Two Phase Flows Sample

Single and Two-Phase Flows on Chemical
and Biomedical Engineering

Editors
Ricardo Dias
Department of Chemical and Biological Technology,
ESTiG, Braganca Polytechnic, C. Sta. Apolonia, 5301-857
Bragança, Portugal
CEFT - Transport Phenomena Research Center,

Faculty of Engineering, University of Porto (FEUP),
Rua Dr. Roberto Frias, s/n, 4200-465 Porto Portugal
Antonio A. Martins
CEFT-Transport Phenomena Research Center,
Faculty of Engineering, University of Porto (FEUP)
Rua Dr. Roberto Frias, s/n, 4200-465 Porto, Portugal
Rui Lima
Department of Mechanical Technology, ESTiG,
Braganca Polytechnic, C. Sta. Apolonia, 5301-857
Bragança, Portugal
CEFT - Transport Phenomena Research Center

Teresa M. Mata
LEPAE-Laboratory for Process,
Environmental and Energy Engineering
eBooks End User License Agreement
Please read this license agreement carefully before using this eBook. Your use of this eBook/chapter constitutes your agreement
to the terms and conditions set forth in this License Agreement. Bentham Science Publishers agrees to grant the user of this
eBook/chapter, a non-exclusive, nontransferable license to download and use this eBook/chapter under the following terms and
conditions:
1. This eBook/chapter may be downloaded and used by one user on one computer. The user may make one back-up copy of this
publication to avoid losing it. The user may not give copies of this publication to others, or make it available for others to copy or
download. For a multi-user license contact permission@benthamscience.org
2. All rights reserved: All content in this publication is copyrighted and Bentham Science Publishers own the copyright. You may
not copy, reproduce, modify, remove, delete, augment, add to, publish, transmit, sell, resell, create derivative works from, or in
any way exploit any of this publication’s content, in any form by any means, in whole or in part, without the prior written
permission from Bentham Science Publishers.
3. The user may print one or more copies/pages of this eBook/chapter for their personal use. The user may not print pages from
this eBook/chapter or the entire printed eBook/chapter for general distribution, for promotion, for creating new works, or for
resale. Specific permission must be obtained from the publisher for such requirements. Requests must be sent to the permissions
department at E-mail: permission@benthamscience.org
4. The unauthorized use or distribution of copyrighted or other proprietary content is illegal and could subject the purchaser to
substantial money damages. The purchaser will be liable for any damage resulting from misuse of this publication or any
violation of this License Agreement, including any infringement of copyrights or proprietary rights.
Warranty Disclaimer: The publisher does not guarantee that the information in this publication is error-free, or warrants that it
will meet the users’ requirements or that the operation of the publication will be uninterrupted or error-free. This publication is
provided "as is" without warranty of any kind, either express or implied or statutory, including, without limitation, implied
warranties of merchantability and fitness for a particular purpose. The entire risk as to the results and performance of this
publication is assumed by the user. In no event will the publisher be liable for any damages, including, without limitation,
incidental and consequential damages and damages for lost data or profits arising out of the use or inability to use the publication.
The entire liability of the publisher shall be limited to the amount actually paid by the user for the eBook or eBook license
agreement.
Limitation of Liability: Under no circumstances shall Bentham Science Publishers, its staff, editors and authors, be liable for
any special or consequential damages that result from the use of, or the inability to use, the materials in this site.
eBook Product Disclaimer: No responsibility is assumed by Bentham Science Publishers, its staff or members of the editorial
board for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any
use or operation of any methods, products instruction, advertisements or ideas contained in the publication purchased or read by
the user(s). Any dispute will be governed exclusively by the laws of the U.A.E. and will be settled exclusively by the competent
Court at the city of Dubai, U.A.E.
You (the user) acknowledge that you have read this Agreement, and agree to be bound by its terms and conditions.
Permission for Use of Material and Reproduction
Photocopying Information for Users Outside the USA: Bentham Science Publishers grants authorization for individuals to
photocopy copyright material for private research use, on the sole basis that requests for such use are referred directly to the
requestor's local Reproduction Rights Organization (RRO). The copyright fee is US $25.00 per copy per article exclusive of any
charge or fee levied. In order to contact your local RRO, please contact the International Federation of Reproduction Rights
Organisations (IFRRO), Rue du Prince Royal 87, B-I050 Brussels, Belgium; Tel: +32 2 551 08 99; Fax: +32 2 551 08 95; E-mail:
secretariat@ifrro.org; url: www.ifrro.org This authorization does not extend to any other kind of copying by any means, in any
form, and for any purpose other than private research use.
Photocopying Information for Users in the USA: Authorization to photocopy items for internal or personal use, or the internal
or personal use of specific clients, is granted by Bentham Science Publishers for libraries and other users registered with the
Copyright Clearance Center (CCC) Transactional Reporting Services, provided that the appropriate fee of US $25.00 per copy
per chapter is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers MA 01923, USA. Refer also to
www.copyright.com
CONTENTS
Foreword i
Preface ii
List of Contributors iii
Acknowledgements xii
Part I: Single Phase Flows on Chemical Enginnering
1. Spacims-Probing the Internal Behaviour of 3D Structured

Materials 3
Jacinto Sá, Cristina-Elena Stere and Alexandre Goguet
2. Modelling of a Monolithic Reverse Flow Reactor for Selective

Catalytic Reduction of NO by Ammonia 26
Emilio Muñoz, David Lesser, Pablo Marín, Salvador Ordóñez and
Fernando V. Díez
3. Mesoscopic Simulation of Rarefied Gas Flow in Porous Media 52

Alexandros N. Kalarakis, Eugene D. Skouras and Vasilis N.
Burganos
4. Mixing Through Half a Century of Chemical Engineering 79

Ricardo J. Santos, Madalena M. Dias and José Carlos B. Lopes
5. Application of the Probability Density Function Method to

Turbulent Mixing with Chemical Reaction 113
Andrei Chorny
6. Activated Sludge Models Coupled to CFD Simulations 153

Pereira J.P., Karpinska A., Gomes P.J., Martins A.A., Dias M.M.,
Lopes J.C.B. and Santos R.J.
7. Using CFD to Estimate External Mass Transfer Coefficients and
Intra-Particle Diffusional Effects on the Supercritical
Hydrogenation of Sunflower Oil 174
A. Guardo, E. Ramírez, M.A. Larrayoz and F. Recasens
8. Mass Transfer Around a Single Soluble Solid with Different

Shapes Buried in a Packed Bed and Exposed to Fluid Flow 196
J.M.P.Q. Delgado
9. Advances on Viscoelastic Fluid Flow Simulation 233

Jovani L. Favero, Argimiro R. Secchi, Nilo S. M. Cardozo and
Hrvoje Jasak
10. Tree-Shaped Flow Structures Viewed from the Constructal

