The lung tissue goes through several
centrihgation and flotation processes, extraction with ethyl acetate to reduce NLs,
and additional extraction with chloroform-methanol to remove ethyl acetate. The
extract is then supplemented with the aforementioned synthetic additives (Fig. 8.5).
The final product contains a weight distribution of 84% of DLs, 8% of palmitic acid,
7% of tripalmitin, and 1% of hydrophobic surfactant proteins SP-B and SP-C as
shown in Fig. 8.6. Surfacten is produced as a suspension sterilized by high-pressure
filtration, lyophilized, stored in vials, and then suspended in sterile 0.15 M of NaCl
at 30 mg/mL prior to, and its recommended use is 4 mL/kg (100 mg/kg), 1-4 doses
every 6 hours. Surfacten is produced by Mitsubishi Tanabe Pharma Corporation,
Tokyo, Japan.

Survanta
Survanta (beractant) is a clinical exogenous lung surfactant made by organic solvent
extraction of processed, supplemented bovine lung tissue (Bloom et al., 1997; Horbar
et al., 1989; Speer et al., 1995). Survanta is prepared using similar methods and
identical synthetic additives as in Surfactant TA (Fig. 8.5). The final product contains
slightly less than 1% of hydrophobic surfactant proteins (including a small amount of
SP-B) as shown in Fig. 8.6. Survanta is produced as an autoclave sterilized suspension
in 0.15 M of NaCl at 25 mg/mL. The recommended dosage is 4 mL/kg (100 mg/kg),

200 0 EJ. Acosta et al.

1-4 doses every 6 hours. Survanta is produced by Abbott Laboratories, Abbott Park,
Illinois, United States.

Synthetic Surfactan ts
One can also subdivide synthetic surfactants (Bengt-Robertson et al., 2000; SeurynckServoss et al., 2006; Wu et al., 2003) into various categories. The first involves synthetic
exogenous surfactants with no proteins (ALEC, Exosurf); the second utilizes synthetic
peptide analogs (Surfaxin); and the third contains recombinant human apoproteins
(Venticute). A comparison of the composition of the four synthetic surfactants is
shown in Fig. 8.7 ; hrther details are found in recent reviews (Curstedt & Johansson,
2006; Mingarro et al., 2008; Walther et a]., 2007).

ALEC
ALEC (artificial lung expanding compound) is a clinical exogenous lung surfactant
containing synthetic DPPC and egg PG in a 7:3 molar ratio (Bangham et al.,
1979; Morley, 1987; Wilkinson et al., 1985) as shown in Fig. 8.7. DPPC is mainly
responsible for lowering surface tension; the egg PG is added to improve adsorption
and spreading. ALEC was produced by Britannia Pharmaceuticals Ltd., United
Kingdom, but was withdrawn from the market.

Hexadecanol: 9%

Fig. 8.7. Dry-base composition (wt%) of synthetic lung surfactant formulations. DPPC: dipalmitoyl
phosphatidylcholine, POPG palmitoyloleyl phosphatidyl glycerol, rPC-: recombinant surfactant protein C.

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 20 1

Exosurf (colfosceril palmitate) is a synthetic exogenous lung surfactant formulated

with DPPC, hexadecanol, and tyloxapol in a 1:0.111:0.075 weight ratio (Anzueto et
al., 1996; Durand et al., 1985; Hudak et al., 1997) as shown in Fig. 8.7. DPPC is the
main surface active species, while tyloxapol (nonionic additive) and hexadecanol are
used to improve adsorption and spreading. Exosurf is produced as a sterile lyophilized
powder in vacuum-sealed vials, and then suspended in 8 mL of saline solution. The
recommended dosage is 5 mL/kg (67.5 mg/kg). It is produced by GlaxoSmithKline,
North Carolina, USA. Exosurf is no longer available in the United States and
Canada.
Suflm'n
Surfaxin (KL4 or lucinactant) is a clinical exogenous lung surfactant containing
synthetic DPPC, palmitoyl-oleoyl-phosphatidyl-glycerol(POPG), palmitic acid, and
a 21 amino-acid synthetic hydrophobic peptide with repeated sequences of one lysine
(K) and four leucine (L) residues (Cochrane & Revak, 1991; Moya et al., 2005;
Saleem et al., 2008). DPPC and POPG are present in a 3:l weight ratio. Palmitic
acid is present at 15% by weight relative to PLs, while the hydrophobic (KL) peptide
is present at 3% by weight as shown in Fig. 8.7. Surfkin is produced by Discovery
Laboratories Inc., Pennsylvania , USA. Recently, an aerosol version of Surfkin was
developed by the same producer, and named Aerosurf. Surfkin and Aerosurf are still
in their developmental stages, and are not yet approved for clinical use.
Vmticute
Venticute (recombinant SP-C or lusupultide) is a clinical exogenous lung surfactant
containing synthetic DPPC, POPG, palmitic acid, and human sequence or modified
human-sequence recombinant surfactant apoprotein (rSP-C) (Benson, 1993; Markart
et al., 2007; Spragg et al., 2004). DPPC and POPG are present in a 7:3 weight ratio.
Palmitic acid is present at 5% by weight relative to PLs, and the rSP-C protein is
present at 2% by weight (Fig. 8.7). It is produced by Altana Pharma AG, Germany.
Venticute is not yet approved for clinical use.

Evaluation of Lung Surfactants

In Vitro Evaluation
In vitro techniques are the most commonly used techniques for assessing the
performance of lung surfictants. In such techniques, the surface properties (such as
minimal surface tensions and adsorption rates) of lung surfactants are assessed at
specific cycling and environmental conditions. In this section, the most common in
vitro techniques are discussed.
Langmuitc Wilbelmy Balance
In a Langmuir-Wilhelmy Balance (LWB) (Dynarowicz-Lath et al., 2001; Langmuir,
1917; Tab& & Notter, 1977), an insoluble surfactant monolayer is spread on top of

202 0 EJ. Acosta et al.

a liquid, and compressed by means of hydrophobic barriers placed on the surface.

During compression, surface pressure (n = y,-y, the difference between the original
surface tension of the pure solvent-yo-minus the surface tension in the presence of
the surfactant film-?) is monitored by a Wilhelmy plate as shown in the schematic
presented in Fig. 8.8. The surface pressure is usually calculated using a simplified
equation:

Eq. 8.1.
where W and T are the width and thickness length of the slide, 8 is the contact angle,
AF is the difference in force (measured by a balance) between the case where the slide
in pure solvent (no surfactant) and the case where the slide is inmersed in the solvent
containing the surfactant film. Equation 8.1 is usually hrther simplified by assuming
the slide thickness T is neglected with respect to the slide width W , and assuming
complete wetting so that the contact angle is zero, giving cose = 1.
The Langmuir trough with the associated Wilhelmy balance is the standard
method to study insoluble surface active films. The surface pressure isotherms of
fatty acids, PLs, proteins, and other surface active material were studied using this
methodology. To produce these isotherms, the surface active material is dissolved in an
organic volatile solvent like hexane. This solution is then placed on the surface of the
aqueous (subphase) solution in the trough. Once the organic solvent has evaporated,
the surface coverage
of the insoluble surfactant is calculated based on the mass of
the surfactant deposited divided by the area of the trough = masdarea). The mobile

(r)

(r

Film collapse

Phase
transitions

a, -1 /r
Fig. 8.8. Schematic of a Langmuir trough (cross-sectionalview), and a typical surface-pressureisotherm
obtained from these experiments.

Lung Surfactants: Formulation,Evaluation, and Polymeric Additives 0 203

barrier(s) of the trough is(are) slowly moved to gradually reduce the area of the trough
and compress the surfactant film. Upon compression, the surfactant film excludes
water molecules from the interface, producing lower surface tensions (higher surface
pressures). The experiment produces a set of data of surface pressure (n)versus the
area per molecule of surfactant (u, -l/r).
This is used to study the phase transitions
of the film, and the ultimate surface pressure and area at collapse. This output is
illustrated in Fig. 8.8.
A major disadvantage ofthe LWB is that it requires a large volume of the subphase.
Also, only slow compression rates (dcl/dt)can be used to avoid the formation ofwaves.
Furthermore, unless special measures are taken, most LWBs experience film leakage
(spreading over the boundaries of the trough), and this prevents the surfactant film
from reaching the low surface tension values of 0-5 mJ/m2 (or surface pressures of
67-72 mJ/m2) required to confirm proper lung surfactant function.

Pulsating-Bubbk Sulfitetometer
In a pulsating-bubble surfactometer (PBS) (Enhorning, 1977; Hall et al., 1993;
Seurynck et al., 2005), an air bubble is formed in a surfactant suspension by drawing
air from the atmosphere through a capillary tube. After surfactant adsorption reaches
equilibrium, the bubble is pulsated between two set positions. The radius of the
bubble is usually monitored, and the pressure in the subphase is measured using a
pressure transducer. The surface tension, y, is calculated from the measured values of
the pressure drop, AP , and the bubble radius, R , using the Laplace equation and
assuming a sherical shape,

=z
Eq. 8.2.

This method has several advantages including simplicity of use. It requires only a
small volume of subphase, it can be operated at physiologically relevant cycling rates,
and it is commercially available. However, still two major disadvantages remain
to be accounted for. The design is prone to film leakage into the capillary at low
surface tensions, and the spherical shape assumption is not appropriate at low surface
tensions. The output from these experiments is the minimal and maximal dynamic
surface tensions during a cycle with given compression ratio (reduction in surface
area) and compression period (duration of the compression-expansion cycle).

Grptive-bubbk Suflactometer
In a captive-bubble surfactometer (CBS) (Codd et al., 2002; Prokop et al., 1998;
Schurch et al., 1989), an air bubble is formed inside a closed chamber filled with a
surfactant suspension. The air bubble floats against a hydrophobic ceiling coated with
1% of agarose gel (similar to the ADSA-CB setup in Fig. 8.9). After the surfactant
film is formed and it reaches its equilibrium surface tension, the bubble is cycled

204 0 EJ. Acosta et al.

(dynamic compression-expansion) by injecting and withdrawing subphase liquid.

The bubble is not expected to be spherical because it is much larger than in the PBS
setup. In this case, the shape of the interface is controlled by the balance between
surface tension forces and gravity as given by the Laplace equation:

Eq. 8.3.
where R, and R, are the principal radii of curvature at a point on the interface, R,
is the symmetric radius ofcurvature at the apex point of the interface, Ap is the density
difference across the interface, g is the local gravitational acceleration, and z is the
vertical distance between the apex point and the point on the interface. Therefore, the
surface tension value, y, can be determined from the shape of the interface. In CBS,
the height-to-diameter ratio of the bubble is used to calculate surface tension, bubble
area, and volume (Schoel et al., 1994).
The CBS has all the advantages of the PBS setup, but the confined bubble
solves the problem of leakage, and does not assume a spherical drop. The CBS does,
however, introduce another set of challenges-is difficult to use and can only handle
dilute surfactant solutions (c2 mg/mL) due to the obscuration of the optical path. No
way is available to appropriately control the composition of the air entrapped in the
bubble, and the large liquid-to-air volume ratios are not compatible with physiological
conditions.

Axiymmetric Dropshape Analysis (ADSA)-Captive Bubble (CB)

Axisymmetric drop-shape analysis (ADSA) is a technique for measuring surface
tension and contact angles (Cheng et al., 1990; Davidson et al. , 1997; Rotenberg et
al., 1983). Currently, three types ofADSA setups exist: a captive bubble (ADSA-CB),
a pendant drop (ADSA-PD), and a constrained sessile drop (ADSA-CSD). These are
illustrated in Fig. 8.9. ADSA software extracts an experimental drop profile from the
digitized image, and uses an optimization procedure to match the extracted profile
to a theoretical profile calculated from the Laplace equation. The surface tension of
the bubble/drop is then calculated based on the matched theoretical profile. Direct
outputs of ADSA are the radius of curvature and the volume and surface area of the
drop. The first embodiment ofADSA, working in conjunction with a captive bubble
(ADSA-CB) (Prokop et al., 1998; Schurch et al., 1998; Zuo et al., 2007), has a very
similar setup to CBS. However, ADSA-CB uses the more accurate and automatic
ADSA software for determining the surface tension, surface area and volume from
a bubble image, rather than just using the height-to-diameter ratio of the bubble as
in the case of CBS. ADSA-CB has the same advantages of the CB with the added
advantage that it provides a greater accuracy in the calculation of the surface tension
and faster data acquisition and processing than CSB. ADSA-CB has the same
limitations of the CSB setup.

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 205

Stirring bar

Fig. 8.9. Schematic diagram of the ADSA setup (top)and the three different drop configurationsthat can
be used (bottom):captive bubble (ADSA-CB), pendant drop (ADSA-PD),and constrained sessile droplet
(ADSA-CSD).

ADSA-Pendant Drop (ADSA-PD)

In ADSA-DD (Cabrerizo-Vilchez et al., 1999; Kwok et al., 1994; Lu et al., 2003), a
pendant drop of a surfactant preparation is formed at the end of a teflon capillary tube
as shown in Fig. 8.9. Surface tension, surface area, and drop volume are calculated
using ADSA s o h a r e . Unfortunately, this setup can experience film leakage, and at
surface tensions of 15 mJ/m* or less, the drop tends to detach from the capillary.

ADSA-Constrained Sessib Drop (ADSA-CSD)

ADSA-CSD (Kanget al., 2008; Wulfet al., 1999; Yu et al., 2004a; Zuo et al., 2006) is
now the preferred method-in our laboratory-for determining the dynamic surface
tension of lung surfactants. In this setup, a sessile drop is formed on top of a small flat
(circular) pedestal which has a sharp knife-edge preventing uncontrolled spreading
and hence film leakage (see Fig. 8.9). Surface tension, surface area, and drop volume
are again calculated by using ADSA software. The advantages of the ADSA-CSD

206 0 EJ. Acosta et at.

configuration are that it is a simple setup and easy to use. It requires only small
subphase volumes (1-3 mL), and can simulate physiologically relevant conditions
including the cycling rate and compression ratio. The environmental conditions of the
surrounding atmosphere (humidity, CO, content, 0, content) and temperature are
readily controlled. A major advantage of this method is that no.optica1 limits exist to
the opacity of the liquid, and therefore no limitation exists as to the concentration of
the surfactant in the preparation. One can use ADSA-CSD for adsorbed and spread
films, and it produces accurate and fast measurements of surface tensions (especially
at values below 5 mJ/m). The main disadvantage of the technique is that it requires
more sophisticated hardware that is not commercially available, and so it is still in a
developmental mode.
Also important is to highlight this: by controlling the humidity in ADSA-CSD to
100% of RH (relative humidity), we have determined that tests carried out in LWB,
PB, CB, ADSA-CB, and ADSA-PD may not represent the physiological condition
in the alveoli. Particularly, we observed that in saturated ( I 00% of RH) air at 37C
numerous lung surfactant preparations fail to reach low surface tensions. However,
the same preparations evaluated at room temperature or at 37C but in unsaturated air
( ~ 1 0 0 %of RH) produce low surface tensions and are considered good formulations
(Acosta et al., 2007; Zuo et al., 2005). The discussion on lung- surfactant additives
in the last part of this chapter concentrates on experiments carried out using ADSACSD in an environment of 100% of RH at 37C.

In Situ Evaluation
In situ techniques are used to estimate the alveolar surface tension in excised lungs
(i.e., in its natural place). In situ techniques are usually considered intermediate
benveen in vitro and in vivo.

Microdroplet Method
In the microdroplet method (Craster & Matar, 2006; Im Hofet al., 1997; Schiirch et
al., 1976), a test fluid droplet is formed on the tip of a micropipette inside an alveolus.
After deposition onto the alveolar surface of an excised lung, the drop spreads to a
diameter determined by the surface tension of the alveolar fluid, the surface tension
of the test liquid, and the volume of the test liquid. The diameter of the liquid lens
is usually monitored via microscopy, The film surface tension is determined from a
calibration of the liquid lens diameter (for a constant volume) of various test liquids
versus the surface tension of these liquids. This technique is difficult to implement,
but it was essential in demonstrating that the surface tension of the alveolar film at the
end of expiration approaches a near-zero value.

Pressure- Volume (P- Method

In the P-V method (Bachofen et al., 1970, 1987; Fisher et al., 1970; Wilson, 198l),
The alveolar surface tension is indirectly determined from the analysis of P-V relations

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 207

of excised lungs ventilated dynamically. In this method, the excised lung is filled with
air, and is ventilated (dynamic compression-expansion cycles) to obtain P-V data.
The P-V work necessary for this cycling has two components-the work required to
overcome tissue forces and the work required to overcome surface tension forces. The
same excised lung is later filled with a saline solution, and is cycled using the same
protocol. In this case, the P-V work only accounts for the tissue forces. The difference
in the P-V isotherms of air-filled and saline-filled lungs indicates the contribution of
surface tension forces. The P-V curve associated with surface tension forces can be
related to surface tension using an energy balance approach. In its simple form, the
equation used to estimate the surface tension is (Bachofen et al., 1970):

Q. 8.4.
where Vis the volume of the air in the lung; P, is the difference between the pressure
required to achieve P for an air-filled lung minus the pressure required to fill the
saline-filled lung; R is a constant that must be determined by using the maximal
inflation pressure and volume and assuming a value of surface tension for that
maximal surface tension. The overall accuracy of the method depends on the accuracy
of that assumption.

In Vivo Evaluation
In vivo techniques involve experimentation done on a living animal. Contrary to the
surface tension as an output from in vitro and in situ techniques, the main outputs
in this approach are usually the lung compliance (the ability of the lungs to stretch
their volume-AVin response to an applied pressure-AP; C =AVlAP) and blood
oxygenation (the partial pressure of oxygen in blood sample). Higher compliances
and blood oxygen levels are indicative of the success of the preparation. Different
animal models were used to evaluate the efficiency of lung surfactant preparations on
premature and adult animals.

lmMtunAnimal Modcl
The premature animal model (Kobayashi et al., 1989; Pinkerton et al., 2000; Sun et
al., 1997) is usually used to evaluate surfactant efficacy in premature animals (e.g.,
premature rabbits or lambs) with natural surfactant deficiency. In these experiments,
the animal fetuses are delivered at an early gestational age to ensure that the lungs are
still surfactant deficient. The premature animal is connected to a ventilator system, and
an exogenous surfactant preparation is administered to the lungs through an injection
via an endotracheal tube. The tidal volume (V)and the peak inspiratory pressure
(PIP) are continuously monitored during ventilation. Dynamic lung compliance is
the main output from this model.

208 0 EJ. Acosta et at.

Saline-lungLavage Model
This model (Bailey et al., 2004; Hafner et al., 1998; Lewis et al., 1996) is also used to
evaluate surfactant efficacy in dealing with a natural surfactant deficiency. However,
in this model, adult animals (e.g., rats, rabbits, pigs, or sheep) are used. To ensure
that the lungs are surfactant-deficient, the endogenous lung surfactant is removed by
repeatedly lavaging the lungs with a saline solution. During this process, the adult
animal is connected to a mechanical ventilator. After a complete depletion of the
endogenous surfactant, an exogenous surfactant preparation is administered to the
lungs through a bolus injection via an endotracheal tube. Throughout the experiment,
the arterial partial pressure of oxygen (PaO) is measured in arterial-blood samples.
The main output from this model is the blood-oxygen level.
In addition to the methods testing surfactant deficiency, other methods are used
to study surfactant dysfunction or inactivation. In these cases, the objective is to
investigate the efficacy of an exogenous surfactant in the therapy of lung injury. One
can induce a lung injury by using methods such as meconium aspiration (Lu et al.,
2005b; Moses et al., 1991; Sun et al., 1993), acid aspiration (Davidson et al., 2005;
Eijking et al., 1992; Lamm et al., 1990), and high-stretch ventilation (Bailey et al.,
2006; Nakamura et al., 2002; Wakabayashi et al., 2006).

Lung Surfactant Additives

Nonionic Polymers
The introduction ofpolymeric additives into lung surfactant formulations was inspired
by the knowledge that SP-A, a large molecular weight hydrophilic protein, enhanced
the surface properties of natural lung surfactants (fast film formation and reverse/
resist protein inhibition). Taeusch et al. (1999) first introduced the use of nonionic
polymers in an attempt to simulate the CRD of SP-A. They found that polyethylene
glycol (PEG), dextran, and polyvinylpyrrolidone (PVP) (with molecular weights of
3.3 to 500 KDa) when used in concentrations of 10 mg/mL to 100 mg/mL, improved
adsorption time and were capable of resisting surfactant inhibition by meconium,
serum, and lysophosphatidylcholine. Parallel to Taeuschs work, Kobayashi et al.
(1999) also studied the effect of dextran on reversing albumin-induced inhibition.
Numerous publications supporting these initial observations (J. Lu et al., 2002,2003,
2005; K. Lu et al., 2000, 2001, 2005 ; Sarin et al., 1999; Taeusch, 2000; Yu et al.,
2004b).
The mechanism of action of nonionic polymers was explained on the basis of
a depletion attraction mechanism (Yu et al., 2004b). Figure 8.10 illustrates this
mechanism, which is well-documented in the polyrner/colloid literature. According
to this mechanism, large aggregates (which are more surface active) are formed after
fusing small surfactant aggregates. This fusion occurs because the polymer can not
penetrate the excluded volume around the surfactant aggregates (e.g., vesicles) which
creates a differential in polymer concentration between the excluded volume and

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 209

the continuous aqueous phase. This difference in polymer concentration produces

an osmotic pressure that drives the water out of the space between the surfactant
aggregates, inducing the fusion of the aggregates (Asakura & Oosawa, 1954).
Despite the excellent in vitro surface activity of these systems and the success
of some in vivo experiments, in some cases PEG was detrimental to the surhctant
therapy (Campbell et al., 2002). With high concentrations of PEG, the viscosity of
the suspensions is high ( > I 0 cp), and this causes difficulty for these formulations to
spread after instillation. Also, the concentration of the polymer can be high enough
to initiate a large enough osmotic pressure difference which dries the epithelial
cells. This explains the cytotoxicity observed in tissue models (Zuo et al., 2006).
Furthermore, when PEG is used in BLES solutions at a concentration of 20 mg/mL,
the oxygen mass transfer coefficient increases by nearly 30%, but when used at 50
mg/mL, the mass transfer coefficient is reduced to almost one-half the original value
(Zuo ,2006). The same trend in oxygen transfer is observed in PEG-only solutions.
The concentration of nonionic polymers in exogenous surfactant preparations is
three orders of magnitude larger than the concentration of SP-A in natural surfactants,
suggesting that nonionic polymers do not properly mimic the role of SP-A well, as
demonstrated by a series of studies in our laboratory. Figure 8.1 1 shows a typical
surface tension-relative area of a cycle obtained during dynamic compression and
the expansion of various lung surfactant formulations. Figure 8.1 l a presents a system
of 0.5 mg/mL of BLES inhibited by 2.5 mg/mL of albumin in the preparation. An
inhibited surfactant cannot achieve a surface tension of 5 mJ/m2 or less during a
cycling protocol that simulates the physiological conditions (3 s/cycle period, 20%
of compression ratio, 100% of RH). When a similar formulation is prepared with
50 mg/mL of PEG, then the formulation does achieve a minimal surface tension of
nearly 5 mJ/m2 (Fig. 8.1 Ib). However, when the 0.5 mg/mL of BLES formulation
is exposed to a more severe inhibitor like serum (which contains albumin, free fatty
acids, and other inhibitors) at a level of 10 pL/mL, then the addition of 50 mg/
mL of PEG cannot reverse the inhibitory action of serum components. Serum is a
more realistic example of inhibitory agents because these agents penetrate the alveolar
space at the onset of pulmonary edema or hemorrhage where blood or filtered blood
components infiltrate the alveolar fluid.
The cumulative evidence on the lack of the beneficial effects of nonionic
polymeric additives in typical inhibitory conditions, the difficulty in delivering these
formulations, the in vitro cytotoxicity, and the poor gas transfer at high polymer
concentration has led the surfactant community to focus on more efficient, charged
polymeric additives.

Anionic Polymers
K. Lu et al. (2005 ) and Taeusch et al. (2005) introduced the use of the anionic
glucosamine and hyaluronan as an alternative to improving the surface activity of
lung surfactants. Hyaluronan consists of the N-acetyl-glucosamine conjugated by a

2 10 0 EJ. Acosta et al.

Fig. 8.10. Schematic of the depletion-attraction mechanism. The polymers in solution (with a given
gyration radius, Rg) cannot penetrate the excluded volume between the surfactant aggregates. This
difference in concentration induces an osmotic pressure difference that "pushes" the surfactant
aggregates together.

+ PEG 50 mg/ml

0.5 mg/ml BLES

'

08

10

Relative Area

(2-b)

1 .o

0.8

Relative Area

Fig. 8.1 1. Surface tension (y)-relative-area compression (open symbol)-expansion (solid symbol)
cyclic isotherms for lung surfactant systems containing adult serum or albumin.The arrows indicate film
collapse. Fig. (1-a)and (1-b) are adapted from Zuo et al. (2006).Cycling conditions: 37"C, 100% of RH, 20%
of compression, 3 skycle compression.

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 2 11

p-1,4 linkage to glucuronic acid (See Fig. 8.12 for its dissociated form). The original
inspiration for its use is its presence in various body fluids, where it helps control
the colloids stability. In keeping with the idea of using nonadsorbing polymers
to promote the depletion-attraction of surfactant aggregates, since the soluble
hyaluronan molecule and lung surhctant aggregates are negatively charged, polymer
adsorption should be minimized and the depletion-attraction mechanism enhanced.
This strategy produced formulations that are 20 times more effective than PEG-based
formulations in albumin-inhibited systems (Taeusch et al., 2005). Hyaluronan was
successfully used in vivo to recover the injury produced by meconium instillation in
rats (K. Lu et al., 2005b).
More recent work on the use of hyaluronan as a lung surfactant additive
corroborated that this polymer is about one order of magnitude more active than
PEG in producing fist-adsorbing formulations (Taeusch et al., 2008). The authors
propose that the lower concentration of hyaluronan should decrease the concerns
of the high viscosity and cytotoxicity observed in formulations containing a high
concentration of PEG. Furthermore, because hyaluronan is a polymer synthesized in
the body, no concerns exist of undesirable interactions between hyaluronan and the
alveolar tissue.
To develop potential low-cost alternatives to hyaluronan, we conducted
preliminary studies using porcine mucine (an anionic glycoprotein that mimics the
mucosal secretion in lungs) to evaluate its potential impact on the dynamic surface
tension of lung surfactant preparations. Figure 8.13 presents one experiment where
a preparation of 2.0 mg/mL of BLES containing 10 mg/mL of porcine mucin was
exposed to 50 pL/mL of serum (a ten-times higher inhibitor content to that evaluated
by Taeusch et al., 2005). As shown in Fig. 8.13, the minimal surface tension during
the dynamic compression (conducted at 3 slcycle, 20% of compression, 100% of
RH, and 37C using the ADSA-CSD device) is 2 mJ/m*, which is an exceptionally
low value for this level of serum content. The actual concentration of mucine is about
eight times that of hyaluronan used in the work ofTaeusch et al. (2005). Quite likely,
at this concentration, porcine mucine also follows the depletion-attraction model
suggested for hyaluronan.
Anionic polymeric additives appear to offer a valid option for the improvement of
the surface activity of lung surfactant preparations, and they continue to be considered
by various research groups.

Cutionic Polymers
Recently, our group introduced the use of a cationic polymer chitosan, and showed
that it can produce positive effects similar to PEG, but uses 1000- times less polymer
(Zuo et al., 2006). Chitosan is a polymer of p(l+4)-linked D-glucosamine (Fig.
8.14), usually derived from chitin (the skeletal material of invertebrates) by progressive
deacetylation in alkaline suspension. Chitosans amino (-NH,) groups, with a pKa
of 6.5, become fully protonated when dissolved in an acidic environment (pH <6).

212 0EJ.Acostaetal.

n
Fig. 8.12. Molecular structure of the repeating unit in hyaluronan (dissociated hyaluronic acid).

._

50
40

3 30
E

* 20
10

0.80

0.90

1.oo

Relative Area

Fig. 8.13. Surface tension (y)-relative-area cyclic compression isotherms for 2 mg/mL BLES and 10
mg/mL of porcine mucin (extractedfrom intestine lining, and used as an anionic polymeric additive) in
the presence of a high-serum concentration (50VVmL). Cycling conditions: 37T, 100% of RH,20% of
compression, 3 s/cycle compression.

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 21 3

The ionized chitosan produces a linear polyelectrolyte with a high positive-charge

density (Henriksen et al., 1994). Chitosan is extensively used in pharmaceuticals. It
can stabilize liposomes as drug-delivery vehicles, selectively bring drugs to their target,
and serve as a substrate material for cell transplantation and tissue regeneration.
Furthermore, chitosan is considered safe in the human body and degradable by
ubiquitous enzymes (Kumar et al., 2004).
Two reasons made us look at chitosan as a potential alternative to anionic and
nonionic polymers. First, cationic polymers are commonly used in another process
to promote the flocculation of negatively charged colloids to form large aggregates
by using a relatively small concentration of the polymer. Although the detailed
mechanisms are unknown, large surfactant aggregates are typically correlated with
hnctional surfactants that produce low surface tension under compression. Secondly,
the protein SP-B is cationic, and is considered to be an essential protein since SP-Bdeprived mice cannot survive due to lung collapse (Tokieda et al., 1997).
Zuo et al. (2006)showed that chitosan could reverse the inhibition caused by 2.5
mg/mL of bovine albumin in 0.5 mg/mL of a BLES preparation using as little as 0.05
mg/mL of chitosan. One must compare this chitosan concentration with the 50 mg/
mL of PEG ( 1 0 KDa) needed to achieve the same effect. In that article, hyaluronan
(1240 m a ) was also tested, but in that case although 1 mg/mL of hyaluronan was
able to overcome the inhibition induced by 1 mg/mL of albumin, it did not overcome
the inhibition caused by 2.5 mg/mL of albumin. This also suggests that the cationic
polymers might be more efficient that anionic polyelectrolytes.
Previous to the work of Zuo et al. (2006 ), the only references to chitosan and
lung physiology suggested that high concentrations of chitosan (about 200 mg/mL of
an intravenous injection) might produce lung injury (Khanala et al., 2001). However,
in vitro cytotoxicity tests carried out in bronchoalveolar cultured tissue showed that
cells exposed to a preparation containing 0.05 mg/mL of chitosan had a viability
of 83%, near the 95% of viability reported for cells exposed to BLES alone and
superior to the 46% of viability for cells exposed to a BLES-PEG formulation (Zuo

Fig. 8.14. Molecular structure of chitin (left) and chitosan-deacetylated chitin (right).

2 14 0 EJ. Acosta et al.

et al., 2006 ). Another important finding in that article is that morphological changes
occur in the lung surfactant preparation after chitosan treatment. In the presence of
chitosan, numerous small vesicles flocculate to form large aggregates. The presence of
these larger aggregates seems, once more, to be correlated with the ability to obtain
low surface tensions, even in the presence of inhibitors.
Kang et al. (2008) studied the mechanism of action of chitosan in BLES
formulations. The dynamic surface tension of these preparations in 100% of RH at
37C was evaluated as a function of BLES-chitosan ratios using 0.5 and 2.0 mg/mL
of BLES. These studies were carried out using the ADSA-CSD, employing a period
of 3 s/cycle, and a compression ratio of 20%. Kang et al. (2008) also determined
the <-potential of the aggregates in these formulations in an attempt to evaluate the
binding of chitosan to BLES lipids.
Figure 8.15a presents the <-potential of the aggregates of formulations prepared
with 0.5 mg/mL and 2.0 mg/mL of BLES as a function of the total chitosan
concentration used in the preparation. The <-potential of the aggregates increased
from -15 mV to +25 mV as the chitosan concentration increased. The pH of these
preparations was approximately 5.5. 'This change in <-potential reflects the binding of
the positively charged chitosan to the anionic lipids, in particular PG, in BLES.
The binding isotherm for BLES and chitosan has remained elusive to our group
due to the analytical challenges in measuring the unbound chitosan in mixtures with
BLES. However, the binding isotherm can be estimated using the c-potential values
in Fig. 8.15a and the method of Mezei and MCszdros (2006). The principle of the
method is that the binding between surfactants and oppositely charged macromolecules
produces a shift in the <-potential, and that two systems with the same <-potential (if
they depart from approximately the same <-potential value) have the same amount
of bound surfactant. Three important assumptions are built into this analysis: (i)
the system is at equilibrium; (ii) all the ionic charges of the polymer are available
for binding; and (iii) the excess unbound solute does not affect the (-potential of
the surfactant-polymer aggregates. The method has worked well (i.e., less than 20%
of difference with respect to the mass-balance data) for polyethyleneimine -sodium
dodecyl sulfate systems (Mezei & Mtszdros, 2006).
The method involves producing two <-potential curves in the systems with two
concentrations of the substrate-Cs. In this case, the substrate is BLES, and the
actual binding (n+/n-)is the moles' positively charged glucosamine groups of chitosan
bound to one mole of anionic lipid (PGs) in BLES. The shift of potential (A<) is
proportional to the value of an actual binding ratio:

<-

where C1 and CL are the initial (dose) concentrations of the polymer (chitosan)
expressed in terms of moles of cationic groups for a given A<, and Csl and Cs2 are

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 2 15

u)

s?
(0

Pw

(32-2.0mg/ml BLES

w
lu

lu

*.a

Total chitosan (mg/ml)

s?
0

.-0
U

e
F
5
aC
ir

0.5 mg/ml BLES

2.0mg/ml BLES

Unbound (free) chitosan concentration (mg/ml)

Fig. 8.15. (A): c- potential of BLES-chitosanpreparations as a function of chitosan concentration using

two different BLES concentrations: 0.5 mg/mL and 2.0 mg/mL (B): Binding isotherm (25C) for BLES and
chitosan estimated from the 3- potential measurements, and calculated using the formulation of Mezei
and M&z&ros(2006).The word'actual"in Fig. 8.1 5b is used to differentiatebetweenthe value calculated
from the methodof Mezei and Medros and the values reportedby Kang et al. (2008) based on the mass
of chitosan and lipids initiallyadded into the preparation..

2160EJ.Acostaetal.

the molar ionic lipid concentrations in the BLES preparations (0.5 mg/mL and 2.0
mg/mL). These points are illustrated, for example, in Fig. 8.1 5a. Ce is the equilibrium
concentration of the unbound polymer in the solution, and is assumed to be the
same in two mixtures with the same & potential (Demarger-Andre & Domard, 1994;
Mezei & Mtszdros, 2006). To estimate the actual binding isotherm, the last equality
of Equation 8.5 is used to determine Ce (for a given A&), and then n+/n-is calculated
using either one of the last two terms of Equation 8.5. The change from mass units
(used in Fig. 8.15a) to molar units requires knowledge of the anionic lipid content
in BLES, an assumption regarding their molecular weight, and knowledge of the
degree of deacetylation and dissociation of chitosan. Further details in this change of
mass to mole ratio are found elsewhere (Kang et al., 2008). Figure 8.15b presents the
estimated binding isotherm for the BLES-chitosan system.
The estimated binding curve is consistent with the hypothesis of the binding
of the cationic groups in chitosan to the anionic lipids in BLES, and we directed
our efforts to determine the effect that different chitosan/BLES ratios have on the
properties of the formulation in the presence and absence of serum. Figure 8.16
presents the surface tension-relative compression area curves for systems formulated
with 2.0 mg/mL of BLES and chitosan (0,0.1, and 0.25 mg/mL) in the absence (top
row) and in the presence (bottom row) of 50 pL/mL of serum. Figure 8.16 presents
three main features of a dynamic compression cycle. The first feature is film relaxation
in which the surface tension of the film at the end of the compression stage increases
while holding the area constant. Film relaxation is linked to the hydration of some
of the compressed lipid molecules in the surfactant film (Acosta et al., 2007). The
second feature of the compression cycles in Fig. 8.16 is the dilatational elasticity of the
film ( E = (RA0)[4/d(RA0)]),
which is proportional to the slope of the compression
stage of the graphs in Fig. 8.16. The higher the elasticity (the more pronounced the
slope), the more solid-like (better) is the surfactant film, and the lower the surface
tensions can be reached for a given compression. The concept of dilatational elasticity
is related to that of lung compliance introduced in the discussion of the in vivo tests.
The last feature observed in some cycles of Fig. 8.16 is film collapse, in which case the
progressive reduction in surface area (compression) does not result in further reduction
of surface tension. Film collapse indicates the onset of film inhibition when it occurs
at relatively high surface tensions (e.g., 10 to 20 mJ/m). In summary, the desired film
properties are: no relaxation, high elasticity, and film collapse (if it happens at all) at
near-zero surface tensions.
Figure 8.16 shows that the addition of chitosan, up to a certain concentration,
helps improve the properties of BLES, and achieves low minimal surface tensions
even in the presence of serum. However, a large dose of chitosan induces surfactant
inhibition. To understand the behaviors observed in Fig. 8.16, the elasticity ( E ) and
relaxation rate (dyldt at the end of the compression stage) were plotted as a function
of the binding ratio (nil#-) between the cationic groups in chitosan and the negatively
charged lipids in BLES (Fig. 8.17) for BLES-chitosan formulations (no serum).
In both cases, the maximal elasticity and minimal relaxation are found at the n+/n-

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 2 17

binding ratio of 1.5 to 2.0. For binding ratios larger than 2.0, the surfactant film
collapses at high surface tensions. Similar trends were observed in the presence of
serum. Kang et al. (2008) also observed that although batch-to-batch performance of
manufacturer-supplied BLES samples was variable in 3TC, 100% of RH air, when all
these systems were formulated at the optimal BLES-chitosan ratio, they all reached
the same desirable performance.
To understand the optimal chitosan-BLES ratio, Kang et al. (2008) proposed
the "patch model" illustrated in Fig. 8.18. The patch model implies that, due to the
presence of the cationic polymer, anionic lipids form patches as they bind to these
polymers on the surface of the vesicles and on the multilayer lipid structures adsorbed
at the air-water interface. In that case, the area occupied by the anionic charges of
the lipid (n-) should be equal to the area of the cationic groups in chitosan (n').
Therefore, n+(a+)= n-(a-), where a+ and a- are the area per molecule of the cationic
groups in the polymer and a- is the area per molecule of the anionic li id. In the case
of chitosan, the area per each ionized glucosamine group is about 25
and the area
of the most abundant anionic lipid in BLES, PG, is 40 A2/molecule when it is hlly
compressed, just before film collapse (Kang et al., 2008). This predicts that at optimal
conditions (full compression of the anionic lipid) the ratio of chitosan to the anionic
lipid is n+/n- = 40 A2/25A2 3 1 . 6 This is in agreement with the data presented in Fig.

R,

1 .o

Fig. 8.16. Surface tension (y)-relative-area cyclic compression isotherms for 2.0 mg/mL of BLES
formulationscontainingdifferent levelsofchitosan(0,0.l.and0.25 mg/mL) in theabsence/presence of 50
pUmL of serum. Cycling conditions:37"C, 10096 of RH, 20% of compression, 3 sec/cycle compression.

2 18 0 EJ. Acosta et al.

200
N
-

Film collapse

:
150

100

.-c0
m

-am

50
0

0.5

1.5

2.5

Actual n+/n- binding ratio

18
h

3 12
E

Film collapse

0
0

0.5

1.5

2.5

Actual n+/n- binding ratio

Fig. 8.1 7. (A) Dilatational elasticity ( E ) and (6) relaxation rate at the end of compression (dy/dt) for BLESchitosan systems (no added serum, cycled at 100% of RH, 37"C,20% of compression, and 3 sedcycle
period) as a function of the actual n+/n-binding ratio.The dashed line corresponds to 0.5 mg/rnL of BLES
systems, and the solid lines correspond to 2.0 mg/mL of BLES systems.

Lung Surfactants: Formulation, Evaluation,and Polymeric Additives 0 2 19

8.17. The same model was successhlly used to formulate BLES-cationic polymers by
using different polyelectrolytes and cationic peptides (articles in preparation).

Conclusions and Future Trends

The greatest accomplishment in lung surfactant research is, undoubtedly, the
development of exogenous surfactant formulations that have contributed to saving
the lives of numerous premature babies. The clinical practice revealed that synthetic
surfactants are not as hnctional as surhctants extracted from the lungs of large
mammals such as pigs and cows, and the quest for better surfactants for neonatal
applications still continues. The issues of batch- to-batch variability, immune response
to natural surfactants, some cases of meconium aspiration as well as cases ofventilatorinduced lung injury need hrther resolution. The ultimate frontier for lung surfactant
formulations in terms of their potential to save lives and the largest medical need
are still in the treatment for ARDS. While, at this point, whether surfactant therapy
would ever be effective for ARDS is unclear, but clearly, the current formulations are
not the answer. The effort of several research groups, including ours, on surfactant
additives may open the door for a breakthrough in ARDS therapy. Out of the

Air

Liquid
Chitosan - anionic lipid patches

Chitosan
Fig. 8.18. Schematic of the binding between the ionized glucosamine groups of chitosan and the
anionic lipids in BLES to produce a patch on the surface of the vesicles and on the multilayer structures
adsorbed at the air-water interface.

220 0 EJ. Acosta et al.

three possible additives presented in this chapter, increasing agreement exists in the
surfactant community that nonionic polymers are not effective enough to produce
surfactant preparations that can counteract the action of potential inhibitors found
in the alveolar fluid of ARDS patients. Anionic polymers are a promising alternative.
The potential breakthrough change may come from some form of cationic additive
whether that is chitosan or recombinant SP-C or SP-B, o r peptide analogs of SP-B and
SP-C. As we continue to understand the interaction between these cationic additives
with the lipids in the surfactant preparation, we will be in better shape to formulate
more effective surfactant preparations.

References
Acosta, E.J.; R. Gitiafroz; Y.Y. Zuo; Z. Policova; PN. Cox; M.L. Hair; A.W. Neumann. Effect of
humidity on lung surfactant films subjected to dynamic compression/expansion cycles. Resp.
Physiol. Neurobiol. 2007, 155, 255-267.
Adhikari, N.; K.E.A. Burns; M.O. Meade. Pharrnacologic therapies for adults with acute lung
injury and acute respiratory distress syndrome. Cochrane Database Rev. 2004, 4. Art. No.
CD004477.
Anzueto, A.; R.I? Baughman; K.K. Guntupalli; J.G. Weg; H.I? Wiedemann; A.A. Ravent6s; F.
Lemaire; W. Long; D.S. Zaccardelli; E.N. Pattishall. Aerosolized surfactant in adults with sepsisinduced acute respiratory distress syndrome. N. Engl. /. Med. 1996,334, 1417-1421.
Avery, M.E.; J. Mead. Surface properties in relation to atelectasis and hyaline membrane disease.
Am. /. Dis. Chikf. 1959,97, 517-523.
Asakura, S.; F. Oosawa. On interaction between two bodies immersed in a solution of macromolecu1es.J. Chem. Phys. 1954,22, 1255-1256.
Bachofen, H.; J. Hildebrandt; M. Bachofen. Pressure-volume curves of air- and liquid-filled
excised lungs-surface tension in situ. /. Appl. Physiol. 1970, 29, 422-431.
Bachofen, H.; S. Schiirch; M. Urbinelli; E.R. Weibel. Relations among alveolar surface tension,
surface area, volume, and recoil pressure. J. Appl. Physiol. 1987, G2(5), 1878-1887.
Bailey, T.C.; A.A. Maruscak; A. Petersen; S. White; J.F. Lewis; R.A.W. Veldhuizen. Physiological effects of oxidized exogenous surfactant in vivo: effects of high tidal volume and surfactant
protein a. Am. J. Physiol. 2006, 291, L703-L709.
Bailey, T.C.; K.A.D. Silva; J.F. Lewis; K. Rodriguez-Capote; F. Possmayer; R.A.W. Veldhuizen.
Physiological and inflammatory response to instillation of an oxidized surfactant in a rat model
of surfactant deficiency./. Appl. Physiol. 2004, 96, 1674-1680.
Bangham, A.D.; C.J. Morley; M.C. Phillips. The physical properties of an effective lung surfactant. Biochim. Biopbys. Acta 1979, 573, 552-556.
Bartmann, I?; U. Bamberger; F. Pohlandt; L. Gortner. Immunogenicity and immunomodulatory
activity of bovine surfactant (sf-ri 1). Acta Paediatrica, Int. J, Paediatr. 1992, 81, 383-388.
Been, J.V.; L.J.I. Zimmermann. Whats new in surfactant? A clinical view on recent developments
in neonatology and paediatrics. Eur. J. Pediatr. 2007, IG6, 889-899.
Bengt-Robertson, B.R.; T. Curstedt; J. Johansson. Prospects for a new synthetic surfactant. Acta
Biomedica de PAteneo Parmense 2000, 71, 409-4 12.

Lung Surfactants:Formulation, Evaluation, and Polymeric Additives 0 221

Benson, B.J. Genetically engineered human pulmonary surfactant. Clin. PerinutoL 1993,20(4),
791-8 1 1.
Bernhard, W.; J. Mottaghian; A. Gebert; G.A. Rau; H. von der Hardt; C.F. Poets. Commerical
versus native surfactants. Am. /. Crit. Cure Med. 2000, 162, 1524-1 533.
Bevilacqua, G. et al . Prophylaxis of respiratory distress syndrome by treatment with modified
porcine surfactant at birth: A multicentre prospective randomized trial. J Perinutul Med. 1996,
24(6), 609-620.
Blanco, 0.;J. PCra-Gil. Biochemical and pharmacological difference between preparations of
exogenous natural surfactant used to treat respiratory distress syndrome: Role of the different
components in an efficient pulmonary surfactant. Eur. J. Pburmucol. 2007,568(1-3), 1-1 5.
Bloom, B.T. et al . Comparison of infasurf (calf lung surfactant extract) to survanta (beractant) in
the treatment and prevention of respiratory distress syndrome. Pediutrics. 1997, 100(1), 31-38.
Cabrerizo-Vflcha, MA.; HA. Wege; ].A. Holgado-Terriza;A.W. Neumann. Axisymmetric drop
shape analysis as penetration langmuir balance. &. Scientifir Instr. 1999, 70(5), 2438-2444.
Campbell, H.; K. Bosma; A. Brackenbury; L. McCaig; Li-J. Yao; R Veldhuizen; J. Lewis. Polyethylene glycol (PEG) attenuates exogenous surfactant in lung-injured adult rabbits. Am. J Rrspir.
Crit. Cure Med. 2002, 165, 475-480.
Capetole, R. EX-99.1 Interim report. (http://www.secinfo.com/duwTa.733c.d.htm),
1998.
Cheng, I?; D.Li; L. Boruvka; Y.Rotenberg; A.W. Neumann. Automation of axisymmetric drop
shape analysis for measurements of interfacial tensions and contact angles. Colloids Surj 1990,
43, 151-167.
Clements, J.A. Surface tension oflungextracts. Proc. Sor. Etp. Biol. Med, 1957,95, 170-172.
Cochrane, C.G.; S.D. Rev&. Pulmonary surfactant protein B (sp-B): Structure-function relationships. Science 1991,254,566-568.
Codd, J.R; S.Schurch; C.B. Daniels; S. Orgeig. Torpor-associated fluctuations in surfactant activiry in goulds wattled bat. Biocbim. Biopbs. Arta 2002, 1580, 57-66.
Craster, RV.; O.K.Matar. On the dynamics of liquid lenses. J Cohidlntefuce Sci. 2006,303,

503-516.
Creuwels L.; L. van Golde; H. Haagsman. The pulmonary surfactant system-Biochemical and
clinical aspects. Lung 1997, 175, 1-39.
Curstedt, T.; J. Johansson. New synthetic surfactant-how and when? Biol. Neonate 2006,89,

336-339.
Dargaville, PA.; J.F. Mills. Surfactant therapy for meconium aspiration syndrome: Current status.
Drug 200565,2569-2591.
Davidson, B.A.; ER. Knight; Z. Wang; E R Chess; BA. Holm; T.A. Russo; A. Hutson; 0.1. Del
No;A.W. Neumann. hisymmetric drop shape analysis: Computational methods for the measurement of interfacial properties from the shape and dimensions of pendant and sessile drops. 1.
CoUoidlntefaeSci. 1997, 196, 136-147.
Demarger-Andre, S.; A. Domard. Chitosan behaviours in a dispersion of undecylenic acid. Structural parameters. Curbobydx Polym. 1994, 24, 177-184.
Dunn, M.S.; A.T. Shennan; D. Zayack; E Possmayer. Bovine surfactant replacement therapy in

222 0 EJ. Acosta et al.

neonates of less than 30 weeks gestation: A randomized controlled trial of prophylaxis versus
treatment. Pediatrics 1991,87, 377-386.
Durand, D. J.; R.I. Clyman; M.A. Heymann. Effects of a protein-free, synthetic surfactant on
survival and pulmonary function in preterm lambs. /. Pediatr 1985, 107, 775-780.
Dynarowicz-Latka, I?; A. Dhanabalan; O.N. Oliveira, J . Modern physicochemical research on
langmuir monolayers. Adv. ColloidZnterface Sci. 2001,91, 221-293.
Eijking, E.E; D. Gommers; J. Kullander; E.D. Buijzer; R.W. Beukenholdt; B. Lachmann. Different surfactant treatment strategies for respiratory failure induced by hydrochloric acid aspiration
in rats. Adv. Exp. Med. Biol. 1992,317, 349-355.
Enhorning, G. Pulsating bubble technique for evaluating pulmonary surfactant. J Appl. Physiol.
1977,43, 198-203.
Enhorning, G.; A.T. Shennan; E Possmayer. Prevention of neonatal respiratory distress syndrome
by tracheal instillation of surfactant: A randomized clinical trial. Pediatrics 1985, 76 145-153.
Fisher, M.J.; M.F. Wilson; K.C. Weber. Determination of alveolar surface area and tension from in
situ pressure-volume data. Respir. Physiol. 1970, 10,159-171.
Frerking, I.; A. Giinther; W. Seeger; U. Pison. Pulmonary surfactant: functions, abnormalities and
therapeutic options. Intensive Care Med. 2001,27, 1699-1717.
Fujiwara, T.; M. Konishi; S. Chida; K. Okuyama; Y. Ogawa; Y.Takeuchi; H. Nishida; H. Kito;
M. Fujimura; H. Nakamura; T. Hashimoto . Surfactant replacement therapy with a single postventilatory dose of a reconstituted bovine surfactant in preterm neonates with respiratory distress
syndrome: Final analysis of a multicenter, double-blind, randomized trial and comparison with
similar trials. Pediatrics 1990, 86, 753-764.
Goerke, J. Pulmonary surfactants: functions and molecular composition. Biochim. Biophys. Acta
1998,1408,79-89.
Gortner, L.; I? Bartmann; F. Pohlandt; U. Bernsau; F. Porz; H.H. Hellwege; R.C. Seitz; G.
Hieronimi; C. Bremer; G. Jorch. Early treatment of respiratory distress syndrome with bovine
surfactant in very preterm infants: a multicenter controlled clinical trial. Pediatr. PulmonoL
1992, 14, 4-9.
Gortner, L.; U. Bernsau; H.H. Hellwege; G. Hieronimi; G. Jorch; H. L. Reiter. A multicenter
randomized controlled clinical trial of bovine surfactant for prevention of respiratory distress
syndrome. Lung 1990, 168, 864-869.
Hgner, D.; EG. Germann; D. Hauschke. Effects of rSP-C surfactant on oxygenation and histology in a rat-lung-lavage model of acute lung injury. Am. /. Respir. Crit. Care Med. 1998, 158,
270-278.
Hall, S.B.; M.S. Bermel; Y.T. KO; H.J. Palmer; R.H. Notter. Approximation in the measurement
of surface tension with the oscillating bubble surfactometer. 1.Appl. Physiol. 1993, 75, 468-477.
Halliday, H.L. Overview of clinical trials comparing natural and synthetic surfactants. Biol. Neonate. 1995, 67, 32-47.
Henriksen, I.; G. Smistad; J. Karlsen. Interactions between liposomes and chitosan. Znt. /. Pharm.
1994,101,227-236.
Hoekstra, R.E.; J.C. Jackson; T.F. Myers; I.D. Frantz; M.E. Stern, W.F. Powers; M. Maurer; J.R.
Raye; S.T Carrier; J.H. Gunkel; A.J. Gold . Improved neonatal survival following multiple

Lung Surfactants:Formulation, Evaluation,and Polymeric Additives 0 223

doses of bovine surfactant in very premature infants at risk for respiratory distress syndrome,
Pediatrics 1991, 88, 1&18.
Horbar, J.D. et al . A multicenter randomized, placebo-controlledtrial of surfactant therapy for
respiratory distress syndrome. New Eng. J. Med. 1989,320, 959-965.
Hudak, M.L. et al . A multicenter randomized masked comparison trial of synthetic surfactant
versus calf lung surfactant extract in the prevention of neonatal respiratory distress syndrome.
Pediatrics 1997, 100, 39-50.
Im Hof, V.; I? Gehr; V. Gerber; M.M. Lee; S. Schurch. In vivo determination of surface tension in
the horse trachea and in vitro model studies. Rrspir. Pbysiol. 1997, 107, 81-93.
Jacobs, S.E.; K. O'Brien; S. Inwood; E.N. Kelly; H.E. Whyte. Outcome of infants 23-26 weeks'
gestation pre and post surfactant. h a Pacdatr. 2000, 89, 1149-1 151.
Kang, N.; Z. Policova; G. Bankian; M.L. Hair; Y.Y. Zuo;A.W. Neumann; E.J. Acosta. Interaction between chitosan and bovine lung extract surfactants. Biochim. Biopbys. h a 2008, 1778,
291-302.

Kendig, J.W. et al . Comparison of rwo strategies for surfictant prophylaxis in very premature
infants: A multicenter randomized trial. Pediatrics 1998, 101, 1006-1012.
Khanala, D.R; Y. Okamotoa; K. Miyatakea; T. Shinobub; Y. Shigemasab; S. Tokurac; S. Minamia.
Protective effects of phosphated chitin (P-chitin) in a mice model of acute respiratory distress
syndrome (ARDS). Carbohydr. Polym. 2001,44, 99-106.
Kobayashi, T.; T.Curstedt; G. Grossmann; B. Robertson. Inhibition of exogenous surfactant in
ventilated immature newborn rabbits. Rcspir. Pbysiol. 1989, 76 1-12.
Kobayashi, T.; K. Ohta; K. Tashiro; K. Nishizuka; W. Chen; S. Ohmura. Dextran restores
albumin-inhibited surfice activity of pulmonary surfactant extract./. Appl. Pbysiol. 1999,86
1778-1 784.

Konishi, M.; T. Fujiwara; S. Chida; H. Maeta; S. Shimada; T. Kasai; Y. Fuji; Y. Murakami. A prospective, randomized trial of early versus late administration of a single dose of Surfictant-TA.
Earb Human DmcIop. 1992,29, 275-282.
Kumar, M.N.; RA. Muzzarelli; C. Muzzarelli; H.Sashiwa; A.J. Domb. Chitosan chemistry and
pharmaceutical perspectives. Chem. Rev. 2004, 104, 6017-6084.
Kwok, D.Y.; D. Vollhardt; R. Miller; D. Li; A.W. Neumann. Axisymmetric drop shape analysis as
a film balance. Col Su$ A Pbysicocbem. Aspects. 1994, 88, 51-58.
Kwong, M.S.; E.A. Egan; R H . Notter; D.L. Shapiro. Double-blind clinical trial of calf lung surfactant extract for the prevention of hyaline membrane disease in extremely premature infants.
Pediatrics 1985, 76 585-592.
Lamm, W.J.E.; RK. Albert. Surfactant replacement improves lung recoil in rabbit lungs after acid
aspiration. Am. Rev. Rcspir. Dis. 1990, 142, 1279-1283.
Langmuir, 1. The constitution and fundamental properties of solids and liquids. ii. Liquids. J.Am.
Chcm. SOC.1917,37, 1848-1906.
Lewis, J.F.; A. Brackenbury. Role of exogenous surfactant in acute lung injury. Crit. Care Med.
2003,31, S324-328,
Lewis, J.E; J. Goffin; I? Yue; L.A. McCaig; D. Bjameson; R.A.W. Veldhuizen. Evaluation of

224 0 EJ. Acosta et al.

exogenous surfactant treatment strategies in an adult model of acute lung injury. /. Appl. Physiol.
1996,80,11561164.
Lu, J.J.; WW.Y Cheung; L.M.Y. Yu; Z. Policova; D. Li; M.L. Hair; A.W. Neumann. The effect
of dextran to restore dynamic surface tension properties of pulmonary surfactant inhibited by
albumin. Resp. Physiol. Neurobiol. 2002, 130, 169-179.
Lu, K.; J. Goerke; J. Clements; H. Taeusch. Hyaluronan decreases surfactant inactivation in vitro.
Pediatz Res. 2005a, 55, 237-241.
Lu, K.W.; J. Goerke; J.A. Clements; H.W. Taeusch. Hyaluronan reduces surfactant inhibition and
improves rat lung function after meconium injury. Pediatz Res. 2005b, 58, 206210.
Lu, K.W.; B. Robertson; H. Taeusch. Dextran or polyethylene glycol added to curosurf for treatment of meconium lung injury in rats. Biol Neonate. 2005c, 88, 46-53.
Lu, K.W.; H.W. Taeusch; B. Robertson; J. Goerke; J.A. Clements. Polymer-surfactant treatment
of meconium-induced acute lung injury. Am. 1.Respir. Crit. Care Med. 2000, 162, 623428.
Lu, K.; H. Taeusch; B. Robertson; J. Goerke; J. Clements. Polyethylene glycol/surfactant mixtures
improve lung function after HCI and endotoxin lung injuries. Am. /. Respir. Crit. Care Med.
2001,164, 1531-1536.
Lu, J.J.; L.M.Y.Yu; W.W.Y. Cheung; LA. Goldthrope; Y.Y. Zuo; Z. Policova; I?N. Cox; A.W.
Neumann. Polyethylene glycol (PEG) enhances dynamic surface activity of a bovine lipid extract
surfactant (BLES). Colloidc Surf: B, 2005d, 41, 145-151 .
Lu, J.J.; L.M.Y. Yu; W.W.Y. Cheung; Z. Policova; D. Li; M.L. Hair; A.W. Neumann. The effect
of concentration on the bulk adsorption of bovine lipid extract surfactant (BLES). Coll. Surf: B
Biointefaces. 2003,29, 1 19-1 30.
Markart, I?; C. Ruppert; M. Wygrecka; T. Colaris; B. Dahal; D. Walmrath; H . Harbach; J.
Wilhelm; W. Seeger; R. Schmidt; A. Guenther. Patients with ARDS show improvement but
not normalisation of alveolar surface activity with surfactant treatment: Putative role of neutral
lipids. Borax 2007,62, 588-594.
McCormack, EX. Structure, properties and processing of surfactant protein-A. Biochim. Biopbys.
Acta 1998,1408, 109-131.
McCormack, EX. Functional mapping of surfactant protein A. Pediatr. Pathol. Mol. Med. 2001,
20,293-318.
McDorman, M.F.; H.M. Rosenberg. Trends in infant mortality by cause of death and other characteristics, 1960-1989. Etaland Health Statistics (DDHS publication PHS 93-1857); National
Center for Health Statistics. 1993; pp. 1-51.
McIntyre, R.C.; E.J. Pulido; D.D. Bensard; B.D. Shames; E. Abraham. Thirty years of clinical
trials in acute respiratory distress syndrome. Crit. Care Med. 2000,28, 33 14-333 1.
McMillan, D.; V. Chernick; N. Finer; D. Schiff; H. Bard; J. Watts; R Krzeski; W. Long. Effects
of two rescue doses of synthetic surfactant in 344 infants with respiratory distress syndrome
weighing 750 to 1249 grams: a double-blind, placebo-controlled multicenter Canadian trial. /.
Pediatr. 1995,126; S90-98.
Mezei, A.; R. MkszPros. Novel method for the estimation of the binding isotherms of ionic surfactants on oppositely charged polyelectrolytes. Langmuir 2006,22, 7148-71 5 1.
Mingarro, I.; D. Lukovic; M. Vilar; J. Pkra-Gil. Synthetic pulmonary surfactant preparations:

Lung Surfactants:Formulation, Evaluation,and Polymeric Additives 0 225

new developments and future trends. Cum Med Cbem. 2008, 15,393-403.
Morley, C.J. Ten centre trial of artificial surfactant (artificial lung expanding compound) in very
premature babies. Brit. Med 1.1987,294, 991-996.
Morrow, M.R.; N. Abu-Libdeh; J. Stewart; K.M.W. Keough. Interaction of pulmonary surfactant
protein SP-A with DPPClegg-PG bilayers. Biophys./. 2003,85, 2397-2405.
Morrow, M.R.; J. Stewart; S. Taneva; A. Dico; K.M. Keough. Perturbation of DPPC bilayers by
high concentrations of pulmonary surfactant protein SP-8. Eur. Biophys.J. 2004,33, 285-290.
Moses, D.; B.A. Holm; I? Spitale; M. Liu; G. Enhorning. Inhibition of pulmonary surfactant
function by meconium. Am. J. Obstet. Gynecot! 1991, 164, 477481.
Moya, ER.; J. Gadzinowski; E. Bancalari; V. Salinas; B. Kopelman; A. Bancalari; M.K. Kornacka; TA. Merritt; R segal;C.J. Schaber; H. Tsai; J. Massaro; RDAgostino. A multicenter,
randomized, masked, comparison trial of lucinactant, colfosceril palmitate, and beractant for the
prevention of respiratory distress syndrome among very preterm infants. Pediutrin 2005, 115,
1018-1029.
Moya, F.; A. Maturana. Animal-derived surfactants versus past and current synthetic surfactants:
Current status. Clin. PerinatoL 2007,34, 145-177.
Nag, K.; M. Frinen-Garcia; R Devraj; A.K. Panda. Interfacial organizations of gel phospholipid
and cholesterol in bovine lung surfactant films. Langmuir 2007,23, 44214431.
Nag, K; J. Perez-Gil; M.L. Ruano; LA. Worthman; J. Stewart; C. Casals; K.M. Keough.
Phase transitions in films of lung surfactant at the air-water interface. BiopLys. 1.1998, 74,
2983-2995.
Nag, K.; K. Rodriguez-Capote;A.K. Panda; L. Frederick: S.A. Hearn; N.O. Petersen; S. Schiirch;
F. Possmayer. Disparate effects of two phosphatidylcholine binding proteins, C-reactive protein
and surfactant protein A, on pulmonary surfactant structure and function. Lung Cell MoL
PbysioL 2004,287, L1145-1153.
Nakamura, T.; M. Tamura. Partial liquid ventilation with low dose of perflubron and a low stretch
ventilation strategy improves oxygenation in a rabbit model of surfactant depletion. BioL Neonate 2002, 82, 66-69.
Noack, G.; I? Berggren; T. Cuntedt; G. Grossmann; I? Herin; W. Mortensson; R Nilsson; B.
Robertson. Severe neonatal respiratory distress syndrome treated with the isolated phospholipid
fraction of natural surfactant. Actu Puediutr Scund 1987, 76 697-705.
Notter, R H . Lung Surfactant, Batic Science and CfinicufApplication; Marcel Dekker, New York,
2000.
Notter, RH.; S.A. Tabak; R D . Marvis. Surface properties of binary mixtures of some pulmonary
surfactants components. J. Lipid Res. 1980,2I, 10-22.
Pattle, RE. Properties, function and origin of the alveoli lining layer. Natun 1955, 175,
1125-1 126.
Pattle, RE.; L.C. Thomas. Lipoprotein composition of the film lining the lung. Natun 1%1,
189, 844.
Persson, A.; D. Chang; K. Rust; M. Moxley; W. Longmore; E. Crouch. Purification and biochemical characterization of CP4(SP-D), a collagenous surfactant-associated protein. Biocbem. 1989,
28,6361-6367.

226 EJ. Acosta et al.

Petty, T.L. Overview. Acute Respiratory Distress Syndrome; M. Mattay, Ed.; Marcel Dekker: New
York, 2003; pp. 1-6.
Pinkerton, K.E.; M. Ikegami; L.M. Dillard; A.H. Jobe. Surfactant treatment effects on lung structure and type I1 cells of preterm ventilated lambs. Biol. Neonate 2000, 77, 243-252.
Possmayer, F.A. Proposed nomenclature for pulmonary surfactant-associated proteins. Am. Rev.
Respir. Dis. 1988,138, 990-998.
Possmayer, E; K. Nag; K. Rodriguez; R. Qanbar; S. Schiirch. Surface activity in vitro: role of
surfactant proteins. Comp. Biochem. Physiol. A. 2001, 129, 209-220.
Prokop, R. M.; A. Jyoti; M. Eslamian; A. Garg; M. Mihaila; 0.1. del No; S.S. Susnar; Z. Policova;
A.W. Neumann. A study of captive bubbles with axisymmetric drop shape analysis. Coll. Surf: A
Physicochem. Aspects. 1998, 131, 231-247.
Robertson, B. Factors influencing the clinical response to surfactant replacement therapy in babies
with severe respiratory distress syndrome. Eur. 1.Pediatr. 1991, 150, 433-439.
Robertson, B.; H.L. Halliday. Principles of surfactant replacement. Biochim. Biophys. Actu 1998,
1408,346361.
Rotenberg, Y.; L. Boruvka; A.W. Neumann. Determination of surface tension and contact angle
from the shapes of axisymmetric fluid interfaces.]. ColloidZntel.faceSci. 1983,93, 169-183.
Rubenfield, G.D.; M.J. Neff. Epidemiology of acute lung injury. A public health perspective.
Acute Respiratory Distress Syndrome; M. Mattay, Ed.; Marcel Dekker: New York, 2003; pp.
37-54.
Riidiger, M.; A. Tolle; W. Meier; B. Riistow. Naturally derived commercial surfactants differ in
composition of surfactant lipids and in surface viscosity. Am. 1.Physiol. 2005,288, L379-L383.
Saleem, M.; M.C. Meyer; D. Breitenstein; H.J. Galla. The surfactant peptide KL4 in lipid
monolayers: phase behavior, topography, and chemical distribution. J Biol. Chem. 2008,283,
5 195-5207.
Sarin, I?; H. Taeusch; K. Lu; J. Clements; J. Goerke. Polyethylene glycol results in concentrationdependent sedimentation of Survanta and improvement of its function. Pediatr. Res. 1999, 45,
3 19A.
Schoel, W.M.; S. Schiirch; J. Goerke. The captive bubble method for the evaluation of pulmonary
surfactant: surface tension, area, and volume calculations. Biochim. Biophys. Actu 1994, 1200,
281-290.
Schiirch, S.; H. Bachofen; J. Goerke; F. Possmayer. A captive bubble method reproduces the in
situ behavior of lung surfactant monolayers. 1.Appl. Physiol. 1989, 67,2389-2396.
Schiirch, S.; J. Goerke; J.A. Clements. Direct determination of surface tension in the lung. Proc.
Nut. Acad. Sci. 1976, 73, 4698-4702.
Schiirch, S.; F.H.Y. Green; H. Bachofen. Formation and structure of surface films: Captive bubble
surfactometry. Biochim. Biophys. Acta 1998, 1408, 180-202.
Scott, J.E. Phosphatidylcholine synthesis, secretion, and reutilization during differentiation of
the surfactant-producing type I1 alveolar cell from fetal rabbit lungs. Eup. Lung Res. 1992, 18,
563-580.
Seeger, W.; C. Grube; A. Giinther; R. Schmidt. Surfactant inhibition by plasma proteins: differen-

Lung Surfactants:Formulation, Evaluation,and Polymeric Additives 0 227

tial sensitivity of various surfactant preparations. Eur. Respir.J. 1993,6, 971-977.

Seurynck, S. L.; N.J. Brown; C.W. Wu; K.W. Germino; E.K. Kohlmeir; E.P Ingenito; M.R
Glucksberg A.E. Barren; M. Johnson. Optical monitoring of bubble size and shape in a pulsating bubble surfactometer.J. Appl.. Phyiot! 2005,77,624433.
Seurynck-Servoss,S. L;M.T. Dohm; A.E. Barron. Effects of including an n-terminal insertion
region and arginine-mimetic side chains in helical peptoid analogues of lung surfactant protein
B. Biochm. 2006,45,11809-1 1818.
Smyth, J.; I. Metcalfe; I? DufFy; F, Possmayer; M. Bryan; G. Enhoring. Hyaline membrane disease
treated with bovine surfactant. Prdiutnks 1983, 71, 913-9 17.
Speer, C.P; 0. Gefeller; I? Groneck; E. Laufkotter; C. Roll; L. Hanssler; I. Harms; E. Herting; H.
Boenisch; J. Windeler; B. Robertson , Randomised clinical trial of two treatment regimens of
natural surfactant preparations in neonatal respiratory distress syndrome. Arch. Dis. Childhood
1995,72,F8-Fl3.
Spragg, RG.; J.F. Lewis; H.D. Walmrath; J. Johannigman; G. Bellingan; PF.Laterre; M.C. Witte;
G.A. Richards; G. Rippin; F. Rathgeb; D. Hafher; F.J.H. Taut; W. Seeger . Effect of recombinant
surfactant protein C-based surfactant on the acute respiratory distress syndrome. N . Engl J.
Med. 2004,351,884-892.
Sun, 8.;T. Curstedt; G. Lindgren; 8. Frandn; A.A. Alaiya; A. Cakovska; B. Robertson. Biophysical and physiological properties of a modified porcine surfactant enriched with surfactant
protein A. Eur. Respir.J. 1997,10, 1967-1974.
Sun, B.; T. Curstedt; G.W. Song; B. Robertson. Surfactant improves lung hnction and morphology in newborn rabbits with meconium aspiration. Biol. Neonate 1993,63,96104.
Tabak, SA.;R H . Noner. Modified technique for dynamic surface pressure and relaxation measurements at the air-water interface. h.
Sci. Ins& 1977, 48. 1196-1201.
Taeusch, H.W. Treatment of acute (adult) respiratory distress syndrome: the holy grail of surfactant therapy. Biol Neonate 2000, 7,2-8.
Taeusch, H.W.; J.B. De La Serna; J. Perez-Gil; C. Alonso; J.A. Zasadzinski. Inactivation of
pulmonary surfactant due to serum-inhibited adsorption and reversal by hydrophilic polymers:
experimental. Biopbys. J. 200587, 1769-1779.
Taeusch, H.W.; E. Dybbro; KW. Lu. Pulmonary surfactant adsorption is increased by hyaluronan
or polyethylene glycol. CohkfSu$, B 2008,62,243-249.
Taeusch, H.; K. Lu; J. Goerke; J. Clements. Nonionic polymers reverse inactivation of surFactant
by meconium and other substances. Am.j. Respir. Crit. Can Med. 1999,157,1391-1395.
Taeusch, H.W.; K.M.W. Keough; M. Williams. Characterization of bovine surfactant for infants
with respiratory distress syndrome. Pediuma. 1986,71,572-58 1.
Taneva, S.G.; K.M. Keough. Differential effects of surfactant protein A on regional organization of phospholipid monolayers containing surfactant protein B or C. Biophys. 1.2000, 79,
2010-2023.
Tokieda, K.; J. Whitsett; J. Clark; T.Weaver; K Ikeda; K. McConnell; A. Jobe; M. Ikegami; H.
Iwamoto. Pulmonary dysfunction in neonatal SP-B-deficient mice. Am. J. PtryZiol 1997,273,
L875-L882.
Valta, P;A. Uusaro; S. Nunes; E. Ruokonen; J. Takala. Acute respiratory distress syndrome: fre-

228 0 EJ. Acosta et al.

quency, clinical course, and costs of care. Crit. Care Med. 1999,27, 2367-2374.
Veldhuizen, R.; F. Possmayer. Phospholipid metabolism in lung surfactant. Su6-cell. Biochem.
2004337,359-388.
Von Neergaard, K. Neue Auffassungen iiber einen Grundbegriff der Atemmechanik. Z. Ges. Eup.
Med. 1929, 66,1.
Wakabayashi, T.; M. Tamura; T Nakamura. Partial liquid ventilation with low-dose
perfluorochemical and high-frequency oscillation improves oxygenation and lung compliance in
a rabbit model of surfactant depletion. Biol. Neonate 2006, 89, 177-182.
Walther, F.J.; A.J. Waring; M.A. Sherman; J.A. Zasadzinski; L.M. Gordon. Hydrophobic surfactant proteins and their analogues. Neonatol. 2007,9I, 303-310.
Weaver, T.; J. Whitsett. Function and regulation of expression of pulmonary surfactant-associated
protein. Biochem. J. 1991,273, 249-264.
Wilkinson, A.; PA. Jenkins; J.A. Jeffrey. Two controlled trials of dry artificial surfactant: Early effects and later outcome in babies with surfactant deficiency. Lancet. 1985,2, 287-291.
Willson, D.F.; A. Zaritsky; L.A. Bauman; K. Dockery; R.L. James; D. Conrad; H. Craft; W.E.
Novotny; E.A. Egan; H. Dalton. Instillation of calf lung surfactant extract (calfactant) is
beneficial in pediatric acute hypoxemic respiratory failure. Crit. Cure Med. 1999, 27, 188-195.
Wilson, T.A. Relations among recoil pressure, surface area, and surface tension in the lung. J. Appl.
Physiol. 1981,50, 921-926.
Withsett, J.; L. Nogee; T. Weaver; A. Horowitz. Human surfactant protein B: structure, function,
regulation and genetic disease. Physiol. Rev. 1995, 75, 749-757.
Worthman, L.A.; K. Nag; N. Rich; M.L. Ruano; C. Casals; J. Perez-Gil; K.M. Keough. Pulmonary surfactant protein A interacts with gel-like regions in monolayers of pulmonary surfactant
lipid extract. Biophys. /. 2000, 79, 2657-2666.
Wu, C.W.; S.L. Seurynck; K.Y.C. Lee; A.E. Barron. Helical peptoid mimics of lung surfactant
protein C. Chem. Biol. 2003, 10, 1057-1063.
Wulf, M.; S . Michel; K. Grundke; 0.1. del Rio; D.Y. Kwok; A.W. Neumann. Simultaneous
determination of surface tension and density of polymer melts using axisymmetric drop shape
analysis. J. Colloid Interfce Sci. 1999,210, 172-1 8 1.
Yu, L.M.Y.; J.J. Lu; Y.W. Chan; A. Ng; L. Zhang; M. Hoorfar; Z. Policova; K. Grundke; A.W.
Neumann. Constrained sessile drop as a new configuration to measure low surface tension in
lung surfactant systems. J. Appl. Physiol. 2004a, 97, 704-71 5.
Yu, L.M.Y.; J.J. Lu; I.W.Y. Chiu; K.S. h u n g ; Y.W. Chan; L. Zhang; Z. Policova; M.L. Hair; A.W.
Neumann. Poly(ethy1ene glycol) enhances the surface activity of a pulmonary surfactant. Coll.
SuI$ B. Biointerfaces 2004b, 36, 167-176.
Zasadzinski, J.A.; J. Ding; H.E. Warriner; F. Bringezu; A.J. Waring. The physics and physiology of
lung surfactants. Cur% Opin. CollOidInterf:Sci. 2001,6, 5 0 6 5 13.

Zuo, Y.Y. Biophysical and Gas Transfer Properties of Lung Surfactant and Polymer-Enhanced
Lung Surfactant Films. Ph.D .Dissertation. University ofToronto, 2006.
Zuo, Y.Y.; H. Alolabi; A. Shafiei; N. Kang; Z. Policova; PN. Cox; E. Acosta; M.L. Hair; A.W.
Neumann. Chitosan enhances the in vitro surface activity of dilute lung surfactant preparations

Lung Surfactants:Formulation, Evaluation, and Polymeric Additives 0 229

and resists albumin-induced inactivation. Pediatr. Rrs. 2006, 60,125-130.

Zuo, Y.Y.; C. Do; A.W. Neumann. Automatic measurement of surface tension from noisy images
using a component labeling method. Colloid Surf: A, 2007,299, 109-1 16.
Zuo, Y.Y.; R Gitiafroz; E. Acosta; 2.Policova; EN. Cox; M.L.Hair; A.W. Neumann. Effect of

humidity on the adsorption kinetics of lung surfactant at air-water interfaces. Langmuir 2005,
21, 10593-10601.

Zuo, Y.; R. Veldhuizen; W. Neumann; N. Petersen; E Possmayer. Current perspectives in pulmonary surfactant-Inhibition, enhancement and evaluation. Biocbim. Biopbys. Acta 2008 (in
press).

of
Their
Functional Developments
Dai Kitamoto, Tomotake Morita, Tokuma Fukuoka, and Tomohiro lmura
Bio-ChemicalMaterials Group, National Institute ofAdvanced Science and Technology (AIST),Central 5-2,
Higashi 1-1, Tsukuba,Ibaraki 305-8565,Japan.

Introduction
A variety of amphiphilic compounds, such as complex lipids and proteins, are known
in living organisms. These compounds play an important role in creating a wellordered biological system by relating to the exchange of energies, substances, and
signals at difference interfaces (Holmberg, 2001; Dembitsky, 2005). For example,
phospholipids self-assemble to form the basic structure of biomembranes, where
membrane proteins and glycolipids work as an essential platform for molecular
recognition, signal transduction, transmembrane transportation, cell adhesion, and
so on (Hakomori, 2002; Vankar & Schmidt, 2000). The unique properties of these
amphiphilic compounds arise from the co-existence of a hydrophobic headgroup and
a hydrophilic part, which often prompt the formation of highly organized structures
via self-assembly.
Amphiphilic compounds are generally called surfactants in the industry, and
more than 10 million tons of surfactants are produced each year for the world
market. Surfactants are used in various industries like textile, paper, polymer, plastic,
cosmetics, pharmaceuticals, food, and machinery manufacture (Luk & Abot, 2002).
They are definitely one of the most frequently used chemicals in our daily lives. The
development of the chemical industry provided a wide variety of petroleum-based
chemical surfactants. In addition to chemical surfactants, some biologically based
amphiphilic compounds, such as soaps (fatty acid salt), lecithin (phospholipid),
and saponin (glycolipid) have long been consumed for home and industrial use
before chemical surfactants were produced and became widespread. With increasing
environmental awareness and emphasis on a sustainable society in harmony with
the global environment, these biobased surfactants are becoming much more
important.
Among biobascd surfactants, ones of microbial origin are especially classified
into biosurfactants (BS) (Kitamoto et al., 2002; Banat et al., 2000). BS were first
231

232 0 D. Kitamoto et al.

discovered as extracellular amphiphilic compounds in research into fermentation
processes using petroleum as the starting material, which started in the late 1960s. BS
are produced by a variety of microorganisms from different renewable resources like
vegetable oils and carbohydrates. When BS were first discovered, they mainly attracted
attention as alternatives to chemical surfactants due to their higher biodegradability
and safety. During the last decade, unique properties of BS that are not observed at all
in conventional chemical surfactants, like versatile self-assembling and biochemical
properties, have been found one after another (Kitamoto et al., 2005). In addition
to these advantages, the production efficiency of BS using microorganisms has been
improved along with the progress of biotechnology (Lang, 2002).
In recent years, the development of new functional structures andlor systems using
self-assembly ofamphiphilic molecules has evolved into a dynamic and rapidly growing
area of nanotechnology (Shimizu et al., 2005). There have been numerous reports on
the design and synthesis of such molecules leading to the bottom-up method (Ariga
et al., 2007; Lazzari et al., 2006), different from the top-down method, where we
often suffer from tedious and complicated synthesis with several steps of separation
and purification. As mentioned previously, many of biological amphiphilic molecules
have a tendency to self-assemble into hierarchically ordered structures by elegantly
using hydrogen bonding, hydrophobic forces, and van der Waals interactions. The
possibility of developing intracellular biomolecules into practical functional materials
in nanotechnology, however, is generally far from straightforward due to their limited
amounts and heterogeneity. In addition, the stereo- and regioselective synthesis of
these complicated compounds is generally difficult; this has limited widespread
application of self-assembled biomolecules, especially of glycolipids. Accordingly,
BS, which are able to be abundantly produced from inexpensive natural resources
and exhibit a variety of structures and functions, have been increasingly attracting
attention not only in nanotechnology but also in various fields from energy-saving
technology to advanced biomedical technology (Kitamoto et al., 2005; Mukherjee at
al., 2006).
With special focus on a recent study on glycolipid type biosurfactants, this
chapter deals with the present status of the biosurfactants and discusses the future
possibilities ofsuch biobased functional materials from the viewpoints ofbiotechnology
and nanotechnology. Before providing the physico-chemical properties and potential
applications of BS, the authors will briefly overview the biological aspects of BS. The
reader should peruse other reviews (Ongena & Jacques, 2007; Ron & Rosenberg,
2002; Banat et al., 2000) that provide more detail about the structures and properties
of other types of BS.

Biological Aspects of Biosurfactants

Nature of Biosurfactants
BS are of microbial origin and have the following properties when compared with
chemical surfactants: (i) one or more functional groups and chiral centers, (ii) bulky

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments0 233

but sophisticated structures, (iii) higher biodegradability and lower toxicity, (iv) lower
critical micelle concentration and higher surfice activity, (v) gradual adsorption and
continuing activity, (vi) superior ability to form molecular assembly and liquid crystal,
and (vii) versatile biological activity (antimicrobial and antitumor actions, etc.).
The structures ofsurfactants are largely reflected in that ofinterfacial characteristics.
Typical hydrophilic groups of BS are carbohydrates, peptides, etcetera, while typical
hydrophobic groups are various fitty acids, such as saturated, unsaturated, branched,
or hydroxylated ones.
The combinations of these hydrophilic and hydrophobic groups are very
efficient and elegant in BS. The entire BS structures have probably been refined and
optimized during the integration process of biological evolution in microorganisms.
Moreover, BS are regio- and stereo-selectively synthesized by enzymatic reactions in
microorganisms from simple biomolecules, most of which are chiral compounds
having a unified molecular configuration. BS are thus able to exhibit excellent
orientation and packing properties in their interfaces. These structural features allow
BS to perform more surprising actions than conventional chemical surfactants.
BS are mainly classified into four categories, (i) glycolipid type (Kitamoto et al.,
2002; Lang, 20021, (ii) fatty acid type (Zhang et al., 2004a), (iii) lipopeptide type
(Ongena &Jacques, 2007) and (iv) polymer type (Ron & Rosenberg, 2002), based
on the structure of their hydrophilic part. The chemical structures of representative
glycolipid BS and their producers are presented in Figures 9.1 to 9.7 and Table 9.1,
respectively.

MEL-A: R = R = A c
1
2
MEL-B: R = A c , R = H
1
2
MEL-C: R = H, R = AC
(n = 6 to 10)
Fig. 9.1. Chemical structures of mannosyletythritollipids.

234 0 D. Kitamoto et al.

Acidic forms

HOaH O o

Lacmne forms

R 1 =R2 =HorAc

Fig. 9.2. Chemical structuresof sophorolipids.

R L-A: R = 2-decenoyl

R L-B: R 3 = 2decenoyl

R2=xcooH
R4=Tc0
CH3

Fig. 9.3. Chemical structures of rhamnolipids.

CH 3

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 235

roR

TL-1: RI

OH

=R2=

H3

H 3C
TL-2: RI =

CH3 R 2 = H

(m + n = 27 to 31)
Fig. 9.4. Chemical structures of trehaiose lipids.

HO

OR3

OR1

OH

STL-1: R,-R, = 2 x succinoyl + 2 x hexadecanoyl

STL-2: Rl-R, = 1 x succinoyl + 2 x hexadecanoyl, R, = H
STL-3: R1-R4 = 1 x succinoyl + 3 x hexadecanoyl
Fig. 9.5. Chemical structures of succinoyl trehalose lipids.

236 0 D. Kitamoto et al.

O
&
$
HHO

!OH
4,
"

CL-A:R,=OHorH,

OH

R2=R3=

R4=H

CL-B:R,=OHorH, R 2 = H , R3=Ac

0
R4=

OH
(

H2)nCH3 (n = 2 or 4)

C L-C: R = OH or H, R 2 = CH,, R , = Ac

R4=

M ( C H 2 l n C H 3 (n = 2 or 4)

Fig. 9.6. Chemical structures of cellobiose lipids.

.OH

OH
G L-1: R 1 = CO(C H,),C H=CH(C Hz,),CH,
(n m = 12, 14)
R, =CO(CH,)nCH, (n = 0 , 2, 4)

GL-2: R = H, R, = CO(CH,)nCH, (n = 6 to 8)
GL-3: R , =P-galactose, R,=CO(CH,),CH,

Fig. 9.7. Chemical structures of oligosaccharidelipids.

( n = 6 to81

Self-assemblingPropertiesof Glycolipid Biosurfactants and Their Functional Developments0 237

Table 9.1. Major Types of Glycolipid Biosurfactants

Biosurfactant

Microorganism

Reference

Mannosylerythritol lipid

Pseudozyma aphidis

Rau et al., 2005

Pseudozyma antarctica

Kitamoto et al., 2002

Sophorolipid

Pseudozyma graminicola

Morita et al., 2008a

Pseudozyma hubeiensis

Konishi et al., 2008a

Pseudozymaparantarctica

Morita et al., 2007a

Pseudozyma rugulosa

Morita et al., 2006

Pseudozyma siamensis

Morita et al., 2008b

Pseudozyma shanxiensis

Fukuoka et al.. 2007a

Pseudozyma tsukubaensis

Fukuoka et al., 2008

Ustialgo maydis

Spoeckner et al., 1999

Candido apicola

Hommel et al., 1994

Candido bombicola

Van Bogaert et al, 2007

Candidobatistae

Konishiet al., 2008b

Rhamnolipid

Pseudomonos aeruginosa

Soberth-Ch6vezet al., 2005

Trehalose lipid

Rhodococcuserythropolis

Lang & Philp, 1999

Rhodococcusopacus

Niescher et al., 2006

Rhodococcus ruber

Philp et al., 2002

Succinoyl-trehaloselipid

Rhodococcuserythropolis

Lang & Philp, 1999

Cellobiose lipid

Ustilago maydis

Spoeckner et al, 1999

Oliaosaccharidelipids

TsukamurellaSO.

Lanaer et al.. 2006

Uchida et al., 1989

Physiological Role of Biosurfactants

Assimilation of WatclcinsolubkSubstintu
The physiological function of BS in a BS-producing microorganism is not fully
understood. However, there has been speculation about their involvement in
emulsification of water-insoluble substrates. When microorganisms are cultivated on
n-alkanes or vegetable oils, some growth-stimulating compounds, that is BS, are often
accumulated in the culture medium (Kitamoto et al., 2002). BS appear to have roles
emulsihing the substrate, extending the interfacial area between the microorganism
and the substrate, and facilitating mass transfer on the surface of microorganism.
In fact, the growth of various bacterial strains, such as Pseudomom amginosa, on
n-alkanes is stimulated by addition of a glycolipid BS, namely rharnnolipid (RL,Fig.
9.3). It specifically activates Psnrdomonas type of bacteria and has no effect on other
genera of bacteria (Nitschke et al., 2005; Soberbn-Chdva et al., 2005).

238 0 D. Kitamoto et al.

In addition to emulsification of water-insoluble carbon sources, BS is estimated

to support the adhesion of microbial cells to such substrates. Another glycolipid BS,
trehalose lipids (TL Fig. 9.4), which are one of the cell wall-associated compounds
(Lang & Philp, 1998), are certainly involved in cellular adaptation to the presence of
highly water-insoluble substrate like n-alkanes. Trehalose lipids render the cell surface
hydrophobic, which may then facilitate the attachment and subsequent passive
transport of the substrates into the cell. Trehalose lipids appear to work as a wetting
agent between the substrate and bacterial cell surfaces.
The generalization that BS are prerequisite for the water-insoluble substrate
uptake, however, does not explain all physiological observations. For example, there
have been many reports on the production of glycolipid BS, such as rhamnolipids
(Nitschke et al., 2005; Sober&-Chdvez et al., 2005) and sophorolipids (Van Bogaert
et al., 2007; Lang, 2002), from water-soluble substrates, such as glucose.

Protection Against Osmotic Stress

Yeasts belonging to the genus Candida produce BS that have been suggested to play a
physiological role different from the previously mentioned view to act as a prerequisite
for substrate uptake. A yeast strain of Candida apicola produces sophorolipids (SL,
Fig. 9.2) when grown on glucose or fructose (Van Bogaert et al., 2007). The lipid
production may thus be regulated by carbohydrate and not relate to the uptake of
water-insoluble substrates.
As C. bombicola and C. apicola by nature occur in high sugar concentration
environments with high osmotic strength, sophorolipid production may be a way
of dealing with the environment. Halotolerant yeasts like Pichia strains in fermented
foods ofien produce polyol compounds, such as glycerol and mannitol, to regulate
the osmotic pressure of the growth environment (Saha, 2003). Sophorolipids are thus
considered to act as intracellular and extracellular antiosmosis materials and to lead
to adaptation to such osmotic stress.

Storage of Carbon and Energy Sources

A yeast strain of Pseudozyma antarctica efficiently produces mannosylerythritol lipid
(MEL, Fig. 9.1) from vegetable oils, but only minor amounts from carbohydrates
(Morita et al., 2007b). The yeast, however, accumulates a large amount of MEL in the
cell regardless of carbon sources supplied. The accumulated level of MEL reaches 10%
or more of the dry weight of the yeast, which is a surprisingly high level compared with
the usual amounts of cellular glycolipids in yeasts (Kitamoto et al., 20O2), indicating
that the glycolipid BS would serve as energy storage materials.
Sophorolipid synthesis is associated with nitrogen starvation. The glycolipid
formation is thus likely to be some sort of overflow metabolism, by means of
extracellular storage material. This hypothesis is supported by the finding of Hommel
et al. (1994), regarding the biosynthesis of the sophorose moiety, which resembles
trehalose synthesis of Saccharomyces cerevisiae under anaerobic conditions. In fact,

Self-assernblingPropertiesof Glycolipid Biosurfactantsand Their Functional Developments0 239

sophorose can be used as the sole carbon source by C. bombicokz (Van Bogaert, 2007).
The proposed physiological roles of BS are illustrated in Fig. 9.8 according to the
previous studies (Kitamoto et al., 2002; Van Bogaert, 2007; Nitschke, 2005). These
roles very likely originate from their unique amphiphilic structures.

PhysicochemicalAspects of Biosurfactants
Interfacial Properties of Glycolipid Biosurfactants
BS are surface-active compounds capable of reducing surface and interfacial tension
at the interfaces between liquids, solids, and gases, thereby allowing them to mix or
disperse readily as emulsions in water or other liquids. BS, especially glycolipid BS,
show versatile surface-active properties including emulsiGing, dispersing, solubilizing,
foaming, penetrating, and wetting actions. Table 9.2 shows some examples of surface
and interfacial tension-lowering activities of glycolipid BS.

Mannosykrythritol Lipid
MEL-Aand MEL-B (Fig. 9.1) exhibit excellent surface and interfacial tension-lowering
actions and low critical aggregate concentrations (CAC), although the hydrophobic
parts consist of medium-chain fatty acids ranging from C, to C , , (Kitamoto et al.,
2002; Imura et al., 2006). O n the Wilhelmy method, MEL-A shows a CAC at 2.7
x lo4 M and can reduce the aqueous surface tension and interfacial tension against
n-hexadecane to about 28 and 2 mN/m, respectively. The CAC of MEL-A and
MEL-B obtained by the pyrene fluorescence method are 4.0 x 104 and 6.0 x 104 M,
respectively (Imura et al., 2006).

Fig. 9.8. Physiologicalroles of biosurfactants.

240 0 D. Kitamoto et al.

Table 9.2. Surface and InterfacialTension of Glycolipid Biosurfactants

Biosurfactant

Surface tension

Interfacial
tension

cmc
(M)

ycmc
(mN/m)

(mN/m)

Reference

Mannosylerythritol lipid
MEL-A (f?antarctica)

2 . 7 ~lod

28.4

2.1 a

Kitamoto et al., 2002

MEL-B (f! antarctica)

4 . 5 ~l o 6 28.2

2.4a

MEL-C (f! hubeiensis)

6 . 0 ~lo4

25.1

Kitamoto et al., 2002

Konishi et al., 2008a

Lang et al., 2002

lshigami et al., 1987

SoDhoroliDid

SL-5 (C. batistae)

138d

39.3

SL-mixture (C.batistae)

366d

40.2

RL-A (Na salt)

6.2~

28.0

O.Zb (0.1%)

RL-B (Na salt)

1.5 x

28.0

3.2 (0.05%) lshigami et al., 1987

SL-1 (C.bomibocola)

40d

35

2-tetradecyl-SL

40

30

n-butvlamide-SL
n-decylamide-SL

80

45
60

5d

Lang et al., 2002

Lanq et al., 2002
Lang et al., 2002
Konishi et al., 2008b
Konishi et al., 2008b

Rhamnolipid

lo4

RL-1

20d

26

4=

Lang & Wullbrandt, 1999

RL-2

lod

27

<1C

Lanq & Wullbrandt, 1999

RL-3

200d

25

<1

Lang & Wullbrandt, 1999

RL-4

ZOOd

30

<lC

Lang & Wullbrandt, 1999

TL- 1

4d

36

17<

Lang & Philp, 1999

TL-2

4 d

32

14

Lanq & Philp, 1999

26

<1

Lang & Philp, 1999

9 . 6 ~ 1 0 30
~

8b(0.2%)

Lang & Philp, 1999

1 . 5 ~ 1 0 35
~

<lo'
<lo'

Vollbrecht et al., 1999

<10e

Vollbrecht et al., 1999

Trehalose lipid

Trehalose-2,2:3,4-tetraester
STL-1

+ STL-2 (Na salt)

Oligosaccharide lipid
GL-1

15

GL-2

1.1 x i 0 4

GL-3

9.1 x 10-5 24

cmc: critical micelle concentration.

n-Tetradecane at 25C.
Kerosene at 30C.
n-Hexadecane at 40C.
mg/L.
en-Hexadecaneat 25C.

23

Vollbrecht et al., 1999

Self-assembling Propertiesof Glycolipid Biosurfactantsand Their Functional Developments0 241

O n high-level MEL producers, such as I! antarctica (Kitamoto et al., 2002), I!

aphidis (Rau et al., 2005), and I! rugulosa (Morita et al., 2006), MEL-A invariably
consists of more than 70% of the total MEL produced in the culture medium.
Recently, Konishi et al (2008a) reported that I! hubken& KM-59 mainly produces
MEL-C (Fig. 9.1) together with MEL-A and -Bas the minor components. The newly
identified MEL-C, which has a different fatty acid profile compared to MEL-A or
-B, exhibits a CAC at 6.0 x 10" M and can reduce the aqueous surface tension to
about 25 mN/m. The MEL-C shows a higher CAC and hydrophilicity compared to
conventional MEL, retaining an excellent surface-tension-lowering activity.
The emulsifying activity ofMEL-A towards soybean oil and n-tetradecane is much
higher than that of Tween 80 (polyoxyethylene sorbitan monooleate). The molecular
occupation area of MEL-A is about 60 A2/molecule at the air-water interface (25C);
it thus has an excellent packing property despite the bulky structure. Moreover,
MEL improves the rheological characteristics of flour products, like bread, and its
carbohydrate backbone, mannosylerythritol, shows a moisturizing effect (Kitamoto
et al., 2002).
Recently, Worakitkanchanakul et al. reported (2008a) the formation of stable
water-in-oil W/O) microemulsions based on the single component of MEL-A.
Dynamic light scattering (DLS)and freeze fracture electron microscopy (FF-EM)
measurements revealed that the diameter of the microemulsion increases with
an increase in water-to-surfactant mole ratio (WJranging from 20 - 60 nm, and
the maximum W,value was found to be 20, which is as high as that of soybean
lecithin (Fig. 9.9). Thus, MEL-A has great potential value for the formation o f W / O
microemulsion without using any co-surfactants.

SophotvIipircs
Sophorolipids produced by Candida bombicola, which are inevitably a mixture of
acid- and lactone-form sophorolipids (Fig. 9.2), lower the aqueous surface tension
below 40 mN/m with a critical micelle concentration of 40 - 100 mg/L; they are not
so effective at stabilizing oil-in-water emulsions. They can be modified into various
derivatives having a different hydrophilic-hydrophobic balance. Their derivatives
show a wide range of surface activities, such as emulsifling, wetting, cleaning, and
solubilizing (Shete et al., 2006). In particular, the propylene glycol adduct shows
an excellent hygroscopic activity, and is commercialized as a skin moisturizer and
softener in cosmetics.
Zhang et al. (2004b) recently synthesized different sophorolipid alkyl (methyl,
ethyl, propyl, and butyl) esters, and examined the effect of n-alkyl ester chain length
on the interfacial properties of the corresponding analogs. The CMC of the esters
decreases to about 1/2 per additional CH, group to the alkyl ester moiety. Interestingly,
these surfactants absorb strongly on alumina but weakly on silica. O n these esters, the
average area per molecule at the aidwater interface is around 80 A*, and is larger
than those of other surfactants with structure similar to sugar-based n-dodecyl-P-Dmaltoside of 50

A2.

242 0 D. Kitamoto et al.

100
c

80

E
E

L.
0)
*,

60
40

E
20
0
0

12

16

100 nm

Fig.9.9. Mean diameter of colloidal structures obtained from ternary mannosylerythritollipid-A (MEL-A)/
waterln-decane system (n-decanelMEL-A molar ratio was 4.8) as a function of water-to-surfactant mole
ratio (W,) at 25C (upper figure). Freeze fracture electron micrographs of colloidal structures obtained
from the above ternary MEL-A/water/n-decane system (lower figure). Bar length is 100 nm. (a) W, = 1.5,
(b) W, = 8.3, and (c) W, = 14.3.

Self-assernbling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 243

The surface and interfacial activities of sophorolipid homologs become stronger

as the lactone ring is opened. The interhcial tension of acid-form sophorolipids is
similar to that of sodium dodecyl sulfate (SDS). Removal of trioleoylglycerol from
aqueous solutions by the acidic sophorolipids is the same level as for n-dodecyl-P-Dmaltoside, SDS, and linear alkylbenzene sulfonate ( U S )(Kitamoto et al., 2002).
Konishi et al. (2008b) recently reported that Candih batistae mainly produces
acid-form sophorolipids (more than 60% of the total glycolipids), considerably
different from conventional sophorolipids that mainly constitute those that produce
lactone. The CMC and yCMC of the present sophorolipid mixture are 366 mg/L
and 40.2 mN/m, respectively. In contrast, those of conventional sophorolipid
mixture were 17 mg/L and 36.4 mN/m, respectively. From these results, the present
sophorolipids is likely to show much higher hydrophilicity and water solubility
compared to conventional surfactants, and should be expected to contribute a broad
range of applications of glycolipid BS.
Sophorolipids are commercially used not only for moisturizers and softeners in
cosmetics (Kitamoto et al., 2002) but also for dishwashing detergents. SARAYA Co.,
Ltd. (Osaka, Japan) recently developed the efficient production of sophorolipids from
glucose and soybean oil, and is now providing a new type of dishwashing detergent
including sophorolipids; this detergent shows much higher biodegradabilitycompared
to conventional dishwashing agents like ethylene-propylene block co-polymers.

Rhamnofipith
Rhamnolipids (Fig. 9.3) show good surface activities, including emulsifying,
dispersing, foaming, and penetrating actions. They display a lower CMC (10' to lov5
M) despite being an anionic surfactant (Lang & Wullbrandt, 1998). They reduce the
surface tension of water from 72 mN/m to values below 30 mNlm and the interfacial
tension of waterloil systems from 43 mN/m to values below 1 mNlm. Sodium salts
of RL-A and RL-B have the micellar weight of 38,000 and 7,000 in phosphate buffer
saline (pH 7.35) at 3VC, respectively. ?he molecular occupation area of sodium salt
of RL-B is 79.1 AVmolecule at the air-water interface (30C). These data show that
the lipids have an excellent packing property despite their bulky and complicated
structure (Ishigami et al., 1987).

TmbaloseLip&
Trehalosc lipids (Fig. 9.4) are chemically stable and their surface activities are
independent over a wide range of temperature, pH values, and salt concentrations.
Trehalose di-corynomycolate (TL-1) and mono-corynomycolate (TL-2) reduce the
aqueous surface tension from 72 mN/m to 36 and 32 mN/m with a CMC of 4 mg/L,
and reduce the interhcial tension of waterln-hexadecane system from 43 mN/m to
17 and 14 mN/m, respectively.
Succinoyl-trehaloselipids (STL- 1 and -2), which are produced from n-hexadecane
by R. clythrupulfi SD-74 (Uchida et al., 1989), have both carboxyl and hydroxyl

244 0 0.Kitarnoto et al.

groups in the molecule, and thus exhibit high surface activities; especially dispersing
and dispersion-stabilizing activities towards solid particles such as red iron oxide
(a-Fe,O,), carbon black, and a-copper phthalocyanine blue (Lang & Philp, 1998).

Environmentally Advanced Surjactants

Currently, sugar-based surfactants as synthetic glycolipids are being manufactured
(Stubenrauch, 20O1), and play important roles in the surfactant and detergent
industry. Typical examples of the surfactants are alky(po1y)glycosides and sucrose fatty
acid esters, which are now widely used in formulations for personal care products,
surface cleaners, laundry detergents, and agricultural applications. The industrial
interest in sugar-based surfactants lies in the fact that they can be synthesized from
renewable resources: for example, glucose from starch on the one hand and fatty
alcohol from vegetable oils on the other. As described previously, glycolipid BS are
directly produced by microbial processes from renewable resources, and thus have
a great potential for environmentally advanced surfactants as well as sugar-based
surfactants.
Glycolipid BS are thus invaluable materials because they are not only highly
functional but also highly biocompatible compared to conventional petroleumbased surfactants. In the near future, the risk in the use of such chemical surfactants
should be greatly reduced along with the progress of glycolipid BS.

Self-assembling Properties of Glycolipid Biosurfactants

As indicated previously, glycolipid BS are highly advantageous for large-scale
production and have a wider range of applications compared to other types of BS.
In this section, the self-assembling properties of glycolipid BS will be discussed in
detail. The saccharide parts of glycolipid BS are generally soluble in water, while
the hydrocarbon parts are insoluble in water. The insolubility of the hydrocarbon
chain in water causes glycolipids to concentrate at the air-water interface with their
hydrocarbon chain oriented toward the vapor phase.
The stereochemistry of the saccharide headgroups displays versatile influences
on the self-assembly of glycolipid/water systems (Kitamoto et al., 2005; Hato,
2001). In addition to the type of saccharide, the number of saccharide residues
and the type of linkages yield important consequences for the self-assembly. For
example, n-octyl-P-D-glucopyranoside, n-decyl- and n-dodecyl-P-D-maltoside, and
n-octanoyl-N-methylglucamideare soluble at room temperature, while n-octyl-a-Dglucopyranoside, n-dodecyl-P-D-cellobioside,
and N-octyl-glucamide are insoluble at
room temperature and are difficult to handle (Kitamoto et al., 2005).
Some glycolipid-type surfactants, which possess relatively large hydrophilic
headgroups as compared to the hydrophobic parts (e.g., BS with a single and relatively
short alkyl chain attached to mono- or oligo-saccharide headgroups), generally form
micelles in a dilute aqueous solution (Soderman & Johansson, 2000). In addition
to spherical micelles, they also form oblate (disk-like) and prolate (rod-like) ones
(Stubenrauch, 2001). As the surfactant concentration further increases, glycolipidl

Self-assembling Properties of Glycolipid Biosurfactants and Their Functional Developments0 245

water systems start to form a range of liquid crystalline phases. In particular, glycolipid
BS spontaneously self-assemble into a variety of molecular assemblies with welldefined and/or unique structures: sponge (coacervate, LJ, cubic (V,), hexagonal (HJ,
or lamellar (La) to name a few (Imura et al., 2007a). These structures are restricted
by a glycolipids need to keep its hydrophobic and hydrophilic parts surrounded in a
favorable way, thus minimizing the free energy of the system (Hato, 2001).
Among these molecular assemblies, vesicles are one of the most intensively
studied and are prepared from natural glycolipids such as macrocyclic tetraether
glycolipids in archaebacteria and digalactosyldiacylglycerols in plants. Vesicles made
from phospholipids (liposomes) have been studied over the last three decades. In
contrast to liposomes prepared from natural phospholipids, investigations on vesicles
made from glycolipids, both naturally occurring and synthetic, have been relatively
rare. This appears due to the limited amount and heterogeneity of natural glycolipids,
and to the high values of 7K for synthetic glycolipids (Kitamoto et al., 2005).
As mentioned previously, glycolipid BS are easy to prepare in large amounts by
microbial processes compared to membrane glycolipids or complicated synthetic
surfactants. In addition, the hydrophobic part of glycolipid BS constitutes a variety of
fatty acids, such as short or medium-chain, unsaturated, and/or branched ones. This
enables the BS to lower the 7K and to be always in a fluid state, and then makes it
easier to control their liquid crystal structures at low temperatures (Hato et al., 1999).
Accordingly, glycolipid BS-based vesicles or bilayer membranes appear to be very
promising and may open new avenues for exploiting usehl nanostructured materials
and/or systems.

Mannosyhythritol Lipid
The self-assembling manner of MEL-A and MEL-B is illustrated in Fig. 9.10.
Kitamoto et al. recently reported that MEL-B and -C (Fig. 9.1), which possess alkyl
chains at the C-2 and C-3 positions, spontaneously form giant unilamellar vesicles
of diameters larger than 10 pm (Kitamoto et al., 2000). Giant vesicles can serve
as biophysical model systems; they have been used as cell models for studying the
dynamic and structural features of many cellular processes, including endcytosis,
exocytosis, cell fusion, viral infection, and transport phenomena (Dobereiner, 2000).
However, it is generally difficult to obtain giant vesicles from glycolipids, because
the vesicle structure requires strictly balanced hydrophobic and hydrophilic groups.
This indicates that these MEL have an excellent molecular orientation property
and superior balance between hydrophilic and hydrophobic groups. As described
below, vesicle-forming lipids such as MEL-B or MEL-C can be applied to various
kinds of drug- and gene-delivery systems, coupling with other carrier materials like
phospholipids and polymers.
In contrast, MEL-A (Fig. 9.1) self-assembles to form a sponge phase at
concentrations above 1 mM, which has been often called L3 phase, plumbers
nightmare, or blue I phase (Imura et al., 2004, 2006). Menger et al. (2000)

246 0 D. Kitarnoto et al.

MEL-A

MEL-B
Fig. 9.10. Self-assernblingmanners of rnannosylerythritol lipid-A and -B.

recently demonstrated the spontaneous formation of the sponge phase from a single
compound of a synthetic gemini surfactant, and they proposed that the surfactant
phase of sponge morphology should be redefined as coacervates. Coacervates are
well known as the origin of life, from which all present-day cells bearing ordered
bilayer membranes are generated. The study on the structure-property interaction of
coacervates is thus of great interest, considering how the ordered bilayer membranes
of cells have been constructed from coacervates. However, coacervates now known
are generally prepared from complicated multicomponent systems such as surfactants
with salt/co-solvent or two oppositely charged polyelectrolytes (De Kruif et al., 2004;
Maldonado et al., 2007). This makes their structural characterization difficult.
Nevertheless, MEL-A is the first natural compound to display the formation of
coacervates by themselves, while MEL-B and -C form vesicles. The manner reveals
that that only a slight decrease in spontaneous curvature resulting from the absence
of the 4-0- or 6-O-acetyl group induces a drastic morphological change in the selfassembled structure from coacervates to vesicles, namely ordered bilayer membranes
(Imura et al., 2004). The FF-TEM observation on the MEL-A assemblies clearly
shows the typical coacervate morphology of a sponge phase composed of a randomly
connected 3D network of the bilayers. In many parts of the micrograph, we can

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 247

follow the bilayer over a range of several micrometers, indicating the bilayer is indeed
continuous.
The spontaneous curvature (HJ is a usehl parameter to characterize the lipidassembled structures. It is given by H,= 114,where H, is the curvature of one
monolayer which forms a bilayer and 4 is the spontaneous radius of curvature.
When the spontaneous curvature is nearly zero, the lipids self-assemble into the
lamella phase (La) (Imura et al., 2006). The spontaneous curvature of MEL-B or
MEL-C assemblies is thus nearly zero since vesicles are obtained by the dispersion of
the La phase. These La phases seem to be stabilized by the hydrogen-bonding network
between the headgroups resulting from the C-4 or C-6 hydroxyl group because
La phase is less dynamic than L, phase (Imura et al., 2004). O n the other hand,
the presence of an acetyl group on the mannose moiety is likely to induce a slightly
negative spontaneous curvature of the glycolipid assemblies; this should lead to the
formation of the L, phase.
L, phase obtained from MEL-A would be a potential host for water-soluble
proteins and DNA using the interior water channels of the phase, because the L, phase
resembles the bicontinuous cubic phase. Although the cubic phase acts as a better
solvent for membrane protein bacteriorhodopsin crystallization, it is a stiff liquid
crystal, making it difficult to handle in high throughput screening systems (Ridell et
al., 2003). In contrast, the L, phase is fluid, giving it the potential to overcome this
problem.
Another interesting feature of sponge phase (LJ is the basis of thermodynamically
stable vesicles. Watanabe et al. (2001) reported that the thermodynamically stable
vesicle phase (La,) exists next to the sponge phase (L,) by the phase diagram
determination study of surhctant mixed systems. Imura et al. (2005) recently
demonstrated the formation of thermodynamically stable vesicles with a high
dispersibility from the above-mentioned MEL-A sponge phase by adding L-a-Ddilauroyl-phosphatidylcholines (DLPC). They investigated the mixed system of
MEL-A and DLPC and confirmed the existence of three regions of self-assembly; (1)
droplets with a sponge phase (L,) with diameters from 2 to 20 pm &
, I O.I), (2)
thermodynamically stable vesicles (La,) (0.2 I %LK 5 0.7), (3) multilamellar vesicles
(La)with diameters from 2 to 10 pm
2 0.8). Figure 9.1 1 shows the trapping
efficiency for calcein and visual observation of MEL-NDLPC mixed self-assemblies.
Interestingly, the average size of the vesicles at the composition of
= 0.3 was
633.2 nm, which is remarkably small compared to other compositions (Fig. 9.1 1).
Moreover, the obtained vesicle solution was slightly bluish and turbid and kept its
dispersion stability at 25C for more than 3 months. The asymmetric distribution
of MEL-A and DLPC in the two vesicle monolayers caused by the difference in
geometrical structures is very likely to have changed their self-assembled structure
from a sponge phase (L,) to a thermodynamically stable vesicle (La,) (Imura et al.,

KLPc

qLpC

2005).
Recently, the self-assemblingproperties of MEL-A and -Bhave been characterized
in more detail using the fluorescence probe method, DLS analysis and synchrotron

248 0 D. Kitamoto et al.

FF-fEM

Fig. 9.1 1. Thermodynamic vesicle formation from mannosylerythritol lipid-A and a-D-dilauroylphosphatidylcholine.

small/wide- angle X-ray scattering (SMWAXS) (Imura et al., 2006). Figure 9.12
shows the schematic diagram for the self-assembled nanostructures prepared from
MEL-A and -B in aqueous solutions.
Surfactants generally form spherical micelles above CMC, and the radius of micelle
is several nanometers (almost equal to its hydrophobic chain length). Surprisingly,
both MEL-A and -B exhibit excellent self-assembling properties at extremely low
concentrations; they self-assemble into large unilamellar vesicles ( L W ) of diameter
lager than 160 nm, just above their critical aggregation concentration (CAC,). The
CAC, value is 4.0 x l o 6 M for MEL-A and 6.0 x lo4 M for MEL-B, respectively.
Moreover, the self-assembled structure of MEL-A over CAC,, of 2.0 x l o 5 M
drastically changes into sponge structures (L,) composed of randomly connected
bilayers network as indicated above. The average water channel diameter (6) of the
sponge structure is approximately 100 nm and is relatively large compared with those
obtained from synthetic surfactant multi-component systems (24 nm), that is,
cetylpyridinium/hexanoI/dextrose/brinesystem (Imura et al., 2006).
Meanwhile, MEL-B which has a hydroxyl group at the C-4 position on mannose
instead of an acetyl group gives only one CAC; the self-assembled structure of MEL-B

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 249

Fig. 9.12. Nanostructuresof self-assemblies prepared from mannosylerythritollipid-A and -6.

seems to gradually move from L W to multilamellar vesicle (MLV) whose lattice

constant is 4.4 nm, depending of the concentration. The bilayer thickness is 3.2 nm
for both MEL-A and -B.
Figure 9.13 illustrates the lyotropic liquid crystal phases of MEL-A and -B
obtained by the water penetration method. MEL-A forms various liquid crystals that
are clearly different from those of MEL-B. Therefore, the difference in spontaneous
curvature of assemblies between the two MEL, which is caused by only one acetyl
group on the headgroup, is very likely to decide the direction of the molecular selfassembly. The unique and complex molecular structures that are naturally engineered
by microorganisms through evolution would lead to the sophisticated self-assembling
properties of the yeast glycolipid BS.
Based on SAXS measurements, Imura et a1 (2007a) recently obtained the binary
phase diagram ofMEL-A/water system (Fig. 9.14). MEL-A self-assemblesinto a variety
of distinctive lyotropic liquid crystals including sponge (LJ, bicontinuous cubic (V,),
and lamella phases (La). The proposed space group of the bicontinuous cubic phase
(V,) is la34 and the estimated lattice constants were 11.39 nm for V, and 3.58 nm for
La. DSC measurement revealed that the phase transition enthalpies from the liquid
crystal to the fluid isotropic phase (FI) are in the range of 0.22-0.44 kJ/mol. Very
interestingly, the MEL-A sponge phase region (L,) is spread considerably over a wide
temperature range (2045C) compared to those of hydrophobic polyoxyethylene or
fluorinated surfactants. This is probably due to the stabilization effect of the hydrogen
bonding networks of the sugar moiety.

250 0 D. Kitamoto et al.

MEL-B

@EL-A

Fig. 9.13. Formation of lyotropic liquid crystals; water penetration scans of mannosylerythritol lipid-A
and -B.

Recently, Worakitkanchanakul et al. (2008b) reported the binary phase diagram

of MEL-B/water system. For conventional MEL producers such as l? antarctica, l?
rugufosa (Morita et al., 2006), and l? aphidis (Rau et al., 2005), the amount of MEL-B
in the culture medium is much lower than that of MEL-A; this makes the phase
diagram analysis of MEL-B very difficult. l? tsukubaensis was recently demonstrated
to predominantly produce MEL-B that has a different sugar configuration ( I-O-P-Dmannopyranosyl-D-erythritol) from that (4-O-P-D-mannopyranosyl-~-erythritol)of
conventional MEL-B (Fukuoka et al., 2008; Morita et al., 2007a).
Figure 9.15a shows the temperature dependence of the phase diagram based
on the newly identified MEL-B/water system. The present MEL-B efficiently selfassembles into a lamellar (La) phase over remarkably wide concentration and
temperature range. According to S A X S measurements, the interlayer spacing (d) is
almost constant (about 4.7 nm) for the low MEL-B concentration (160 wt%) region,
where the La phase is in equilibrium with the excess water phase (La+ W). O n the
other hand, at high MEL-B concentration (>GO wt%) region, the d-spacing gradually
decreases to 3.1 nm with an increase in the MEL-B concentration. The La phase
obtained is stable up to 95C below a MEL-B concentration of 85 wt%; the melting
temperature of the liquid crystalline phase dramatically decreases with an increase in
MEL-B concentration (above 85 wt%). Furthermore, the MEL-B efficiently gives
large vesicles (1-5 pm) at low concentrations (Fig. 9.15b). These results suggest that
the newly identified MEL-B would be useful in various applications as an La phaseand/or vesicle-forming lipid.

Self-assemblingProperties of Glycolipid Biosurfactants and Their Functional Developments 0 25 1

Sponge
(L3)

E3

Bicontinuous
cubic
(V,)

Lamella
(La)
MEL-A concentration (wt%)
Fig. 9.14. Temperature dependence of the binary phase diagram of mannosylerythritol lipid-A/water
system.
L;, sponge phase, V;, bicontinuous cubic phase, La; lamellar phase, FI; isotropic phase.

80
W

65
3

f 50
Q

5 35
I-

20
0

20 40 60 80 100
MEL -B concentration (wt%)

Fig. 9.1 5. a) Temperature dependence of the binary phase diagram of rnannosylerythritol lipid-8 (R

tsukubaensis)/water system.
La; lamellar phase, FI; isotropic phase.
b) Confocal laser scanning micrograph of mannosylerythritol lipid-6 (f?tsukubaensis) vesicles (1
wt% in water).
Bar length is 5 pm.

252 0 D. Kitamoto et al.

Sophorolipid
Vesicle forming glycolipids often give ribbon and tube structures. Several groups
reported formation of ribbon, tube, myelin figure and gel structures, using acyclic
single chain glycolipids (Kitamoto et al., 2005; Shimizu et al., 2005). The resultant
morphology depends on the carbohydrate headgroup structure. From the analysis
of crystallographic structure and other spectral data, the hydrogen bonding of the
linkage amide bond is shown to stabilize the formation of the unidirectional threedimensional structures. A ribbon structure is known to often be a precursor of
tubules.
As described previously, acidic sophorolipid (Fig. 9.2) molecules represent a novel
type of asymmetrical bolaamphiphile due to their unique structural features that
include an asymmetrical polar head size (disaccharide vs. carboxylic acid), a nicked
hydrophobic core (oleic acid), and a non-amide polar-nonpolar linkage. Zhou et al.
(2004)recently investigated the supramolecular structures of self-assembled aggregates
of the BS using light microscopy, DLS, SAXS/WS,
and FT-IR spectroscopy.
Figure 9.16 shows the self-assembling manner of sophorolipids under different
conditions. Interestingly, acidic sophorolipids self-assemble into giant twisted and
helical ribbons of 5-11 pm width and several hundreds of micrometers length in
acidic conditions (pH c 5.5). An increase in the solution pH decreases the yield of
solidlike ribbon products. It also slows the formation of giant ribbons, and increases
the helicity and entanglements of the giant ribbons. The sophorolipid molecules form
highly crystallized lamellar structures at room temperature but with a much shorter
long period of 2.78 nm compared to oleic acid molecules. This lamellar spacing
excludes the possibility that a similar bilayer molecular packing is formed in which
the sophorolipid carboxylic acid groups form hydrogen-bond-stabilizing dimers.
The layered structures with a long period of 2.78 nm within the giant ribbons
are likely to be formed by interdigitated packing, which is stabilized by both the
strong hydrogen bonding between the intermolecular disaccharide headgroups
and the strong hydrophobic-hydrophobic interactions between the closely packed
hydrocarbon chains (Zhou et al., 2004). Neutralization of acidic sophorolipids with
dilute NaOH to pH >5.9 produces clear solutions with the formation of short range
ordered micelles. The self-assembly of the sodium salt molecules strongly depends on
the solution concentration. At dilute concentrations, the size of the micellar aggregates
increases with increasing concentration. When the concentration of the sodium salt
is above 1.O mg/mL, narrowly distributed large micellar aggregates with a constant
hydrodynamic radius of about 100 nm are formed.
These data clearly indicate that the supramolecular assemblies arise from the
unique structure of both hydrophobic and hydrophobic parts, which are elegantly
designed and refined by yeasts through a long period of evolution.

Rbamnolipid
Rhamnolipids also self-assemble into variety of structures, and their assembled
structures drastically change with slight variation in the headgroup. Because of the

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 253

lnterdigitated
lamellar packing

Acidic sophorolipid
Carboxylic group
Acidic conditions
(

Sophrose group

< pH 3.6)

neutralization
With NaOH
Giant ribbons

Higher concentrations
(>
.OImglmL)

A few hundred
micrometers

5 to 10 pm

Large micelles
(100 nm)
Helical ribbons
Fig. 9.16. Self-assembly manner of sophorolipid.

carboxylic acid on the headgroups, rhamnolipids reversibly change their self-assembly

structures depending on the solution pH.
Rhamnolipids (RL-A and -B; Fig. 9.3), bearing two alkyl chains at the C-1 and
C-2 position, form vesicles that can change their morphology to lamellar or micelles
within a narrow pH range (Ishigami et al., 1987). Namely, the rhamnolipids form
micelles at pH over 6.8, lipid particles at pH 6.6-6.2, lamellar structures at pH 6.5-6.0,
and finally vesicles of diameter 50-100 nm at pH 5.8-4.3. Fig. 9.17 illustrates the
self-assembling manner of rhamnolipids under different pH conditions.
The producers of rhamnolipids, which are mostly hydrocarbon-assimilating
bacteria, grow well under a narrow pH range around 7.0, but hardly grow under
acidic conditions. The pH-dependent conversion of molecular aggregates of the
BS would thus be associated with the biological hnctions inside and outside the
bacterial membrane under weakly alkaline or neutral conditions. In the case of acidic
conditions, the bacterial cell membrane seems to be protected by the BS in the state
of lamellar and vesicle.
The detailed study for understanding the physico-chemical properties ofglycolipid/
water systems is still in its infancy. However, as seen previously, the recent efforts
have revealed several features of glycolipids, which are distinct from phospholipids or
other conventional surfactants and are not able to be inferred by simple extrapolation

254 0 D. Kitamoto et al.

Rhamnolipid
Carboxylic group

Rhamnosyl group
Vesicles
pH 4.3-5.8

Micelles
pH > 6.8

Lipid particles
pH 6.2-6.6

Lamellar (La)
pH 6.0-6.5

Fig. 9.17. Self-assembly manner of rhamnolipid.

of the knowledge of other amphiphile/water systems (Kitamoto et al., 2005). These

appear to arise mainly from the strong cohesive forces between the oligosaccharide
headgroups involving hydrogen bonds. The orientation of some key hydroxyl groups
in the headgroups is thus crucial to controlling the molecular assemblies of the
systems. This is in marked contrast to the conventional amphiphile/water systems,
where the inter-headgroup interactions are in most cases simply repulsive (Hato et
al., 2001). These points are important for a better application of glycolipid BS.

Potential Applications of Glycolipid Biosurfactants

The ability of BS to spontaneously form molecular assemblies, their versatile
biochemical functions, and environmentally friendly features make them particularly
attractive tools in a wide range of industrial fields. Possible examples are those for
cold thermal storage (Kitamoto et al., 2001), soil remediation (Singh et al., 2007;
Mulligan, 2005), cell cycle regulation (Kitamoto et al., 2002), protein sensing and/or
separation (Konishi et al., 2007; Imura et al., 2007b; Ito et al., 2007), gene delivery
(Nakanishi, 2003; Rodrigues et al., 2006), and skin care (Fukuoka et al., 2007b).

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 255

Here the authors would like to focus special attention on the application of BS in
biomedical area.

Antimicrobial and Antiviral Compounds

Generally, surfactants having high surface-active properties show certain antimicrobial
activities to some extent. Indeed, many glycolipid BS show various biological activities
reflecting their carbohydrate structures (Cameotra & Maker, 2004; Singh et al.,
2004). The effect of the BS on bacteria appears more strongly with gram-positive
bacteria than gram-negative bacteria because of the different cell wall structures.
Moreover, some of glycolipid BS inhibit not only growth of microorganisms but also
that of viruses and tumor cells. Table 9.3 shows the antimicrobial activities of some
glycolipid BS and chemical glycolipid surfactants.
Mannosylerythritol lipids (Fig. 9.1) show a high antimicrobial activity particularly
against gram-positive bacteria. The minimum inhibitory concentrations (MIC) of
MEL against the bacteria are much smaller than those of sugar-based surfactants such
as sucrose monodecanoate and sorbitan monododecanoate (Span 20) (Kitamoto et
al., 2002).
Ustilagic acids produced by I? fis$ormata, which are derivatives of cellobiose
lipids (Fig. 9.6) and act as fungicides, show growth inhibition against more than
80% of the 280 yeast and yeast-like species tested under acidic conditions (pH
4.0) at 20-30C. The purified glycolipids enhance non-specific permeability of the
cytoplasmic membrane in sensitive cells, which results in ATP leakage (Kulakovskaya
et al., 2005).
Sophorolipids (SL-1 and -2, Fig. 9.2) inhibit the growth of B. subtilis, Staphylococcus
epidermidisand Streptococcusfaeciumat concentrations of6-29 mg/L. SL-2 also inhibits
the germination of conidia of the fungus GLomereLh cingukzta at a concentration of 50
mg/L (Langet al., 2002). Different sophorolipids (lactone forms) produced by Candidz
apicola IMET 43733 inhibit the growth of not only gram-positive bacteria but also
gram-negative bacteria such as Escherichia coli and Serratia marcescens (Hommel et
al., 1994). Shah et al. (2005) recently examined in vitro spermicidal and anti-HIV
virucidal activities of a series of sophorolipid analogs. Sophorolipid diacetate ethyl
ester derivative is the most potent spermicidal and virucidal agent of the series of
analogs studied. Its virucidal activity against HIV and sperm-immobilizing activity
against human semen are similar to those of nonoynol-9.
Rhamnolipids (RL-I and -2, Fig. 9.3) inhibit the growth of Bacillus subtilis in
concentrations of 10-35 mg/L. The lipids show antiphytoviral effects for the virus/
host combinations of tobacco mosaic viruslNicotiana glutinosa and potato X virus/
Nicotiana tabacum. Rhamnolipids can also be used as biological control products; they
exhibit zoosporicidal activity on species of three representative genera of zoosporic
phytopathogens: Pythium aphanidermatum, Phytophtora capsici, and Phmopara
lactucaeradicis, and on the harmful algal bloom species. At concentrations of 5-30
mg/L, both lipids bring about a cessation of motility and lysis of the entire zoospore

256 0 D. Kitamoto et al.

Table 9.3.Antimicrobial Activities of Glycolipid Biosurfactants
Compound
Microorqanism