Theory Perspective 266
António F. Miguel
11. Determination of Effective Transport Properties of Metallic

Foams: Morphology and Flow Laws 292
J. Vicente, E. Brun, J.M. Hugo, J.P. Bonnet and F. Topin
Part II: Two Phase Flows on Chemical Enginnering
12. Flow Visualization in Gas-Solid Packed Beds by Spatially

Resolved Near-Infrared Imaging 332
Aiouache Farid, Nic An tSaoir Méabh and Luis Abreu Fernandes
Daniel
13. Interfacial Area Modelling in Two-Phase Flow Studies 361

Christophe Morel
14. Mass Transfer Models for Oxygen-Water Co-Current flow in

Vertical Bubble Columns 386
Valdemar Garcia and João Sobrinho Teixeira
15. Characterization of Turbulence and Flow Regimes in Bubble
Columns Based on Nonlinear Chaos Analysis of Various Data 412
Stoyan Nedeltchev
16. Characterization and Modeling of Flotation Processes 440

G.G. Kagramanov and V.A. Kolesnikov
Part III: Single and Two Phase Flows on Biomedical Enginnering
17. Pulse Wave Propagation in Large Blood Vessels Based on Fluid-

Solid Interactions Methods 460
Tomohiro Fukui, Kim H. Parker and Takami Yamaguchi
18. An-Harmonic Modeling of the Peripheral Distortion of the

Arterial Pulse 472
Panagiotes A. Voltairas, D.I. Fotiadis, A. Charalambopoulos and
L.K. Michalis
19. Numerical Analysis of Blood Flow in Stenosed Channels 489

Stéphanie Ferreira, Ricardo P. Dias, Carlos Balsa and Carla S.
Fernandes
20. Blood Flow Behavior in Microchannels: Past, Current and Future

Trends 513
R. Lima, T. Ishikawa, Y. Imai and T. Yamaguchi
21. A Survey of Microchannel Geometries for Mixing of Species in

Biomicrofluidics 548
Francesco Pennella, Francesco Mastrangelo, Diego Gallo, Diana
Massai, Marco A. Deriu, Giuseppe Falvo D'Urso Labate, Cristina
Bignardi, Franco Montevecchi and Umberto Morbiducci
22. Endothelial Cell Responses to Fluid Shear Stress: From

Methodology to Applications 579
Toshiro Ohashi and Masaaki Sato
23. Micro-Flow Visualization of Magnetic Nanoparticles for
Biomedical Applications 600
R. Lima, R.J. Joseyphus, T. Ishikawa, Y. Imai and T. Yamaguchi
24. A Computational Study on the Possibility of the Initialization And

Development of Intracranial Aneurysms Considering Biofluid and
Biosolid Mechanics 613
Yixiang Feng, Shigeo Wada and Takami Yamaguchi
Index 634
i
FOREWORD
I am delighted to write the Foreword for this book, which captures the recent advances
in both analytical and experimental techniques of describing accurately the single and
multiphase flow phenomena in chemical engineering practice and biomedical systems.
In the chemical engineering arena, there is an interesting combination of chapters

covering both fundamental and other applied studies. Fundamental studies focus on a
wide variety of topics such as simulation of rarefied gas flow in porous media, mass
transfer for soluble solids in a packed bed, tree shaped flow structures and constructal
theory, interfacial area modeling in two-phase flows, etc. Other studies address the
environmental topics of NOx abatement by selective catalytic reduction of NO by
ammonia, and modeling of waste water treatment by combining CFD simulation with
activated sludge models. Other CFD studies on chemical engineering side include
super critical hydrogenation of edible oil and simulation of a viscoelastic fluid.
Important area of mixing is covered by two articles, one dealing with the evolution of
mixing science and its impact on chemical engineering, and the other dealing with a
specific topic of turbulent mixing with chemical reaction. There are some interesting
experimental studies reporting work on bubble column and packed beds.
In the biomedical arena a number of chapters cover many fascinating articles

related to blood flow, ranging from pulse wave propagation in blood vessels to
numerical simulation of blood flow in stenosed channels. Other chapters in this
area which describe application of microfluidics to biological systems and
biomedical applications are equally fascinating.
I would like to congratulate the editors for bringing these separate but
complementary chapters together in a book form. It is also nice to see many multi-
disciplinary teams from countries across the globe working together to produce these
chapters. I hope that the reader will benefit from the breadth and depth of the range
of chemical engineering and biomedical topics covered in these chapters.
Dr. Vishwas V. Wadekar

Technology Director, HTFS Research
AspenTech Ltd, Reading
United Kingdom
ii
PREFACE
Single or two-phase flows are ubiquitous in most natural process and engineering
systems. Examples of systems or process include packed bed reactors, either
single phase or multiphase, absorber and adsorber separation columns, filter beds,
plate heat exchangers, flow of viscoelastic fluids in polymer systems, or the
enhanced recovery of oil, among others.
In each case the flow plays a central role in determining the system or process
behaviour and performance. A better understanding of the underlying physical
phenomena and the ability to describe is crucial to design, operate and control
processes involving the flow of fluids, ensuring that they will be more efficient
and cost effective.
Growing areas such as microfluidics, nanomedicine and the modelling and simulation
of complex flow in living systems such as the blood flow in microvascular networks
rely upon a good description of the flow. One way of studying the blood flow
behavior is in the context of blood flow in large arteries, and another is in small
vessels. In the former case, blood may be treated as a homogenous fluid and
Newtonian constitutive equations are generally accepted as a good approximation to
express the rheological property of blood. In small vessels, however, the scale of
generated flow field sometimes becomes comparable to the scale of a blood cell. The
chapters from the area of biomedical engineering will describe several physiological
and pathological events that happen in both large and small vessels.
Recent advances either in computational and experimental techniques are

improving the existing knowledge of single and multiphase flows in engineering
and physical systems of interest. This book reviews the state of the art and recent
advances in various key areas of fluid mechanics and transport phenomena in the
fields of chemical and biomedical engineering.
Ricardo Dias
CEFT - Transport Phenomena Research Center
iii
List of Contributors
Jacinto Sá
School of Chemistry and Chemical Engineering, David Keir Building, Queen’s
University Belfast, Stranmillis Road, Belfast, BT9 5AG, Northern Ireland, UK.
Cristina-Elena Stere
Alexandre Goguet
Emilio Muñoz
Department of Chemical Engineering and Environmental Engineering, University
of Oviedo, C/Julián Clavería 8, 33006-Oviedo, Spain.
David Lesser
Institut für Chemische Verfahrenstechnik, Technische Universität Clausthal,
Leibnizstr. 17, 38678 Clausthal, Germany.
Pablo Marín
Salvador Ordóñez
Fernando V. Díez
Ricardo Dias, Antonio A. Martins, Rui Lima and Teresa M. Mata (Eds)
All rights reserved-© 2012 Bentham Science Publishers
iv
Alexandros N. Kalarakis
Institute of Chemical Engineering and High Temperature Chemical Processes,
Foundation for Research and Technology, Hellas, Greece.
Eugene D. Skouras
Vasilis N. Burganos
Ricardo J. Santos
Laboratory of Separation and Reaction Engineering, Departamento de Engenharia
Química, Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto Frias,
4200-465 Porto, Portugal.
Madalena M. Dias
José Carlos B. Lopes