MEL-Aa

MEL-Ba SLb

RLc

SEafd

Span20a

MIC (mg/L)
Gram-positives

Arthrobacter oxidans

-*

Azo tobacter chroococcum

16

1.95

Bacillus subtilis

6.2

25

0.12

64

800

400

Micrococcus luteus

3.1

12.5

0.48

32

400

400

Mycobacrerim phlei
Mycobacterium rhodochrous

16

25

25

>800 400

Mycobacterium rubrum

0.12

StaDhv/ococcusaureus

12.5

25

>800

128

>800

Staphylococcus edipermidis

Streptococcus faecalis

64

~~~

Gram-negatives

Alcaligenes faecalis
Borderella bronchiseotica

32
128

Escherichia coli

>400

>400

7.8

32

>800 >800

Proteus vulgaris

>31.3

Pseudomonas aeruginosa

>400

>400

7.8

>256

Pseudomonas rivoflavina

12.5

25

>800 S O 0
>800 S O 0

Serratia marcescens

1.95

16

>256

>800 S O 0

>256

16

S O 0 >800

16

32

Yeasts

Candida albicans

>400

>400

Saccharomyces cerevisiae

Fungi

Aspergillus niger

>400

>400

Gliocadium virens
Penicillium chrysogenum

Botrytis cinerea

Rhizotecnia solani

18
18

MIC, Minimum inhibitory concentration.

a Data from Kitamoto et al(2002).
bSophorolipidsproduced by Candida apicola IMET43747 (Hommel et al., 1994).
Rhamnolipids produced by Pseudomonas aeruginosa AT10 (Abalos et al., 2001).
Sucrose monodecanoate.
'Not available.

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments0 257

population, although sophorolipids and trehalose lipids show no activity up to 1000

mg/L (Lang & Wullbrandt, 1999; Nitschke et al., 2005).
Newly identified rhamnolipids from I? areruginosa AT10, which are a mixture of
seven homologs, show excellent growth inhibition activities not only for gram-positive
and -negative bacteria but also for filamentous fungi in a concentration range of 16 to
32 mg/L. (Abalos et al., 2001). In addition, the lipid mixture inhibits the growth of
the phytopathogenic fungi, Botytis cinerea and Rhizotecnia sokzni at a concentration
of 18 mg/L.
Trehalose lipids (TL-1 and TL-2; Fig. 9.4) show no growth inhibition against
gram-negative bacteria and yeasts. However, TL-1 inhibits the conidia germination of
the hngus Glomerelkz cingukzta at a concentration of 300 mg/L. O n the other hand,
succinoyl-trehalose lipids (STL- 1 and -2; Fig. 9.5) produced by R. erythropolis inhibit
herpes simplex virus and influenza virus at concentrations of 11-33 mg/L (Kitamoto
et al., 2002).
Oligosaccharide lipids (Fig. 9.7) produced by Tsuhmurelkz sp. show some
inhibition activity against gram-positive and gram-negative bacteria. GL-1 and GL-3
inhibit the growth of Bacillus megaterium at a concentration of 150 and 50 mg/L,
respectively (Vollbrecht et al., 1999).

Antitumor and Cell Differentia tion-inducing Probes

Recently, some microbial products have attracted attention as low-molecular bioprobes
that can control a variety of mammalian cell functions. They are considered to
participate in various intercellular molecular recognitions such as signal transduction,
cell differentiation, and cell immune response (Kitamoto et al., 2005).
Glycolipid BS exhibit unique actions on mammalian cells in addition to
antimicrobial and antiviral activities as mentioned previously (Rodrigues et al., 2006;
Singh et al., 2004). Mannosylerythritol lipids (MEL-A and -B, Fig. 9.1), succinoyltrehalose lipids (STL-1 and -3; Fig. 9 3 , sophorolipids (SL; Fig. 9.2) and polyol lipids
(PL) show excellent growth inhibition and differentiation-inducing activities against
human leukemia cells such as myelogenous leukemia cell K562, promyelocytic
leukemia cell HL60, and basophilic leukocyte KU812 (Kitamoto et al., 2002).
MEL-A and MEL-B inhibit the growth of HLGO cells at concentrations of
5- 1OpM and induce drastic morphological changes-that is, adhesion to the bottom
of a cultivation flask. Interestingly, the differentiation direction of these induced cells
depends on the saccharide structure of BS. With respect to HL6O cells, MEL-A,
MEL-B and PL (10 mg/L) induce granulocytic differentiation, while SL (10 mg/L),
STL-1 (2.5 pM) and STL-3 (5 pM) induce monocytic differentiation. Figure 9.18
illustrates an outline of the differentiation-inducing activities of these glycolipid BS.
More significantly, all of these six glycolipid BS inhibit the activity of phospholipidand Ca2+-dependentprotein kinase C (PKC) of HL6O cells.
MEL-A, MEL-B, and SL induce significant neurite outgrowth from rat
pheochromocytoma PC-12 cells and partial cellular differentiation. MEL-A can

258 0 D. Kitamoto et at.

K562 (myeloaenous leukemia cell)

SL, STL-3

STL-1

Monocvtes

G ranulocvtes

Meaa kawocvtes

H L 6 0 (promyelocytic leukemia cell)

STL-1, STL-3
SL

P K C inhibition

G ranulocvtes

Monocvtes
U937 (monocytoid leukemia cell)
STL-1

Monocvtes -ma croDha aes

KU812 (basophilic leukemia cell)

STL-1

Monocvtes

SL

G ranulocvtes
S L, sophorolipid; STL, succinoyl uehalose lipid

Fig. 9.18. Differentiation-inducingactivities of glycolipid biosurfactantsagainst human leukemia cells.

MEL-A, mannosylerythritol lipid-A; MEL-8, mannosylerythritol lipid-6; SL, sophorolipids; STL, succinoyltrehalose lipids; PKC, protein kinase C.

induce neurite outgrowth even in the presence of an anti-nerve growth factor (NGF)
receptor antibody that obstructs NGF action. These results indicate that MEL and
NGF induce neurite outgrowth from PC-12 cells by different signal transduction
pathways (Wakamatsu et al., 2001).
MEL-A has been recently demonstrated to inhibit the growth of mouse melanoma
B 16 cells in a dose-dependent manner. Exposure of B 16 cells to the lipid (10 pM)
causes the condensation of chromatin, DNA fragmentation, and sub-GI arrest, all of
which are hallmarks of cells that are undergoing apoptosis. The lipid treatment also
enhances expression of PKC,, suggesting that the lipid triggers the differentiation of
the cells through a signal pathway that involves PKC, (Zhao et al., 2001).
It is truly surprising that many glycolipid BS show certain activities against various
organisms ranging from prokaryotes to mammalians. The complicated but naturally
engineered structures of the BS allow them to display versatility over conventional
chemical surfactants. Therefore, the functional development of the BS should also be
carried out from the viewpoints of biochemistry and molecular biology.

Imm unoligan ds
A large variety of glycolipids can be found in all species from bacteria to mammals as
constituents of cell membranes (Kitamoto et al., 2005), where they are positioned to
interact with extracellular signal transducers such as immunoglobulins, lectins, and
enzymes or receptors on the surfaces of other cells. For example, glycosphingolipids,

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 259

such as gangliosides, participate in vital functions including signal transduction, cell

cycle, antigenicity, cell adhesion, and proliferation through protein-carbohydrate or
carbohydrate-carbohydrate interactions (Hakomori et al., 2002; Vankar & Schmidt,
2000). Glycolipids are thus information-rich molecules vital to recognition processes
in many physiological events. Hence, glycolipid BS are also expected to show binding
interaction with the above signal transducers.
Immunoglobulin G (IgG), which is the most dominant and essential
immunoglobulin in mammals, is widely used in immunodiagnostics and therapeutic
applications. Because of its high affinity, protein-A is the usual choice for immunoaffinity chromatography of IgG (Roque et al., 2007). Chromatography using
protein-A, however, has two major drawbacks: the high cost of the ligand protein, and
the highly acidic conditions needed to elute IgG. Gangliosides exhibit high affinity
for immunoglobulins, and thus are focused on as a new affinity ligand for IgG (Evans
& MacKenzie, 1999). The possibility of developing these membrane glycolipids
into new practical ligands, however, is far from straightforward due to their limited
amounts and heterogeneity. Im et al. recently focused their attention on glycolipid BS
as an alternative ligand for IgG, and investigated the binding beween MEL (Fig. 9.1)
and human IgG (HIgG) (Im et al., 2003).
MEL-A shows nearly the same binding affinity towards HIgG as that of
bovine ganglioside GM,. Interestingly, MEL-A noncovalently attached onto poly
(2-hydroxyethyl methacrylate) beads exhibits a significant binding constant of
1.43 x lo6 (M-I) for HIgG, which is approximately four times greater than that of
protein-A previously reported (Teng et al., 2000). The binding amount of HIgG
to the composite increases with increased applied concentration, reaching over 100
mg (per g of composite). Figure 9.19 shows the binding of HIgG toward the MELpolymer complex. This is the first report on the binding of a natural human antibody
to a yeast glycolipid, and may facilitate the application of glycolipid BS for an affinity
ligand for immunoglobulin.
'The recognition of carbohydrate moieties at the surface of the monolayers by
specific receptors can provide quantitative information about the binding and/or
multivalent effects (Kitamoto et al., 2005; Goodby et al., 2007). Surface plasmon
resonance (SPR) spectroscopy is a useful technique to measure the binding properties
of proteins and the ligands immobilized on a gold surface (Hernaiz et al., 2002).
.
Recently, Imura et al. (2007b) conducted SPR study using the monolayer of
MEL-A attached on a gold surface through alkanethiol self-assembled membrane
(SAM),and confirmed that HIgG efficiently binds to the glycolipid monolayer by
the multiple association between the carbohydrate recognition sites of HI@ and
the carbohydrate headgroup of MEL-A. Interestingly, the monolayer also shows
high binding affinity toward human IgM (HIgM). Figure 9.20 illustrates the selfassembled monolayer of MEL-A employed for SPR study and sensorgram between
the monolayer and HIgG.
Based on SPR, the estimated binding constants were K, = 9.4 x 1O6 M-' for HIgG
and 5.4 x lo6 M-' for HIgM, respectively. 'The binding site of HIgG toward the

260 0 D. Kitamoto et al.

120

100

.-caJ,
8

sPI

80

0
0

2
4
6
8
10
12
IgG solution concentration (mg / mi)

Fig. 9.19. Binding of human immunoglobulin G to MEL-polymer composite. The polymer composite
(0.33f 0.08 g) bearing MEL-A (7.1 pmol/g composite) was suspended in 50 mM phosphate buffer (pH
4.6, 1 M Na,SO, ) in a polypropylene tube. Different amounts of human immunoglobulin G were added
to the tube, and then the tube was incubated for 1 hr. Ka,apparent binding constant.

MEL-A monolayer was estimated to be the Fab region. Imura et al. (2007b) also
demonstrated a large amount of HIgG and HIgM bind to the monolayer with a
markedly high density using atomic force microscopy.
Ito et al. (2007) recently reported kinetic studies on the interactions between the
monolayers of MEL (MEL-A, -B, and -C) and various classes of immunoglobulins
(HIgG, IgA, and IgM) using SPR. Interestingly, the monolayer of MEL-A gives high
affinity (K, = 1.7 x 104 M) toward HIgG, while those of MEL-B or MEL-C give
little affinity. The MEL-A monolayer also gives high affinity toward HIgA (K, = 2.4
x l o 7M) and HIgM (K, = 2.2 x l o 7M). Interestingly, the binding manner between
MEL-A and these immunoglobulins was estimated as not being the monovalent

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments0 261

Lectins
Immunoglobulins
MEL-A
self-assembled
monolayer

.Octadecanethiol (C1
HPA chip
(Biacore)

800
n

3
QC
c

.-c

MEL-A

600

3
400
0

2 200
Protein-A
0

100

200
300
Time (sec)

400

Fig. 9.20. Binding of glycoproteinstoward the self-assembled monolayersof mannosylerythritol lipid-A

on the surface plasmon resonance study.
MEL-A, mannosylerythritol lipid-A; HIgG, human immunoglobulin G.

mode but the bivalent mode. Figure 9.21 shows the images for the binding of
these immunoglobulins to the surface of the MEL-A monolayer. These results clearly
demonstrate that the yeast glycolipid monolayers would be useful as noble affinity
ligand system for various immunoglobulins.
MEL-A shows binding affinity not only towards immunoglobulins but also
towards lectins, which are one of the most important signal transducers (Konishi
et al., 2007). O n SPR (Fig. 9.20), the monolayer of MEL-A exhibits high binding
affinity to Concanavalin A (Cod)and Maackia amurensis lectin-I (MAL-I). The
observed affinity constants for C o d and MAL-I were estimated to be 9.48 f 1.31
x 106and 3.13 f 0.274 x 10 Me, respectively; the value was comparable to that of
a-Man-( 1-3)-[a-Man( l-G)]-Man, which is the most specific probe to Cod.
More significantly, a-methyl-D-mannopyranoside (1 mM) exhibited no binding
inhibition between MEL-A and C o d . MEL-A is thus likely to self-assemble to give
a high affinity surface, where C o d binds to the hydrophilic headgroup in a different

262 0 D. Kitamoto et al.

Fig. 9.21. Tapping mode of atomic force microscopy images of different immunoglobulins bound to
the self-assembled monolayers of mannosylerythritol lipid-A (MEL-A). IgG, immunoglobulin G; DPPC,
a-dipalmitoyl-phosphatidylcholine.

manner from that generally observed in lectin-saccharide interactions. The binding

manner may be related to the previously mentioned biochemical actions of MEL
toward mammalian cells.

Gene Carriers
Gene delivery across cell membranes into the nucleus, that is, gene transfection,
is a fundamental technology not only for bioscience, but also for the clinical gene
therapy of generic and acquired diseases like cancer. The success of gene therapy, in
particular, is highly dependent on the development of transfection vectors that are
safe and efficient. Although the most efficient methods for gene transfection involve
the use of viral vectors, there are still arguments about risks regarding propagation
and immunogenicity. A variety of nonviral gene-delivery systems have thus been
investigated (Kawakami et al., 2008).
Among nonviral vectors hitherto reported, cationic liposomes are one of the
most efficient vectors for the delivery of plasmid D N A and antisense oligonucleotides
into mammalian cells (Dass & Choong, 2006; Nakanishi, 2003). However, cationic
liposomes alone nonspecifically interact with cellular membranes by electrostatic
interactions; targeted gene transfection systems have been strongly desired. Recently,
glycolipids have received much attention and frequently been employed as key
materials for targeted gene delivery using liposomes, providing liposomes with
enhanced colloidal stability in blood circulation, and specific affinity towards targeting

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments0 263

cells (Hashida et al., 2001).Indeed, liposomes consisting ofspecific glycolipids show a

high stability in a dispersed state or in a frozen state, due to the exclusive volume effect
of the extending and rotating oligosaccharide chains on the surface of the liposomes
(Kitamoto et al., 2005).
Taking the previously mentioned results into consideration, glycolipid BS is very
likely to improve the performance of cationic liposomes for gene transfection. In
addition, the use of glycolipid BS is highly advantageous compared to the use of
cellular glycolipids such as gangliosides. Inoh et al. (2001)focused on the distinctive
biological and self-assembling properties of MEL and examined their effect on the gene
transfection. MEL-A (Fig. 7.1)has dramatically demonstrated the ability to increase
the efficiency of the gene transfection mediated by the liposomes including a cationic
cholesterol derivative, cholesteryl-3~-carboxyamindoethylene-N-hydro~ethylamine
(Chol-OH), and a-D-dioleoylphosphatidylethanolamine (DOPE).
Figure 7.22 illustrates the scheme of gene transfection into mammalian cells by
cationic liposomes including MEL-A. Compared to previous commercially available
cationic liposomes like Lipofectin, the new liposome including MEL-A enables a 50to 70-fold increase in the efficiency for delivery of plasmids encoding luciferase into the
target cells (NIH3T3, COS-7, and HeLa). The authors also studied the localization of
FITC-conjugated antisense DNA and the cationic liposomes containing rhodamineconjugated phosphatidylethanolamine in the target cells (Inoh et al., 2004).In the
case of liposomes with MEL-A, fluorescence-labeled liposome and antisense DNA
complexes are temporarily on the plasma membrane of the target cells, and the DNA
is then rapidly transferred to the nucleus. The distribution of fluorescence along the
plasma membrane suggests the membrane fusion between liposomes and the plasma
membrane. MEL-A would induce the membrane fusion to accelerate the efficiency of
gene transfection (Ueno et al., 2007).
Igarashi et al. (2006) recently prepared the cationic liposomes consisting of
MEL-A and investigated its transfection efficiency in human cervix carcinoma HeLa
cells. Interestingly, the size of the MEL liposomes is about 40 nm, and the complex
of the liposomes with plasmid DNA remains an injectable size (167 nm). MEL-A
induces a significantly higher level of gene expression, compared to commercially
available liposome (TWO).
In this transfection system, the amount of DNA associated with the cells is
rapidly increased by addition of MEL-A to the liposome. Also, the liposome-DNA
complex is widely distributed in the cytoplasm, and the DNA is detected strongly
in the cytoplasm and around the nucleus, compared with control liposomes. These
results suggested that MEL-A increases gene expression by enhancing the association
of the liposome-DNA complex with the cells. Therefore, the yeast glycolipid BS,
MEL, have a great potential as a vector for gene transfection.
Gene delivery systems mediated by nonviral vectors are still at an early stage, and
remain an inefficient process compared to systems mediated by viruses. However,
there appear to be a number of promising ways for developing them into advanced

264 0 D. Kitamoto et al.

Cationic liposornes

Cationic cholesteKEi
t

Low all toxicity

Gene (DNA)
Liposomes efficienty
bind to DNA

Lipowme-DNA complexes
efficiently bind to cell surface

High efficiency

70-foM higher than with

convsntind liposwnes

Gene expression

Fig. 9.22. Scheme of gene delivery into mammalian cells using liposomes prepared from cationic
cholesterol and mannosylerythritol lipid-A.

nonviral vectors. Together with the unique mechanism for the gene transfection,
glycolipid BS are a critical lipid formulation of the desired vectors, and will facilitate
the development of novel gene delivery systems in terms of efficiency, time, and
toxicity.

Skin Care Materials

Sophorolipid converted to glycolipid ester has been used as an essential component
in a cosmetic composition (Shete et al., 2006). Glycolipid ester acts an excellent
moisturizer, imparts a pleasant finishing touch to the skin and hair, and eliminates the
adverse properties attributable to conventional moisturizers. Sophorolipids have antiradical and anti-elastatic properties, which is desirable in compositions for cosmetic,
hygienic, or pharmaco-dermatological use, in particular for the protection and care
of skin. They stimulate dermal fibroblast metabolism and collagen neosynthesis.
Sophorolipids are also used for skin treatments, as an activator of macrophages, and
as agents that restructure, repair, or tone the skin (Van Bogaert et al., 2007).

Self-assemblingPropertiesof Glycolipid Biosurfactants and Their Functional Developments 0 265

Recently, Kitagawa et al. (2007) have investigated the moisturizing properties

of MEL and evaluated the recovery effect on damaged skin. O n assays using a threedimensional cultured skin model, MEL exhibit excellent moisturizing characteristics
and are equivalent to those of natural ceramides (Fukuoka et al., 2007b). As the
MEL structure resembled natural ceramides, it was expected to easily penetrate
the intercellular spaces in the stratum corneum. Also, MEL would be effective in
moisture retention and maintenance at the skin intercellular level, because it easily
forms lyotropic liquid crystals like lamella (La) as indicated previously.
Natural ceramides have been praised for their moisturizing properties, and
together with hyaluronic acids they have become a crucial material for skin care
applications. However, the use of natural ceramides is extremely expensive due to
the limited amount in plants and to the complicated extraction and purification
processes. Therefore a safe product with characteristics similar to natural ceramides at
a lower cost has been strongly desired. Considering the high-level production of MEL
by Pseudozyma strains, the manufacturing cost of MEL would be much lower than
that of natural ceramide products. The yeast glycolipid BS is likely to contribute to
functional skin care products.

Conclusions
As described in the above sections, BS show unique properties compared to their
chemical counterparts. The numerous advantages of BS have prompted applications
not only in the food, cosmetic, and pharmaceutical industries but in environmental
protection and energy-saving technology as well. Among known BS, glycolipid types
are the most promising, due to high productivity from renewable resources and
versatile biochemical properties.
Development of nanotechnology requires nanoscale materials, which should be
easy to prepare and handle. To this end, the molecular self-assembly process is one of
the most promising routes for creating a new class of nanomaterials (Shimizu et al.,
2005; Ariga et al., 2007). There have been a lot of natural and synthetic amphiphiles
that are able to form self-assembled nanostructures. However, the advantages of
using glycolipid BS for novel nanomaterials are their ability to spontaneously form
molecular assemblies, as well as their environmentally friendly features.
During the past decade, the research and development of glycolipids has
undoubtedly rushed into a new era. Glycolipid BS display versatile performance over
that attained by conventional chemical surfactants. In addition, recent advances in
molecular biology and biotechnology have enabled processes that drastically improve
the productivity of the BS and control their chemical structures (Mukherjee at al.,
2006). Total syntheses of mannosylerythritol lipid (Crich et al., 2002), sophorolipids
(Furstner et al., 2000), and rhamnolipid (Duynstee et al., 1998) have already been
attained. These synthetic approaches will also facilitate the functional development of
glycolipid BS along with the progress of biotechnology.

266 0 D. Kitamotoet al.

Different glycolipid BS provide a variety of self-assembled nanostructures both

in the pure form as well as in the mixed or conjugated form (Fig. 9.23). Although
the current development of BS is still in its infancy, these nanostructures have a
great potential for a variety of industrial fields. Various programs are now being put
into action all over the world, aiming at constructing a sustainable society. Among
the programs, introduction of green technology into all areas of industry is one
of the most important challenges. Considering the current social and technological
backgrounds, utilization of BS, which are highly environmentally friendly as well as
highly functional, is in a favorable position. Therefore, BS may become the flagship
of biobased materials leading green technology in the near future.

Acknowledgments
The authors would like to thank Professor Yoichi Nakatani at the University of Louis
Pasteur (Strasbourg, France) and Professor Mamoru Nakanishi at Aichi Gakuin
University (Nagoya, Japan) as excellent collaborators. We also thank Dr. Masaru
Kitagawa of TOYOBO Co., Ltd. (Osaka, Japan) for providing information on the
use of glycolipid BS in cosmetics.

Fermentation orocesses

Self-assembly & Bottom-ura

Biobased materials
Fig. 9.23. Self-assembled nanostructuresof glycolipids biosurfactants and their potential applications.

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments0 267

References
Abalos, A,; A. Pinazo; M.R. Infante; M. Casals; F. Garcia; A. Manresa. Physicochemid and
Antimicrobial Properties of New Rhamnolipids Produced by Pseuhmonas aeruginosa AT10 from
Soybean Oil Refinery Wastes. L a n p u i r . 2001,17,1367-1371.
Ariga, K.; T. Nakanishi; J.I? Hill. Self-Assembled Microstructures of Functional Molecules. Curr.
Opin. Colloid Interface Sci. 2007,12,106120.
Banat, I.M.; R.S. Makkar; S.S. Cameotra. Potential Commercial Applications of Microbial Surfactants. Appl. MicrobioL Biotechnol. 2000,53,495-508.
Cameotra, S.S.; R.S. Makkar. Recent Applications of Biosurfactants as Biological and Immunological Molecules. Curr. Opin. Microbiol. 2004,7,262-266.
Crich, D.; M.A. de la Mora; R. Cruz. Synthesis of the Mannosylerythritol Lipid MEL A; Confirmation of the Configuration of the Meso-Erychritol Moiety. Gtrahedron. 2002,58,35-44.
Dass, C.R.; P.M.F. Choong. Selective Gene Delivery for Cancer Therapy Using Cationic Liposomes: In Vivo Proof of Applicability.J. Control. Rehase. 2006,113,155-163.
De Kruif, C.G.; E Weinbreck; R. de Vries. Complex Coacervation of Proteins and Anionic Polysaccharides. Cum Opin. Colloid Interface Sci. 2004,9,340-349.
Dembitsky, V.M. Astonishing Diversity of Natural Surfactants. Lipid. 2005,39,933-953.
Dobereiner, H.-G. Properties of Giant Vesicles. Cum Opin. ColbidInterfce. Sci. 2000,5,

256263.
Duynstee, H.; M.J. van Vliet; G.A. van der Marel; J.H. van Boom. An Efficient Synthesis of
(R)-3-{(R)-3-[2-O-(a-L-Rhamnosyl)a-L-Rhamnosyl] Oxydecanoyl}Oxydecanoic Acid, a
Rhamnolipid from Pseudomonas aeruginosa. Eur. J. Org. Chem., 1998,303-307.
Evans, S.V.; C.R. MacKenzie. Characterization of Protein-Glycolipid Recognition at the Membrane Bi1ayer.J. Mol. Recognit. 1999,12,155-168.
Fukuoka, T.; T. Morita; M. Konishi; T. Imura; D. Kitamoto. Characterization of New Types of
Mannosylerythritol Lipids as Biosurfactants Produced from Soybean Oil by a Basidiomycetous
Yeast, Pseudozyma shanxiensis. J. Oleo Sci. 2007% 56,435-442.
Fukuoka, T.; T. Morita; M. Konishi; T. Imura; D. Kitamoto. A BasidiomycetousYeast, Pseudozyma
tsukubaensis, Efficiently Produces a Glycolipid Biosurfactant; Identification as a New Diastereomer of Conventional Mannosylerythritol Lipid-8. Carbohydrate Res. 2008,343,555-560.
Fukuoka, T.; T. Morita; M. Konishi; T. Imura; D. Kitamoto. Development of Highly Functional
Biosurfactants Produced by Yeasts. Home. Pers. Care Today. 2007b,1,74-76.
Fiirstner, A.; K. Radkowski;J. Grabowski; C. Wirtz; R. Mynott. Ring-Closing Alkyne Metathesis. Application to the Total Synthesis of Sophorolipid Lactone.J. Org. Chem. 2000,65,

8758-8762.
Goodby, J.W.; V. Gortz; S.J. Cowling; G. Mackenzie; I? Martin; D. Plusquellec;T. Benvegnu; I?
Boullanger; D. Lafont; Y.Queneau; S. Chamberte; J. Fitremannf. Thermotropic Liquid Crystalline Glycolipids. Chem. SOC.Rev. 2007,36,1845-2128.
Hakomori, S. The Glycosynapse.Proc. NatL Acaa! Sci. USA.2002,99,225-232.
Hashida, M.; M. Nishikawa; F. Yamashita; Y. Tikakura. Cell-Specific Delivery of Genes with
Glycosylated Carriers. Adv. Drug Delivery Rev. 2001,52,187-196.

268 0 0.Kitarnoto et at.

Hato, M. Synthetic Glycolipid/Water Systems. Curr Opin. Colloid Interface. Sci.2001, G,

268-276.
Hato, M.; H. Minamikawa; K. Tamada; T. Baba; Y. Tanabe. Self-Assembly of Synthetic GlycoIipidlWater Systems. Adu. Colloid Interface Sci. 1999, 80, 233-270.
Hernaiz M.J.; J.M. de la Fuente; A.G. Barrientos; S. Penades. A Model System Mimicking
Glycosphingolipid Clusters to Quantify Carbohydrate Self-Interactions by Surface Plasmon
Resonance. Angew. Chem. Int. Ed. 2002, 41, 1554-1557.
Holmberg, K. Natural Surfactants. Curr. Opin. Colloid Interface. Sci. 2001, 6, 148-159.
Hommel, R.K.; L. Weber; A. Weiss; U. Himmelreich; 0. Rilke; H.P. Kleber. Production of
Sophorose Lipid by Candidu (Torulopisis)apicola Grown on Glucose. J. Biotechnol. 1994,33,
147-155.
Igarashi, S.; Y. Hattori; Y. Maitani. Biosurfactant MEL-A Enhances Cellular Association and Gene
Transfection by Cationic Liposome. J. Control. Release. 2006, 112,362-368.
Inoh, Y.; D. Kitamoto; N. Hirashima; M. Nakanishi. Biosurfactant MEL-A Increase Gene Transfection Mediated by Cationic Liposomes. Biochem. Biophys. Res. Commun. 2001, 289, 57-61.
Inoh, Y.; D. Kitamoto; N. Hirashima; M. Nakanishi. Biosurfactant MEL-A Drastically Increase
Gene Transfection Via Membrane Fusion. /. Contr. Release. 2004,94, 423-43 1 .
Im, J.H.; T. Ikegami; H. Yanagishita; Y. Takeyama; Y. Idemoto; N. Koura; D. Kitamoto. Mannosylerythritol Lipids, Yeast Glycolipid Biosurfactants, Are Potential M n i t y Ligand Materials for
Human Immunoglobulin G.J Biomed. Muter. Res. 2003, G5, 379-385.
Imura, T.; H. Yanagishita; D. Kitamoto. Coacervate Formation from Natural Glycolipid: One
Acetyl Group on the Headgroup Triggers Coacervate-to-VesicleTransition. 1.Am. Chem. Sor.
2004,126, 10804-10805.
Imura, T.; J. Ohira; H. Sakai; M. Abe; D. Kitamoto. Thermodynamically Stable Vesicle Formation
from Glycolipid Biosurfactant Sponge Phase. Colloid Surf: B. Biointerfaces 2005, 43, 1 1 5-121.
Imura, T.; N. Ohta; K. Inoue; N. Yagi; H. Negishi; H. Yanagishita; D. Kitamoro. Naturally Engineered Glycolipid Biosurfactants Leading to Distinctive Self-Assembled Structures. Chem. EUY.J.
2006, 12,2434-2440,
Imura,T; Y. Hikosaka; W. Worakitkanchanakul; H. Sakai; M. Abe; M. Konishi; H. Minamikawa;
D. Kitamoto. Aqueous Phase Behavior of Natural Glycolipid Biosurfactant, Mannosylerythritol
Lipid-A Sponge, Cubic, and Lamellar Phases. Langmuir. 2007a, 23, 1659-1663.
Imura, T.; S. Ito; H. Yanagishita; H. Sakai; M. Abe; R. Azumi; D. Kitamoto. Monolayers Assembled from a Glycolipid Biosurfactant from Pseudozyma (Candida)antarctica Serve as a HighAfXnity Ligand System for Immunoglobulin G or M. Biotechnol. Lett. 2007b, 29, 865-870.
Ishigami, Y.; Y. Gama; H. Nagahora; M. Yamaguchi; H. Nakahara; T. Kamata. The-pH Sensitive
Conversion of Molecular Aggregates of Rhamnolipid Biosurfactant. Chem. Lett. 1987, 763-766.
Ito, S.; T. Imura; T. Fukuoka; T. Morita; H. Sakai; M. Abe; D. Kitamoto. Kinetic Studies on the
Interactions Between Glycolipid Biosurfactant Assembled Monolayers and Various Classes of
Immunoglobulins Using Surface Plasmon Resonance. Colloid Sul-f:B. Biointerfaces 2007, 58,
165-1 71.
Kawakami, S.; Y. Higuchi; M. Hashida. Nonviral Approaches for Targeted Delivery of Plasmid
DNA and Oligonucleotide. J. Pharm. Sci.2008,97,726-745.

Self-assembling Propertiesof Glycolipid Biosurfactants and Their Functional Developments 0 269

Kitagawa, M.; M. Suzuki; S. Yamamoto; A. Sogabe; D. Kitamoto; T. Imura; T. Fukuoka; T.

Morita. Skin Care Cosmetic for Treating Dermatological Disorders Such as Rough Skin, Acne,
Enema, Asteatosis and Senile Xerosis, Contains Biosurfactants as Main Ingredients. Patent
number: W02007060956-A1,2007.
Kitamoto, D.; G. Sangita; G. Ourisson; Y.Nakatani. Formation of Giant Vesicles from Diacylmannosylerythritols, and Their Binding to Concanavalin A. Chem. Commun. 2000,861-862.
Kitamoto, D.; H. Yanagishita; A. Endo; M. Nakaiwa; T. Nakane; T. Akiya. Remarkable Antiagglomeration Effect of a Yeast Biosurfactant, Diacylmannosylerythritol, on Ice-Water Slurry for
Cold Thermal Storage. BiotechnoL Prog. 2001, 17, 362-365.
Kitamoto, D.; H. Isoda; T.Nakahara. Functions and Potential Applications of Glycolipid Biosurfactants. /. Biosci. Bioeng. 2002,94, 187-201.
Kitamoto, D.; K. Toma; M. Hato. Glycolipid-BasedNanomaterials. Handbook of Nunostructured
Biomateriah and n e i r Applications in Nanobiotechnology; Nalwa, H.S., Ed.; American Science
Publishers; California; 2005; Vol. 1. 239-271.
Konishi, M.; T. Morita; T. Fukuoka; T. Imura; D. Kitamoto. A Yeast Glycolipid Biosurfactant,
Mannosylerythritol Lipid, Shows High Binding Affinity Towards Lectins on a Self-Assembled
Monolayer System. Biotechnol Lett. 2007,29,473-480.
Konishi, M.; T. Morita; T. Fukuoka; T. Imura; D. Kitamoto. Efficient Production of Mannosylerythritol Lipids with High Hydrophilicity by Pseudozyma hubeiensis KM-59. AppL Microbiol.
Biotechnol. 2008a, 78, 37-46.
Konishi, M.; T. Morita; T.Fukuoka; T. Imura; D. Kitamoto. Production of New Types of
Sophorolipids by Candida batistae.J. Olco Sci. 2008b, 57, 359-369.
Kulakovskaya, T.; A. Shashkov; E. Kulakovskaya; W. Golubev. Ustilagic Acid Secretion by
Pseudozymaf;.i$rrnata Strains. FEh4S East Res. 2005,5, 9 19-923.
Lang, S. Biological Amphiphiles (Microbial Surfactants). Curr. Opin. Colloid Interface Sci. 2002,
7, 12-20.
Lang, S.; J.C. Philp. Surface-Active Lipids in Rhodococci. Antonie van Leeuwenhoek. 1998, 74,
59-70.
Lang, S.; Wullbrandt, Rhamnose Lipid Biosynthesis,Microbial Production and Application
Potential. Appl. Microbiol. Biotechnol. 1999,51,22-32.
Langer, 0.;0. Palme; V. Wray; H. Tokuda; S. Lang. Production and Modification of Bioactive
Biosurfactants. Process Biochem. 2006,41, 2138-2145.
Lazzari, M.; C. Rhodrigua-Abreu; J. Rivas; M.A. Lopez-Quintela Self-Assembly: A Minimalist
Route to the Fabrication of Nanomaterials.J. Nanosci. Nunotechnol. 2006, 6,892-905.
Luk, Y.-Y.; N.L. Abbott. Application of Functional Surfactants. Cum Opin. Colloid Interface Sci.
2002,7,267-275.
Maldonado, A.; R. Ober; T. Gulik-Krzywicki;W. Urbach; D. Lanbevin. The Sponge Phase of a
Mixed Surfactant System.J. Colloid Interface Sci. 2007,308, 485-490.
Menger, EM.; A.V. Peresypkin; K.L. Caran; R,I?Apkarian. A Sponge Morphology in an Elementary Coacervate. Langmuir. 2000, I G,9 1 13-9 1 16.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; D. Kitamoto. Discovery of Pseudozyma rugulosu

270 0 D. Kitarnoto et al.

NBRC 10877 as a Novel Producer of the Glycolipid Biosurfactants, Mannosylerythritol Lipids,

Based on rDNA Sequence. Appl. Microbiol. Biotecbnol. 2006, 73, 305-313.
Morita, T; M. Konishi; T. Fukuoka; T. Imura; H. Kitamoto; D. Kitamoto. Characterization of the
Genus Pseudozyma by the Formation of Glycolipid Biosurfactants, Mannosylerythritol Lipids.
FEMS Yeast Research. 2007a, 7,286-292.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; D. Kitamoto. Physiological Difference in the Formation of Glycolipid Biosurfactants, Mannosylerythritol Lipids, Between Pseudozyma antarctica
and Pseudozyma apbidis. Appl. Microbiol. Biotecbnol. 2007b, 77,307-3 15.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; S. Yamamoto; M. Kitagawa; D. Kitamoto. Identification of Pseudozyma graminicola CBS 10092 as a Producer of Glycolipid Biosurfactants,
Mannosylerythritol Lipids. 1.Oleo Sci. 2008%57, 123-131.
Morita, T.; M. Konishi; T. Fukuoka; T. Imura; D. Kitamoto. Production of Glycolipid Biosurfactants, Mannosylerythritol Lipids, by Pseudozyma siamensis CBS 9960 and Their Interfacial
Properties. /. Bioeng. Biotechnol. 2008b, 105, (in press).
Mukherjee, S.; I? Das; R. Sen. Towards Commercial Production of Microbial Surfactants. Trends
Biotecbnol. 2006,24, 509-5 15.
Mulligan, C.N. Environmental Applications for Biosurfactants. Enuiron. Pollution. 2005, 133,
183- 198.
Nakanishi, M. New Strategy in Gene Transfection by Cationic Transfection Lipids with a Cationic
Cholesterol. Cum Medic. Chem. 2003, 10, 1289-1296.
Niescher, S.; V. Wray; S. Lang; S.R. Kaschabek; M. Schlomann. Identification and Structural
Characterization of Novel Trehalose Dinocardiomycolates from n-Alkane-Grown Rbodococm
opacus lCl? Appl. Microbiol. Biotechnol. 2006, 70,605-61 1.
Nitschke, M.; S.G.V.A.O. Costa; J. Contiero. Rhamnolipid Surfactants: An Update on the General Aspects of These Remarkable Biomolecules. Biotecbnol. Prog. 2005,21, 1593-1600.
Ongena, M.; I? Jacques. Bacillus Lipopeptides: Versatile Weapons for Plant Disease Biocontrol.
Trends Microbiol. 2007, 16, 115-125.
Philp, J.C.; M.S. Kuyukina; L.B. Ivshina; S.A. Dunbar; N. Christofi; S. Lang; V. Wray. Alkanotrophic Rhodococcus ruber as a Biosurfactant Producer. Appl. Microbiol. Biotechnol. 2002,59,
318-324.
Rau, U.; L.A. Nguyen; S. Schulz; V. Wrya; M. Nimtz; H. Roeper; S. Lang. Formation and Analysis of Mannosylerythritol Lipids Secreted by Pseudozyma apbidis. Appl. Microbiol. Biotecbnol.
2005, GG,551-559.
Ridell, A.; K. Ekelund; H. Evertsson; S. Engstron. O n the Water Content of the Solvent/MonooIeinlWater Sponge (L,) Phase. Colloids Su$ A. 2003,228, 17-24.
Rodrigues, L.; I.M. Banat; J. Teixeira; R. Oliveira. Biosurfactants: Potential Applications in Medicine. /. Antimicrobial. Chemotherapy. 2006, 57, 609-61 8.
Ron, E.Z.; E. Rosenberg. Biosurfactants and Oil Bioremediation. Cur%Opin. Biotecbnol. 2002,
13,249-252.
Roque, A.C.A.; C.S.O. Silva; M.A. Taipa. Mnity-Based Methodologies and Ligands for Antibody Purification: Advances and Perspectives. /. Chromatogr A. 2007, 1160,44-55.

Self-assembling Properties of Glycolipid Biosurfactants and Their Functional Developments 0 27 1

Saha, B.C. Production of Mannitol. ACSsymposium Series. 2003, 862, 67-85.

Shah, V.; G.F. Doncel; T. Seyoum; K.M. Eaton; I. Zalenskaya; R. Hagver; A. h i m ; R. Gross.
Sophorolipids, Microbial Glycolipids with Anti-Human Immunodeficiency Virus and SpermImmobilizing Activities. Antimicrobial Agents Chemotherafi. 2005, 49, 4093-4 100.
Shete, A.M.; G. Wadhawa; I.M. Banat; B.A. Chopade. Mapping of Patents on Bioemulsifier and
Biosurfactant: A Review.]. Sci. Ind. Res. 2006, 65, 91-1 15.
Shimizu, T.; M. Masuda; H. Minamikawa. Supramolecular Nanotube Architectures Based on
Amphiphilic Molecules. Chem. RPY.2005, 105, 1401-1443.
Singh, I?; S.S. Cameotra, Potential Applications of Microbial Surfactants in Biomedical Sciences.
Trench. Biotechnol. 2004,22, 142-146.
Singh, A,; J.D. Van Hamme; 0.P:Ward. Surfactants in Microbiology and Biotechnology: Part 2.
Application Aspects. BiotechnolAdv. 2007,25, 99-121.
Soberbn-Chiva, G.; E Upine; E. DCziel. Production of Rhamnolipids by Pseudomonas aeruginosa. Appl. Microbiol. Biotechnol 2005, 68, 7 18-725.
Soderman, 0.;I. Johansson. Polyhydroxyl-Based Surfactants and Their Physico-Chemical Properties and Applications. Cum Opin. Colloid Inte6ace. Sci. 2000, 4, 391-401.
Spoeckner, S.; V. Wray; M. Nimtz; S. Lang. Glycolipids of the Smut Fungus Ustilago maydis from
Cultivation on Renewable Resources. Appl. Microbiol. Biotechnol. 1999, 51, 33-39.
Stubenrauch, C. Sugar Surfactants-Aggregation, Interfacial, and Adsorption Phenomena. Cum
Opin. ColloidInte$ace. Sci. 2001, G,16170.
Teng, S.F.; K. Sproule; A. Husain; C.R. Lowe. f i n i t y Chromatography on Immobilized Biomimetic Ligands Synthesis, Immobilization and Chromatographic Assessment of an Immunoglobulin G-Binding Ligand. J Cbromatogx B, 2000,740, 1-1 5.
Ueno Y.; Y. Inoh; T. Furuno; N. Hirashima; D. Kitamoto; M. Nakanishi. NBD-Conjugated
Biosurfactant (MEL-A) Shows a New Pathway for Transfection. J Control. Release. 2007, 123,
247-2 53.
Uchida, Y.; S. Misawa; T. Nakahara; T. Tabuchi. Factors Mecting the Production of Succinoyl
Trehalose Lipids by Rhodococcw erythropolis SD-74 Grown on n-Alkanes. Agric. Biol. Chem.
1989,53,765-769.
Van Bogaert, I.N.A.; K. Saerens; C. De Muynck; D. Develter; W. Soetaert; E.J. Vandamme.
Microbial Production and Applications of Sophorolipids. Appl. Microbiol. Biotechnol. 2007, 76,
23-34.
Vankar, Y.D.; R.R. Schmidt. Chemistry of Glycosphingolipids-CarbohydrateMolecules of Biological Significance. Chem. SOC.Rev. 2000,29,201-216.
Vollbrecht, A.E.; U. Rau; S. Lang. Microbial Conversion of Vegetable Oils into Surface-Active
Di-, Tri- and Tetra-Saccharide Lipids (Biosurfactants) by the Bacterial Strain Tsukamurella sepc.
FettLipid. 1999, 101,389-94.
Wakamatsu, Y.; X. Zhao; C. Jin; N. Day; M. Shibahara; N. Nomura; T Nakahara; T. Murata;
K.K. Yokoyama. Mannosylerythritol Lipid Induces Characteristics of Neuronal Differentiation
in PC12 Cells Through an ERK-Related Signal Cascade. Eur. J Biochem. 2001,268,374-383.
Watanabe, K.; Y. Nakama; T. Yanaki; H. Hoffmann. Novel Vesicle and Sponge Phase Prepared

272 0 D. Kitamoto et al.

in Amphoteric Surfactant/Anionic Surfactant/Oleic Acidmater System. Langmuir. 2001, 17,

72 19-7224.
Worakikanchanakul, W.; T. Imura; T. Morita; T. Fukuoka; H. Sakai; M. Abe; R. Rujiravanit; S.
Chavadej; D. Kitamoto. Formation of W / O Microemulsion Based on Natural Glycolipid Biosurfactant, Mannosylerythritol Lipid-A. J Oleo Sci. 2008a, 57, 55-59.
Worakikanchanakul, W.; T. Imura; T. Fukuoka; T. Morita; H. Sakai; M. Abe; R. Rujiravanit; S.
Chavadej; H. Minarnikawa; D. Kitamoto. Aqueous-Phase Behavior and Vesicle Formation of
Natural Glycolipid Biosurfactant, Mannosylerythritol Lipid-B. Colloid Surf: B. Biointefaces
2008b, online published March 28.
Zhang, Y.J.; Y.Ishigami; EX. Ji; Y. Mitsutsuka; H. Hidaka; E Ishii. Organized Assembly from
Biosurfactants. Composite Zntefuce. 2004a, I I , 2 13-22 1.
Zhang, L.; l? Somasundaran; S>K> Singh; A.l? Felse; R. Gross. Synthesis and Interfacial Properties
of Sophorolipid Derivatives. Colloid Su$ A. 2004b, 240,7542.
Zhao, X.; T. Murata; S. Ohno; N. Day; J. Song; N. Nomura; T. Nakahara; K.K. Yokoyama.
Protein Kinase Ca Plays a Critical Role in Mannosylerythritol Lipid-Induced Differentiation of
Melanoma Bl6 cel1s.J Biol. Chem. 2001,276,39903-39910.
Zhou, S.; J. Wang; W. Gao; R. Akhverdiyeva; V. Shah; R. Gross. SupramolecularAssemblies of a
Naturally Derived Sophorolipid. Langmuir. 2004,20,7926-7932.

Sugar-, Polyol-, and Aminobased Lipids: Biodegradable and

BiocomDatible Surfactants for
Foods, health Care Products,
and Pharmaceuticals
I

Basic Properties of Sucrose Fatty

Acid Esters and Their Applications
Naoya Otomo
Mitsubishi-KagakuFoods Corporation

Introduction
Sucrose fatty acid esters (SEs) are recognized as very functional emulsifiers. They
are mainly used in the food industry because they are safe and environmentally
friendly. Cosmetic, pharmaceutical, and industrial uses are also growing steadily. The
environmental friendlinessis a consequence of the naturalness of the raw materials,
namely sucrose from sugar cane and fatty acids derived from vegetable fats.
The production of SEs was first commercialized by Dai-Nippon Sugar
Manufacturing Co. Ltd. in Japan in 1959, and the business was subsequently
developed by Mitsubishi Chemical Co. and Dai-Ichi Kogyo Seiyaku Co., both also
Japanese companies. These two companies dominate the worldwide manufacture of
SEs, and the total production capacity was estimated as more than 5500 tonnes in
2007.
This chapter describes the application of commercial SE products manufactured
by Mitsubishi-Kagaku Foods Corporation (Tokyo, Japan) with the brand name
Ryoto Sugar Ester. Table 10.1 shows the list of them and their ester compositions.

Chemical Structure of Sucrose Esters

SEs are made by the interesterification reaction between sucrose and methyl esters of
fatty acids in the presence of a solvent and an alkaline catalyst. Some basic research
was conducted on the enzymatic synthesis of saccharide esters (Dang et al., 2005), but
chemical synthesis is currently the only viable method for commercial production.
The sucrose molecule has eight free hydroxyl groups (Fig. 10.1). By changing the
degree of esterification (DE) and the fatty acids esterified, one can make a wide range
of SEs. The resultant products are mixtures of esters having different DE (see Fig.
10.2). One can classi& SEs according to their HLB value, where HLB stands for
hydrophilic-lipophilic balance. HLB is a numerical index indicating the relative
hydrophilicity and lipophilicity of emulsifiers. For example, sucrose stearate with
an HLB of 16 (high-HLB, hydrophilic) contains about 78% of monoester, 19%
of diester, and 2-3% of tri- and higher esters. Low-HLB esters have a high DE,
275

276 0 N. Otomo

Table 10.1, Typical Grades of Ryoto Sugar Ester Manufactured by Mitsubishi KagakuFoods CorDoration
Auuroximate ester comuosition, wt%
Fattyacid

Type

HLB

mono-

di-

tri-

tetra- and higher

Lauric (C12)

L-595

30

40

25

L-1695

16

80

18

Mvristic(C14)

M-1695

16

83

15

Palmitic (C16)

P-170

10

87

P-1670

16

80

18

Stearic (C18)

5-170

10

85

5-370

20

30

30

20

5-570

30

35

25

10

5-1170

11

55

30

10

5-1670

16

77

20

Behenic (C22)

8-370

15

25

25

35

Oleic IC18:l)

0-170

10

85

0-1570

15

75

22

ER-290

20

72

Erucic (C22:l)

Source: Mitsubishi Chemical, unpublished data.

0 C6

C6'

P -D-Fructofuransy1 a -D-glucopyranoside 6- octadecanoate

Fig. 10.1. Chemical structure of sucrose monostearate.

Basic Propertiesof Sucrose Fatty Acid Esters and Their Applications 0 277

HLB
16

Mono
Di
Tri
Tetra
Penta
Hexa
Hepta
Octa

11

7
3
1
0%

50%

100%

Fig. 10.2. Relationship between HLB and ester composition of sucrose stearates (Mitsubishi Kagaku
Foods Corp.,Technical information brochure, 2000).

and contain a small percentage of filly esterified octaester. Fig. 10.3 shows how the
combination of different esterified fatty acids and HLBs creates esters suitable for
different applications.
Hydrophilicity is derived from various groups (e.g., -OH, -0-,and -COOH),
and is affected by many environmental factors, like temperature, salt concentration,
pH, and so on. Also, hydrophilicity is affected by these factors to different extents.
For example, hydrophilicity of the -OH group (SE) is not affected by temperature,
while that of the -0-group (polyoxyethylene) becomes lower at a high temperature.
Historically, the HLB system was developed for linear alkyl ethoxylates. It is a useful
index for an approximate classification of these emulsifiers. However, if one mixes an
SE with an HLB of 16 and a linear alkyl ethoxylate (AE) with an HLB of 10 at the
ratio of 1:I (w/w), the behavior of the surfactant mixture would not necessarily reflect
that of a surfactant with an HLB value of 13.
In the production of SEs, the use of solvents such as dimethylformamide (DMF)
or dimethylsulfoxide (DMSO) is necessary to mix the raw materials uniformly. Since
both solvents have high boiling points, very sophisticated purification processes are
required to reduce the content of solvent to conform with the F A O M 0 (Food and
Agriculture Organization/World Health Organization) limits (DMF: not more than
1 mg/kg; DMSO: not more than 2 mg/kg).
The primary -OH group on C6 of the glucose moiety is the first to be esterified.
In monoesters, more than 70 mol% of ester is that of C6. After the C6 hydroxyl

278 0 N. Otomo

HLB

Fig. 10.3. Application map of sucrose fatty-acid

information brochure, 2000).

esters. (Mitsubishi-Kagaku Foods Corp., Technical

group of the glucose moiety, the C6 and C1 hydroxyls of the fructose group are the
next to be esterified. Thus, sucrose diesters are composed mainly of C6 and CG-ester
linkages. Because of the chemical structure, pure diester is easily crystallized. The
melting point of sucrose distearate is over 100C (Fig. 10.4; Otomo, 2003). Foodprocess temperatures are rarely above this temperature, except for heat sterilization
processes. Therefore, the pure ester, having such a high melting point, is not suitable
for use in foods. However, commercial sucrose stearate is a mixture of esters with
different DE (see Table 10.1). The resultant melting point is around 55-65C, and
therefore is suitable for applications in food processing.

Molecular Shape of Sucrose Fatty Acid Esters

When we try to understand the hnctionality of emulsifiers, we should consider the
chemical structure and shapes of the molecules. Surface pressure measurements are
usehl for estimating the molecular shape of emulsifiers. Fig. 10.5 shows the schematic
illustration of a Langmuir film balance with a Wilhelmy Plate. The surface pressuresurface area isotherm, or simply isotherm is recorded by reducing the area within
the barriers at a constant rate while continuously monitoring the surface pressure.
From these measurements, we were able to determine the available surface area of
each molecule (KSV Instruments USA, 2009).
After we tested SEs by this method, we were able to estimate the molecular shapes
of mono- to tetrastearate. Fig. 10.6 (Otomo, 2003) shows the estimated molecular

Basic Properties of Sucrose Fatty Acid Esters and Their Applications 0 279

o 3.0
w

3 2.0
E
1.o

Fig. 10.4. Differential scanning calorimetry (DSC) thermogram of sucrose stearates: (a)puredistearate,
and (b)monostearate/distearate l/lg/g.Temperature program: 0 to 150C, 5"C/min; sample weight of
2mg. Apparatus: Seiko Instruments Inc. EXSTAR 6000DSC. Otomo (2003)(Reprintedwith permission
from the Japan Society for Food Engineering).

Fig. 10.5. Schematic illustration of a Langmuir film balance with a Wilhelmy plate electrobalance
measuring the surface pressure,x , and barriers for controlling the available surface area, A.

shapes of these esters. The relationship between the average DE and molecular area is
shown in Fig. 10.7 (Otomo, 2003).
The results are shown in Fig. 10.6. In this figure, the top of each structure is
the hydrophilic part. As shown in Fig. 10.6, sucrose monoester is conical, and the
di- and tri-esters are cylindrical in shape. As mentioned above, commercial SEs are
a mixture of esters with different DE. SEs with a high content of monoester tend to
form micelle structures. These SEs are used for oil in water ( O W ) emulsions, and
solubilization of oil in water.
SE mixtures containing mono-, di-, and triesters tend to form lamellar structures.
A lamellar structure is very important for the stability of 0" emulsions. SEs with an
average DE around 2.7 are most suitable for the formation of lamellar structures. The

280 0 N. Otomo

monostearate distearate

tristearate

tetrastearate

monolaurate monooleate

Fig. 10.6. Schematic illustration of molecular shapes of sucrose esters at 293 K.The top of each
structure is the hydrophile, sucrose. Otomo (2003)(Reprinted with permission from the Japan Society
for Food Engineering).

Average Degree of Esterification

Fig. 10.7. Relationship between average
. * . . degree of esterification and molecular area of sucrose stearates
&T+
at 2 9 3 K . m water layer, and
oil layer. Otomo (2003)(Reprinted with permission from the Japan
Society for Food Engineering).

iii~~i~~~:

Basic Properties of Sucrose Fatty Acid Esters and Their Applications 0 281

thickness of the internal water layer of multilayer lamellar systems can be increased
by adding a small amount (5-10 wt% to total SE) of ionic surfactant. Fig. 10.8 is a
schematic representation of the effect on the internal water layer of adding an ionic
surfactant (e.g., monoglyceride citrate or succinyl monoglyceride) to a system using
sucrose stearate as the principal emulsifier. Using SAXS (small-angle x-ray scattering)
analysis, we observed the expansion of the water layer to 25 nm (Mitsubishi Chemical
Corp., unpublished data). In this test, we used S-1170 (seeTable 10.1), and the total
emulsifier concentration was 30% in water. The test was done at 25 and 60C by
using X-ray apparatus of 5OkV, 200 mA, Ni-filter (Anton-Paar) with a Kratky camera
U-slit.
This expansion of the water layer helps to prevent coalescence of the oil droplets.
This is comparable to the thickness of protein film formed when using caseinate. That
is why ionic surfactants, especially anionic, play a significant role in emulsion stability.
SEs with an average DE higher than 4 tend to form reverse micelle structures.

Surface Activity of Sucrose Fatty Acid Esters

The sucrose moiety of SEs is the key factor in their unique interfacial properties. Fig.
10.9 shows the surface-tension measurement of sucrose monopalmitate (Takagi et al.,

sucrose stearate
only

sucrose stearate +
anionic surfactant

Fig. 10.8. Schematic illustration of the expansion of internal water layer thickness in a multilamellar
system by the addition of an anionic surfactant to an aqueous sucrose stearate solution.

282 0 N. Otomo
1995). This shows the quite unique behavior of SEs when compared with conventional
linear AEs (Ueno et a]., 1981). The surface tension of sucrose-monopalmitate
solutions equals that of pure water at a low concentration, and then drops sharply
near the critical micelle concentration (CMC), whereas with AEs, the surface tension
decreases gradually as the concentration increases.
This behavior means that in the region between a and b of Fig. 10.9, SE molecules are not adsorbed at the interface. It is more thermodynamically stable for the
SE molecules to exist as monomers rather than as components of aggregates. In other
words, SE molecules have a strong affinity to the water structure. Above point b, SE
molecules begin to form pre-micelles, lose their affinity to water structure quickly,
and begin to be adsorbed at the interface. This explains the strong hydrophilicity of
SE molecules that leads to the high efficiency of adsorption at interfaces.
We also found this property of high adsorption efficiency in water-in-oil (W/O)
emulsion formation (Fig. 10.10, Otomo et al., 2000). We made a W / O high internal
phase emulsion (HIPE), which had an extremely high-volume fraction of dispersed
water, by using a hydrophobic sucrose erucate in the oil phase (i.e., ER-290) (see
Table 10.1). Also a small amount of a hydrophilic sucrose stearate S-1670 (see Table
10.1) was added to the water phase. This combination of two types of emulsifiers is
especially useful to improve the stability at lower temperatures. We assume that the
combination makes a molecular assembly with a stiffer interfacial membrane. This
synergistic effect is because sucrose stearate is adsorbed at the interface very efficiently.
When we used polyoxyethylene sorbitan monolaurate (Tween 20) in place of the
sucrose stearate , the emulsion became very unstable and collapsed in the course of
preparation. We observed the same instability by using polyoxyethylene sorbitan
monopalmitate (Tween 40). The hydrophilicity of the Tween 20 and 40 decreases at
a higher temperature in aqueous solutions, and the surfactant phase separates as an
oily liquid from water due to the dehydration of the ether group. This temperature
is called the cloud point. This means ethoxylates can dissolve in the oil phase, and
explains why the polyoxyethylene sorbitan fatty acid ester did not stabilize the
interfacial membrane.
This strong hydrophilicity remains, even in oil- soluble SEs having DE of around
5. Oil-soluble SEs can be dissolved in oil, but if an O/W interface is present, the SEs
will readily orientate at the interface due to the lipophobicity of the remaining -OH
groups on the sucrose moiety.
Lipophobicity is a relatively new concept when considering surfactant properties.
A surfactant is considered efficient when it has a low saturation concentration of its
monomeric form in water or oil solvents. Hydrophilicity is strongly dependent on
alkyl chain length, while lipophobicity is mainly related to the length and type of
hydrophile group. Fukuda and Shinoda (1999; Fukuda, 2005) tried to measure the
lipophobicity of different surfactants. They concluded that the lipophobicity of each
-OCH,CH(OH)CH,- unit (one secondary hydroxyl group) corresponds to about 5.6
times greater than that of -OCH,CH,- (oxyethylene group). Because oil-soluble SEs

Basic Properties of Sucrose Fatty Acid Esters and Their Applications 0 283

80

-b 60
P

20

-8

-7

-6

-5

-4

-3

logC(mo1 L-l)
Fig. 10.9. Measurement of surface tension, y, for (0)
sucrose monopalmitate and (4octaethyleneglycol
monotetradecyl ether as a function of concentration at 303.1 K. Positionsa and bare referred in the
text.Takagi et al. (1995) and Ueno et al. (1981) (Reprintedwith permission from Japan Oil Chemists
Society and Springer).

Fig. 10.10. Stability test for water in oil emulsions formed by the system water/oil-soluble surfactant/
water soluble surfactanthqualene 89.9/1.0/0.1/9.0 w/w/w/w at 5C. From left to right: ER-290 1% only,
ER-290 1% 5-570 0.1%; ER-290 P-1670 0.1%; ER-290 1% +Tween 20 (polyoxyethylene(20) sorbitan
monolaurate)0.1%, respectively. Stored at 5C for 5 weeks. ER-290,5-570 and P-1670 are sucros ester
products identified inTable 10.1. Otomo et at. (2000) (Reprinted with permission from the Society of
Cosmetic Chemists of Japan).

284 0 N. Otomo

have many hydroxyl groups, they are strongly lipophobic. Thus, they are very efficient
in many emulsion systems. Furthermore, the lipophobicity of SEs is not affected by
temperature change, because the hydration of the' hydroxyl group is much stronger
than that of an ether group.
Fig. 10.1 1 shows the temperature dependence of the interfacial tension between oil
and water (Otomo, 2007). In the case of glycerol monostearate (GMS), the molecule
is dissolved in the bulk oil phase at a high temperature due to the structural similarity
to the triglycerides, and thus interfacial tension is not low. As the temperature drops,
its solubility decreases and begins to adsorb at the interface.
The interfacial tension of sucrose pentastearate S- 170 and pentaoleate 0-170
(see Table 10.1) is not temperature-dependent because of the sugar esters' strong
lipophobicity. Fig. 10.12 shows the concentration dependence of interfacial tension
(Otomo, 2007). GMS (0.1%) is necessary to decrease the interfacial tension; but in
the case of sucrose pentastearate, 0.01% is sufficient for a temperature less than 20C.
Again, this is due to the strong lipophobicity and high efficiency of SEs.
The reduction of interfacial tension in ONV emulsions has great importance in
food systems because proteins and emulsifier molecules show competitive adsorption
at the interface (Dickinson et al., 1993). Consequently, they influence emulsion
stability and rheology.

Applications of Sucrose Esters

Bacteriostatic Activity of Sucrose Monopalmitate
Sucrose monopalmitate has bacteriostatic activity against spores of heat-resistant
bacteria. This activity is important for retorted long-life foods, such as canned coffee

.
%

-25
f
Y

GMS:Glycerol monostearate
0-170 : Sucrose pentaoleate
S-170 : Sucrose pentastearate

20

$8 15

s?

S-170

0
0

10

20
30
40
Temperature
["CI

50

Fig. 10.1 1.Temperature dependence of interfacial tension between canola oil and water For three
different nonionic surfactants. Emulsifier concentrations: 0.1% in oiLTemperature programming: 50C -+
5C by -O.Ol"C/s.
Apparatus: model K100MK2 surface tensiometer from Kruss, Helsinki, Finland. Otomo
(2007) (Reprintedwith permission from Science &Technology, Inc., Tokyo, Japan).

Basic Propertiesof Sucrose Fatty Acid Esters and Their Applications 0 285

S-170 (sucrose pentastearate)

Glycerol monostearate
25

Blank

Blank

ZE

E 20

,E I5
C

3 10

5
0
0

10

20

30

Temperature [TI

40

50

0
0

10

20
30
40
Temperature [TI

50

Fig. 10.12. Concentration dependence of interfacial tension between canola oil and water for two
nonionic surfactants. Temperature programming and instrumentation given in Fig.lO.11. Otomo (2007)
(Reprinted with permission from Science &Technology, Inc., Tokyo, Japan).

sold in Japan, Korea, and other Asian countries. In winter, the cans are bought hot
from vending machines. In the vending machine, the canned coffee is held hot, about
55C (131"F), until sold. This temperature is ideal for the growth of thermophilic
bacteria; also these beverages contain high levels of sugar, proteins, oils and other
nutrients, providing an ideal environment for bacterial growth (Fig. 10.13).
These canned beverages are sterilized under very severe conditions, such as 121"C
(250F) for 30 - 40 minutes. Table 10.2 shows the heat resistance of the spores of
thermophilic bacteria. D-value is defined as the time necessary to reduce the number
of bacteria to one-tenth of the original value. In the example, the temperature is
121C. For commercial sterilization, at least four times the D-value is required. Table
10.2 shows the sterilization times for six times the D-value for a range of microbial
organisms and spores. Some spores are extremely heat- resistant, and they can survive
the sterilization conditions. For example, if we use Moorefa thermoacetica as a target
bacterium of a canned food, it requires a heat treatment of 121C (250F) for 270
minutes for commercial sterilization. Such a long sterilization process results in badquality food, poor production efficiency, and big energy consumption.
SE inhibits the germination process at a very low dosage. Suwa et al. (1989)
showed that sucrose palmitate was the most effective of the different fatty acid esters
(Fig. 10.14) (Otomo, 2007). Also, they showed that monoesters were more effective
than more highly substituted esters. In canned coffee, 300 to 600 ppm of Ryoto' SE
P-1670 (80% of monoester content) is the standard dosage. However, this depends on
the ingredients (fadoil and protein content), homogenization pressure, sterilization
conditions, and so on. In a pure buffer system, sucrose monopalmitate inhibits spore
germination at as low as 10 ppm.

286 0 N. Otomo

Without SE

Sucrose monopalmitate
500ppm

Fig. 10.13. Flat sour spoilage of milk coffee stored at 40C for 2 months. UHT, ultra high temperature
sterilization, condition; 137C for 60s. Inoculated with Geobacillus stearothermophilus spore (Mitsubishi
Chemical, Unpublished data).

Starch forms complexes with fatty acid esters such as monoglycerides. Similarly

SEs also form this type of complex with starch. Thus SEs effectiveness against
bacterial spores is reduced in starch-containing foods. Recently, Sashida et al. (2007)
reported that a mixture of sucrose dicaprylate and tricaprylate (C8 fatty acid) had
high activity toward thermophilic bacterial spores, and the activity was not inhibited
by the presence of starch. We assume that di- and tricaprylate are more sterically
hindered than monopalmitate, and do not form complexes with starch.
Noteworthy is that this activity of SEs is, unlike that of sorbic-acid or benzoicacid esters, not bactericidal but is bacteriostatic. Sucrose monopalmitate inhibits the
germination process of the life cycle of spore-forming bacteria (Fig. 10.15). If the
spores are washed and transferred to an ester-free culture broth, they will germinate
and grow. Vegetative growth of these bacteria is only slightly affected by the ester.
Moriyama et al. (1996) tried to establish the mechanism of this activity. They found
that esters became bound to the spore coat; but this does not fully explain the mode
of action.

Basic Propertiesof Sucrose Fatty Acid Esters and Their Applications 0 287

Table 10.2. Effect of Sucrose Monopalmitate (P-1670) on the Heat Stability

of Microbesat 121OC.
Species

Heating
Sugar Ester
Time' (min) Effectb
Notes

Bacilluspolymyxa

0.005

spore

Clostridium Dasteurianum

0.14

spore

Clostridium perfringens

1.4

++
+

Clostridium botulinumType A

1.3-3.1

spore, food
poisoning

Clostridium botulinum Type Ec

spore, food
poisoninq

Bacillus licheniformis

1.8

Bacillussubtilus

2.6-3.2

Clostridium sDoroo4enes

2.0-4.0

spore, food
poisoning

spore

spore

+
++

spore
spore

spore

Bacilluscoagulans

19.8

Geobacillusstearothermophilus

18-65

Thermoanaerobacter
thermohydrosulfricus

42

++
+

Moorelra thermoacerica

180-270

++

spore

spore

'6 times of D value(Decima1reduction time) a t the indicated temperature

(106+<1); b++: suppressed below 100 ppm +: suppressed between 100-1000
ppm, f:suppressed a little : suppressed very little in broth: '82C
(MitsubishiChemical. Unpublisheddata)

Dispersion of Solid in Aqueous Media

Calcium is an essential element for humans, especially young children during their
growing period and the elderly, who need to maintain their skeletal strength. In Asian
countries, the intake of dairy products is much lower than in Europe or the Americas.
Thus, calcium intake is sometimes insufficient. To cover the nutritional shortage,
calcium-fortified foods are becoming more popular. For example, a considerable
market for calcium-fortified reconstituted milk was developed in Japan.
In this type of milk, calcium carbonate is dispersed in aqueous media. Since
the specific gravity of the salt is more than 2.7, hydrocolloids, such as carrageenan,
are commonly used to maintain the suspension of the solid particles. However, the
consistency of this type of reconstituted product is not ideal.
Sucrose monostearate is effective in providing a good dispersion of solids in
aqueous systems. Fig. 10.16 shows the electrostatic repulsive potential between
particles according to DLVO theory, (named after Derjaguin, Landau, Venvey, and

288 0 N. Otomo

L- 1695

m 250PPm
200ppm

M- 1695
P- 1695

S- 1695

0-1695

CONTROL]
0

10

15

20

25

Apparent D,,,.c value ( min.)

Fig. 10.14. Reduction of apparent D value of Moorela thermoacetico in milk coffee by sucrose fatty acid
esters. Formulation: whole milk powder: 1.896, skim milk powder: 1.696, coffee extract: 2.096, sucrose:
8.0%. P-1695,5-1695 and 0-1695 are laboratory scale products made with highly purified fatty acids
with monoester content approximately equal to 8096, with nomenclaturelisted inTable 10.1. Otomo
(2007) (Reprinted with permission from Science &Technology ,Inc., Tokyo, Japan).

Spore

Sucrose monopalmitate inhibits the

germination of spores

Sporangium

Q;mination

Sporulation

(23
*I-*,

+:7

&
L
(
J
Vegetative Growth

Vegetative Cell

Fig. 10.1 5. Life cycle of heat-resistant spore-forming bacteria.

Outgrowth

Basic Propertiesof Sucrose Faay Acid Esters and Their Applications 0 289

Overbeek who developed it in the 1940s). In this figure, the x-axis represents the
distance between two droplets, and the y-axis is the electrostatic potential, where
+ represents a repulsive force. The bold line is the resultant potential which is the
cumulative effect of the repulsive potential and the van der Waals attractive potential.
The resultant potentid is mostly negative (i.e., attractive over small distances), but
may be positive (repulsive) at intermediate distances, leading to a stable colloidal
system (Hiemenz, 1986).
Although sucrose monostearate is a nonionic emulsifier, it confers a negative
charge to particles and stabilizes the dispersion. SE forms a bilayer around calciumcarbonate particles. Fig. 10.17 shows the results of an experiment that can help
illustrate the nature of this adsorption (Otomo, 2003). We added an equal volume
of hexane to aqueous dispersions containing various concentrations of SE, from 0.2
to 1.5%, and containing 10% of solid calcium carbonate. At a low level of SE, SE
molecules are adsorbed to form a monolayer. The hydrophobic tails faced the bulk
water phase. Thus the surface becomes hydrophobic, and particles migrate into the
hexane phase. As the concentration of SE increases, a double layer is formed around
the particles, and the surface becomes hydrophilic.
We found that the &-potential of untreated calcium-carbonate particles was
positive. After adding sucrose stearate, it turned negative. The subsequent addition of
anionic surfactants (e.g., succinyl monoglyceride and monoglyceride citrate) enhanced
the negative charge (Fig. 10.18). The difference in the absolute value of the potential
did not appear to be significant, but the effect on the stability of the dispersion was
remarkable (Fig. 10.19). This is because the &-potential is the electrical potential of
the sliding surfaces of the particles (Microtec, 2009), and does not reflect the charge
of particles directly. Assumably, the addition of anionic surfactants increased the
hydration radius of the particles leading to the much greater stability of the dispersion
(Otomo, 2007).
Viscosity Control of Molten Chocolate
Viscosity control is important in chocolate manuhcturing. Viscosity is affected by
many factors. The selection of emulsifiers is especially important (Babin et al., 2005;
Lee et al., 2002; Sevais et al., 2004). Sucrose pentaerucate ER-290 (see Table 10.1)
has unique viscosity-reducing properties for liquid chocolate.
In the manufacture of chocolate, low viscosity is desirable when the chocolate is
liquid. Once the liquid chocolate is poured into the mold, solid-like properties are
desirable. Hence, ideally, chocolate should have low plastic viscosity, and its yield
value should be almost unaffected by the presence of the viscosity-reducing agent (Fig.
10.20). Lecithin is widely used as a viscosity-control agent, but although it reduces
the plastic viscosity, it increases yield value significantly at a high dosage. Polyglycerol
polyricinoleate (PGPR) is also a well-known viscosity- reducing agent in chocolate.
It has the ability to strongly reduce the yield value, even to zero, but has little effect
on plastic viscosity. However, the sucrose pentaerucate efficiently reduces the plastic

290 0 N. Otomo

P
(f
+ repulsive

PL

Electrostatic Repulsive
Potential

The resultant potential

Distance
Distance between
two dropleis

- attractive
Stable Colloidal System

van der Waals

attractive
potential

Unstable System (Coagulation)

Fig. 10.16. An electrostatic repulsion force leads to stable colloidal suspensions (DLVOTheory).

Hexane layer

Aqueous layer

Low SE concentration

High SE concentration

Fig. 10.1 7. Effect of sucrose ester content on the phase behavior of an aqueous suspension of calcium
carbonate particles (10 ml) mixed with hexane (10 ml) at room temperature. Concentrationof 5-1670
(sucrose monostearate; seeTable 10.1) in original aqueous solution were: 0.2,0.6,1.0, and 1.5 wt.%
from left to right. Otomo (2003) (Reprinted with permission from Japan Society for Food Engineering),

Basic Properties of Sucrose Fatty Acid Esters and Their Applications 0 29 1

::-

>
E

CHzO C

(CHz)iCH3

CH-OH

CHz-O-$-(Oi

Zh-COOH

Main component of
succinyl monoglyceride
CaC0,only

S-1670

S-I670/succinylmonoglyceride=l/l w/w

Fig. 10.18. Zeta potential of calcium carbonate particles (10 wt.%) in an aqueous solution that contained 1 wt.% total emulsifier, measured at a dilution of 1/100 and 293K 5-1670 sucrose monostearate
(seeTablelO.1); Succinyl monoglyceride: An anionic emulsifier consisting of a complex mixture of esters made from succinyl anhydride and monostearate. Otomo (2007) (Reprinted with permission from
Science & Technology Inc., Tokyo, Japan).

30

3.

.-paEg
aJ

15

.
I

10
0

S-16701 Succinyl Monoglyceride

i
a

= 111 wlw

12

18

24

30

36

42

48

54

60

66

Storage period (days)

Fig. 10.19. Stability of an aqueous dispersion of calcium carbonate (10% w/w) in the presence of
Iwt. % total emulsifier during storage at 5C. Emulsifiers described in Fig. 10.18. Apparent increase
of diameter was due to aggregation of calcium carbonate particles. Otomo (2007) (Reprinted with
permission from Science &Technology Inc., Tokyo, Japan).

292 0 N.Otomo

tan 8 ; Plastic viscosity

The force required to maintain
constant flow in a fluid mass

f ; Yield value

Shear rate

The force required to initiate

flow in fluid mass

Fig. 10.20. Flow curve of molten chocolate at 45'C.

Yield value
_-

Plastic viscositv

(The force required to initiate flow)

(The force required to maintain constant flow)

Plastic viscosity at 40C (104F)

Yield value at 40C (104F)

30

400

Lecithin
300

PGPR

Q 20

Lecithin

ER-290

ER-29U

'$ 100
i

PGPR
0

0
0

0.5

Added emulsifier %

0.5

Added emulsifier %

Fig. 10.21.The effect of emulsifiers on the plastic viscosity and yield value of semi-sweet chocolate
(cocoa butter 32%, Lecithin 0.45%). ER-290: Sucrose pentaerucate (seeTable 10.1); PGPR: polyglycerin
polyricinoleate. (Mitsubishi Chemical, Unpublished data.)

Basic Propertiesof Sucrose Fatty Acid Esters and Their Applications 0 293

viscosity without a great decrease of yield value (see Fig. 10.21). This property is
greatly advantageous to chocolate manufacturers, especially when they want to make
a low-calorie chocolate. The practical difference between sucrose pentaerucate and
PGPR emulsifiers is shown in Fig. 10.22 (Otomo, 2007). Some works were produced
on the effect of emulsifiers on chocolate viscosity (Babin et al., 2005; Rousset et al.,
2002); however, the mechanism of emulsifiers is not fully understood. Consequently,
manufacturers should select emulsifiers according to their practical requirements.

Crystal Control of Fat in Oil-in-Water Emulsion

In O/W emulsions containing solid fat, like coffee cream and whipping cream, crystal
growth in the oil droplets has a great influence on emulsion stability (Kaneko et
al., 1999; Katsuragi et al., 2000). The surfaces of dispersed oil droplets are covered
by surface-active substances, like dairy proteins or emulsifiers, immediately after
the preparation. Fat crystals can grow inside the oil droplets, and sometimes they
protrude through the surface membranes when the emulsion is stored at inappropriate
temperature conditions. The surfaces of protruding fat crystals are hydrophobic; thus
a hydrophobic interaction causes the coalescence of droplets.
Arima et al. (2007) reported that the combination of hydrophobic sucrose
pentapalmitate, P-170, and hydrophilic sucrose monopalmitate, P-1670 (see Table
1O.I), dramatically retarded the crystallization-induced destabilization of the OIW
emulsion that contains palm-mid-fraction (PMF) as the oil phase (Fig. 10.23, Sato,
2007). They assumed that sucrose pentapalmitate caused the interfacial heterogeneous
nucleation of PMF crystals. The pentapalmitate molecules may serve as a template
for nucleation when the temperature of the emulsion was lowered. The addition of

Fig. 10.22. Chocolate coating of biscuits. Left chocolate with 0.5% of ER-290 (sucrose pentaerucate: see
Table 10.1). Right: chocolate with 0.5% of PGPR. Otomo (2007) (Reprinted with permission from Science
&Technology Inc.,Tokyo, Japan).

294 0 N. Otomo

+P-16?0

2%

+P-170 1% + P-1670 2%

5 ~ m

Fig. 10.23. Microscopic images of oil in water ( O N )emulsion of palm-midfraction (PMF). OMI ratio:
20/80vlv. Emulsification achieved using a microfluidizer at 8.5 kg/cm2for 6 consecutive repetitions.
Stored at 5C for 2 days. PURE: polyoxyethylenesorbitan monolaurate (TweenZO) at 2 wt. % P-170
sucrose pentapalmitate; P-1670 sucrose monopalmitate (seeTable 10.1). Sato (2007) (Reprintedwith
permission).

the less highly esterified P-1670, (average DE is approximately 1.2) did not cause
heterogeneous nucleation. However, it is reasonably assumed is that the addition of
P-1670 retarded both the a + p transformation and the growth of large p crystals,
as evidenced by the X-ray diffraction study (Arima et al., 2007). The enhancement
of interfacial crystallization by sucrose pentapalmitate, P- 170 (forming tiny PMF
crystals and retarding the polymorphic transformation by Ryoto@Sugar Ester P-1670),
prohibiting the growth of needle-shaped crystals of PMF, combined to retard the
crystallization-induced destabilization of the emulsion.

Pharmaceutical Applications
SEs are also widely used in pharmaceutical applications in solid and liquid products.
Magnesium stearate is a widely used lubricant for tabletting. SEs are also effective
tabletting lubricants. In addition SEs are nonionic, and very little danger of them
reacting with medicinal chemicals is likely, unlike magnesium stearate. Therefore, SEs
are increasingly used in tablet production as a lubricant and sometimes as a releasecontrolling agent. Also, because these low HLB stearates are tasteless, one can use
them in chewable tablets (Shibata et al., 2002).

Basic Propertiesof Sucrose Fatty Acid Esters and Their Applications 0 295

Surfactants are commonly used to enhance the skin penetration of lipophilic

medicinal chemicals. Generally, highly hydrophilic surfactants are used as solubilizing
agents. Okamoto et al. (2005) reported that sucrose dilaurate had a remarkable effect
on skin penetration. The test was done by using a flow-through diffusion cell made of
Teflon (Fig. 10.24). The test material was applied to the upper-surface skin of hairless
mice , and the material that had penetrated became dissolved in phosphate buffer and
was detected continuously.
Fig. 10.25 shows the results for a test of lidocaine, a local anesthetic, and an alkaline
compound. Generally, skin penetration is improved if a compound is unionized. In
the case of lidocaine, pH 9 or 10 would be required to obtain this properly. This pH
is not realistic for skin applications. Instead of aqueous media, lidocaine was dissolved
in propylene glycol with SEs. High-HLB sucrose laurate showed almost no effect, but
sucrose dilaurate showed a remarkable enhancement of skin penetration.
The mechanism is not precisely understood, but supposedly the lamella structureforming tendency of sucrose dilaurate is the key factor, because skin has a lamellar
structure, too.

Conclusions
In conclusion, I hope that I demonstrated that SEs are powerful emulsifiers with
a wide range of applications in foods and nonfood systems. The sucrose molecule,
having eight hydroxyl groups as an acyl acceptor, enables the production of a wide
range of compounds. These range from the monosubstituted esters that are extremely
hydrophilic and are used to produce oil in water, to highly substituted esters that
facilitate water-in-oil systems. In addition SEs have unpredictable properties such as
being bacteriostatic against spore-formingorganisms and having the ability to facilitate
the suspension of calcium-carbonate particles. I hope the information contained in
this chapter is useful to readers who work in many industrial fields.
.Donor Phase
D w
Skin

Sampling

PBS

Drug

Sampling

PBS

Stirrc

Receptor Phase
Fig. 10.24. Flow-throughdiffusion cell used to investigate the penetration of skin.PBSrefers to phosphate buffer saline. Okamoto (2001) (Reprinted with permission).

296 0 N. Otomo

1. 2
1
n

8 0.8

0.4
0. 2
0
0

Time (hr)

Fig. 10.25. Effect of sucrose esters on lidocaine penetration through a propylene glycol matrix at 37C.
Conditions: lidocaine(5% in propylene glycol)with 1.5%emulsifier: (0)
sucrose L-595 (seeTable10.1);
(A)L-1695; or (0)no emulsifier (control). Okamoto (2001)(Reprintedwith permission).

References
Arima, S.; T. Ueji; S. Ueno; A. Ogawa; K. Sato. Retardation of crystallization-induceddestabilization of pmf-in-water emulsion with emulsifier additives. Colloids and Surfaces B 2007,55,

98-106.
Babin, H.; E. Dickinson; H. Chisholm; S. Beckertt. Interactions in dispersions of sugar particles
in food oils: influence of emulsifier. Food Hydrocolloids 2005, 19,513-520.
Dang, H.T.; 0. Oriana; D.G. Hayes. Feed batch addition of saccharide during saccharide-fatty
acid esterification catalyazed by immobilized lipase: time course, water activity, and kinetic
model./. Am. Oil Chem. SOC.2005,82(7), 487-493.
Dickinson, E.; G. Iveson; S. Tanai. Competitive adsorption in protein-stabilized emulsions
containing oil-soluble and water-soluble surfactants. Food Colloids and Polymer4 E. Dickinson, I?
Walstra, Eds.; The Royal Society of Chemistry: Cambridge, UK, 1993;p. 312.
Fukuda, M. 'The importance of lipophobicity in surfactants: methods for measuring lipophobicity
and its effect on the properties of two types of nonionic surfactant. J Colloid and Interface Sci.

2005,289,512-520.

Basic Propertiesof Sucrose Fatty Acid Esters and Their Applications 0 297

Fukuda M.; K. Shinoda. Importance of the lipophobicity of hydrophilic groups of surfactants and
evaluation. Nippon Yukagaku ffiishi, 1999,48(6), 587-594.
Hiemenz, I?C. Principles of Colloid and Surface Chemisty, 2ded.; Marcel Dekker Inc.: New York,
1986.
Kaneko, N.; T. Horie; S. Ueno; J. Yano; T. Katsuragi; K.Sato. Impurity effects on crystallization
rates of n-hexadecane in oil-in-water emulsions. J. C y t a l Growth 1999 197,263-270.
Katsuragi, T; N. Kaneko; K. Sato. DSC study of effects of addtion of sucrose fatty acid esters on
oil phase crystallization of oil in water emulsion. Nippon Yukagaku ffiishi 2000 49(3), 255-262.

KSV Instruments USA (Monroe, CT, USA) Langmuir and Langmuir-Blodgett Films: WHAT
and HOW? AN#107; 2009, http://www.ksvinc.com/LB.htm
Lee, S.; M. Heuberger; I? Rousset; N.D. Spencer. Chocolate at a Sliding 1nterface.J. Food
Sci.2002 G7(7), 2712-2717.
Microtec Co., Ltd. (Chiba, Japan) What is zeta potential?, 2009, http://nition.com/en/producdzeecom-s.htm
Moriyama, R.; K. Sugimoto; S. Miyata; T. Katsuragi; S. Makino. Antimicrobial action of sucrose
esters of fatty acids on bacterial spores.J. Antibact. Antt$ng. Agents 1996 24(I). 3-8.
Okamoto, H. Meijo University, Faculty of Pharmacy, Nagoya, Japan, Personal Communication,
2001.
Okamoto, H.; T. Sakai; K. Danjo. Effect of sucrose fatty acid esters on transdermal permeation of
lidocaine and ketoprofen. Biol. Pharm. Bull. 2005 28(9), 1689-1694.
Otomo, N. Self-organizedstructure of food emulsifiers and their applications. Nippon Shokuhin
Kougaku Kaishi 2003, 4(I), 1-9.
Otomo, N. Applications in food industry. Selertions andApplications of Surfactants; Science &
Technology Inc.: Tokyo, Japan, 2007; pp. 346-368.
Otomo, N.; Y. Watanabe; M. Tsuruta. A novel method for the stabilization of high internal phase
water-in-oil emulsions. J. SOC.Cosrnet. Chem. Jpn. 2000,34(3),,299-306.
Rousset, I?; I? Sellappan; I? Daoud. Effect of emulsifiers on surface properties of sucrose by inverse
gas chromatography.]. Chromatop A. 2002,9G9,97-101.
Sashida, R.; M. Tomida; K. Motoda; N. Matsumoto;T Kaji. Growth inhibitory effect of sucrose
fatty acid esters in starch containing foods. Kitnzume-Jihou 2007, 8G(10), 1162-1 163.
Sato, K. Hiroshima University, Faculty of Applied Biological Science, Hiroshima, Japan, Personal
Communication, 2007.
Servais, C.; H. Ranc; I.D. Roberts. Determination of chocolate viscosity.J. Terture Stuu, 2004,
34,467-497.
Shibata, D.; Y. Shimada; Y. Yonaawa; H. Sunada; N. Otomo; K. Kasahara. Application and
evaluation of sucrose fatty acid esters as lubricants in the production of pharmaceuticals. Yakuzaigaku 2002,G2(4), ,133-1 45.
Suwa, N.; H. Machida; A. Nishimura. Effect of sucrose fatty acid esters on spoilage of canned
coffee caused by thermophilic spore forming bacteria. Mitsubishi k k e i R6.0 Rev. 1989 3(2),
2633.
Takagi, K.; S. Hirai; Y. Fujinuma; H. Fujimatsu; H. Usami; S. Ogasawara; Y. Kasahara; A. Yuki.

278 0 N. Otomo

Surface properties in aqueous dilute solutions of sucrose monopalmitate. Yukagaku,1995,44(3),

207-210.
Ueno, M.; Y. Takasawa; H. Miyashige; Y. Tabata; K. Meguro. Effects of alkyl chain length on
surface and micellar properties of octaethyleneglycol-n alkyl ether. Colloid Pulym. Sci. 1981,259,

761-766.

Selective Enzymatic Synthesis

of N-Acylated Alkanolamine
Emulsifiers
Cristina Otero
lnstituto de Catdlisis, CSIC. Campus Universitario,Cantoblanco (28049)' Madrid, Spain.

Introduction
Biodegradable surfactants containing amide functional groups are of great interest
for applications requiring relatively stable emulsifiers because amide linkages are very
stable chemically and are not easily degraded in alkaline media. These emulsifiers are
nonionic and are completely biodegradable. They are a particularly attractive class
of compounds that are potential substitutes for emulsifiers derived from petroleum.
They are characterized by their skin tolerance, good biological degradability, and low
toxicity (Maag, 1984; Garrison, 1968).
The skin tolerance of surfactant formulations can be improved by adding fatty
acid alkanolamides.
Esters of amides have excellent antimicrobial activities. They are employed in
the manufacture of foods, cosmetics, pharmaceutical products, and a wide variety of
specialty chemicals that exploit the surface active properties of these materials. The
biodegradability and the relatively nonallergenic character of mono- and diethanol
amides of fatty acids depend on both the chemical nature of the substituents on the
nitrogen atom and the chain length of the fatty acid residue.
Table 1I. 1 summarizes the applications of alkanolamines whose constituent fatty
acid residues have different chain lengths. Amides with short-chain fatty acid groups
may be used as foam stabilizers, while other amines may be used as conditioning agents
or other purposes. Alkanolamines are frequently employed in formulating personal
care products to serve as foam boosters and viscosity enhancers in aqueous-based
cleansers. In addition, alkanolamines are utilized commercially in the manufacture
of textiles and in the formulation of industrial grade detergents. These compounds
perform a variety of functions including viscosity enhancement, foam stabilization,
emulsification, corrosion inhibition, and detergency.
Routes for the chemical synthesis of these compounds are not very specific and
require multiple steps for protection/deprotection of the hydroxyl groups (Maag, 1984).
9

299

300 0 c. Otero

Table 11.1. Functional Properties of Alkanolarnines

Number of Carbon Atoms
of the Fattv Acid Chain
<8

Ethanolamine

8-10

Good foam stability

Little viscosity improvement
Good foam boosters
Good viscosity builders

12-16

18+

Unsaturated(oleic, linoleic
acids)

Diethanolamine
~

Humectant agents
Hair conditioners

~~

_-___-

------

Good foam boosters

Some emollient properties
Some conditioning
properties
Reduced foam stabilization
Thickening agents in
formulation of hair dyes
Good viscosity builders
Some emollient properties
Lubricants in textile industry
Corrosion inhibitors
Reduced foam formation and stabilization
Good viscosity builders

A potential alternative to current technology is based on the use of biocatalysts.

Enzymatic reactions can be conducted under mild conditions (Khemelnitsky et al.
1988; Wescott & Klibanov, 1994). By conducting these reactions at relatively low
temperatures and pressures, one can increase the reaction selectivity using fundamental
principles of reaction and enzyme engineering (Arcos et al., 1998a, 1998b;Torres &
Otero, 1999; Torres et al., 1999).
It is assumed that formation of an amide bond can be catalyzed by proteases, but
these enzymes are generally highly specific for a given amino acid and are sensitive to
the organic solvent (Matos et al., 1987). Lipases are better catalysts for these reactions
than are proteases, acylases, and esterases (Furutani et al., 1996).
Lipases (triacylglycerol acyl ester hydrolases, EC 3.1.1.3) catalyze the hydrolysis
and interesterification reactions of their natural substrates (fats and oils) (Paiva et al.,
2000). Moreover, lipases are enzymes widely employed in synthetic organic chemistry
because of their ability to accept a great variety of substrates, including long-chain
fatty acids and substrates that do not occur naturally (e.g., alkanolamines) (Zaks &
Klibanov, 1984, 1985).
Enzymes catalyze the selective 0-acylations of several alkanolamines with fatty
acids (Kanerva et al., 1992a), and selective N-acylations of different P-alkanolamines
(Gotor et al., 1988; Kanerva et al., 1992b; Furutani et al., 1996; Maugard et al.,
1997). They also catalyze enantioselective (Gotor et al., 1988; Kanerva et al., 1992b)
and chemoselective reactions of ethanolamine (Ferndndez-Pirez & Otero, 200 I),
L-serine (Furutani et al., 1996), and N-methyl-glucamine (Maugard et al., 1997).
Diethanolamine is a precursor which contains two hydroxyl groups and an amine
group. Ethanolamine has only one hydroxyl and one amine groups. All these groups
are susceptible to acylation in the presence of a lipolytic enzyme. For substrates

Selective Enzymatic Synthesisof N-Acylated Alkanolamine Emulsifiers 0 30 1

containing several moieties susceptible to acylation, kinetic strategies designed to

produce selective acylation could be developed using lipases to effect these reactions
under mild conditions. However, under mild conditions the relatively high viscosity
and low solubility of diethanolamine, and the precipitation of the ionic pair complex
formed by the alkanolamine (or dialkanolamine) and the fatty acid moiety could limit
the extent of these two biotransformations.

0bjectives
This report describes the potential of enzymatic technology for selective N-acylation
of alkanolamine and dialkanolamine, namely ethanolamine and diethanolamine.
The strategies developed to accomplish this goal also consider and/or minimize
problems associated with the high viscosity and/or low solubility of the precursor
alkanolamine.
In this work we have employed different lipolytic enzymes as biocatalysts for
the acylation of alkanolamines with fatty acids, and explored the associated reaction
engineering to determine the optimum conditions for N-acylation ofethanolamine and
diethanolamine so as to accomplish these reactions in a short time while maximizing
both yield and selectivity. This work focuses on the lipase-catalyzed synthesis of
N-acylated alkanolamines of different fatty acids at low and high substrate loads.
The authors wanted to maximize the volumetric productivity of the process without
significant loss in yield or selectivity. To achieve this goal the authors employed high