Andrei Chorny
A.V. Luikov Heat and Mass Transfer Institute, 15 P.Brovka Str, Minsk, 220072,
Belarus.
J.P. Pereira
LSRE-Laboratory of Separation and Reaction Engineering, Faculdade de
Engenharia da Universidade do Porto, Portugal.
v
A. Karpinska
P.J. Gomes
A.A. Martins
A. Guardo
Fluid Mechanics Department, Universitat Politècnica de Catalunya, Av. Diagonal
647, ETSEIB, 08028, Barcelona, Spain.
E. Ramírez
Chemical Engineering Department, Faculty of Chemistry, Universitat de
Barcelona, Martí i Franquès 1, 08028, Barcelona, Spain.
M.A. Larrayoz
Chemical Engineering Department, Universitat Politècnica de Catalunya, Av.
Diagonal 647, ETSEIB, 08028, Barcelona, Spain.
F. Recasens
Chemical Engineering Department, Universitat Politècnica de Catalunya, Av.
Diagonal 647, ETSEIB, 08028, Barcelona, Spain.
J.M.P.Q. Delgado
LFC-Laboratório de Física das Construções, Departamento de Engenharia Civil,
Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto Frias, s/n,
xii
ACKNOWLEDGEMENTS
The editors acknowledge all the authors for their contributions and express our
sincere appreciation to the assistance of all parts involved in the preparation of the
book. We also acknowledge the support from the following grants: Grant-in-Aid
for Science and Technology (PTDC/SAU-BEB/108728/2008, PTDC/SAU-
BEB/105650/2008 and PTDC/EME-MFE/099109/2008) from the Science and
Technology Foundation (FCT) and COMPETE, Portugal.
Finally, we give our special thanks and gratitude to the editorial assistants,
Tomoko Yaginuma and Carla Fernandes, who critically read the text, identified
mistakes and omissions and helped us throughout the preparation of the present
book.
Part I: Single Phase Flows on Chemical Enginnering
Single and Two-Phase Flows on Chemical and Biomedical Engineering, 2012, 3-25 3
CHAPTER 1
Spacims-Probing the Internal Behaviour of 3D Structured
Materials
Jacinto Sá*, Cristina-Elena Stere and Alexandre Goguet*

University Belfast, Stranmillis Road, Belfast, BT9 5AG, Northern Ireland, UK
Abstract: 3D structured materials, especially monoliths have been used for decades in
the removal of pollutants in automotive and stationary stations, bulk chemical
production and so on. Their applications are expected to rise due to the ever growing
concern of global warming. Despite the clear advantages of using materials with a well
defined 3D structure, their implementation is often hindered by the absence of detailed
information of what happens inside the channels. SpaciMS is a minimum invasive
spatially resolved capillary-inlet mass spectroscopy system, often combined with other
analytical probes such as thin thermocouples. The probes can be positioned at multiple
axial and/or radial locations within the working monolith, enabling for the generation of
detailed spatio-temporal maps of the reactions and breakthrough fronts. This antagonist
approach from the conventional ‘end-pipe’ analysis which can be often misleading to
describe the internal behaviour of structured materials, offers the possibility to test the
validity of a pre-conceived kinetic and/or hydrodynamic model instead of the common
occurrence of fitting a model from ‘end-pipe’ measurements.
Keywords: SpaciMS, monolith, mass spectroscopy, catalyst, kinetic oscillations,

temporal resolution, spatial resolution, high sensitivity, operando, flux, model.
INTRODUCTION
Spatially structured systems are easily found in nature, however in Catalysis and
Chemical Reaction Engineering poorly structured catalytic systems have long
been the dominant technology. The driving force for the development and
implantation of structured catalytic solutions are new paradigms in the field, such
as minimization of energy consumption and of waste formation, which are related
to the ever growing concern of global warming. In fact it is increasingly important
*Address correspondence to Jacinto Sá and Alexandre Goguet: School of Chemistry and Chemical
Engineering, David Keir Building, Queen’s University Belfast, Stranmillis Road, Belfast, BT9 5AG,
Northern Ireland, UK; E-mails: jacinto.sa@psi.ch; a.goguet@qub.ac.uk
4 STP Flows on Chemical and Biomedical Engineering Jacinto Sá et al.
to drastically reduce the environmental impact of human activities, so that

sustainability can be achieved. Catalysis is a major player in the future green
economy; and structured catalysts and reactors have a large potential in reaching
the desired sustainability objectives.
Three dimensional (3D) structured materials have been used for decades in the
removal of pollutants in automotive and stationary applications, bulk chemical
production, just to mention a few [1-4]. Many structured systems, especially
monoliths, currently applied or considered for application offer high degree of
freedom, high geometric surface area, low-pressure drop and good mass transfer
performance. There may also bring advantages in kinetics and mass transfer in
gas/liquid phase applications due to better catalyst wetting, higher effectiveness
factors, and thin films that are beneficial in terms of high mass transfer rates [1].
Despite the clear advantages of using materials with a well defined 3D structures,
their implementation is often hindered by the absence of detailed information
regarding spatial chemical mapping, hydrodynamics and flow distribution of
multi-phase flow inside the channels, i.e., reactor bed (Fig. 1) [1-4].
Unknown gas
phase composition
Unknown gas
phase composition
Figure 1: Sketch of a common catalytic converter.
The SpaciMS is a minimally invasive spatially resolved capillary-inlet mass

spectrometry system, often combined with other analytical probes such as thin
Spacims-Probing the Internal Behaviour STP Flows on Chemical and Biomedical Engineering 5
thermocouples [1-6], which enables the determination of chemical composition of

the reactants and products and hydrodynamic data inside the monolith channels.
The development of the SpaciMS began in 1997 in a response to the need to
quantify the fast transient emissions associated with the development and
operation of Lean NOx Trap (LNT, i.e., NOx Storage and Reduction: NSR)
catalysts. The probes can be positioned at multiple axial and/or radial locations
within the working monolith, enabling for the generation of detailed spatio-
temporal maps of the reactions and breakthrough fronts.
Figure 2: Conventional approach (gold) versus the approach created with the development of
SpaciMS (blue).
The SpaciMS approach compared to the conventional ‘end-pipe’ analysis strategy