substrate loads and explored the synthesis reactions using both commercial and tailormade immobilized lipases as biocatalysts.

N-Acylated Ethanolamine Biosurfactants

Experimental
Materials
A lipase from Candidz antarctica (Lipase B, a nonspecific lipase, immobilized on a
macroporous acrylic resin, designated Novozym 435), Novozym 868 (lyophilized
Candida antarctica A) and LypozymeTM(a lipase from Mucor miehei immobilized on
Duolite 568) were kindly provided by Novo Nordisk A/S (Bagsvaerd, Denmark).
Lyophilized Candidz rugosa lipase type VII was purchased from Sigma-Aldrich (St.
Louis, MO). A lipase from Pseudomonas sp. and Lipase G (lyophilized lipase from
Penicillium sp.) were supplied by Amano Pharmaceuticals Co. (Nagoya, Japan).
Ethanolamine (more than 98% pure), lauricacid (> 99% pure), and myristicacid (98%
pure) were purchased from Merck (Darmstadt, Germany). Capric acid (99-1 00%
pure), palmitic acid (95% pure), and lauric acid ethyl ester were purchased from
Sigma-Aldrich. Acetonitrile, chloroform, and methanol were all of HPLC quality and
were supplied by Scharlau (Barcelona, Spain). Before use, all solvents were dried with
molecular sieves with an effective pore diameter of 4A.

302 0 c. Otero

Enzymatic Reactions
For trials involving a 1:1 mole ratio of reactants, 0.1 mmol of both the ethanolamine
(6.1 mg) and the fatty acid (17 mg of capric acid, 20 mg of lauric acid, 22.8 mg of
myristic acid, or 25.6 mg of palmitic acid) were mixed with 0.5 mL of solvent in
stoppered glass bottles. Precipitation of the reactant mixture was observed. Additional
experiments were conducted with different mole ratios of lauric acid to ethanolamine.
The solution was placed in a thermoconstanter orbital shaker (Stuart Scientific, Surrey,
England) and held at the reaction temperature for 5 min. Subsequently 25 mg of
Novozym 435 were added and the mixture was shaken at 200 rpm for the indicated
time in the corresponding experiment. Molar yields were calculated with respect to
the limiting reactant.
Analysis of the Reaction Mixtures
The reaction was quenched at the reaction times indicated in the figures and tables,
and a mixture of methanol and chloroform [50:50 (vlv)] was added until a transparent
solution was formed. The immobilized enzyme was then removed by filtration with
a 0.1 mm sieve, followed by centrifugation. The volume of final solution was then
increased to 5 mL by addition of a mixture of methanollchloroform [50:50 (v/v)].
The composition of the solution was quantified with the HPLC using a Kromasil
W - 1 8 column (250 x 4.6mm) as previously reported (Ferndndez-Pkrez & Otero,
2001). For the lauric acid ethyl ester, the mobile phase employed was a 45/45/10
(v:v) mixture of acetone, acetonitrile, and water fed at 2 mL/min. The retention time
was 5.5 min.

Results and Discussion

Structural Elwidation
FTIR, NMR, and molecular mass data for the reaction products are given in Table 1 1.2.
Screening of Biocatalysts
Of the native and immobilized lipases investigated (Novozym 435, Pseudomonas
sp.lcelite, C. antarctica A, Mucor mieheilduolite 568, Pseudomona sp., C. rugosa,
Penicillium rp., C. antarctica Alcelite, Penicillium sp./celite, C. antarctica Alsilica,
and Penicillium sp./silica ), only Novozym 435 produced an appreciable amount of
reaction within 24 hours. This enzyme is an immobilized C. antarctica B lipase that
was used in subsequent experiments.
E#ect of Molar Ratio of Reactants
Trials were conducted using different molar ratios of reactants. Some trials involved
use of an excess of fatty acid while another series of trials involved a molar excess
of ethanolamine (Fig. 1 1 . 1 ) . The 0-acylated ethanolamine was not formed in a
significant amount in these experiments. These trials indicated that when acetonitrile
is the solvent, the reaction is completely selective to the intermediate N-acylated

Selective Enzymatic Synthesis of N-AcylatedAlkanolamine Emulsifiers0 303

Table 11.2. Structural Elucidation of N-acylated Ethanolamine Biosurfactants

N-Caproyl ethanolamine

IR Spectra

'H-NMR

(OH)= 3300 cm',

(CH)=2800-2900cm1,
and (CO-N)=1641-1560

'H-NMR (CDCI,, 6):5.93

(bs., 1 H, NY), 3.72 (m,
2H, -CH20),3.42 (m, 2H,
CH,-N), 2.70 (b.s, 1 H,
OH), 2.20 (t, 2H, J=7.6
Hz, CH2-CO), 1.63 (m, 2H,
-CH,-CH,-CO), 1.25 (m,
16H, CH, chain), 0.88 (t,
3H, J=7.05 Hz, CH,)
values within f 0.01
ppm of those found for
N-lauroyl ethanolamine

Cm"

N-Lauroyl ethanolamine

N-myristoyl ethanolamine

(OH)=3400 cm",
(CH)=2800-2900 cm'
and (CO-N)=1642-1561
cm-'
(OH)=3300 cml,
(CH)=2800-2900 cm"
and (CO-N)=1641-1555

values within f 0.01

ppm of those found for
N-lauroyl ethanolamine

Molecular
Masses
172.2

243.3

271.4

Cm-'

N-palmitoylethanolamine

(OH)=3300 cm",
(CH)=2800-2900 cm',
(CO-N)=1640-1554cm-'
(CH)=2800-2900cm-',
(CO-O)=1737cm-' and
(CO-N)=1641-1552cm'

values within f 0.01

299.47
ppm of those found for
N-lauroyl ethanolamine
1-0,2-~-dicaproyl
369.3
'H-NMR (acetoned6,6 1:
ethanolamine
7.20 (bs., lH, NH),4.09 (t,
2H, J=5.7 Hz, CH,O), 3.41
(4,2H, J=5.7 Hz, CH2N),
2.28 (t, 2H, J=7.5 Hz, CH2COO), 2.1 3 (t, 2H, J=7.5
Hz, CH;CO-N), 1.57 (m,
4H, 2~ CH,-CH,-CO), 1.29
(m, 32 H, 2x CH,- chain),
0.88 (t, 6H, J=6.9 Hz,
ZxCH,).
1-0,2-~-dilauroyl
(CH)=2800-2900cm-l,
values within f 0.01
425
ethanolamine
(CO-0)=1737 cm-' and
ppm of those found
(CO-N)=1639-1551cm'
for l-O,2-~-diIauroyl
ethanolamine
l-o,2-~-dimyristoyl
(OH)=3300 cm',
values within f 0.01
481.4
ethanolamine
(CH)=2800-2900cm-',
ppm of those found
(CO-0)=1737 cm-',
for l-O,2-~-dilauroyl
(CO-N)=1640-1552cm'
ethanolamine
l-o,2-~-dipalmitoyl
(CH)=2800-2900cm',
values within f 0.01
537.5
ethanolamine
(C0-0)=1736cm',
ppm of those found
(CO-N)=1640-1551cm'
for 1-0,2-~-dilauroyl
ethanolamine
In the 'H-NMR spectra the values of the chemical shifts are referred to the residual signal of
the sovent; 6:7.2 H (CDCI,), 2.0 H (acetone-d,) or 3.30 H (methanol-d,). Abreviations: s=singlet;
b.s=broad singlet; d=doublet; t=triplet and q=quartet.

304 0 c. Otero

n)

Tiiiie (hours)

Time (hours)

Fig. 11.1. Effect of the molar ratio of lauric acid to ethanolamine. Conditions:4PC, 0.5 ml of acetonitrile
and 25 mgof Novozym435.A).Excessofacid:6.1mg (0.1 mmo1)ethanolamine.B) Excessofethanolamine:
20 mg (0.1 mmol) lauric acid. A/E = amide-ester. Results expressed as percentage molar yield with
respect to the limiting reagent. Experimental values correspond to the mean of at least 2 independent
experiments. Experimental errors were between 0.4 and 4%.

alkanolamine only when an excess of ethanolamine is employed. However, excess

acylating agent lowers the reaction selectivity to the desired intermediate N-acylated
ethanolamine and considerable amounts of the diacylated product are also formed.
However, in this case, the reaction is relatively slow and the rate decreases as the
percentage excess of ethanolamine increases.
A complementary study under comparable operating conditions (Fig. 1 1.2)
indicated that this result is not a consequence of deactivation of the enzyme by
ethanolamine. However, the reaction is inhibited by both reactive and nonreactive
amines that may be present in the reaction medium. Figure 11.2 depicts the inhibitory
effects of different amounts of triethanolamine.
Consequently, the results of these preliminary trials indicated that use of an
equimolar ratio of reagents is preferable for selective preparation of the N-acylated
alkanolamine. For a molar ratio of one: (i) the selectivity towards the N-acylated
alkanolamine is very high, (ii) recycle of substrates present in excess is not necessary
if semiquantitative yields are obtained, and (iii) the reaction is faster under these
conditions than when an excess of ethanolamine is employed.

Efect of the Solvent

The author studied two routes to synthesize the desired products: first, direct acylation
and second, transacylation (Fig. 11.3). A variety of solvents were investigated in each

Selective Enzymatic Synthesisof N-Acylated Alkanolamine Emulsifiers 0 305

Triethylamine (mmol)
Fig. 11.2. Inhibition by non reactive triethylamine. Conditions: 6.1 mg (0.1 mmol) ethanolamine, 20 mg
(0.1 mmol) lauric acid, 0.5 ml acetonitrile, 40% and 0.1-0.5 mmol triethylamine.

I SYNTHETIC ROUTES I
I

\r.
DIRECT ACYLATION
ETHANOLAMINE
+ FATTY ACID

TRANSACYLATION
ETHANOLAMINE
ETHYL-ESTER

F=-=?
Cprecipitat4
Ion-pair complex

Free precursor reagents

Fig. 11.3. Scheme of reaction systems corresponding to direct acylation of ethanolamine with a free
fatty acid and to the transacylation of the amide with the correspondingfatty acid ethyl ester.

306 0 C. Otero
case. The two consecutive acylation steps generate products of different polarities.
Consequently, the solvent could determine the product distribution in the final
mixture. Three solvents of different polarities [i.e., dioxane (Log P = -1.1), acetonitrile
(Log P = -0.33), and n-hexane (Log P = 3.5)] were compared. Log P i s the logarithm
of the partition coefficient of the indicated solvent in the octanol-water two-phase
system.
For this particular biotransformation the two synthetic routes differ not only with
respect to the type of the precursor reagent employed, but also with respect to the
nature of the intermediates formed from the original reagents. For direct acylation
in dioxane, acetonitrile, and hexane, rapid precipitation of a significant fraction of
the precursor reagents was observed when the liquid ethanolamine was added to
the solution of the fatty acid. Figure 11.4 depicts the IR spectra of lauric acid and
the corresponding ion pair formed with the amine. Interpretation of the infrared
spectrum of the mixture of the two precursor reagents indicates the formation of an
ionic complex between these reactants. The carboxylic acid band that appears in the
spectrum of lauric acid at 1700-1715 cm-' has disappeared from the spectrum of the
mixture, whereas a band corresponding to a carboxylate ion appeared at 1558-1560
cm-I.
In some solvents the solubility of the ion pair complex of reagents formed
when the reaction proceeds via the direct acylation route and the solubility of free
ethanolamine present when transacylation is the route of choice are very small (Table
11.3). The solubilities of the precursor reagents determined in this work permit one
to identify whether or not partial solubilization of the precursor reagents affects the
reaction rates associated with these two alternative routes to synthesize the desired
products. In polar solvents, such as acetonitrile and dioxane, free ethanolamine is 1-2
orders of magnitude more soluble than the corresponding ion complex of reagents
(Table 1 1.3, see scheme depicted in Fig. 11.3).
The selectivity of the reaction and the time required to achieve the maximum yield
of the desired intermediate product depend on both the solvent and the temperature
employed. More precisely, the selectivities of these reactions depend very much on
the solubility of the desired product (Table 11.4). The solubility of the N-acylated
alkanolamine is 1-2 orders of magnitude lower in n-hexane than in acetonitrile and
dioxane (Table 11.4). Hence, at 40C in n-hexane continuous precipitation of this
intermediate product prevents subsequent acylation to the ester of alkylamine (Fig.
11.5).
Although the selectivity of the reaction is best in n-hexane, use of this solvent
suffers from two disadvantages, namely: the reaction rate is slow because of the low
solubility of the reagents, and elimination of hexane from the final product is more
difficult.
To increase the selectivity of reaction in polar solvents, one needs to achieve
maximum precipitation of the intermediate N-acylated alkanolamine. In polar
solvents the selectivity for formation of the N-acylated alkanolamine can be increased

Selective Enzymatic Synthesisof N-Acylated Alkanolamine Emulsifiers 0 307

Wavelength (ern-')

Absorbance

Wavelength (ern-')
Fig. 11.4. infrared spectra of (A) free lauric acid and (6) the ionic complex formed by ethanolamine and
lauric acid at 40C.

308 0 c. Otero

Table 11.3. Solubility of Ethanolamine and the Ion-pair Formed by this Reagent with
Lauric Acid in Different Solvents at 40C
Product
Ethanolamine
Ethanolamine

Solvent
Dioxane
Acetonitrile

Ethanolamine
Ion complex
ion complex
Ion complex

n- Hexane
Dioxane
Acetonitrile
n-Hexane

mg/mL
>250
>250

Solubility
mol/L
4.20
>4.20

1.25

0.02
0.205
0.025
0.01 8

53
6.5
4.7

ea

0
2

Time (houls)
Fig. 11.5. Selective mono-acylation of ethanolamine in n-hexane at 400C. Conditions: 25 mg Novozym
435, 0.5 ml n-hexane, 0.1 mmol ethanolamine, equimolar ratio of reagents. Results expressed as
percentage molar yield.

Selective Enzymatic Synthesisof N-Acylated Alkanolamine Emulsifiers 0 309

by carrying out the reaction at relatively low temperatures (namely 2O-3O0C, Fig.
11A). However, this method markedly increases the time required to achieve high
levels of reaction. Hence, alternative reaction conditions were further explored in
efforts to increase the selectivity to the intermediate N-acylated alkanolamine in polar
solvents.

ComparativeS d y of DirectAcylationAnd TransacylationReactions at Relatively

High Loadr of Substrates and 40C
A primary objective of engineering reactions requiring the use of solvents is to
determine conditions under which maximum concentrations of reagents can be
employed. One wishes to do this without significantly decreasing either the yield or
the selectivity to the desired product. Use of a minimum amount of solvent not only
minimizes its potential toxicological impact on the final product, but also reduces the
reactor volume and the associated capital costs.
Use of high substrate loads was also explored using equimolar concentrations of
the precursor reactants. The amount of biocatalyst was increased in proportion to the
amount of reagents employed (e.g., 25 mg biocatalyst for 0.2 mol/L of substrate). As
the concentrations of reactants increased, so too did the percentage of the intermediate
products which precipitated. This fact should permit one to increase the reaction
selectivity in polar media.
For the case of direct acylation, according to the solubilities of the precursor
reagents in acetonitrile at 40C (Table 11.3), systems containing charges of substrates
above 0.4 mol/L are characterized by slower rates and lower conversions (Table 11.5).
By contrast, the corresponding limit on substrate loading for transacylation is much
larger at 2 mol/L (Table 11.5).Via transacylation in acetonitrile where the solubility of
ethanolamine is sufficiently high (Table 11.3), the selectivity increases as the substrate
load increases. Completely selective synthesis of the N-acylated alkanolamine was
obtained for reagent loadings above 0.6 mol/L. Consequently, via transacylation the
concentrations of reagents in acetonitrile can be increased by one order of magnitude
(up to 2 rnol/L) without having a significant effect on the reaction kinetics and
selectivity.
However, because the ethanol produced in the transesterification route decreases
the stability of the enzyme, the author tried to improve this strategy by using free fatty
acids which are cheaper and more readily available than their corresponding alkyl
esters.

DirectAcykztionand TransacylationReactions at Relatively Hi& Substrate Load

and 60C
Once the factors which limit the conversion and selectivity of the reaction were
identified, the author utilized the fundamental principles of reaction and enzyme
engineering to overcome these limitations at high substrate loads.
The comparatively high solubility of the reagents at the increased temperature (at
least up to 5 mollL at 60C) permits one to increase the loading of reagents up to 3

3 10 c. Otero
Table 11.4. Solubility of N-acylethanolamines in Different Solvents at 30,40, and 60C
Solubility
40C

Product
N-Caproylethanolamine
N- Lauroylethanolamine

Solvent
Acetonitrile
Dioxane

N- Lauroylethanolamine
Acetonitrile"
N-Lauroylethanolamine
n-Hexane
N-Mvristovlethanolamine
Acetonitrile
N-palmitoy let hanolamine
Acetonitrile
LI Solubility a t 3OoC = 9.6 mg/mL (0.039 mol/L).

60C
mg/mL mol/L

mg/mL
77
>48

mol/L
0.356

>0.200 >82

>0.340

23
1.o

0.095
0.004
0.030
0.011

1.5

>0.330
0.006

8.0
3.5

>80

Tm (hours)

Tm (hours)
Fig. 11.6. Effect of the temperature in direct acylation ofethanolaminewith lauric acid in a polar solvent.
Conditions: 25 mg Novozym 435,6.1 mg (0.1 mmol) ethanolamine, equimolar ratio of reagents. Results
30C(A),
40C (A)~60C
(O).The y-axis
expressed as percentagemolar yield. 5C (*), 10C(W), 20C (O),
labels areexplained in Fig. 11.1.

Table 11.5. Comparative study of the two synthetic routes using different charges of substrates in equimolar concentrations of
reactants (ethanolamine and lauric acid acyl groups) and lipase at 40C"

Reaction Route
Direct Acylation

Transacylation

Solvent
Acetonitrile

Acetonitrile

ReactionTime
(hr)
24

Total
Conversion
(mole %)
95

0.2
0.4

Precursor Dissolvedb Amide Dissolved'

(mole%)
(mole %)
20
12
50
6
24

96

93

0.6
0.2
0.6
1.o

4
100
100
100

15
1.5
1.5
1.5

50
97

Substrate
(mol/L)
O.Zd

16
50
16
9

96
94

Amide-Ester
(mole %)
0
6
4
4
4
0
0

2.0
100
5
1.5
92
0
The amount of C. anrafcrica lipase was increased in proportionto the amount of reagents employed (e.g., 25 mg biocatalyst for 0.2 mol/L of
substrate).
bMolepercent of ethanolamine or the corresponding ion-pair.
Mole percent of N-acyl ethanolamine dissolved in the reaction medium when the ethanolamine is completely transformed to this product.
Reactionat 30C.

wl

a.

m
rn
2
.c
N

E.3
Y
In

2.
T
R
2
0
h
)

0
0)

3.
3

3 12 0 C.Otero
mol/L, without having any significant decrease in the reaction conversion. But, the
reaction conversion dramatically diminishes when the presence of solids (the reagents
and N-acylated ethanolamine) becomes excessively large at reagent loads higher than
3 mol/L.
In reactions with reagent loads above 0.4 mol/L, the percentage of the desired
product (N-acylated ethanolamine) precipitated in the reaction medium, is sufficiently
high for obtaining maximum selective formation of this intermediate product
(92-95%, Table 11.6). Note that the selectivity towards the intermediate product
(N-acylated ethanolamine) remains constant independent of the load of reagents
utilized in the range of loadings studied (0.4-3 mol/L).
Use of a higher temperature (60C) permitted the author to increase the
concentrations of reagents from 0.4 mol/L to 3 mol/L while minimizing problems
associated with the solubilities of reagents and intermediates, thereby minimizing the
limitations on the rate (Table 11.6). Consequently, the increase of temperature from
40 - 60C permits one to increase the volumetric productivity of these reaction routes
by one order of magnitude.

N-Acylated Diethanolamine Biosurfactants

Diethanolamine is a precursor which contains several groups susceptible to acylation,
namely two hydroxyl and one amine group. For substrates containing several moieties
susceptible to acylation, kinetic strategies designed to produce selective acylation could
be developed using lipases to effect these reactions under mild conditions. However,
under mild conditions the relatively high viscosity and low solubility ofdiethanolamine
could limit the extent of the biotransformation. This report describes strategies that
Table 11.6. Direct Acylation of ethanolamine at high substrate loadings in equimolar
concentrations of reactants at 60C
Immobilized Biocatalyst
Novozym 43Sb

Substrate
(mol/L)

Biocatalyst
(mg)

250

92

91

2.5
3

312.5
375
500
625
71.5
89.5
143

94
94

98
93
83
78
96
97
95

C.unturcticu 6-Silica 600

5
2
2.5
4.0

Selectivity Total Conversion

(96)
(mole %)

95
92
94
95
95

Selectivity to the amide.

Cundidu unturctica B lipase immobilized on a macroporousacrylic resin (Novo Nordisk A/S,
Bagsvaerd, Denmark).
C. unturcticu B lipase immobilized on a chemically modified silica with octyltriethoxysilane
designated as Silica 600.

Selective Enzymatic Synthesis of N-Acylated Alkanolamine Emulsifiers0 3 13

the author has developed for selective synthesis of N-acylated diethanolamines in two
different media. These strategies also consider and/or minimize problems associated
with the high viscosity and low solubility of the precursor diethanolamine.

Experimental
Materaah
Diethanolamine (> 98% pure), lauric acid (> 99% pure), myristic acid (98% pure),
and dioxane were purchased from Merck (Darmstadt, Germany). Other reagents
were the same as those used in the ethanolamine study. Before use, all solvents were
dried with molecular sieves with an effective pore diameter of 4A.

Enzymatic Reactions
For trials involving a 1/1 mole ratio of reactants, 0.1 mmol (0.2 M) of both the
diethanolamine (10.5 mg) and the fatty acid (20 rng of lauric acid, 22.8 mg of
rnyristic acid, or 25.6 mg of palmitic acid) were mixed with 0.5 mL of solvent in
stoppered glass bottles. Additional experiments were conducted using 2/1 or 3/1 mole
ratios of lauric acid (40 mg or 60 mg, respectively) to diethanolamine (10.5 mg). In
experiments in which the mole ratio of lauric acid to diethanolamine was 1/2 or 1/3,
the quantity of diethanolamine that was employed was increased to 21 mg or 31.5
mg, respectively. For reactions involving solutions which were intended to be 0.5,
0.8, 1.2, 1.5, or 2 M for both reagents, the amounts of lauric acid added were 50,
80, 120, 150, and 200 mg while the corresponding amounts of diethanolamine were
26.2, 42, 63, 78.7, and 100.5 mg. For the transacylation reactions, the quantities of
lauric acid ethyl ester used to produce 0.2, 0.5, 0.8, and 1.2 M solutions were 22.8,
57, 91.2, and 136.8 mg, respectively. The resulting solution was equilibrated at the
reaction temperature for a few minutes prior to addition of the enzyme. Subsequently
2 5 ,6 2 5 , 100, 150, 187.5, and 250 mg of Novozym 435 were added to the solutions
of the acidic precursor (corresponding to concentrations of 0.2,0.5,0.8, 1.2, 1.5, and
2 M, respectively). Then the mixture was shaken at 200 rpm for 5 min to 96 hr at the
temperature of interest.
At predetermined reaction times, dimethylformamide (DMF) was added until
precipitated reagents and/or products were dissolved and a transparent solution was
obtained. The immobilized enzyme was then rapidly removed by filtration with a 0.1
mm sieve, and the filtrate was centrifuged. DMF increased the volume of final solution
to 10 mL (when the reactant concentration was 0.2 M). For reactant concentrations
of 0.5, 0.8, 1.2, and 2 M, DMF increased the final volume to 25, 50, 50, and 100
mL, respectively. An HPLC system consisting of an autosampler L-7200, a L-7100
pump, connected to a Kromasil C18 250 x 4.6 mm column (Eka Chemicals AB,
Bohus, Sweden) and a SEDEX 55 light-scattering detector (Sedere Sa., Alfortville,
France) was used to analyze the product mixture. Molar yields were calculated with
respect to the limiting reactant.

Results and Discussion

Elucidation of the Structures of the Reaction Products
FTIR, NMR, and molecular mass data for the reaction products are given ..I Table 1 7.

Table 11.7. Structural Elucidation of N-acylated Diethanolamine Biosurfactants.

IR Spectra
H-NMR
Molecular
Masses
287.4
N-Lauroyl diethanolamine (OH)= 3300 cm, (CH)= 6: 3.69 (m, HH, CH,O),
2800-2900 cm- and
3.50 (m, HH, CH2N),2.43
(CO-N)= 1625 cm-
(t, 2H, J=7.4, CH,CO),
1.60 (m, 2H, CH2-CH2CO), 1.40 (m, 16H, CH,chain), 0.89 (t, 3H, J=7.0,
CH.).
N-myristoyl
(OH)= 3300 cm-, (CH)= values within f
315.5
0.01 ppm of those
diethanolamine
2800-2900 c m and
(CO-N)= 1620 cm
found for N-lauroyl
diethanolamine
N-palmitoyl
(OH)= 3300 cm-l, (CH)= values within f
343.5
diethanolamine
2850-2900 cm and
0.01 ppm of those
(CO-N)= 1624 cm-
found for N-lauroyl
diethanolamine
6: 4.23 (m, 2H, CH,469.7
(OH)= 3300 cm,
1-0,2-~-dilauroylZN(2-hydroxyethy1)(CH)= 2850-2930 cm,
0-CO), 3.68 (m, HH, 2
ethanolamine
(CO-0)= 1743 Em- and x CH,N), 3.49 (m, 2H,
(CO-N)= 1628 cm
CH,-OH), 2.42 (m, 2H,
C!j,-COO), 2.30 (m, 2H,
CH2-CO-N),1S9 (m, HH,
2 x CH,-CH,-CO), 1.29
(m, 32 H, 2 x CH,- chain),
0.89 (t, 6H, J=6.8 Hz, 2
x CH,)
1-0,2-~-dimyristoyl(OH)= 3380 cm,
values within f 0.01
525.9
ZN(2-hydroxyethyI)(CH)= 2850-2930 cm-,
ppm of those found
(CO-o)= 1730 c m and for l-O,2-~-dilauroylethanolamine
(CO-N)= 1630 cm-
ZN(2-hydroxyethy1)ethanolamine
1-0,2-~-dipalmitoyl(OH)= 3300 cm,
values within f 0.01
581.9
ZN(2-hydroxyethyl)(CH)= 2850-2930 cm,
ppm of those found
(COO)= 1730 cm and for l-O,2-~-dilauroylethanolamine
(CO-N)= 1609 cm-
ZN(2-hydroxyethy1)ethanolamine
For the H-NMR spectra, the values of the chemical shifts are referred to the residual signal of
the solvent; 6: 3.30 H (methanol-dd. Abbreviations: $=singlet;d=doublet and t=triplet.

Selective Enzymatic Synthesis of N-Acylated Alkanolamine Emulsifiers0 315

The corresponding chemical shifts for 1-0,2-N-dimyristoyl2N (2- hydroxyethy1)ethanolamine and 1-0, 2-N-dipalmitoyl 2N (2- hydroxyethy1)- ethanolamine were
within f 0.0 1 ppm of those obtained for 1-0, 2-N-dilauroyl-2N(2-hydroxyethyl)ethanolamine.
The molecular masses ofN-lauzoyl diethanolamine, N-myristoyl, and N-palmitoyl
diethanolamine were 287.4, 315.5, and 343.5, respectively. Values for 1-0,2-Ndilauroyl-diethanolamine, 1-0,2-N-dimyristoyl diethanolamine, and 1-0,2-Ndipalmitoyl diiethanolamine were 469.7, 525.9, and 581.9, respectively.

Ee'ct
of the Solvent a n d Temperature
In some cases, enantio-, regio-, prochiral, chemo-, and substrate selectivities can be
governed by the solvent (Shinichirou & Klibanov, 1993). Moreover, the extent of
ionic dissociation of the precursor reagents can vary from one solvent to another.
Hence, the reaction was studied in dioxane and n-hexane.
In both solvents, infrared spectra of the mixture of the two precursor reagents
indicated the formation of an ionic complex between the two reactants. When
diethanolamine was added to solutions of the fatty acid, the carboxylic acid band
at 1700-1715 cm-' disappeared and a band corresponding to a carboxylate ion
appeared at 1558-1562 cm-'. The ion-pair complex formed is soluble in dioxane,
but is not completely soluble in rt-hexane. The reaction mixture was a transparent
solution in dioxane, whereas a milk-like emulsion of two liquid phases was observed
in n-hexane.
For an equimolar ratio of substrates, formation of the ester intermediate is favored
at short reaction times in dioxane but this species is not formed to any appreciable
extent in n-hexane. This intermediate product was converted to the amide and/or the
amide-ester, via either spontaneous O+N acyl migration or a subsequent enzymatic
acylation reaction, respectively (Fig. 11.7).
In dioxane, the initial reaction rate (V, = VN+ V, ; where V, is the initial rate
of 0-acylation, VN is the initial rate of N-acylation, and V, is the total initial rate
of acylation) increases as the temperature increases, but the rate is always slower in
n-hexane than in dioxane (Fernindez-P6ra & Otero, 2003). These results can be
attributed to the low solubility of the ion-pair formed from the reactants in n-hexane
and to the fact that 0-acylation is a very fast reaction that takes place in dioxane but
not in n-hexane.
In n-hexane at 30C the rate is very low, but at 60C one may obtain significantly
higher conversions (69 mole %) and yields of the N-acylated alkanolamine (59 mole
Yo)(Ferninda-Pdrez & Otero, 2003). This result is associated with the high viscosity
of the solution of the reactants in n-hexane at 30C (5.52 0.13 cSt in n-hexane
and 1.30 f 0.01 cSt in dioxane). Limitations associated with the viscosity of the
medium may be avoided by increasing the temperature to 60C. At this temperature,
the observed conversions (69-74%) and the viscosities of the solutions (0.87-0.90
* 0.01 cSt) are similar in both solvents. At 60"C, the low solubility of the ion-pair

3 16

C. Otero

A) Dioxane

B) Hexane

Fig. 11.7. Reaction courses of direct acylation of diethanolamine with lauric acid at 4PC in polar (A) and
apolar (B) solvents. Conditions: 25 rng Novozyrn 435,O.l mmol of both diethanolamine and lauric acid
and 0.5 ml solvent.

formed from the reactants does not seem to be a limiting factor in n-hexane, and
the direct acylation reaction in n-hexane remains more selective to the N-acylated
alkanolamine intermediate than is the case in dioxane (86 and 75 % selectivity to the
amide intermediate at 60C, respectively) (Ferndndez-Pkrez & Otero, 2003).

Eflect of the Chain Length of the Fany Acid

In these two solvents, the rate or the selectivity of the reaction does not vary with
the chain length of the fatty acid. This behavior reflects the high/complete solubility
of the fatty acids of interest in the two solvents at 60C (consequently, the chain
length of the fatty acid does not affect the solubility of the corresponding ion-pairs in
n-hexane), and the similar values of the viscosities of the solutions of different fatty
acids in both dioxane and n-hexane (0.87-0.91 cSt for solutions of lauric, myristic,
and palmitic acids in both solvents).
Eflect of the Amount of Biocatalyst
The influence of the amount of enzyme on the rate of the reaction between lauric
acid and diethanolamine at 60C was studied in hexane. The time required to obtain
approximately 50% total conversion varies with the amount of enzyme added (Fig.

Selective Enzymatic Synthesis of N-Acylated Alkanolamine Emulsifiers0 3 17

11.8). The rate decreases when amounts of enzyme greater than 50 mg are employed.
A similar effect was reported by the author's group for biotransformations of
heterogeneous mixtures of sugar esters and a-hydroxy acids. This effect is explained
by the failure of the shaker to maintain uniform suspensions of the solids (ion pair of
reagents + biocatalyst) in the reaction mixtures with high reagent loads (Arcos et al.,
1998a, 1998b; Torres & Otero, 1999; Torres et al., 1999).

Direct Acyhtion at High Substrate Load

The effect of the substrate load on both the reaction kinetics and selectivity was
performed with equimolar concentrations of reactants of 0.2, 0.5, 0.8, 1.2, 1.5, and
2 M and the corresponding amounts of biocatalyst of 2562.5, 100, 150, 187.5, and
250 mg, 0.5 mL solvent (acetonitrile or n-hexane) and 60C.
The results summarized in Table 11.8 indicate that the potential for increased
volumetric productivity is limited to that corresponding to reactant concentrations
of 0.8 M. Both the conversion and the selectivity to the N-acylated alkanolamine can
be maintained at 60C when the reactants concentrations increase from 0.2 - 0.8 M.
Reactant concentrations above 0.8 M lead to decreases in conversion from 73 - 38%
and from 73 - 46% in n-hexane and dioxane, respectively. This limitation can be
attributed to the presence of an excessive amount of solids (biocatalyst) in the reaction
mixture in a fixed volume of solvent and/or problems associated with the increased

25

60
n

20

40

I?

15

.-0
G2
3

10

20

.E
b

12.5

25

50

75

100

150

200

Loading of Biocatalyst (mg)

Fig. 11.8. Effect of the amount of the biocatalyst on direct acylation of diethanolamine in n-hexane.
Conditions:0.1 mmol of both diethanolamine and lauric acid, 0.5 ml n-hexane, 60C.

3 18 c. ~ t e r o
viscosity of the reaction mixture (Table 11.8). In the case of dioxane these results
can not be attributed to solubility problems, because neither precipitation nor phase
separation of the precursor reagents was observed in the more concentrated dioxane
solutions.

Eflect of the Type of Solvent on the Transacylationof Diethanolamine with a Fatty

Acid Ethyl Ester
The transacylation of diethanolamine with a fatty acid ethyl ester is an alternative
synthetic route of N-acylated alkanolamine emulsifiers. Transacylation does not
involve formation of an ion-pair complex from the reactants and some kinetic
consequences should be expected.
At 60C transacylation is faster in dioxane (9.6 x
mmol/min) than in
n-hexane (1.0 x
mmol/min). These results can be explained by noting that only
4% (w/w) of the liquid diethanolamine precursor is soluble in n-hexane at 60C
At this temperature, the solubility of diethanolamine in n-hexane is 0.8 mg/mL.
However, diethanolamine is completely soluble in dioxane. The ethyl esters of fatty
acids are highly soluble in these solvents.
Transacylation is more selective for formation of the intermediate amide in
n-hexane than in dioxane (Table 11.9). In n-hexane, transacylation is more selective
towards formation of the N-acylated alkanolamine than direct acylation (Tables 11.8
and 11-9). In both polar and apolar media, the transacylation reaction produces
higher conversions than does direct acylation (Tables 11.8 and 11.9).
Table 11.8. Conversion, Selectivity, and Viscosity of Reaction System of Direct Acylation
of Diethanolamine with Lauric Acid in Equimolar Concentrations at 60C with Higher
Concentrations of Substrates in Both, Polar, and Apolar Media
Solvent

C. antarctica Amide
Conversion Selectivitp Viscosity
Reactant
Concentration (M) Lipase
(mole Yo) (mole %)
(%)
(CSt)
0.2
0.5
0.8
1.2
1 .5

25
62.5
100
150
187.5

56
71
67
51
45

74
73
73
55
50

76
97
92
93
90

2
0.2
0.5
n-Hexane 0.8
1.2
1.5

250
25
62.5
100
150
187.5

41
59
64
69
35
36

46
69
68
73
39
39

89
86
94
94
90
92

Dioxane

Reaction time: 24 hours.

Conversion of the limiting reactant.
Selectivity to the amide.

(I

0.87 (* 0.01)

1.38 (* 0.01)
1.87(* 0.01)

0.90 (* 0.01)

1.40 (k 0.08)
2.00 (* 0.011

Selective Enzymatic Synthesis of N-AcylatedAlkanolamine Emulsifiers 0 3 19

Table 11.9. Conversion, Selectivity, and Viscosity of Reaction System of Transacylation

of Diethanolamine with Lauric Acid Ethyl Ester in Equimolar Concentrationsat 60C with
Higher Concentrations of Substrates in a Polar and an Apolar Solvent
Reactant
C. antarctica
Concentration (MI Lipase
0.2
25
0.5
62.5
Dioxane
0.8
100
1.2
150
0.2
25
0.5
62.5
n-Hexane
0.8
100
1.2
150
Reaction time: 24 hours.
Conversion of the limiting reactant.
bSelectivityto the amide.

Ivent

Amide
(mole%)
57
55
62
60
67
67
69
51

Conversion
(mole%)
77
75
71
65
71
71
73
53

Selectivityb Viscosity
(CSt)
74
0.87 (k 0.01)
73
87
0.87 kt 0.011
92
0.92 (* 0.01)
94
94
94
96
(%)

Transacylationat Higb Substrate Loads

Table 1 1.9 summarizes the results obtained for the reaction at equimolar concentration
of reagents and 60C with increased substrate loads (0.2-1.2 M). The transacylation
route permits one to increase the reactant concentrations from 0.2 - 0.8 M without
incurring a significant decrease in either conversion or selectivity. The adverse effects
noted for the most concentrated systems should be attributed to the presence of an
excessive amount of solids (biocatalyst), because in dioxane at 60C the viscosities
of the more concentrated solutions (1.2 M) are not prohibitive (Table 1 1.9) and the
concentration of diethanolamine is below its solubility limit.
Complete conversions of precursors are obtained with the primary amine (Fig.
11.1, 11.5, and 11.6), but not with the secondary amine (Tables 11.8 and 11.9).
These results are attributed to the fact that the products of the ethanolamine reaction
precipitate in the reaction medium, but precipitation of the desired product does not
occur in the case of diethanolamine. Continuous extraction of the products from the
liquid reaction phase permits one to shift the thermodynamic equilibrium position
of the ethanolamine reaction. Moreover, the high viscosity of the diethanolamine
precursor leads to high viscosities of the corresponding reaction mixtures. This is not
the case for the reactions of ethanolamine.
Etbanolamine and Diethanolamine Reactions Usinga Tailor-madeBiocatalystfor
Increaing %lumetric Productivity
It is always desirable to achieve maximum volumetric productivity, but the concomitant
increase in the amounts of solid materials present in the reaction medium is a limiting
factor of these bioprocesses (Tables 11.8-1 1.10). Consequently, it would be desirable
to work at high volumetric productivities using a more active biocatalyst (higher

320 0 c. Otero

Table 11.10.Transacylation of Dethanolaminewith Ethyl Laurate at High Substrate Loads

in Equimolar Concentrations of Reactants at 6O0CC"

Biocatalyst Enzymeb
Amide
Biocatalyst
(mg)
(Activity Units) (mole %)
Novozym 435 150
395.1
60
Silica 600
32.2
394.6
75
aConditions:[Substrates]= 1.2 M; 0.5 mL dioxane; 24 hr.
bActivity units on p-nitrophenyl butyrate hydrolysis.

Amide Ester Total Conversion

(mole 96)
(mole 96)
5

65

81

ratio of enzyme to support). One must also consider that the viscosity of the reagent
mixtures can increase significantly when increased substrate loads are employed.
The author has prepared different tailor-made immobilized lipases according to
the requirements of the reaction. Namely, the author employed an immobilization
procedure using a support with a high porosity, an appropriate mean pore size to get
a high specific surface area (surface area 300-320 m2/g, average pore diameter 30-40
nm), a chemically modified matrix surface to facilitate adsorption of the enzyme in
an activated form that is accessible to the substrate, and a mesoporous silica modified
with octyltriethoxysilane (Blanco et al., 2004). The author selected a lipase with
high specificity for the substrates. More precisely, Candida antarctica B lipase was
immobilized on the aforementioned relatively hydrophobic silica.
The hydrolytic activities of both commercial (Novozym 435) and tailor-made
(Silica 600) immobilized derivatives were 2634 and 9200 unidg, respectively. The
activity units of these two solid biocatalysts employed in biotransformations of
ethanolamine were similar, namely 197.5 and 197.9 activity units, respectively. The
amount of grams of solid biocatalyst added to the reaction mixture was lower in the
case of the tailor-made catalyst than in the case of the commercial one (Table 11.6).
The new biocatalyst gives results comparable to those of the commercial biocatalyst
at an intermediate substrate load (0.6 mol/L) and temperature (50C). The new
biocatalyst also gives excellent reaction conversions and selectivities to the N-acylated
alkanolamine at the highest substrate load studied (namely 2.5-4 mol/L; Table 11.6).
These values are much better than those of the best commercial catalyst.
The results of the comparative study of transacylation of diethanolamine with
ethyl laurate catalyzed by Novozym 435 and the new tailor-made biocatalysts at
60C and 1.2 M reagents concentration are summarized in Table 11.10. In this
study, the amount of activity units of the immobilized catalysts were approximately
the same, but the reaction mediated by the most efficient lipase contained a lower
amount of the solid enzyme preparation (Table 11.10). The conversion obtained at
24 hr with new biocatalyst (81%) is significant higher than that obtained with the
commercial immobilized lipase (65%). These results demonstrate that use of a more
efficient biocatalyst permits the reduction of the amount of solids in reaction systems
with high substrate loads. Consequently, higher conversions are obtained with the
more efficient biocatalyst when high substrate loads are employed. This study also

Selective Enzymatic Synthesis of N-Acylated Alkanolamine Emulsifiers0 32 1

demonstrates that the presence of high amount of solids in the medium limits the
reaction conversion that can be achieved. The more efficient biocatalyst permits to
work at much higher reagent concentrations. Under these conditions the reaction
selectivity can be increased (from GO% to 75% amide, Table 11.10).

Concl usions
A general strategy has been developed for optimization of biotransformations at
high substrate concentrations, especially, for reactions that involve species with low
solubilities and relatively high viscosity of the reaction medium. Optimal reaction
conditions for the synthesis of N-acylated ethanolamine and diethanolamine
emulsifiers have been identified. The optimum corresponds to maximum conversion
of substrate and maximum selectivity to the desired product. Factors which limit the
conversion and the selectivity of this reaction have been identified. These limitations
have been overcome at high substrate loads. The new immobilization procedure
permits one to increase the volumetric productivity of ethanolamine reaction to 4
moles/L. The high catalytic efficiency of the new biocatalyst permits a reduction in
the amount of solids present in the reaction system, and increases the selectivity and
reaction conversion of the transacylation of diethanolamine. Under these conditions
the volumetric productivity is also increased.

References
Arcos, J.A.; M. Bernabt; C. Otero. Quantitative Enzymatic Production of 6-O-Acylglucose Esters.
Biotechnol. Bioeng. 1998a,37, 505-509.
Arcos, J.A.; M. Bernabe; C. Otero. Quantitative Enzymatic Production of 1,6-Diacyl Sorbitol
Esters. Biotechnol. Bioeng. 1998b, GO, 53-60.
Blanco, RM.; I!Terreros; M. Ferndndez-Perez; C. Otero. Functionalization of Mesoporous Silica
for Lipase Immobilization. Characterization of the Support and the Catalysts./. Mol Catal.
2004,30, 83-93.
Ferninda-Perez, M; C. Otero. Enzymatic Synthesis of Amide Surfactants from Ethanolamine.
Enzyme Micro&. Zchnol. 2001,28, 527-536.
Ferndnda-Perez, M.; C. Otero. Selective Enzymatic Synthesis of h i d e Surfactants from Diethanolamine. Enzyme Micro&. Technol. 2003,33,650-660.
Furutani, T.; M. Furui; H. Ooshima; J. Kato. N-Acylation of P-Amino Alcohol by Acyl Migration
Following Enzyme Catalyzed Esterification. Enzyme Micro&. Technol. 1996, 19, 578-584.
Garrison, J. Monoethanolamides. Detergent Age 1968, 27-29.
Gotor, V.; R. Brieva; F. Rebolledo. EnantioselectiveAcylation of Amino Alcohols by Porcine Pancreatic Lipase. /. Chem. Soc. Chem. Commun. 1988, 957-58.
Khemelnitsky, Y.U.; A.V. Levashov; N.L. Klyachko; K. Martinek. Engineering Biocatalytic Systems in Organic Media With Low Water Content. EnzymeMicrob. Zchnol. 1988, 10,710-724.
Kanerva, L.T.; M. Kosonen; E. Mnttinen; T. Huuhtanen; M. Dahlqvist. Studies of the Chemoand Enantio- Selectivity of the Enzymatic Monoacylations of Amino Alcohols. Acta Chemica

322 0 c. Otero
Scand. 1992a, 4G, 1101-05.
Kanerva, L.T.; K. Rahiala; E. Mnttinen. Lipase Catalysis in the Optical Resolution of 2-Amino-lPhenylethanol Derivatives. J. Chem. SOC.Perkin Trans. 1992b, I, 1759-62.
Maag, H. Fatty Acid Derivatives: Important Surfactants for Household, Cosmetic and Industrial
Purposes. J. Am. Oil Chem. SOC.1984, GI, 259-267.
Matos, J.R.; J.B. West; C.H. Wong. Lipase Catalyzed Synthesis of Peptides. Preparation of a Penicillin G Precursor and Other Peptides. Biotechnol. Lett. 1987, 9, 233-236.
Maugard, T.; M. Remaud-Simeon; D. Petre; I? Monsan. Lipase-Catalyzed Chemoselective NAcylation of Amino-Sugar Derivatives in Hydrophobic Solvent: Acid-Amine Ion-Pair Effects.
Tetrahedron 1997,53, 7587-94.
Paiva, A.L.; V.M. Balcao; EX. Malcata. Kinetics and Mechanisms of Reactions Catalysed by Immobilized Lipases. Enzyme Microb. Technol. 2000,27, 187-204.
Shinichirou, T.; A.M. Klibanov. Chemoselectivity of Enzymes in Anhydrous Media is Strongly
Solvent Dependent. Biocatal 1993, 8, 3-19.
Torres, C.; M. Bernabt; C. Otero. Part 2. Two Enzymatic Procedures for the Synthesis of Malic
Acid Monoesters. Enzyme Microb. Techol. 1999B, 25, 753-761.
Torres, C.; C. Otero. Part 1. Enzymatic Synthesis of Lactate and Glycolate Esters of Fatty Alcohols. Enzyme Microb. Technol. 1999,25, 745-752.
Wescott, C.R; A.M. Klibanov. 'The Solvent Dependence of Enzyme Specificity. Biochim. BiophyJ.
Acta 1994, 1206, 1-9.
Zaks, A.; A.M. Klibanov. Enzymatic Catalysis in Organic Media at 1OOoC. SCI 1984,224,
1249- 125 1.

Zaks, A.; A.M. Klibanov. Enzyme Catalyzed Processes in Organic Solvents. Proc. Nut. Acad. Sci.
1985, 82, 3192-3196.

Synthesis of Saccharide
Fatty Acid Ester Biosurfactants
Catalyzed by Lipase
Sang-Hyun Pyo and Douglas G. Hayes
DepartmentofBiosystemsEngineeringand SoilScience,Universityof Tennessee,Knoxville, TN 37996-4531

Saccharide Fatty Acid Esters:

Properties and Applications
Due to factors relating to human health, environmental safety, and usage of petroleumbased raw material, interest for using natural resources to produce commercial
products has greatly increased during last decade. For example, saccharide fatty acid
esters (value-added products derived from natural feedstocks such as corn or other
plant oils and starch, cellulose, or other biobased polysaccharides) have recently been
employed in the foods, cosmetics, and pharmaceutical industry as biobased and
biocompatible surfactants or emulsifiers (Nakamura, 1997; Szuts et al., 2007). They
are fully biodegradable, odorless, flavorless, non-toxic, do not irritate skin, and can
be easily digested by the stomach, producing a sugar-fatty acid mixture (Allen & Tao,
1999; Pinna et al., 2004). They find widespread employment as water-in-oil (W/O)
emulsifiers in food products (Sabeder et al., 2006).
Selection of the proper emulsifier is very important in the manufacture of food
additives; the emulsifier must possess suitable functional properties to confer stability
against droplet coalescence during the shelf-life and biocompatibility (Partal et al.,
1999). Because of the wide rage of hydrophilic-lipophilic balance (HLB) values that
can be produced, sucrose fatty acid esters can be employed for many applications in
the food, cosmetic, and pharmaceutical industries. Examples of the latter include
liquid, plastic, semisolid and solid dosage forms as emulsifiers, solubilizing agents,
lubricants, penetrating enhancers, and pore-forming agents (Ntawukulilyayo et
al., 1995, 1996; Thevenin et al., 1996). The type of fatty acid and the degree of
esterification determine the HLB value and the melting point of these materials
(Cs6ka et al., 2007). In personal care and cosmetic products they can be found in
tooth paste, lotions, shampoos, and lipsticks. Recently, some of the fatty acid esters
have shown anti-tumor and antibiotic activity, and plant growth inhibition (Kohya et
al., 1986; Woods & Swinton 1991; Guillemard et al., 1993; Otomo, Chapter 10 in
this book).
323

324 0 5-H. Pyo and D.G. Hayes

Saccharide-fatty acid esters are amphiphilic and mainly non-ionic type surfactants
having the saccharide serve as polar head group and one or more fatty acyl group are
apolar tail moieties (Fig. 12.1; Sabeder et al., 2006; Cs6ka et al., 2007).
?he chemical properties of the saccharide and fatty acyl groups control the surface
energy related properties of the esters. Saccharides, such as glucose, sucrose, fructose
and xylose, contain more than 5 hydroxyl groups as possible reactive sites with acyl
donor. The reaction mainly occurs at the more reactive primary hydroxyl groups
among them (typically 1-2 primary hydroxyl per molecule) in accordance with lipases
regioselectivity.
Taking sucrose as an example, the enzymic approach appears to be functionally
distinct from chemical acylation in at least two respects (Rich et al., 1995). First,
chemical acylation with acid chlorides, acid anhydrides, acyl azides, acyl cyanides,
and N-acylimidazoles results in an acylation preference of 6-OH 2 6-OH > 1-OH >
secondary-OH. In contrast, enzymes nearly universally acylate with 1-OH = 6-OH
> secondary-OH >> 6-OH. Second, enzymes often differ in their regioselectivities.
For example, subtilisin Carlsberg (alkaline serine protease from B. lichenifuormis)
preferentially catalyzes acylation of sucrose at the 1-OH (with small yields of 6-OH)
in pyridine with short-chain ester donors, whereas lipase from Tbermomyces lanuginusus
(Novozymes Inc., Franklinton, NC USA) is selective toward the 6-OH (Rich et al.,
1995, Ferrer et al., 2005). Resultant products possess various HLB values with a wide
range according to the properties and number of saccharide and fatty acyl groups.
Sucrose-fatty acid esters have been produced commercially by Mitsubishi-Kagaku
Foods Corporation (Tokyo, Japan) and Sisterna B.V. (Roosendaal, Netherlands)
by chemical processing. Sucrose has eight free hydroxyl groups, which allow the
formation of mono- to octa-esters using different fatty acids (stearic, lauric, myristic,
and oleic acid) ( C d ka et al., 2007). The commercial sucrose fatty acid esters contain
mixtures of the previously mentioned fatty acyl groups and partial esters, to obtain an
HLB range from 1 to 16 HLB with the HLB decreasing as the chain length of fatty
acyl group and degree of esterification increase. (See Table 10.1 of Otomos Chapter
10, in this book.)

Fig. 12.1. Lipase-catalyzed synthesis of fructose-fatty acid esters (R = acyl group) Sabeder et al. (2006).
Reproduced with permission from Elsevier.

Synthesisof SaccharideFatty Acid Ester BiosurfactantsCatalyzed by Lipase 0 325

Motivation for Lipase-catalyzed Synthesis of

Saccharide Fatty Acid Esters
Although several different approaches have been employed for their production, most
saccharide fatty acid esters are produced by chemical methods resulting in costly and
environmentally unsafe conditions, such as high temperature, high pressure, alkaline
pH, and use of organic solvents (Feuge et al., 1970; Yan et al., 2001; Vulfson, 2003;
Hoydonckx et al., 2004). Additionally, chemical reactions can result in a broad
distribution of partial esters in the product, and the employment of unsaturated fatty
acid acyl donor reactants might produce additional by-products via degradation of
the double bonds that might be harmhl or toxic (Tarahomjoo & Alemzadeh, 2003;
Sabeder et al., 2006).
In contrast, the biocatalytic approach using immobilized lipase provides milder
and more environmentally-friendly operating conditions and leads to minimal
chemodegradation of double bonds and a very narrow product distribution of
typically 1 to 3 mono- and di-ester species due to the inherent regioselectivity of
enzyme (Sarney & Vulfson, 2001; Vulfson, 2003; Kennedy et al., 2006; Ballesteros
et al., 2007). However, there are several hurdles to overcome for the biocatalytic
approach, particularly the stability and reusability of the enzyme, and very poor
miscibility of acyl donor and acceptor substrates due to their contrasting lipophilic
and hydrophilic natures, respectively, leading to low reaction rates. Among this list,
the latter provides the biggest challenge. The main approach to improve miscibility
has been to use organic solvents and/or ionic liquids. An additional approach has
been to derivatize the mono- or di-saccharide substrate through covalent attachment
or complexation agents. Alternatively, multiphasic systems that utilize conditions
promoting precipitation of ester product upon its formation (i.e., solid-phase
synthesis) have been used successfully. In addition, the effect of water activity has
been investigated as a key parameter to control the degree of completion and enzyme
stability in lipase-catalyzed esterification. The approaches used to enhance miscibility
and the effect of water concentration are both discussed in detail below.

Synthesis of Saccharide Fatty Acid Ester

in Organic Solvents
The most desirable strategy is to carry out the reaction in solvent-free conditions.
Solvents have been commonly employed to enhance partial co-solubilization of the
acyl donor and acceptor substrate. The solvent ideally should be biocompatible, not
inactivate the biocatalyst, should be easily removable and recyclable, and chemically
inert toward the reaction of interest (Villeneuve, 2007).
The value of the solvent systems hg defined as the logarithm of its partition
coefficient benveen water and octanol, is widely used as a measure of solvent polarity;
it is also an important parameter for enzymatic synthesis in nonaqueous media and
for saccharide ester synthesis in particular (Tables 12.1 and 12.2; Laane et al., 1987;

326 0 5-H. Pyo and D.G. Hayes

Table 12.1. Activity of Lipase (Novozym SP435*, Novozymes, Inc, Franklinton, NC USA),
Glucose Solubility, and Enzyme Stability for the Esterification of Glucose and Myristic
Acid as a Function of Solvent Hydrophobicity
Enzyme
activity
Glucose
Solvent hydrophobicitf
Solvents
OmoVmin g) solubilityb (mM) Residual activity (%) (log P)
DMSO
0
29
0
-1.3
Dioxane
1.1
7.5
53
-1.1
DMF
0
12
0
-1 .o
-0.33
27
Acetonitrile
0
1.1
Acetone
3.O
2.6
46
0.23
THF
1.6
2.1
46
0.49
0
0.69
Pvridine
0
134
tert-8utanol
3.7
12
75
0.80
10
71
1.4
tert -Pentanol 3.6
0
0.6
54
2.5
Toluene
Hexane
0
0
80
3.5
Measured by reacting 150 mg glucose, 1 H,O and 750 mg myristic acid in 5 mL of solvent in
the presence of 35 mg lipase and molecular sieves (3A; 0.5 g) at 45C and rotary shaking (250
rpm) for 24 h.The reaction was carried out at 45C (250 rpm).
bDeterminedafter the initial 24 h of incubation at 45"C, before enzyme addition.
The enzymes were recoveredafter 24 h of reaction to determine their residual activity. After
decanting the solvent, the immobilized enzyme preparation and the molecular sieves were
washed 4 times with warm (45C) ten-butanol followed by drying in vacuo for 2 h. Additional
molecular sieve (500 mg) was added before reuse.
dFromLaane et al. (1987)
Degn & Zimmermann (2001). Reproduced with permissionJohn Wiley and Sons.

Yahya et al., 1998; Kim et al., 2004; Kennedy et al., 2006). The highest enzyme
activities are found when using nonpolar solvents (Degn & Zimmermann, 2001);
for example, solvents with log Pvalues >3 in Table 12.1 are widely used in reactions
such as the lipase-catalyzed interesterification of oils and fats to obtain new fats with
improved physical and/or nutritional properties (Villeneuve, 2007). However, to
synthesize fatty acid esters of saccharide and other polyols, acyl acceptor substrates are
poorly soluble in these solvents, which makes them unsuitable.
The stability of the enzyme is usually higher in the more hydrophobic solvents.
In contrast, solvents with lower log P values (e.g., pyridine, dimethylformamide
(DMF), tert-butanol, and acetone) can partially co-solubilize acyl donors and many
polar saccharide molecules (Table 12.2) and thus can be employed for polyol ester
synthesis (Villeneuve, 2007). However, these solvents often inactivate the enzyme by
their ability to remove water molecules of hydration, promote accumulation of water
in the reaction media, leading to hydrolysis and hence reduced product yield and/
or the formation of by-product and are incompatible with food applications (Degn
& Zimmermann, 2001; Ganske & Bornscheuer, 2005a; Hayes, 2004). For example,

Table 12.2. RegioselectiveSynthesisofFatty Acid Esters of Monosaccharidesand RelatedCompounds Using Lipases and Proteases

S no.
1
2
3
4
5

Acyl Acceptor
Arabinose
D-Glucose
DGlucose
D-Glucose
D-Glucose

Solvent
Diisopropylether
Pyridine
Pyridine
Heptane
Diisopropylether

Reaction
Time
Enzyme"
72 h
PPL

Yield

18 h
5d
46h
72 h

Subtilisin
Subtilisin
LipozymeIM
PPL

Product
1-0-Acetylarabinofuranoside
6-0-butyrylglucose
6-0-butyrylglucose
1- or 6-0-stearate glucose
6-OAcetylglucopyranoside

(%)

68

64
60
9.32
62

Ref.
Sharma & Chattopadhyay, 1993
Riva et al., 1988
Riva et al., 1988
Oguntimein et al., 1993
Sharma & Chattopadhyay, 1993

Y
3
3.
2

D-Glucose

Pyridine

l h

Subtilisin

6-Omonoacylglucose

96

Park et al., 1996

7
8

9
10
11

D-Glucose
D-Fructose
D-Fructose
D-Fructose
D-Fructose

Dimethylformamide
tert-Butanol
2-Methyl-2-butanol
n-Heptane
Diisopropylether

7h
46h
26 h
12 h
72 h

Proteases
SP382
Lipozyme IM
Lipozyme IM
PPL

6-0-Vinyladipoylglucose
1- or 6-0-stearate glucose
Fructoseoleate
Fructosemonostearate
1-0-Acetylarabinofuranoside

56
8.6
83
40
70

Shibataniet al., 1997

Oguntimein et al., 1993
Khaled et al, 1991
Schlotterbecket al., 1993
Sharma & Chattopadhyay, 1993

12

D-Fructose

2-Methyl-2-butanol 24 h

Lipozyme IM

1-0-Monoacylfructose
6-0-monoacylfructose

39

Scheckermann et al., 1995

13

D-Fructose

n-Hexane

1-0-Monoacylfructose

40

Scheckermann et al., 1995

14

D-Fructose

tert-Butanol

ByssOChlumys'lVu Mixture of esters

71.3

Ku & Hang, 1995

15

D-Fructose

Dimethylformamide 3 h

Pseudomonus sp

1-0-Laurylfructose

80

Sin et al., 1998

60-ButyrylD-glucose

Tetrahydrofuran

40 h

Chmobacteriurn viscosurn
,ipase

2
W

3.6-Di-0-butyrylglucose

50

Therisod & Klibanov, 1987

3
in
x

74 h

C. viscosum lipase

Mixture of sorbitol esters

80

Janssen et al., 1991

48h

SP382

Mixture of esters

56

Akoh & Mutua, 1994

17

Sorbitol
2-Pyrrolidone
Methyl a-DBenzene/pyridine
l8 glucopyranoside (21 v/v)

12h
24h

LipozymeIM
lipase

Gi.

m
W
n

s-

s.

P
b

,$

g
rn

m
%

Y
CT

l9 lsopropylidenc
D-xylofuranose Dimethylformamide 30 h

Novozym 435

Xylose-5-arachidonate

83

Ward et al., 1997

PPL = porcine pancreatic lipase, Lipozyme IM = immobilized Rhizopmucormiehei lipase (Novozymes, Inc, Frankllinton, NC), SP 382 and Novozym 435 =
immobilized CundiduuntarcticuA+B and B lipases, respectively(Novozymes)
Kennedy et al. (2006). Reproducedwith permission by John Wiley and Sons.

2
t

w
h,
w

328 0 5-H. Pyo and D.G. Hayes

pyridine, which, in the early literature on employment of lipases in organic media

was often used as solvent for esterification of saccharide and other polyols (Therisod
& Klibanov, 1986; Mutua & Akoh, 1993; Sin et al., 1998; Chopineau et al., 1998),
should be avoided owing to the high toxicity of this solvent (Villeneuve, 2007). The
solvents log P value can also influence the rate and extent of acyl migration, in which
acyl groups of polyol-fatty acid esters migrate between different hydroxyls of the acyl
acceptor moiety (e.g, isomerization of 1- and 2-monoglycerides) (Compton et al.,
2007).
The effect of organic solvent type on the lipase catalyzed esterification of
saccharide and fatty acid will now be illustrated by discussing some recent examples
in the literature. A recent report on the synthesis of glucosylmyristate by Novozym
435 (Candidz antarctica B lipase immobilized onto nylon beads, Novozymes Inc.)
also describes the influence of the pre-reaction treatment of the enzyme and reaction
mixture and the use of adsorbents to remove water generated by the reaction on the
rate, conversion, and selectivity (Cauglia & Canepa, 2008). Myristic acid is solublized
in the reaction solvent at 0.5 M concentration at 60C. An equal molar amount
of glucose is added and the reaction mixture is equilibrated overnight at 60C. The
enzyme preparation is equilibrated in the reaction solvent overnight under the same
condition in a separate container. The reaction starts when pretreated Novozym 435
is added to the substrate solution at the concentration of 6.7 mg/mL. The use of tertbutanol (2-methyl-2-propanol) with molecular sieves (100 mg/mL), which is much
more efficient at dehydrating than salts (such as CaSO,, CaCl,, and MgSO,), gives
higher ester yields (43% in an 8 h reaction) in comparison with the use of hexane and
diethyl ether mainly due to the low solubility of acyl acceptor. Acetone, a solvent of
intermediate polarity (Table 12.2), gives 28% ester yield in a 7 h reaction time.
Consistent with previous literature (reviewed in Hayes, 2004), a recent
investigation of the mono-esterification of glucose with stearic acid catalyzed by
immobilized lipases from Candida sp in acetone in the presence of molecular sieves
demonstrated the kinetics described by a 2-substrate Michaelis-Menten-based model,
the Ping-Pong bi bi mechanism, which predicts a time course consisting of a linear
rapid conversion of substrate during the initial few hours, followed by a gradual
reduction of the reaction rate as time is increased (Yu et al., 2008). The activation
energies for the formation of the acyl-enzyme complex and for the formation of the
monoester were calculated to be 52.9 kJ/mol and 19.4 kJ/mol, respectively.
When apolar solvent is used, it is important to provide a means to remove the
released hydroxyl-containing products, water, and n-alkanol derived from free fatty
acid, FFA, and alkyl ester acyl donors, respectively, to allow for a high polyol-ester
yield. 6-0-P-D (+)-Glucose fatty acid monoesters were synthesized from p-D(+)glucose and FFA or fatty acid methyl esters (FAME) of saturated acyl groups (Cs,
CIB,CIS)with Novozym SP525 (free lipase B from C. antarctica B, Novozymes,
Inc.) immobilized onto polypropylene resin (Yan et al., 1999). Highest yields (up to
90%) were achieved in ethyl methyl ketone or acetone as solvent by conducting the
reactions under reduced pressure at 60C in order to remove the by-products, water

Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 329

(FFA acyl donor) or methanol (FAME acyl donor), by creating an azeotropic mixture
that was subsequently treated by distillation.
Mixtures of high and low hg P solvents, which represent a compromise between
high enzyme activity and saccharide solubility, can be employed for lipase-catalyzed
polyol ester synthesis (Degn & Zimmermann, 2001; Ferrer et al., 2005). The highest
activity for the Novozyme 435'-catalyzed synthesis of saccharide-myristic acid esters
was observed for the relatively nonpolarlpolar solvent mixture tert-butano1:pyridine
55:45 v/v (Degn & Zimmermann, 2001). For solvent systems that yielded low glucose
solubility (< 60 mM), the measured enzyme activity retention was generally high;
in contrast, solvent systems that achieved high glucose solubility led to low activity
retention, reflecting a loss of enzyme stability due to the relatively polar solvent, and/
or to substrate inhibition. The effect of the addition of dimethyl sulfoxide (DMSO),
a polar solvent known to inactivate enzymes, on the activity of the lipase was also
investigated for a ternary (tert-butano1:pyridine:DMSO) solvent system. With > 2%
(v/v) DMSO in the reaction mixture the enzyme activity and stability decreased,
indicating a denaturing effect of DMSO on the enzyme (Degn & Zimmermann,
2001).
Ferrer et al. (2005) employed apolar/polar solvent systems (2-methyl-2-propanol/
DMSO, 4/1 v/v) as reaction media for the acylation of sucrose, maltose, and glucose
catalyzed by granulated I: kznuginosus lipase (Novozymes, Inc.), which is selective
toward the 6-OH group of sucrose. In this solvent system, over 80% of the initial
activity was retained after 20 successive 6-h batch reactions (Ferrer et al., 2005). Ferrer
et al. (1999) demonstrated with I: kznuginosus lipase that the DMSO percentage
in the solvent mixture substantially modified the final esterification degree. Thus,
at DMSO concentrations 510% the formation of diesters was favored, whereas at
percentages higher than 15% the formation of diesters was minimized (Ferrer et al.,
1999). Moreover, the increased polarity of the reaction medium promoted by the
higher concentration of DMSO led to an increased proportion of the more polar
monoester product.
Although there have been many successes on a laboratory scale, use of organic
solvents for the synthesis of sugar fatty acid esters has certain limitations for largescale synthesis. There are only a few solvents, such as acetone, acetonitrile, tertbutanol, pyridine, DMF, or DMSO, which can solubilize saccharide to a significant
extent; but, the toxicity and environmental safety of these solvents is a concern.
Also, many enzymes do not retain their catalytic activity in these solvents. Therefore,
alternate methods that reduce the solvent amount or eliminate their need have been
recommended recently.

Synthesis of Saccharide Fatty Acid Esters in

Supercritical CO,
Supercritical carbon dioxide (CO,), which may provide an interesting alternative,
has several advantages over organic solvents since it is inert, nontoxic, nonflammable,

330 0 S.-H. Pyo and D.G.Hayes

and inexpensive. Since enzymes are insoluble in supercritical CO,, the catalyst can
be easily separated from the reaction mixture (Habulin et al., 2008). The Novozym
435"-catalyzed synthesis of fructose palmitate, fructose laurate, sucrose palmitate
and sucrose laurate were carried out at 60C in 2-methyl-2-butanol at atmospheric
pressure and in supercritical CO, at lOMPa (Habulin et al., 2008). The highest
conversions for fructose palmitate synthesis were obtained in 2-methyl-2-butanol
and in supercritical CO, (65% and 61%, respectively). Tsitsimpikou et al. (1998)
investigated the acylation of glucose with lauric acid at a mole ratio of 1:5 catalyzed
by Novozym 435" (12.5 mg/cm3) in supercritical CO,. Glucose conversions up to
60% at 60C have been observed. Since the only substrate or product that is soluble
in the supercritical phase is lauric acid, the system offers the potential advantage of
easy separation of the glucose laurate product from remaining substrate and enzyme
upon completion of the reaction.
Although supercritical CO, is a promising alternative to organic solvents, its
employment has limitations: only non-polar compounds are soluble at an acceptable
level (Tsitsimpikou et al., 1998), and capital costs for systems that include a high
pressure reactor and controller are high.

Synthesis of Saccharide Fatty Acid Esters via

Derivatization
Reaction rate and conversion can be enhanced in non-aqueous media by the
derivatization of fatty acyl groups and/or saccharides in organic solvents; the former
through the "activation" provided by their covalent attachment to good leaving
groups and the latter to enhanced solubilization (Coulon et al., 1999). In early work
in the field, activated fatty acids in polar organic solvents were used for the onestep, preparative regioselective enzymatic acylation of primary hydroxyl groups of
various underivatized monosaccharides (Therisod & Klibanov, 1986). D-Glucose
was dissolved in warm anhydrous pyridine at 45"C, followed by the addition of
2,2,2-trichloroethyl laurate and dried porcine pancreatic lipase, and then subjected
to shaking at 250 rpm. 6-O-Laurylglucose was obtained at 40% conversion and 95%
selectivity (Therisod & Klibanov, 1986).
Derivatization of monosaccharides, sugar alcohols, and other polyols by
isopropyldene, a common protective group for sugars, has led to enhanced
solubilization of the polyol acyl acceptors (yielding higher rate and conversion) and
to improved selectivity by the blockage of hydroxyl groups covalently bound to
isopropylidine. 1-0-lauroyl-D-mannitol, a non-ionic surfactant, was synthesized via
a chemo-enzymatic procedure that utilizes I ,2:4,5-di-O-isopropylidene-D-mannitol
and vinyl laurate as substrates (Fig. 12.2; Pinna et al., 2004). The acyl acceptor was
synthesized by reacting D-mannitol with 2 mole equivalents of 1,2-dimethoxypropane
in 1,2-dimethoxyethane under neutral condition, yielding a product that has only
two free hydroxyl groups (Chittenden, 1980). The reduced polarity of this compound
compared to D-mannitol enables it to solubilize in n-hexane and solvent-free reaction
medium.

Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 33 1

1,2:4,5-di-O-isopropylidene-D-mannitol(O.373
g, 1.40 mmol) and vinyl laurate
(0.771 g, 3.40 mmol) were dissolved in n-hexane (10 mL). After the addition of0.50 g
of Novozym 435",transesterification was carried out at 50Cunder magnetic stirring.
The product of the enzymatic acylation, 1-0-lauroyl-2,3:5,6-di-O-isopropylidene-Dmannitol was produced at a yield of 92%. The same reaction performed in solventless
media provided a 65% yield. The selective acid-catalyzed hydrolysis to remove the
isopropylidene protective groups gave the surfactant 1-0-lauroyl-D -mannitol with
an overall yield of 80% and a high selectivity (Pinna et al., 2004).
The enzymic solvent-free acylation of sugar acetals with a range of long-chain
fatty acids (lauric, myristic, palmitic, and stearic) and their methyl esters has been
demonstrated (Fig. 12.3; Fregapane et al., 1991). The lipase-catalyzed reaction was
carried out at 1:l and 2:1 mole ratios of FFA or FAME to sugar acetal to obtain
mono- or diesters, respectively. A range of 6 monoesters of 1,2-O-isopropylideneD-glucofuranose and 1,2:3,4-di-O-isopropylidene-D-galactopyranoseas well as

"

$G

II

H~C-OC(CH~)~OCHS

HO
H3C

0
II

HzC-OH
Vinyl laurate

Novozym435

H3CXO-CH2

H3C

Acetic acid

H~~~

90%

HO

_____e)

H3CXO--CH2

H~C-OC(CH~)~OCH~

HO-CH2

Fig. 12.2. Synthesis of 1-0-lauroyl-D-mannitol. Pinna et al. (2004). Reproduced with permission from
Elsevier.

Acetalization

OH

b'CH3

0
II

H&-C-CH3

Lipase

Fatty acids
Methyl esters

t
Deacetalization

'.0

XH3
CH3

c,

Fig. 12.3. Synthesis of xylose -fatty acid esters via isopropylidene derivatives. R represents alkyl group.
Fregapane et al. (19911. Reproducedwith permission from Elsevier.

332 0 5-H. Pyo and D.G. Hayes

5-mono- and 3,5-diesters of 1,2-0-isopropylidene-D-xylohranosewere prepared.

Syntheses were performed at 75C in open vials or under vacuum, to provide a means
to remove water or methanol produced in the course of the reaction. Typically yields
of 50-90% were obtained under optimal conditions (Fregapane et al., 1991).
Another derivatization approach, formation of alkyl glucoside acyl acceptors,
allows the formation of new sugar esters and a highly efficient iipase-catalyzed
process for regioselective esterification of the primary hydroxyl group in simple alkyl
glucosides. Alkylation converts reducing sugars with reactive C- 1 anomeric centers to
non-reducing sugars with less reactive anomeric centers (Bjorkling et al., 1989; Mutua
& Akoh, 1993). Regioselective 6-O-esterification of alkyl glucosides with long-chain
fatty acids, yielding more than 95% of 6-O-monoesters, can be achieved using lipases
as catalysts in a solvent-free process (Bjorkling et al., 1989). This procedure allows
for ethyl D-glucopyranoside to be esterified with C,-C,, saturated FFA, catalyzed by
Novozym 435", producing 6-O-monoester at 8 5 9 5 % yield (Bjorkling et al., 1989).
Alkyl glycoside fatty acid esters were successfully synthesized by lipase-catalyzed
transesterification of methyl glucoside, methyl galactoside, and octyl glucoside with
methyl oleate (Mutua & Akoh, 1993). The optimal conditions for the enzymatic
synthesis of alkyl glycoside fatty acid esters were: a mole ratio of alkyl glycoside and
methyl oleate of 1:4; use ofSP382" (immobilized C antarcticaA+Blipase, Novozymes,
Inc.) as biocatalyst; benzene/pyridine (2: 1, vol/vol) with no added water as solvent,
55C;48 h of reaction time; and agitation provided at 200 rpm. Acceptable levels of
oleic acid incorporation (58.6-100 mol %) into the alkyl glycosides were achieved.
In summary, the use of isopropylidene-derived saccharides or alkyl glycoside as
acyl acceptor substrates led to high yield and selectivity. However, this approach has
several disadvantages, including extra protection and/or deprotection steps, additional
material cost, and perhaps modified emulsifier performance for the alkyl glycoside
derivatives.

Synthesis of Saccharide Fatty Acid Esters Using

CompIexat ion Agents
Employment of phenylboronic acid, a solubilizing agent for hydrophilic substrates in
nonpolar solvents, results in enzymatic acylation of a single primary hydroxyl group of
a saccharide (Schlotterbeck et al., 1993; Oguntimein et al., 1993; Ikeda & Klibanov,
1993; Scheckermann et al., 1995). Organoboronic acids are known to solubilize sugar
by forming a carbohydrate-boronate complex through reversible condensation with
carbohydrates (Oguntimein et al., 1993). Of the many organoboronic acids capable
of forming reversible lipophilic complexes with sugar (Ferrier, 1978), phenylboronic
acid (2 mole equivalents) was found to effectively complex with a-D-glucose (1 mole
equivalent) in benzene to give the corresponding 1,2:3,5-bis(phenyIboronate) (Fig.
12.4). This complex, which, in contrast to free glucose, is soluble in a variety of organic
solvents, was employed as a target for enzyme-catalyzed acylation in early work in the
field (Ikeda & Klibanov, 1993). For a typical reaction, a molar excess of vinyl acrylate

Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 333

(0.3 M) was co-dissolved with glucose-phenylboronic acid complex (0.1 M) into 10

mL anhydrous tert-butyl alcohol; subsequently, lyophilized Pseudomonas sp. lipase
(1 g) was added, The reaction was allowed to proceed for 4 days under continuous
agitation. After removal of the lipase by filtration, distilled water (10 mL) was added
to remove phenylboronic acid, The overall yield was 77% (Ikeda & Klibanov, 1993).
Scheckermann et al. (1995) carried out the lipase-catalyzed monoacylation of
phenylboronic-complexed fructose with long-chain (e.g., palmitic and stearic) FFA in
hexane or 2-methyl-2-butanol. The reaction using Lipozyme IM" (Rhizomucormiehei
lipase immobilized onto anion exchange resin, Novozymes, Inc.) or Novozym 435" in
n-hexane at 60C was favored by a 1.0:3.0:4.5 mole ratio of fatty acid to fructose to
phenylboronic acid, with only the monoester at the C-1 position of fructose formed
at 40% yield with respect to the acyl donor (Scheckermann et al., 1995). In contrast,
in the absence of phenylboronic acid, a mixture of the C-1 and C-6 monoacylated
fructose fatty acid esters were formed. Similarly, Schlotterbeck et al. (1993) in a
one-pot-process for the Lipozyme IW-catalyzed monoacylation of fructose (1 mole
equivalent) with stearic acid (3 mole equivalents) in n-hexane using phenylboronic
acid as a solubilizing agent achieved a 40% yield, with two monoester isomers formed
(Fig. 12.5).
In conclusion, the employment of phenylboronic acid complexation agent
enhances the rate of reaction through the increased solubilization of the acyl acceptor,
and improves the selectivity toward monoesters by blocking many of the hydroxyl
groups of the acyl acceptor through complexation as depicted in Fig. 12.4. However,
this approach has several disadvantages, including use of organic solvents, additional
material cost, and processing steps to add and remove the complexation agent.

II

::

CH20CCH=CH2

Fig. 12.4. Synthesis of 6-0-acryloylglucose via complex formation of D-glucose and

1,23,5-bi(phenylboronate) at a 1:2 mole ratio in benzene. lkeda & Klibanov, (1993). Reproduced with

permission from John Wiley & Sons, Inc.

334 0 S.-H. Pyo and D.G. Hayes

Fructose

+
Stearic
acid

1. Phenylboronicacid
Lipozyme, 60C

2. Mild hydrolysis

OH

*
OH

OH

Fig. 12.5. Reaction scheme for the enzymatic monoacylation of fructose with stearic acid in a one-potprocess. Schlotterbeck et al. (1993).Reproducedwith permission from Springer.

Synthesis of Saccharide Fatty Acid Esters

in Solid Phase System
As an alternate approach, lipase-catalyzed synthesis of sugar fatty acid esters can be
carried out in a mainly solid-phase system consisting of saccharide, fatty acid, and
product in the presence of a small amount of organic solvent (e.g., tert-butanol or
acetone) at relatively low, often near ambient, temperature. The solvent serves mainly
as adjuvant to maintain a catalytic phase for the action of the enzyme (Cao et al.,
1996). Although saccharides (such as glucose, mannose, and galactose), polyols, and
sugar alcohols (such as sorbitol) are almost insoluble in the system's liquid phase,
high conversion to saccharide ester, mostly monoester, was achieved, due in part to
the rapid crystallization of the latter from the liquid phase. Complete conversion
at 60C after 48 h, with a productivity of 0.4 mmol sugar ester/g lipase h for the
esterification of P-D(+)-gIucose with stearic acid were achieved using solvents having
low log Pvalues and product solubility such as acetone, Novozym 435" as biocatalyst,
and saturated FFA with chain lengths from C,, to C,, as acyl donor (Cao et al., 1996).
For the Novozym 435"-catalyzed synthesis of 6-O-glucose palmitate, the reaction
mixture consisted of equimolar amounts of glucose and palmitic acid (typically
0.5 mmol), 2 mass equivalents of adjuvant (tert-butanol) (w/w of substrates), and
0.2 mass equivalents of activated molecular sieves (4 A, 10 mesh) to adsorb water
generated during esterification (Cao et al., 1999). The reaction mixture was incubated
in a capped vial, placed in an oil bath, thermostated to 60C and stirred by a magnetic
bar (250 rpm). High conversion (84%, 24 h) and productivity (0.69 mmol product/g
lipase h) were achieved.
For the latter reaction, Novozym 435" was reused in six subsequent batch reactions
and conversion decreased by approximately 25% for reactions employing long-chain
fatty acyl donors, thus demonstrating good enzyme stability. As described previously,
the polarity of the liquid phase can strongly influence enzyme stability, with more

Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 335

polar solvents reducing stability. In the solid-phase system, the polarity of the liquid
phase is strongly influenced by the dissolved fatty acid; however, the influence of the
sugar or sugar fatty acid ester on lipase stability can be neglected due to their low
solubility (Cao et al., 1999). The lower conversion and lipase stability observed using
short-chain htty acid can be explained by the higher polarity (i.e., lower log P) of the
short-chain FFA, producing a lower overall logPvalue for the liquid phase (Cao et al.,
1999).
In summary, a solid-phase system results in quantitative saccharide fatty acid
ester yield and a relatively high reaction rate, provides an enhanced selectivity toward
monoesters, and can utilize near-ambient temperatures. However, this approach
possesses disadvantages, such as the requirement of organic solvent and its inherently
batch nature, and its translation into a continuous process is difficult (Ganske &
Bornscheuer, 2005a). The separation of the monoester, unreacted saccharide, and
immobilized biocatalyst from the solid phase may also be challenging.

Synthesis of Saccharide Fatty Acid Esters in Ionic Liquid

Phase System
Ionic liquids (IL) are considered to be environmentally friendly alternatives to
organic solvents for organic synthesis in general and catalytic processes in particular
because of their very low vapor pressure and good chemical and thermal stability
(Sheldon et al., 2002; Ganske & Bornscheuer, 2005a). This interest stems from their
potential as green solvents (Sheldon et al., 2002). Their non-volatile character and
thermal stability makes them potentially attractive alternatives for environmentally
unattractive volatile organic solvents, notably chlorinated hydrocarbons. Moreover,
their hydrophobicity and hydrophilicity can be tuned by appropriate modification
of the cation and/or anion, which has earned them the accolade designer solvents
(Freemantle, 2001). Depending on their chemical composition, they can be miscible
with water or alkanes, which renders them useful for performing catalytic reactions
in biphasic media, thus facilitating catalyst recovery and recycling. Most studies
involving enzymes have involved the use of 1,3-dialkylimidazolium salts (e.g., l-butyl3-methylimidazolium tetrafluoroborate, [bmim] [BF4], and hexafluorophosphate,
[bmim] [PF6]) which are miscible and immiscible with water, respectively (Fig. 12.6;
Sheldon et al., 2002).
Glucose fatty acid ester synthesis with poly(ethy1ene glycol)-modified C.antarctica
B lipase (Chirazyme L2 C2, Roche Diagnostics, Germany) was performed in pure
[bmim] [BF4] and [bmim] [PF6], yielding 30% and 35% conversion, respectively
(Ganske & Bornscheuer, 2005b). However, in a solvent system composed of ionic
liquid, [bmim] [BF4], and 40 vol. Yo tert-butanol using fatty acid vinyl esters as the
acyl donor operating at 60C, 90% conversion and 89% overall yield for the recovered
biobased surfactant were achieved. Although the literature clearly demonstrates tertbutanol is a useful solvent for sugar ester synthesis (Cao et al., 1996), the solubility of
mono- and di-saccharides is very low (0.34 mg/mL for glucose at 25C; see also Table

336 0 S.-H. Pyo and D.G. Hayes

+ x
CH3

R =CnH~n+i
X = PF6, BF4,MeS04, octylsulfate(OS04),
dimethylphosphate(DMP), etc

Fig. 12.6. Structures of ionic liquids based on methylimidazolium salts.

l2.l), and an acceptable conversion rate could only be achieved for the solid-phase
approach. Ionic liquids (IL) yield several-fold higher values of solubility (Table 12.3);
thus, their addition to tert-butanol results in an increased liquid-phase saccharide
concentration, In addition, the same authors also successhlly employed FFA as acyl
donor under the same conditions to achieve a high yield and reaction rate (Ganske &
Bornscheuer, 2005a).
Novozym 435" catalyzed synthesis of 6-0-lauroyl-D-glucose using supersaturated
glucose solution in IL was investigated (Table 12.3; Lee et al., 2008a). The highest
lipase activity was obtained in water-miscible 1-butyl-3-methylimidazolium
trifluoromethanesulfonate, [bmim] [TfO], which can dissolve high concentrations
of glucose, while the highest stability of lipase was shown in hydrophobic l-butyl3-methylimidazolium
bistrifluoromethylsulfonylimide, [bmim] [TQN]
The
supersaturated solution is usually prepared by dissolving excess solute at high
temperature and then slowly cooling to low temperature. However, Lee et al. recently
developed an alternate dissolution process that uses water as a mediator (Lee et al.,
2008b). First, saccharide (50 mg) was dissolved in water (0.3 mL) and mixed with IL (1
mL) at room temperature. After clear solutions were obtained, the water in the mixtures
was removed by vacuum evaporation for 12 h at 60C. The saturated sugar solutions
were slowly cooled to 25C and then incubated for 2 h at 25C. After centrifugation,
the supernatant was obtained and its sugar concentration measured. Supersaturated
solutions prepared by this method yielded apparent glucose concentrations of 113
and 46.3 g/L at 25C for [emim][TfD] and [bmim][TfO], respectively, 19 and 10
times higher than the true solubility (6.1 and 4.8 g/L, respectively). A supersaturated
[bmim] [TfD]:[bmim] [TON] (1: 1 v/v) mixture produced an even higher solubility
at 25"C, 222 mM. Furthermore, the supersaturated glucose solutions in IL were
maintained over several hours without any significant precipitation of glucose.
The optimal activity and stability of lipase were obtained in a [bmim][TfO]/
[bmim][TBN] (1:1, v/v) mixture. Specifically, the activity of lipase was increased
from 1.1 to 2.9 pmol/min g by using a supersaturated glucose solution in this mixture,
compared with reaction using a saturated glucose solution (Table 12.3). To measure
enzyme stability, lipase (Novozym 435") was employed and then subsequently
recovered from 5 subsequent batch reactions. After the fifth batch reaction, Novozym
retained 86% of initial activity when the [bmim] [TfO]/[bmim] [TON] (1: 1, v/v)
mixture was employed; in contrast, the residual activity using the pure polar IL [bmim]
[TfO] was only 36%. Therefore, the productivity obtained by using IL mixtures was

Synthesis of Saccharide Fatty Acid Ester BiosurfactantsCatalyzed by Lipase 0 337

Table 12.3. Dissolved Glucose Concentration, Enzyme Activity, and Residual Enzyme
Activity in Ionic Liquids (IL) and IL Mixtures

IL'

Dissolved Concentration
Enzyme Activity"
(pmol/min/g)
of Glucose at 40C (mM)
Saturated Supersaturated Saturated Supersaturated
Solutionc Solutiond
Solution Solution

Residual
enzyme
Activitpb(%)
Supersaturated
Solution

[bmiml[BFJ

6.3

43.4

0.65

1.29

66.1

[bmiml[TfOl

30.6

363.3

1.31

4.2 1

70.5

[bmim][TfO]
:[bmim] [Tf,N],
25/75 vlv
[bmiml
[TfO]:[bmim]
[Tf,N], 50/50 v/v
[bmim]
[TfOl:[bmiml
[Tf,Nl, 75/25 v/v

11.1

127.2

1.27

3.75

87.9

6.1

50.0

1.08

2.85

96.0

1.7

18.3

0.98

1.89

95.9

[bmiml[Tf,N]

0.6

1.7

0.69

0.77

95.7

[bmiml
[TfOl:[bmim]
[PF,], 50/50 v/v

6.1

50.0

0.92

1.94

95.0

[bmiml[PF,l

0.6

1.7

0.67

0.76

96.0

Reaction conditions: 222mM glucose, 444mM vinyl laurate, 0.5mL IL, 50 mg Novozym 435,
40"C.The water contents in IL were less than 0.1%. A t the end of the reaction, deionized
water ( 1 mL) was added to the reaction vials in order to remove unreacted glucose. The
concentration of unreacted glucose in the supernatant after centrifugation was determined
by the dinitrosalicylic acid (DNS) method with a glucose standard. Enzyme activity was
obtained from the glucose content consumed after 6 h reaction.The precipitated product
(6-0-lauroyl-d-glucose)was dissolved in tetrahydrofuran and the concentration was
determined by HPLC (Lee et al., 2008b).
bAfter reaction for 6 h, Novozym 435 was washed with water and tetrahydrofuran and then
dried in desiccator before reuse. The residual activity was determined by the relative glucose
conversion in subsequent reaction.
Ionic liquids and glucose crystal were directly mixed and incubated for 12 h at 40C.
Ionic liquids and glucose solution in water were mixed. The mixture was then evaporated to
remove water at 60C and slowly cooled to 40C.
5olvents: [bmiml[BF.lI, 1-butyl-3-methyl imidazolium tetrafluoroborate; [brnim][TfO],lbutyl-3-methylimidazoliumtrifluoromethanesulfonate;[bmim][TfZN], 1-butyl-3-methyl
imidazolium-bis-trifluoromethylsulfonylamide;[bmimI[PF61,1-butyl-3-methyl imidazolium
hexafluorophosphate.
Lee et al. 2008a. Reproduced with permission from Elsevier.

338

S.-H. Pyo and D.G. Hayes

higher than those in pure IL. Figure 12.7 plots the time course for the enzymatic
synthesis of 6-0-lauroyl-D-glucose in pure IL and the IL mixtures. Lauroyl glucose
concentrations of 190 and 153 mM were obtained after 11 h of incubation in the
pure IL and the IL mixture, respectively (Lee et al., 2008a).
Biocatalytic conversions in IL can be beneficial with regard to activity, selectivity,
and stability in comparison with more classical organic solvents (Sheldon et al., 2002;
Villeneuve, 2007). Indeed, the use of enzymes in IL opens up new possibilities for
nonaqueousenzymology (Sheldon et al., 2002). However, there remainseveralproblems
to overcome, such as enzyme inactivation, poor solvent toxicity, and biocompatibility
for many IL. To date, the use of IL on an industrial scale for biocatalyzed reactions is
limited due to the high price of such compounds and can be seen as a realistic solution
only for the production of value-added products (Villeneuve, 2007). However,
material costs for ILs are predicted to decline as new applications for IL continue to
develop.

100
80

20
0
0

10

12

Time (h)
Fig. 12.7.Time courses for the lipase-catalyzed esterificationof glucose with vinyl laurate in pure IL and
IL mixtures Reaction conditions: 222 mM glucose, 444 mM vinyl laurate, 0.5 mL IL, 50 rng Novozym
435,4OoC(+, [bmiml[TfOl: @,SO% [bmiml[TfOl with 50% [bmimltTf,Nl: A,[bmiml[Tf,NI). Filled symbols
represent the reaction with supersaturated glucose solution. Unfilled symbols represent the reaction
with saturated solution with undissolved glucose crystals present. Lee et al. (2008a) Reproduced with
permission from Elsevier.

Synthesis of SaccharideFatty Acid Ester Biosurfactants Catalyzed by Lipase 0 339

Synthesis of Saccharide Fatty Acid Esters

in Solventless System
Lower cost, higher substrate concentration, improved process safety, improved
environmental impact, and greater volumetric production are some of the advantages
of solvent-free systems for conducting biocatalytic reactions (Foresti et al., 2007).
Although several different approaches reduce organic solvent use and maintain high
yield, problems persist with even low solvent amounts, including additional material
costs and pre-reaction, purification, and recycling steps. Alternatively in solventfree systems, the absence of solvents Edcilitates downstream processing since faver
components would be present in the reaction mixture at the end of the reaction;
moreover, the elimination of solvents from the production step offers significant
cost savings (Yahya et al., 1998) and can improve process safety and product
biocompatibility.
In some ways, solvent-free systems are similar to solvent-based systems with the
substrate(s) and product(s) serving as solvent. A distinguishing feature of these
systems is the elevated impact of the substrates and products on the characteristics
of the reaction media. As the reaction progresses, the composition (and hence, the
properties) of the reaction medium can change dramatically. One drawback of
solventless systems is the possibility of high liquid-phase viscosity, especially for longer
chain alcohols, which may result in poor solute mass transfer rates to the biocatalytic
solid phase and hence, slow reaction rates (Yahya et al., 1998). For reaction systems
where the substrates are miscible, for example, the lipase-catalyzed esterification of
FFA and n-alkanols (e.g., ethanol), the solvent-free approach can be quite successhl