(often misleading to describe the internal behaviour of material), offers the
possibility to validate pre-conceived kinetic models based on the real intra-
catalyst reaction maps instead of the common approach where postulated models
are fitted to the limited “end-pipe” data (Fig. 2). Clearly the SpaciMS approach
allows developing models derived with a higher level of accuracy and details due
to the high data volume and precision.
THE SPACIMS SET-UP
Figs. 3 and 4 report a sketch of the QUB (Queen’s University Belfast) SpaciMS
set-up and photographs of the experimental setup, respectively.
The equipment is subdivided in 5 parts:
 A mass spectrometer (MS),

26 Single and Two-Phase Flows on Chemical and Biomedical Engineering, 2012, 26-51
CHAPTER 2
Modelling of a Monolithic Reverse Flow Reactor for Selective
Catalytic Reduction of NO by Ammonia
Emilio Muñoz1,*, David Lesser2, Pablo Marín1, Salvador Ordóñez1 and
Fernando V. Díez1
1
Department of Chemical Engineering and Environmental Engineering,
University of Oviedo, C/Julián Clavería 8, 33006-Oviedo, Spain and 2Institut für
Chemische Verfahrenstechnik, Technische Universität Clausthal, Leibnizstr. 17,
38678 Clausthal, Germany
Abstract: One of de most important environmental problems nowadays is the abatement

of NOx emissions. NOx are involved in photochemical reactions that produce smog and
acid rain, and (together with CO2, CH4 and H2O), take part in the greenhouse effect.
One of the most efficient ways to reduce NOx emissions is Selective Catalytic
Reduction (SCR). In SCR NOx are reduced to N2 and H2O by reaction with ammonia or
urea. Ammonia is most commonly used and is the reactant considered in this study.
Conventionally, this reaction is performed in steady fixed bed reactors, but there are
other reactor types that can be used, including those operating in dynamic conditions,
such as Reverse Flow Reactors (RFR). RFR consist of a fixed bed catalytic reactor in
which the direction of the inlet flow is reversed periodically. Among other advantages,
the periodic change in the inlet flow allows autothermal operation even for weakly
exothermal reactions. This study is devoted to the modeling a monolithic RFR used for
NH3-SCR. An unsteady one-dimensional heterogeneous model has been developed for
simulating the reactor behaviour. The model is formed by differential equations
corresponding to conservation equations, applied separately to the gas and solid phases,
and algebraic equations used for estimating the physical and transport properties. The
model considers internal and external mass and heat transfer resistances, and axial
dispersion. Experimental validation of the model allows its use for the optimization of
the most important variables that affect the process.
Keywords: Reverse flow reactor (RFR), selective catalytic reduction (SCR),

dynamic modeling, NOx abatement, pilot scale reactor, autothermal operation,
transient respond method, non-stationary reactors, structured catalysts, greenhouse
gases abatement.
*Address correspondence to Emilio Muñoz: Department of Chemical Engineering and Environmental

Engineering, University of Oviedo, C/Julián Clavería 8, 33006-Oviedo, Spain; E-mail:
emilio.emv@gmail.com
Modelling of a Monolithic Reverse Flow STP Flows on Chemical and Biomedical Engineering 27
INTRODUCTION
During the last 50 years, both traffic (including sea, air and road transport) and
industrial activities (i.e., power plants or dairy industries) have grown rapidly. As
a result, the use of fossil fuels and biomasses combustions, and the concentration
of anthropogenic pollutants in the atmosphere have increased.
Nitric oxides (NOx), which include nitrogen oxide (NO), nitrogen dioxide (NO2)
and nitrous oxide (N2O), are nowadays considered as one of the most dangerous
air pollutants. NOx are known for their contribution to the greenhouse effect,
besides participating in photochemical reactions that cause acid rain and the
formation of troposphere ozone (“smog”). Moreover, they have an important role
in lakes and rivers eutrophication. Regarding human health, NOx can damage the
respiratory system.
The abatement of NOx has been studied widely during the last decades [1-4] and
their emissions in the industrialized countries is regulated by restrictive laws, i.e.,
in Europe the Directive 2001/81/EC, establishes national emission ceilings for
several pollutants (an extract for the limits is shown in Table 1).
Table 1: National emission ceilings for SO2, NOx and NH3 to be attained by 2010 (Extract)
Country NOx [Kilotonnes] NH3 [Kilotonnes] SO2 [Kilotonnes]

Austria 103 66 39
Belgium 176 74 99
France 810 780 375
Germany 1051 550 520
Italy 990 419 475
Spain 847 353 746
Sweden 148 57 67
United Kingdom 1167 297 585
EC 15 (European Countries 15) 6519 3110 3850
The European Environmental Agency, in the status report up to 2008

(Nº11/2009), indicates that the aggregate of EU-27 projections for NOx emissions
28 STP Flows on Chemical and Biomedical Engineering Emilio Muñoz et al.
will be 6% above the Directive ceilings. The comparison between the projections
and the ceilings for every state member is shown in Fig. 1.
Figure 1: Comparison of NOx projections and ceilings in 2010.
The three major techniques used for the abatement of the nitric oxides are lean
NOx traps (LNT), ammonia-or urea-Selective Catalytic Reduction and
Hydrocarbons Selective Catalytic Reduction (HC-SCR).
LNT consists basically of an alumina support on which a metal oxide (e.g., Ba)
and a noble metal (e.g., Pt) are deposited. By operating in cyclic conditions, NOx
are stored in the lean step, whereas NOx are oxidized in the rich step. This
technique is well established for mobile sources. NH3-SCR is a well established
technique, has been mainly used for the elimination of NOx in stationary sources,
due to the ability of ammonia to react selectively with NOx forming N2, together
with the resistance to poisoning and aging of the catalysts used.
NH3-SCR Reaction
NH3-SCR is based on the reaction of NH3 with NO to form N2 and H2O:
4 NH3 + 4 NO + O2 → 4 N2 + 6 H2O (1)

CHAPTER 3
Mesoscopic Simulation of Rarefied Gas Flow in Porous Media
Alexandros N. Kalarakis, Eugene D. Skouras and Vasilis N. Burganos*

Foundation for Research and Technology, Hellas, Greece
Abstract: The accurate description of flow in nano-scale pores or channels is very

important for the reliable design of materials and processes in the areas of MEMS,
mesoporous media, and vacuum technologies. Use of classical flow equations fails in
this regime since the continuum assumption is not valid. This is due to the fact that the
mean free path is comparable to the characteristic dimensions of the system, and
rarefaction effects dominate the process. Such a difficulty arises notably in the
intermediate Knudsen number regime (Kn=0.1 to 10), commonly referred to as the
“transition” flow regime. To remedy this, slip flow conditions have been adopted in the
literature, following the simple first-order approach of the velocity near the walls given
by Maxwell, and extended to higher-order treatments. Alternatively, direct deterministic
or stochastic atomistic and mesoscopic techniques have been employed for the flow
description, which solve the Boltzmann or the Burnett equations and use kinetic theory
approaches pertinent to this flow regime. A description of recent advances in simulation
techniques, namely, the “continuum” slip approaches, and some direct mesoscopic
techniques are presented in this chapter. Illustrative simulation results of permeability
and viscosity coefficients in mesoporous media using the DSMC and LB methods are
also given, followed by comparisons with classical continuum formulations.
Keywords: Rarefied flow, porous media, transition regime, direct simulation

monte carlo, lattice-boltzmann, reconstruction, fractional brownian motion, slip
flow, nanoscale pores, knudsen number, mesoscopic methods.
INTRODUCTION
Recent advances in microtechnology and, in particular, in microelectromechanical

systems (MEMS) and nano porous media have necessitated the elucidation of
flow and transport processes in small dimensions. This is also the case with
several other industrial applications, which rely on low-pressure conditions, or
*Address correspondence to Vasilis N. Burganos: Institute of Chemical Engineering and High

Temperature Chemical Processes, Foundation for Research and Technology, Hellas, Greece; E-mail:
vbur@iceht.forth.gr
Mesoscopic Simulation of Rarefied Gas STP Flows on Chemical and Biomedical Engineering 53
involve features of small characteristic size, including vacuum processes, high-

altitude and space applications, ultrasonic gas flows, membrane distillation, fuel
cells, etc. Flows in such systems require special consideration due to the facts that
the continuum assumption is not applicable and phenomenological equations
cannot be applied safely.
Typically, the distinction among the various flow regimes is based on the value of
the Knudsen number, Kn, which is defined as the ratio of the mean free path,  ,
of the fluid molecules to a characteristic length, L [1-5]. It is common ground
that the continuum limit is attained for Kn values less than 0.001, whereas for
0.001<Kn< 0.1 some slip flow boundary condition is usually employed in
continuum descriptions and the flow regime is commonly termed “slip flow”
regime. For higher Knudsen numbers the continuity assumption breaks down and
the flow can no longer be described by the traditional transport equations (Navier-
Stokes) even with the aid of a slip boundary condition at the fluid-solid interface
[2]. More specifically, Kn values between 0.1 and 10 correspond to “transition
flow”, whereas for Kn values greater than 10 the fluid is considered to be in a
“free molecule flow” state (collisionless transport).
Even though it is widely used, such a classification oversimplifies the underlying

phenomena during flow and, thus, special attention should be paid to the
peculiarities of the various flow regimes and the validity of the continuum
approaches, such as the Navier-Stokes equations with or without appropriate slip
boundary conditions. According to Gad-el-Hak [6], there are three fundamental
conditions that must be satisfied in order for the Navier-Stokes equation to be
valid: (i) an underlying Newtonian framework of mechanics, (ii) continuum
conditions, i.e., statistical invariance of local intensive field properties (such as
velocity and temperature) with the size of the elementary volume, and (iii) local
thermodynamic or quasi-equilibrium that permits linear relation between stress
and rate of strain. Although the first condition is almost always guaranteed in non-
relativistic systems [6, 7], the other two limit the applicability of the continuum
approach in rarefied conditions. More specifically, the continuum approach breaks
down when the spatial scales of the flows are comparable to the mean distance
between molecules, whereas the thermodynamic quasi-equilibrium breaks down
when time is not sufficient for equilibrium to be attained, or, equivalently, when
54 STP Flows on Chemical and Biomedical Engineering Kalarakis et al.
the molecular time and length scales are not sufficiently small compared to the
characteristic macroscopic flow scales.
Therefore, it is essential to define additional parameters in order to classify the flow

regimes, as well as the fluid state. Gases are considered dilute when the mean
molecular spacing,  , is at least 7 times greater than the mean molecular diameter,

 0 , [8] that is, ~ 7 [6]. On the contrary, if the aforementioned ratio is smaller
 0
than ~7, the gas is considered as dense. Alternatively, the ratio of a macroscopic
L
characteristic flow dimension scale, L , to the gas molecule distance,  , that is, ,

can be used to bound the continuum approach. More specifically, in order for the
molecular chaos to be ensured and the molecular fluctuations to be statistically
insignificant, this ratio has to be greater than 100, which means that at least 106
particles reside within the elementary fluid volume. The various flow regimes and
the corresponding governing equations are summarized in Fig. 1 (see also [1]).

Navier-Stokes Equation Burnett
Euler No-slip Slip Equation
 Boltzmann Equation 
Continuum Flow Slip Flow Transition Flow Free-molecule

Flow
Kn 10-3 10-2 10-1 100 10+1
Figure 1: Flow regimes and governing equations.
In the context of this work a description of recent advances in simulation

techniques, namely, the “continuum” slip approaches [1, 9], and direct
mesoscopic techniques, such as the Direct Simulation Monte Carlo (DSMC)
method [8, 10-12], the Information Preservation (IP) method [13-16], and the
Lattice Bolzmann (LB) method [17, 18], are presented. Illustrative simulation
results of permeability and viscosity coefficients in mesoporous media using the
CHAPTER 4
Mixing Through Half a Century of Chemical Engineering
Ricardo J. Santos, Madalena M. Dias and José Carlos B. Lopes*
Laboratory of Separation and Reaction Engineering, Departamento de

Engenharia Química, Faculdade de Engenharia da Universidade do Porto, Rua Dr.
Roberto Frias, 4200-465 Porto, Portugal
Abstract: The first mixing studies on chemical reactor engineering date from
approximately fifty years ago. There are several reviews on mixing that are generally
focused on a specific topic, such as chaotic mixing. This chapter presents an overall
picture of the evolution on mixing science and its impact on chemical reactor
engineering. The works reviewed in this chapter are single-phase flows although some
of the concepts presented in this chapter are also applied to multiphase reactors.
Keywords: Mixing, micromixing, macromixing, residence time distribution,

micromixing test reaction, chaotic mixing, turbulent mixing, micromixing models,
lamellar model, CFD.
INTRODUCTION
This chapter reviews mixing over the last fifty years of chemical engineering and
it is an addition to several short reviews on this subject that can be found among
the lectures and papers of some of the most prominent researchers in this field [1-
5]. A more complete review can be found in [6].
Up to the early 50s the literature published on mixing was scarce [1]. The turning
point occurred in 1952 when Danckwerts made a first approach to non-ideal
reactors by introducing the concept of Residence Time Distribution [7], RTD, and
defining two measures for fluid non-homogeneities [8]. In addition to the
measurement of mixing, either by tracer or chemical reaction, some approaches to
model the mixing mechanism have been tried, and in this review, the proposed
models will be separated into phenomenological models that do not take into
account the fluid mechanics of the process, and mechanistic models [5].
*Address correspondence to José Carlos B. Lopes: Laboratory of Separation and Reaction Engineering,
Departamento de Engenharia Química, Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto
Frias, 4200-465 Porto, Portugal; E-mail: lopes@fe.up.pt
80 STP Flows on Chemical and Biomedical Engineering Santos et al.
The evolution of computational resources has also allowed some model free
simulations of mixing [9, 10] by numeric resolution of the equations that govern
the flow and mass transfer. The huge computational resources required for model
free simulations of mixing most often requires the introduction of some degree of
modelling of the flow field and for mass transfer mechanisms, in the resolution of
real and complex problems. This type of approach to mixing is nowadays very
common and has been encouraged by the wide spread of commercial
Computational Fluid Dynamics, CFD, codes to solve the governing equations of
flow and mass transfer. The last sections of this chapter are dedicated to the
review of mixing and flow field simulation by CFD techniques.
Residence Time Distribution