(Foresti et al., 2007). For saccharide-fatty acid ester synthesis, the poor miscibility of
the substrates makes the solventless approach challenging. As noted by Hayes and coworkers, the saccharide-fatty acid ester products are good solvents to co-solubilize
saccharide and FFA (Dang et al., 2005). The apparent solubility of saccharide in a
mixture of oleic acid substrate and fructose mono- and di-esters products at 60Cwas
reported in the form of a ternary phase diagram demonstrating the presence of ester
greatly increased the solubility to an extent that co-solvent would not be required;
moreover, fructose solubility increased linearly from 0.002 to 0.13 g/g as the ester
mass fraction increased from 0.00 to 0.80, respectively (Fig. 12.8; Dang et al., 2005),
in agreement with previously published results for fructose and glucose (Zhang &
Hayes, 1999; Tsavas et al., 2002). However, recent work by the authors demonstrated
that suspensions of fructose in oleic acid/fructose oleate were formed at the previously
mentioned conditions rather than true solutions, with the true solubility limit being
a factor of 2-3 smaller (Pyo & Hayes, 2008).
The large increase of fructose solubility as the proportion of ester increased was
utilized for the Lipozyme 1M-catalyzedesterification of saccharide (fructose, sucrose,
and xylose) and oleic acid at 65C in batch mode, using near-stoichiometric amounts
of substrates (Dang et al., 2005). A solvent, tert-butanol, was employed only during

340 0 S.-H. Pyo and D.G. Hayes

Oleic Acid

Fructose

Monoester

Fig. 12.8.Ternary phase diagram for fructose/oleic acidhechnical-grade monoester (5% diester and 95%
monoester) at 60C.An apparent one-phase liquid mixture exists to the right of the phase boundary,
two-phase media to the left of the boundary. Dang et al. (2005). Reproduced with permission from
Springer.

the initial period to enhance fructose solubility, and (along with the product, water)
was allowed to evaporate away, with 100% solvent removal achieved upon reaching
~ 2 5 %conversion. The acyl acceptor was added in fed-batch mode. A conversion of
80-93% was achieved (Fig. 12.9), with the product consisting of mono- and di-ester
at a ratio of 9: 1 g/g. Due to the high conversion and the use of a stoichiometric feed, a
technical-grade bio-based surfactant product was formed, with the only downstream
processing required being the simple centrifugation or filtration-based removal of
immobilized lipase. Lipozyme IM" exhibited excellent operational stability within 1
month of continuous use (Dang et al., 2005).
These results provide the basis for the development of bioreactor system design
for Lipozyme 1M"-catalyzed solvent-free synthesis of saccharide fatty acid esters,

Synthesisof Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 341

Time, h
Fig. 12.9.Time course of lipase-catalyzedfructose ester synthesis ( A conversion of oleic acid, 6 product
distribution, and C fructose concentration on a tert-butanol-free basis) using a reaction medium
saturated with fructose and nearly stoichiometrically equal amounts of the two substrates. Initial
conditions: 100 mmol(28.247 g) oleic acid, 25.00 g (32.05 mL) tert-butanol, 1.00 g Lipozyme IM, fructose
at saturation (y-intercept of panel C), 65T, and a 400-rpm stir rate.The reaction medium was resaturated
with fructose periodically (indicated in panel C by increases of fructose concentration). t-BuOH and
water were allowed to freely evaporate during the time course of esterification. The curve in panel A
represents the prediction by the Ping-Pong Bi Bi kinetic model. Dang et al. (2005). Reproduced with
permission from Springer.

under investigation in the authors laboratory (Fig. 12.10). The bioreactor systems
main components are the bioreactor and a packed column containing saccharide as a
major component of its stationary phase to deliver saccharide at saturated conditions.
The desorption of saccharide from a fructose/silica gel packed column to a solventfree oleic acid/fructose oleic acid ester liquid solution at 65C increases as the fructose
oleate concentration in the liquid phase increases and the saccharide mass fraction

342 0 S.-H. Pyo and D.G. Hayes

F
C

E
D

Fig. 12.1 0. Bioreactor System Designs for Solventless Synthesis of Fructose-Oleic Acid Ester, FOE. A.
Reservoir tank "(no lipase) or Stirred Tank Bioreactor (STBR, with lipase, designated as A'); B. Peristaltic
Pump (0.1 mL/rnin); C. Fructose desorption column (DC; 100 x 10 mm ID); D. Molecular sieves column
(MSC; lg, 50 x 10 mm ID); E. Packed-Bed Bioreactor (PBBR; 50 x 10 mm ID, Lipozyme RM IM.); F.Tempcontrolled Oven; 1. STBR /fed-batch addition of fructose: A'; 2. STBR / DC: A' B C A'; 3. PBBR/ DC: A B C
E A ;4. PBBR / DC / MSC: A B C D E A. Pyo & Hayes, submitted 2008. Reproduced with permission from
Springer.

for the stationary phase increases (up to 70 wt Yo), and follows the well-known
Freundlich isotherm (Pyo & Hayes, 2008). Preliminary results demonstrate that
the reaction rate and conversion are maximized with use of a packed-bed bioreactor
(80285% conversion), and that control of water activity, most readily achieved using
free evaporation, is a critical parameter in the bioreactor system's performance. (Pyo
& Hayes, submitted 2008).
However, the resultant rates of reaction for the bioreactor systems were slower by
a factor of 2-3 compared to the batch-mode reactions, presumably due to the fructose
concentration produced by the packed fructose/silica gel column being a factor of
2-3 times smaller than that produced in the batch-mode reactions (Pyo & Hayes, in
preparation). Through analysis using dynamic light scattering and optical density it
was discovered that under magnetic stirring above 200 rpm suspensions of fructose
crystals formed in oleic acid/fructose oleate mixtures (Pyo & Hayes, 2008), akin to
the supersaturated solutions of saccharides formed in IL as discussed previously. Such

Synthesisof Saccharide Fatty Acid Ester BiosurfactantsCatalyzed by Lipase 0 343

conditions mirror those used for the batch-mode reaction that yielded the results of
Fig. 12.9. In contrast, light scattering of the packed column effluent demonstrated
the absence of aggregates, suggesting this approach yielded saccharide concentrations
corresponding to the true solubilization limit.

Effects of Water Activity and Content in Synthesis of

Saccharide Fatty Acid Esters
Water plays multiple roles on lipase-catalyzed esterification performed in nonconventional media (Foresti eta]., 2007) since its presence promotes hydrolysis, which
will limit the final conversion in accordance with thermodynamic equilibrium. The
removal of the alcohol by-product is desirable when fatty acid esters serve as acyl donor.
Furthermore, the water content influences the enzyme's conformational stability in
the presence oforganic solvents (Sakaki et al., 2006). Enzymes need a small amount of
water to retain their active three-dimensional conformational state for both dispersed
enzyme powder prepared via lyophilization or other means (Zaks & Klibanov, 1988;
Yahya et al., 1998) and immobilized enzyme preparation (Carta et al., 1991). Water
contributes to the structural integrity, active site polarity, and protein stability. It forms
bonds with surface amino acid groups which would otherwise interact with each other
to produce an inactive conformation (Triantafyllou et al., 1995). The actual amount
of bound water needed varies significantly depending on the origin and composition
of the enzyme preparation (Yahya et al., 1998). Water removal in organic solvent or
solvent-free systems is often readily achieved through free evaporation from reaction
vessels open to the atmosphere, the addition of molecular sieves, vacuum pressure,
azeotropic distillation, pervaporation using special membranes such as tubular zeolite
NaA (BNRI, Japan), and dry gas bubbling (Fregapane et al., 1991; Napier et al.,
1996; Kim et al., 1998; Sakaki et al., 2006).
Yan et al. (1999) removed the by-product, water, during lipase-catalyzed sugar
ester synthesis by azeotropic distillation with the intention to develop a process that
is practical on a large scale (Yan et al., 1999). Using an organic solvent system and
either C,, C,6,or C,, FFA or FAME as an acyl donor, 6-O-P-D(+)-glucose fatty acid
monoesters were synthesized with Novozym 435". The reaction mixture was incubated
in a 50-mL two-neck round-bottom flask equipped with a Soxhlet extractor/condenser
apparatus, the latter controlled by a vacuum controller. Molecular sieves, activated
by heating overnight to 250C under reduced pressure, were placed in the Soxhlet
extractor to remove the by-products (water, 3 A; methanol, 5 A). The condensed
solvent was dried by passing through activated molecular sieves before being returned
to the reaction system. This approach provides constant removal ofwater or methanol
generated in the reaction and drives the equilibrium towards sugar ester synthesis.
The highest yields (up to 90%) were achieved using ethyl methylketone or acetone as
solvent and by conducting the reactions under reduced pressure at 60C.
Chamouleau et al. (2001) investigated the effect of both water activity and water
content on fructose-palmitic acid esterification catalyzed by Novozym 435". ?he

344 0 S.-H. Pyo and D.G. Hayes

lipase-catalyzed synthesis was carried out using an equimolar mixture of fructose and
palmitic acid in 2-methyl 2-butanol at 60C. The initial water activity strongly affected
the time course of reaction with the lowest water activity (a, = 0.07) leading to the
highest conversion yield (28.5%) and initial rate (4.9 glL h). The use of molecular
sieves enhanced yield, however the selectivity of the enzyme was affected. Whereas only
fructose monopalmitate was produced in the absence of molecular sieves, the addition
of molecular sieves as a drying agent promoted the synthesis of fructose dipalmitate,
by conversion of up to 18.3% of the acyl acceptor. The change of selectivity may be
explained by the reduction of the hydration layer surrounding the proteins. 'The lack of
water near the enzyme promotes an increase of the hydrophobicity and consequently
decreases the local fructose solubility. In these conditions, monopalmitate becomes
a better substrate than fructose for the nucleophilic attack on the acyl-enzyme
intermediate (Chamouleau et al., 2001). The increase of conversion yield and initial
rate due to the reduction of water activity can be explained by the adsorption curve
of Novozym 435", which indicates that a small increase ofwater amount from 0 - 20
mg water/g dry catalyst promotes an increase of water activity, aw,from 0.0 - 0.8, the
latter value promoting hydrolysis over esterification (Fig. 12.1 1; Chamouleau et al.,
200 1).

100
80
60
40

20

0
0

0.2

0.4

0.6

0.8

measured wate r activity

Fig. 12.11. Water adsorption curve for Novozym 435'at 20C.Chamouleau et al. (2001). Reproduced
with permission from Elsevier.

Synthesisof SaccharideFatty Acid Ester BiosurfactantsCatalyzed by Lipase 0 345

Although the proper amount of water for a given enzymatic reaction depends
on many factors (the selected enzyme, support, solvent system, and polarity and
quantities of substrates), some authors have proposed the existence of an optimum
water content, generally in the range of 0 . 2 4 % (Foresti et al., 2007; Rocha et al.,
1999; Manj6n et al., 1991; Yadav & Piyush, 2003; Is0 et al., 2001). This supposition
is based on two opposing trends: an increase of enzyme activity as water content is
increased (and the essential water molecules of hydration for the enzyme are restored)
versus the increased equilibrium conversion of the reaction as the water content is
decreased (Rocha et al., 1999).
In summary, water, which helps maintain the enzyme activity when at the
proper amount and results from the synthesis reaction of saccharide fatty acid esters,
should be removed to maintain the increased equilibrium conversion of the reaction.
Although several different approaches have been successfully implemented, it is
necessary for new methods to be developed that are applicable to bioreactor systems
upon scaleup.

Conclusio ns
Saccharide-fatty acid esters are value-added products derived from natural feedstocks
that have recently been employed in the foods, cosmetics, and pharmaceutical industry
as green, biodegradable, and biocompatible nonionic surhctants or emulsifiers. Most
commercial saccharide fatty acid esters are produced by chemical methods that require
costly and environmentally unsafe conditions. In contrast, the biocatalytic approach
using immobilized lipase provides milder and more environmentally friendly operating
conditions, lower operating costs, reduced downstream separation requirements, and
enhanced regioselectivity of reactions. Thus lipases have been employed as tools for
organic reactions due to their catalytic behavior for numerous applications in various
areas of industrial microbiology and biotechnology. In particular, their catalytic
behavior has been very effective for the esterification of saccharide and fatty acids.
Various parameters, such as solubility of saccharide in reaction media, control of
water activity and content, reaction temperature, nature of acyl donor, and substrate
ratio, have been varied in a systematic fashion to improve reaction conversion yields
and rates. Although there are still difficulties for their employment at an industrial
scale and in establishing cost-effective scale-up and downstream processing protocols,
additional research will yield improvements in these area and more widespread use of
the biocatalytic approach is anticipated.

Acknowledgment
The authors thank the U.S. Department of Agriculture, Grant 2006-35504-1 7262,
for supporting their research in the bio-based surfactant area.

346 0 5-H.

Pyo and D.G. Hayes

References
Akoh, C.C.; L.N. Mutua. Synthesis of Alkyl Glucoside Fatty Acid Esters: Effect of Reaction Parameters and Incorporation of n-3 Polyunsaturated Fatty Acids. Enzyme Microb. Techno1 1994,
16, 115-119.
Allen, D.; B. Tao. Carbohydrate-Alkyl Ester Derivatives as Biosurfactants. /. Sufact. Deters 1999,
2, 383.
Ballesteros, A.; F.J. Plou; M. Alcalde; M. Ferrer; H. Garcia-Arellano; D. Reyes-Duarte; I. Ghazi.
Enzymatic Synthesis of Sugar Esters and Oligosaccharides from Renewable resources. In Biocatalysis in the Pharmaceutical and Biotechnology Indwtries, Patel, R.N., Ed.; CRC Press: Boca
Raton, FL, 2007,463-488.
Bjorkling, F,; S.E. Godtfredsen; 0. Kirk. A Highly Selective Enzyme-Catalysed Esterification of
Simple Glucosides. J. Chem. Soc. Chem. Commun. 1989, 934-935.
Cao, L.; U.T. Bornscheuer; R.D. Schmid. Lipase-catalyzed solid-phase synthesis of sugar esters.
Influence of immobilization on productivity and stability of the enzyme. J. Mol. Catal. B: Enz.
1999,6,279-285.
Cao, L.; U.T. Bornscheuer; R.D. Schmid. Lipase-Catalyzed Solid Phase Synthesis of Sugar Esters.
Fett. Lipid 1996, 98, 332-335.
Carta, G.; J.L. Gainer; A.H. Benton. Enzymatic Synthesis of Esters Using an Immobilized Lipase.
Biotechnol. Bioeng. 1991, 37, 1004-1009.
Cauglia, F.; l? Canepa. The Enzymatic Synthesis of Glucosylmyristate as a Reaction Model for
General Considerations on Sugar Esters Production. Bioresource 2chnol. 2008, 99, 4065-4072.
Chamouleau, F.; D. Coulon; M. Girardin; M. Ghoul. Influence of Water Activity and Water Content on Sugar Esters Lipase-Catalyzed Synthesis in Organic Media. J. Mol. Catal. B: Enz. 2001,
1 1 , 949-954.
Chittenden, G.J.F. Isopropylidenation of D-Mannitol under Neutral Condition. Carbohyd. Res.
1980, 87, 219-226.
Compton, D.L.; K.E. Vermillion; J.A. Laszlo. Acyl Migration Kinetics of 2-Monoacylglycerols
from Soybean Oil via H NMR. J. Amer. Oil Chem. Soc. 2007, 84,343-348.
Chopineau, J.; ED. McCaffery; M. Therisod; A.M. Klibanov. Production of Biosurfactants from
Sugar Alcohols and Vegetable Oils Catalyzed by Lipases in a Non Aqueous Medium. Biotechnol.
Bioeng. 1998, 31, 208-214.
Coulon, D.; M. Girardin; M. Ghoul. Enzymic Synthesis of Fructose Monooleate in a Reduced
Pressure Pilot Scale Reactor Using Various Acyl Donors. Process Biochem. 1999, 34, 913-918.
Csbka, G.; S . Marton; R. Zelko; N. Otomo; I. Antal. Application of Sucrose Fatty Acid Esters in
Transdermal Therapeutic Systems. Eur. J. Pharrn. Biopharm. 2007,65,233-237.
Dang, H.T.; 0.0.Obiri; D.G. Hayes. Feed Batch Addition of Saccharide During Sacharide-Fatty
Acid Esterification Catalyzed by Immobilized Lipase: Time Course, Water Activity, and Kinetic
Model. J. Am. Oil Chem. Sor. 2 0 0 5 8 2 , 4 8 7 - 4 9 3 .
Degn, l?; W. Zimmermann. Optimization of Carbohydrate Fatty Acid Ester Synthesis in Organic
Media by a Lipase from Candida antarctica. Biotechnol. Bioeng. 2001, 7 4 , 483-491.
Ferrer, M.; M.A. Cruces; M. Bernabt; A. Ballesteros; F.J. Plou. Lipase-Catalyzed Regioselective

Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 347

Acylation of Sucrose in Two-Solvent Mixtures. Biotechnol. Bioeng. 1999, 65, 10-16.

Ferrer, M.; J. Soliveri; F.J. Plou; N. L6pa-Cortbs; D. Reyes-Duarte; M. Christensen; J.L. CopaPatiho; A. Ballesteros. Synthesis of Sugar Esters in Solvent Mixtures by Lipases from fiermomyces knuginosus and Candida antarctica B, and Their Antimicrobial Properties. Enzyme Microb.
Technol. 2005, 36, 391-398.
Ferrier, R.J. Carbohydrate Boronates. Adv. Carbohydr. Chem. Biochem. 1978, 35, 31-80.
Feuge, R.O.; H.J. Zeringue; Jr., T.J. Weiss; M. Brown. Preparation of Sucrose Esters by Interesterification. J. Am. Oil Chem. SOC.1970,47, 56-60.
Foresti, M.L.; M. Pedernera; V. Bucala'; M.L. Ferreira. Multiple Effects of Water on Solvent-Free
Enzymatic Esterifications. Enzyme Microb. Technol. 2007,4 1, 62-70.
Fregapane, G.; D.B. Sarney; E.N. Vulfson. Enzymic Solvent-Free Synthesis of Sugar Acetal Fatty
Acid Esters. Enzyme Microb. Technol. 1991, 13, 796-800.
Freemantle, M. New Horizons for Ionic Liquids. Chem. Eng. News 2001,79, 21-25.
Ganske, F.; U.T. Bornscheuer. Optimization of Lipase-Catalyzed Glucose Fatty Acid Ester
Synthesis in a Two-Phase System Containing Ionic Liquids and t-BuOH. J. Mol. Catal. B:
Enzyme 2005a, 36,4042.
Ganske, F; U.T. Bornscheuer. Lipase-Catalyzed Glucose Fatty Acid Ester Synthesis in Ionic Liquids. 0%.Lett. 2005b, 7, 3097-3098.
Guillemard, E.; M. Geniteaulegendre; B. Mabboux; I. Poilane; R. Kergot G. Lemaire, et al.
Antiviral Action of Trehalose Dimycolate Against EMC Virus: Role of Macrophases Interferon
AlphdBeta. Antiviral Res. 1993, 20, 201-213.
Habulin, M.; S. Sabeder; 2. K n a . Enzymatic Synthesis of Sugar Fatty Acid Esters in Organic
Solvent and in Supercritical Carbon Dioxide and Their Antimicrobial Activity. /. Supercrit.
Fluids 2008,45, 338-345.
Hayes, D.G. Enzyme-Catalyzed Modification of Oilseed Materials to Produce Eco-Friendly Products./. Am. OilChem. SOC.2004,81, 1077-1103.
Hoydonckx, H.E.; D.E. De Vos; S.A. Chavan; PA. Jacobs. Esterification and Transesterification of
Renewable Chemicals. Top. Catal. 2004, 27, 83-96.
Ikeda, I.; A.M. Klibanov. Lipase-Catalyzed Acylation of Sugars Solubilized in Hydrophobic Solvents by Complexation. Biotechnol. Bioeng. 1993,42,788-79 1.

Iso, M.; B. Chen; M. Eguchi; T. Kudo; S. Shrestha. Production of Biodiesel Fuel from Triglycerides and Alcohol Using Immobilized Lipase.J Mol. Catal. B: Enz. 2001, 16, 53-58.
Janssen, A.E.M.; C. Klabbers; M.C.R. Franssen; V.K. Riet, Enzymatic Synthesis of Carbohydrates
Esters in Aqueous Media. Enzyme Microb. Technol. 1991, 13, 565-572.
Kennedy, J.F.; H. Kumar; PS. Panesar; S.S. Marwaha; R. Goyal; A. Parmar; S. Kaur. EnzymeCatalyzed Regioselective Synthesis of Sugar Esters and Related Compounds. /. Chem. Echnol.
Biotechnol. 2006, 81, 866-876.
Khaled, N.; D. Montet; H. Pina; J. Graille. Fructose Oleate Synthesis in a Fixed Bed Catalyst Bed
Reactor. Biotechnol. Lett. 1991, 13, 167-172.
Kim, J.E.; J.J. Han; J.H. Yoon; J.S. Rhee. Effect of Salt Hydrate Pair on Lipase-Catalyzed Regioselective Monoacylation of Sucrose. Biotechnol. Bioeng. 1998, 57, 121-125.

348 0 5-H.

Pyo and D.G. Hayes

Kim, J.; S. Haarn; D.-W. Park; I.-S. Ahn; T. G. Lee; H.-S. Kim; W.-S. Kim. Biocatalytic Esterification of P-Methylglucoside for Synthesis of Biocompatible Sugar-Containing Vinyl Esters. Cbem.
Eng. J. 2004,99, 15-22.
Kohya, H.; F. Ishii; S. Takano; T Katori; T. Ebina; N. Ishida. Antitumor Effect of a Synthetic
Cord Factor, 6,6-Di-0 Decanoyl-a, a-Trehalose (SS 554) in Mice. ]pn. J. Cancer Res. 1986,
77,602-609.
Ku, M.A.; Y.D. Hang. Enzymatic Synthesis of Esters in Organic Medium with Lipase from Byssocblamysfilva. Biotecbnol. Lett. 1995, 17, 1081-1084.
Laane, C.; K.V. Boeren; C. Veeger. Rules for Optimization of Biocatalysis in Organic Solvent.
Biotecbnol. Bioeng. 1987,30, 81-87.
Lee, S.H.; S.H. Ha; N.M. Hiep; W.-J. Chang; Y.-M. Koo. Lipase-Catalyzed Synthesis of Glucose
Fatty Acid Ester Using Ionic Liquids Mixtures. J. Biotecbnol. 2008a, 133, 486-489.
Lee, S.H.; D.T. Dang; S.H. Ha; W.-J. Chang; Y.-M. Koo. Lipase-Catalyzed Synthesis of Fatty
Acid Sugar Ester Using Extremely Supersaturated Sugar Solution in Ionic Liquids. Biotecbnol.
Bioeng. 2008b,99, 1-8.
Manjbn, A.; J.L. Iborra; A. Arocas. Short-Chain Flavour Ester Synthesis by Immobilised Lipase in
Organic Media. Biotecbnol. Lett. 1991, 13, 339-344.
Mutua, L.N.; C.C. Akoh. Synthesis of Alkyl Glycoside Fatty Acid Esters in Non-Aqueous Media
by Candida sp. Lipase. J. Am. Oil Cbem. Soc. 1993,70,43-46.
Nakarnura, S. Using Sucrose Esters as Food Additives. INFORM 1997, 8, 866-874.
Napier, I?E.; H.M. Lacerda; C.M. Rosell; R.H. Valivety; A.M. Vaidya; PJ. Halling. Enhanced
Organic-Phase Enzymatic Esterification with Continuous Water Removal in a Controlled AirBleed Evacuated-Headspace Reactor. Biotecbnol. Prog. 1996, 12,47-50.
Ntawukulilyayo, D.; C. Demuynck; J.I? Remon. Microcrystalline Cellulose-Sucrose Esters as
Tablet Matrix Forming Agents. h t . J. Pbarm. 1995, 121, 205-221.
Ntawukulilyayo, J.D.; S.C. De Smedt; J. Demeester; J.l? Remon. Stabilisation of Suspensions Using Sucrose Esters and Low Substituted n-Octenylsuccinate Starch-Xanthan Gum Associations.
Int. J. Pbarm. 1996, 128,73-79.
Oguntimein, G.B.; H. Erdmann; R.D. Schmid. Lipase Catalysed Synthesis of Sugar Ester in
Organic Solvents. Biotecbnol. Lett. 1993, 15, 175-180.
Park, O.J.; H.G. Park; J.W. Yang. Enzymatic trans-Esterification of Monosaccharides and Amino
Acid Esters in Organic Solvents. Biotecbnol. Lett. 1996, 18, 473-478.
P a d , I?; A. Guerrero; M. Berjano; C. Gallegos. Transient Flow of O W Sucrose Palmitate Emulsions. J. Food Eng. 1999,4 1, 33-4 1.
Pinna, M.C.; A. Salis; M. Monduzzi. Novel Mannitol Based Non-Ionic Surfactants from Biocatalysis Part Two: Improved Synthesis. J. Mol. Catal. B: Enz. 2004, 27,233-236.
Pyo, S.-H.; D.G. Hayes. Desorption of Fructose from a Packed Column to an Oleic AcidFructose Oleate Mixture for Employment in a Bioreactor System. J. Am. Oil Cbem. SOC.2008, 85,
1033-1 040.
Rich, J.O.; B.A. Bedell; J.S. Dordick. Controlling Enzyme-Catalyzed Regioselectivity in Sugar
Ester Synthesis. Biotecbnol. Bioeng. 1995,45, 426-434.

Synthesis of Saccharide Fatty Acid Ester Biosurfactants Catalyzed by Lipase 0 349

Riva, S.; J. Chopineau; A.P.G. Kiebomm; A.M. Klibanov. Selective Enzymatic Acylations and
Deacylations of Carbohydrates and Related Compounds. /. Am. Chem. SOC.1988, 110,
584-589.
Rocha, J.M.S.; M.H Gil; F.A.I? Garcia. Optimisation of the Enzymatic Synthesis of n-Octyl
OIeate with Immobilised Lipase in the Absence of Solvents. /. Chem. Technol. Biotechnol. 1999,
74,607-612.
Sabeder, S.; M. Habulin; 2. Knez. Lipase-Catalyzed Synthesis of Fatty Acid Fructose Esters. /.
Food Eng. 2006, 77, 880-886.
Sakaki, K.; A. Aoyama; T. Nakane; T. Ikegami; H. Negishi; K. Watanabe; H.,Yanagishita. Enzymatic Synthesis of Sugar Esters in Organic Solvent Coupled with Pervaporation. Desalination
2006, 193, 260-266.
Sarney, D.B.; E.N. Vulfson. Enzymatic Synthesis of Sugar Fatty Acid Esters in Solvent-Free Media. In Enzymes in Nonaqueous Solvents: Methods and Protocols (Method in Biotechnology GL.15),
Vulfson, E.N., PJ. Halling, H.L. Holland, Eds.; Humana Press: Totawa, NJ, 2001, 531-543.
Scheckermann, C.; A. Schlotterbeck; M. Schmidt; V. Wray; S. Lang. Enzymatic Monoacylation of
Fructose by Two Procedures. Enzyme Microb. Technol. 1995, 17, 157-162.
Schlotterbeck, A.; S. Lang; V. Wray; F. Wagner. Lipase-Catalyzed Monoacylation of Fructose.
Biotechnol. Lett. 1993, 15,61-64.
Sharma A.; S. Chattopadhyay. Lipase Catalysed Acetylation of Carbohydrates. Biotechnol. Lett.
1993, 15, 1145-1 146.
Sheldon, R.A.; R.M. Lau; M.J. Sorgedrager; F. van Rantwijk; K.R. Seddon. Biocatalysis in Ionic
Liquids. Green Chem. 2002,4, 147-1 5 1 .
Shibatani, S.; M. Kitagawa; Y. Tokiwa. Enzymatic Synthesis ofvinyl Sugar Ester in Dimethylformamide. Enzyme Microb. Technol. 1997, 19, 5 1 1-5 14.
Sin, Y.M.; K.W. Cho; T.H. Lee. Synthesis of Fructose Esters by Pseudomonm sp. Lipase in Anhydrous Pyridine. Biotechnol. Lett. 1998, 20, 91-94.
Szuts, A.; E. Pallagi; G. Regdon; Jr., 2. Aigner; I? Szabo-Revesz. Study of Thermal Behaviour of
Sugar Esters. Int. J. Pharm. 2007, 336, 199-207.
Tarahomjoo, S.; I. Alemzadeh. Surfactant Production by an Enzymatic Method. Enzyme Microb.
Echnol. 2003, 33, 33-37.
Therisod, M.; A.M. Klibanov. Facile Enzymic Preparation of Monoacylated Sugars in Pyridine. J.
Am. Chem. SOC.1986, 108, 5638-5640.
Therisod, M.; AM. Klibanov. Regioselective Acylation of Secondary Hydroxyl Groups in Sugars
Catalyzed by Lipases in Organic Solvents. /. Am. Chem. SOC.1987, 109, 3977-398 1 .
Thevenin, M.A.; J.L. Grossiord; M.C. Poelman. Sucrose EsterslCosurfactant Microemulsion
Systems for Transdermal Delivery: Assessment of Bicontinuous Structures. Int. /. Pharm. 1336,
137, 177-186.
TriantaFyllou, A.O.; E. Wehtje; I? Adlercreutz; B. Mattiasson. Effects of Sorbitol Addition on the
Action of Free and Immobilized Hydrolytic Enzymes in Organic Media. Biotechnol. Bioeng.
1995,45,406-414.
Tsavas, I?; S. Polydorou; I. Faflia; E.C. Voutsas; D. Tassios; M.V. Flores; K. Naraghi; I?J. Halling;

350 0 5-H. Pyo and D.G. Hayes

F. Chamouleau; M. Ghoul; et al. Solubility of Glucose in Mixtures Containing 2-Methyl-2Butanol, Dimethyl Sulfoxide, Acids, Esters, and Water.]. Chem. Eng. Data 2002, 47, 807-810.
Tsitsimpikou, C.; H. Stamatis;V. Sereti; H. Dafos; EN. Kolisis. Acylation of Glucose Catalysed
by Lipases in Supercritical Carbon Dioxide. ]. Chem. Technol. Biotechnol. 1998,71, 309-314.
Villeneuve, I! Lipases in Lipophilization Reactions. Biotechnol. Adv. 2007, 25, 5 15-536.
Vulfson, E. Enzymatic Synthesis of Surfactants. In Novel Surfactants, Holmberg, K., Ed.; MarcelDekker: New York, 2003, 257-278.
Ward, 0.P; J. Fang; Z. Li. Lipase-Catalyzed Synthesis of Sugar Ester Containing Arachidonic
Acid. Enzyme Microb. Zchnol. 1997,20, 52-56.
Woods, L.F.J.; S.J. Swinton. Enzymes in the Starch and Sugar Industries. In Enzymes in Food
Processing, Tucker, G.A., L.F.J. Woods, Eds.; Blackie: Glasgow, 1991,221-238.
Yahya, ARM.; W.A. Anderson; M., Moo-Young. Ester Synthesis in Lipase-Catalyzed Reactions.
Enzyme Microb. Zchnol. 1998, 23,438450.
Yadav, G.D.; S.L. Piyush. Kinetics and Mechanism of Synthesis of Butyl Isobutyrate Over Immobilised Lipases. Biochem. Eng. ]. 2003, 16,245-25.
Yan, Y.; U.T. Bornscheuer; L.U. Cao; R.D. Schmid, Lipase-CatalyzedSolid-Phase Synthesis of
Sugar Fatty Acid Esters Removal of Byproducts by heotropic Distillation. Enzyme Microb.
Technol. 1999, 25,725-728.
Yan, Y.; U.T.Bornscheuer; G. Stadler; S. Lutz-Wahl; M. Reuss; R.D. Schmid. Production of Sugar
Fatty Acid Esters by Enzymatic Esterification in a Stirred-Tank Membrane Reactor: Optimization of Parameters by Response Surface Methology.]. Am. Oil Chem. SOC.2001, 78, 147-1 52.
Yu, J.; J. Zhang; A. Zhao; X. Ma. Study of Glucose Ester Synthesis by Immobilized Lipase from
Candida sp. Catal. Comm. 2008, 9, 1369-1374.
Zaks, A.; A.M. Klibanov. Enzymatic Catalysis in Nonaqueous Solvents.]. Biol. Chem. 1988, 263,
3194-3201.
Zhang, X.; D.G. Hayes. Increased Rate of Lipase-CatalyzedSaccharide-Fatty Acid Esterification
by Control of Reaction Medium. ]. Am. Oil Chem. SOC.1999,76, 1495-1 500.

Synthesis, Aggregation Properties,

and Applications of Biosurfactants
Derived from Arginine
MaRosa Infante, Lourdes Perez, Carmen Morin,
Ramon Pons, and Aurora Pinazo
lnstiruto de QulmicaAvanzada de Catalunya.IQAC. CSlC. Jordi Girona 18-26.08034Barcelona,Spain

Introduction
Surfactants are amphiphilic molecules composed of a polar group (ionic or non-ionic)
and one or more hydrophobic chains (usually hydrocarbon). This duality confers a
unique range of properties to surfactants, which are used in a variety of processes.
These compounds have garnered ever increasing interest owing to their interfacial
activity, ability to self-aggregate into myriad supramolecular motifs on the nano scale,
biological activities, and diverse applications. Nonetheless, research and development
of new surfactants has shifted towards compounds with multifunctional benefits, and
must be carried out in accordance with acting regulations for human health and the
environment (Holmberg, 2003).
One of the key strategies to minimize the toxicity and environmental impact of
synthetic surfactants is to mimic natural surfactant structures, such as lipoaminoacids
or their analogs (i.e., surfactant-like peptides) all of which are found in cell membranes
(Epand et al., 1998;Yasuhara et al., 2005; Adams et al., 2007). Lipoaminoacids constitute
an important class of natural surface active bio-molecules of great interest to organic and
physical chemists as well as to biologists; these molecules have an unpredictable number
of basic and industrial applications (Xia & Nnanna, 200 1). Structurally, lipoaminoacids
are a very heterogeneous group of compounds but with a common advantage, in
that they are relatively easy to design and synthesize. Often these molecules combine
charged, or noncharged residues [i.e., glutamic acid (Glu), lysine (Lys), arginine (Arg),
serine (Ser), leucine (Leu); phenylalanine (Phe), and alanine (Ala)] as the hydrophilic
head group with an hydrophobic tail of different structure, length, and number [i.e.,
fatty acids, fatty alcohols, and fatty amines] as synthons for the amphiphilic structure
(Takehara 1989; Boyat et al., 2000; Zhang et al., 2005; Gerova et al., 2008; Vijay et al.,
2008). This fact explains the diversity of amino acid/peptide-based surfactants and the
variety of their physicochemical and biological properties (Roy & Dey, 2006; Das et al.,
2006; Varka et al., 2006; Ohta et al., 2008; Capone et al., 2008).
351

352 0 M.R. Infante et al.

With the aim ofdeveloping biocompatible surfactants, our group, in collaboration

with others, has synthesized and studied new monodisperse and chiral lipoaminoacidtype surfactants ofdiverse structure and ionic character (Infante et al., 1985; Pinazo et
al., 1993; Allouch et al., 1996; Infante & Moses 1994; Infante et al., 1997; Castillo et
al., 2004). Of particular interest are the arginine-based cationic surfactants which have
been designed in accordance with three different amphiphilic structures: Single chain
or monocatenary arginine surfactants, one arginine residue bearing one hydrophobic
tail (Fig. 13.1, 1); gemini arginine surfactants, two arginine polar heads and two
hydrophobic tails per molecule linked by a spacer chain (Fig. 13.1,2); and glycerolipid
arginine surfactants, one arginine polar head and one or two hydrophobic moieties
linked together through a glycerol skeleton (Fig. 13.1, 3). These three structures are
characterized by the presence of weak amide and/or ester bonds anywhere in the
molecule. (Infante et al., 1984; Pirez et al., 1996; Clap& et al., 1999; Mordn et al.,
2004a).

. .. ....
........_
arginine
\,___..
...............
'... .
........

............................
...... _,..

._-

..I

.........

... ....

__

.....____
areinin:

____

v.'

......................I ......._. ...

!

.....

arginine

........- .
.......

........

../

.........

'

_--

C
0
0

-\

-i

arginine ,

- -

__

c o
d

,
3

Fig. 13.1. Schematic structure of arginine-based surfactants. Reproduced from (Moran et a1.2004~)with
permission of the Royal Society of Chemistry (RSC).

Synthesis,Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 353

In terms of surfactant behavior, the authors have evaluated and compared their
physicochemical properties of adsorption and self aggregation in aqueous media at a
range of concentrations and in the presence and absence of other components.
The presence of the positive charged arginine in such amphiphilic structures gives
an extensive series of compounds with rich phase behavior properties and strong
antimicrobial activity (Mordn et al., 2004b).
These compounds are water soluble, nontoxic if orally administered, nonirritating,
biodegradable, and have a minimal aquatic impact, thus guaranteeing their ultimate
commercial development in the food and cosmetic sector and highlighting their
potential for biochemical applications (Pkrez et al., 2002; Benavides et al., 2004;
Pkrez et al; 2005; Martinez et al., 2006). They can be regarded as an alternative to
conventional synthetic surfactants in which fundamental requirements for industrial
development are present: biodegradability, low toxicity, multifunctional performance,
and renewable sources of raw materials. The authors' work in this field contains basic
science as well as technology elements, and has been of the most interdisciplinary
nature possible, drawing on the expertise of researchers from chemical synthesis,
biocatalysis, physical chemistry of colloids and surfaces, ecotoxicity, toxicology,
microbiology, and has relied on numerous industrial collaborators. In this chapter
the authors present and analyse the data obtained from our group on the synthesis,
physicochemical properties, and some potential applications of these three types
of biosurfactants derived from arginine with an emphasis on the self-aggregation
properties.

Synthesis
Monocatenary Arginine Surfactants
Amino acids are linked to long aliphatic chains through the a-amino, a - C O O H ,
or side chain groups. Thus, fatty acids or alkyl halides can react with amino groups
yielding the corresponding N-acyl and N-alkyl derivatives, respectively. Alternatively,
the carboxyl group of the amino acid can be condensed with alkyl amines or aliphatic
alcohols to give N-alkyl amides or 0-alkyl esters. Among the different types of linkages
between the long aliphatic chain and the amino acid, the N a -acyl, 0 a - alkyl esters,
and 0" -alkyl amides of arginine have attracted much interest in order to prepare
low toxic, biodegradable antimicrobial cationic surfactants. They selectively disrupt
bacterial membranes at submicellar concentrations, but not erythrocytes or skin cell
membranes. It has been demonstrated that the incorporation of ester functionality
accelerates biodegradation. (Infante et al., 1984, 1992a; Mordn et al., 2001 b; Seguer
et al., 1994).
Thus Nu-acyl-arginine-methyl ester hydrochloride (Fig. 13.2), arginine-Oa-alkyl
ester dihydrochloride (Fig. 13.3) and arginine-Oa-alkyl amide dihydrochloride (Fig.
13.4) salts of different alkyl chain lengths have been synthesized by our group using
chemical and biotechnological methodologies. Fig. 13.2-1 3.4 indicate the structure
and acronyms of the different homologs for each series.

354

M.R.Infante et a\.

H2N
Fig. 13.2. Nn-acyl-arginine-methylester hydrochloride.CAM n = 8; LAM n = 1 0 M A M n = 12; PAM n = 14.

C1'

Fig. 13.3. Arginine-0 a-alkyl ester dihydrochloride.AOE: n = 6; ACE n = 8; ALE: n = 10.

Synthesis,Aggregation Properties, and Applications of BiosurfactantsDerived from Arginine 0 355

H2N
Fig. 13.4. Arginine-0 a-alkyl amide dihydrochloride.ACA n = 8; ALA n = 10; AMA n = 12.

Nu-acyl-arginine-methyl ester hydrochloride salts (CAM, LAM, MAM, and

PAM) were prepared by Nu-acylation of the amino terminal arginine. Fatty acids
were condensed to arginine methyl ester hydrochloride using classical chemical
methods (Infante et al., 1997). The application of biotechnological procedures was
not efficient for these compounds (ClapCs & Infante, 2002). Hovever, papain from
Carica papaya latex immobilized by deposition onto Polyamide-6 or Celite-R63
support materials was found to be a suitable catalyst for the formation of amide and
ester bonds between N-benzyloxycarbonyl-Arg-OMe (Cbz-Arg-OMe) and various
long-chain fatty alcohols and amines to prepare in organic media arginine-0-alkyl
ester dihydrochloride salts (AOE, ACE, and ALE) (Fig. 13.3) and arginine-0-alkyl
amide dihydrochloride salts (ACA, ALA, and AMA) (Fig. 13.4), respectively (Clapis
et al., 1999). Changes in enzymatic activity and product yield were studied for the
following variables: organic solvent, aqueous buffer content, support for the enzyme
deposition, presence of additives, enzyme loading, substrate concentration, and
reaction temperature. The best yields (81-89%) of arginine N-alkyl amide derivatives
were obtained at 25C in acetonitrile with an aqueous buffer content ranging from 0
to I % (v/v) depending on the substrate concentration. The synthesis of arginine alkyl

356 0 M.R. Infante et al.

ester derivatives was carried out in solvent-free systems at 50 or 65C depending on
the fatty alcohol chain length. In this case, product yields ranging from 86 to 89%
were obtained with a molar ratio Cbz-Arg-OMe/Fdtty alcohol of 0.01. In all cases,
papain deposited onto polyamide gave the highest enzymatic activities and yields.
Under the best reaction conditions the syntheses were scaled up to produce 2 g
of final product. The crude products were loaded onto a preparative PrepPak column
filled with Delta-Pak C4 stationary phase (Waters, Milford, MA). Products were
eluted using CH,CN gradients (0.2% CH,CN/min) in 0.1% aqueous TFA. The flow
rate was 100 mL/min, and the products were detected at 220 nm. The pure fractions
were pooled and lyophilized in the presence of the calculated amount of HCl to
obtain the products in hydrochloride form. The overall yields, which include reaction,
N-Cbz deprotection, and purification by preparative HPLC, varied from 53 to 77%
of pure (99.9% by HPLC analysis) product.

Gemini A rginine Surfactants or bis(A rgs)

With the aim of obtaining surfactants that are environmentally acceptable and have a
high degree of performance, Pkrez first described the synthesis of a novel class ofgemini
cationic surfactants derived from the monocatenary long-chain N -acyl-arginine
derivatives: the bis (N a -acyl-L-arginine) a -a-polymethylenediamide dihydrochloride
salts named bis(Args) (Fig. 13.5) (Pirez et al., 1996). These compounds consist oftwo
symmetrical monocatenary Nu acyl-arginine structures of 8, 10, and 12 carbon atoms
(Fig. 13.5, II = 6,8,10), connected through the a-carboxylic groups of the arginine
residues by amide covalent bonds to an a -W diaminoalkane spacer chain of varying
length (Fig. 13.5, s = 1-10). This particular diaminoalkane spacer chain was chosen
to control the distance between the charged sites of the cation, which modifies the
inter- and intrahydrophilic-hydrophobicinteractions.
The method used for the first approach involves chemical protecting groups,
organic solvents, and chemical catalysts. Later, a strategy to reduce the environmental
impact was developed. To this end, a novel chemo-enzymatic synthesis of bis(Args)
was described (Piera et al., 2000). Nu acyl-L-arginine alkyl ester derivatives (Fig.
13.2) were the initial building blocks for the synthesis. The best strategy found
consisted of two steps: First, the quantitative acylation of one amino group of the a
o diaminoalkane spacer by the carboxylic ester of the N-a-acyl-arginine took place
spontaneously, at the melting point of the a, w diaminoalkane, in a solvent-free
system. The second step was the papain-catalyzed reaction between another N-a-acylarginine alkyl ester and the free aliphatic amino group of the derivative formed in
the first step. Reactions were carried out in solid-to-solid and solution systems using
low-toxic potential solvents. Changes in reaction performance and product yield were
studied for the following variables: organic solvent, support for enzyme deposition,
and substrate concentration. The best yields (70%) were achieved in solid-to-solid
systems and in ethanol at a water content of 0.80% (w/w) equivalent to water activity
of 0.07.

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 357

CH3

CH3

c1-

c1-

Fig. 13.5. Structure of bis(Args) gemini cationic surfactants Cs(XA)2 where s refers to the length of the
spacer chain, 2 refers to the presence of two symmetrical long-chain N a-acyl-arginine residues, X refers
tothelengthofthefattyacylchains;X=O,C,andLrefertofattyacylchainsoflength8(n=6), 10(n = 8 ) ,
and 12 (n = lo), respectively. C3 (OA)2 n = 6, s = 1; C3 (CAI2 n = 8,s = 1; C4 (CA)2: n = 8,s = 2; C6 (CA)2: n =
8 , ~ = 4 ; C 9 ( C A ) 2 n = 8 , ~ = 7 ; C l O ( C A ) Z : n = 8 , ~ =C3(LA)2n=
8;
1O,s= l;C4(LA)Z:n= 1O,s=2; C6(LA)2
n = 10,s =4;C9(LA)2 n = 10, s = 7; ClO(LA)2: n = lO,s=8C12(LA)2: n =8, s = 10.

Bis (Args) homologs of 8, 10, and 12 carbon atoms using 1,3-diaminopropane

and 1,3-diamino-2-hydroxy-propane
as hydrocarbon spacers were prepared at the 6-7
g level employing the methodology developed. Products were purified using cationexchange chromatography. The preparative procedure involves simple equipment,
low cost materials, and minimal amounts of solvent (watedethanol), with low toxicity
(Torres et al., 2001). The overall yields which include reaction and purification varied
from 5 1 - 65% of pure (97 - 98% by HPLC analysis) product.

Glycerolipid Arginine Surfactants

Amino acid glyceride conjugates (i.e., glycero amino acids) constitute a novel class of
lipoamino acids, which can be considered analogs of mono and diacylglycerides and
phospholipids. They consist of one or two aliphatic chains and one amino acid, as the
polar head, linked together through ester bonds to the glycerol backbone.
Our group has synthesized mono and diacyl glyceride derivatives from N-aacetyl-arginine (Fig. 13.6a,b) and arginine (Fig. 13.7a,b) using chemical and chemoenzymatic methodologies (Morin et al., 2001a, 2002; Ptirez et al., 2002a).

358 0 M.R. Infante et al.

The enzymatic preparation of mono- and diacyl glyceride acetyl-arginine esters
in Fig. 13.6 started with the preparation of the polar head 1-0-(Na-acetyl-arginyl)
glycerol, obtained by enzymatic methodology using hydrolases (Mordn et al., 200 1a).
Proteases and lipases were found to be a versatile catalyst for this reaction. A variety
of protected amino acid glyceryl ester derivatives were obtained in 4 6 9 8 % yield
under mild and selective conditions. In a second step, the free hydroxyl groups of the
glyceryl moiety were acylated with fatty acids using lipases as catalyst. The authors
have developed a novel methodology to obtain both 1-monoacyl- and 1,2-diacyl3-aminoacyl glycerol (Mordn et al., 2002, 2004a). Mono- and diacylation of amino
acid glyceryl ester may be carried out using selective lipases by taking advantage of the
spontaneous intramolecular acyl-migration reaction that occurs in partial glycerides.
Thus, the 1(3)-acylated product may undergo intramolecular 1(3)+2 acyl migration
and the resulting 1,2(2,3)-isomer subsequently can be acylated at the free primary
hydroxyl group by the lipase. Accordingly, the yield of diacylated product will depend
on both the rate of intramolecular acyl-migration and the enzymatic esterification
of the newly free primary hydroxyl of the monoacylated derivative. Both processes
are influenced by the reaction conditions, such as solvent, support for enzyme
immobilization, buffer salts, and by the amino acid glyceryl ester derivative. All the
enzymatic acylations were carried out in solvent-free media, at a temperature around
0

Fig. 13.6. a)l-0-acyl-rac-glycero-3-O-(Na-acetyl-L-arginine)hydrochloride salts.12ORAc: n = 10; 140RAc:

n = 12 .b) 1,2-di-O-acyl-rac-glycero-3-O-(Na-acetyl-L-arginine) hydrochloride salts. 1212RAc: n = 10;
1414RAc: n = 12.

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 359

the melting point of the corresponding fatty acid. The 1,2-diacyl-3-aminoacyl glycerol
derivatives were in fact a mixture of two regioisomers: 1,2-diacyl-racglycero-3-(amino
acid) derivative as the major one, and 1,3-diacyl-glycero-2-(aminoacid) derivative.
With this methodology a series of mono and dilauroylated glycerol derivatives
of acetyl-arginine, aspartic acid, glutamic acid, asparagine, glutamine, and tyrosine
were prepared. Purification of the amino acid glyceride conjugates was achieved
by preparative HPLC on a C4 reversed-phase column eluted with gradients of
acetonitrile in aqueous trifluoroacetic acid or with flash chromatography on silica.
Both methodologies provided similar recovery yields. Reversed-phase HPLC on a C4
column was better suited for ionic derivatives while silica was preferred for neutral
products. Both methodologies provided highly pure products.
?he Na-arginine compounds (Fig. 13.7a,b) were prepared using chemical
methodologies. The synthesis of these surfactants consists of three steps: Step 1
corresponded to the preparation of 1-0-(N-Cbz-L-arginyl) rac-glycerol monochloride
(OORZ) by chemical esterification of the a-carboxyl group of N-Cbz-L-arginine
oHC1 with the primary hydroxyl function of glycerol using boron trifluoroetherate as
catalyst. The overall reaction yield was 95%. Step 2 consisted of the synthesis of the
1-acyl-3-O-(N-Cbz-L-arginyl)rac-glycerol~HCl
(XORZ) and 1,2-diacyl-3-O-(N-CbzL-arginyl)rac-glycerol.HCl (XXRZ) by acylation of one hydroxyl group of (OORZ)

II

CI-

Fig. 13.7 a) l-O-acyl-rac-glycero-3-O-(Na-L-arginine)dihydrochloridesalts.100R n = 8; 12OR n = 10;

140R n = 12. b) 1,2-di-O-acyl-rac-glycero-3-O-(Na-L-arginine) dihydrochloride salts.88R n = 6; 1010R n
=8;1212Rn=12;1414Rn=12.

360 0 M.R. Infante et al.

with the corresponding long-chain acid chloride. A 1 :2 (OORZ/acid chloride) ratio
was used to obtain the monoglyceride derivatives and 1:3 ratio to obtain the diacylated
products. Using acetonitrile, sodium bicarbonate as the base, and molecular sieves
to remove the water from the medium, a 92% conversion of the starting reagent
was obtained. The last synthetic step to obtain the target compounds XOR and XXR
consists of a catalytic hydrogenation of the Cbz group using Pd over charcoal. The
reaction was carried out controlling the p H to prevent the hydrolysis of the ester
linkages present in these compounds. Pure compounds were obtained after several
crystallizations in methanoVacetonitrile.
All three types of arginine-based surfactants, monocatenaries, gemini, and glycerol
type, were prepared at multigram scale with a purity higher than 97% by HPLC
analysis, allowing us to perform a systematic study of their properties.

Physicochemical Properties
Molecular self-assembly of biomimetic molecules has recently attracted considerable
attention for its use in the design and fabrication of advanced biocompatible materials
with a wide range of applications in nanotechnology, medicine, and drug delivery
systems (Gorbitz, 2007). The aggregation morphologies of the acyl amino acids in
water and especially the effect of the molecular structure of the amphiphile on the
formation and type ofself-aggregatesis a fascinating task. It has been extensivelystudied
by different groups, particularly during the last fav years, for their potential uses in
advanced industrial technologies (Imae et al., 2000;Yamashita et al., 2007; Roy &
Dey, 2007;Oshimura et al., 2007; Gerova et al., 2008, Lee et al., 2008). In this
section the authors review the surface and aggregation properties of the three types of
arginine-based surfactants whose acronyms are provided in Table 13.1,

Monocatenaries from Arginine

The micellization properties in water solutions of the hydrochloride salts of Nu -acylarginine-methyl esters (Fig. 13.2),arginine-Oa-alkyl esters (Fig. 13.3),and arginineO'-alkyl amides (Fig. 13.4)has been studied by our group in collaboration with
others. Notice that compounds in Fig. 13.3 and 13.4 have two positive charges
in the hydrophilic moiety, one in the primary amino group and the other in the
guanidine group whereas the hydrochloride salts of Nu-acyl-arginine-methyl ester in
Fig. 13.2 has only one positive charge in the guanidine group at pH 7. The early
studies of the micellization process of these cationic surfactants involved surface
tension measurements with the dual purpose of investigating their behavior at the airsolution interfice and determining the critical micelle concentration (CMC) values
(Morbn et al., 2OO1b; Pks, 1993). Surface tension as a function of concentration
was measured using a Model K-12 tensiometer with a Wilhelmy plate from Kruss
(Hamburg, Germany). From the surface tension/concentration curves at 25 O C the
CMC and surface tension at the CMC),,y,(
were determined. Using the Gibbs
adsorption isotherm, the saturation adsorption (Tm)at the air water interface and the
area per molecule (A,) were obtained (Rosen, 1988).

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derivedfrom Arginine 0 361

Table 13.1. Acronyms and Chemical Names of the Three Types of Arginine-Based
Surfactants Described in this Chapter
Acronym

Chemical Name

Figure

CAM
LAM
MAM
PAM

N a-caproyl-arginine-methylester.HCI (n = 8)
N a-lauroyl-arginine-methylester.HCI (n= 10)
N "-myristoyl-arginine-methyl ester.HC1 (n = 12)
N "-palmitoyl-arginine-methylester.HCI (n = 14)

13.2

AOE
ACE
ALE

Arginine-O"-octyl ester.2HCI (n = 6)
Arginine-O"-caprylester.2HCI (n= 8)
Arginine-0" -1auryl ester. 2HCI (n = 10)

13.3

ACA
ALA
AMA

Arginine-0" -capryl amide. 2HCI (n = 8)

Arginine-0" -1auryl amide. 2HCl (n = 10)
Arginine-0"- myristoyl amide. 2HCl (n = 12)

13.4

'C,(XA),
C,(XA),
C,(XA),
C,(XA),
120RAc
140RAc
1212RAc
1414RAc

Bis (N "-acyl-arginine)a-w propylene diamide. 2HCl (s = 1)

Bis((N"-acyl-arginine) a-w butylene diamide. 2HCl (s = 2)
Bis (N "-acyl-arginine)a-w hexylene diamide. 2HCI (s= 4)
Bis (N "-acyl-arginine)a-w nonylene diamide. 2HCI (s = 7)
1-0-Iauroyl-glycero-3-0-(Na-acetyl-arginine).HCI(n = 10)
1-0-myristoyl -glycero-3-0-(Na-acetyl-arginine).HCI (n = 12)
1,2-di-0-lauroyl-glycero-3-0-(Na-acetyl-arginine).HCl (n = 10)
1,2-di-O- myristoyl -glycero-3-0-(No-acetyl-arginine). HCI (n = 12)

1OOR
12OR
140R

1-0-decyl-glycero-3-0-(Na-arginine).2HCI (n = 8)
1-0-lauroyl-glycero-3-0-(Na-arginine).2HCl (n = 10)
1-0- myristoyl -glycero-3-0-(Na-arginine).2HCI(n = 12)

13.5

13.6a
13.6b
13.7a

1,2-di-O-octyl-glycer0-3-O-(N~-arginine).2HCI(n = 6)
88R
1010R
1,2-di-O-decyl-glycer0-3-0-(N"-arginine) 2HCI (n = 8)
13.7b
1212R
1,2-di-O-lauroyl-glycero-3-0-(N"-arginine)2HCI (n = 10)
1414R
1,2-di-0- myristoyl -glycero-3-0-(Na-arginine)2HCl (n = 14)
X=O, C and L refer to fatty acyl chains of length 8 )n=6), 10 (n=8), and 12 (n=lO),
respectively

As expected, the surfactant activity of all these compounds was similar to those
with conventional quaternary cations, CMC values in the range 1-30 mM and ycMc
values being in the range 30-37 mN/m. The CMC values of the three families of
monocatenary surfactants depend of the alkyl chain length; the more hydrophobic
the molecule, the lower the CMC value (Fig. 13.8). Contrarily, the nature of the
hydrophilic group does not significantly affect the CMC and ycMc.
The effect of the hydrophilic group on Amis most interesting. The authors found
that the Am values for the arginine-Oa-alkyl ester dihydrochlorides (Fig. 13.3) and
arginine-Oa-alkyl amides dihydrochloride salts (Fig. 13.4) (0.62-1.14 nm', and
0.96- 1.22 nm'. respectively) were higher than that for Na -acyl-arginine-methyl
ester compounds with the same alkyl chain length (0.67-0.62 nm') (Mordn et al.,
2001 b). These results indicate that the arginine-Oa-alkyl ester and arginine-Oa-alkyl
amides structures are packed less densely at the interface. The two charged groups
contained in their molecular structure tend to spread them out on the interface due
to electrostatic repulsion forces.

362 0 M.R. Infante et al.

Arginyl amides
Arginyl esters
P q i arginines

Carbon number
Fig. 13.8. Relationship between CMC and the alkyl chain length for the arginine-alkyl amides (Fig. 13.4),
arginine-alkyl esters (Fig. 13.31, and acyl-arginine (Fig. 13.2) surfactants. CMC were determined in water
at 25C using Wilhelmy plate technique. Taken from MorBn et al., 2001 b. (Reprinted with permission
from ACS Press)

A first study on self-aggregation of LAM, the hydrochloride salt of N" -1auroyl

arginine methyl ester (Fig. 13.2) at different concentrations was carried out by Talmon's
group (Weihs et d., 2005). They examined the microstructures appearing in LAM
solutions (typical of a single-tailed surfactant with a large head-group) by cryogenictemperature-transmission electron microscopy (cryo-TEM) to image microstructures
appearing as a function of the concentration. Data showed that LAh4 forms spheroidal
micelles at low concentration; as the concentration increases and approaches the
phase-transition concentration, elongated structures (cylindrical micelles) in an
ordered array appear (data not shown).
A deeper insight into the micellization process of the LAh4 has been carried out
using the Pulsed Grandient Spin Echo NMR (PGSE-NMR) experiment and small
angle x-ray scattering (SAXS) (Pirez et al., 2007). The best form factor-structure factor
model fits for the S A X S scattering patterns were obtained for a form factor based on
polydisperse spheres with a core-shell structure. The structure factor was modelled
according to the rescaled mean spherical approximation model (Hansen et al.. 1982).
The main parameters which characterize the micelle shape (i.e., aggregation numbers

Synthesis, Aggregation Properties, and Applications of BiosurfactantsDerivedfrom Arginine 0 363

and geometrical dimensions) have been evaluated from the analysis of the SAXS data
of LAM solutions at different concentrations above the CMC (Table 13.2).
The area per molecule value, 0.65 f 0.01 nm2, agrees roughly with those obtained
from surface tension. The polar head and hydrophobic core electron densities are
consistent with the molecular structure of the surfactant. Moreover, hydration per
molecule obtained from the fits, 27 5 moles of water per mole of surfactant, is also
reasonable because in the molecular structure of the head group several hydrogen
bonding groups are present, the guanidinium cationic group and the chloride anion.
The PGSE-NMR was used to determine the self-difision coefficients of LAM at
25C using standard procedures (Stilbs, 1987), Table 13.3.
Fitting the two-site exchange model (Lindman et al., 1984) to the diffusion/
concentration curve we calculated the CMC, the intradiffusion coefficient of free
monomers, Djcc,and that of the micellized molecules, D,,.

Table 13.2. Structural Parameters of Micelles Formed by N "-Lauroyl-Arginine-Methyl

Ester HCI, LAM, and bis(Args)Gemini Surfactants Dissolved in Water at 25C ' J
LAM
Parameters' (24.6 mM)

LAM
(49.2 mM)

LAM
(73.8 mM)

LAM
(98.4 mM)

C,(LA),
(12.1 mM)

C,(LA),
(11mM)

T. (nm)

0.87 f 0.02

0.93 f 0.02

0.94 f 0.02

1.03 f 0.02

1.04 f 0.02

0.83

R,(nm)

1.52fO.02

1.46f0.02

1.51 f0.02

1.47 f0.02

1.24f0.02

1.2

6.1 f0.2

PI

0.13fO.03

0.12k0.03

0.12f0.03

0.11 f0.03

Nw4
A_ (nm35

21.5f1

27.2 f 1

26.0 f 1

33.0 f 1.5

32.8 f 2

0.64fO.01

0.67f0.01

0.65 fO.O1

0.66fO.01

1.05 20.02

L (nm)

1.1

45f2
40*2
44f2
41 f 2
45 f 1
Nw6
'Nomenclature for surfactants provided in Table 13.1.
'The scattering models were obtained as developed in L. Perez et at., 2005. For the LAM
derivative a model of polydispersedspheres has been used as the form factor to fit the X-ray
data while for the Gemini surfactants the best fits have been obtained using a two shell
cylindrical model.The structure factor was modelled according to the rescaled mean spherical
approximation (RMSA) model (Hansen et at., 1982).
3T,, hydrophilic shell thickness; R, radius of inner hydrophobic core; L, Length of the cylinders;
PI, polydispersityindex; PI=<r>/r,,,-l for a Schultz distribution of R;, Nw,water molecules per
head group; A,,surface area per surfactant molecule; Nw aggregation number.
'Number of water molecules per head group are taken from the volume of the hydrophilic
shell, the aggregation number, the hydrophilic head group volume, and the volume of one
water molecule.
SThesurface area per surfactant molecule i s obtained from the surface of the hydrophobic core
divided by the aggregation number.
6Aggregationnumbers are calculated using the volume of the hydrophobic core and the
volume of the hydrophobic chain per molecule.
From Pirez et al., 2007.

364 0 M.R. Infante et al.

The CMC values obtained by NMR agree with those obtained previously by
surface tension and the structural micelar parameters determined by N M R are in
accordance with those calculated from the SAXS data.
Phase behavior, including structural characterization of the monocatenaries, Naacyl-arginine-methyl esters hydrochloride salts CAM, LAM, MAM, and PAM (Fig.
13.2), in binary watedsurfactant and ternary water/surfactant/co-surfactant systems
have been systematically investigated as a function of the alkyl chain length of the
surfactants, and co-surfactants. These studies were carried out using phase diagrams,
optical microscopy observations, light scattering, spectrofluorimetry, Fourier
transform PGSE-NMR, and dielectric spectroscopy (Solans et al., 1989,1990; Infante
et al., 1992b; Fordedal et al., 1993). Solans showed by polarized light microscopy
that lyotropic liquid crystals of hexagonal (CAM, LAM, MAM, PAM), cubic (LAM,
MAM, PAM), and lamellar (MAM, PAM) structure appear as a function of the chain
length along with the hexagonal liquid crystal the more favored structure in LAM
in binary systems that exceed the solubility limit of micelles (Krafft temperature) at

25C.
The phase behavior of monocatenary arginine surfactants in multicomponent
systems was also investigated. Thus, the ternary systems were studied using n-alkanols
of different lengths (C5-Cl6) with the four N-acyl arginine homologs. Generally
in all systems three monophasic regions were identified: micelles, reversed micelles,
and lamellar liquid crystals. Microemulsion formation in presence of hydrocarbon
Table 13.3. Critical Micelle Concentration (CMC), Free Monomer Diffusion Coefficient
(DJ,
Micelle Diffusion Coefficient (DmJ and Derived Parameters, Obtained from
PGS-NMR at 25 " C for N "-Lauroyl-Arginine-MethylEster HCI, LAM, and 6is(Args)Gemini
Surfactants
CMC
,,D, x 1010 D
,, x 1010 R(T,)*
'Rmon
Rrnic'
Compounds' (mM)
(m29)
(m2s1)
(nm)
(nm)
(nm)
LAM
4.3
3.6
0.90
0.53
0.55
2.2
48
CJLA),
0.3
2.6
0.59
0.67
0.77
3.4
88
C,(LA),
0.1
2.7
0.6 1
0.68
0.74
3.3
79
C,(LA),
0.2
2.2
0.53
0.69
0.89
3.8
116
' Nomenclature referred to Table 13.1.
Theoretical hydrodynamic radius of the monomer based on monomer volume (Vmon),which
was calculated from the molecular weight divided by the density and by Avogadro's number.
The radius was calculated from Vmn assuming spherical geometry. Rmonand Rmlc are the
average hydrodynamic radii of the monomer and micelle, respectively, calculated from Dmon
and Dmk,respectively, using the Stokes-Einsteinequation;
Aggregation number calculated by dividing the approximate volume of a micelle, 4/3 i~
Rm,:i
by the approximatevolume of a surfactant molecule plus its water molecules of hydration:
V, plus the volume per moleculefor D,O multiplied by 10, the approximatedegree of
hydration (Lindmanet al., 1984)
From Perez et al.. 2007.

Synthesis, Aggregation Properties,and Applicationsof Biosurfactants Derived from Arginine 0 365

components (hexadecane, squalane, and toluene) was also studied by Solans (Solans
et al., 1990; P&, 1993). More interestingly, reversed vesicles in a system containing
lecithin-LAM/squalene/water were also described by Kunieda as biocompatible
systems to be applied in cosmetic and pharmaceutical formulations (Kunieda et al.,
1992).
Cationic liposomes are now recognized as potent vehicles for the delivery of genes
and other nucleic acids to cells. Cationic liposomes are formed from either a double
chain cationic surfactant in water or from a combination of a single cationic and
single anionic amphiphile, (catanionic systems) which at a critical concentration and
ratio can form cationic vesicles (dispersions of a lamellar phase). Combinations of
different mixtures of ALA, the hydrochloride salt of argininyl O-lauryl amide (Fig.
13.4, n = lo), and LAM with different anionic surfactants in water were studied. The
ability of these systems to spontaneously form vesicles, cubosomes (dispersions of a
cubic liquid crystalline phase), and hexosomes (dispersed particles with a hexagonal
internal structure) has recently been described. (Rosa et al., 2006).

Gemini Surfactants: bis(Args)

Gemini surfactants (two polar headdtwo tails) appear to be better in certain important
properties than the corresponding and more conventional monomeric surfactants (one
polar head/one tail). They tend to have much lower CMC, can produce lower surface
tensions than monomeric surfactants at the same molar or mass concentrations, and
have better wetting properties (Zana & Xia, 2003). In this section, adsorption at the
aidwater interface and aggregation behavior of bis(Args) are reviewed.
The equilibrium surface tension of bis(Args): C, (LA),, C,(LA), C,(LA),, and
C,(LA), gemini surfactants with two lauroyl hydrophobic tails and a spacer chain of
3 , 4 , 6 , or 9 methylene groups (Fig. 13.5: n = 10; s = 1 , 2 , 4 , and 7 respectively), were
studied and compared to those of the monomeric surfactant LAM. Results show that
bis(Args) have a CMC (calculated from surface tension versus concentration plots)
two or three orders of magnitude smaller than that of LAM and that the minimum
tensions at the CMC are comparable (Pkrez et al., 1998; Pinazo et al., 1999). Use of
other methods, primarily ion activity (of Ck),conductivity, fluorescence using pyrene as
fluorophore, and NMR, show that the aggregates formed at surfactant concentrations
above the surface tension CMC, or first CMC (CMC,), are not traditional micelles
(Pirez et al., 1998; Pinazo et al., 1999, Table 13.3). These methods revealed a higher
second CMC (CMC,), beyond which traditional micelles capable of solubilizing
fluorophores and having substantial counterion binding were present. The minimal
observed surface tensions for LAM and bis(Args) surfactants were similar, indicating
similar surface densities of hydrocarbon chains and polar groups at the surface. The
surface area of bis(Args) showed a maximum of 0.6 nm2/molecule with a spacer carbon
number of 4-6 (Pirez et al., 1998).
The nonclassical behavior of these bis(Args) surfactants, that show two CMC, may
be due to the formation of two types of aggregates. These two types of aggregates may

366 0 M.R.Infante et al.

play a role in the solution behavior and the practical properties of these surfactants.
For practical applications an important consideration is the performance of bis(Args)
under dynamic conditions. The dynamic surface tension was studied under constant
area and pulsating area for three bis(Args): C,(LA),, C,(LA),, and C,(LA),, and
compared to that of their monomer, LAM (Pinazo et al., 2001). Below the CMC,,
adsorption is quite slow, but becomes faster for thegemini surfactants between the first
and second CMC, and even faster above the CMC,. The time parameter ty the time
needed for the surface tension to drop 50% from the initial value (i.e., tl% of pure
water) compared to the equilibrium value, decreases with increasing concentration of
each surfactant, and correlates best with the dimensionless concentration, c/CMC,,
or c/CMC for LAM. The tension equilibration slows down significantly at longer
times, with ty95 being much larger than tys0 compared to what would be expected
for diffusion-controlled adsorption. Dynamic surface tensions at pulsating area are
consistent with the above inferences.
For new surfactants, one of the most studied properties to determine the possible
application scope is the foaming power. Foam stability and break up depend on a
series of complex phenomena, including hydrodynamic liquid drainage, aqueous
film thinning, and bubble coalescence. These phenomena are affected by equilibrium
surface tension, the equilibrium surface equation of state, Gibbs elasticity, surface
charge, film disjoining pressure, and also complex surface deformation and dynamic
surface tension properties (Prudhomme, 1996; Langevin, 1999). When surfactant
solutions are shaken, a large surface area free of surfactant is created. Then, the
molecules adsorb from the bulk solution, and the foam film surface is covered by
monolayers which protect against film rupture.
The ability of C (LA), C6(LA),,and C9(LA)*surfactants and their monomeric
counterpart LAM to form stable foams was measured from the maximum foam height
with solutions confined in glass vials and shaken vigorously (Pinazo et al., 2001). The
data correlate best when plotted as dimensionless foam height, H/H,, where H, is the
maximum possible height, versus the dimensionless concentration C/CMC, for LAM
or C/CMC, for bis(Args). Gemini surfactants were more efficient foam stabilizers
than LAM.
The effect of the spacer length on the shape and size of the micelles formed by C,
(LA),, C6(LA),, and C9(LA), has been studied by SAXS and PGSE-NMR and their
geometries compared with those obtained for their single chain homolog LAM (Pkrez
et al., 2007). Figure 13.9 shows the self-diffusion coefficient versus concentration
curve for the C,(LA),.
From the PGSE-NMR data the CMC, the monomer diffusion coefficient (Dmon)
and the micelle diffusion coefficient (DmiJwere obtained (Table 13.3). Deuterated
water (95%) was used to prepare these samples in order to have comparable signals
from water and surfactant. For LAM the theoretical value of the monomer size
Rmon(T),
calculated based on the surfactants molecular geometry (see Table 13.3 for
further details), is similar to R,,,, the hydrodynamic radius calculated from Dmon;
in

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 367

contrast, for the gemini compounds, Rmonis higher than Rmo,(T).This suggests the
presence of small aggregates, dimers, and trimers for bis(Args) concentrations between
CMC, and CMC,. These observations agree with those obtained by surface tension,
fluorescence, and conductivity (Pkrez, et al., 1998, Pinam et al., 1999).
The aggregation number has been calculated using Rmon(T),as shown in Table
13.3. The high aggregation number obtained for the bis(Args) indicates the formation
of large aggregates. This compares well with the electronic microscopy results
published previously (Weihs et al., 2005). The authors obtained electron micrographs
of these four surfactants. It was observed that at low concentrations the single chain
LAM forms spheroidal micelles while the gemini surfactants form twisted-ribbons,
flat-ribbons, and thread-like ribbons. PGSE-NMR showed that for the C9(LA)*the
intensity of the signals of the simple proton experiment decreases as a function of time.
The main peak of the spectrum, corresponding to the CH, groups of the alkyl chain,
loses one-half of the intensity after 1 day and about 80% in 6 days. This reduction in

C,( LA),Concentration
Fig. 13.9. Self-diffusioncoefficients obtained using PGSE-NMR versus concentrationfor C3(LA)2.The line
corresponds to the best fit of the two-site exchange model (equation 3 of Perez et al., 2007) Reproduced
from Perez et al., 2007 with permission of the American Chemical Society (ACS). Nomenclature of
surfactant is referred to in Table 13.1.

368 0 M.R.Infante et al.

intensity can be explained by the formation of large aggregates and the subsequent
broadening of the PGSE-NMR peaks due to the slow local anisotropic motions of the
alkyl chain, as it is usually observed in vesicular systems (Cocquyt et al., 2004).
The analysis of the SAXS data indicates that while the LAM compound forms
spherical micelles, thegemini surfactants C3(LA)2and C9(LA)2
form cylinders. The area
per molecule and water molecules per head group obtained for the gemini surfactants
are consistent with those obtained for the single chain surfactant (Table 13.2). In both
cases the value obtained for the gemini surfactants is approximately twice that of the
single chain surfactant.
For the C, (Q) a change with time of the S A X S scattering curves in the region
of the structure factor was detected; moreover, the intensity at small scattering vector,
q, decreases with increasing time. Concurrently, an increase of the viscosity of the
solution is observed with increasing time to give, at long times, a gel appearance
of the aqueous samples. These results could indicate a development of a long-range
order with cylindrical aggregates. This agrees with the PGSE-NMR observation
of a reduction of the signal intensity with the time due to the formation of large
aggregates.
In general the aggregation behavior shown by PGSE-NMR agrees with those
obtained by SAXS. Classical spherical micelles have been observed for LAM. Moreover,
larger micelles, which can be considered of cylindrical shape, have been observed for
the three gemini surfactants. This type of aggregation has been also described for bis
quaternary ammonium surfactants or bis(Quats) with short spacer chain (Danino et
al., 1995).
The microstructures formed in aqueous solutions of bis(Args) have been studied
by cryogenic-temperature-transmission electron microscopy (cryo-TEM) (Weihs al.,
2005). The effect of a perturbation of the local arrangement of polar heads, by the
spacer length, on the micellar and mesomorphic properties of surfactants in aqueous
solutions were examined. Cryo-TEM was employed to image microstructures appearing
as a hnction of spacer length and concentration. The results show that bis(Args) tend
to form aggregates of lower curvature than the corresponding monomeric surfactant
LAM. For the bis(Args) molecule with a short spacer, C,(LA), (Fig. 13.5, s = l),
the microstructures observed were spheroidal micelles that changed to thread-like
micelles and disk-like structures as the concentration was increased. The bis(Args)
molecules with longer spacers exhibited lower-curvature microstructures, mainly flat
and twisted-ribbons. Those microstructures appeared at lower concentrations as the
spacer became longer and more hydrophobic. When the spacer no longer inhibits
head-group separation (s > I), low curvature and twisted ribbons structure occur. In
all likelihood, ribbons form because of chirality of the amphiphiles and enhanced
hydrogen bonding of the spacer with the surrounding water that leads to rigid
filament-like structures.

Synthesis,Aggregation Properties,and Applications of BiosurfactantsDerived from Arginine 0 369

Arginine Mono- and Diacylglyceride Conjugates

The micelle formation of mono- and diacylglyceride surfactants from arginine has
been evaluated by conductivity, surface tension, and fluorescence measurements. The
capabilityofthe monoacylatedarginineglycerides 1OOR, 120R, and 140Rsurfactants
(Fig. 13.7a) to form micellar aggregates was studied by electrical conductivity (P6rez
et al., 2004a, 2004b). The calculated CMC values for the three homologs, 6mM for
1 0 0 R , 1.3 mM for 120R, and 0.2 mM for 140R, are approximately twice that
of the lysophosphatidylcholines (Yamakana et al., 1997) with the same alkyl chain
length and one order of magnitude lower than the CMC corresponding to the
12-carbon-straight-chain conventional cationic surfactants (Rosen, 1988). They also
have slightly lower CMC than the N*-acyl-arginine surfactants (Fig. 13.8) (Morin et
al., 2001 b). The difference in molecular architecture between the monoglycerides and
Nu -acyl-arginine derivatives is the glycerol group located between the arginine and
the alkyl chain groups for the former, leading to an increased distance between the
ionic group and the a-carbon atom of the hydrophobic group, which often produces
a lower C M C (Rosen, 1988).
From the conductivity/concentration curves the CMC of the diacyl arginine
glyceride surfactants 88R, 1010R, 1212R, and 1414R (Fig. 13.7b) was determined
and the following values were obtained: 5 mM for the 88R, 1.1 mM for 1010R, 0.3
mM for 1212R, and 0.25 mM for 1414R. The CMC of the diacyl arginine glyceride
surfactants are one order of magnitude higher than those published for the shortchain phospholipids (Tausk et al., 1974) with similar alkyl chain lengths. The second
alkyl chain increases the hydrophobic content of the molecule and consequently
the C M C of these compounds is lower than those corresponding to the monoacyl
arginine glycerides with the same alkyl chain. Figure 13.10 shows the variation of
the log CMC with the surfactant chain lenth for mono- and diacyl glycerides from
arginine. The plots are nearly parallel. These results indicate that the value of the free
energy of transfer from water to the micelle per methyl group (AGo(CH,)) is similar
for the two families. This behavior has also been described for gemini surfactants and
their corresponding monomeric surfactants, such as the bis(Args) and LAM (P6rez et
al., 1998) and for the biQuats and monoQuats (Zana et al., 1991).
The CMC was also measured by surface tension using the Wilhelmy plate method
and fluorescence techniques for the 88R, 101OR, and the 1212R surfactants. Whereas
the values obtained by fluorescence agree with those obtained by conductivity for all
cases, the surface tension technique gave lower CMC values for the diacylglyceride
arginine derivatives of Fig. 13.7b, 0.07 mM for the 88R, 0.006 mM for the 1010R,
and 0.008 mM for the 1212R. A similar trend was obtained for the bis(Args)
surfactants depicted in Fig. 13.5. As discussed elsewhere (Pinazo et al., 1999), the
bis(Args) compounds at very low concentrations form substantial aggregates that
can reduce the surface tension at these low concentrations. To further investigate
the underlying cause of the difference of CMC values obtained by the Wilhelmy
plate and fluourescent methods, Static Light Scattering was carried out to establish

370 0 M.R. Infante et al.

4ov
Alkyl chain length
Fig. 13.10. Relationship between log (CMC/mM) and the alkyl chain length for the mono acyl glyceride
( W ) and the diacyl glyceride ( 0 )compounds from arginine as obtained from unbuffered solutions at
25C by surface tension measurements. Taken from Perez et al., 2004b. (Reproduced by permission
of the Royal Society of Chemistry (RSC) on behalf of the Centre National de la Recherche Scientifique
(CNRS))

the aggregate type formed for these compounds at these low concentrations. The
IOlOR compound in particular showed a vesicle to ribbon transition as a function of
concentration. The scattered intensity of lOlORat concentrations as low as 0.001mM
showed that aggregates were already present in the solution. The form and size of
the aggregates was evaluated to be polydisperse vesicles at low concentrations and
ribbons at millimolar concentrations. This change in form was interpreted in terms
of charging of the bilayer. At low concentration the surfactant molecules dissociate
giving a chloride ion and the cationic species. The latter can further dissociate, with
a proton being released. The highly charged surface that forms is unfavorable and a
pK, shift as large as 7 units is observed (Hagslatt et al., 1991) producing a more acidic
behavior than that typically observed for arginine. As the concentration increases the
proportion of charged species increases with a concomitant increase in preferred area
per molecule, thus inducing a change in the curvature of the aggregates. (Pinazo et
al., 2004). This change induced by the intrinsic pH change was independently proven
by acidification of a vesicle-forming concentration. This is exemplified in Fig. 13.1 1

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 371

where the scattered intensity as a function of squared scattering vector q is shown. The
upper curve corresponds to a 0.05mM concentration in the vesicle-forming regime
while the lower curve corresponds to the same sample afier acidification with 0.04M
HCI. The evident decrease in intensity implies a reduction in the overall size of the
aggregates.
The physicochemical properties of diacyl glycerides from acetyl-arginine 12 12RAc
and 1414RAc, Fig. 13.6b, have been studied based on their ability to form monolayers
and multilayers. Because they have two hydrocarbon chains, their aggregation in
aqueous solutions starts at very low concentrations. A change in the slope of specific
conductivity versus concentration was observed for 1212RAc and 1414RAc around
0.1 mM. Because there was no significant difference between the two compounds,
it was considered that this change did not correspond to a true CMC but to some
change in the aggregate form (Pkrez et al., 2004a), probably a transformation from
vesicles to ribbons (Pinazo et al., 2004). Dilauroyl glycerol acetyl-arginine conjugates
can be considered analogs of partial glycerides and phospholipids. During their
preparation, spontaneous intramolecular acyl-migration reactions were observed and
both possible regioisomers were obtained: 1,2-dilauroyl rac- glycero-3-(Nu-acetyl-Larginine) (1212RAc) and 1,3-dilauroylglycero-2-(Nu-acetyl-L-arginine) (12RAc12).

1XI

o-~

1X I 0"

2XIOl4

4~10'~ 6~10'~ 8~10'~ IxIO'~

Fig. 13.1 1. Scattered light intensity measured using Static Light Scattering as a function of the square of
the scattering vector, q, for a O.05mM aqueous solution of 1,2-di-0-decyl-glycero-3-0-(Na-arginine)
2HCl
(1010R) at 2S'C. Full symbols: aqueous solution prior to acidification. Open symbols: aqueous solution
acidified with 0.04M HCI The lines correspond to a vesicular model and ribbon model respectively.
Adapted from Pinazo et al., 2004. Nomenclature of surfactant is referred to in Table 13.1. (Reproduced
with permission of the PCCP Owner Societies)

372 0 M.R. infante et al.

The presence of both regioisomers influences the phase behavior. The study of the
thermotropic phase behavior in the dry state of pure 1,2-dilauroyl-rac-glycero-3-(N
a-acetyl-L-arginine) and two mixtures of both regioisomers showed that they arrange
in multilamellar stacks. When observed by optical polarized microscopy the typical
texture for smectic systems was found to coincide with a characteristic peak ordering
of the small angle x-ray scattering curves. At low temperature, the lamellar distance
of the pure 1,2-compound was compatible with that of fully extended, all-trans, alkyl
chains. For the mixtures, the difference in bilayer thickness was associated with a
tilting of the hydrocarbons chains away from the bilayer normal caused by difficulties
of packing. The higher the temperature, the shorter the lamellar distance. This change
was associated with the introduction of a kink into the hydrocarbon chain that marks
the onset of melting. Above the transition temperatures, all the samples show the
same repeat distance that would correspond to the melted hydrocarbon chains in the
lamellar liquid crystal phase (Mordn et al., 2004~ ).When comparing this behavior
with that of the monoacyl derivative, the immediate difference corresponds to the
repeat distance of the corresponding smectic gel phases. While the diacyl derivatives
present repeat distances around 4.5nm, the monoacyl derivative (Fig. 13.6a) presents
a repeat distance of only 3.8nm. This is shown in Fig. 13.12. The coexistence of two
lamellar orders is seen for the diacylated derivative, curve a. This was attributed to the
coexistence of two lamellar phases caused by the crystallization from the solvent of the
product that had two enantiomers.
The difference in repeat distances for the monoacyl and diacyl compounds is
due to the preservation of a large area per molecule while there is a strong reduction
in hydrophobic volume. Therefore, the monoacyl derivative self-assembles into a
hydrocarbon interdigitated lamellar phase while the diacylated counterpart forms a
normal bilayer arrangement. The self-assembly structures influence the headgroup
conformation that in the diacylated compound expands to a shorter length (0.8 nm)
than in the monoacylated derivative (1.1 nm). (Mordn et al., 2005). Monoacylated
derivatives melt (80C) only slightly above the formation of the liquid crystalline phase
(70C) while the diacylated derivatives form liquid crystals at a lower temperature
(19C) and melt at a higher temperature (93C). This is probably due to the more
compact packing of the chains in the monoacylated derivative compared to the
diacylated derivative in which pairs of hydrophobic chains are restricted by their polar
head binding.
Concerning the monolayer behavior, it was shown that 1,2-diacyl glycerol arginine-based surfactants exhibits a behavior similar to that found in natural phospholipids. The use of BAh4 (Brewster Angle Microscopy) image analysis of the inner textures
revealed that condensed phases of the dimyristoyl glycerol compounds exhibit hexatic
order. Variations in the chain length introduce similar changes as those commonly
found in lipid monolayers (Abalat et al., 2003).
The compatibility studies of the diacylated surfactants 1212RAc and 1414RAc
with phospolipids in water surface monolayers show that the behavior of 1414RAc is

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 373

Fig. 13.12. SAXS scattering intensity as a function of scattering vector modulus for a) l212RAc and b)
12ORAc dry products at 25C. Arrows and double arrows show the position of the reflectionsattributed
to the two coexisting lamellar phases of 1212RAc. From (MorBn et al., 2004~).Nomenclature is referred
to inTable 13.1. (Reprinted with permission from the American Chemical Society (ACS))

similar to that of 1,2-dipalmitoyl phosphatiylcholine (DPPC); that is, both products

exhibit the gas, expanded liquid, compressed liquid, solid, and collapsed phase
behaviors with increasing surface pressure. The behavior of 1212RAc is similar to that
of 1,2-dimyristoyl phosphatidylcholine (DMPC) showing the same phase transition
sequence as the longer chain homologues but with the absence of the solid phase. The
behavior of the long-chain homologues corresponds to that of an insoluble monolayer
while that of the shorter chain corresponds to a fluid monolayer. The mixtures of
the phospholipids with these lipoaminoacids show miscibility over the full range
of compositions, except for the 1212RAc/DMPC mixtures, which were insoluble
(Lozano eta]., 2008).
The surfactants 1212RAc and 1414RAc were able to stabilize both water in oil
and oil in water droplets (Pkrez, 2004a). This resulted in the stabilization of both
types of droplets concurrently when 50/50 w/w oil/surfactant mixtures were sheared
in the presence of 0.2% surfactant. The emulsions were of the W/O/W type, being
the small water droplets of around 5 pm in diameter and were dispersed in 50 pm
diameter oil droplets dispersed in a water medium. This ability to stabilize both types

374 0 M.R. Infante et al.

of droplets is not usual and could be related to the ability to form lamellar type liquid
crystalline phases.

A pplicat ions
Due to their interesting properties, research on arginine-based surfactants has moved
in the last years from fundamental research to first applications in life sciences. In
this section three different biological applications are discussed antimicrobial activity,
sequestration of membrane lipopolysaccharide,and DNA compaction.

An timicrobial Properties
After years of overuse for antibiotics and biocides, Gram-positive and Gram-negative
organisms have developed a broad range of mechanisms to evade antimicrobial agents,
resulting in a potential global health crisis (Heir et al., 1995). In order to overcome
this rapid development of drug resistance, development of new classes of antimicrobial
compounds has stimulated substantial research interest (Tan el al., 2008; Ozdemir
et al., 2007). Available microbicides and formulations differ considerably in their
physical and chemical properties, effectiveness, and spectra of activity although all
these variables have to be compatible with the toxicological demands of the society.
One important milestone in our research activity is the design and development
of amino-acid-based surfactants with a low toxicity profile and antimicrobial activity.
(Infante et al., 1985, 1992). Arginine derivatives present the best activity against
bacteria due to the presence of the protonated guanidine group at a wide range of
pH. The antimicrobial activity of all type of arginine derivatives was systematically
determined in vitro on the basis of the minimum inhibitory concentration (MIC)
values (Jones et al., 1980), defined as the lowest concentration of antimicrobial agent
which inhibits the development of visible growth of microorganisms after 24 h of
incubation at 37C. The results obtained for monocatenary arginine surfactants,
gemini arginine surfactants, and glycerolipid arginine surfactants, are summarized in
Tables 13.4, 13.5a and 13.5b, and 13.6, respectively.
Data show that all the cationic surfactants from arginine showed inhibition
activities against a wide range of microorganisms. These substances showed a
moderate activity level against bacteria with MIC values of 4-256 mglL. Structureactivity relationships indicated that the higher the antimicrobial activity the higher
the toxicity against the Daphnia magna and the Photobacteriumphosphoreum bacteria,
determined using the procedures described by Pdrez el al. (2002b). It is worth nothing
that a consequence of these MIC values is the low toxic and ecotoxic activity of these
compounds. In general, Gram-negative bacteria were more resistant than the Grampositive bacteria, which can make them suitable for the subsequent biodegradability
process of these surfactants (Pirez et al., 2002b, 2005; Morin et al., 2001b). It is
known that some Gram-negative bacteria are relatively insensitive because their outer
membranes are impermeable to some hydrophobic-hydrophilic compounds (Rosen
et al., 1999) the antibacterial activity of these being higher against Gram-positive

Table 13.4. Minimum Inhibitory Concentration (mg/L) of Monocatenary Arginine Surfactants

Gram-positive

Gram-negative

Microorganisms

ACA'

ALA'

AMA'

AOE'

ACE'

ALE'

CAM'

LAM'

MAM'

Bacillus cereus var. mycoides

128

32

64

256

32

64

128

64

256

Bacillus pumilus
Staphylococcus aereus

32
16
16

256

64

R
32

16

256
256

128
16

Staphylococcus epidermidis

32
32
32

256
32
128

128
128

Candida albicans

64

16

32

2%

128

64
64

256
128
128

64

128

Alcaligenes faecalis

32

16

32

256

64

32

128

64

128

Bordetella bronchiseptica
Citrobacter freundii

16
64

8
32

R
32

64

64

32
64

Serratia marcenses

64

32

64

128

64
64

128
128

64

128

Salmonella typhimurium
Streptococcus faecalis
Escherichia coli
Klebsiella pneumoniae

64

32
R
R
16

32

256

64

R
128
R
256
R
128

64

R
32

Pseudomonas aeruginosa
128
Arthrobocter oxidans
64
Nomenclature of surfactants referred to in Table 13.A.
From Moran et al., 2001b.

'

R
R

64

64

256

256
256

R
R
R
R

R
128

256
32

64

128

128

128
8
256
128
128

64

256
256
256
256
256
256

Table 13.5a. Minimum Inhibitory Concentration (mg/L) of C,(OA), and CJCA), ,C,(CA),, C,(CA),
Microorganisms

C,(CA),

o\

C,(CA),

C$A),

Bacillus cereus vor. mvcoides

128

16

64

>12a

Bacillus subtilis

256
256
32
-

64

64

8
8
32

64
64

128
64
128
32

16
16

16
8

32
128

16
16
64
16

128
32
128
32

128
128
64
64
32

16

>256

64

64

32
32

Stophylococcusoereus
Staphylococcus epidermidis
Micrococcus luteus
Condido albicons
Alcoliqenes foecalis
Bordetello bronchiseptico
Citrobocter freundii
Enterobocteraerqenes
Solmonella typhimurium
Gram-negative
StreDtococcus foecolis
Escherichia coli
Klebsiella Dneumoniae
Pseudomonos oeruginoso
Arthrobacter oxidons
Nomenclature referred to in Table 13.1.
Data from PPrez et al., 2002b.

Gram-positive

C,(OA),

128
64

32
256
256
128

256

16

16

256
8

64

64

32

64

128
64

g
nr

Table 13.5b. Minimum InhibitoryConcentration(mglL) of C,(LA), C,(LA),

Microorganisms
Bacillus cereus var. mycoides
Bacillus subtilis