The first approach to the mixing problem was the analysis of the different periods
of residence of clumps of fluid, or individual fluid elements, inside a vessel. The
heterogeneity of the reactor composition, regarding the age of fluid elements, at a
macroscopic scale is classified as macromixing. The macroscale heterogeneities
are assessed by the Residence Time Distribution, RTD [7], which accounts for the
presence of backmixing, shortcuts and dead zones inside the vessel. The RTD,
E  t   , where t  is the residence time or age, can be obtained from a step tracer
experiment, from the fraction of tracer flux in the outflow, F  t   [7],
dF  t  
E t    (1)
dt 
Another important function is the distribution of ages of material in the system,

I  t   , which can de obtained from [7],
q
I t   
V
1  F  t    (2)
where q is the inlet flow rate and V the reactor volume.
The RTD can be applied to the design of steady-state reactors with first order
reactions [7]. For reactions of other orders, two reactors with the same RTD can
be associated to different conversions depending on factors such as the earliness
Mixing Through Half a Century of Chemical STP Flows on Chemical and Biomedical Engineering 81
of mixing. A simple example where the same RTD yields different final
conversions, is the case of a plug flow reactor, PFR, in series with a continuous
stirred tank reactor, CSTR, where alternating the order of the reactors keeps the
RTD constant but changes the final conversion of the system [11].
The RTD can be used for the prediction of the reactor yield for a state of complete
segregation, which assumes that each clump of fluid at the inlet with the same life
expectancy is a batch reactor completely segregated from the other clumps and
that mixing of clumps of different ages occurs only at the outlet. In 1959
Zwietering [11] introduced the concept of maximum mixedness, which assumes
that mixing occurs as soon as possible. The model of Zwietering consisted of a
PFR with: a common inlet and multiple outlets for the case of complete
segregation (see Fig. 1a); and multiple entrances and a common outlet for the case
of maximum mixedness (see Fig. 1b). For the same RTD the maximum
mixedness and complete segregation models establish the upper and lower
conversion bounds.
Figure 1: Scheme proposed by Zwietering (1959) for a) complete segregation, b) maximum

mixedness.
In 1967 Weinstein and Adler [12] introduced a mixing parameter that consists on
the division of the reactor into the two situations described by Zwietering [11]:
fluid with lower residence times is described by the complete segregation model;
and fluid of higher residence times is described by the maximum mixedness
model. Weinstein and Adler [12] used two models to divide the fluid according to
its different life expectancies, one with consecutive reactors and the other with
parallel reactors.
Nowadays the RTD is the most common approach to model the non-idealities of
chemical reactors as can be seen from the most popular reference books of
chemical reactors courses. The RTD is also one of the tools for mixing studies in
chemical reactors, as can be seen from a number of recent papers on topics such
CHAPTER 5
Application of the Probability Density Function Method to
Turbulent Mixing with Chemical Reaction
Andrei Chorny*
A.V. Luikov Heat and Mass Transfer Institute, 15 P.Brovka Str, Minsk, 220072,
Belarus
Abstract: In studies of a turbulent flow with mixing-sensitive chemical reactions, the

equations for the PDF of scalars (temperature and mixture component concentrations) are
applied in combination with the conventional turbulence models that contain the equations for
statistical moments. The PDF method advantage is an accurate representation of the
chemistry influence in model equations. However, to calculate correlations responsible for an
averaged chemical reaction rate, one needs a more thorough description of micromixing. As
micromixing is governed by the small-scale flow structure, the latter can be considered
statistically homogeneous. So, the well-developed homogeneous turbulence theory is used for
closing the micromixing models. Just the adequate description of micromixing connected not
only with the successful modeling of the chemistry influence, but first with a reasonable
analysis of the entire mixing process, remains a stumbling block for the PDF method. This
problem has received particular attention in the present work where the problem on mixing
with chemical reacting in the homogeneous turbulent flow is stated by means of both the
method of statistical moments for turbulence parameters and the mixture fraction PDF
method for closing unknown correlations responsible for chemical reacting. Several known
micromixing models (LSME/IEM, Langevin, multi-zone PDF) applicable in the PDF transfer
equation are compared to explain their influence on an averaged reaction rate to be calculated.
Keywords: Turbulent flow, mixing-sensitive chemical reaction, micromixing,

statistical moment, probability density function, reynolds number, damköhler
number, mixture fraction, progress variable.
To my parents and teachers
INTRODUCTION
To progress in new technological processes and engineering facilities one needs a
more detailed description of the turbulent flow properties with regard not only to
averaged values of velocity, temperature or concentration, but also to fluctuation
*Address correspondence to Andrei Chorny: A.V. Luikov Heat and Mass Transfer Institute, 15 P.Brovka
Str, Minsk, 220072, Belarus; E-mail: anchor@hmti.ac.by
114 STP Flows on Chemical and Biomedical Engineering Andrei Chorny
characteristics of considered parameters. Problems of assembling process flow

diagrams for operation of different heat/mass transfer apparatus are the examples
of technical gaps imposing heavy demands on turbulent transfer theory. Demands
to enhance more complete combustion of fuel, to decrease an amount of pollutants
and also possibilities to use alternative fuel resources generate a need for
combustion optimization in engine combustion chambers, furnaces, power plants,
and chemical reactors, and this specifies their energy and ecological properties. In
environmental protection problems it is important to correctly forecast the
turbulent flow behavior with objectives to evaluate consequences of natural and
artificial pollutions of air and water basins, as well as those of smog formation
due to the activity of industrial enterprises.
The problems mentioned above are being solved by obtaining and collecting
detailed data on occurring physical phenomena. In doing so, the first estimation of
the processes of fluid-and thermodynamics is usually made by mathematical
modeling procedures.
Turbulent transfer and mixing studies are based on extensive experimental

experience and theoretical approaches invoking a broad spectrum of mathematical
apparatus-from simple algebraic relations for averaged turbulent flow parameters
to probability methods. In the context of the concept reflecting the modern state-
of-art-of this area of knowledge the methods and notions of classical turbulence
theory [1] are combined with the ideas of statistical fluid mechanics and
computational mathematics [2-4].
A mathematical description of turbulent reacting flows starts from the equations of

continuity, Navier-Stokes, energy conservation and conservation of mixture
components, as well as from thermodynamic and kinematic relations. Solving such a
system with appropriate boundary conditions yields total information on all space-
time distributions of velocity, pressure, temperature, density, and chemical
composition. Yet practically, this is difficult to realize because of a wide range of
turbulence length and time scales. Chemical reaction determines scales associated
with the kinetics of elementary reactions (their spatial scale is usually less than the
Kolmogorov length scale). This circumstance imposes additional limits on a
computational grid resolution. Up to now, developing in connection with computer
Application of the Probability Density STP Flows on Chemical and Biomedical Engineering 115
technology advances, direct numerical simulation (DNS) [5] and large eddy
simulation (LES) [6] of complex turbulent flows cannot be used in full measure first
of all for chemical engineering problems [3, 4]. The methods mentioned here find
use mainly with research objectives for simple geometrical flow configurations at
moderate Reynolds number Re. So, the statistical methods are under development
including three main aspects of modeling: chemical kinetics, mutual influence of
flow dynamics and chemical processes, and also directly turbulence [4].
A necessity to model chemical kinetics is caused by a scarcity of CPU time. An

account of a detailed mechanism of reactions is justified for chemical systems
with a small number of reactants. Otherwise, assumptions should be considered
which allow simplifying the description of such systems: a local chemical
equilibrium assumption should be introduced for fast reactions; a state of a
reacting medium should be characterized using special parameters-mixture
fraction, Shvab-Zeldovich variables, progress variable, etc. [3].
Of late, the mathematical apparatus of turbulence theory has been developed to

such a level, at which it is in principle possible to state and solve practically
important turbulent mixing problems with the use of the equations of fluid-and
thermodynamics [2-4]. In this situation, dynamic parameters (velocity, pressure,
density, temperature, concentration, etc.) are regarded as random space and time
functions. Therefore, turbulence studies use statistical characteristics of random
fluid-and thermodynamic fields. Depending on the parameters chosen, different
approaches for studying turbulent transfer [7] are known and are cited below,
since the completeness of a statistical description of random fields diminishes:
 functional approach using a probability functional of a random field [1];
 approach based on a finite-dimensional probability density function

(PDF) [3, 4, 8, 9];
 approach considering energy spectra as a function of wave number or

frequency [1, 2];
 correlation approach dealing with the simplest characteristics of a random

field-one-point correlation functions and statistical moments [2-4, 7].
CHAPTER 6
Activated Sludge Models Coupled to CFD Simulations
Pereira J.P., Karpinska A., Gomes P.J., Martins A.A., Dias M.M., Lopes
J.C.B. and Santos R.J.*

Engenharia da Universidade do Porto, Portugal
Abstract: Wastewater treatment modelling combining hydrodynamics, mass transfer and kinetic
issues remains one of the major goals of chemical engineering. The reactor design is frequently
based on Activated Sludge Models (ASM) that describe the biological processes in an Activated
Sludge (AS) process. The ASM are generally implemented on ideal models of chemical reactors
that do not account for the actual hydrodynamics of the reactor. Nowadays, it is possible to obtain
hydrodynamic information of an AS tank from computational fluid dynamic (CFD) simulations
that provide information about the mass transfer in the flow field of the AS tank and enable
determining the residence time distribution (RTD). This chapter focuses on the use of CFD data
coupled to the state-of-the-art ASM1, and resumes the work developed by many researchers on
this issue. The influence of the RTD on the results of the simulations using ASM1, is analyzed
from the simulations of the ASM1 using various reactor models. A procedure is proposed for the
coupling of the CFD simulations information with the ASM1 biological model.
Keywords: Activated sludge, oxidation ditch, macromixing, residence time

distribution, activated sludge models, CFD, large eddy simulation, averaged
reynolds navier-stokes equations, turbulence models, compartmental approach.
INTRODUCTION
One of the major challenges in the wastewater treatment industry is to assure that
the treatment process meets the outflow quality requirements, while both
investment and operating costs are kept as low as possible [1]. Although the
design of wastewater treatment plants (WWTP) is a key step to assure the goals of
the process, the design is usually based on general guidelines and designers’
experience. The wastewater treatment modelling combining hydrodynamics, mass
transfer and the biochemical reactions kinetics remains one of the major goals of
chemical engineering [2, 3].
*Address correspondence to Santos, R.J.: LSRE-Laboratory of Separation and Reaction Engineering,

Faculdade de Engenharia da Universidade do Porto, Portugal; E-mail: rsantos@fe.up.pt
154 STP Flows on Chemical and Biomedical Engineering Pereira et al.
The biological reactions that remove organic pollutants from the wastewater are
performed in an aerated reactor that as suspended flocks of active biomass that
use the organic matter in the wastewater for metabolic activities and cellular
growth. The biomass flocks suspended in the water in the aerated reactor form a
sludge that is called the mixed liquor. The biomass flocks are separated from the
water in a settler where the settled portion forms a concentrated sludge that is
recirculated into the aerated reactor. This biological process for wastewater
treatment is called Activated Sludge (AS) process. In many WWTPs the AS
process is the step where the non-settable pollutants are removed. The AS
processes have many implementations, a particular one is the oxidation ditch; a
scheme of an oxidation ditch is shown in Fig. 1. The tank with two channels in
Fig. 1 is the AS biological reactor where the mixed liquor is circulated and
aerated, the circular tank is the settler.
This chapter reviews the implementation of the Activated Sludge Models (ASM)
on chemical reactor models and assesses the effect of hydrodynamic on the results
of the AS processes simulation using the case study of an oxidation ditch. A
procedure is proposed for the assessment of the hydrodynamics in AS processes,
based on Computational Fluid Dynamics Simulations (CFD) using the same case
study: the oxidation ditch. A procedure to couple the actual AS process
hydrodynamics into the simulation of the ASM is proposed and the influence of
the reactor model in the ASM simulation results is assessed.
Figure 1: Scheme of an AS process: the oxidation ditch.