Staphylococcus aereus
Gram-positive
Staphylococcus epidermidis
Micrococcus luteus
Candida albicans
Alcaliaenes faecalis
Bordetella bronchiseptica
Citrobacter freundii
En terobacter aerogenes
Salmonella typhimurium
Gram-negative
Streptococcus faecalis
Escherichia coli
Klebsiella pneumoniae
Pseudomonas aeruginosa
Arthrobacter oxidans
Nomenclature referred to in Table 13.1.
Data from Perez et at., 2002b.

CJLA),

32
4
64

> 128
8
16
>128
>128
>128
>128
64
8
>128
8
>128
8

C,(W,
32
8
8
>128
16
16
32
128
>128
>128
128
128
>128
8
>128
8

and C,(LA),

$(LA),
64
64

>128
>128
16
128
32
128
>128
>128
>128
128
>128
128
>128
128

Table 13.6. Minimum Inhibitory Concentration(mg/L) of the Mono- and Di-Glyceride Arginine Surfactants

Gram-positive

88R

1010R

1212R

1414R

100R

12OR

140R

Bacillus cereus var. mycoide

64

16

256

>256

32

128

128

Bacillus subtillis
Staphylococcus aereus
Staphylococcus epidermidis
Micrococcus luteus
Candida albicans
Salmonella typhimurium
Pseudomonas aeruginosa

>256

>256

128

64

128

256

128

64

64

Escherichia Coli
Arthrobacter oxidans

Streptoccocus faecalis
Bortedella bronchiseDtica
Citrobacter freundii
Alcaligenes faecalis
Enterobacteraerqenes
Klebsiella pneumoniae v.
preumonial
Nomenclature referred to inTable 13.1.
Data from Perez et al, 2002a, 2004b.

Gram-negative

Mirrooraanicmc

T
4

16

>256

>256

>256

16

128

128

64

16

128

64

64

16

32

64

>256

32

64

128

16

32

>256

>256

32

128

128

>256

64

>256

>256

256

256

>256

>256

128

64

128

32

16

>256

>256

128

64

64

05

128

128

128

0.25

0.25

0.25

64

64

64

32

025
>256

>256

256

256

128

128

>256

256

256

256

128

32

>256

>256

16

256

128

128

16

>256

128

64

128

64
6

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 379

than against Gram-negative bacteria. Moreover, given that the MIC values occur at
concentrations below the CMC of surfactants in water, it may be inferred that the
surfactant monomers and not the aggregates are the species that interact with the cells
(Rosen et al., 1999). These compounds are particularly interesting as preservatives for
food and pharmaceutical formulations as well as active ingredients in dermatology and
personal care products. The antimicrobial action of cationic surfactants is based on
their ability to disrupt the integral bacterial membrane by a combined hydrophobic
and electrostatic adsorption phenomenon at the membranelwater interface followed
by membrane disorganization. The antimicrobial activity of the arginine surfactants
depends on their structure and size, the chain length being a critical structural
parameter for their effectiveness.
Monocatenary Arginine Sugactants
In all instances, both, the alkyl chain length as well as the polar head nature affect the
bactericidal activity. Data in Table 13.4 show that arginine 0-alkyl amides (ACA,A M ,
AMA) Fig. 13.4, and 0-alkyl esters (AOE, ACE, ALE) Fig. 13.3, with two positive
charges in the polar head show the lowest MIC values (Morbn, et al., 2001b). The
molecules are probably strongly adsorbed due to the presence of two ionic charges in
the molecule, which enhances their interaction with the polyionic components of the
charged surface of the microbial cell, consequently triggering membrane-disrupting
properties in the cell bacteria. O n the other hand, for the three series the better activity
was obtained for the surfactants with 12 carbons atoms in the alkyl chain. The best
biological effect at alkyl chain of 12 carbon atoms appears on numerous occasions for
the monocatenary coumpounds (Morbn et al., 2001b; P&ez et al., 2005).
?his optimum can be attributed to the combination of several physicochemical
parameters: hydrophobicity, adsorption, CMC, and aqueous solubility. However, the
higher active alkyl chain length for every homologous series depends on the surfactant
structure (Thorsteinsson et al., 2003; Birnie et al., 2000,;Tsatsaroni et al., 1987).

Gemini Sugactants
Results in Table 13.5a and 13.5b demonstrate that thegemini surfactants with spacer
chain lengths of 3-9 and alkyl chains of 10 carbon atoms, exhibit higher antibacterial
activity than the corresponding single chain homolog (CAM) whereas the bis(Args)
with alkyl chains of 12 carbon atoms exert lower activity than their homolog LAM
(Pirez et al., 2002b). When keeping the alkyl chain length constant, activity seems
to decrease with values of s25. When keeping the spacer chain length constant at 5
= 1, the relationship between the alkyl chain length and the activity is not linear,
showing a maximum for the hornologs C,(CA),. (Pirez et al., 1996, 2002b; Piera et
al., 2000).
Mono- and Di-dycerideArginine Sugactants
All of the mono- and diglyceride arginine surfactant derivatives showed antimicrobial
activities against a wide range of microorganisms, that is they inhibited the growth

380 0 M.R. Infante et al.

of all the microorganisms tested (Table 13.6) (Pirez et al., 2004a, 2004b, 2005). The
power of monoglycerides from arginine does not change drastically with the alkyl
chain. For diglyceride compounds, the antimicrobial activity decreases when the alkyl
chain increases. Nevertheless, the activity of the shorter alkyl chain diglycerides, 88R
and lOlOR (average MIC value of 0.081 mM), is considerably superior to the activity
of monoglycerides (average MIC value of 0.235 mM). The 88R and lOlOR (Pkrez
et al., 2OO2a, 2005) were able to inhibit bacterial growth and even kill some bacteria
at concentration as low as 4 mg/L which is comparable to the activity showed by the
benzalkonium chloride, a well known biocide compound, and that of new analogs
of benzalkonium chlorides (Pernak et al., 1999).This demonstrates that the presence
of two alkyl chain of 8-10 carbon atoms is effective in improving the intensity of the
antimicrobial activity.
Since the perturbation of the cell membrane by these compounds is directed
primarily by physicochemical processes, model membrane systems can provide
valuable information for understanding the mechanism of action of these molecules.
The relationship between the antimicrobial activity of bis(Args), LAM, and other
agents and the physicochemical process involved in the perturbation of the cell
membrane, has been studied (Castillo et al.,2004). To this end, the interaction of
these surfactants with two biomembrane models, DPPC multilamellar lipid vesicles
(MLV) and monolayers of DPPC, 1,2 dipalmitoyl phosphatidylglycerol sodium salt
(DPPG), and ficherichia coli total lipid extract was investigated using Differential
Scanning Calorimetry (DSC) and Langmuir monolayers. DSC results show that
variations in both the transition temperature and the transition width at one-half of
the height of the head absorption peak were consistent with the antimicrobial activity
of the compounds. Penetration kinetics and compression isotherm studies indicated
that both steric hindrance effects and electrostatic forces explained the antimicrobial
agent-lipid interactions.

Sequestration of Lipopolysaccharide
Gram-negative sepsis is a common clinical problem (Gasche et al., 1995) and the
mortality due to septic shock reflects the absence of specific therapy aimed at the
underlying pathogenetic mechanisms. Cationic hydrophobic compounds can interact
with the toxic portion of the lipopolysaccaride (constituent of the outer membranes
of the gram-negative bacteria) and reduce this serious problem. Bis(Args) bind to
lipopolysaccharide and neutralize endotoxic activity in in vitro tumor necrosis factor-a
and nitric oxide release assays (David et al., 2002). The presence in thegemini structure
of two highly basic protonatable guanidinium hnctionalities separated by a spacer chain
provides for excellent recognition of the bis-phosphates on the lipopolysaccharide. In
spite of these results, bis(Args) are themselves unlikely to be of therapeutic d u e due to
their cytotoxicity. However, this class of compound offers an excellent point of departure
with which refine the design and development of less toxic analogs for the treatment of
Gram-negative sepsis.

Synthesis, Aggregation Properties,and Applications of Biosurfactants Derived from Arginine 0 38 1

DNA Compaction
DNA packaging in the living cellular environment is a very important phenomenon.
DNA compaction by polyamines, like spermidine and spermine, are examples of
events that occur in cells and are believed to be important in the regulation of cell
proliferation and differentiation. In the literature one can find many studies of DNA
compaction in aqueous solution. DNA molecules are known to undergo a discrete
conformational transition from an extended to a collapsed state by interacting with
single or double chain cationic amphiphiles. Cationic surfactants associate strongly to
DNA and produce compaction but are often toxic. Since one of our main motivations
consists of the development of new nontoxic biocompatible and biodegradable
systems, we studied the interaction of the single chain arginine-based surfactant ALA
(Fig. 13.4, n = 10) with DNA (Rosa et al., 2007). The ability of this surfactant alone
to compact DNA is compared by fluorescence microscopy studies to classical cationic
surfactants. Furthermore, toxicity studies revealed that the incorporation of ALA in
catanionic vesicle systems transformed them into cell viable systems, extending their
use to drug and gene delivery systems.
A precondensation step of DNA as a viable approach for liposome-based gene
delivery has been also addressed (Rosa et al., 2008). To the best of our knowledge,
ALA is the first cationic amphiphile based on an amino acid structure used in gene
delivery. This approach consists in both the precondensation of plasmid DNA with
an arginine-based cationic surfactant, ALA, and the incorporation of the blood
protein transferrin (Tf) into the formulations. Two cationic liposome formulations
were used, one composed of a mixture of dioleoyl trimethylammoniopropane and
cholesterol (D0TAP:Chol) and the other a pH sensitive formulation constituted of
DOTAP, Chol, 1,2-Dioleoyl phosphatidylethanolamine (DOPE), and cholesteryl
hemisuccinate (CHEMS). The lipidic composition played an extremely relevant role in
transfection efficiency. The pair D0PE:CHEMS enhanced transfection in comparison
with the complexes composed of D0TAP:Chol liposomes. Remarkable transfection
results were obtained for ALA-CatpH-complexes. Correlation between formulations
that transfect poorly and large mean sizes was made. Overall, we demonstrate that the
presence of ALA improves the transfection efficiency when conjugated with cationic
liposome systems.

Conclusions
The proposed biobased surfactants will contribute to the field of biocompatible
surfactant research and ultimately lead to the commercial advancement of biochemical
products for industrial use. Moreover, we believe that the results on the membrane
interaction studies will elucidate new functions of the surfactants, thus expanding
their list of potential applications in biochemistry. Finally, the surfactants proposed in
this review could be of great interest to field of biology, specifically as substitutes for
natural phospholipids.

382 0 M.R. Infante et al.

References
Adams, D.J.; K. Holtzmann; C. Schneider; F. Michael. Self-Assembly of Surfactant-Like Peptides.
Langmuir 2007,23, 12729-12736.
Albalat R.; J. Claret; J. IgnCs-Mullol; F. SaguCs; C. Morin, L. Ptrez; I? ClapCs, A. Pinazo. Langmuir monolayers of Diacyl Glycerol Amino Acid-Based Surfactants. Effect of Substitution Pattern of the Glycerol Backbone, Langmuir 2003, 19, 10878-10884.
Allouch, M; J. Seguer; M.R Infante; C. Selve. Nonionic Amphiphile Compounds from Aspartic
and Glutamic Acids as Molecular Mimics of Lecithins.]. Am. Oil Chem. Soc. 1996, 7 3 , 8 7 4 6 .
Benavides, T.; V. Martinez; M. Mitjans; M.R. Infante; M.C. Morin; I? ClapCs; R. Clothier; M.I?
Vinardell. Assessment of the Potential Irritation and Photoirritation of Novel Amino Acid-Based
Surfactants by in vitro Methods as Alternative to the Animal Tests. Toxicol. 2004,201,87-93.
Birnie, C.R., D. Malamud, R.L. Schnaare, Antimicrobial Evaluation of N-Alkyl Betaines and
N-Alkyl-N,N-Dimethylamine Oxides with Variations in Chain Length. Antimicrob. Agents
Chemother. 2000, 44, 2514- 2517.
Boyat, C.; V. Rolland-Fulcrand; M.-L. Roumestant; P.H. Viallefont; J. Martinez. Chemo-Enzymatic Synthesis of New Nonionic Surfactants from Unprotected Carbohydrates. Prep. Biochem.
Biotechnol, 2000,30, 28 1-294.
Capone, S ; I? Walde; D. Seebach; T. Ishikawa; R. Caputo. pH-Sensitive Vesicles Containing a
Lipidic Beta-Amino Acid with Two Hydrophobic Chains. Chem. Biodivers. 2008,5, 16-30.
Castillo, J.A.; A. Pinazo; J. Carilla; M.R. Infante; M.A. Alsina; I. Haro; I? Clapes. Interaction
of Antimicrobial Arginine-Based Cationic Surfactants with Liposomes and Lipid Monolayers.
Langmuir 2004,20,3379-3387.
Clap&, E; M.R. Infante. Amino Acid-Based Surfactants: Enzymatic Synthesis, Properties and
Potential Applications. Biocatal. Biotranrform., 2002,20, 2 1 5-233.
Clapts, I?; M.C. Morin; M.R Infante. Enzymatic Synthesis of Arginine-Based Cationic Surfactants. Biotechnol. Bioeng., 1999, 63, 333-343.
Cocquyt, J.; U. Olsson; G. Olofsson; l? Van der Meeren. Temperature Quenched DODAB Dispersions: Fluid and Solid State Coexistence and Complex Formation with Oppositely Charged
Surfactant. Langmuir 2004,20,3906-3912.
Danino, D.; Y.Talmon; R. Zana. Alkanediyl-Alpha,Omega-bis(Dimethylalkyylammonium
Bromide) Surfactants (Dimeric Surfactants).5. Aggregation and Microstructure in AqueousSolutions. Langmuir, 1995, 11, 1448-1456.
Das, D.; A. Dasgupta; S. Roy; R.N. Mitra; S . Debnath; I?K. Das. Water Gelation of an Amino
Acid-Based Amphiphile. Chem-Eur J. 2006, 12, 5068-5074.
David, S . , L. Ptrez; M . R Infante. Sequestration of Bacterial Lipopolysaccharide by bis(Args) Gemini
Compounds. Bioorg. Med. Chem. Lett. 2001, 12,357-360.
Epand, R E ; M.R. Infante; T.D. Flanagan; R.M. Epand. Properties of Lipoarnino Acids Incorporated into Membrane Bilayers. Biochim. Biophys, Acta Biomembr. 1998, 1373, 67-75.
Fordedal, H.; S . Sjoblolm; M.R. Infante. Lipoaminoacid Association in the System Na-LauroylI-Arginine Methyl Ester/l-PentanollWater as Studied by Dielectric Spectroscopy. Coll. Su$ A
Physicochem. Eng. Aspect, 1993, 79,81-88.
Gasche, Y.;D. Pittet; I? Sutter. In Clinical Eiahrfor Treatment ofsepsis, Sibbald, W.J., J.L. Vincent,

Synthesis, Aggregation Properties,and Applicationsof Biosurfactants Derived from Arginine 0 383

Eds.; Sprinter: Berlin, 19951 35-36.

Gerova, M.; F. Rodrigues; J.E Lamere; A. Dobrev; S. Fery-Forgues. Self-Assembly Properties of
Some Chiral N-Palmitoyl Amino Acid Surfactants in Aqueous Solution. J Colloid Znteface Sci.,
2008,319, 526533.
Gorbitz, C.H. Microporous Organic Materials from Hydrophobic Dipeptides. Chem. Eur. J
2007,13,1022-103 1 .
Hansen, J.P.; J.B. Hayter. A Rescaled MSA Structure Factor for Dilute Charged Colloidal Dispersions. Mol. Phys.1982, 46,651-656.
Hagslatt, H.; 0.Soderman; B. Jonsson. Divalent Surfactants. Experimental Results and Theoretical Modeling of SurfactantlWater Phase Equilibria. Langmuir, 1991, 10, 2177-2187.
Heir, E.; G. Sundheim; A.L. Holck. Resistance to Quaternary Ammonium-Compounds in
Stapbyhcoccw spp Isolated from the Food-Industry and Nucleotide-Sequence of the Resistance
Plasmid pst827.J Appl. Bacteriol. 1995,79, 149-156.
Holmberg, K. In Novel Sufactants: Preparation, Applications, and Biodegradability, 2nd. Edn.;
Holmberg, K., Ed.; Marcel Dekker: NY 2003; pp. 1-2.
Imae, T.; N. Hayashi; T. Matsumoto; T. Tada; M. Furusaka. Structures of Fibrous Supramolecular
Assemblies Constructed by Amino Acid Surfactants: Investigation by AFM, S A N S , and SAXS. 1.
Colloid Znteface Sci. 2000, 225, 285-290.
Infante, M A ; A. Pinazo; J. Seguer. Non-Conventional Surfactants from Amino Acids and Glycolipids: Structure, Preparation and Properties. Colloidc Su$, A , 1997, 123-124, 49-70.
Infante, M.R.; J. Molinero; l? Erra. Lipopeptidic Surfactants .2. Acidic and Basic N-AlphaLauroyl-LArginine Dipeptides from Pure Amino-Acids.J Am. Oil Chem. SOC.,1992a, 69,
647-652.
Infante, M.R.; M.A. Pks; A. Vilcha; C. Solans. Sintesis y Propiedades de Tensioactivos Catihicos
Derivados de Arginina. Anaks de Quimica, 1992b, 88, 542-548.
Infante, M.R.; J. Molinero; P. Erra; M.R. Julii; J.J. Garcia Domingua. A Comparative Study on
Surface Active and Antimicrobial Properties of Some N"-lauroyl-I-a-a-Dibasic Amino Acid
Derivatives. Fette Seifen Ansn: 1985, 8,309-313.
Infante, M.R.; V. Moses. Synthesis and Surface-ActivityProperties of HydrophobiclHydrophilic
Peptides. Znt. J Pept. Protein Res., 1994, 43, 173-1 79.
Infante, M.R; J.J. Garcia Domingua; I? Erra; M.R. Julii; M. Prats. Surface Active Molecules:
Preparation and Properties of Long Chain Nu-Acyl-I-a-Amino-o-Guanidine
Alkylic Acid Derivatives. Znt. J Cosm. Science, 1984, 6, 275-282.
Jones, R.N.; A.L. Barry; T.L. Gavan; J.A. Washington. In I1 Manual ofClinical Microbioloa. Lennette, E.H.; A. Ballows; W.J. Hauser; H.J., Shadomy; Eds.; American Society for Microbiology;
Washington, 1980, pp. 176-92.
Kunieda, H.; K. Nakamura; M.R. Infante; C. Solans. Reversed Vesicles from Biocompatible Surfacrants. Aa'u. Mater., 1992,4,291-293.
Langevin, D. In Foams andEmulsions; Sadoc, J.E; N. Revier. Eds.; Kluwer: Boston, 1999, pp.
9-18.
Lee, S.; E. Kim; M.L. Seo; Y. Do; Y-A. Lee; S.S. Lee; J.H. Jung; M. Kogiso; T. Shimizu. Self-

384 0 M.R.Infante et at.

Assembled Helical Ribbon and Tubes of Alanine-Based Amphiphiles Induced by Two Different
Formation Mechanisms. Tetrahedron 2008,G4,1301-1308.
Lindman, B.; M.C. Puyal; N. Kamenka; R. Rymden; l? Stilbs, Micelle Formation ofAnionic and
Cationic Surfactants from Fourier-Transform H- 1 and Li-7 Nuclear Magnetic-Resonance and
Tracer Self-Diffusion Studies, J Phys. Chem., 1984,88, 5048-5057.
Lozano, N:, L. PCrez; R. Pons; J.R. Luque-Ortega; M. Ferninda-Reyes; L. Rivas; A. Pinazo.
Interaction Studies of Diacyl Glycerol Arginine-Based Surfactants with DPPC and DMPC
Monolayers, Relation with Antimicrobial Activity. Cod Su$ A Physicochem. Eng. Aspect 2008,

319,196-203.
Martinez V.; E. Corsini; M. Mitjans; A. Pinazo; M.I? Vinardell. Evaluation of Eye and Skin Irritation of Arginine-Derivative Surfactants Using Different in vitro Endpoints as Alternatives to the
in vivo Assays. Toxicol. Lett. 2006,lG4,259-267.
Morin, M. C.; M.R. Infante; I? ClapCs. Synthesis of Glycerol Amino Acid-Based Surfactants. Part
1. Enzymatic Preparation of rac-1-O-(N-Acetyl-LAminoacyI)Glycerol Derivatives. J Chem.
SOC.,Perkin Trans.I . 2001a,2063-2070.
Morin, M.C.; M.R. Infante; I? ClapCs. Synthesis of Glycerol Amino Acid-Based Surfactants. Part
2. Lipase-Catalysed Cynthesis of 1 -0-Lauroyl-rac-Glycero-3-(Nu-Acetyl-L-Amino
Acid) and
1 ,2-Di-0-Lauroyl-rac-Glycero-3-(Nu-Acetyl-L-Amino
Acid) Derivatives. J Chem. SOC.,Perkin

Trans.I.2002,1124-1 134.
Morin, M.C.; A. Pinazo; L. PCrez; I? ClapCs; R. Pons; M.R. Infante. Enzymatic Synthesis and
Physicochemical Characterization of Glycerol Arginine-Based Surfactants. C R Chim. 2004a,7,

169-176.
Morin, M.C.; l? ClapCs; F. Comelles; M.T Garcia; L. Perez; M.l? Vinardell; M.R. Infante. Chemical Structure/Properties Relationship in Single-Chain Arginine Surfactants. Langmuir 2001b,17,

507 1-5075.
Morin, M.C; A. Pinazo; L. Pkrez; l? ClapCs; M. Angelet; M.T. Garcia; M.l? Vinardell; M.R.
Infante. Green Amino Acid-Based Surfactants. Green Chem. 2004b,G,233-240.
Morin, M.C.; A. Pinazo; I? ClapCs; M.R. Infante; R. Pons. The Effect of Molecular Shape on the
Thermotropic Liquid Crystal Behavior of Monolauroylated Amino Acid Glyceride Conjugates.
J PhyJ. Chem. B, 2005,109,22899-22908.
Morin, M.C.; A. Pinazo; I? ClapCs; M.R. Infante; R. Pons. Investigation of the Thermotropic Behavior of Isomer Mixtures of Diacyl Arginine-Based Surfactants. Comparison of Polarized Light
Microscopy, DSC, and SAXS Observations, J Phys. Chem. B,2004c, 108, 11080-1 1088.
Ohta, A; K. Toda; Y. Morimoto; T. Asakawa; S. Miyagishi. Effect of the Side Chain of N-Acyl
Amino, Acid Surfactants on Micelle Formation: An Isothermal Titration Calorimetry Study.
CoL! Su$ B Biointerfaces, 2008,317,316-322.
Oshimura, E.; Y. Yuji; S. Kazutami. Conformational Behavior of N-Acylamino Acid Oil and NAcylamino Acid Surfactant in Aqueous Solution. J OIpo. Sci. 2007,5G,1 1 5-21.
Ozdemir, A.;G. Turan-Zitouni; Z.A. Kaplancikli; G. Revial; K. Guven. Sintesis and Antimicrobial Activity of 1-(4-Aryl-2-?hiazoyl)-3-(2-Thienyl)-5-Aryl-2-Pyrazoline
Derivatives, Eur. J Med.
Chem., 2007,42,403-409.
PCrez, L.; A. Pinazo; M.J. Rosen; M.R. Infante. Surface Activity Properties at Equilibrium of

Synthesis, Aggregation Properties, and Applications of Biosurfactants Derived from Arginine 0 385

Novel Gemini Cationic Amphiphilic Compounds from Arginine bis (Args). Langmuir 1998, 14,
2307-23 15.
Pkrez, L.; A. Pinazo; M.R. Infante; R. Pons. Investigation of the Micellization Process of Single
and Gemini Surfactants from Arginine by SAXS, NMR Self-Diffusion, and Light Scattering. ].
Pbys. Cbem. B, 2007, I l l , 11379-11378.

Ptrez, L.; J.L.Torres; A. Manresa; C. Solans; M.R. Infante. Synthesis, Aggregation, and Biological Properties of a New Class of Gemini Cationic Amphiphilic Compounds from Arginine,
bis(Args). L a n p u i r , 1996, 12, 5296-5301.
Pkrez, L.; M.R Infante; R. Pons; C. Mordn; M.I? Vinardell; M. Mitjans; A. Pinazo. A Synthetic
Alternative to Natural Lecithins with Antimicrobial Properties. Coll. Su$ B Biointefaces, 2004a,
35,235-242.
Pkrez, L.; I? Clap&; A. Pinazo; M. Angelet; M.I? Vinardell; M.R. Infante. Synthesis and Biological
Properties of Dicationic Arginine-Diglycerides. Nau]. Cbem. 2002a, 26, 1221-1227.

Pkrez, L.; A. Pinazo; M.T. Garcia; C. Mordn; M.R. Infante. Monoglyceride Surfactants from Arginine: Synthesis and Biological Properties. New J. Cbem., 2004b, 28, 1326-1334.
Pkrez, L; M.T. Garcia; I. Ribosa; M.I? Vinardell; A. Manresa; M . R Infante. Biological Properties of
Arginine-Based Gemini Cationic Surfactants. Environ. Toxicol. Cbem. 2002b, 21, 1279-1285.

Ptrez, N.; L. Perez; M.R. Infante; M.T. Garcia. Biological Properties of Arginine-Based Glycerolipidic Cationic Surfactants. Green Chem 2005, 7, 540-546.
Pernak, J.; I. Mirska; R. Kmiecik. Antimicrobial Activities of New Analogues of Benzalkonium
Chloride. Eur. J. Med. Cbem., 1999, 34,765-771.
Pes, A. Ph.D Thesis. Estudio del Comportamiento Fhico de Sistemas con Tensioactivos Derivados de Arninoicidos y Caracterizaci6n Structural de las Diversas Fases Is6tropas y Anis6tropas,
Universidad de Barcelona 1993.
Piera, E.; M.R. Infante; I? Clapts. Chemo-Enzymatic Synthesis of Arg-Based Gemini Surfactants.
Biotecbnol. Bioeng. 2000, 70, 323-33 1.
Pinazo, A,; X. Wen; L Ptrez; M.R. Infante; E.I. Franses. Aggregation Behaviour in Water of
Monomeric and Gemini Cationic Surfactants Derived from Arginine. Langmuir 1999, 15,
3 134-3142.
Pinazo, A,; L. Ptrez; M.R. Infante; E.I. Franses. Relation of Foam Stability to Solution and Surface Properties of Gemini Cationic Surfactants Derived from Arginine. Coil. SuTf:A Physicorhem.
Eng. Aspect 2001,183,225-235.
Pinazo, A.; L. Ptrez; M.R. Infante; R. Pons. Unconventional Vesicle-to-Ribbon Transition Behaviour of Diacyl Glycerol Amino Acid Based Surfactants in Extremely Diluted Systems Induced by
pH-Concentration Effects. Pbys . Chem . Cbem . Pbs, 2004, 6, 1475-1481.
Pinazo, A.; M. Diz; A. Pts; I? Erra; M.R Infante. Synthesis and Properties of Cationic Surfactants
Containing a Disulfide Bond.]. Am. Oil Cbem. SOC.1993, 70,37-42.
Prud'homme, R.K.; S.A. Khan. In Foams: Beory, Measurements, andApplications; Prud'homme,
R.K.; S.A. Khan; Eds.; Marcel Dekker: New York, 1996, pp. 276-279.
Rosa, M.; M.R. Infante; M.G. Miguel; B.Lindman. Spontaneous Formation of Vesicles and
Dispersed Cubic and Hexagonal Particles in Amino Acid-Based Catanionic Surfactant Systems.
Langmuir, 2006; 22, 5588-5596.

386 0 M.R.Infante e t al.

Rosa, M.; M.C. Morin; M.G. Miguel; B. Lindman. The Association of DNA and Stable Catanionic Amino Acid-Based Vesicles. ColL Su$ A: Pbysicocbem. Eng. Aspects, 2007,301, 361-375.
Rosa, M:, N. Penacho; S. Simoes; M.C.I? Lima; B. Lindman; M.G. Miguel. DNA Pre-Condensation with an Amino Acid-Based Cationic Amphiphile. A Viable Approach for Liposome-Based
Gene Delivery. Mol. Membt. BioL, 2008,25,23-34.
Rosen, M.J. In Sufactants and Intefacial Phenomena, 2nd Edn.; Wiley-Interscience Publication:
New York, 1988, pp. 65-67.
Rosen, M.J.; L. Fei; Y.l?Zhu; S.W. Morral. The Relationship of the Environmental Effect of Surfactants to Their Interfacial Properties.J. Su$ Deterg., 1999, 2, 343-347.
Roy, S.; J. Dey. Effect of Hydrogen-Bonding Interactions on the Self-Assembly Formation of Sodium N-( 1 1-Acrylamidoundecanoy1)-L-Serinate,LAsparaginate, and L-Glutaminate in Aqueous
Solution. J. Colloid Interjace Sci. 2007,307,229-234.
Roy, S; J. Dey. Self-organization Properties and Microstructures of Sodium N-( 1 1Acrylamidoundecanoy1)-L-Valinateand -L-Threoninate in Water. Bull. G e m . SOC.
Japan, 2006,
79, 59-66.
Seguer, J.; J. Molinero; A. Manresa; J. CaeUes; M.R. Infante. Physicochemical and Antimicrobial
Properties of N-Alpha-Acyl-I-Arginine Dipeptides from Acid-Hydrolyzed Collagen. JSoc. Cosm.
Cbem., 1994,45, 53-63.
Solans, C.; M.R. Infante; N. Azemar; T. Warnheim. Phase Behaviour of Cationic Lipoaminoacid
Surfactant Systems. Prog. Colloid Polym.Sci. 1989, 79,70-75.
Solans, C.; M.A. Pks; N. Azemar; M.A. Infante. Lipoaminoacid Surfactants: Phase Behaviour of
Long Na'pha-AcylArginine Methyl Esters. Prog. Colloid Polym. Sci. 1990, 81, 144-150.
Stilbs, l? Fourier Transform Pulsed-Gradient Spin-Echo Studies of Molecular Diffusion. Prog.
NMRSpectrosc., 1987, 19,1-45.
Takehara, M. Properties and Applications of Amino-Acid Based Surfactants. Coll. Su$, 1989,38,
149-167.
Tausk, R.J.M.; J. Karmiggelt; C. Oudshoorn; J.T.G. Overbeek. Physical Chemical Studies of
Short-Chain Lecithin Homologues. I. Influence of the Chain Length of the Fatty Acid Ester and
of Electrolytes on the Critical Micelle Concentration. Biopbys Cbem, 1974, I , 175-183.
Tan, H.; H. Xiao. Synthesis and Antimicrobial Properties of Novel L-Lysine Gemini Surfactants
Pended with Reactive Groups. Tetrahedron Lett., 2008,49, 1759-1761.
ThorsteinssonT; M. Masson; K.G. Kristinsson; M.A. Hjalmarsdottir; H. Hilmarsson; T. Loftsson.
Soft Antimicrobial Agents: Synthesis and Activity of Labile Environmentally Friendly Long
Chain Quaternary Ammonium Compounds. J. Med. Cbem., 2003,4154 173-4 18 1.
Torres, J.L.; E. Piera; M.R. Infante; l? Clapks. Purification of Non-Toxic, BiodegradableArginineBased Gemini Surfactants, Bis(Args), by Ion Exchange Chromatography. Prep. Biocbem. Biotecbnol. 2001,31,259-274.
Tsatsaroni, E.; S. Pegiadou; G. Dememis. Synthesis and Properties of New Cationic Surfactants,
11, Odd Homologous Members.]. Am. Oil G e m . SOC.,1987, G4, 1444-1447.
Varka, E.M.; M.G. Heli; E. Coutouli-Argyropoulou; S.A. Pegiadou. Synthesis and Characterization of Nonconventional Surfactants of Aromatic Amino Acid-Glycerol Ethers: Effect of the
Amino Acid Moiety on the Orientation and Surface Properties of These Soap-TypeAmphiphiles.

Synthesis, Aggregation Properties,and Applications of Biosurfactants Derived from Arginine 0 387

Chem.-Eur.J. 2006, 12,8305 -8311.

Vijay, R.; S. Angayarkanny; G. Bhasker. Amphiphilic Dodecyl Ester Derivatives from Aromatic
Amino Acids: Significance of Chemical Architecture in Interfacial Adsorption Characteristics.
Coll. Surf A Physicochem. Eng. Aspect., 2008,317, 643-649.
Weihs, D.; D. Danino; A. Pinazo; L. Pkra, E.I. Frames; Y. Talmon. Self-Aggregationin Dimeric
Arginine-Based Cationic Surfactants Solutions. Coll. Surf A Physicochem. Eng. Aspect, 2005,
255,73-78.
Xia, J.; LA. Nnanna. In Protein-Baed Sufactants. Synthesis, Physicochemical Properties, and&
plications. Nnanna, LA.; J. Xia; a s . ; Marcel Dekker: New York 2001, pp. 1-14.
Yamanaka, T.; N. Ogihara; T. Ohhori; H. Hayashi; T. Muramatsu. Surface Properties of Homologs of Lysophosphatidilcoline and Lysophosphatidylethanolamine in Water. Chem. Phys.
Lipids, 1997,70,97-107.
Yamashita, Y; H. Kunieda; E. Oshimura; E; et al. Formation of Intermediate Micellar Phase
Between Hexagonal and Discontinuous Cubic Liquid Crystals in Brine/N-Acylamino Acid
SurfactantlN-AcylaminoAcid Oil System.]. Cuff.Inter. Sci. 2007,312, 172-178.
Yasuhara, K.; A. Ohta; Y. Asakura; T. Kodama; T.Asakawa; S. Miyagishi. Unique Incorporation
Behavior of Amino Acid-Type Surfactant into Phospholipid Vesicle Membrane. /. Coll. Inter.
Sci., 2005,283,987-993.
Zana, R.; J. Xia. Eds. Gemini Surfactants: Synthesis, Intefacial and Solution-Phme Behavior, and
Applications. Surfactant Science Series; Marcel Dekker, Inc.: New York,2003, 117, 1-8.
Zana, R; M. Benrraou; R. Rueff. Alkanediyl-Alpha,Omega-bis(Dimethylalky1ammoniumBromide) Surfactants .1. Effect of the Spacer Chain-Length on the Critical Micelle Concentration
and Micelle Ionization Degree. Langmuir, 1991, 7, 1072-1075.
Zhang, X.M.; S. Adachi; Y.Watanabe; R Matsuno. Lipase-CatalyzedSynthesis of 0-Lauroyl LSerinamide and O-Lauroyl L-Threoninamide.FoodRes. Int., 2005,38,297-300.

Design of Vegetable Oil

Metalworking Fluid Microemulsions
Using Biobased Surfactants
Fu Zhaol, Kim Hayes2,Steven J. Skerlos
Schoolof Mechanical Engineering, Purdue UniversitK West Lafayette, Indiana, 47907; Department oKivil
and Environmental Engineering and Department of Mechanical Engineering, Universityof Michigan, Ann
Arbor, Michigan, 48105

Introduction
Metalworking fluids (MWF) are engineering materials that optimize metalworking
processes such as metal cutting and metal forming (McCoy, 2006). The primary
function of MWF is to provide lubrication and cooling. Additionally, MWF provide
some secondary functions such as chip transport, corrosion protection, and tool/
work-piece cleaning (Childers, 2006). Worldwide more than 2 billion liters of MWF
are consumed annually (Stanford et al., 2002).

Classification of Metalworking Fluids

In general, commercial metalworking fluids can be divided into two basic categories:
water-immiscible and water-miscible (Childers, 2006). The water-immiscible MWF,
also known as oil-based fluids or straight oils, are petroleum or vegetable oils that
are used without dilution in water. They can be oil alone or oil compounded with
various polar andlor chemically active additives. Due to their higher cost, smoke and
fire hazards, operator health risks, and limited tool life due to inadequate cooling,
water-immiscible MWF only account for a small fraction of MWF used (Iowa Waste
Reduction Center, 2003). The water-miscible MWF are true aqueous solutions or
oil-in-water emulsions, and are sold in the form of concentrates which have to be
diluted with water before use. They can be further classified into soluble oil, semisynthetic, and synthetic MWF according to the amount of oil and water present in
the formulations. Soluble oils typically consist of 60-90% oil in the concentrate,
while semi-synthetic MWF consist of 2-30% mineral oil. Both soluble oil and semisynthetic MWF contain emulsifiers (surfactants) to keep the oil dispersed in water
and to maintain stable oil-in-water emulsions. Synthetic MWF contain only watersoluble ingredients. They are solutions formulated with lubricant additives to achieve

389

390 0 F. Zhao et al.

lubricity. All water-miscible MWF contain auxiliary components such as corrosion

inhibitors, wetting agents, chelating agents, pH buffers, and defoamers (Childers,
2006; Sheng & Obenvalleney, 1997). Due to the fact that semi-synthetic MWF share
the advantages of both soluble oil and synthetic MWF, and enable a balance between
cooling efficiency and lubrication efficiency, they account for more than 40% of
the MWF market (Childers, 2006) and their market share is expected to continue
growing.

Environmental Concerns of MWF and Motivation for Biobased MWF

In North America more than 95% ofMWF consumed are formulated with components
made from petroleum feedstock (Whitby, 2004). Petroleum oils and their derivatives
in MWF have high eco-toxicity and low biodegradability. This leads to high treatment
and disposal costs and raises concerns regarding occupational health risks for workers
exposed to MWF (IAMS, 1995; NIOSH, 1998; Simpson et al., 2003; Raynor et
al., 2005; Cohen & White, 2006). These factors, along with increasing crude oil
prices and the national security concerns associated with imported oil, have served as
motivations for practitioners to consider alternatives to petroleum-based MWF.
Biobased formulations offer a renewable and domestically produced alternative
to petroleum-based MWF. Due to their inherent biodegradability, biobased
formulations can reduce the waste treatment costs required to meet the MWF
effluent limitation guidelines and standards published by the U.S. EPA in the Metal
Products and Machinery Rule ( U SEPA, 2003). Also, biobased formulations may
reduce the occupational health risks associated with petroleum oil MWF due to their
lower toxicity (Raynor et al., 2005). At the same time, biobased MWF perform better
in many manufacturing operations, such as thread cutting (Clarens et al., 2004),
due to the fact that vegetable oils can have better lubricity than petroleum oils.
Moreover, the major economic barrier for the adaptation of vegetable oil lubricants
is diminishing as crude oil prices increase (Ash & Dohlman, 2005), while the major
technical concerns associated with vegetable oil lubricants (i.e., thermal and oxidative
stability) have been addressed through genetic modification of oil-seed bearing crops,
chemical modification, and the use of various additives (Rose & Rivera, 1998). Since
worldwide only 5% of the annually consumed lubricants (about 40 million tons) are
used to formulate MWF (Mang, 2007), switching to vegetable oil-based MWF will
not significantly increase the world vegetable oil demand which is close to 130 million
tons (Soy Stats, 2008). Due to the increasing popularity of semi-synthetic M W , here
we focus on formulating semi-synthetic oil-in-water microemulsions using vegetable
oils.
In general, semi-synthetic MWF are sold as concentrates with 10-30% oil and are
diluted 10-20 times with water before use as metalworking fluids (Childers, 2006).
Table 14.1 gives the composition of a representative semi-synthetic MWF concentrate
used here as a model system (Milacron, Inc., Cincinnati, OH). The diluted fluids
are stable, translucent and ofien called microemulsions,with emulsified oil droplet

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 39 1

sizes less than 100 nm. The selection of surfactants to disperse the oil and other
hydrophobic additives in water is critical for producing a stable microemulsion.
Recent efforts have reported on the use of vegetable oils to replace the petroleum oil
in MWF formulations (Belluco & De Chiffre, 2004; John et al., 2004; Abdalla &
Patel, 2006; Oliveira & Alves, 2006). However, most of these formulations contain
surfactants made from petroleum feedstock and therefore are not completely biobased.
According to the German Blue Angel label standard for environmentally friendly
lubricant packages, 70% (by weight) of the components must be biodegradable and
made from biobased feedstock (Bartz, 1998). With the goal of moving towards 100%
biobased MWF formulations, biobased surfactants should also be considered. Similar
to converting petroleum-based MWF to vegetable oil-based formulations, switching
to biobased surfactants would have negligible impact on world vegetable oil demand
since about 0.1% of the worldwide crude oil consumption (i,e,, about 3.5 million
tons) is used for the production of surfactants (Saouter, 2003). Although other
biobased constituents/additives of a fully hnctional MWF could also be targeted, in
this chapter only biobased oils and surfactants are considered.
The challenge of formulating vegetable oil MWF with biobased surfactants exists
because, even for traditional petroleum-based fluids, the creation of new formulations
has long been carried out using guidelines developed through trial-and-error experience
(Childers, 2006). A systematic process to select biobased surfactants with appropriate
concentrations to disperse vegetable oil into water to form MWF microemulsion
systems with maximum stability has yet to be developed using current advances in
emulsion science. As a result this chapter develops a set of structure-based surfactant
selection guidelines and a model-based tool for optimizing surfactant concentrations
that can be used to design highly stable biobased semi-synthetic metalworking fluids
without extensive trial-and-error experiments.
Table 14.1. Composition of a representative semi-synthetic MWF
Constituent

w/w

Petroleum oil

15%

Sodium Detroleum sulfonate (anionic surfactant)

6.0%

Diisoprooanolamfne (co-surfactant)

8.0%

Corrosion inhibitor

6.0%

Coupler

1.5%

Fattv acids

3.7%

--

__

Monoethanolamine

1.9Yo

Chelatinq aqent

0.15%

Biocides

2.0Yo

water

balance

392 0 F. Zhao et al.

Surfactant Selection for MWF Formulations

Surfactants Used in M WF
Surfactants, or surface acting agents, are usually organic compounds that are
amphipathic, meaning that they contain both hydrophobic (the tail) and hydrophilic
(the head) groups in the molecule (Myers, 2006). Therefore, they tend to reside at the
gas-water or oil-water interface. The most widely used surfactants can be classified into
four categories according to the type of charged head group. A nonionic surfactant
has no charge in the hydrophilic head group. If the head group is negatively charged,
the surfactant is called anionic. If the head group has positive charge, the surfactant
is called cationic (Rosen, 2004). Another group of surfactants called zwitterionic
surfactants have both positive and negative charges in the head group. Zwitterionic
surfactants are generally considered to be specialty surfactants, and their use is usually
limited to cosmetic applications. Compared with anionic and nonionic surfactants,
cationic surfactants are rarely used in MWF formulations since they may be lost to
metal surfaces that are typically negatively charged at the pH of 8-10 commonly
found in MWF formulations (Zimmerman et al., 2003a).
Due to its low cost, sodium sulfonate (anionic) is currently the most predominant
surfactant used in MWF formulations (Childers, 2006). Sodium sulfonate tends
to precipitate out from water when hardwater cations (Ca*+and Mg+) of high
concentration are present. To counter this adverse effect, nonionic surfactants such
as nonylphenol ethoxylates, polyethylene glycol esters, and alkanolamides are used
to improve hardwater stability (Zimmerman, 2003b; Childers, 2006). Anionic and
nonionic surfactant mixtures are commonly supplemented by fatty acid soaps, esters,
and coupling agents to hrther increase emulsion stability (Childers, 2006). Most of
the surfactants used in MWF formulations, such as sodium sulfonate, nonylphenol
ethoxylate, and alkanolamide are made from petroleum oil (Patel, 2004; Hauthal,

2004).
In order to develop guidelines to select biobased surfactants for vegetable oil
MWF, considered here are the commercially available anionic and nonionic surfactants
that can be used in MWF from the major classes (Myers, 2006). Specifically, anionic
surfactants from six different classes are investigated, viz.: fatty acid soaps, alcohol
sulfates, alcohol ether sulfates, alkane sulfonates, alkyl aryl sulfonates, and sulfocarboxylic esters. Nonionic surfactants have been investigated from four classes:
ethoxylated alcohols, ethoxylated glyceryl esters, polysorbitan esters, and alkyl
polyglucosides. Table 14.2 lists all the surfactants evaluated in this chapter. ?he table
lists the average carbon chain lengths in the tail and the degree of ethoxylation (as
indicated by head group ethylene oxide numbers) as provided by manufacturers and
adopted for data analysis as suggested by Kibbey and Hayes (1998).Among all the
surfactants listed in Table 14.2, only the anionic surfactants of the sulfonate class
must be manufactured from petroleum feedstock. The remainder, particularly the
nonionic surfactants, can be made from biobased feedstock, though many are still at
least partially derived from petroleum (Patel, 2004; Hauthal, 2004).

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants0 393

Table 14.2. RepresentativeSurfactants that can Potentially be Used in MWF Formulations

surfactant class
ethoxylated
fattty acid

chemical structure
--&,-..-)&o&-.~~~

leF:th

l2

8080

..-

alkane
sulfonate

+/,-kSO,Na

sulfocarboxylic
ester

ethoxylated
alcohol

361

22

NinexMT603

15

865

26

Nlnex MT 615

232

40

288

39

Polystep 829
POlYSteP B5

H+fiO~f-,'OH

vendor
Stepan
Stepan

1
4

332
464

39 Polystep B130S
40 polystep B430S Stepan

13

434

38

Stetol FS406

14

328

11

Bioterge AS40

216
222
208
455

Stepan
15 Bioterge PAS8S
13.7 Stepanate SCS
14.2 Stepanate SXS Stepan
8.2
Biosoft D40

542

25.6 Calfax 1OLA-75

CEE;cal

14

340

16 PolystepMC48 Stepan

10

298

20

Fluka86146

tljTh

10

3
6

425

8.5 Tomadol 91-2.5

12.4 Tomadolgl-6

12

3
7

322
484

7.9
12

Tomadol23-3
Tomadol 23-6.5

Tomah

2
10

540
760

6.5
11.5

Brij 52
Brij 56

:;jzh

20

1110 13.2

Brij58

1780

4.2

Ethox-5

Stepan

20

2500

8.4

TagatV2O

Degussa

40

3300 14.0 Toximul8244

16

18

12
18

polysorbitan
ester

HLB trade name

12
12

~ ~ M r o ~ ~ + o s o , N 8

azrr

alcohol
sulfate
alcohol
ether
sulfate

EO #

12

281

346

8.6

Span 20

429

4.3

Span 80

1228

16.9

Tween 20

1310

15

Tween 80

310

8.5

Agnique 265

440

10.5

TritonCG 110

Stepan

20
18

alkyl
polyglucoside

14

10

Cognls
Dow-

Chemical structure, average tail length, average EO#, HLB provided by vendor.
Reprinted with permission from Zhao, F. et al, Environ. Sci.Technol., 40,24,7930-7937,2006. Copyright
2006 American Chemical Society.

394 0 F. Zhao et al.

Hl.B Method
Despite the long history in developing M W , formulating petroleum-based MWF
is still highly empirical (Childers, 2006). Consequently, it is often unclear why a
particular surfactant mixture has been selected for a given MWF, and why certain
combinations of surfactants at specific concentrations form more stable MWF than
others. Among the approaches used to quantitatively correlate surfactant structures
with their emulsification effectiveness, the HLB (hydrophile-lipophile balance)
method has achieved some degree of acceptance for selecting surfactants by M W
formulators (Myers, 2006; Canter, 2005). This method was proposed by Griffin in
1949 for the design of emulsion systems using nonionic surfactants, and has been
extended to systems with anionic and nonionic surfactants (Griffin, 1949; Myers,
2006).
In the HLB method, a number (0-40) indicative of emulsification behavior is
assigned to a surfactant. The HLB number can be calculated from the structure of the
surfactant molecules or it can be determined experimentally. For a binary surfactant
mixture, the composite HLB can be determined by (Myers, 2006):
(Eq 14.1)
wheref;, is the wt% of the first surfactant. In addition, commonly used hydrophobic
substances such as benzene, castor oil, and soybean oil are assigned nominal HLB
numbers based on emulsification experiments. For example, benzene has a nominal
HLB of 15 and paraffin has a required HLB of 10 (Myers, 2006). In theory, any
surfactant or surfactant combination with an HLB number that matches the nominal
HLB should be able to emulsify the hydrophilic substance (Rosen, 2004).
While the HLB method is widely used in industrial applications as an initial guide
for surfactant selection, it has serious limitations when applied to MWF, especially
vegetable oil M W . For instance, Zimmerman et al. (2003b) designed both mineral
oil- and canola oil-based semi-synthetic MWF using surfactant mixtures consisting
of a nonionic ethoxylated alcohol and either an anionic alkylbenzene sulfonate or
an alkyldiphenyl oxide disulfonate. The data revealed that stable canola oil-based
formulations could be achieved with a wider range of HLB values (6-18) than for
petroleum oil-based formulations (6-12). Also, at a given HLB value, it was found
that both stable and unstable formulations could be created. This is consistent with
industry experience suggesting that a single indicator of surfactant properties, such as
HLB, is not predictive ofstable MWF formulations. O n the other hand, Zimmerman
et al. (2003b) suggested that other physical characteristics, such as head and tail group
surfactant structures, could be considered when trying to predict the stability of an
emulsifier system during the design of M W formulations as discussed h r t h e r in the
next section.

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants 0 395

Structure Based Surfactant Selection

MWF Mimemuhion as Swollen Micehs
In considering how surfactant structure affects MWF microemulsion stability, it
is beneficial to review the basic principles and nomenclature used for describing
the stability of oil-in-water microemulsion systems. Although little research on
semi-synthetic MWF microemulsions has been documented, the formulation of
microemulsions has been extensively investigated in other applications, such as oil
recovery, soil remediation, cosmetics, and pharmaceutical product manufacturing
(Paul & Moulik, 1997). The term microemulsion was first introduced in 1958 by
Schulman to describe a specific class of colloidal systems consisting of oil, water or
an aqueous salt solution, and surfactant (usually with a co-surfactant such as fatty
alcohol). Such a system is macroscopically a single phase and thermodynamically a
stable isotropicsolution. However, controversy still surrounds the suitability ofthe term
microemulsion in certain applications. Some researchers view microemulsions as a
specific class of emulsion, which implies the presence ofseparated oil and water phases
and a favorable free energy associated with forming the oil-water interface (Shinoda
& Lindman, 1987). Given the small droplet size associated with microemulsions
(i.e., 10-100 nm) and the assumed thermodynamic stability, the interfacial tension
between the oil and water phase has to be zero or even negative. As pointed out by
Hiemenz and Rajagopalan (1997), quantitatively verifying thermodynamically stable
microemulsion phase regions is difficult due to the challenge associated with applying
macroscopic interfacial tension measurements to oil-water interfaces that approach
the nanoscale dimensions in microemulsion systems.
Another perspective is to view stable microemulsions as systems formed by
micellular solubilization of the dispersed phase (Hiemenz & Rajagopalan, 1997;
Morrison &Ross, 2002). Due to theiramphiphilic characteristics, surfactant molecules
tend to form aggregates or micelles above a certain concentration (called critical
micelle concentration, CMC). For oil-in-water emulsions, the hydrophobic tails of
surfactant molecules are oriented toward the interior of the micelle. A substance (e.g.,
oil) that is normally insoluble or only slightly soluble in water can be dissolved in the
hydrophobic core of a micelle. In Fig. 14.1, the small aggregate on the lefi represents a
micelle with little or no solubilization of oil. From left to right, the droplets increase
in size owing to an increasing amount of oil solubilized in the micelle, and are often
called swollen micelles. As larger droplets are formed, the system eventually becomes
a macroemulsion with dimensions exceeding the micron size domain (Adamson &
Gast, 1997).
The continuum of states suggested by Fig. 14.1 may not be physically attainable.
Colloid systems exist which, instead of transitioning to macroemulsions, remain as
microemulsions with the quantity of dispersed oil existing well above an apparent
solubilization capacity of spherical micelles as defined by the amount of oil solubilized
per micelle. Shah et al. (1977) noted that for microemulsions of the swollen micelle
type the ratio of solubilized molecules to surfactant molecules rarely exceeds 2, while

396 0 F. Zhao et al.

Fig. 14.1. Progressionof a micelle to a swollen micelle in oil-in-water microemulsion upon the addition
of oil to surfactant aqueous solution.

for many other microemulsion systems, in which the phase separation is more complex
(e.g., lamellar or bicontinuous phases), the ratio can exceed 100. In semi-synthetic
MWF dilutions, the molar ratio of oil to surfactant is typically around 1:1 and the
volume fraction of oil can be as low as 0.5% (Childers, 2006). Therefore, semisynthetic MWF microemulsions can be thought of as swollen micelles, and as shown
below, developing surfactant selection guidelines consistent with this perspective is
useful.

Eflect of Surjactant Structure on Micelle Solubilization

The molecular structure of a surfactant can greatly affect its colloidal properties such
as the CMC, aggregation number (defined as the number of surfactant molecules
that form a micelle), and micelle shape. The structure of a surfactant will also, to a
certain degree, serve as a predictor of the surfactants ability to solubilize hydrophobic
substances. In an aqueous system, micelle-forming surfactants will present their polar
head group at the water interface with the hydrocarbon tail forming a hydrophobic
micellar core. When low water solubility compounds with long-chain hydrophobic
functional groups, such as vegetable oils, are introduced, the micelles will favorably
solubilize them, with the hydrophobic portion of the low solubility compound
extended deep within the interfacial layer or in the inner core of the micelle interior
(Myers, 2006). In this case, the solubilization capacity of the micelle will increase
as the size of the micelle increases. Therefore, any factor that increases either the
diameter of the micelle or its aggregation number will increase the solubilization
capacity of the micelles. From the surfactant structure perspective, one would expect
the solubilization capacity of a micelle to increase as the length of the surfactant
hydrocarbon tail increases or the size of its polar head group decreases (Myers,
2006).
Likewise properties of the oil being solubilized can be important. For a given
surfactant system, the micelle solubilization capacity generally decreases as the
carbon chain length of the oil molecule increases. Weiss et al. (1 997) investigated the
solubilization of n-tetradecane, n-hexadecane, and n-octadecane in polyoxyethylene
sorbitan monolaurate solutions, and found the amount (in moles) of solubilized

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 397

n-octadecane is one order of magnitude lower than that of a-tetradecane. Measurements

of the change in size of polyoxyethylene sorbitan monolaurate micelles before and
after solubilization of n-hexadecane indicated that there was only a slight increase in
their diameter (<5%), which suggested that the molecules pack between the chains of
the surfactant molecules. If a hydrocarbon molecule is significantly longer than the
surfactant tail, it may not be completely accommodated in the hydrophobic interior
of a micelle and its end may protrude into the more polar interfacial layer or even
into the aqueous phase, resulting in a thermodynamically less favorable solubilization
scenario (Weiss et al., 1997; Myers, 2006).

Selecting Sut$actants Based on Molecular Structure and Micelle Solubilization

The manufacturing performance of a MWF is determined primarily by the
concentration of oil in the formulation (Childers, 2006). Therefore, to formulate
a stable semi-synthetic MWF, a surfactant or combination of surfactants has to be
selected such that the oil at the required concentration can be solubilized within the
micelles. This oils solubility limit is determined by the product of the solubilization
capacity of the micelle (moles of oil per micelle) and the micelles own solubility
(above which coalescence or aggregation occurs). For semi-synthetic MWF, it is
generally desirable to select a surfactant system that produces micelles with a high oil
solubility (in molesll) at low surfactant concentration so the total surfactant usage
can be reduced to avoid high costs, excess foaming, and difficult wastewater treatment
issues.

Experimental Method
A protocol was developed to establish concentration conditions under which a given
surfactant system was able to produce a stable MWF. First a set of standard oils was
used that included a canola oil (AgriPure 75, Cargill, Inc), a soybean oil (Technical
Grade, Cargill, Inc.), and a trimethylolpropane (TMP) tetra-ester (Priolube 1427,
Uniqema, Inc.). Canola oil, soybean oil, and synthetic esters made from vegetable
oils are representatives of three biobased feedstocks that can be used as major base
lubricant substitutes for petroleum oil. As seen in Table 14.3, all three oils have a fatty
acid composition predominantly in the 16- to 18-carbon chain length range. Since
the focus here is on surfactant selection, for convenience all MWF were formulated
to have a fixed oil concentration of 0.019 mole/L, which is a representative oil
concentration found in semi-synthetic MWF.
For each formulation of surfactants selected (see single and binary system
descriptions below), theMWF fluidsampleswereagedfor 12-15 hoursat approximately
25C and then subjected to stability measurements based on visual transparency, light
transmittance, and oil droplet size (Byers, 2006). The measurements were interpreted
as a score with possible values limited to 1, 3, or 9, with 9 corresponding to the
most stable. For light transmittance, 0-50% was assigned 9, 50-90% was assigned
3, and 90-100% was assigned 1. For particle size, a mean droplet size of 0-100

398 0 F. Zhao et al.

Table 14.3. Fatty Acid Composition of Canola Oil, Soybean Oil, and TMP Ester

Fatty Acid Composition*

Canola Oil

Soybean Oil

Trimethylolpropane
tetra-ester

C160

4%

5%

0%

C180

2%

5%

1%

C181

74%

61%

58%

C182

12%

7%

24%

C183

4%

3%

10%

Other
Trade Name
Vendor

4%

1 9%

7%

AgriPure 75

Technical Grade

Priolube 1427

Cargill Industrial Oils & Lubricants,

Minneapolis, MN

Uniqema, New
Castle,. Delaware

*As provided by vendors.

Reprinted with permission from Zhao, F. et al, Environ.Sci.Technol., 40,24,7930-7937,2006.
Copyright 2006 American Chemical Society.

nm was assigned 9, 100-500 nm was assigned 3, and >500 nm was assigned 1. A

composite score was calculated as the sum of the three stability metrics measured after
7 days, with a maximum of 27 and a minimum of 3. As an operational definition of
emulsion stability, only fluids with a score of 27 were considered to be stable. In order
to determine the long-term stability of the fluids that were stable in the short term,
all three measurements were performed again afier 7 days and with a visual inspection
after 6 months. The visual inspection after 1 week consistently gave the same score
as obtained afier 6 months, suggesting that stability metrics determined after 7 days
are indicative of long-term stability. In the results presented below, only surfactants
or surfactant systems that led to at least one stable formulation received significant
consideration.

2.2.3.3.2 Single surkctant

For each surfactant out of the 31 anionic and nonionic surfactants listed in Table
14.2, single surfactant MWF were prepared with a surfactant to oil molar ratio of 1:8,
1:3.5, 1:2, 1:1.25,1.25:1,2: 1,3.5: 1, and 8: 1. According to the previous discussion on
micellar oil solubilization, it would be expected that vegetable oil molecules can only
be effectively solubilized in the interior of a micelle if a surfactant with a tail longer
than 16 carbons is selected. The experimental data proved that this was indeed the
case. Among the 31 anionic and nonionic surfactants investigated, only the following
nonionic surfactants produced stable emulsions: the ethoxylated alcohols with 16
carbon atoms in the tail (denoted as C=16) and 20 ethylene oxide groups in the head
(denoted as EO=20), the ethoxylated glyceryl esters with C=18 and EO=40, and

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants0 399

the polysorbitan esters with C=18 and EO=20. For all these systems, the required
surfactant to oil ratio was at least 3.51. However, this high surfactant to oil ratio
would be impractical for manufacturing operations due to the high cost of surfactants
and difficulties in controlling foam and treating the waste emulsions that are associated
with high concentration nonionic surfactant systems (Schott, 1988; Byers, 2006).
The high surfactant to oil ratio required to achieve stable h4WF microemulsions
when using only one nonionic surfactant was derived from the fact that there exists
an inherent trade-off between micelle size and micelle solubilization capacity. As
discussed previously, if the surfactant tail length is held constant, the head group of
the surfactant needs to be small enough so that large micelle size and high micelle
solubilization capacity can be achieved (Myers, 2006). This way the amount of
surfactant needed to emulsify a given amount of oil is reduced. However, a small
head group size leads to low micelle solubility in water and may not produce enough
micelles to achieve a stable system for the given oil concentration. O n the other hand,
a large head group can lead to high micelle solubility but produces smaller micelles
with lower and insufficient oil solubility. Even if the oil solubility is high enough,
it may be that an impractically large amount of surfactant will be needed in order
to maintain enough of the small micelles in the system. 'Therefore, for vegetable
oil MWF, the trade-off between micelle size and micelle solubility has to be taken
into consideration when determining the size of the surfactant head. To reduce the
limitations of a given surfactant chain length or head group size for oil solubilization,
it is customary to consider a binary mixture of surfactants.

Binary Surfactant Mixture

Treiner et al. (1990) found that binary surfactant mixtures, especially an ionic
combined with a nonionic, can have higher oil solubility than those obtained with
either component alone if the total surfactant concentration is kept the same. That is,
a surfactant mixture can have a synergistic effect on oil solubility. For a surfactant that
forms large micelles with only limited micelle solubility, the introduction of a watersoluble surfactant can increase the micelle solubility (i.e., number of rnicelles), and
hence the oil solubility limit. When a water-soluble surfactant is properly selected,
mixed micelles consisting of molecules from both surfactants may be formed. The
mixed micelles have similar size relative to the micelle originally formed but increased
micelle solubility due to the presence of the more highly water soluble component.
The greater repulsion of the more water-soluble surfactant head groups within the
mixed micelle also can enhance access of the oil into the micellular core. Because
of their larger micelle solubilization capacity when compared to ionic surfactants,
nonionics are preferred when selecting surfactants for vegetable oil MWF (called
the primary surfactant). The additional, water-soluble surfactant (often called the
secondary surfactant or the co-surfactant) can be either anionic or nonionic.
To determine whether or not the binary surfactant mixture leads to a stable
microemuslion, one can test all the possible combinations of oil and surfactant

400 0 F. Zhao et al.

concentrations. Results from such a test can be plotted to create a formulation

triangle such as that shown in Fig. 14.2, where each point within the triangle
represents a MWF formulation with a different oil and surfactant molar fraction.
By selecting a subset of system conditions within the triangle one can reasonably
well represent compositions of interest. For this work, ten formulations uniformly
distributed within the triangle formulations were considered sufficient to evaluate the
system. Any stable formulation that could not be detected by the ten points would be
essentially too sensitive to composition variations that would be unavoidable in field
applications, and therefore, would have very limited practical significance. With the
ten points, the triangle can be divided into ten sub-regions. Denotingiil,&, andfc as
the molar fractions of oil, primary surfactant, and co-surfactant, respectively, for every
formulation point in Fig. 14.2

Oil

100%

Co-Surfactant

Surfactant Fraction

Primary
Surfactant

Fig. 14.2. Formulation triangle representing different oil/surfactant molar ratios for a system of two
surfactants and one oil. (Reprinted with permission from Zhao, F. et al, Environ. Sci. Technol., 40, 24,
7930-7937,2006.
Copyright 2006 American Chemical Society)

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants 0 401

xi,

+G

+fc

(Eq. 14.2)

=1

(MYi,),

Given the molecular weight of oil

primary surfactant (MW ), and cosurfactant ( M W c ) ,the weight fraction required to make the surfactantJoi1 mixture
can be determined by:

(Eq. 14.3)
Since in practice semi-synthetic MWF are sold as concentrates and diluted 10-20
times with water before use as metalworking fluids (Childers, 2006), the mixtures
of surfactants/oil have to be diluted before stability testing. Here ASTM I deionized
water adjusted to pH=9.5 with sodium hydroxide was used to make the resultant
diluted fluids with a pH consistent with a typical value found in commercial MWF.
Given that all MWF microemulsions were tested at a fixed oil concentration of 0.019
mole/L, the weight based dilution ratio R was calculated as:

R=

W .

M Y i l R'b.0 19

(Eq. 14.4)
'lCrwF

where pMW is the specific weight of the diluted MWF microemulsions, which is close
to 1000 g/L.
Figure 14.3 shows the number of observed stable formulations (out of the ten
possible formulations indicated in Fig. 14.2) when using the selected binary surfactant
systems to emulsify canola oil. For some nonionic surfactants, no stable formulation
was achieved regardless ofwhich anionic was used in combination. This indicates that
the nonionic surfactant is more important than the anionic surfactant with respect
to microemulsion stability (which is why it serves as the primary surfactant). Figure
14.3 also indicates that in order to obtain at least one stable formulation, the tail
length of the nonionic surfactant should be at least 16, which is close to the carbon
chain length of the fatty acid compositions in canola oil. These observations are in
agreement with the principles of micelle solubilization.
From Fig. 14.3 it can be seen that ethoxylated glyceryl esters, ethoxylated alcohols,
and polysorbitan esters with tail lengths longer than 16 carbon atoms serve as effective

402 0 F. Zhao et al.

Co-Surfactant
sulfo-carboxylicester C=15
alkane sulfonate C=14
alcohol ether suifate C=12,EO=4
alcohol sulfate c=12

Fig. 14.3. Number of stable formulations out of ten achieved for each surfactant combination when
nonionic surfactants with different tail lengthsare usedto emulsify canola oil (Stabilitymetrics measured
at 25C). (Reprinted with permission from Zhao, F. et al, Environ. Sci. Technol., 40,24, 7930-7937,2006.
Copyright 2006 American Chemical Society)

primary surfactants. Based on these results, surfactants with tail lengths of at least 16
carbon atoms but different head groups were investigated for the effect of head group
size. The results in Fig. 14.4 show that stable formulations were achieved when the
head group in the nonionic surfactant had an EO number of at least 10. One possible
explanation for this is that when the head group is very small (e.g., the number of
EO units is less than lo), the nonionic surfactant is too hydrophobic, leading to the
formation of a relatively low number of large micelles and low surfactant solubility.
As a result the total amount of oil solubilization is small and no stable formulation
can be achieved. When the size of the head group is too large, however, such as for
the ethoxylated glyceryl ester with C=18 and EO=40, stable formulations were only
observed when a very high concentration of nonionic surfactant was present (e.g.,
oi1:nonionic surfactant molar ratio of at least 1:3.5). As the head group size increases,
the micelle size decreases, leading in turn to a reduced micelle solubilization capacity,
?he micelles on the whole become more hydrophilic and surfactant solubility increases,
leading to stable formulations only if high surfactant concentrations are utilized.

Effect of Co-surfactant.
Figure 14.5 shows the number of stable formulations produced when the ethoxylated
glyceryl esters, ethoxylated alcohols, and polysorbitan esters with tail lengths longer
than 16 carbon atoms are paired with various co-surfactants (either anionic or

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 403

Co-Surfactant
L
0)

E,C

illfo-carboxylic ester C=15

alkane sutfonate C=14
alcohol ether suffate C=12,EO=4
alcohol sulfate C=12
fattyacid soap C=12,EO=15
alkyi aryl disuifonate C=10

Fig. 14.4. Number of stable formulations out of ten achieved for each surfactant combination when
nonionic surfactants with different head group size are used to emulsify canola oil (Stability metrics
measured at 259). Only successful surfactant combinations from Fig. 14.3 are considered. (Reprinted
with permission from Zhao, F. et al, Environ. Sci. Technol., 40, 24, 7930-7937, 2006. Copyright 2006
American Chemical Society)

nonionic) representing different head structures and tail lengths. The results illustrated
in Fig. 14.5A reveal that at most one stable formulation out of ten is achieved when
the co-surfactant used has a small head group, corresponding to low water solubility.
The results illustrated in Fig. 14.5B reveal that with the same primary surfactant, the
number of stable formulations out of the ten defined in Fig. 14.2 is larger when the
co-surfactant tail length is close to that of the primary (e.g., the tail length difference
is less than 6) than when the tail length is significantly different from the tail length
of the primary (e.g., the difference is larger than 6).
The data suggest that when the oil-soluble surfactant is applied alone, the oil
solubility is limited by the surfactant (micelle) solubility in water. When the watersoluble co-surfactant is applied alone, the oil solubility is limited by the micelle
solubilization capacity. When a mixture of one oil-soluble and one water-soluble
surfactant is applied, molecules of the two surfactants can form mixed micelles with
increased surfactant solubility and unchanged micelle size, which in turn leads to a
significant increase in the oil solubility.

404 0 F. Zhao et al.

A.

Co-Surfactant: Anionic or Nonionic

Fig. 14.5. Effect of co-surfactant on number of stable formulations achieved from each surfactant
combination (Stability metrics measuredat 25C). (Reprinted with permissionfrom Zhao, F. et al, Environ.
Sci.Technol., 40,24,7930-7937,2006. Copyright 2006 American Chemical Society)

Design ofvegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants0 405

Efect of Vegtuble Oil Type

In all of the above experiments, canola oil was used as the dispersed phase. To
investigate the impact of vegetable oil type on stability, 60 of the surfactant systems
using ethoxylated alcohols, ethoxylated glyceryl esters, and polysorbitan esters as the
primary surfactant were randomly selected from the 150 systems tested for canola oil.
For each, the number of stable formulations out of ten was determined for soybean oil
and TMP ester and compared with the corresponding number of stable formulations
observed for canola oil. As shown in Table 14.4, there were at most two formulations
out of ten that showed a difference from canola oil. Over 80% of the combinations
either showed identical formulation stability or had only one formulation out ten that
showed a stability difference. This is expected since all three vegetable oils have similar
molecular structures and fatty acid compositions. Considering that the TMP ester is
less similar to the other two oils due to its additional ethyl group, the larger difference
in the number of stable formulations out of ten could be expected between TMP ester
and canola oil relative to the difference between soybean oil and canola oil.

Surjactant Selection Guidelines

Based on the results presented in Fig. 14.3-14.5, the following guidelines can be used
to formulate vegetable oil semi-synthetic MWF:
1. Surfactant systems consisting of one nonionic surfactant and one water-soluble
co-surfactant (either anionic or nonionic) are preferable to single surfactant

systems in order to reduce the amount of surfactant needed.

2. A nonionic surfactant with a tail length of at least 16 and an intermediate size

of head group, such as ethoxylated glyceryl ester with C=18 and EO=20, is
preferable.

3. The co-surfactant should have a tail length similar

to that of the nonionic

surfactant (e.g., the tail length difference should be less than 6).

Since only the anionic surfactants of the sulfonate class are currently manufactured
exclusively from petroleum feedstock, one should avoid the use the sulfonates (currently
the most popular class) when selecting biobased surfactants for vegetable oil M W . To
replace sulfonates, an anionic surfactant from the classes of fatty acid soaps, alcohol
sulfates, alcohol ether sulfates, or sulfo-carboxylic esters can be used. Alternatively,
one can use a suitable nonionic surfactant available from any of the surfactant classes
considered here, taking into consideration that MWF microemulsions formulated with
only nonionic surfactants are difficult to demulsify at the end of life and can increase
wastewater treatment cost (Byers, 2006). Consequently, these results illustrate that the
combinations of vegetable oil and biobased surfactant systems investigated here can
serve as a starting point for the development of 100% petroleum-free formulations,
and help facilitate the transition from petroleum-based MWF to renewable biobased
MWF in the machine-tool industry.

Table 14.4. Similarity of MicroemulsionStability for Three Vegetable Oils Over 10 Formulation Points Identified in Fig. 14.2

0
7

Oils compared

Soybean vs Canola

Canola vs. TMP

Soybean vs. TMP

1

# of formulations out of ten

showing stability difference

ethoxylated alcohol
# of combinations=20

I-7l2

>2

>2

I
1

20% 20% 0% 70% 20% 10% 0%

25% 16% 0% 66% 17% 17% 0%

pobsorbitanester
# of combinations=16

total number of systems=60

/ / I / /

87% 13% 0% 0%

62% 25% 13% 0% 75% 12% 13% 0%

I

84% 13% 3% 0% 60% 23%117% 0% 70% 17% 13% 0%

(Stabilitymetrics measured at 25OC)

Reprinted with permissionfrom Zhao, F. et al, Environ. Sci. Technol., 40,24,7930-7937,2006. Copyright 2006 American Chemical Society.

9
nl

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 407

Manufact uring Performance EvaIuat ion

For any newly formulated MWF to have practical significance, manufacturing
performance has to be maintained, if not improved, relative to the state of the
art. Currently there are a variety of tests available for evaluating the performance
of metalworking fluids (Byers, 2006). These tests can be roughly classified into
two groups: one group of tests is based on rubbing and rolling benveen surfaces of
simple geometry and the other group of tests tries to simulate the tool-workpiece
interactions under real conditions. Tests from the first group, such as the Falex Pin
and V-block test (ASTM D 2670 and D3233), the Four Ball Wear test (ASTM D
4172), and the Block on Ring test (ASTM D2714 and D2782), have gained widespread use. However, results from these tests have been found to be poorly correlated
with manufacturing performance under real-world manufacturing conditions. This is
not a surprise since the test conditions are quite different from that of real machining
operations. In fact, much evidence exists in the literature to suggest that one can
only reasonably predict the lubrication performance of MWF in cutting operations
through the use of a real machining operation such as reaming, drilling, or tapping
(Axinte et al., 2003; Zimmerman et al., 2003~).Naturally the closer a test condition
is to the actual manufacturing condition being simulated, the better its prediction.
Of the many MWF performance tests from the second group, the thread cutting
test ASTM D5619, The Standard for Comparing Metal Removal Fluids Using the
Tapping Torque Test Machine, has been gaining wide acceptance among MWF
formulators and end-users because the test offers advantages such as operation
simplicity, cost-effectiveness, short testing time, good repeatability, variable machining
severity, and good correlation with field results (De Chiffre & Belluco 2000). In
general, low tapping torque means long tool life, good surface integrity of the thread,
and an effective metalworking fluid. As stated in ASTM D5619, [the tapping torque
test] method can be used to more accurately predict the lubricating properties of a
metal removal fluid than previously available laboratory scale tests. While the text of
the standard is not reason enough to believe it, there is strong anecdotal evidence to
support its accuracy (Zimmerman et al., 2003~).
To evaluate the comparative performance of M W , such as biobased M W vs.
petroleum-based M W , the tapping torque test developed by Zimmerman et al.
(200%) has been employed here. This tapping torque test is a modified version of
the ASTM D 5619. Comparative tests have been performed using a MegaTap G8
machine tool with 1018 cold-rolled steel workpieces that were predrilled and prereamed with 240 M6 holes (Maras Tool, Schaumburg, IL). The tapping tools used
were titanium nitride-coated high-speed steel taps with GO pitch and 3 straight flutes.
The cutting speed was set to be 1000 rpm (revolutions per minute). Figure 14.6 shows
a representative tapping torque curve as a function of the depth of cut. The torque
data in the plateau region (after full engagement of the tap and before retrieving) was
used for analysis.

408 0 F. Zhao et at.

250
200
n

region to calculate
average torque

9 150

Fig. 14.6. Representative tapping

6
8
10
depth of cut (mm)

12

14

16

torque curve vs. depth of cut.

To minimize the effect of variation on workpiece hardness and tool wear, the
MWF to be tested were randomized over the workpiece. 25 holes were tapped for
each fluid. MWF machining performance is reported as tapping torque efficiency
which is determined by normalizing the statistical average of tapping torque over all
the tapping torque data points collected in the plateau region from all the 25 tests to
that of the benchmark fluid. This protocol eliminates the wide variation in tapping
torque test results reported in the literature (Byers, 2006) and leads to statistically
significant results that are reasonably correlated with field data (Zimmerman et al.,
2003~).
Figure 14.7 shows the tapping torque efficiency of representative formulations
developed here using surfactant systems of ethoxylated glyceryl ester (C=18, EO=20)
and sulfated alcohol (C=12, EO=4) normalized to a representative petroleum semisynthetic MWF with same oil molar concentration. Interestingly, the vegetable
oil-based fluids have a small, but statistically significant improvement on tapping
performance relative to the high-performance petroleum-based MWF. This is
consistent with previous observations by Belluco and De Chiffre (2001 and 2004)
and Clarens et al. (2004) and follows from the higher lubricity of vegetable oils.

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 409

110
n

105

80
Petroleum

Canola

Soybean

TMP Ester

Fig. 14.7. Tapping torque efficiency of vegetable oil MWFs with biobased surfactants. (Reprinted with
permission from Zhao, F. et al, Environ. Sci. Technol., 41, 3, 1016-1023, 2007. Copyright 2007 American
Chemical Society)

Optimization of Surfactant Concentrations for

Maximum Stability of Vegetable Oil MWF
So far we have developed guidelines that can be used to select biobased surfactants
for vegetable oil-based semi-synthetic MWF microernulsions. As seen in Figures
14.3-14.5, stable microemulsions can be achieved with different concentrations
of primary surfactant and co-surfactant. Although all of these microemulsions are
stable with respect to the stability metrics defined earlier (i.e., visual transparence,
light transmittance, and oil droplet size), their capability to resist aggregation and
coalescence caused by external disturbance may be different. Therefore, it becomes
essential to determine whether it is possible to optimize the concentrations of the
primary surfactant and co-surfactant for maximum stability. In other words, the
goal is to maximize resistance to coalescence under external influences by optimizing
surfactant concentrations. In the following subsection, the optimal surfactant
concentrations for a binary surfactant system are defined via the molar ratio of oil
to total surfactants (denoted as O ) and the molar ratio of primary surfactant to cosurfactant (denoted as 0).

410 0 F. Zhao et al.

CoalescenceActivation Energy and Microemulsion Stability

At the microscopic level, microemulsion droplets or swollen micelles are highly
dynamic. They collide, aggregate, coalesce, and then break apart again. Due to the
presence of a surfactant monolayer at the oil-water interfice, microemulsion droplets
have to overcome an energy barrier (or activation energy) before coalescence occurs.
The higher the activation energy is, the higher is the microemulsions ability to resist
coalescence. One way to optimize surfactant concentrations for maximum stability
is to maximize the activation energy through the development of a quantitative
relationship between surfactant concentration and the activation energy. A common
approach to determine the activation energy is through a laser-induced temperature
jump approach (Fletcher et al., 1995). In this approach, energy from a light pulse
is absorbed by the microemulsion and a temperature rise follows. This in turn leads
to an increase of droplet size (volume) due to coalescence, and consequently, the
turbidity increases. By measuring the turbidity change following this temperature
jump, the rate of droplet volume change and the activation energy can be determined
by using the Smoluchowski equation (Yamaguchi et al., 1995).

CoalescenceActivation Energy Estimation Based on Microfiltration

The laser-induced temperature jump method requires precise and fast turbidity and
droplet size measurements, which makes the required instrumentation expensive.
Also,when one anionic or cationic surfactant is used in combination with a nonionic
surfactant, no droplet size change may be observed (Fletcher & Suhling, 1998). In this
case, the temperature jump based method cannot be applied to estimate activation
energy. As an alternative approach, we have developed a method for coalescence
activation energy estimation based on coaxial microfiltration. The method does not
require expensive instruments nor does it depend on the temperature effect. Therefore
it can be used for microemulsions formed by combinations of anionic/cationic and
nonionic surfactants, such as the MWF discussed in this chapter.
Microfiltration belongs to the family of pressure-driven membrane separation
technologies that utilize a semi-permeable barrier or membrane capable ofseparating
desired feed stream components from contaminants according to size exclusion (Ho
& Sirkar, 1992). Membranes with pore size ranging from 0.2 to 5.0 pm are generally
used for microfiltration (Cheryan, 1998). When performing microfiltration of
microemulsions, the permeation rate (also called microfiltration flow rate or flux) of
oil-in-water microemulsions is much lower than the permeation rate of water, even
though the droplet size of the microemulsion is one or two orders of magnitude
smaller than the membrane pore size. This is due to the membrane pores being blocked
by coalesced oil droplets as time evolves. While this lowers the permeation rate of
microemulsions, it also creates an opportunity to estimate the droplet coalescence
activation energy using flow rate data collected from the microfiltration process. It
also implies that microemulsions with higher stability will also have higher filtration
rates, which is usehl to lower costs of microfiltration-based recycling of MWF. Such

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants 0 4 1 1

recycling can cost-effectively increase MWF lifetime significantly and nearly eliminate
manufacturing process contamination from particulates and microorganisms (Skerlos
& Zhao, 2003; Rajagopalan et al., 2004).
To measure the activation energy of microemulsions using microfiltration data, a
model that describes how the coalescence kinetics affects pore-blocking behavior and
eventually the microfiltration flow rate was developed (Zhao et al., 2004, 2007 ). In
short, the microfiltration flow rate at time t can be expressed as:
2

(Eq. 14.5)
where Jo is the microfiltration flow rate measured under the same trans-membrane
pressure when water (instead of MWF microemulsions) is applied, c (l/s) is a
lumped parameter dependent on queuing characteristics and micelle concentration
but independent of surfactant chemistry and concentrations, AG is the coalescence
activation energy, Rmais the maximum equivalent pore radius reduction when steadystate is achieved between adsorption and desorption of surfactants, k, is the surfactant
adsorption rate (L/s/mole), kd is the surfactant desorption rate (l/s), and c is the
total molar surfactant concentration that can be calculated by dividing the oil molar
concentration co by the molar ratio of oil to total surfactants w.
For a binary surfactant mixture, Razavizadeh et al. (2004) suggest that AG can be
expressed as:

AG=G,-

[LG,+
1+8
1 G,+ KRT --In-+e
e
1+e
i+e i+e

l 1n-]I1
i+e 1+e
(Eq. 14.6)

where Go is the threshold free energy when droplet coalescence occurs, G, and G, are
the nominal free energy values when the surfactants are applied individually, and K is a
proportionality constant that determines the synergisticeffect between the surfactants.
When a surfactant mixture is used, the free energy decreases in accordance with lower
surface tension at the oil-water interface. This leads to a higher AG.
In the microfiltration flow rate model above (Equations 5 and 6) there are six
parameters that are dependent on surfactant concentrations. Since these parameters
are not known in advance for a given MWF formulation/membrane combination, six
microfiltration experiments using MWF microemulsion ofdifferent w-qcombinations
are required to calibrate the model. Although only four of the six parameters (G,
G,, G,, and K) are needed to estimate the activation energy using Equation 6, six
experiments are needed since all six parameters are presented in Equation 5, either

41 2 0 F. Zhao et al.

implicitly or explicitly.A parameter estimation method combining Genetic Algorithms

and Sequential Quadratic Programming was developed and implemented for this
purpose. The approach was validated using a number of benchmark optimization test
problems routinely utilized in the literature (Houck, 1995). After G, GI ,G,, and K
are determined, Equation 6 can be used to identify the optimal oi1:surfactant molar
ratio (w) and primary:co-surfactant molar ratio (q)that lead to the highest activation
energy AG and maximum stability. This process is demonstrated in the case study
below.

Case Study of Vegetable Oil-based M WF Microemulsions

To demonstrate how to optimize surfactant concentrations for maximum coalescence
stability, we considered the vegetable oil-based MWF using disulfonate (C=10) and
ethoxylated glyceryl ester (C=18, EO=20) as the binary surfactant mixture. For a
given combination of primary and co-surfactant, stable microemulsions could only
be achieved within defined ranges of o i h r f a c t a n t molar ratio (w) and primary:cosurfactant molar ratio (q)as implied previously by the formulation triangles. Figure
14.8 shows the stable w-q formulation region. In order to reduce total surfactant
usage, the maximum surfactant to oil molar ratio was selected to be 1:2.5. At this
total surfactant level, the foaming tendency of the MWF microemulsion measured
according to ASTM D35 19-C88 is close to the maximum that could be tolerated by a
commercial MWF, Moving to a higher total surfactant concentration (by reducing w
below 2.5) could lead to foaming problems. This might in turn 1) increase operating
costs due to fluid loss, 2) shorten pump life due to cavitation, and 3) reduce the
cooling and lubricating capability of the MWF.
As shown in Fig. 14.8, six points are strategically placed with uniformity
throughout the stable region above w = 2.5. For each of the six w-q combinations
in this sub-region of the plot, microfiltration was repeated once and the flow rate
variation was found to vary by less than 5%. Figure 14.8 lists the six parameters (G,
GI, G, ,K, R m ,and k / kJ estimated using microfiltration flow rate data. With four of
them (G, GI, G2,an8K), the activation energy AGat all w-qcombinations within the
stable regin was determined, with the highest AG (139 kJ/kmol) achieved at w=1:2.5
and q=0.57. Future research is needed to compare and interpret the activation energy
derived from the microfiltration method developed here relative to measurements
produced from other technologies, such as the laser temperature jump method.

From Biobased MWF to Sustainable MWF Systems

Sustainable M WF Systems
Concep-l Approach
We have so far demonstrated how to formulate vegetable oil-based semi-synthetic
MWF using biobased surfactants, and how to optimize surfactant concentrations
to achieve maximum stability and microfiltration flux. Besides oil and surfactants,
generally other components are also present in semi-synthetic MWF formulations

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants 0 4 13

such as couplers, extreme pressure (EP) additives, corrosion inhibitors, and chelating
agents. Research is ongoing to develop biobased alternatives for these components
(Tandy et al., 2004; Pedisic, et al., 2003).
Even with the establishment of 100% biobased formulations, there are still
major challenges to be addressed. While in service, trap oils from leaking hydraulic
systems in machine tools, as well as particulates from the surrounding environment,
accumulate in the fluid (Foltz, 2006; Dick, 2006). In addition, microbial growth and
the chemicals (i.e., biocides) used to control microbial growth in MWF can present
significant occupational health risks to exposed workers (NIOSH, 1998; Mathias,
2006; Passman, 2006; Howell et al., 2006; Sondossi et al., 1999,2001; Skerlos et al.,
2003; Kleber et al., 2004). Also, increasing attention has been paid to bacterial by-

4.0

..

= 2.0
0

I.5
I.o

0.2
0.4
0.6
0.8
1.o
cosurfactant : primary surfactant molar ratia

Fig. 14.8. Stable region for vegetable oil based semi-synthetic MWF formulations using a disulfonate/
glyceryl ester surfactant mixture. (Reprinted with permission from Zhao, F. et al, Environ. Sci.Technol., 41,
3, 1016-1023,2007. Copyright 2007 American Chemical Society)

414 0 F. Zhao et al.

products in MWF, such as endotoxins present in MWF mists (Simpson et al., 2003;
Kreiss & Cox-Gamer, 1997). In total the accumulation of particulates, tramp oil,
microorganisms, heat, water hardness, and water evaporation is known to deteriorate
the quality of metalworking operations over time until the MWF can no longer be
used (Foltz, 2006). At this point, MWF must be disposed of, leading to significant
acquisition and disposal costs. Acquisition, maintenance, and disposal operations
create such high costs that one German study estimated that MWF systems account
for 7-17% of total metals manufacturing costs, an amount significantly higher than
tool costs (Klocke & Eisenblatter, 1997).
The aforementioned economic, environmental, and occupational health concerns
have created an interest to develop sustainable metalworking fluid systems. For aqueous
systems, sustainable metalworking fluids systems minimize life cycle environmental
impact by 1) minimizing the materials, energy, and toxicity of system inputs and
outputs, and 2) maximizing MWF lifespan by maintaining physical, biological, and
chemical parameters within limits appropriate to system function (Skerlos et al.,
2001 b). Figure 14.9 shows a conceptual approach to achieving a sustainable aqueous
MWF system. Basically, a sustainable MWF system can be thought of as the union of
an environmentally benign MWF chemical formulation and an appropriate control
system for this formulation that maximizes the MWF lifetime on the shop floor.
The research presented in this chapter has introduced biobased MWF formulations
that are designed for use with microfiltration, which is one approach to achieving
a more sustainable MWF systems. For the rest of this section, microfiltration as a
contaminant control strategy that maximizes MWF lifetime is discussed.
Although oil skimming, centrifugation, pasteurization, coalescence, settling,
depth filtration, magnetic separation, and flotation technologies have been used for
contaminant removal in the metalworking industry (Dick, 2006), these technologies
have yet to be shown to economically control the range of contaminants present in
MWF. Since the dimension of contaminants present in MWF is usually larger than
0.1 pm (Skerlos et al., 2000a, 2000b; Brandt, 2006), and 10-100 times larger (or
more) than microemulsion droplet sizes, the microfiltration approach shown above
for optimizing surfactant concentrations can also be used to cost effectively extend
the usable life of MWF by removing all major contaminants while allowing only the
hnctional MWF components to pass through (Skerlos et al., 2000a, 200Ob, 2OO1a,
2OO1b; Skerlos & Zhao, 2003; Rajagopalan et al., 2004). In this way microfiltration
can be viewed as a highly advanced form of recycling that features the additional
benefit of bacterial control without a heavy reliance on biocides.

Microfitration- based MWI; Recycling

The economic feasibility of applying microfiltration to MWF is determined by the
process productivity, or flux, which is defined as the volumetric flow rate at which the
feed stream is passed through the membrane divided by the membrane surface area.
Higher flux is always desired in order to achieve better system economics. Although

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 4 15

Sustainable
MWF Formulation
Benign, Renewable Ingredients
Robust to destabilization

Physiochemical
Sensor Application

Bio-sensor
Application

Contaminant Control
Recycling
Treatment Systems

Data Acquisition,

Artificial Intelligence
Control Algorithms
System Optimization

Life Cycle Assessment

& System Economics
Fig. 14.9. Conceptual approach of a sustainable MWF system.

increasing total surfactant concentration can lead to higher microemulsion stability

and less pore blocking, at the same time higher surfactant concentration drives the
adsorption/desorption equilibrium towards the adsorption side and results in pore
constriction. Here we address how to manage this trade-off by investigating how to
select surfactant concentrations for maximum flux. The approach is based on the
microfiltration flux model presented in Equations 5 and 6 that was developed to
optimize surfactant concentrations for maximum stability.
Equation 5 describes flux decline due to pore constriction and pore blocking.
With this equation, the flux can be selected at a point where both pore constriction
and pore blocking have fully developed. In this case, optimal surfactant concentrations
corresponding to maximum flux are determined by taking derivatives on /(t) from

4160F.Zhaoetal.

Equation 5 with respect to w and q, and then solving the following system of
equations:

(Eq. 14.7)
The solution to this set of equations yields the surfactant concentrations that maximize
the microfiltration flux of a given microemulsion at a fixed oil concentration. The
solution to the system of equations in Equation 7 results in the following iteration
formulas that yield optimal values of anionic:nonionic molar ratio (q*) and
oihrfactant molar ratio (w*)for maximizing microfiltration flux:

(Eq. 14.8)
and,

=-

1+9'

9 w , +G,)

4k,RL, X c , ( l + O * )
RTIn
*
x @ + k, l k , C , ) I G , + G , ) @ R ~ , c ,- a *- k, lk,c$
.t

(Eq. 14.9)

Again, six parameters (G, G,, G, , K, R",,,

and k,/ k) are estimated from
microfiltration flow rate data collected in six microfiltration experiments using MWF
microemulsion of different w-q combinations. After estimating the six parameters
from the microfiltration flux data of the six different surfactant concentration
combinations within the stable region (Fig. 14.8), Equations 8 and 9 are used to
calculate the surfactant concentrations that lead to highest flux.
For MWF microemulsions based on disulfonate/glyceryl ester, the optimal
oi1:surfactant ratio (09 is 3.7 and the optimal anionic:nonionic ratio (ex) is 0.44
(this formulation is called YO in Fig. 14.10). For YO,the flux predicted by the model

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants0 41 7

4.0

.0

2.0
I.5
1.o
0.2
0.4
0.6
0.8
1.o
cosurfactant : primary surfactant molar ratio

Fig. 14.10. Flux Optimization for MWF formulations using a disulfonate/glyceryl ester surfactant mixture.
(Reprinted with permission from Zhao, F. et al, Environ. Sci.Technol.,41, 3, 1016-1023, 2007. Copyright
2007 American Chemical Society)

agrees with experimental results to within 7%, with flux of 3110 LMH (L/m2/
hr). This flux is higher than any of the six formulations originally selected within
the stable region, which have flux values ranging from 1800 LMH to 2810 LMH.
These values are well above the "rule-of-thumb" reference value of 100 LMH that
generally leads to the economical application of microfiltration to MWF for advanced
recycling and contaminant control (Skerlos & Zhao, 2003). Moreover, flux of the
optimized formulation YO is about three times higher than that of the benchmark
semi-synthetic MWF, which is petroleum oil based with the composition given in
Table 14.1. Economic analysis of general microfiltration implementation strategies
for MWF recycling has indicated that the increase of microfiltration flux can lead to
significant reductions of overall system cost (Skerlos & Zhao, 2003).

418 0 F. Zhaoet al.

Conclusions
While it would seem desirable to eliminate MWF in all manufacturing processes to
address their economic, environmental, and occupational health concerns, research
over the past two decades has demonstrated that this goal is extremely challenging and
perhaps impossible (Skerlos et al., 2008). Although machining under a completely
dry condition, or applying sprays of compressed air and a small amount of oil
(called minimum quantity lubrication), has proven successful in certain operations,
these approaches still face significant challenges in severe operations (Filipovic &
Stephenson, 2006; Li & Liang, 2007). For instance, in machining processes such
as titanium alloy milling, gun drilling, and compacted graphite iron boring, the
absence of cooling provided by MWF results in accelerated tool wear, residual stresses,
dimensional errors, and poor surface finish (Skerlos et al., 2008). While research
proceeds with the goal of addressing these concerns without the use of MWF, aqueous
MWF microemulsions will continue to be used all over the world. This chapter has
illustrated that combining biobased formulations with advanced recycling techniques
can prove to be an economically and environmentally sound approach until such
time as superior M W replacements with lower environmental and health costs are
found.

Acknowledgements
The authors would like to express their appreciation to the faculty, graduate students,
and undergraduate students who contributed to the results presented here: Professor
Julie Zimmerman (Yale, University), Professor Andres Clarens (University ofVirginia),
Ye Eun Park, Carlos Aguilar, Heather Landis, Ashley Murphree, Ashley Earle, David
Delind, and Marcy Urbance. The authors also appreciate the financial support from
the US National Science Foundation (DMII-0084796 and DMII-00935 14),the US
Environmental Protection Agency (R831457),
and Ford Motor Company.
Any opinions, jndings, and conclusions or recommenhtions expressed in this material
are those of the author@ and do not necessarily reject the views of the National Science
Foundation or the Environmental Protection Agency.

References
Abdalla, H.S.;
S. Patel. The Performance and Oxidation Stability of Sustainable Metalworking
Fluid Derived from Vegetable Extracts. Proc. Inst. Mech. Eng. Part B. J Ens Man. 2006, 220,
2027-2040.
Adamson, A.W.; A.P. Gast. Physical Chemistry ofsurfaces, 6th edition, John Wiley & Sons, Inc.,
New York, 1997.
Axinte, D.; M. Axinte; J.D.T. Tannock. A Multicriteria Model for Cutting Fluid Evaluation. Proc.
Inst. Mech. Eng., Part B:]. Eng. Man. 2003,217, 1341-1353.
Ash, M.; E. Dohlrnan;. Oil Crops Situation and Outlook Yearbook 2005, Economic Research Service, United States Department of Agriculture, Washington, DC, 2005.

Design of Vegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 4 19

Barn, W.J. Lubricants and the Environment. Trib. Internat. 1998,31, 35-47.
Belluco, W.; L. De Chiffre. Performance Evaluation of Vegetable-Based Oils in Drilling Austenitic
Stainless Steel. ].Mat. Proc. Tech. 2004, 148, 171-176.
Belluco, W.; L.De Chiffre. Testing of Vegetable-Based Cutting Fluids by Hole M&ing Operations,
Lub. Eng. 2001,57, 12-16.
Brandt, RH. Filtration Systems for Metalworking Fluids. Metalworking Fluids, 2nd Edn., Byers,
J.P Ed.; Taylor & Francis: New York, 2006; pp. 231-252.
Byers, JJ? Laboratory Evaluation of Metalworking Fluids. Metalworking Fluid, 2nd Edn., Byers,
JJ? Ed.; Taylor & Francis: New York, 2006; pp. 147-174.
Canter, N. H L B A New System for Water-Based Metalworking Fluids. Trib. Lubr. Tech. 2005,
GI,10-12.
Cheryan, M. Ultrafiration and Microf%ration Handbook, 2nd Edition, CRC Press, Boca Raton,
FL, 1998, Chapter 1, pp 1-30.
Childers, J.C. The Chemistry of Metalworking Fluids. Metalworking Fluid, 2nd Edn. Byers, JJ?
Ed., Taylor & Francis: New York, 2006; pp. 127-146.
Clarens, AX; J.B. Zimmerman; H.R. Landis; K.F. Hayes; S.J. Skerlos. Experimental Comparison
of Vegetable and Petroleum Base Oils in Metalworking Fluids Using the Tapping Torque Test.
Proceedings of the japan/USA Symposium on Flexible Manufacturing, Thomas Kurfess Ed.; Georgia
Institute ofTechnology. Denver, Colorado, July 19-21,2004.
Cohen, H.; E.M. White. Metalworking Fluid Mist Occupational Exposure Limits: A Discussion
of Alternative Methods. /. Occ. Environ. Hyg. 2006,3, 501-507.
De Chiffre, L.; W. Belluco. Comparison of Methods for Cutting Fluid Performance Testing. Ann.
CIRP 2000,49, 5 7 4 0 .
Dick, R.M. Recycling of Metalworking Fluids. Metalworking Fluids, 2nd Edn. Byers, J.E Ed.
Taylor & Francis: New York, 2006; pp. 279-300.
Filipovic, A.; D.A. Stephenson. Minimum Quantity Lubrication (MQL) Applications in Automotive Power-Train Machining. Mach. Sci. Tech. 2006, IO,3-22.
Fletcher, ED.; J.S. Morris. Turbidity of Oil-in-Water Microemulsion Droplets Stabilized by Nonionic Surfactants. ColL Su$ A.: Physiocochem EngAspect 1995, 98, 147-154.
Fletcher, PD.; K. Suhling. Interactions Between Weakly Charged Oil-in-Water Microemulsion
Droplets. Langmuir. 1998, 14,4065-4069.
Foltz, G.J. Metalworking Fluid Management and Troubleshooting. Metalworking Fluid, 2nd
Edn., Byers, J.E Ed., Taylor & Francis: New York, 2006; pp. 253-278.
Griffin, W.C. Classification of Surface Active Agents by HLB. ]. SOC.Cosm. Chem. 1949, I ,
31 1-326
Hauthal, H.G. CESIO 2004-Dynamic Surfactants and Nanostructured Surfaces for an Innovative Industry, Gth World Surfactants Congress, Berlin, 2004. SOW-J. 2004, 130, 3-17.
Hiemenz, EC.; R Rajagopalan. Principles ofcolloid and Sulface Chemistg grd Edn. Marcel
Dekker: New York, 1997. pp. 355-404.
Ho, W.S.W.; K.K. Sirkar. Membrane Handbook. Van Nostrand Reinhold: New York, 1992. pp.
1-16.

420 0 F. Zhao et al.

Houck, C., et al., A Genetic Algorithm for Function Optimization: A Matlab Implementation,
Technical Report No. NCSU-IE TR 95-09, published by North Carolina State University,
Raleigh, NC. 1995. pp 1-14
Howell, J.K.; WE. Lucke; E.M. White. Health and Safety Aspects in the Use of Metalworking
Fluids. Metalworking Fluid, 2nd Edn. Byers, J.l? Ed. Taylor & Francis: New York, 2006; pp.
337-376.
Institute of Advanced Manufacturing Sciences Incorporated, IAMS. Pollution Prevention Guide to
Using Metal Removal Fluid in Machining Operations. IAMS: Cincinnati, OH, 1995. pp. 46-58.
Iowa Waste Reduction Center. Cutting Fluid Management for Small Machining Operations: A Practical Pollution Prevention Guide, grd edn. University of Northern Iowa, 2003. pp. 5-13.
John, J.; M. Bhattacharya; PC. Raynor. Emulsions Containing Vegetable Oils for Cutting Fluid
Application. Coll. Sut$ A.: Physiocochem Eng Aspect 2004,237, 141-1 50.
Kibbey, T.; K.F. Hayes. Predictive Numerical Thermodynamic Model of Mixed Nonionic Surfactant Sorption in Natural Systems. 2. Application to Broadly Distributed Mixtures.J Coll. Inter.
Sci. 1998, 197, 210-220.
Kleber, M.; W. Follmann; M. Blaszkewicz. Assessing the Genotoxicity of Industrial Cutting Fluids
under Conditions of Use. Toxicol. Lett. 2004, 151,211-217
Klocke, F,; G. Eisenblatter. Dry Cutting. Ann. CIPl? 1997, 46 1-8.
Kreiss, K.; G. Cox-Ganser. Metalworking Fluid-Associated Hypersensitivity Pneumonitis: A
Workshop Summary. Am. J Ind. Med. 1997,32,423-432.
Li, K-M; S.Y. Liang. Performance Profiling of Minimum Quantity Lubrication in Machining.
Internat. J Adv. Man. Techn. 2007, 35, 226-233.
Mang, T. Lubricants and Their Market. Lubricants and Lubrication, 2nd Edn. Mang, T.; W Dresel
Eds., Wiley-VCH: Berlin, 20071 pp. 1-6.
Mathias, C.G.T. Contact Dermatitis and Metalworking Fluids. Metalworking Fluid, 2nd Edn.
Byers, J.P Ed. Taylor & Francis: New York, 2006; pp. 325-336.
McCoy, J.S. Introduction: Tracing the Historical Development of Metalworking Fluids. Metalworking Fluid, 2nd Edn. Byers, J.l? Ed. Taylor & Francis: New York, 2006; pp. 1-18.
Morrison, I.D.; S . Ross. Colloidal Dispersions: Swpensions, Emulsions, and Foams. John Wiley &
Sons: Inc., New York, 2002. pp. 420-455.
Myers, D. Surfctant Science and Tecbnologv, jrdEdn. Wiley-Interscience: Hoboken NJ, 2006; pp.
1-28.
NIOSH, National Institute of Occupational Safety and Health, Criteriafor a Recommended Standard: Occupationalfiposure to Metalworking Fluid. NIOSH: Cincinnati, OH, 1998; Chap.4-5.
http://www.cdc.gov/ nioshl98- 1O2.html
Oliveira, J.F.G.; S.M. Alves. Development of Environmentally Friendly Fluid for CBN Grinding.
CIRP Ann., 2006,55,343-346
Passman, F.J. Microbiology of Metalworking Fluids. Metalworking Fluid, 2nd Edn. Byers, J.l? Ed.
Taylor & Francis: New York, 2006; pp. 195-230.
Patel, M. Surfactants Based on Renewable Raw Materials: Carbon Dioxide Reduction Potential
and Policies and Measures for the European Union. J Ind. Ecol. 2004, 7, 47-62.

Design ofvegetable Oil Metalworking Fluid Microemulsions Using Biobased Surfactants 0 42 1

Paul, B.K.; S.P. Moulik. Microemulsions: An 0verview.J Dispers. Sci. Technol. 1997, 18,
30 1-367.
PediSiC, L.; M. Sarit; S. Bielen. Application Possibilities of New AW/EP Additive Types in Watermiscible Metalworking Fluids. Ind. Lubr. Trib. 2003,55, 23-31.
Rajagopalan, N.; T. Rusk; M. Dianovsky. Purification of Semi-synthetic Metalworking Fluids by
Microfiltration. Eib. Lub. Tech. 2004, GO, 38-44.
Raynor, P.C.; S.W. Kim; M. Bhattacharya. Mist Generation from Metalworking Fluids Formulated Using Vegetable Oils. Ann. Occ. Hyg.2005, 47,283-293.
Razavizadeh, B.M.; M. Mousavi-Khoshdel; H. Gharibi; R. Behjatmanesh-Ardakani; S. Javadian;
B. Sohrabi. Thermodynamic Studies of Mixed IoniclNonionic Surfactant Systems.]. Colloid
Interface Sci. 2004,276, 197-207.
Rose, B.; P. Rivera. Replacement ofPetroleum Based Hydraulic Fluid with a Soybean BasedAlternative. Sandia National Laboratory Technical Report DE-ACD4-94Al85000. Sandia National
Laboratory: Albuquerque, NM, 1998. pp. 1-10,
Rosen, M.J. Surfactants and Interfacial Phenomena. grd Edn. John Wiley & Sons, Inc.: Hoboken,
New Jersey, 2004. pp. 1-33.
Saouter, E. Natural and Synthetic Surfactants:Which One Is Better? P&G Technical Article,
2003.
Schott, H. Foaming of Nonionic Surfactant Solutions: Effect of Surfactant Concentration and
Temperature.]. Am. OilChem. SOC.1988,65, 816-819.
Shah, D.O.; R.S. Schechter. Improved Oil Recovery by Surfactant and Polymer Flooding. Academic
Press: New York, 1977. pp. 28-54.
Sheng, P.S.;S. Obenvalleney. Life-Cycle Planning of Cutting Fluids-A Review. ASMEJ Manuf:
Sci. Eng. 1997,119,791-800
Shinoda, K.; B. Lindman. Organized Surfactant Systems: Microemulsions. Langmiur 1987,3,
135-149.
Simpson, A.T.; M. Stear; J.A. Groves; M. Piney; S.D. Bradley; S. Stagg; B. Crook. Occupational
Exposure to Metalworking Fluid Mist and Sump Fluid Contaminants. Ann. Occ. Hyg.2003, 47,
17-30
Skerlos, S.J.; A.F. Clarens; K.E Hayes; E Zhao. Current Advances in Sustainable Metalworking
Fluids Research. Inter. J Swt. Man. 2008, in print.
Skerlos, S.J.; F. Zhao. An Optimization Approach to Implementing Microfiltration for Microbial
Control and Recycling in Metalworking Fluids.]. Man. Syst. 2003,22, 202-210.
Skerlos, S.J.; L.A. Skerlos; C.A. Aguilar; F. Zhao. Expeditious Identification and Quantification of
Mycobacteria Species in Metalworking Fluids Using Peptide Nucleic Acids. J Man. Syst. 2803,
22, 137-147.
Skerlos, S.J.; N. Rajagopalan; R.E. DeVor; S.G. Kapoor; V.D. Angspatt. Microfiltration of PolyoxyalkyleneMetalworking Fluid Additives Using Aluminum Oxide Membranes./. Man. Sci.
Eng. Trans. ASME. 2001a, 123,692-699.
Skerlos, S.J.; I! Adriaens; K. Hayes; A. Rihana; K. Kurabayashi; S. Takayama;J. Zimmerman; F.
Zhao. Challenges to Achieving SustainableAqueous Systems: A Case Study in Metalworking

422 0 F. Zhao et al.

Fluids. Proceedings of EcoDesign 2001: 2nd International Symposium on Environmentally Conscious

Design andInverse Manufarturing, IEEE Press: Tokyo, December 13-16,2001b, 1 4 6 153.
Skerlos, S.J.; N. Rajagopalan; RE. DeVor; S.G. Kapoor; V.D. Angspatt. Ingredient-Wise Study
of Flux Characteristics in the Ceramic Membrane Filtration of Uncontaminated Synthetic Metalworking Fluids: Part 1: Experimental Investigation of Flux Dec1ine.J Man. Sci. Eng. Eans.
ASME, 2000a, 122,739-745.
Skerlos, S.J.; N. Rajagopalan; RE. DeVor; S.G. Kapoor; V.D. Angspatt. Ingredient-Wise Study
of Flux Characteristics in the Ceramic Membrane Filtration of Uncontaminated Synthetic Metalworking Fluids: Part 2: Analysis of Underlying Mechanisms. /. Man. Sci. Eng. Trans. ASME,
2000b, 122,746752.
Sondossi, M.; H.W. Rossmoore; R. Williams. Relative Formaldehyde ResistanceAmong Bacterial
Survivors of Biocide-Treated Metalworking Fluid. Inter. Biodeter. Biodegrad 2001,48,286300.
Sondossi, M.; H.W. Rossmoore; E.S. Lashen. Influence of Biocide Treatment Regimen on Resistance Development to Methylchloro/Methylisothiazolone in Pseudomonas aeruginosa, Inter.
Biodeter. Biodegrad. 1999, 43, 85-92.
Soy Stats, Soy Stats 2008, http://www.soystats,com/2008/default.htm,
accessed July 23, 2008.
Stanford, M.; P.M. Lister. The Future Role of Metalworking Fluids in Metal Cutting Operations.
Ind. Lubr. Trib. 2002, 54, 11-19.
Tandy, S.; K. Bossart; R. Mueller; J. Ritschel; L. Hauser; R. Schulin; B. Nowack. Extraction of
Heavy Metals from Soils Using Biodegradable Chelating Agents. Environ. Sci. EchnoL 2004,38,
937-944.
Treiner, C.; M. Nortzi; and C. Vautioni. Micellar Solubilization in Strongly Interacting Binary
Surfactant Systems. Langmuir. 1990, 6, 1211-1216.
US EPA. E$uent Limitations Guidelines and New Source Pe@rmance Standardsfor the Metal
Products and Machinery Point Source Category; Final Rule, 40 CFR Part 438, Federal Register,
Vol. 68, No. 92, 2003. pp. 71014-26. Also available at http://www.epa.gov/fedrgstr/EPA-WATElU2003 /May/Day-13/~4258.pdf
Weiss, J.; J.N. Coupland; D. Brathwaite; D.J. McClements. Influence of Molecular Structure of
Hydrocarbon Emulsion Droplets on Their Solubilization in Nonionic Surfactant Micelles. Coll.
Surf: A: Pbsicochem. Eng. Asp. 1997, 121, 53-60.
Whitby, R.D. Will Regulation Help Improve the Bio-Lubricant Market? Trib. Lubr. Techn. 2004,
I, 60-64.
Yamaguchi, T.; K. Nishizaki; S. Itai; H. Hayashi; H. Ohshima. Physicochemical Characterization
of Parental Lipid Emulsion: Determination of Hamaker Constants and Activation Energy of
Coalescence. Pharma. Res. 1995, 12,342-347.
Zhao, F.; M. Urbance; S.J. Skerlos. Mechanistic Model of Coaxial Microfiltration for a SemiSynthetic Metalworking Fluid Microemulsion.J. Man. Sci. Eng. 2004, 126,435-444,
Zhao, E; A.F. Clarens; S.J. Skerlos. Optimization of Metalworking Fluid Microemulsion Surfactant Concentrations for Microfiltration Recycling. Environ. Sci. Echnol. 2007, 41, 10161023.
Zhao, F.; A.E Clarens; A. Murphrey; K.E Hayes; S.J. Skerlos. Structural Aspects of Surfactant
Selection for the Design of Vegetable Oil Semi-synthetic Metalworking Fluids. Environ. Sci.
Ecbnol. 2006,40,7930-7937.

Design of Vegetable Oil Metalworking Fluid MicroemulsionsUsing Biobased Surfactants 0 423

Zimmerman, J.; K.F. Hayes; S.J.Skerlos. Influence of Ion Type and Concentration on the Emulsion Stability and Machining Performance of Two Semi-Synthetic Metalworking Fluids. Environ.
Sci. Technol. 2003%38, 242-2430.
Zimmerman, J.; A.F. Clarens; K.F. Hayes; S.J.Skerlos. Design of Hard Water Stable Emulsifier
Systems for Petroleum and Biobased Semi-Synthetic Metalworking Fluids. Environ. Sci. 2chnol.
2003b, 37, 5278-5288.
Zimmerman, J.; S . Takahashi; K.F. Hayes; S.J. Skerlos. Experimental and Statistical Design Considerations for Economical Evaluation of Metalworking Fluids Using the Tapping Torque Test.
Lubr. Eng, 2 0 0 3 ~59, 17-24.

Polyol and Amino Acid-based

Biosurfactants, Builders, and
Hydrogels
Kenneth M. Doll and Sevim 2. Erhan
food and lndustrid Oil Research Unit; National Center for Agricultural UtilizationResearch, United States
Department ofAgriculture,Agricultural Research Service, 18 75 N. UniversitySt., Peoria, /L 6 1604 USA

Introduction
A modern detergent is a meticulously balanced blend of surfactants (to solubilize and
emulsify), builders (to remove hard water deposits), bleaches (to treat stains), and
other ingredients (Adami, 2004). According to the Soap and Detergent Association,
the combined global market of laundry, personal care, and dishwashing detergent was
valued at $70 billion in 2005 with growth projected to $78 billion by 2010 (SDA,
2007b). The United States alone spent over $12 billion on household detergents in
2005 (SDA, 2007b). With this market size comes tremendous competition where
over 200 companies are competing for consumer dollars (Spitz, 2004). Successful
products not only have to work, but they must have some innovative advantage over
competitive offerings (McCoy, 2006b).
In todays world, the use of natural products in the detergent industry makes
sense for environmental, economic, and regulatory reasons (Hogue, 2006; McCoy,
2008; Short, 2007; Suszkiw, 2007). Originally, all surfactants were made from natural
materials (Rosen, 2004; Spitz, 2004), where these carboxylate-based soaps were used
for thousands of years. However, the 20rhcentury saw the rise of synthetic detergents
which gave superior performance in hard water due to their better foaming capability
and lower tendency to leave deposits. Other advantages of synthetic detergents
were favorable pH stability, ease of manufacture, and cost. Due to these advantages,
synthetic detergents had captured the majority of some markets before 1955. This
trend continued to the end of the 20th century, where the European production of
traditional soap dropped nearly in half between 1972 and 2000 (Berna et al., 2004).
However, there are reasons why the trend may be slowly changing.

425

426 0 K.M. Doll and S.Z. Erhan

Environmental Factors
Water Quality
Along with their success, synthetic detergents have had negative environmental
impacts. The contamination of major rivers in the United States and Europe was
a high profile example of this problem (Sullivan & Evans, 1968). At the height
of this problem in the early 1960s, the water of the Illinois and Mississippi rivers,
large rivers in the United States, was contaminated to the point where foaming was
becoming a serious problem (Barber et al., 1995). The mean value of the surfactant
concentration of samples taken from the Illinois River between 1959-1965 was 0.54
mg L'. Following a voluntary industry switch to more biodegradable surfactants in
1965, the surfactant level measured in the river went down to a more reasonable value
of 0.05 mg L-' by 1968 (Sullivan & Evans, 1968; Sullivan & Swisher, 1969). Further
improvements were mandated by the Federal Water Pollution Control Act of 1972
and the series of following amendments, Clean Water Acts, and Water Quality Acts.
Detergent builders, which function to remove metal ions from solution and inhibit
the formation ofprecipitates, have had negative high profile water quality issues as well.
The common builders in the detergents of the 1950s and 1960s were polyphosphate
compounds such as sodium diphosphate and sodium triphosphate. 'The problem with
these compounds was not that they couldn't be utilized by the environment, but that
they were too readily used by the environment. In many environmental systems, a low
level of inorganic phosphate is a key factor in regulating algal growth. When additional
phosphate is added to the environment, algal growth can increase dramatically. 'The
resultant algae decay causes low oxygen levels which can destroy desirable aquatic life
in a process called eutrophication. Because manmade sources contribute 75% of the
inorganic phosphate to the environment (Ferguson, 1968), elimination of phosphates
from synthetic detergents became a target of environmentalists and regulators.
Starting in 1972, major detergent manufactures reduced the level of phosphate used
(Knud-Hansen, 1994) and in 1994, they completely discontinued its use in laundry
detergent in the United States (Zini, 1995). Reduction of phosphate in dishwashing
detergents has been slower, but a recent policy of'The Soap and Detergent Association
will limit phosphate content to 0.5% or less, in all detergents made after July 2010
(SDA, 2007a, 2007b).
Because a detergent builder serves several functions, direct replacement of
phosphate with a single additive has not been achieved. Instead, a combination of
ion-sequestering chelators (Fig. 15.1) with crystallization-suppressing polymers, like
polyacrylic acid, are.used. However, these substitutes also have their own problems.
Polyacrylic acids are usually not biodegradable and some chelators, like nitrilotriacetic
acid, are suspected human carcinogens. There is clearly a further need for innovation
in this area.

Polyols and Amino Acids 0 427

Sodium tripolyphosphate

Sodium citrate

Sodium nitriloacetate (NTA)

Sodium ethylenediaminetetraacetate (EDTA)

Fig. 15.1. The structure of some common detergent builders. Note that effective detergent builders
usually have two or more metal bonding oxygen groups in close proximity to each other.

Biodegradation
Biodegradation can be classified as either primary biodegradation; which means that
the substance degrades to the point where its physical and chemical properties are
altered, or ultimate biodegradation; where the substance is reduced to CO,, H,O,
and inorganic minerals (Struijs & Stoltenkamp, 1994). Biodegradation of a material
will depend on several factors including its chemical structure, its molecular weight,
and its solubility. The poor biodegradability of surfactants which contain branching
in their alkyl chain, such as branched alkylbenzene sulfonates (ABS; Fig. 15.2), caused
the water quality problems in the early 1960s. This problem is made even worse if
the branching is adjacent to the terminal methyl groups in the hydrophobe (Rosen &
Dahanayake, 2000). Poor biodegradability caused an accumulation of the compounds
in the water which were not effectively removed by the water purification systems of
the era. A switch to linear compounds, such as linear alkylbenzene sulfonate (LAS; Fig.
15.2), helped alleviate this problem. It was shown to be is 5-8 times more biodegradable
than ABS (Sullivan & Evans, 1968), a finding which has been confirmed in recent
studies where LAS was found to have a high rate of primary biodegradability in sea
water and an even higher rate in river water (Perales et al., 2007).
The use of natural product based surfactants is an indicator, but not a guarantee,
that the resultant material will be biodegradable. Soybean oil itself is completely
biodegradable and shows greater than 90% biodegradation in less than 5 days under
soil-emulating conditions. Soybean oil which has been modified via heat bodying also

428 0 K.M. Doll and S.Z.Erhan

Sodium 6-dodecylbenzene sulfonate (6-NaDBS)

An A B S surfactant of limited biodegradability

Sodium dodecylbenzene sulfonate

A popular LAS surfactant
Fig. 15.2. The structure of a common detergent alkylbenze sulfonate (ABS) surfactant (top) and an
environmentally more acceptable linear alkylbenzene sulfonate (LAS) surfactant (bottom).

degrades, but at a slower rate (Erhan et al., 1997). However when larger polymeric
substances are made from soybean oil, the type of intramolecular covalent linkage
is important. Biodegradation studies of soybean oil based polymers show that
biodegradability is achieved if degradable linkages such as ester bonds occur, but not
if the linkages are poorly-degradable, such as amine linkages (Shogren et al., 2004).
This points to a strategy where soybean oil based hydrophobes for surfactant use
should be either linear, or have hydrolyzable bonds in their structure. It also suggests
that the biodegradation of a surfactant is a parameter which must be studied before
any material should be discharged into environment.
The polymeric materials used to replace functions of phosphate builders in
detergents also have biodegradability issues. For example, sodium polyacrylate is an
effective dispersing agent and metal ion chelator, with poor biodegradability, whereas
citric acid has rapid biodegradability but it is a poor dispersing agent and only a
fair chelator. The manufacturers must weigh the environmental cost of using a small
amount of a non-biodegradable compound with high performance properties, or
using a larger amount of a biodegradable compound with lower performance (McCoy,
2008).
Detergent builders are another product type where a biobased source of monomer
is suggestive of, but not necessarily a guarantee of, biodegradability. An interesting and

Polyols and Amino Acids 0 429

well studied example is polyaspartic acid. This polymer is formed by the condensation
of aspartic acid, a natural amino acid, which can form a simple linear peptide of
high biodegradability. Because it has two available carboxylate groups, aspartic acid
yields many polycondensation products (Scheme 15.1). When a simple acidic catalyst
like phosphoric acid is used (Chang & Swift, 1999; Swifc, 2002), both carboxylate
groups condense with the same amine group forming an imide structure which
can be hydrolyzed to form linear polyaspartic acid. However, when the synthesis is
conducted thermally without catalyst, the carboxylates can react with different amine
groups forming a branched structure. The branching density of this structure can be
high enough that the product loses biodegradability (Roweton et al., 1997; Swift,
2002). Because of the hidden complexity in this seemingly simple reaction, work on
these polymerization reactions is still being researched. In a recent development, a salt
of aspartic acid was incorporated into the polymerization. This change has resulted
in the production of more soluble products which eases further derivatization of the
polymer (Sikes, 1999; Swift & Redlich, 2005).

Toxicity
Both acute toxicity and environmental toxicity are important considerations for
the surfactant and detergent industries. Alcohol ethoxylates are common nonionic
surfactantsconsumed in the UnitedStatesataboutone-halftherateofanionicsurfactant
consumption. The United States consumed 0.380 MMT of anionic surfactants in
2005 (SDA, 2007b). Both alcohol ethoxylates and akylphenol ethoxylates, depicted
in Fig. 15.3, biodegrade in a relatively short time (Perales et al., 2007). However,

n+m

Scheme 15.1. The synthesis of either linear polyaspartic acid (top),or branched polyaspartic acid/
polysuccinimide (bottom).

430 0 K.M. Doll and S.Z.

Erhan

these compounds illustrate the importance of consideration of the biodegradation

pathway of the compounds. The initial biodegradation of ethoxylates is relatively fast,
but the resultant reaction products may actually be more toxic to fish than the parent
surfactant (Yoshimura, 1986). Other surfactants, such as ethoxysulfonates (Sibila et al.,
2008) have also been scrutinized, but the nonylphenol ethoxylates have resulted in the
most high profile problems. It was determined that the concentration of nonylphenol
formed from the primary biodegradation of nonylphenol ethoxylates builds to levels
which can be considered highly toxic to aquatic life (Giger et al., 1984) and perhaps
unhealthy for humans. This potential for estrogen-like disruption of the endocrine
systems (Watkins, 2006) have led to the development of new detection methods
(Loyo-Rosales et al., 2003), an EU ban on nonphenol ethoxylates in 2005, and recent
consideration of EPA regulations in the United States (Hogue, 2007). Perhaps the use
of natural products, such as sugar-, amino acid-, or glycerol-based surfactants, could
lower the chances of this type of unintended effects in future surfactants.
The acute toxicity of surfactants is also a consideration, especially if they are
ingredients of products that come into direct contact with consumers. This is another
area where nonionic ethylene oxide-based surfactants have potential problems. During

Polyoxyethylene lauryl ether (POE lauryl ether)

an alcohol ethoxylate

Polyoxyethylene stearate (POE stearate)

an ester ethoxylate

R
Nonylphenol ethoxylate (NPE)
various isomers can be present
Fig. 15.3. Common nonionic surfactants including alcohol ethoxylates and ester ethoxylates. Shown are
Brij 30. and Mryj 45*, products of Croda Industrial Specialties* formerly Uniqema. (top two structures,
respectively). A nonylphenol ethoxylate (bottom) which has a nonylphenol group that can be released
during biodegradation. R refers to an n-alkyl group.

Polyolsand Amino Acids 0 43 1

the synthesis of ethoxylates, a potential byproduct, 1,Cdioxane can be formed.

Because of this possibility, it has been recommended that cosmetic products made
by directly polymerizing a fatty alcohol with ethylene oxide should not be applied
to damaged skin (Hooker, 2004). In contrast, an evaluation of 20 nonionic cosmetic
surfactants synthesized from the natural product, glycerol, did not find encounter this
problem. These surfactants were determined to be safe when used in conjunction with
present cosmetic practices (Johnson, 2004).

Energy Consumption
Another major environmental consideration for the detergent industry is to reduce the
amount of energy consumed by washing machines. 'The United States Department of
Energy estimates that about 90% of the energy consumed in washing is used to heat
water (DOE, 2008). Their Energy Star program recommends that consumers operate
washing machines using warm water rather than hot water and use high efficiency
detergents to significantly reduce energy consumption. Proctor and Gamble has
capitalized on this environmental concern by recently developing the Tide Coldwater"
brand. Other companies have patented similar products which may lead to market
growth in this area Achieving equivalent detergent performance using colder water
is difficult for two reasons. First, a surfactant's ability to solubilize oily material is
significantly increased with temperature, due to higher solubility and soil melting.
Second, the deposits which form from hard water are an increased problem in cold
water.

Economic Factors
Of course one needs to keep sight of the reasons which facilitated the rise of synthetic
detergents in the first place, high performance and traditionally low cost. However,
the large increase of oil prices have left a smaller and more costly supply of n-paraffins
for the industry resulting in price increases. For example, over the last year, the LAS
price has increased from = $0.50 per pound to over $0.71 per pound (Graff, 2008).
Although biobased feedstocks are also seeing cost increases in 2008 as well, the larger
increase for petrochemical cost has opened an opportunity for biobased surfactants.
The market has shown that consumer friendly green technologies can be
successful. For example, Seventh Generation (Burlington, Vermont, USA), a company
specializing in environmentally friendly cleaners, experienced 28% growth in 2006
and now commands nearly $100 million in annual sales (Case, 2007; Watkins, 2007).
More traditional detergent companies are also entering the green market. Clorox's
Green Works",and Proctor and Gamble's Pure Essentials" are product lines with big
potential value in 2008. Dow Chemical Company is also supporting the demand
by selling customers on its EcosurP surfactants, made from naturally derived palm
kernel oil.
The Soap and Detergent Association has backed sustainable development as
well, by maintaining a "Sustainability Central" online forum where companies can

432 0 K.M. Doll and S.Z. Erhan

demonstrate their corporate responsibility. Even retailers like Walmart, by encouraging

efficient packaging and environmentally sound ingredients, are marketing the green
aspects of their detergent products, helping to ensure the marketplace for naturally
based detergents (McCoy, 2007,2006a)
However, not all of these products will be successful, as was shown by the
poor sales of Henkel's Persil" laundry detergents which used fatty acid sulfates and
alkylpolygulcosides as its major surfactants. Due to the competitive industry, naturally
based detergents must be able to compete with their counterparts from traditional
industry on performance and cost bases.

Synthesis of a Polyaspartic Acid Polymer Using

Supercritical Carbon Dioxide
Purpose
Aspartic acid, along with glutamic acid, are interesting amino acids due to the
carboxylate groups on their side chains. This allows them to form linear polymers, but
still have an additional hnctional group which can control the polymer's solubility,
hydrophilicity, and other surface active properties. Aspartic acid can be made by
enzymatic or fermentation processes or by chemical treatment of maleic acid with
ammonia. It is a component of the widely used aspartame sweetener with a world
production of more than 7000 tons per year (Kumagai, 2006). Polymers of this amino
acid have been made in the past, with potential applications as absorbing molecules,
dispersants, or detergent additives (Chang & Swifi, 1999; Roweton et al., 1997).
Currently, acrylic polymers are used in these applications. Even though they are
water soluble, these acrylics are generally not biodegradable at molecular weights over
1000 Daltons (Swifi, 2002). Because the ability of a polyacrylate to sequester Cat*
ions nearly doubles as the molecular weight is increased from 1000 to 4500 Daltons
(Zini, 1995), manufacturers must chose between builders that posses either high
performance (i.e., high molecular weight), or high biodegradability (low molecular
weight).
Aspartic acid has three groups that can participate in polymerization: an amino
group and two carboxylate groups. A linear condensation polymer of aspartic acid
will have 1 free carboxylate for every 3 backbone atoms. Although this is less than
the maximum ratio of 1 free carboxylate for every 2 backbone atoms possessed by
polyacrylic acid, the overall structure is similar. More importantly, because it has
amide linkages in its backbone, natural polyaspartic acid is considered completely
biodegradable. However, as mentioned in the introduction, the non-catalyzed
condensation of aspartic acid (Scheme 15.1) does not produce just the linear form,
but also the branched form which is not completely biodegradable.
Another complication of the thermal synthesis is that both carboxylates can react
with the amine group of another monomer which will form polysuccinimide (PSI).
This material must be hydrolyzed with sodium hydroxide in order to form a soluble

Polyols and Amino Acids 0 433

polymer. This inconvenience led to a patented method of synthesizing a soluble form

of polyaspartic acid without first forming the insoluble PSI (Sikes, 1999). In its
preferred embodiment, the method utilizes a pre-established mixture of ammonium
or sodium salts under conditions where, instead of PSI or branched compounds,
soluble linear compounds are formed.
There are several reason why this reaction can be effectively run utilizing
supercritical carbon dioxide as solvent (Doll et al., 2005; Swift et al., 2005). First,
carbon dioxide is effective in the removal ofwater from the system, so neither organic
solvent nor vacuum pressure were required. Second, carbon dioxide was effective in
finely dispersing the reagents. Finally, carbon dioxide could be removed from the
system without distillation resulting in a purification process which only required
some additional drying.

Results and Discussion

Using the patented methodology, a co-poly(succinimide-aspartate) copolymer
product (1:1 molar ratio of succinimide to aspartate) was synthesized. Aspartic acid,
(13.3 g, 0.1 mol ), sodium hydroxide (5.1 mL of 9.83 M solution, 0.05 mol sodium
hydroxide), and ammonium hydroxide (3.25 mL of 15.43 M solution, 0.05 mol
ammonium hydroxide) were all mixed together in 100 mL of water and then dried
at 80C overnight. This material was ground and 5.53 g was transferred to a 450 mL
high pressure reactor (Fig. 15.4) where carbon dioxide was introduced. The reactor
was heated with careful monitoring of the pressure and stirring was set to 400 rpm.
The polymerization was run at 7.63 MPa at 200 "C for 4 h. The resultant product was
titrated to determine its carboxylate content which was 0.45 carboxylate equivalents
per 100 g polymer (Stevens, 1990).
Bulk polymerization of aspartic acid without the patented salt procedure was
also run. In this reaction, L-aspartic acid was simply added to the reactor and the
polymerization was run. Gel permeation chromatography experiments conducted
on the resultant products showed that at 200C, material with an average degree of
polymerization of=36was synthesized. Syntheses run at the lower temperature of 150C
showed little reaction. The same method could also be used to synthesize polyglutamic
acid, or a condensation polymer of adipic acid with triethylenetetramine.
Overall, this system demonstrated a viable alternative method for the synthesis
of these potentially valuable products. The supercritical carbon dioxide was effective
at dispersing the reactants and growing polymer as a suspension and removing the
produced water from the suspended particles through dissolution. This enabled the
running of the polymerization without a vacuum system, and more importantly,
without the use of organic solvents. Further work using supercritical carbon dioxide
may lead to other useful derivatives of these materials as well (Doll et al., 2005; Swift
et al., 2005; Swift & Redlich, 2006).

434 0 K.M. Doll and S.Z. Erhan

Pressure transducer
The rmocou ple
Stainless steel
reactor body

Fig. 15.4. A picture of the reactor used in the synthesis of polyaspartic acid in supercritical CO,.

Aspartic Acid D-Sorbitol Copolymer Builder

Purpose
'The potential advantages of using aspartic acid polymers as detergent builders have
been discussed in the previous chapter. However, the addition of a carbohydrate
component to the system would allow formulators to expand their potential
applications by increasing their molecular weight, yet preserving their solubility.
D-sorbitol is a reduced form of glucose which is commonly used as a polyol in
the multi-billion dollar polymer market. It was chosen as the co-monomer in the
system for several reasons. It is available in large quantities at a reasonably low cost.
Because it is the reduced form of glucose, it has 6 alcohol groups, all of which are
potentially available for covalent attachment. It can form a 1,4-anhydride, or even
a 1,4-3,6-dianhydride (Scheme 15.2), yet still have reactive alcohol groups available
for esterification. Finally, a copolymer of L-aspartic acid and D-sorbitol had not been
reported prior to the authors' work (Swifi et al., 2007).

Results and Discussion

An acid-catalyzed thermal method was initially chosen for the synthesis of polyaspartic
acid-sorbitol copolymers. The thermal method allowed reaction at all of the hydroxyl
groups, instead of selectively esterifying only the 1 and 6 hydroxyl groups (Scheme
15.3). Using a simple form of an equation developed by Carothers, the percentage of
functional groups which will react before an insoluble gel will form, was calculated
from the molar ratio of L-aspartic acid to D-sorbitol (Stevens, 1990). As shown in
Table 15.1, the calculations show that 100% conversion of the carboxylate groups can
be achieved without gelation in reactions with 1:1 and 5:l molar ratios of monomers,
but a lower % conversion occurs at ratios in-between. 'This allows control over the
desired products. If a more soluble polymer is desired for detergent application, then
a different reagent ratio than that used to make an absorbing gel is called for.

Polyols and Amino Acids 0 435

sorbitan

isosorbide

sorbitol

Table 15.1. The Extent of Reaction at Which an Insoluble Gel i s Expected t o Form in
Thermal Syntheses of L-Aspartic Acid:D-sorbitol Copolymersa
Molar Ratio L-Aspartic Acid:D-Sorbitol

Extent of Reactionwhere Gelation is

Expectedb(%)
100

5:l
100
%e calculationswere done using the Carothers equation (Stevens, 1990).
bThepercent conversion of reactive groups at which an insoluble gel is expected to form. The
product a t this point will have a theoretical degree of polymerization approaching infinity.

436 0 K.M. Doll and S.Z. Erhan

Because D-sorbitol has a convenient melting point, the addition of solvent
was not necessary in this system. D-sorbitol (0.1 mol) was melted at 15OoC, then
L-aspartic acid (0.1 mol for the 1:1 molar ratio experiment) and phosphoric acid
catalyst (0-0.5 molar equivalents) were added with vigorous stirring. The mixture
was heated to 170C under vacuum. The extent of reaction was verified by FT-IR
spectroscopy, which showed a large peak at 1730 cm-, an expected value for an
ester structure. Insoluble materials could be made at polymerization temperatures of
200C. These materials showed water absorbance properties using either pure water
or 0.1 M sodium carbonate solution.
A similar synthesis was performed using ammonium hydroxide catalyst instead
of phosphoric acid. With 1 equivalent of added base, a soluble polymer was formed
when excess sorbitol was used and an insoluble solid was formed when excess aspartic
acid was used. When studied by FT-IR spectroscopy, this insoluble polymer had a
large absorbance at 1716 cm- indicating the presence of the imide structure in PSI.
Overall, the addition of a carbohydrate to this system allowed materials with
considerably different properties to be synthesized. The potential applications for
these materials range from detergents to absorbent fillings for the medical industry.
Because they are made from natural products, they may also be more environmentally
friendly than acrylic acid-based alternatives.

Citric Acid-sorbitol Copolymer Builder

Purpose
Citric acid was also used as the natural carboxylic acid source in a copolymer builder
systems as well (Swift et al., 2007). Citric acid is a well-know component of many
cleaning products already available to consumers giving it a green reputation. Citric
acid presents some advantages over aspartic acid from a chemistry perspective as well.
First, it contains three carboxylate groups instead of only two. Second, it is highly
water soluble, making it easy to prepare citrate and sorbitol solutions. Finally, citric
acid can undergo a dehydration reaction (Scheme 15.4) and form an anhydride which
has a significantly higher reactivity towards alcohols than a normal carboxylic acid.
This allows citric acid to undergo polymerization without additional catalyst.
Although polymers of citric acid and D-sorbitol were patented many years ago,
as additives in the food industry (Centolella & Razor, 1972), most work on citric acid
carbohydrates concentrated on larger acyl acceptor molecules such as starch (Agboola
et al., 1991; Gaffar, 2002; Sikora et al., 1997; Wing, 1996b), cellulose (Wing, 1997),
and corn fiber (Sessa & Wing, 1998, 1999; Wing, 1996a). There is also one patent
where citric acid was reacted with a variety of polyhydroxy compounds, including
sorbitol derivatives (Kappes et al., 1996). From this earlier work, the ability of citric
acid to chelate metal ions, a primary function of detergent builders, is well known.
Citric acid derivatives have demonstrated affinity for Cdt, Cot, Cut, Fet2, Pb+2
Mnt2,Nit2, Ag, and Zn ions (Sessa & Wing, 1999)

Polyols and Amino Acids 0 437

Results and Discussion

As in theasparticacidsystem, the reagent ratio 0facitricacid:D-sorbitol polymerization
reaction is a key variable in determining the properties of the observed products. Not
only does sorbitol have six hydroxy groups, but citric acid also contains a hydroxyl
group as well. Molar ratios of citric acid:sorbitol from 1:l to 6:1, which could
theoretically form structures containing up to 12 free carboxylates (Doll et al., 2006),
were used. The synthesis was run by melting the D-sorbitol, stirring in the citric acid,
and heating to 150C under vacuum (Scheme 15.4). This reaction was followed by
titrating the remaining acidity of samples removed from the reaction at various times.
The product was identified by looking at its FT-IR spectrum, which contained bands
at 1735 cm-' and 1708 cm-', confirmingvarious C=Ostructures in the product. A I3C
NMR experiment also showed citrate esters were present where peaks in the carbonyl
region had shifted upfield from the 6 177.6 and 6 174.2 peaks observed in free citric
acid. The new peaks were broad and in the ranges of 6 176-1 77 and 6 169-1 74.
A comparison of the products made using a 4 hour reaction time (Table 15.2)
showed that as the amount of citric acid increased, the amount of free carboxy groups
increased, as expected; and the solubility of the resultant product also increased,
which was not expected. Gel permeation chromatography on the material showed
an average degree of polymerization benveen 3.5 and 6, depending on the amount of
citric acid that was used.
The Ca+2ion-sequestering ability of the copolymers was tested using an ion
selective electrode. A quantity of 0.65 mmol of Ca+2per g of builder had to be added
in order for the free ion to be present at 0.4 ppm. This is a good level of sequestering,
but not at the level of many commercial builders.
The insoluble part of the samples made at the lower citric acid to sorbitol ratios
was also interesting. The water absorbance index of these samples was taken by adding
water to a sample and stirring for 30 min, then removing the supernatant with a

Citric acid

Citric anhydride

1,4-anyhdrosorbitan
D-sorbitol

Scheme 15.4. A possible structure of the citric acid sorbitol copolymer. Anhydride structures of both
starting materials can form resulting in a complex final product.

438 0 K.M. Doll and S.Z. Erhan

Table 15.2.The Amount of Residual Acid and the Water Solubility and Water Absorbance
Indices o f Citric Acid-Sorbitol Copolymers, Synthesized under Vacuum at 150C for 4 h
Remaining
Water
Carboxylate
Water Solubility Absorbance
Water
at Room
Index Without DH Absorbance Index
Molar Ratio Citric Grows a
Acid:D-Sorbitol
(meq. q-l)
Temperature (W) Adjustmentb
at pH 7
1:l
2.6
66
3.4
2.9
2:1
3.9
51
5.1
11.8
3:1
6.9
54
6.0
9.4
4:1
9.5
100
ND
ND
5:l
10.2
100
ND
ND
61
10.8
100
ND
ND
"Carboxylategroups available per gram of polymer material as determined by titration of
samples with 0.2 M NaOH solution
bSamplesof the products were weighed and then suspended in water for 30 minutes.They
were then centrifuged, the unabsorbed water was removed with a pipette and the sample
was reweighed.The water absorbance index was calculated as the ration of the mass of wet
sample over the weight of the dry sample.
T h i s procedure was identical to the water absorbance index, except the solution was adjusted
to pH 7 with 1 .O M NaOH solution before stirring in the solution for 30 minutes.

pipette. The sample was weighed, dried under vacuum, and weighed again. The ratio
of these weights is the water absorbance index. This procedure was also repeated on
samples where the solution had been adjusted to pH 7 with a 1.O M sodium hydroxide
solution. The results (Table 15.2) show that the copolymer is capable of absorbing
almost 10 times its weight in water. Although this value is too low for application as a
superabsorbing polymer, it is of significant interest for agricultural or pharmaceutical
applications.
In another method (Shogren et al., 2007; Swift et al., 2007), reactive extrusion
was used in place of the vacuum oven synthesis. Citric acid, or sodium salts of
citric acid, were mixed with D-sorbitol and fed through a twin screw extruder.
Experimental parameters were varied including sodium:citrate ratio, sorbito1:citrate
ratio, temperature, feed rate, and extruder screw speed. Most of the samples were
soluble and had molecular weights, measured by light scattering, from 1,080-26,000
Daltons. These products were tested for their ability to inhibit CaCO, precipitation.
The best results were for samples of a molecular weight ~ 8 , 9 0 0Daltons, as determined
by light scattering. They were able to inhibit CaCO, precipitation, from a 0.003 M
CaCI, / 0.003 M NaCO, solution, for at least 10 min at a builder concentration of
only 5-6 ppm. ?his is sufficient for use in formulations which typically use polyacrylic
acid.
Thus natural polycarboxylic acids have potential use in detergent builders (Shogren,
2007), or a variety of other products. It also shows that a natural co-monomer, such
as a reduced sugar, can be used to enhance the properties of the polymer material.
Because they can sequester ions or inhibit crystallization, these types of systems have
a place in the environmentally friendly detergent products of the future.

Polyols and Amino Acids 0 439

Synthesis of a Surfactant from Epoxidized Methyl

Oleate and Glycerol
Purpose
Unlike builders, discussed in the previous three sections, surfactants have been based
on natural products for years. The use of sugar in surfactants dates back to 1885
(Ames, 1960). One recent area of opportunity is in nonionic surfactants. Alkyl
ethoxylates (AE) are often made from large quantities of ethylene oxide. Because
this compound is usually derived from ethylene, its price is affected by many factors
such as petroleum cost and transportation issues. Just last year, a European rail strike
caused a severe shortage of this commodity chemical. The price is now over $0.75 lb',
up from only $0.45 lb', a couple of years ago. With high petroleum prices likely to
continue, the price of ethylene oxide is likely to remain high (Finfacts, 2008).
Glycerol, because it is a coproduct of the large biodiesel market, is significantly
less expensive and will likely remain a cost-effective alternative to ethylene oxide
(Rattay, 2006). Glycerol-based surfactants and emulsifiers are already on the market,
but more study is needed to broaden the use of these versatile molecules. Different
surfactant applications require different physical properties. For example, a good
oil-in-water emulsifier will have a hydrophile to lipophile balance (HLB) between 8
and 18 whereas a water-in-oil emulsifier should have a lower HLB number (Griffin,
1949). Because it has a reactive site which allows further modification, we chose
epoxidized methyl oleate (EMO) as a hydrophobe for our work (Doll & Erhan,
2006). It can be made from either of two commercially viable processes (Scheme
15.5). The first is the epoxidation of methyl oleate, which is straightforward and can
use an inexpensive oxidant like hydrogen peroxide (Findley et al., 1945; La Scala &
Wool, 2002; Nowak et al., 2004; Schmits &Wallace, 1954). The second is through
the transesterification of commercially available epoxidized soybean oil (ESO, Holser,
2008) which is the same reaction type commonly practiced at biodiesel plants. Either
Methanol
'

)-. Catalyst
Purification

I,*

Methyl oleate

Vegetable oil

I
"

Methanol
Catalyst

H f

Formic Acid

,~
"

"C

Purification

Epoxidized Soybean Oil (ESO)

Scheme 15.5. Two alternative ways to synthesize epoxidized methyl oleate (EMO) from soybean oil.
Either a transesterification of commercially available epoxidizedsoybean oil (bottom), or an epoxidation
of methyl oleate (top).The two methods utilize the same chemistries, but with reversed order of steps.

440 0 K.M. Doll and S.Z. Erhan

method produces a versatile hydrophobic molecule which can be easily attached to
a hydrophilic polyglyceride chain. The epoxide functionality can be left available for
further derivatization increasing the potential range of properties available in this
system.

Results and Discussion

Glycerol was first polymerized to form the hydrophobic part of the surfactant. It was
stirred with sodium hydroxide (10 wt %) and heated to 140C for 2 h. The resultant
oligomer had a viscosity of 23,000 f 3,000 mPa s as measured by a Brookfield
viscometer at room temperature ( ~ 2 3 C )E. M 0 was prepared by the epoxidation
of methyl oleate using hydrogen peroxide and formic acid catalyst (Doll & Erhan,
2005). Varied amounts of E M 0 were then added to the polymer which was heated
to 70C for 10-16 hrs without any additional catalyst (Scheme 15.6). The resultant
surfactants were viscous gels that could reduce aqueous surface tension. Their HLB
was measured by the following method and then they were tested in emulsification
applications.
The amount of glycerol incorporated into each surfactant was calculated from
integration of the 'H NMR spectra of the samples. They ranged from 2- 7 glyceride
units. The minimum surface tension which could be achieved in an aqueous solution
(Table 15.3) was measured by the duNouy ring method and found to be around 34
mN m-l. The HLB was measured by a literature method (Piispanen et al., 2004) which
tests the ability of a surfactant to solubilize water in an organic mixture of ethylene
glycol, dimethyl ether, and toluene. The resultant HLB, from 7-13, correlated well
with values calculated (Griffin, 1954; Rosen, 2004) from molecular weights of the
hydrophobe and hydrophile using Griffin's equation.
These values are similar to AE of the formula C,2Hz5(OCzH4)9-,z,
and are in the
range of surfactants commonly used for oil-in-water emulsifiers. They were tested
for their ability to emulsify soybean oil in water. A 1% SBO in water emulsion was
prepared in the small volume sample presentation unit of a Malvern Mastersizer E
particle size analyzer, and the volume mean diameter of the droplets was measured.
Because this presentation unit uses an ordinary propeller at relatively low rpm, it is a
low shear process and leaves the droplets fairly large. Comparison of the results (Table
15.3) with control experiments utilizing commercial nonionic surfictants show that
these surfactants are comparable with an ethylene oxide-based polymeric surfactant,
BASF's Pluronic" L43.
This work shows that there is a potential value for utilizing glycerol as a replacement
for ethylene oxide in emulsification surfactants. The system also merits hrther study
since the range of materials which can be made using the epoxy fatty acyl group have
not been fully explored. Overall, this is another way to increase the use of natural
products and eliminate the world's dependence on petroleum resources.

Polyols and Amino Acids 0 44 1

NaOH

- H20

E M 0 polyglyceride surfactant
Scheme 15.6.The synthesisof a nonionic surfactant based on glycerol and epoxidized methyl oleate (EMO).

ESO Polymerizationto Form a Surfactant or a Hydrogel

Purpose
One effective method of chemically modifying soybean oil is through epoxidation
(Sharma et al., 2006), which removes unsaturation from the soybean oil and increases
its oxidative stability. Another positive attribute of ESO is that it can be polymerized,
either directly to form polyether structures (Scheme 15.7), or with cross-linking agents
to form more complex structures (Liu & Erhan, 2003). The polyether structure can
be hydrolyzed to a soluble material. Using this method, surfactants and hydrogels
based almost entirely on soybean oil can be prepared (Biresaw et al., 2008; Liu &
Erhan, 2007).

442 0 K.M. Doll and S.Z.Erhan

Table 15.3. Physical Properties of the EM0 Polyglyceride Surfactants

Surfactant
EMOGly I f
EMOGly 2f
EMOGly3f
Glycerol
Alone
Pluronic

Volume Mean
Diameter
(pm) of 1%
SBO Droplets,
Surfactant
Concentration

Volume Mean
Diameter
(pm) of 1%
5-60 Droplets,
Surfactant
Concentration
42
420

Minimum
Aqueous
Number of Surface
Glyceride Tension
HLB
HLB
Unitsa
(mN m-l)b Measuredc Calculatedd O.l%e
7.0
5.7

33.9
34.7

2.0
ND

ND
ND

ND

ND

0.5%'

13.1
9.2
9.0

13
12
7

42
42
20

ND

ND

68

18
68

ND

ND

46

66

L43gQ

Caprol
ND
ND
ND
ND
55
1
MPGOh@
'Calculated from NMR
bLowestachievablesurface tension of water using each surfactant
'Measured by a relative solubility number method (Wu et al., 2004)
dCalculatedusing the Griffin equation (Griffin, 1954)
eMeasuredby a Malvern Mastersizer@particle size analyzer
The EM0 gly structures all contain 1 EM0 chain and between 2 and 7 glyceride units.
9Commercial product of BASF
hCommercialproduct of Abitec Corporation

Results and Discussion

Commercially available ESO was used in these experiments. It was dissolved in
methylene chloride and the solution was cooled to 0C (Biresaw et al., 2008). Boron
trifluoride etherate (1.3% wt. with respect to ESO) was added dropwise and the
reaction was stirred for 3 h, resulting in polymerized epoxidized soybean oil. The
hydrolysis of the polyether structure can be done by refluxing in 0.4 M sodium
hydroxide solution for 24 h resulting in hydrolyzed polymerized epoxidized soybean
oil (HPESO).

Sugactant Properties
Low concentration solutions of HPESO are soluble, and the resulting material is an
anionic surfactant. It was studied with sodium, potassium, or tetraethonalammoniom
counterions. All of the systems were able to reduce the surface tension of water to a
range of 20-24 mN m-' as measured by a pendant drop tensiometer at 23C. The
interfacial tension of an aqueous solution of these surfactants with hexadecane were
found to be in the range of 12-17 mN m-'.

Polyols and Amino Acids 0 443

Epoxidized Soybean Oil (ESO)

BF, Purification

Polymerized Epoxidized Soybean Oil (PESO)

NaOH

- glycerol

Hydrolyzed Polymerized Epoxidized Soybean Oil (HPESO)

Scheme 15.7.The synthesis of a soluble polymer from epoxidized soybean oil (ESO).

Controlled Release Properties

A highly interesting and potentially high value application of these polymers is in
controlled drug release (Liu & Erhan, 2007). A study was performed incorporating
the drug, doxorubicin (Dox),into HPESO either as a solution, or as part ofa polymersolid lipid nanoparticle hybrid system (Wong et al., 2006). 'The results show that the
polymer-lipid hybrid nanoparticle system was the most effective at delivery to the
multi-drug resistant cells. The percentage of the drug that was retained by the cells

444 0 K.M. Doll and S.Z. Erhan

after 2 h was nearly twice as high as in those treated with ordinary drug solutions. In
other words, the use of HPESO improved the quality on the solid lipid nanoparticle
drug delivery system resulting in an enhanced performance.
These results have shown that soybean oil can be used in both the high volume
surfactant market and also in the high value pharmaceutical market as well. This is
an area where natural materials not only match the performance of petrochemical
alternatives, but actually exceed them.

Conclusions
Natural materials are well suited for applications in the surfactants, detergents,
absorbents, and pharmaceutical delivery. Other industries, such as fuel and lubrication,
have already replaced a portion of their petroleum consumption with biobased
materials. Although they have many advantages from environmental and sustainability
standpoints, natural materials will only truly gain customer acceptance if they can also
compete on a cost and performance basis as well. Much work remains to be done and
innovative products, such as those discussed in this book, need to be developed before
there is a truly sustainable biobased economy for future generations.
Ihe use of trade, jirm, or corporation names in thispublication isfor the information and
convenience of the reader. Such use does not constitute an official endorsement or approval
by the United States Department OfAgricultureor the Agricultural Research Service of any
product or service to the exclusion of others that may be suitable.

References
Adami, I. Household Detergents Production: Technological Evolution and Trends. Sodeopec,
Soaps, Detergents, Oleochemicals, and Personal Care Products, L. Spitz, Ed.; AOCS Press, Champaign, 2004; 10,288-323.
Agboola, S.O.; J.O. Akingbala; G.B. Oguntimein; 0. State Production of Low Substituted Cassava Starch Acetates and Citrates. Starch-Sti!rke 1991 43, 13-15.
Ames, G.R. Long-Chain Derivative of Monosaccharides and Oligosaccharides. Chem. Rev. 1960

60,541-553.
Barber, I., B. Larry; J.A. Leenheer; W.E. Pereira; T.I. Noyes; G.K. Brown; C.F. Tabor; J.H. Writer
Organic Contamination of the Mississippi River from Municipal and Industrial Wastewater.
US.Geological Survey Circular 1133. http://water,er.usgs.gov/pubs/circ1133/organic.html. Accessed 2008.
Berna, J.L.; C. Bengoechea; A. Moreno. Laundry Products in Bar Form. Sodeopec, Soaps, Detergents, Oleochemicals, and Personal Care Products, L. Spitz, Ed.; AOCS Press, Champaign, 2004;

6,189-211.
Biresaw, G.; Z.S.Liu; S.Z. Erhan. Investigation of the Surface Properties of Polymeric Soaps Obtained by Ring-Opening Polymerization of Epoxidized Soybean Oil. /. Appl. Polym. Sci. 2008
108,1976-1985.
Case, F. The Emergence of Green Chemistry. INFORM2007 17,751-753.

Polyols and Amino Acids 0 445

Centolella, A.P.; B.G. Razor. Polyesters of Citric Acid and Sorbitol. US, 3661955, Polyesters of
Citric Acid and Sorbitol, May 9, 1972.
Chang, C.J.; G. Swift. Poly(AsparticAcid) Hydrogel. ]Macromol Sci Pure Appl Chem 1999 301,
963-970.
DOE, www.energystar.gov. Accessed 2008.
Doll, K.M.; S.Z. Erhan. Synthesis and Performance of Surfactants Based on Epoxidized Methyl
Oleate and Glycerol. /. Sufact, Deterg. 2006 9, 377-383.
Doll, K.M.; S.Z. Erhan. Synthesis of Carbonated Fatty Methyl Esters Using Supercritical Carbon
Dioxide. J Agric. Food Chem. 2005 53,9608-9614.
Doll, K.M.; RL. Shogren; R.A. Holser; J.L. Willett; G. Swift. Polymerization of L-Aspartic Acid
to Polysuccinimide and Copoly(Succinimide-Aspartate)in Supercritical Carbon Dioxide. Let.
Organ. Chem. 2005 2,689-689.
Doll, K.M.; R.L. Shogren; J.L. Willett; G. Swift. Solvent Free Polymerization of Citric Acid and
D-Sorbitol. /. Polym. Sci., Part A: Polym. Chem. 2006 44, 4259-4267.
Erhan, S.Z.; M.O. Bagby; T.C. Nelsen. Statistical Evaluation of Biodegradation of News Ink
Vehicles and Ink Formulations./. Amer. Oil Chem. Soc. 1997 74,707-712.
Ferguson, EA. Features. A Nonmyopic Approach to the Problem of Excess Algal Growths. Environ. Sci. Technol. 1968 2, 188-193.
Findley, T.W.; D. Swern; J.T. Scanlan. Epoxidation of Unsaturated Fatty Materials with Peracetic
Acid in Glacial Acetic Acid Solution.]. Am. Chem. Soc. 1945 67, 412-414.
Finfacts, Big Industrial Users of Energy in Huge Price Hikes; Dow Chemical to Also Idle Plants in
Europe and US. June 25, 2008., http://www.finfacts,ie/irishfinancenews/article-1014008.shtml
(2008).
Gaffar, M.A. Preparation and Utilization of New Carboxyl Group Containing Cation Exchangers
Based on Starch Using a Dry Reaction Method. Starch 2002 54, 185-192.
Giger, W.; EH. Brunner; C. Schdner. 4-Nonylphenol in Sewage Sludge: Accumulation of Toxic
Metabolites from Nonionic Surfactants. Science 1984 225, 623-625.
G r d , G., Petroleum, Plant Oils Send Surfactant Prices through the Roof. January 17, 2008,
http://www.purchasing.comlarticlelCA6518717.html (2008).
Griffin, W.C. Calculation of HLB Values of Non-Ionic Surfactants. /. SOC.Cosm. Chem. 1954 5,
249-256.
Griffin, W.C. Classification of Surface-Active Agents by HLB. .] SOC.Cosm. Chern. 1949 1,
312-326.
Hogue, C. Greener Cleaners. Chem. Eng. News 2006 84, 14.
Hogue, C. Surfactant Scrutiny. Chem. Eng. News 2007 85, 33-35.
Holser, R.A. Transesterification of Epoxidized Soybean Oil to Prepare Epoxy Methyl Esters. Ind.
Crops Prod. 2008 27, 130- 132.
Hooker, E. Final Report of the Amended Safety Assessment of PEG-5, -10, -16, -25, -30, and -40
Soy Sterol. Internat. J Toxicol. 2004 23, 23-47.
Johnson, W., J . Final Report of the Amended Safety Assessment of Glyceryl Laurate, Glyceryl
Laurate Se, Glyceryl Laurate/Oleate Etc. Internat. /. Toxicol. 2004 23, 55-94.

446 0 K.M. Doll and S.Z. Erhan

Kappes, E.; A. Oftring; R. Baur; A. Kud; D. Boeckh; H. Hartmann; V. Schwendemann. USA,
5508394, Citric Esters of Polyhydroxy Compounds and Use Thereof in Detergents, April 6,
1996.
Knud-Hansen, C., Historical Perspective of the Phosphate Detergent Conflict. http://www.colorado.edu/conflict/~ll~text~search/~CRCDo~/94-54.htm/.
Accessed 2008.
Kumagai, H. Amino Acid Production. B e Prokaryotes, a Handbook on the Biology OfBacteria, M.
Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt as.;
Springer, New
York, 2006; Volume 1 : Symbiotic associations, Biotechnology,Applied Microbiology, 756-765.
La Scala, J.; Rl? Wool. Effect of FA Composition on Epoxidation Kinetics ofTAG. ]. Amer. Oil
Chem. SOC.2002 73,373-378.
Liu, Z.; S.Z. Erhan. Soy-based thermosensitivehydrogels for controlled release systems. US,
2007077298 A1 ,Soy-Based Thermosensitive Hydrogels for Controlled Release Systems,
20050930.
Liu, Z. S.; S.Z. Erhan Ihe United States-Japan Cooperative Program in Natural Resources Food and
Agricultural Panel., Tsukuba Ivaraki, Japan, p 394-400. 2003.
Loyo-Rosales, J.E.; I. Schmin-Afonso; C.l? Rice; A. Torrents. Analysis of Octyl- and Nonylpheno1 and Their Ethoxylates in Water and Sediments by Liquid Chromatography/Tandem Mass
Spectrometry.Anal. Chem. 2003 75,4811-4817.
McCoy, M. Going Green. Chem. Eng. News 2007 85, 13.
McCoy, M. Greener Cleaners. Chem. Eng. News 2008 86, 15-23.
McCoy, M. Soaps and Detergents. Chem. Eng. News 2006a 84, 13- 19.
McCoy, M. Surfactant Firms End a Tough Year. Chem. Eng. News 2006b 84,21-22.
Nowak, J.A.; T.A. Zillner; I. Mullin, Leslie Patrick. Thin-film epoxidation of an unsaturated oil or
alkyl fatty acid ester. US, US 6740763, Thin-Film Epoxidation of an Unsaturated Oil or Alkyi
Fatty Acid Ester, May 25,2004.
Perales, J.; M. Manzano; M. Garrido; D. Sales; J. Quiroga. Biodegradation Kinetics of Linear
Alkylbenzene Sulphonates in Sea Water. Biodegrau! 2007 18,63-70.
Piispanen, RS.;M. Persson; l? Claesson; T. Norin. Surface Properties of Surfactants Derived From
Natural Products. Part 1 : Syntheses and Structure/Property Relationships - Solubility and Emulsification.]. Surfact. Deters 2004 7, 147-159.
Rattay, J.B. Glycerine-How Sweet It Is. INFORM2006 17, 285.
Rosen, M.J. Surfactants and Interfacial Phenomena; 3rd. ed.; John Wiley and Sons, Inc.: Hoboken,
2004; 444.
Rosen, M.J.; M. Dahanayake. Industrial Utilization of Surfactants: Principles and Practice; AOCS
Press: Champaign, Illinois, 2000; 176.
Roweton, S.; S.J. Huang; G. Swift. Poly(AsparticAcid): Synthesis, Biodegradation, and Current
Applications.]. Environ. Polym. Degrd. 1997 5, 175-181.
Schmits, W.R.; J.G. Wallace. Epoxidation of Methyl Oleate with Hydrogen Peroxide.]. Arner. Oil
Chem. Soc. 1954 31,363-365.
SDA, 2006-2007 Report to the Membership. www.cleaningIOl.com. 2007a.
SDA, The Soap and Detergent Associations Convention News Daily. Chemical WeekJanuary 31,

Polyols and Amino Acids 0 447

2007, www.chemweek.com, w . c l e a n i n g l 0 l .corn. 2007b.

Sessa, D.J.; R.E. Wing. Metal Chelation of Corn Protein ProductslCitric Acid Derivatives Generated Via Reactive Extrusion. Znd. Crops Prod. 1999 10, 55-63.
Sessa, D.J.; R.E. Wing. Thermochemically-Modified Soybean and Corn Protein Products with
Enhanced Metal-Binding Properties. Nahrung Food 1998 42,266-268.
Sharma, B.K.; A. Adhvaryu; Z. Liu; S.Z. Erhan. Chemical Modification of Vegetable Oils for
Lubricant Applications.]. Amer. Oil Chem. SOC.2006 83, 129-136.
Shogren, R.; S. Gonzala; J.L. Willet; D. Graiver; G. Swift. Preparation of Sorbitol Citrate Polyesters by Reactive Extrusion and Application as Inhibitors of Calcium Carbonate Precipitation./.
Biobased Materials Bioenergy 2007 1,229-237.
Shogren, R.L. Biobased Detergent Builder. Znd Bioprocess 2007 29, 5.
Shogren, R.L.; Z. Petrovic; 2. Liu; S.Z. Erhan. Biodegradation Behavior of Some Vegetable OilBased Polymers. /. Polymers Environ. 2004 12, 173-178.
Short, I?L. New Regulatory Era in Europe Begins. Chem. Eng. News 2007 85,33-35.
Sibila, M.A.; M.C. Garrido; J.A. Perales; J.M. Quiroga. Ekotoxicity and Biodegradability of an
Alkyl Ethoxysulphate Surfactant in Coastal Waters. Sci. Total Environ. 2008 394, 265-274.
Sikes, C.S. hide-free and mixed amidelimide thermal synthesis of polyaspartate. US, 5981691,
hide-Free and Mixed Amidelhide Thermal Synthesis of Polyaspartate, November 9, 1999.
Sikora, M.; I? Tomasik; K. Pielichowski. Chemical Starch Modification in the Field of Microwaves. 11. Reaction with Alpha-Amino and Alpha-HydroxyAcids. POL/. FoodNutr. Sci. 1997
6/47,23-29.
Spitz, L. The History of Soaps and Detergents. So&opec, Soaps, Detergents, Oleochemicals, and
Personal Care Products, L. Spitz, Ed.; AOCS Press: Champaign, 2004; 1 , 1-72.
Stevens, M.I? Polymer Chemistv an Introduction; Second Edn.; Oxford University Press: New
York, 1990; 633.
Struijs, J.; J. Stoltenkamp. Testing Surfactants for Ultimate Biodegradability. Chemosphere 1994
28, 1503-1523.
Sullivan, W; R. Evans. Features. Major U. S. River Reflects Surfactant Changes. Environ. Sci.
Echnol. 1968 2, 194-200.
Sullivan, W.T.; R D . Swisher. Methylene Blue Active Substances and Linear Akylate Sulfonate
Surfactants in the Illinois River, 1968. Environ. Sci. Echnol. 1969 3, 481-483.
Suszkiw, J. Corn: A New Ingredient for Detergents?Agric. Res. 2007 55, 9.
Swift, G. Environmentally BiodegradableWater-Soluble Polymers. Degrahble Polymers, 2; G.
Scott, Ed.; Kulwer Academic Publishers, Netherlands, 2002; 1 1 , 379-412.
Swift, G.; K.M. Doll; R.L. Shogren; R.A. Holser; J.L. Willett. Synthesis of polysuccinimide and
copoly (succinimide-aspartate)in a supercritical fluid. USA, 6887971, Synthesis of Polysuccinimide and Copoly(Succinimide-Aspartate)in a Supercritical Fluid, May 3, 2005.
Swift, G.; G.H. Redlich. Methods of synthesis of polysuccinimide and derivatives thereof, US,
6903 18 1 , Methods of Synthesis of Polysuccinimide, Copolymers of Polysuccinimide and Derivative Thereof, June 7,2005.
Swift, G.; G.H. Redlich. Methods of synthesis of polysuccinimide and derivatives thereof, US,

448 0 K.M.

Doll and S.Z.Erhan

707488 1, Methods of Synthesis of Polysuccinimide, Copolymers of Polysuccinimide and Derivatives Thereof, July 11, 2006.
Swift, G.; D.G. Westmoreland; J.L. Willett; R.L. Shogren; K.M. Doll. Methods of synthesis of
polymers and copolymers from natural products. US, 7256251, August 14,2007.
Watkins, C. Detergents Go Green. INFORM2007 18,780-783.
Watkins, C. Endocrine Disruption. INFORM 2006 17, 6-8.
Wing, R.E. Cellulosic Citrates: Preparation and Ion Exchange Properties. /. Polym. Mater. 1997
14, 303-309.
Wing, R.E. Corn Fiber Citrate: Preparation and Ion-Exchange Properties. Ind. Crops Prod. 1996a
5, 301-305.
Wing, R.E. Starch Citrate: Preparation and Ion Exchange Properties. Starch 1996b 48, 275-279.
Wong, H.L.; A.M. Rauth; R. Bendayan; J.L. Manias; M. Ramaswamy; Z. Liu; S. Erhan; X.Y.
Wu. A New Polymer-Lipid Hybrid Nanoparticle System Increases Cytotoxicity of Doxorubicin
Against Multidrug-Resistant Human Breast Cancer Cells. Pharma. Res. 2006 23, 1574-1 585.
Wu, J.; Y. Xu; T. Dabros; H. Hamza. Development of a Method for Measurement of Relative
Solubility of Nonionic Surfactants. Colloids Surfaces A: Physicochem. Eng. Aspects 2004 232,
229-237.
Yoshimura, K. Biodegradation and Fish Toxicity of Nonionic Surfactants. /. Amer. Oil Chem. SOC.
1986 63, 1590-1 596.
Zini, I? Polymeric Additivesfor High Performing Detergents;Technomic Publishing Company: Basel,
Switzerland, 1995; 273.

Interfacial Properties of
Sugar-based Surfactants
Orlando 1. Rojas, Cosima Stubenrauch: Lucian A. Lucial, and Youssef Habibi
ForestBiomaterials, North Carolina State University,Campus Box 8005,Raleigh, NC27695 and2University
CollegeDublin, School of Chemical and Bioprocess Engineering,Belfield, Dublin 4, Ireland

Introduction
Motivation
Interest is growing in the development of material technologies that are based on
renewable resources. The use of biomass as feedstock is of expanding industrial
significance, particularly in the energy and commodities sectors. Relative to these
arenas, mono- and polysaccharides, very flexible and tunable chemical starting
materials, are estimated to make up three-quarters of the worlds biomass. The efficient
use of this resource is now recognized as a major future objective in a wide range of
technology applications.
The need exists to identify secondary streams of saccharide by-products and to
use these materials as the basis for higher, value-added surfactant chemistries. Several
research groups in the world are creating a platform for advancing the knowledge
on the structure-performance properties and for facilitating the penetration of
sugar-based surfactants in traditional markets dominated by nonrenewable nonionic
surfactants. The general goals of the research work performed in these groups are to
propose and test new structures for targeted applications as well as to provide the basis
for future strategies aimed at enhancing cost-effectiveness. More specific aims involve
both identifying molecular factors that govern the surface activity and facilitating
the design and application of sugar-based surfactants as substitutes for conventional
poly(ethy1ene oxide) (EO) surfactants and others.
Significant groundwork was already done in this area (Balzer & Luders, 2000;
Kiraly & Findenegg, 2000; Kocherbitov et a]., 2002; Kumpulainen et al., 2004a,b,
2005; Liljekvist et al., 2001; Matsson et al., 2004; Muruganathan et al., 2003, 2004;
Nickel et al., 1996; Person et al., 2002, 2003a,b, 2004; Ruiz, 2009; Stubenrauch,
2001). Three classes of surfactants with sugar or a polyol derived from sugar as polar
head group were widely researched: alkyl polyglucosides (APGs) (also known as
alkylpolyglycosides), alkyl glucamides, and sugar esters (Holmberg, 200 1). Both these
surfactants and biosurfactants produced by microorganisms and other surfactants
derived from renewable raw materials are coming progressively onto the market.
449

450 0 O.J. Rojas et al.

Deleu and Michel Paquot (2004) provided an interesting summary of trends, and
reported for APGs a share of ca. 3% of the total production. The reader is referred to
the book edited by Hill et al. (1 997) for more detailed information on this subject.
Overall, APGs have established themselves as natural surfactants of choice for diverse
applications-from emulsifiers for skin care to foaming agents.
The possibility of future development in the area of biofuels from cellulose may
open new opportunities, especially for secondary streams in related processes. We
believe that four critical work packages need pursuing to successfully expand the
potential of sugar-based surfactants from such sources:
Understanding and rationalizing the state-of-the-art in the area of saccharidebased surfactants. The existing, highly fragmented activities need to be
systemized and analyzed to adequately respond to current changes in economic
and environmental aspects of surfactant production and use.
Identifying molecular and structural factors that govern the surface activity of
these materials so that we can facilitate their design and application in areas that
are currently utilizing petroleum-based surfactants. One can support this by
studies with model systems which will expand our understanding and allow us to
target specific applications.
Designing and characterizing new sugar surfactants based on natural products and
on new motifi (Blunk et al., 2006) to demonstrate the economic and technical
viabilities in different applications.
Exploring new synthetic approaches to utilize saccharides for the synthesis of new
surfactants. One could harvest saccharides from biomass processing (and related
secondary streams) or from agricultural/food by-products. One can then combine
this with an understanding of the current and projected economics ofsaccharideand petroleum-based surfactant technologies to facilitate the identification of key
future markets for these materials.
Recent research results show that glucose-based and sucrose-based surfactants have
a range of beneficial physical and performance properties, including high levels of
surfactancy (surface and interfacial activity), very rapid biodegradability, low human
and animal toxicity, effective emulsification properties, and surface interactions.
Therefore, we anticipate that we can rationally design saccharide-based surfactant
structures, centered on the combination of potentially low-cost (based on natural
starting materials), renewable saccharide components and appropriate renewable
(natural fatty acids) hydrophobic pendant groups that will allow the understanding
of the fundamental interactions governing their hnctionalities. Based on this
understanding as well as on the synthetic approaches to generating sugar-based
surfactants, we hope to promote interest in, and the prospects for, their utilization in a
number of surfactant markets, which would ultimately lead to a significant reduction
in the use of petroleum-based materials in these sectors.

Interfacial Properties of Sugar-based Surfactants 0 45 1

Nonionic Surfactants
Nonionic surfactants represent a major component material for applications ranging
from personal care to a wide range of industrial uses. Structurally based on the molecular
combination of hydrophilic and hydrophobic substructures, nonionic surfactants are
effective as wetting and spreading agents, as emulsifiers, and as foaming agents (while
having minimal skin and eye irritation effects). Furthermore, they are characterized
by a wide range of critical secondary-performance properties.
The hydrophilic component of these materials is currently based largely on EO,
which is petroleum-derived. In addition, a significant portion of the hydrophobic
components of these materials is also petroleum-derived. The production costs of
these commodity chemicals are closely linked to the highly volatile petroleum prices,
which one can regard as an additional motivation to search for substitutes. In view
of the large volume in the consumption of EO-based nonionic surfactants, further
replacement of the established market would be very attractive for a high-value
utilization of sugars from biomass.
However, extensive replacement of EO-based surfactants can be challenging,
and a market barrier may exist that prevents more widespread penetration of a new
generation of surfactants. Nevertheless, in our opinion, the time is right to further
advance efforts by starting with high-end applications such as drug emulsification,
pharmaceutical formulations, etcetera, which are industries that require these
technologies and can temporarily tolerate higher prices.
In summary, a need is evident to broaden the application of surfactants to replace
petroleum-based products toward new classes of highly biodegradable, low-irritating,
low-toxic, completely naturally-derived nonionic surfactants. As superior-performance
properties of current and novel sugar-based surfactants are demonstrated, they will
become further established in the future green market.

Environmental and Product Performance Concerns

Environmental and toxicity issues are prompting an increased consideration of a
wider use of sugar-based surfactants. For example, two of the most common types
of surfactants used in large-scale applications are EO alkyl ethers and fatty-acid
soaps. In particular, the use of nonylphenol (NP) ethoxylates is problematic. NP is
a mixture of isomeric monoalkyl phenols, predominantly para-substituted, found in
the environment as a result of the biodegradation of NP ethoxylates. NPs that are
used as nonionic surfactants are released to the environment through various waste
streams in industrial processes.
The National Library of Health reports these surfactants as being severely
irritating to skin and eyes. Vapors cause a slight irritation of the eyes or respiratory
system if present in high concentrations (ILO, 1998; Lewis, 1997; NIOSH, 1983).
Furthermore, NPs are suspected of being endocrine disruptors, that is, they have
adverse effects on the workings of the endocrine system in humans and animals
(Ren et al., 1997). Many European countries have banned the use of NPs. Recent

452 0 O.J. Rojas et al.

EU legislation on the use of NPs (Directive 2003/53/EC) states that NP and NP
ethoxylates may not be placed on the market or used as a substance or constituent
of preparations in concentrations equal to or higher than 0.1% by mass (EPHA,
2005).
Therefore, an important need exists for replacing this and other toxic surfactants
with environmentally-friendly alternatives. Our objective in this chapter is to
introduce the interfacial characteristics of surfactants based on renewable materials
(natural surfactants). In this way, one can evaluate the hndamental properties of
sugar surfactants and consider them for specific applications. Overall, understanding
the adsorption of biobased surfactants will not only solve an important industry and
environmental problem, but will also open an avenue for a number of products with
unique properties that one can generate from biomass.

Significance of Biobased Surfactants

A significant demand is apparent for new, less- expensive, nontoxic, and biodegradable
surface-active materials. Their use, as conventional surfactants, is aimed at stabilizing
liquid-liquid (emulsions), solid-liquid (dispersions), and gas-liquid (foams)
interfaces. The economic impact that the penetration of sugar-based surfactants can
have is mirrored by the substantial consumption of emulsions worldwide in which
the surfactant composition amounts to 0.1-5% of the total mass of the emulsion.
However, the use of surfactants is not limited to emulsions; vast numbers of other
applications involve large quantities of (amphiphilic) stabilizers, including household,
cleaning, food, cosmetic, pharmaceutical, and other specialty products.
Numerous studies recently were carried out to examine the viscoelastic properties,
surface tension, and surface properties of sugar surfactants. Most of those studies point
to the overall feasibility of employing various sugar surfactants to replace their EObased counterparts because of their superior-surface physical responses and application
performance, environmental compatibility, and chemical tunability (functionalization
and modification). Thus, the sections that follow provide a summary of results that
one can use as a basis for the hrther development of sugar-based surfactant systems.
The final goal is to design biobased surfactants with properties superior to those of
the traditionally-used nonionic surfactants and thus to replace the latter whenever
possible and feasible.

Interfacial Properties of Sugar-based Surfactants

Numerous hndamental studies were conducted in the area of sugar-surfactant
properties. Four milestones that serve as recent evidence for the superior characteristics
of sugar-based surfactants compared with EO-based surfactants are discussed below
in terms of XirNater Interfaces, Solid/Liquid Interfaces, Comparison between
EO- and Sugar-Based Surfactants: Structural Aspects, Interfacial Aspects, Packing at
the Interface, and Viscoelasticity Properties of Isomeric Sugar-based Surfactants.

Interfacial Properties of Sugar-basedSurfactants 0 453

Before discussing these subjects, let us briefly outline the tools that were utilized in
these experiments since most of these are not widely available in typical laboratories.
Results from techniques used to measure surface tension, solution rheology, contact
angle, detergency, and emulsion properties are not covered in this chapter. Instead, we
describe the interfacial properties of adsorbed layers of sugar surfactants by using the
methods described below.

Techniques
Measurementsand Analysis of SurfaceInteractionsand Forces (MASIF)
Force measurements were conducted using the noninterferometric surface force
apparatus (Parker et al., 1989, 1994). This device, commonly known as MASIF
(Measurements and Analysis of Surface Interactions and Forces), employs a bimorph
force/deflection sensor that, after calibration of the spring (deflection) constant, yields
the interaction force. One of the surfaces (bottom surface) is mounted on the edge
of the bimorph, and the other (the top one) at the end of a piezoelectric tube. The
assembly is enclosed in a stainless-steel cell of ca. a 10-mL volume, and mounted on
a translation stage that is isolated from electrical and sound noise. During a typical
force measurement, the surfaces are driven closer until contact, and then they are
separated further apart. This is done by applying a triangular voltage wave to the piezo
crystal. Simultaneously, the charge produced upon any deflection of the bimorph,
due to repulsive or attractive forces, is recorded. Once the surfaces come into hardwall contact, the linear movement of the piezo deflects the bimorph, thus enabling
the force sensor to be calibrated against the known piezo-crystal expansion and
contraction as measured by a linear variable differential transformer (LVDT) sensor.
Provided the deflection and the spring constant of the bimorph are known, the data are
used to calculate the force-distance curves from Hookes law. The noninterferometric
surface force apparatus does not allow an absolute determination of the zero surface
separation; however, one can obtain the adsorbed surfactant-layer thickness from the
magnitude of inward jumps that typically occurs when a surfactant layer is pushed
out from the contact area upon compression.
Ederth et al. (1998) show that flame-polished glass surfaces are smooth enough
to enable accurate measurements of surface forces down to molecular separations. The
surfaces used in each experiment were prepared by melting one end of a borosilicate
glass rod (diameter 2 mm, length ca. 25 mm) in a butane-oxygen burner until the tip
formed into a sphere with a diameter of about 4 mm. The normal radii of curvature (rI
and r2)for each surface were determined more accurately at the end of the experiment
by using a micrometer, and the local harmonic mean radius of the interaction, R,
was then calculated as R = 2rlr2/(r,+ r2). The spring constant of the bimorph was
measured at the end of each experiment by placing known weights on the bimorph
spring and measuring the resulting deflection (usually about 100 N/m). The force was
then normalized by the local harmonic mean radius of interaction (flfl.

454 0 O.J. Rojas et al.

The hydrophobic surfaces on which the adsorption of the surfactants was studied
were obtained by silanization and thiolization, respectively, of the hydrophilic glass
surfaces. Despite being hydrophobic, the silanated glass carries a significant net
negative charge. This charge results from the dissociation of unreacted silanol groups
in the glass substrate that is not completely screened by the self-assembled silane
layer. The thiolated surfaces, however, are completely uncharged. One can find details
about the surface preparation in previous reports (Ederth et al., 1998; Stubenrauch et
al., 2004a).
All procedures for assembling the measuring chamber and preparing the
solutions were carried out in a laminar-flow cabinet. At the beginning of each set of
experiments, the interaction profiles were first determined in air to ensure that the
system showed no signs of contamination. Then a background electrolyte solution
was introduced into the measuring chamber, and the interaction profiles were again
determined. A stock surfictant solution was then introduced through a 0.2-pm poly
(tetrafluoroethylene) (PTFE) filter until the desired concentration inside the chamber
was attained. All measurements were carried out at 22 f 1C.

Thin-Film Pressure Balance (TFPB)

The most prominent method for investigating the interactions between two surfactant
films at the aidwater surface (i.e., the interactions acting in foam films) is the thinfilm pressure balance (TFPB) (Claesson et al., 1996; Exerowa & Kruglyakov, 1998;
Exerowa & Scheludko, 1971; Mysels &Jones, 1966; Stubenrauch et al., 2003) and
its modified versions. In brief, a film is formed in a film holder consisting of a porous
glass disc that is connected to a glass tube. A hole is drilled in the disc in which the
film is formed. This film holder is fixed in a gas-tight cell, a pressure is applied to the
cell via a syringe, and the film thickness h at this particular pressure is determined
interferometrically. By calculating the disjoining pressure from the applied pressure,
one obtains the characteristic n-h curves.

Surface Light Scattering (SLS)

The surface light scattering (SLS) set up referred to in this study was reported by
Rojas et al. (2005b), which is based on the original work of HHrd and Neuman
(1 987). The respective surfactant solution was placed inside a closed 3 16-stainlesssteel double-walled thermostated cabinet sitting on an optical table. The temperature
was monitored by temperature probes located inside the chamber, and maintained
by an external thermostated circulation bath (to *O.l"C). The humidity was set close
to saturation [ca. 70% of relative humidity (RH)] by placing filter papers moistened
with water.
In a typical experiment, the experimental autocorrelation function of the
surface waves was recorded and fitted to an exponentially dampened cosine function
(Bouchiat & Meunier, 1971). The correlograms were Fourier-transformed and then
fitted to a four-parameter Lorentzian function (Bellman & Pennington, 1954; Hdrd

Interfacial Properties of Sugar-based Surfactants 0 455

& Neuman, 1987). From these operations, we obtained the parameters of the power
spectrum. The measured power spectrum was compared with the theoretical power
spectrum (Kramer, 1971) for capillary waves in the presence of air. This theoretical
power spectrum is described, among others, in terms of the surface tension y, the
transverse viscosity p, the sum of interfacial shear elasticity and interfacial dilatational
elasticity E, and the sum of interfacial shear viscosity and interfacial dilatational
viscosity K.
The experimental data [i.e., the central frequency and the dampening coefficient
(after correction for instrumental broadening) together with the bulk properties
of the fluids] were used to calculate the viscoelasticity coefficients from the powerspectrum equation. Hence, the sum of the real and imaginary parts of the complex
modulus (elasticity and viscosity) was obtained. We used polar diagrams for direct
interpretation of the rheological parameters E and K.These plots were constructed from
the dispersion equation for a given temperature, wave number, and surface tension
(or surface pressure) by using as parameters the normalized complex frequency (00/
ad slaw), where o0is the experimental central frequency, and a is the dampening
coefficient. Here the subscript w is used to denote the experimental values for a
film-free surface ( E = 0, K = 0) (i.e., water in our case).

Fundamental Studies on Interfacial Properties of

Sugar-based Surfactants
A key aspect of our studies lies in a systematic comparison between petroleumbased surfactants and sugar-based surfactants. This is intended to highlight the
benefits of sugar-based surfactant platforms. We compared an EO-based surfactant
(hexaoxyethylene dodecyl ether, C,,EG) with a sugar-based surfactant with the same
hydrophobic group (n-dodecyl-P-D-maltoside, P-C,,G,) (see Fig. 16.1) (Claesson et
al., 2006; Rojas et al., 2005a).
We consider here the interactions between nonpolar surfaces coated with either
C,,E, or P-C,,G,. As nonpolar surfaces, we chose the aidwater surface, silanated

n-dodecyl-p-D-maltoside, p-C,,G,

Hexaoxyethylenedodecyl ether, C,E,

HO Fig. 16.1 .The n-dodecyl-(3-Pmaltoside(P-C,,G,, top) and hexaoxyethylenedodecyl ether (C,E,

bottom)
are two nonionic surfactants with the same hydrophobic group that were compared regarding their
interfacial properties.

456 0 O.J. Rojas et al.

glass, and thiolated gold surfaces. The most important results with respect to the
comparison of different surfaces are summarized in this section.

AirMater Interfaces
The adsorption of nonionic surfactants at the airlwater surface leads to a decrease
of the surface charge of the interface (reviewed in Stubenrauch et al., 2003). This
decrease eventually results in a transition from an electrostatically stabilized common
black film (CBF) to a Newton black film (NBF) that is stabilized by short-range
repulsive forces. This phenomenon is illustrated in Fig. 16.2 with data obtained
for the nonionic surfactant hexaoxyethylene dodecyl ether (C,,E6). The formed
NBF consists of two densely packed monolayers, and creates a force barrier that
prevents the film from rupturing, thus stabilizing a foam. Apart from densely-packed
monolayers, a sufficient monolayer cohesion is required for the formation of a stable
NBF (Stubenrauch et al., 2004a,b). From the thickness of the NBF, one can estimate
the thickness of one monolayer.
Comparing Fig. 16.2 with the respective results obtained for n-dodecyl-P-Dmaltoside (P-C,,G,) shown in Fig. 16.3, apparently, the same general trend is observed.
In both cases, film thicknesses ranged from 930 nm to c5 nm, depending on the
surfactant concentration and the applied pressure, which ranges from 200-9000 Pa.
As was the case for C,,E6 (Fig. 16.2), two different kinds of films were observed: thick
CBFs stabilized by electrostatic repulsion, and thin NBFs stabilized by short-range
repulsion. The thicknesses of the CBFs decrease monotonically as increases. While
the slope d(1og )ldh is independent of the surfactant concentration, a significant
shik of the curves toward lower disjoining pressures is observed when increasing
the P-C,,G, concentration from 0.034-0.137 mM. Moreover, at the highest
concentration, no CBF is formed at all, but the foam film drains directly down to the
NBF.
Experimentally (reviewed in Stubenrauch et al., 2003) and only recently also
theoretically (Kudin & Car, 2008), the airlwater surface is negatively charged. This
charge is responsible for the long-range electrostatic repulsive forces observed in
foam films stabilized by nonionic surfactants. An increase in the nonionic surfactant
concentration leads to a decrease of the surface-charge density as more uncharged
molecules (i.e., nonionic C,,E6 and P-C,,G, surfactants) adsorb at an originally
charged surface. The electrostatic forces acting in foam films stabilized by P-C,,G,
were quantified by means of the DLVO theory by using constant-charge boundary
conditions and the theoretical Debye length of K-= 30 nm. These calculations led
to surface-charge densities of qo = 1.55 mC rn- for the 0.035 mM solution and q,, =
0.95 mC rn-, for the 0.137 mM solution, respectively. The decrease in surface-charge
density destabilizes the CBF until no CBF is observed for c > CMC under the given
experimental conditions. At these concentrations, the immediate formation of an
NBF is observed. The NBFs are very thin (ca. 5 nm) with an aqueous core of 1-2 nm
assuming a length of 22 nm for the surfactant. In other words, these films consist of two

InterfacialProperties of Sugar-based Surfactants 0 457

Common Black Film, CBF

Newton Black Film, NBF

h<10nm

1 0 n m < h < l00nm

100

hlnm
Fig. 16.2. (top) Pictures and schematic drawings of a common black film and a Newton black film,
respectively.Thepictureswere taken with an optical microscopeandthecircles represent the film formed
in a small hole with a diameter of 1-2 mm drilled through the disk in the TFPB. (bottom) Disjoining
pressure n as a function of film thickness h measured for three concentrationsof C,,E, in lo4 M of a NaCl
solution.The two lower concentrationsare below; the highest is above the critical micelle concentration
(CMC) (= 0.073 mM).The solid lines are calculated according to the Derjaguin-Landau-Verwey-Overbeek
(DLVO)theory assuming interactionsat constant charge (Stubenrauch et al., 2004a).

458 0 O.J.

Rojas et al.

surfactant monolayers with only small amounts of water separating the head groups
(mainly hydration water). As is the case for the force barrier between nonpolar solid
surfaces (to be presented later), densely-packed monolayers are required to stabilize an
NBF and thus to prevent contact between the two bare surfaces (i-e., film rupture).
Hence, the presence of an NBF signifies the existence of a force barrier between two
aidwater surfaces analogous to the force barrier between two solid surfaces (see Figs.
16.4 and 16.5).
The remarkable similarities with regard to the n - h curves of &G,
and C,,E,
are evident from the fact that the curves nearly lie on top of each other, which means
that the surface- charge densities qo are nearly equal. Indeed, surface-charge densities
of qo = 1.70 mC m-2 at c = 0.01 mM (0.12 CMC) for C,,E, and qo = 1.55 mC m-2
at c = 0.035 mM (0.25 CMC) for P-C,,G, were calculated from the experimental

10000

100
0

10 20 30 40 50 60 70 80 90 100

hlnm
Fig. 16.3. Disjoining pressure ll as a function of film thickness h measured for three concentrations of
P-C,,G, in lo4 M of a NaCl solution. The two lower concentrations are below; the highest is above the
critical rnicelte concentration (CMC) (= 0.14 mM). The solid lines are calculated according to the DLVO
theory assuming interactionsat constant charge(Stubenrauchet al., 2002) (Reproduced with permission
of the PCCP Owner Societies).

Interfacial Properties of Sugar-based Surfactants 0 459

10

20

30

40

50

D/nm
Fig. 16.4. Force F normalized by the radius R as a function of relative surface separation 0 between two
thiolated solid surfaces. The forces were measured across aqueous solutions containing 0.1 mM NaCl
in the absence (filled circle) and in the presence of 0.01 mM non-ionic surfactant C,,E, (open squares)
(Adapted with permission from Stubenrauch et al., 2004 and Rojas et at., 2005a. Copyright 2004 and
2005, American Chemical Society). Similar behavior was observed for p-Cl2G2 (see Fig. 16.5).

data. As the surface-charge density is a property of the bare aidwater surface, these
values mean that a surface concentration of 3.0 x 10" mol rn-, C,,E, (0.12 cmc)
reduces the surface-charge density to the same amount as a surface concentration
of 4.0 x lo4 mol m-2 P-C,,G, (0.25 cmc). Moreover, in both cases, the surface
concentration is enough to stabilize a foam film up to 10,000 Pa. We suggest that the
effective surface cross-section area covered per molecule determines the reduction in
surface-charge density. Since the head group of C,,E, is larger than that of P-C,,G,,
a smaller number of C,,E, molecules would be needed to achieve a given reduction
in the repulsive double-layer force. We conclude that, for each nonionic surfactant,
the surface- charge density is reduced with increasing adsorption. A more thorough
discussion regarding the influence of the head groups and the chain length of the

460 O.J. Rojas et al.

surfactant on film stability is found in Stubenrauch et al. (2002) and Schlarmann &
Stubenrauch (2003).

Solid/Liquid Interfaces
Silanated glass and thiolated gold surfaces were used as hydrophobic solid surfaces.
Upon adsorption, at the lowest surfactant concentration, we noted that the interaction
forces were dominated by a steric barrier at a short-surface separation. By increasing
the surfactant concentration, a more robust steric barrier was formed. Interestingly,
the NBF formation observed for foam films (see Figs. 16.2 and 16.3) corresponded
to the appearance of this force barrier between two solid surfaces (see Fig. 16.4 in the
case of thiolated gold).
The barrier shown in Fig. 16.4 is a measure of the force that is needed to remove
surfactant from between the two surfaces. Because this barrier is located at a distance
corresponding to the thickness of a surfactant bilayer, the analogy to the NBF
formation is obvious. Note that a removal of this bilayer results in film rupturing in
the case of foam films, whereas in the case of solid/liquid interfaces it leads to a direct
contact of the solid surfaces (Rojas et al., 2005a; Stubenrauch et al., 2004a).
Also concluded (data not shown) is that the nature of the surface at which the
surfactant adsorption takes place mainly influences the interaction forces at a lowsurface coverage. Once a densely-packed surfactant layer is formed, the surfactant
itself determines the interaction forces.

Comparison between 0-and Sugar-based Surfactants:Structural

Aspects, Interfacial Aspects, Packing at the Interface
An important conclusion drawn in the summary of results presented before is the fact
that the nature of the surface at which the surfactant adsorption takes place mainly
influences the interaction forces at a low-surface coverage. Once a densely-packed
surfactant layer is formed, the surfactant itself determines the interaction forces.
Therefore, we were able to detect structural effects (i.e., the influence of the polar
head group) on the interaction forces at the molecular level.
Comparing data obtained for P-C,,G2 and C12EB,
we demonstrated that for similar
bulk concentrations the sugar surfactant forms a denser and more robust monolayer
with higher cohesiveness (see Fig. 16.5). O n the other hand, compared with the
shorter tail homolog P-C,,G,, P-C,,G, is anchored more strongly to hydrophobic
surfaces. This is explained by a stronger interaction between the nonpolar tails within
the monolayer.
Our measurements indicated that the thickness of the j3-C12G, monolayer
under compressive loads was marginally lower than 2 nm. This thickness value is in
agreement with the surfactant molecular dimensions but smaller than the (compressed)
monolayer thickness measured for C,,E,. For both surfactants, the most distinctive
feature in the force curve between surfactant-coated solid surfaces is the buildup of a

Interfacial Properties of Sugar-basedSurfactants 0 461

P'C12G2

rt

1 2E6

D/nm

10

15

20

Fig. 16.5. Steric force barriers for 0.035 mM of P-C,,G,and 0.01 mM of C,,E, respectively, between solid
thiolated surfaces. A stronger force barrier is observed for the sugar-based surfactant. This steric force
is responsible for the stabilization of dispersions, where the solid particles are separated by liquid films.
The stronger barrier seen in the case of the sugar surfactant is explained by a larger stiffness of its polar
group as compared to the EO units (Adapted with permission from Rojas, et al., 2005a. Copyright 2005
American Chemical Society).

steep and short-range repulsion as the surfactant adsorption approaches saturation.

Shrestha at al. measured the surface rheology of C,,E, and P-C,,G, under similar
conditions, and found that P-C,,G, has a higher surface elasticity, which, in turn,
explains the observation that P-C,,G, forms much more stable foams than C,,E6
(submitted for publication). This indicates that the head group of the sugar-based
surfactants is stiffer and less compressible, which, in turn, seems to be favorable for
the stabilization of foams (Shrestha, 2008, submitted for publication) and dispersions
(Lu et al., 2007; Lu & Somasundaran, 2008).
In summary, we have so far exclusively considered short-range repulsive
interactions between surfactant layers. They are divided roughly into repulsive forces
due to hydration and steric forces (Claesson et al., 2006). Of course, the net shortrange interaction-that
is the only measurable quantity-is due to a balance of
attractive and repulsive-force contributions. In the next sections, we recapitulate some
findings about the hydration of polar groups, and relate those findings to the shortrange interactions between surfactant layers.

462 0 O.J. Rojas et al.

ViscoelasticityProperties of Isomeric Sugar-based Surfactants

A wealth of information about the surface tension of sugar-based surfactant is available in the literature, some ofwhich are cited in this chapter. As expected, differences
in surface activity, packing density, minimal surface tension, and CMC were observed
when comparing sugar surfactants of different types or different structures. Less obvious is perhaps the fact that subtle changes in the stereochemistry of a given surfactant
can lead to noticeable differences in the same properties mentioned above. Instead of
discussing tensiometry data for sugar surfactants here, we only point out the case of
an interrelated property, namely, the viscoelasticity of adsorbed monolayers.
Surface light scattering by capillary waves was used to investigate the adsorption
behavior of nonionic sugar surfactants at the aidwater surface. The viscoelastic
properties (surface elasticity and surface viscosity) of monolayers formed by the
isomeric surfactants octyl P-glucoside, octyl a-glucoside, and 2-ethylhexyl a-glucoside
(see Fig. 16.6) were quantified at submicellar concentrations (Rojas et al., 2005b).
The same surfactant types were used in an earlier study by Waltermo who presented
surface-tension data and its relationship with the foaming behaviors (Waltermo et al.,
1996). Unfortunately, SLS data for C,,E6 and P-C,,G, that were discussed in previous
sections are not available; however, we used the surfactants illustrated in Fig. 16.6 to
document the conclusion that differences in the configuration of the molecule affect
the surface packing and thus the rheological behavior of the formed monolayers.
Likewise, macroscopic properties such as foam ability, emulsion stability, and others
are also influenced.
The surfactants used (Fig. 16.6) were anomerically pure alkyl glucosides. The
focus of this study was to uncover the difference between the p and a anomers of
octyl glucoside (Figs. I6.6a and I6.6b, respectively). In addition, the effect of the
structure of the surfactants hydrophobic part was addressed by comparing a linear
(Fig. 16.6b) and a branched octyl chain (Fig. 16.6~).
Readers interested in the processing of the SLS data to calculate the surface
rheological properties are referred to (Rojas et al., 2005b). In short, we used a socalled helmet or polar plots in which radial and semicircular lines represent constant
interfacial elasticitiy ( E ) and constant interfacial viscositiy (K), or, in other words,
isoelasticity and isoviscosity curves, respectively (see Fig. 16.7). The distance of an
experimental data point from the origin ( E = 0, K = 0) in a polar plot is a measure of
the surface elasticity and/or viscosity. Therefore, the surface elasticity and viscosity can
simply be estimated by the inspection of the location of the experimental data within
the polar plot.
Figure 16.7 shows the experimental polar loci or the viscoelastic behavior of the
three surfactants described earlier (at a SLS wave number of 78508 m-I). The polar
plots show first the limiting case of a surfactant-free liquid dynamics (no monolayer
present) [i.e., pure water ( E = 0, K = O)]. As the surfactant concentration increases,
the respective viscoelastic behavior can be identified by the direct inspection of the
curve, starting from the value corresponding to zero surfactant concentration ( E = 0,

Interfacial Properties of Sugar-based Surfactants 0 463

Fig. 16.6. Isomeric sugar surfactants used to investigatethe effects of the molecularstructure on surface
viscoelasticty:n-octyl-0-glucoside (a), n-octyl-a-glucoside(b), and 2-ethylhexyl-a-glucoside (c).

0). At low-surfactant concentrations, the tendencies indicate a deviation from

the perfectly elastic film behavior (K = 0, contour line). In these cases, a maximal
propagation velocity and dampening limit is attained.
More importantly, in terms of the present discussion, is the fact that the polar plots
indicate small, yet distinctive, differences in the viscoelasticity of the three isomeric
(sugar) surfactants. Therefore, in agreement with other studies (Neimert-Anderson
et al., 2004; Niraula et al., 2004 ), the stereochemistry of the surfactant plays an
important role in molecular packing, lateral interactions, and overall intermolecular
forces, all of which affect the respective macroscopic surface behaviors.
We note that the area per molecule for octyl a-glucoside and octyl P-glucoside
surfactant was 41.5 and 43.5 A, respectively (Rojas et al., 2005b). Furthermore,
the data presented in Fig. 16.7 demonstrate further important changes in molecular
packing at the aidwater surface for these isomeric surfactants. This is by virtue of very
subtle configuration differences of the head group as well as the tail structure. One can
therefore anticipate that by fine-tuning the molecular architecture of sugar surfactants,
one can develop applications involving different performance requirements. For
K =

464 0 O.J. Rojas et al.

Normalized Central Frequency ~ui$

Normalized Central Frequency 1%

Normalized Central Frequency w ~

Fig. 16.7. Helmet or polar plots showing the experimental surface viscosity and elasticity obtained from
the propagation (centralfrequency and damping)ofcapillarywavesprobed by surface light scatteringfor
n-octyl-&glucoside (left),n-octyl-a-glucoside (center), and 2-ethylhexyl-a-glucoside (right)surfactants.
The loci of the different curves show differences in the behavior of these isomeric surfactants (Adapted
with permission from Rojas et al., 2005b. Copyright 2005 American Chemical Society).

example, one can vary foamability and emulsion stabilization by changing the polar
group (position and stereochemistry), as well as the structure of the hydrophobic tail
(e.g., via branching).
Changes in the stereochemistry of the head group units are also expected to
influence the interaction with the ions present in solution. Such effects were made
clearer in the discussion presented in Rir/Water Interfaces on thin-film pressurebalance experiments for other surfactants. One should stress, however, that dynamics
effects (diffusion, relaxation, etcetera) play important roles as well. At this point, to
speculate further about the relationship between surfactant structure and properties
is meaningless until the nature of surface-charge density, surfactant packing, and
dynamic effects is better understood. An attempt to shed some light regarding the
effects of surfactant structure is made in the next sections.

Molecular Prospects
To obtain experimental information on the structure of water adjacent to solutes is
very difficult. The main reason for this is that in most cases the spectroscopic signal
from the hydration shell is swamped by that of bulk water. However, recently some
progress was made where an inherently surface-sensitive spectroscopic technique,
vibrational sum frequency spectroscopy (VSFS), was employed to investigate the
hydration of sugar groups and oligo(ethy1ene oxide) chains anchored to the aidwater
surface. Some very strong hydrogen bonds are formed between sugar and water,
whereas the hydration of the oligo(ethy1ene oxide) chain is dominated by a clathratelike structure (Claesson et al., 2006; Tyrode et al., 2005).
One can apply related arguments when considering short-range interactions
between surfactant layers. The fact that hydrogen bonds can form between polar

Interfacial Properties of Sugar-based Surfactants 0 465

head groups of, for example, sugar-based surfactants does not directly lead to the
conclusion that the short-range interaction across water should be attractive due to a
hydrogen-bond formation. To compile available data on the depth of the interactionforce minimum is thus of some interest. In this case, the pull-off force needed to
separate surfaces that are in contact (for example, after bringing the surfaces together,
as shown in Fig. 16.4 and Fig. 16.5) is measured. This force is related to the adhesion
energy that, in our case, was measured between adsorbed nonionic surfactant layers.
This will allow us to learn if the possibility of forming hydrogen bonds between two
head groups has a bearing on the short-range attraction observed in these systems (see
Table 16.1).
We note that one cannot form direct hydrogen bonds between dodecyldimethylamine oxide (DDAO) molecules in uncharged form, decyldimethyl phosphine oxide (DDPO), or surfactants with EO head groups (see Table 16.1). However, such
interactions are possible for surfactants with polyhydroxyl head groups and also for
amine head groups. By comparing the data compiled above, we note that the trend is
that head groups that are able to form interlayer hydrogen bonds display a somewhat
larger attractive short-range force component than layers that are unable to form such
bonds. This could indicate that a contribution from interlayer hydrogen-bond formation exists, but the extension of this effect is difficult to determine since the nature of
the head group also influences other short-range interactions, for example, van der
Waals forces.
Due to the lack of systematic data, we cannot compare quantitatively the
hydration numbers obtained for EO and glucose head groups. A qualitative evaluation,
however, might shed some light on the differences between these surfactants. From
extensive Small Angle Neutron Scattering ( S A N S ) measurements in bicontinuous
microemulsions, we know that the area per head group at the water/oil interface is
0.56 nm2 for a glucose unit (Kluge 2000), which is comparable to that of a surfactant
with four EO units which is 0.54 nm2 (Sottmann et al., 1777). 'The hydration number
of the glucose head group, as determined by nuclear magnetic resonance (NMR), is
6 at a surfactant concentration of 3 wt%, whereas NMR estimates the hydration
number to be 6 per an EO unit at a surfactant concentration of 10 wt% (i.e., 24 water
molecules per tetraoxyethylene head group). Keeping in mind that the hydration
number per EO unit is expected to be even larger at 3 wt%, one can conclude that
under similar conditions and for similar head group sizes the hydration of EO-based
surfactants is significantly higher than that of sugar-based surfactants (Claesson et al.,
2006).
The short-range interaction between surfactant layers exposing the polar part
toward solution is due to a complex interplay between van der Waals forces, hydration,
and steric effects. One can only measure the total force, and this force displays both
attractive and repulsive regimes. The van der Waals force is the main attractive force,
while hydrogen bonding within adsorbed layers, again directly or mediated by water
molecules, increases the packing density within adsorbed surfactant layers; this
increases the van der Waals force.

466 0 O.J. Rojas et al.

Table 16.1. Depth of Force Minimum between Various Nonlonic Surfactant Layers

Compound

Technique

Depth of force
minimum
F/R (Mn/M)

n-dodecyl amine oxide (DDAO)

Unchargedform.

SFA

0.25

Bilayer on mica.

n-dodecyl phosphine oxide

(DDPO)

SFA

0.25

Hydrophobied mica. Force

minimum increases with
increasingtemperature.

tetraoxyethylenedodecyl
ether (C,E,)

MASIF

Silanatedglass surfaces.

pentaoxyethylenedodecyl
ether (C,E,)

SFA

0.1 -0.2

Hydrophobized mica.
A t 20C. Force minimum
increases with increasing
temperature.

hexaoxyethylenedodecyl
ether (C,E,)

MASIF

1-1.4

Silanatedglass surfaces.

hexaoxyethylenedodecyl
ether (C,E,)

MASIF

0.2-0.4

Thiolated gold surfaces.

tetraoxyethylenedodecyl
amide
K,,ONHE,)

SFA

0.3-0.4

Hydrophobizedmica and
silanated glass (Herder,
1991)

tetraoxyethylenedodecyl
amine (C,,NHE,)

SFA

0.1 -0.2

Bilayer on mica, pH 9.4-9.9.

Substrate surface and

remarks

n-octyl-p-glucoside(P-C,G,)

SFA

1.2-1.5

Hydrophobized mica.

n-decyl-j3-o-glucoside(j3-C,,G1)

MASIF

Thiolated gold surfaces.

Estimated. (Waltermo, 1993)

n-decyl-P-o-maltoside(p-C,,G,)

MASIF

0.9

Silanated glass surfaces.

n-dodecyl-j3-o-maltoside
(P-C,,G,)
technical sugar-surfactant
mixture

MASIF

1-2

Thiolated gold surfaces.

SFA

Hydrophobizedmica.
Located on the repulsive
side.

n-dodecyl lactobion amide

(LABA), inner sugar ring is
open

SFA

1.2

Hydrophobized mica

n-dodecyl lactobion amide

(LABA), inner sugar ring i s
open

MASIF

1.5

Silanated glass surfaces.

maltose 6-O-dodecanoate
(C,,OGJ
n-dodecyl amine (C,,NH,)
Most molecules in the outer
layer are uncharged.

MASIF

1.o

Silanatedglass surfaces

SFA

0.5-1

Bilayer on mica, pH 9.5-10.1

Data correspond to aqueous surfactant solutions a t or above the respectiveCMC (i.e., to a

saturated monolayer).SFA = surface-forceapparatus, MASIF = measurementand analysis of
surface-interaction forces.
Reproduced with permission of the PCCP Owner Societies.
Claesson et al., 2006 and references therein.

Interfacial Properties of Sugar-based Surfactants 0 467

We note that both hydration and the mobility of surfactants are important for
short-range interactions. The steric interaction between surfactants with oligomeric
head groups also deserves further attention. As the head group size increases, the
short-range repulsion goes from a hard-wall repulsion, via compressible excluded
volume repulsion, to a steric repulsion described by polymer-brush theories (de
Gennes, 1987).
The experimental results discussed in this chapter help to demonstrate hndamental
features of nonionic surfactants based on sugar polar groups (mainly glucoside and
maltoside). These polar groups make them structurally more attractive in terms of
their capacity to stabilize interfaces (liquid-liquid, solid-liquid, and gas-liquid) as
compared to EO-based nonionic surfactants. Note that one can use the OH groups in
the sugar units for chemical reactions, and thus one has the possibility of engineering
the surfactant structure and its stereochemistry (which, as was demonstrated before,
play a leading role for the properties). Key molecular components (head group
configuration, identity, number, tail length, and saturation) lead to subsequent
packing aspects for green polysaccharide streams that one can ultimately use to
manufacture nonionic surfactants.
As mentioned in previous sections, nonionic, saccharide-based surfactants are
a veritable plethora of tangible benefits for state-of-the-art surface, colloid, and
condensed state research. Their solubility and phase behavior are typically insensitive
to temperature changes in contrast to EO-based surfactants. This allows much greater
amplitude in their synthetic design without compromising desirable physical and
chemical features such as small molecule (e.g., blood glucose) biosensor applications,
biophysical recognition (such as antigen-antibody mechanisms), host-guest chemistry,
and a number of other smart functions. One needs to use separations chemistries, and
chemically fine-tune the respective sugar surfactants with the appropriate physical
characteristics (functional groups and/or ligands) to achieve the smart functionality
we have described.
The performance of sugar surfactants (versus EO surfactants) in dispersed systems
(such as foams, emulsions, and dispersions) and their potential for the formation
of new structures in these systems (such as unilamellar vesicles, ribbons, spirals,
tubes, etcetera) depends on the detailed molecular structure. One may synthesize
new surfactants which may offer unique templates for more sophisticated electronic,
biomedical, and optical devices. Some strategies in the design of sugar surfactants to
provide greater functionalities include the introduction of both mono- (glucoside)
and di-saccharide (maltoside) moieties with various modes of attachments to the
head group of synthesized surfactants and also the introduction of different degrees
of hydrophobicity and crystallinity to the sugar surfactants by manipulating their
aliphatic tails.

Structure-Property Relationships
Efforts need to be devoted to the development of new surfactants based on the
structure-property heuristics. In the simplest form, a new sugar surfactant that

468 0 O.J. Rojas et al.

could be synthesized would consist of three discrete parts: (a) polar head group(s),
for example, one or two glucosides or one maltoside, with or without a (b) spacer
(that could display the versatility of allowing the attachment of a second head group
and two tails as well as two different tail placements for one group), and the (c)
aliphatic tail (for example, a fatty acid). How the changes in these components will
affect their performance in model systems (foams, emulsions, dispersions) is then the
key question. Some leads to answering such questions were presented in previous
sections.
What is exciting about all of the possible configurations is the inherent potential
for providing a number of different surface interactions based on the HLB (hydrophilic
and lipophilic balance) parameter (Griffin, 1954), the variation in the packing based
on steric interactions (Piispanen et al., 2004), and the packing variation based on
other interactions (John & Vemula, 2006).
The HLB is useful to characterize the balance between the hydrophilic and
hydrophobic parts of the surfactant molecule. One can vary the HLB by changing
the length of the tail of the surfactant. This can lead to different packing behaviors.
The molecules will have a pronounced ability to engage in discrete regular packing
motifs, as was shown in recent work (John et al., 2003). Typically, surfactant packing
is dominated by the hydrophobic interactions between the hydrophic chains, which,
in turn, influence the flexibility, the orientation, and the surface energy of molecular
assemblies.
A serious need is noted for a systematic study of the structure-property relations
that define sugar-based surfactants as the alternative of choice. Major issues in this
area are:
Development of new products, production practices, and business in the biomass
industry. This requires rapid changes in consumer demand for renewable products
and alternative market solutions.
Replacement (even small fractions) of petroleum-based nonionic surfactants with
sugar-based surfactants, which is important from both an environmental and
energetic point of view.
Synthetic strategies to reduce the costs of sugar-based surfactants since current
barriers for the acceptance of sugar-based surfactants are their costs. Linking
fundamental studies with economics and market analysis may promote this
goal.
Systematic studies of the properties of sugar-based surfactants as a lack of
information about their properties and performance benefits may be an obstacle
for their wide use.

Interfacial Properties of Sugar-based Surfactants 0 469

Conclusions
The use of nonbiodegradable, petroleum-based surfactants is a major drawback, and
therefore the development of biodegradable surfactants derived from agricultural
resources is justified. Sugar-based surfactants represent a proven candidate, but any
effort aimed at formulating dispersed systems by using the saccharide platform entails
that specific technical requirements in terms of their applications must be met. The
performance of (new) surfactants in any given application is expected to be closely
related to molecular factors, with the interfacial behaviors being some of the most
important ones.

Acknowledgments
O.J.R. greatly appreciates the financial support of the National Research Initiative
of the USDA Cooperative State Research, Education and Extension Service, grant
number #USDA 2008-01 5 19. C.S. is grateful for financial support of the European
Commission under the 6th Framework Program, contract No. MRTN-CT-20045 12331-Project SOCON.
References
Balzer, D.; H. Luders (Eds.). Nonionic Sufactants:Alkyl Polygucosides; Surfactant Science; Marcel
Dekker: New York, 2000; Vol. 91.
Bellman, R.; R.H.Pennington. Effects of surface tension and viscosity on Taylor instability. Q.
Appl Math. 1954,12, 151-162.
Blunk, D.; I? Bierganns; N. Bongartz; R. Tessendorf; C. Stubenrauch. New speciality surfactants
with natural structural motifs. Newj. Chem. 2006,30, 1705-1717.
Bouchiat, M.A.; J. Meunier. Spectre des fluctuations thermiques de la surface libre d'un liquide
simp1e.J Phys. 1971,32, 561-571.
Claesson, EM.; T. Ederth; V. Bergeron; M.W. Rutland. Techniques for measuring surface forces.
Adv. Colloid Znteface Sci. 1996, G7, 1 19-1 84.
Claesson, EM.; M. Kjellin; O.J. Rojas; C. Stubenrauch. Short-range interactions between nonionic surfactant layers. Phys. Chem. Chem. Phys. 2006, 8(47), 5501-5514.
de Gennes, EG. Polymers at an interface, a simplified view. Adv. Colloid Znteface Sci. 1987,27,
189-209.
Deleu, M.; M. Michel Paquot. From renewable vegetables resources to microorganisms: New
trends in surfactants. Compt. Rendus Chem. 2004, 7(&7),641446.
Ederth, T.; I?M. Claesson; B. Lindberg. Self-assembled monolayers of alkanethiolates on thin gold
films as substrates for surface force measurements. long-range hydrophobic interactions and
electrostatic double-layer interactions. Langmuir 1998, 14, 4782-4789.
European Public Health Alliance, http://www.epha.org/a/926. Accessed August 22, 2005
Exerowa, D.;Kruglyakov. Foam and Foam Films- aeory, Experiment, Application [Studiesin
ZntefacialScience; Vol. 5, D. Mobius, R. Miller (Series Eds.)]; Elsevier: Amsterdam, 1998.

470 0 O.J. Rojas et at.

Exerowa, D.; A. Scheludko. Porous plate method for studying microscopic foam and emulsion
films. CR Acad Bulg. Sci. 1971,24,47-50.
Griffin, W.C. Calculation of HLB values of non-ionic surfactants./. Soc. Cosr.net. Chem. 1954,5,
249-258.
Hard, S.; R.D.J. Neuman. Accurate laser light-scattering measurements of capillary waves. /. Colloid Interface Sci. 1987, 120(1), 15-29.
Hill, K.; W. von Rybinski; G. Stoll (Eds.). Alkyl Polyglycosisides. Technolou, Properties andApplications; VCH Verlagsgesellschaft: Weinheim, 1997; 242 pp.
Holmberg, K. Natural surfactants. Cum Opin. ColloidInterj2te Sci. 2001, G, 148-159.
ILO (International Labour Office). Encyclopaedia ofOccupational Health and S a j g ; 4th ed.; Geneva, Switzerland: International Labour Office, 1998; Vols. 1-4, p. 104.
John, G.; H. Minarnikawa; M. Masuda; T. Shimizu.Liquid crystalline cardanyl j3-Dglucopyranosides. Liq. Cyst. 2003,30,747-749.
John, G.; I?K. Vemula. Design and development of softnanomaterials from biobased arnphiphiles.
So# Matter 2006,2,909-914.
Kiraly, Z.; G.H. Findenegg. Calorimetric study of the adsorption of short-chain nonionic
surfactants on silica glass and graphite: Dimethyldecylarnineoxide and octyl monoglucoside.
Langmuir 2000, IG,8842- 8849.
Kluge, K. Der Schlussel zum Verstandis von Mikroemulsionen aus Zuckertensiden: Die Interne
Grenzflache, Ph.D. thesis, Universidt zu Koln, 2000.
Kocherbitov, V.;0.Soderman; L.J. Wadso. Phase diagram and thermodynamics of the n-octyl
beta-D-glucoside/watersystem. Phys. Chem. B 2002, 106 2910- 2917.
Kramer, L. Theory of light scattering from fluctuations of membranes and monolayers. /. Chem.
Phys. 1971,55,2097-2105.
Kudin, K.N.; R. Car. Why are water-hydrophobic interfaces charged?/. Am. Chem. Soc. 2008,
130(12), 39 15-391 9.
Kumpulainen, A.J.; C.M. Persson; J.C. Eriksson. Head group and hydrocarbon tail effects on the
surface tension of sugar-based surfactant solutions. Langmuir 2004a, 20,10935-10942.
Kumpulainen, A.J.; C.M. Persson; J.C. Eriksson. n-Decyl-glucopyranoside and n-decyl-maltopyranoside gibbs monolayers. Phase changes in the dilute liquid-expanded range. Langmuir 2004b,
20, 10534-10541.
Kumpulainen, A.J.; C.M. Persson; J.C. Eriksson; E.C. Tyrode; C.M. Johnson. Soluble monolayers
of n-decyl glucopyranoside and n-decyl maltopyranoside. Phase changes in the gaseous to the
liquid-expanded range. Langmuir 2005,21, 305-31 5 .
Lewis, R.J. Sr. (Ed.). Hawleys Condensed Chemical Dictionary; 13th ed.; John Wiley and Sons,
Inc.: New York, NY, 1997; 805pp.
Liljekvist, I?; M. Kjellin; J.C. Eriksson. The surface pressure effect of pentaoxyethylene and maltoside surfactant head groups. Adv. ColloidInterfaceSci. 2001, 89, 293-302.
Lu, S.; I? Somasundaran. Tunable synergismlantagonismin a mixed nonionic/anionic surfactant
layer at the solidlliquid interface. Langmuir 2008,24(8), 3874-3879.
Lu, S.; Y. Bian; L. Zhang; I? Somasundaran. pH Dependence of adsorption of n-dodecyl-beta-D-

Interfacial Properties of Sugar-based Surfactants 0 471

maltoside on solids.J. Colloid Integace Sci. 2007,31G(2), 3 10-3 16.

Matsson, M.K.; B. Kronberg; PM. Claesson. Adsorption of akyl polyglucosides on the solidlwater interface: equilibrium effects of alkyl chain length and head group polymerization. Langmuir
2004,20, 4051- 4058.
Muruganathan, R.M.; R. Krustev; N. Ikeda; H.-J. Miiller. Temperature dependence of the gas
permeability of foam films stabilized by dodecyl maltoside. Langmuir 2003, 19,3062- 3065.
Muruganathan, R.M.; R. Krustev; H.J. Miiller; H. Mohwald; 9. Kolaric; R. von Klining. Foam
films stabilized by dodecyl maltoside. 1. Film thickness and free energy of film formation. Langmuir 2004,20,6352-6358.
Mysels, K.J.; M.N. Jones. Stability and some properties of the second black film. Discws. Farahy
SOC.
1966, 42, 42-50.
Neimert-Anderson, K.; E. Blomberg; P Somfai. Stereoselectivesynthesis of novel polyhydroxyl
surfactants. Stereochemical influence on Langmuir monolayers.J. Org. Chem. 2004, 69,
37463752.
Nickel, D.; T.Forster; W. von Rybinski. Physicochemicd properties of alkyl polyglycosides. Alkyl
Pobglycoside, TechnologJ,Propertie$ andApplications; K. Hill, W. von Rybinski, G. Stoll, Us.;
VCH: Weinheim, Germany, 1976,3949.
NIOSH. National Occupational Exposure Survey (NOES); 1983, http:llwww.cdc.govlnoes/
Niraula, B.B.; T.K. Chun; H. Othman; M. Misran. Dynamic-interfacialproperties of dodecylbeta-D-mdtoside and dodecyl-beta-D-fructofuranosyl-apha-D-glucopyranoside
at dodecanel
water interface. Colloids Surf: A. Physicchem. Asp. 2004,248, 157-166.
Parker, J.L. Surface force measurements in surfactant systems. Prog. Surf: Sci. 1994,47,205-271,
Parker, J.L.; H.K. Christenson; B.W. Ninham. Device for measuring the force and separation
between two surfaces down to molecular separations. Rev. Sci. Instrum. 1989, GO, 3135-3138.
Persson, C.M.; PM. Claesson; K. Lunkenheimer. Interfacial behavior of n-decyl-beta-D-maltopyranoside on hydrophobic interfaces and the effect of small amounts of surface-active impurities.
J. ColloidInteface Sci. 2002, 251, 182-192.
Persson, C.M.; U.R.M. Kjellin; J.C. Eriksson. Surface pressure effect of poly(ethy1ene oxide) and
sugar head groups in liquid-expanded monolayers. Langmuir 2003a, 19,81 52-8160.
Persson, C.M.; A.J. Kumpulainen; J.C. Eriksson. Adsorption of n-decyl-P-D-glucopyranoside
and n-decyl-P-D-maltopyranosidemixtures at the liquid-vapor interface. Langmuir 2003b, 19,
61 10-61 14.
Persson, C.M.; A.J. Kumpulainen. Surface forces measured in sugar surfactant solutions between
two hydrophobic thiolated silica spheres. Colloids Surf: A. Physicchem. Asp. 2004,233,4349,
Piispanen, ES.; M. Persson; E Claesson; T. Norin. Surface properties of surfactants derived from
natural products. Part 1 : Syntheses and structurelproperty relationships-solubility and emulsificati0n.J Surf: Deterg. 2004, 7, 147-159.
Ren, L.; M.A. Marquardt; J. Lech. Estrogenic effects of nonylphenol on pS2, ER and MUCl gene
expression in human breast cancer cells-MCF-7.J. Chem. Biol Interact. 1997, 104(1), 55-64.
Rojas, O.J.; C. Stubenrauch; J. Schlarmann; EM. Claesson. Interactions between non-polar
surfaces coated with the non-ionic surfactant n-dodecyl-P-D-maltoside. Langmuir 2005a, 21,

472 0 O.J.

Rojas et al.

11836-1 1843.

Rojas, O.J.; R.D. Neuman; PM. Claesson. Viscoelastic properties of isomeric alkylglucoside surfactants studied by surface light scattering. J. Phys. Cbem. B. 2005b, 107, 22440-22448.
Ruiz, C.C. (Ed.). Sugar-Bared Surfactants: Fundamentah andApplications; Taylor & Francis: Boca
Raton, 2009.
Schlarmann, J.; C. Stubenrauch. Stabilization of foam films with non-ionic surfactants: alkyl polyglycol ethers compared with alkyl polyglucosides. Tenside Su@ct. Deterg. 2003, 40,
190- 195.
Shrestha, L.K.; K. Aramaki; I. Johansson; PM. Claesson; G. Olanya; C. Stubenrauch. Aqueous
foams stabilized by n-dodecyl-P-D-maltoside, hexaethyleneglycol monododecyl ether and their
1 : 1 mixture (submitted to Soft Matter 2008).
Sottmann, T.; R. Strey; H.-S. Chen. A small-angle neutron scattering study of nonionic surfactant
molecules at the water-oil interface: area per molecule, microemulsion domain size, and rigidity.
J. Cbem. Pbys. 1997, 106,6483-649 1.
Stubenrauch, C. Sugar surfactants-aggregation, interfacial, and adsorption phenomena. Cur%
Opin. Colloid Inte$ace Sci. 2001, G, 160-170.
Stubenrauch, C.; O.J. Rojas; J. Schlarmann; PM. Claesson. Interactions between nonpolar surfaces coated with the nonionic surfactant hexaoxyethylene dodecyl ether C,,E, and the origin of
surface charges at the airlwater interface. Langmuir 2004a, 20, 4977-4988.
Stubenrauch, C.; J. Schlarmann; O.J. Rojas; PM. Claesson. Comparison between interaction
forces at aidliquid and solidlliquid interfaces in the presence of non-ionic surfactants. Tenside
S u ~ a c tDeterg.
.
2004b, 41, 174-1 79.
Tyrode, E.; M.C. Johnson; A. Kumpulainen; M.W. Rutland; PM. Claesson. Hydration state of
nonionic surfactant monolayers at the liquidlvapor interface: structure determination by vibrational sum frequency spectroscopy. J. Am. Cbem. Soc. 2005,127, 16848-16859.
Waltermo, A,; EM. Claesson; S. Simonsson; E. Manev; I. Johansson; V. Bergeron. Foam and thinliquid-film studies of alkyl glucoside systems. Langmuir 1996, 12(22), 5271-5278.

INDEX

Index Terms

Links

A
Acetate esters of monoglycerides

14

Acetyl

43

Acid aspiration
Acryloyl

208
43

Acute respiratory distress syndrome

(ARDS)

160

therapy for

219

N-Acyl-arginine-methyl ester
hydrochloride salts

353

354

355

N-Acylated alkanolamine emulsifiers,

selective enzymatic synthesis of

299

diethanolamine biosurfactants

312

ethanolamine biosurfactants

301

objectives

301

N-Acylated diethanolamine
biosurfactants

312

direct acylation at high substrate loads

317

effect of amount of biocatalyst

316

effect of solvent and temperature

315

effect of the chain length of fatty acid

316

effect of type of solvent on

transacylation of, with
fatty acid ethyl ester

318

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

N-Acylated diethanolamine (Cont.)

elucidation of structures of
reaction products
enzymatic reactions

314
313

ethanolamine and diethanolamine

reactions
structural elucidation of

319
314

transacylation at high substrate

loads

319

N-Acylated ethanolamine
biosurfactants
analysis of reaction mixtures

301
302

direct acylation and transacylation

reactions at relatively high
substrate loads and 60C

309

312

effect of molar ratio of reactants

302

304

effect of solvent

304

306

309

comparative study of
direct acylation and
transacylation reactions
at relatively high loads
of substrates

309

enzymatic reactions

302

screening of biocatalysts

302

solubility of, in different solvents at

310

structural elucidation

302

Acyl chain length

303

ADSA-Constrained Sessile Drop

(ADSA-CSD)

205

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
ADSA-Pendant Drop (ADSA-PD)

Links
205

Adsorption followed by solvent extraction

68

Aerobic fermentations foaming

83

rhamnolipid production

89

Agricultural feedstocks, production

and modification of
sophorolipids from

29

biochemistry/enzymology

39

biosynthesis

30

post-synthetic enzymatic/
chemical modifications of

41

Air/water interfaces

456

Alcohol ethoxylates, toxicity and

429

Alcohol feedstocks

ALEC (artificial lung expanding compound)

160

Algal growth, regulating

426

Algal oils

200

Alkaline hydrolysis or alcoholysis

in ring opening

43

Alkanolamines
applications

299

functional properties

300

Alkyl ethoxylates

13

Alkyl glucamides

449

Alkyl glucoside acyl acceptors, formation of

332

Alkyl glycoside fatty acid esters, synthesis of

332

Alkyl glycosides
synthesis of

9
16

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Alkylphenyl ethoxylates (APES)

need to replace

proposed banning

Alkyl polyglucosides (APGs)

449

Alveofact (bovactant)

197

Amino acid surfactant (lauryl glutamate)

11

Ammonium nitrate

63

Amphiphilic compounds, properties of

Amphiphilic polypeptides

231
10

Animal habitat loss

Animal wastes

Anionic polymers

209

Anionic surfactants

392

Antimicrobial compounds

255

Antimicrobial properties

374

Anti-nerve growth factor (NGF)

258

220

Antitumor and cell

differentiation-inducing probes

257

Antiviral compounds

255

Aqueous media, dispersion of solid in

287

Arginine
monocatenaries from

360

synthesis, aggregation properties,

and applications of
biosurfactants derived from

351

Arginine mono- and diacylglyceride

conjugates, critical micelle
concentration (CMC)values for

369

Arginine-O -alkyl amide

dihydrochloride

353

355

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Arginine-O-alkyl ester
dihydrochloride

353

354

Arthrofactin

131

133

Aspartic acid

432

bulk polymerization of

433

Aspartic acid D-sorbitol copolymer builder

434

Atomic force microscopy

165

Autoinducers

170

84

Axisymmetric drop-shape analysis

(ADSA)-captive bubble (CB)

204

B
Bacillomycin

131

Bacillus Licheniformis, lipopeptides produced by

132

Bacillus megaterium

257

Bacillus mojavenesis JF-2

biosurfactant
Bacillus subtilis

135
131

inhibiting growth of

255

lipopeptides produced by

132

Bacteriostatic activity of sucrose

monopalmitate

284

BASFs Pluronic L43

440

Binary DPPC/Hel 13-5 preparations

166

AFM observations

170

cyclic compression and expansion isotherms

171

FM observations

169

Langmuir isotherms

166

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Binary surfactant mixture
Biobased alkyl ethoxylates, synthesis of

Links
399
16

Biobased amino acid derivatives

Biobased fatty acid

Biobased feedstocks

advantage of, for surfactants

increased interest in utilization of
increased use of

9
vii
6

Biobased products
categories attributed to

defined

Bio based-surfactant commercial products

Biobased surfactants
certification labels for
classification

17
3
6
231

defined

examples and applications

motivation

significance

452

utilization of enzymes for

manufacture
Biochemicals
Biodegradation

16
3
427

of ethoxylates

430

of nonylphenol ethoxylates

430

primary

427

ultimate

427

Biodiesel
development of facilities

20

32

10

12

34

This page has been reformatted by Knovel to provide easier navigation.

35

Index Terms
Biofuels

Links
3

Biological evolution,
integration process of, in
microorganisms
Biomass, use of, as feedstock
Biomaterials
Biomembranes
Bioreactor production process
Biosurfactant flooding experiments

233
449
3
231
70
140

Biosurfactant-mediated oil recovery,

field applications of

141

Biosurfactants (BS)

biological aspects

232

categories

233

functional properties

29

hydrophilic groups

233

nature

232

physicochemical aspects

239

physiological role

237

synthesis

231

51

synthesis, aggregation properties,

and applications of,
derived from arginine
BLES

351
195

197

209

213
binding isotherm for

214

BLES-chitosan system

216

BLES formulations, mechanism of

action of chitosan in

214

This page has been reformatted by Knovel to provide easier navigation.

211

Index Terms
BLES-PEG formulation

Links
213

Block on Ring test (ASTM

D2714 and D2782)

407

Botrytis cinerea

257

Bottom-up method

232

Bovine lung surfactant extract

(BLSE)

197

Bovine spongiform encephalopathy

(BSE)
Branched alkylbenzene sulfonates

160
427

Branched polyaspartic acid/

polysuccinimide, synthesis of
Burkholderia pseudomallei

429
82

C
C7 acid

56

61

C8 acid

56

61

C9 acid

56

61

C10 acid

56

61

C11 acid

56

61

C12 acid

56

61

C14:0 fatty acids

56

C18:l lactones

38

C18:2 ester

56

Calcite, mining of

12

Calystegia sepium

92

Candida apicola

30

238

kinetics of cytochrome
P450 synthesis in

41

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Candida bombicola

Links
30

32

35

38

39

92

93

94

238
effect of dissolved oxygen
concentration on
sophorolipids synthesis by
sophorolipids production by

95
96

Candida magnoliae

59

Candida sp. B-7

53

Candida sp. SY16

56

batch cultivation of

67

Capillary number (NCA)

131

Captive-bubble surfactometer (CBS)

203

Carbodiimide-mediated esterification

44

Carbohydrate-recognition domain (CRD)

241

56

194

Carbon and energy sources,

storage of

238

Carbon dioxide, synthesis of

polyaspartic acid polymer
using supercritical

432

Carboxylate-based soaps

425

Cationic liposomes

262

Cationic polymers

211

Cell cycle regulation

254

Cellobiose lipids, chemical structures of

236

Centrifugation
Chitosan
binding isotherm for

213

68
211

213

214

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Chocolate, viscosity control of molten

289

293

Cholesteryl-3carboxyamindoethylene-Nhydroxyethylamine (Chol-OH)
-Cholesteryl glucoside, biosynthesis of

263
40

Chun-Huh relationship

112

Citric acid in cleaning products

436

Citric acid-sorbitol copolymer

builder

436

Clean Water Acts

426

Climate change

Climate change, potential danger of

Clorox's Green Works

Coacervates
formation of

vii
431
246
246

Coalescence activation energy

estimation based on
microfiltration
microemulsion stability and
Coconut oils
Cold thermal storage

410
410
4
254

Commercial exogenous lung

surfactants

195

Complexation agents, synthesis of

saccharide fatty acid esters
using
Co-product feedstocks
Corn oils

332
32

35

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Corrosion inhibition

299

Co-surfactant, effect of

402

CP-B

161

Critical aggregate concentration (CAC)

239

Critical micelle concentrations (CMCs)

29

of esters

241

of lipopeptides

136

108

282

measurement of, with Wilhelmy

plate method
values

369
360

for arginine mono-and

diacylglyceride
conjugates

369

Critical microemulsion
concentration (CC)

108

Crystal control of fat in oil-in-water

emulsion

293

Crystallization

68

Cuphea

Curosurf (poractant alpha)

160

Cyclic compression and expansion isotherms

171

CYP52E1

41

CYP52E2

41

199

D
Dai-Ichi Kogyo Seiyaku Co.

275

Dai-Nippon Sugar Manufacturing Co.

275

Deacylation reaction
Deforestation

43
6

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Denitrification

Links
88

89

advances in rhamnolipid
production by
Depletion attraction mechanism

87
208

Derivatization, synthesis of
saccharide fatty acid esters via
Detergency

330
299

Detergent builders
aspartic acid polymers as

434

biodegradability and

428

functions of

426

water quality and

426

Detergent industry

vii

environmental goals for

market value within European Union

use of natural products in

425

Detergents
modern

425

synthetic

425

-D-galactose monosaccharides
Diethanolamine

426

38
300

312

effect of type of solvent on

transacylation of, with
fatty acid ethyl ester
Diethylamine
Dilatational elasticity

318
43
216

218

L--D-Dilauroylphosphatidylcholines (DLPC)
asymmetric distribution of

247
247

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Dimethylformamide (DMF) in
production of SEs

277

Dimethylsulfoxide (DMSO)
effect of, on lipase

329

in production of SEs

277

Dimorphecolic acids

Dimorphotheca

-D-Dioleoylphosphatidylethanolamine (DOPE)

263

Dipalmitoylphosphatidylcholine
(DPPC)

157

Disaccharide

30

Disaturated phospholipids

182

194

158

Dishwashing detergent, global

market of

425

DLVO theory

287

DNA compaction

381

Dow Chemical Company

431

DPPC/PG/PA mixture

160

D-sorbitol

434

melting point of
Dynamic light scattering (DLS)

289

436
241

E
Eco-friendliness, environmental
impact and

Eco-friendly products, desire to

purchase

This page has been reformatted by Knovel to provide easier navigation.

201

Index Terms
Emulsification

Links
29

299

Energy, U.S. Department of, Energy

Star program
Energy consumption, standard of living and

431
129

Environmental and product

performance concerns

451

Environmental goals for surfactants

and detergents industry

Environmentally advanced
surfactants

244

Environmentally-sensitive
applications, surfactants for

Environmental remediation,
surfactants for

Enzymes, utilization of, for the

manufacture of biobased
surfactants

16

EO alkyl ethers

451

20

Epoxidized methyl oleate and

glycerol
synthesis of surfactant from

439

synthesis of surfactant from

glycerol and
Epoxidized seed oils

439
4

Equivalent alkane carbon number

(EACN)
Erythritol

115
58

62

ESO polymerization in forming

surfactant or hydrogel

441

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Esterquats

Links
12

Esters, critical micelle concentration of

241

Esters of amides, antimicrobial activities of

299

Ethanolamine
Ethanolamine and diethanolamine reactions

300

319

Ethoxylated fatty acids or

monoglycerides

13

Ethoxylated fatty amines

13

Ethoxylated sorbitan esters

13

Ethoxylates, biodegradation of
Ethylene-glycol esters
Ethylene oxide

15

430
14
5

Ethyoxysulfonates

430

Eutrophication

426

Exogenous lung surfactants

commercial

195

historical overview and medical

needs for

191

Exosurf (colfosceril palmitate)

160

Extraction phenomenon

160

201

F
Falex Pin and V-block test (ASTM
D 2670 and D3233)

407

Fatty acid esters, regioselective synthesis of,

of monosaccharides and related
compounds using lipases and proteases

327

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Fatty acid ethoxylates, uses of

Links
13

Fatty acid methyl esters (FAMEs)

conversion into fatty alcohol

12

synthesis of

16

10

16

169

175

Fatty acid methyl esters (FAMEs)

feedstock.
Fatty acid-polyol esters, applications of

12
13

Fatty-acid soaps

451

Fatty acid type of biosurfactants

233

Fatty alcohols

fatty acid methyl esters (FAMEs)

conversion into
Federal Water Pollution Control Act (1972)
Flotation agents
Fluorescence microscopy (FM)

12
426
12
164

Foam fractionation

68

Foaming

29

in aerobic fermentation

83

Foam stabilization

299

Fossil fuels as energy source

129

Four Ball Wear test (ASTM D4172)

407

Freeze fracture electron microscopy (FF-EM)

241

Furfuryl

G
Galactose

51

Galactose sulphate

52

-Galactosidase activity

38

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Gangliosides

259

Gel permeation chromatography

433

Gemini arginine surfactants or bis (Args)

356

Gemini esterquats

12

Gemini surfactants

379

Gemini surfactants bis(Args)

365

Gene carriers

262

Gene delivery

254

Geotrichum candidum ST 002515

Global oil production

53
129

Global surfactant industry

Global warming

Glomerella cingulata
Gluconeogenesis
Glucosamine
Glucose
in MEL formation

257
58
209
51
61

Glucose/oleic acid fermentations

38

Glucosyltransferase

59

Glucuronic acid

52

Glutamic acid
Glycerine
Glycerol

432
4

16

35

238

synthesis of surfactant from

epoxidized methyl oleate and
Glycerolipid arginine surfactants
Glycol esters, applications of

439
357
13

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Glycolipid biosurfactants

233

antimicrobial activities

256

interfacial properties

239

potential applications

254

self-assembling properties

244

surface and interfacial tension

240

types

237

Glycolipid ester
Glycolipids

264
29

Glycols

30

51

Gl ycosphingolipids

258

Greenhouse gases

Green technology

266

Gujarat Chemicals (Nanpura, India)

13

H
Halotolerant yeasts

238

HEL 13-5, interfacial properties of

162

adsorption isotherms

163

atomic force microscopy

165

170

fluorescence microscopy

164

169

peptide structures

162

175

temperature dependence in
Langmuir isotherms

163

Henkels Persil laundry detergents

432

High-stretch ventilation

208

HL60 cells, MEL-A and MEL-B in

inhibiting growth of

257

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Hookes law
Huish Detergents

Links
453
10

Huntsman Chemical Company

120

Hyaluronan

209

as lung surfactant additive

211

Hydrogel, ESO polymerization in forming

441

Hydrogen bonding

232

Hydrolyzed polymerized epoxidized

soybean oil (HPESO)

442

Hydrophile-lipophile balance (HLB)

method

394

Hydrophilic groups of biosurfactants

233

Hydrophilicity

277

of sophorolipid

282

122

Hydrophilic-lipophilic balance
(HLB) method

114

115

275

468
wide range of values

323

Hydrophobic and hydrophillic

regions and surface activity
Hydrophobic forces

138
232

Hydrophobic groups of
biosurfactants (BS)
13-Hydroxydocosanoic acid

233
40

I
Illinois River, contamination of

426

Immunoglobulin G (IgG)

259

Immunoligands

258

This page has been reformatted by Knovel to provide easier navigation.

277

Index Terms
Infasurf (CLSE or calfactant)

Links
197

199

International Organization for

Standardization protocols ISO 14024

Ionic liquid phase system, synthesis

of saccharide fatty acid esters in
Isopropylamine

335
5

Iturin A

133

Iturins

131

J
Jatropha, oils from

Jeneil Biosurfactant Co.

115

JF-2 lipopeptide

136

K
Kurtzmanomyces sp I-11

53

56

L
Lactate esters of monoglycerides

14

Lactic acid

16

Lactonic sophorolipids

98

purification of

96

Langmuir film balance

278

Langmuir trough

202

Langmuir-Wilhelmy Balance (LWB)

201

Laplace equation

204

Lard

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
L-aspartic acid
melting point of
Laundry, global market of
LC/APCI-MS method

Links
434
436
425
39

Lesquerella

Lesquerolic acids

Levoglucosanyl compounds

Lichenysin A

135

Lichenysins

131

133

Life Cycle Assessment (LCA)

16

Lignin

Lignin sulfonates

13

Linear alkylbenzene sulfonates (LASs)

10

Linear and branched alkyl ethoxylates

Linear polyaspartic acid, synthesis of

12

427

429

Linear variable differential

transformer (LVDT) sensor

453

Lion (Japanese oleochemical

manufacturer)

10

Lipase-catalyzed synthesis of
saccharide fatty acid esters,
motivation for
Lipase P3

325
42

Lipases, regioselective synthesis

of fatty acid esters of
monosaccharides and related
compounds using proteases and

327

Lipoaminoacids

351

Lipofectin

263

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Lipopeptide biosurfactants

Links
131

biosynthesis

133

heterogeneity of structures

133

main structural features

131

regulation
common transcription, with
other starvationinduced phenomena

135

transcriptional, of surfactin
biosynthesis
Lipopeptides

134
29

critical micelle concentrations

136

heterogeneity

133

oil recovery

140

types

131

233

Lipopeptide surface and interfacial

activities
effect of temperature, pH, and salinity

136
138

hydrophobic and hydrophilic

regions and surface activity

138

improving biosurfactant activity

changes in fatty-acid composition

139

changes to amino-acid composition

138

formulating biosurfactant mixtures

140

Lipophilicity of sophorolipid

122

Lipophobicity

282

Lipopolysaccharide, sequestration of

380

Lipopolysaccharide protein complexes

284

30

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Lipoproteins

Links
29

Liquid laundry detergents, demand

for more concentrated
Long chain fatty acids, biosynthesis of

8
40

Lungs, deficiency symptoms in

160

Lung surfactant replacement therapy

192

Lung surfactants

191

additives
anionic polymers

209

cationic polymers

211

nonionic polymers

208

commercial exogenous

195

natural extracts

195

synthetic

200

213

evaluation
in situ

206

in vitro

201

in vivo

207

future trends and

219

historical overview and medical

needs for exogenous
discovery and physiology

192

respiratory distress syndrome

and lung surfactants

191

respiratory distress syndrome

(RDS) and
Lysophosphatidycholine (LPCs)

191
197

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

M
Magnesium stearate

294

Mannitol

238

Mannose

51

Mannosylerythritol lipids (MEL-A and MEL-B)

Mannosylerythritol lipids (MELs)

62

257
5

10

30

51

238

239

241

245

255
analyses

58

biosynthesis

58

chemical structures
downstream processing

233
67

influence of different substrates

on variation of fatty
acid moiety in, using
Pseudozyma (Candida)
antarctica T.

55

moisturizing properties

265

production
bioreactor

64

substrates and shake flask

cultivation
structure and microbial origin

61
52

variation of fatty acid moiety

in, depending on
microorganism and
substrate

54

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Measurements and analysis of

surface interactions and
forces (MASIF)

453

Meconium aspiration

208

MEL-A

52

asymmetric distribution of

247

binding affinity of

261

56

239

56

239

critical aggregate concentration

(CAC) of

239

emulsifying activity of

241

formation of coacervates

246

in inducing neurite outgrowth

257

in inhibiting growth of HL60 cells

257

self-assembling manner of

245

self-assembling properties of

247

MEL-B

52

critical aggregate concentration

(CAC) of

239

formation of vesicles

246

in inhibiting growth of HL60 cells

257

self-assembled structure of

248

self-assembling manner of

245

self-assembling properties of

247

MEL-C
formation of vesicles
MEL-D
MEOR, biosurfactants used in

52

241

246
52
136

This page has been reformatted by Knovel to provide easier navigation.

241

Index Terms
Metalworking fluids (MWFs)

Links
3

classification

389

defined

389

environmental concerns

390

manufacturing performance evaluation

407

microemulsion as swollen micelles

395

motivation for biobased

390

semisynthetic

389

surfactant selection for formulations

392

surfactants used in

392

synthetic

389

water-immiscible

389

water-miscible

389

Methacryloyl
Methyl ester sulfonates (MESs)
synthesizing

390

43
7

10

12

10

Methyl tertiary butyl ether (MTBE)

58

Micellar aggregation

29

Micelle solubilization, effect of

surfactant structure on

396

Microbial surfactants, use of, in

industrial applications

51

Microdroplet method

206

Microemulsions

390

phase behavior and interfacial tension

108

properties of

107

rhamnolipid-based

114

solubilization and Winsor R-ratio

110

This page has been reformatted by Knovel to provide easier navigation.

16

Index Terms

Links

Microemulsions (Cont.)
sophorolipid-based

122

use of biosurfactants in

107

Microfiltration, coalescence
activation energy estimation based on

410

Microfiltration-based metalworking
fluids (MWF) recycling

414

Minimum inhibitory concentrations

(MIC) of MEL

255

Mississippi River, contamination of

426

Mitsubishi Chemical Co.

275

Mitsubishi-Kagaku Foods Corporation

275

grades of Ryoto Sugar ester manufactured by

Monoacylation
Mono- and diacylglycerols
Mono- and di-glyceride arginine surfactants

324

276
42
4
379

minimum inhibitory
concentration (mg/L) of

378

Monocatenaries, from arginine

360

Monocatenary arginine surfactants

353

minimum inhibitory
concentration (mg/L) of
Mono-corynomycolate (TL-2)

375
243

Mono-glucopyranosyl-13-hydroxydocosanote

39

Monoglycerides

14

applications of

13

Monohydroxy fatty acids

31

Monomethyl -sulfosuccinate

12

Moorela thermoacetica

285

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Multicompoent DPPC/PG/PA/Hel 13-5

systems and surfactants

175

179

comparisons to surfactant
(surfactant TA) in hysteresis measurements

182

FM observations

179

hysteresis curves

181

Langmuir isotherms

175

Mycosubtilin

131

Myristic acid

328

181

179

N
Nanotechnology

232

development of

265

Natural ceramides

265

The Natural Step

Natural surfactants' extracts

6
195

Neonatal respiratory distress

syndrome (NRDS)

160

Neutral lipids

29

Niche market

Nicotinamide adenine dinucleotide

phosphate (NADPH)
Nitrate reductase

94
88

Nitrilotriacetic acid

426

Nitrogen starvation

238

Nonionic polymers

208

Nonionic surfactants
Nonribosomal peptide synthetases

392

451

133

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Nonylphenol (NP) ethoxylates
biodegradation of

Links
451
430

42

as biocatalyst

334

Novozyme 435

fructose-palmitic acid
esterification catalyzed by

343

reuse of

334

synthesis of 6-O-lauroyl-D-glucose

336

Oil-in-emulsion, crystal control of fat in

293

Oil production
factors limiting

130

global
U.S.
Oil recovery by lipopeptides

129
129
140

Oleic acid

61

Oleochemical biorefinery

16

Oleochemical feedstocks

Oleochemical prices, growth in

Oleochemical -yielding crops

Oligosaccharide lipids
chemical structures of

257
236

Organic solvents, synthesis of

saccharide fatty acid ester in

325

Osmotic stress, protection against

238

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

P
P. fusiformata, ustilagic acids produced by
PAI2 concentrations
Palm

255
85
4

Palmitoyl-oleoyl-phosphatidylglycerol (POPG)
Palm-kernel oil
Palm-mid-fraction (PMF)
Palm oils
production of

201
4
293
4

Palm stearine oil

Peracetylation
Personal care, global market of
Petroleum

12

Palm stearine

Patch model

217

219

44
425
vii

pH, effect of, on sophorolipids

formation
Phase partitioning
Phosphates

95
29
8

Phosphatidylcholines

192

Phosphatidylethanolamines (PEs)

192

Phosphatidyllinositols (PIs)

192

Phosphatidylserines (PSs)

192

Phospholipids, pharmaceutical
application of
Phytophtora capsici

13
255

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Pichia strains

238

Plasmopara lactucaeradicis

255

Polyacrylic acids

426

Polyaspartic acid, biodegradability and

429

Polyaspartic acid polymer, synthesis

of, using supercritical carbon dioxide

432

Polycarboxylic acid, potential use in

detergent builders

438

Polyethylene glycol

Polyglycerol esters

13

applications of

13

Polyglycerol polyricinoleate (PGPR)

289

Polymer type of biosurfactants

233

15

Polyol and amino acid-based

biosurfactants, builders, and hydrogels

425

aspartic acid D-sorbitol

copolymer builder

434

citric acid-sorbitol copolymer builder

436

economic factors

431

environmental factors
biodegradation

427

energy consumption

431

toxicity

429

water quality

426

ESO polymerization in forming

surfactant or hydrogel

441

synthesis of polyaspartic acid

polymer using supercritical
carbon dioxide

432

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Polyol and amino acid-based (Cont.)

synthesis of surfactant from
epoxidized methyl oleate and glycerol
Polyol compounds

439
238

Polysaccharide-protein-fatty acid complexes

30

Polysorbates

13

Polysuccinimide (PSI)

432

Post-synthetic enzymatic/
chemical modifications of
sophorolipids
Precipitation

41
68

Premature animal model

207

Pressure-Volume (P-V) method

206

Primary biodegradation

427

Primary oil production

130

Proctor & Gamble

431

Pure Essentials of

431

1, 2-Propanediol

Propylene glycol

Propylene-glycol (1,2-propanediol) esters

14

Proteases, regioselective synthesis

of fatty acid esters of
monosaccharides and related
compounds using lipases and
Proteins

327
16

Protein sensing and/or separation

254

Pseudomonas aeruginosa

257

growth of

237

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Pseudomonas putida

131

Pseudomonas sp. MIS38

133

Pseudomonas spp., lipopeptides

produced by

132

Pseudozyma

53

Pseudozyma aeruginosa

82

Pseudozyma antarctica

53

56

Pseudozyma antarctica T-34

57

58

Pseudozyma aphidis

56

59

Pseudozyma aphidis
DSM 70725
Pseudozyma clororaphis

56
82

Pseudozyma graminicola
CBS 10092
Pseudozyma hubeiensis KM-59

56
56

Pseudozyma rugulosa
NBRC 10877

57

Pseudozyma shanxiensis
CBS 10075

58

Pseudozyma sp. KM-160

56

Pseudozyma tsukubaensis

56

Pseudozyma tsukubaensis
JCM 10324

56

Pulmonary surfactants

157

components and compositions

157

deficiency symptoms in lungs

160

peptide designing for replacement

treatments

161

This page has been reformatted by Knovel to provide easier navigation.

238

Index Terms

Links

Pulmonary surfactants (Cont.)

research and development
binary DPPC/Hel 13-5
preparations
interfacial properties of HEL 13-5

166
162

multicomponent DPPC/PG/
PAHel 13-5 systems and

175

179

ternary DPPC/PG/Hel 13-5

and DPPC/PAHel 13-5
systems

172

Pulsating-bubble surfactometer (PBS)

203

Putisolvin

131

Pythium aphanidermatum

255

R
Reactive extrusion

438

Respiratory distress syndrome (RDS)

acute

191

adult

191

lung surfactants and

191

neonatal

191

192

Reverse phase high-performance

liquid chromatography (RPHPLC)

163

Rhamnolipid-alone microemulsions

115

Rhamnolipid-based microemulsions

114

Rhamnolipid-co-surfactant
microemulsions

118

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Rhamnolipids

applications
chemical structures of
production

Links
5

10

30

107

243

252

81
234
82

advances in, by denitrifying

fermentation
technology

87

biosynthesis pathway and

regulations
ongoing and future work

84
90

producing microorganisms and

characteristics

82

properties

79

structures

77

Rhamnose

81

Rhamnosyl--hydroxydecanoyl-hydroxydecanoate (Rha-C10-C10)

77

Rhamnosyl-rhamnosyl-hydroxydecanoyl-hydroxydecanoate (Rha-RhaC10-C10)

77

Rhizotecnia erythropolis

257

Rhizotecnia solani

257

Rhodotorula bogoriensis

30

Rhodotorula bogoriensis sophorolipids

40

Ricinoleic acid
Ring opening
Roundtable for Sustainable Palm Oil

39

4
43
7

This page has been reformatted by Knovel to provide easier navigation.

77

Index Terms

Links

S
Saccharide esters

14

Saccharide fatty acid esters

effects of water activity and
content in synthesis of

343

motivation for lipase-catalyzed

synthesis of
properties and applications

325
323

synthesis
in ionic liquid phase system

335

in organic solvents

325

in solid phase system

334

in solventless system

339

in supercritical CO2

329

using complexation agents

332

via derivatization

330

Saccharide-fatty acid esters

Saccharide/lipidic feedstocks
Saccharides

9
31
9

16

Saccharomyces cerevisiae,
trehalose synthesis of

238

Saline-lung Lavage Model

208

Saponins

13

SARAYA Co., Ltd. (Osaka, Japan)

243

Sasol Chemical Company

119

Schizonella melanogramma

53

Schizonellin

53

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Secondary oil production
Seed oil

Links
130
4

Selective enzymatic synthesis of

N-acylated alkanolamine
emulsifiers

299

diethanolamine biosurfactants

312

ethanolamine biosurfactants

301

objectives

301

Self-assembling properties of
glycolipid biosurfactants

244

Semisynthetic metalworking fluids

(MWFs)
Sequestration of lipopolysaccharide
Serine

389

390

380
5

Serratia marcescens

133

Serrawettin W2

133

Seventh Generation (Burlington, Vermont)

431

Short-chain alkyl polyglucosides

Sierra Club

Skin care

254

Skin care materials

264

Small Angle Neutron Scattering

(SANS) measurements
Small-angle x-ray scattering (SAXS)

465
281

Small/wide-angle X-ray scattering

(SAX/WAXS)

248

Smoluchowski equation

410

Soap and Detergent Association

425

Sustainability Central online forum

431

426

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Soap industry, market value within

European Union
Soapnut, oils from

4
4

Sodium citrate, structure of

427

Sodium diphosphate

426

Sodium dodecyl sulfate (SDS)

243

Sodium ethylenediaminetetraacetate
(EDTA) structure of
Sodium nitrate
Sodium nitriloacetate (NTA), structure of

427
63
427

Sodium polyacrylate as effective

dispersing agent and metal
ion chelator

428

Sodium triphosphate

426

Sodium tripolyphosphate, structure of

427

Soil remediation

254

Solid/liquid interfaces

460

Solid phase system, synthesis of

saccharide fatty acid esters in

334

Solubilization

110

defined

110

Solubilization potential, defined

110

Soluble oils

389

Solvay Chemicals (Deer Park, TX)

13

Solvent extraction

68

Solventless system, synthesis of

saccharide fatty acid esters
Sophorolipid-based microemulsions

339
122

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Sophorolipids

Links
5

10

30

241

243

252

antiradical and anti-elastic

properties

264

applications

81

biosynthesis

30

chemical structures

234

hydrophilicity/lipophilicity

122

introduction and applications

107

30

post-synthetic enzymatic/
chemical modifications of
production

41
92

biosynthesis pathway

92

effects of culture conditions on formation

93

purification and

96

producing microorganisms and characteristics

92

production and modification

of, from agricultural feedstocks

29

biochemistry/enzymology

39

biosynthesis

30

post-synthetic enzymatic/chemical
modifications of

41

properties

79

structures

77

synthesis

30

Sophorose

31

Sorbitan derivatives, applications of

13

Sorbitan esters

15

ethoxylated

238

15

This page has been reformatted by Knovel to provide easier navigation.

77

Index Terms

Links

Sorbitol

Soybean oils

as biodegradable

56

194

195

194

195

161

194

194

195

427

Soybean processing

36

Soy molasses

36

SP-A

157

Span

13

SP-B

157

159

197

213

157

159

195

197

SP-D

158

159

Sphingomyelins (SPHs)

192

SrfAD

134

SP-C

Standard for Comparing Metal

Removal Fluids Using
the Tapping Torque Test
Machine (ASTM D5619)

407

Standard of living, energy

consumption and

129

Staphylococcus epidermidis, inhibiting

growth of

255

Starmella bombicola

92

Stepan

10

Sterols

13

Streptococcusfaecium, inhibiting
growth of
Structure based surfactant selection

57

255
395

Submerged aeration

87

Succinic acid

12

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Succinoyl
Succinoyl-trehalose lipids (STL-1 and-2)
chemical structures of
Sucrose esters

Links
43
243
235
14

applications
bacteriostatic activity of
sucrose monopalmitate

284

crystal control of fat in oil-in-water emulsion

293

dispersion of solid in aqueous media

287

pharmaceutical

294

chemical structure of

275

Sucrose fatty acid esters

molecular shape of

278

surface activity of

281

Sucrose monopalmitate

286

bacteriostatic activity of

284

effect of on heat stability of

microbes at 121C
Sucrose monostearate

287
289

chemical structure

276

in dispersion of solid in aqueous media

287

Sugar-based surfactants

244

interfacial properties

449

289

environmental and product

performance concerns

451

fundamental studies on

455

air/water interfaces

456

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Sugar-based surfactants (Cont.)

comparison between
EO-and sugar-based surfactants:
structural aspects, interfacial aspects,
packing at the interface
solid/liquid interfaces

460
460

viscoelasticity properties of
isomeric sugar-based surfactants

462

measurements and analysis of

surface interactions and forces

453

molecular prospects

464

motivation

449

nonionic

451

significance of biobased

452

structure-property relationships

467

surface light scattering

454

thin-film pressure balance

454

Sugar esters

449

Supercritical CO2, synthesis of

saccharide fatty acid esters in
Surface activity

329
29

Surface light scattering (SLS)

454

Surfactant (Surfactant TA)

199

Surfactant industry

vii

environmental goals for

global

Surfactants

131

acute toxicity

430

anionic

392

biobased, significance

452

133

160

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Surfactants (Cont.)
classification

29

defined

29

environmentally advanced

244

ESO polymerization in forming

441

industrial uses

231

nonionic

392

451

optimization of concentrations
for maximum stability of
vegetable oil metalworking
fluids
poor biodegradability

409
427

selecting, based on molecular

structure and micelle
solubilization

397

structures

233

sugar-based

244

synthesis of, from epoxidized

methyl oleate and glycerol

439

used in metalworking fluids (MWF)

392

use of natural product based

427

Zwitterionic

392

Surfactant selection guidelines

405

Surfactant structure, effect of,

on micelle solubilization

396

Surfactin biosynthesis,
transcriptional regulation of
Surfaxin (KL, or lucinactant)

134
160

201

This page has been reformatted by Knovel to provide easier navigation.

Index Terms
Survanta (beractant)

Links
160

199

Sustainable metalworking fluids

(MWF) systems

412

Synthetic detergents
advantages of

425

elimination of phosphates from

426

negative environmental impact of

426

Synthetic metalworking fluids (MWFs)

389

Synthetic surfactants

200

T
Tallow

Ternary DPPC/PG/Hel 13-5 and

DPPC/PA/Hel 13-5 systems

172

FM observations for DPPC/PA/

Hel 13-5 system

175

hysteresis curves

175

Langmuir isotherms

172

Tertiary oil recovery

Thermal synthesis

432

Thermodynamically stable vesicles

247

Thin-film pressure balance (TFPB)

454

Threonine

Top-down method

232

Torulopsis apicola

92

Torulopsis magnoliae

30

Toxicity, alcohol ethoxylate and

92

429

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Transacylation
effect of type of solvent on, of
diethanolamine with fatty
acid ethyl ester
at high substrate loads

318
319

Transcriptional regulation of
surfactin biosynthesis
Trehalose di-corynomycolate (TL-1)

134
243

Trehalose esters

Trehalose lipids

10

chemical structures of

238

243

58

59

235

Triglycerides, sophorolipid
synthesis from

93

Tsukamurella sp.

257

Tubular myelin

157

Tween
Type II pneumocytes

13
192

U
Ultimate biodegradation

427

U.S. oil production

129

Uridine diphosphate (UDP)

glucose:sterol
glucosyltransferase

40

Ustilagic acids, produced by P.

fusformata

255

Ustilago maydis

52

Ustilipids

52

This page has been reformatted by Knovel to provide easier navigation.

61

Index Terms

Links

V
Vacuum oven synthesis

438

Van der Waals attractive potential

289

Van der Waals force

465

Van der Waals interactions

232

Vegetable oil-based metalworking fluids (MWF)

microemulsions, case study of

412

Vegetable oil metalworking fluid

microemulsions, design of,
using biobased surfactants

389

Vegetable oil metalworking fluids

(MWF) optimization of
surfactant concentrations
for maximum stability of

409

Vegetable oil type, effect of

405

Venticute (recombinant SP-C or lusupultide)

201

Vesicles, formation of

245

Vibrational sum frequency

spectroscopy (VSFS)

464

Viscoelasticity properties of isomeric

sugar-based surfactants

462

Viscosity control of molten chocolate

289

Viscosity enhancement

299

293

W
Walmart

432

Water activity and content, effects of,

in synthesis of saccharide fatty acid esters

343

This page has been reformatted by Knovel to provide easier navigation.

Index Terms

Links

Waterflooding

130

Water-immiscible metalworking fluids (MWF)

389

Water-in-oil (W/O) emulsion formation

282

Water-in-oil (W/O) microemulsions

241

Water-insoluble substrates, assimilation of

237

Water-miscible metalworking fluids (MWF)

389

Water penetration method

249

Water quality

426

impact of synthetic detergents on

143

426

Water Quality Acts

426

Well treatments

141

Wetting

29

Wickerhamiella domericqiae

92

143

Wilhelmy method

239

Wilhelmy Plate method

202

278

Winsor R-ratio

113

115

Winsor Type I microemulsions

108

114

Winsor Type II microemulsions

108

Winsor Type III microemulsions

108

Winsor Type IV microemulsions

108

W/O high internal phase emulsion (HIPE)

282

369

X
Xylitol

This page has been reformatted by Knovel to provide easier navigation.