Activated Sludge Models Coupled to CFD STP Flows on Chemical and Biomedical Engineering 155
BIOLOGICAL MODEL
The removal of biodegradable organics by microorganisms from wastewaters is

usually performed by activated sludge (AS) processes in aerated reactors through
complex physical, biological and chemical interactions between the process
variables. The AS models have become more complex over the years; several
mechanistic mathematical models describing the biological processes have been
introduced. Mathematical modelling is considered to be a very important tool in
AS processes since the lab-scale tests are quite time consuming: for example the
test of each operational condition requires the lab reactor to be run for several
days. The International Association on Water Research and Control (IAWPRC)
established a task group on activated sludge models that introduced the Activated
Sludge Model No. 1 (ASM1) [4].
The ASM1 soon became the most widely used and accepted model of the AS
process. The generalized usage of the ASM1 is due to its structure that is based on
a matrix model that connects all the components and rate expressions, based on
growth and substrate utilization. Furthermore, another important feature of this
model is that chemical oxygen demand (COD) is used as a common measure of
organic substrate and biomass allowing a total COD balance [5]. The fundamental
processes assumed by the model are the growth and decay of biomass, the
ammonification of organic nitrogen, and the hydrolysis of particulate organics.
The ASM1 models consider twelve components of an AS process that are briefly
described in Table 1; these twelve components are involved in eight kinetic
processes that are also considered in the ASM1 and are summarized in Table 2.
Among the ASM1 components is the biomass that is divided into autotrophic,
heterotrophic, and inert fractions. The heterotrophic biomass growth is based on
the consumption of both soluble biodegradable substrate and oxygen, in the
aerobic process, or merely based on nitrate in the anoxic process. The autotrophic
biomass consumes ammonium and oxygen, generating nitrate during the growth.
Both biomasses decay into inert biomass, particulate biodegradable substrate, and
particulate nitrogen. With the exception of the inert biomass, both particulate
fractions are hydrolyzed into soluble biodegradable substrate and soluble
biodegradable organic nitrogen, respectively. The soluble biodegradable organic
CHAPTER 7
Using CFD to Estimate External Mass Transfer Coefficients and
Intra-Particle Diffusional Effects on the Supercritical
Hydrogenation of Sunflower Oil
A. Guardo1,*, E. Ramírez2, M.A. Larrayoz3 and F. Recasens3
1
Fluid Mechanics Department, Universitat Politècnica de Catalunya, Av.
Diagonal 647, ETSEIB, 08028, Barcelona, Spain; 2Chemical Engineering
Department, Faculty of Chemistry, Universitat de Barcelona, Martí i Franquès 1,
08028, Barcelona, Spain and 3Chemical Engineering Department, Universitat
Politècnica de Catalunya, Av. Diagonal 647, ETSEIB, 08028, Barcelona, Spain
Abstract: Hydrogenation of vegetable oils is an important process in the food industry
because of its widespread application to produce margarines, shortenings, and other
food components. Supercritical technology has proven to be a reliable alternative to
conventional hydrogenation process because not only the trans isomer levels can be
reduced, but also offers a clean, economic and environmental friendly process.
Computational Fluid Dynamics (CFD) modeling applied to the supercritical
hydrogenation reaction can be useful in visualizing and understanding the mass transfer
phenomena involved. CFD is applied to the study of the catalytic hydrogenation of
sunflower oil in the presence of a supercritical solvent. A mix of sunflower oil,
hydrogen and supercritical propane (used as a solvent) is the flowing fluid. Their
transport properties at high pressure are incorporated within a CFD commercial code in
order to estimate them online within the simulation process. A 2D CFD model of a
single Pd-based catalyst pellet is presented. Intra-particle and surface concentration
profiles and surface mass fluxes for all species present in the mixture (oil triglycerides
and hydrogen) are obtained and compared against experimental results. Different
temperatures, flow velocities and particle sizes are studied and external and internal
mass transfer phenomena are analyzed. External mass transfer coefficients for hydrogen
and oil triglycerides are obtained and a correlation for estimating them is presented.
Keywords: Computational fluid dynamics, high pressure, supercritical fluids,

hydrogenation, single phase flow, packed bed reactor, vegetable oils, mass
transfer, trans fatty acids, food industry.
INTRODUCTION
The hydrogenation of edible oils is an important process in the fat industry
*Address correspondence to A. Guardo: Fluid Mechanics Department, Universitat Politècnica de

Catalunya, Av. Diagonal 647, ETSEIB, 08028, Barcelona, Spain; E-mail: alfredo.guardo-zabaleta@upc.edu
Using CFD to Estimate External Mass STP Flows on Chemical and Biomedical Engineering 175
because of its wide applications to produce margarine, shortenings and frying oils.
The general purpose of this process is to increase the melting point and oxidation
point of the liquid oil [1]. The quality and physical properties of the final product
during hydrogenation are greatly affected by the number of double bonds present
in the oil, i.e., the iodine value (IV) and the cis-trans C18:1 isomers [2].
The industrial hydrogenation process is typically carried out in a three-phase
dead-end reactor operating at 383-443 K and 0.1-0.5 MPa with a nickel-base
catalyst. In this process, the low reaction rates and the increase of the trans C18:1
content (about 40 wt %) by isomerisation are consequences of the low solubility
of H2 in the oil and the high mass transfer resistance for the hydrogen in the liquid
phase [3].
Trans isomers not only change the physical and chemical properties of
hydrogenated fats but were also found to be responsible for an increased risk of
cardiovascular and heart diseases [4-7]. Nutritionally, these isomers formed
during hydrogenation are regarded like saturated fatty acids.
Apprehension and public awareness have risen regarding the potential health
hazards of trans fatty acids intake in the human diet. United States Food and Drug
Administration (FDA) issued in 2003 a regulation requiring manufacturers to list
trans fat, on the Nutrition Facts panel of foods and some dietary supplements. In
the same year, Denmark became the first country to set an upper limit on the
percentage of industrially produced trans fat in foods, limiting trans fats from
sources other than meats and dairy products to a maximum of 2% of total fat in
each food item. In 2005, Canada became the first country to regulate the
mandatory labelling of trans fats on prepackaged foods. In 2006, the United States
introduced the mandatory declaration of trans fats in foods containing 0.5 grams
or more per serving.
In view of the Danish legislation, the European Food Safety Authority (EFSA)
was asked by the European Commission in 2004 to provide an opinion on the
presence of trans fatty acids in food and their effect on human health. The
provision of nutrition labelling is voluntary within the European Union. When it is
provided, it has to follow a prescribed format which does not include trans fatty
acids, unless the manufacturer voluntarily decides to provide this information for
176 STP Flows on Chemical and Biomedical Engineering Guardo et al.
consumers. Few products do provide information on trans fatty acids, and they are
not usually included in any form within the nutrition information panel. The
European Commission is due to publish proposals for a review of the nutrition
labelling directive, which provides an opportunity to address this.
As a consequence of the aforementioned facts, food industry tries to eliminate

trans-fatty acids in bakery food formulas using palm and canola oils. However,
these oils increase the saturated fatty acid content. Oil blends allow suppliers to
take plastic fats and blend them with oils to create free trans fat shortenings and
margarines with an appropriate plasticity. However, the functionality and melting
profiles of these oils prohibits their uses in certain applications.
Formation of the high-melting unsaturated fats or isomerisation accompanies

hydrogenation and appears to be in proportion to the selectivity of the reaction.
Therefore, compromises must be made between selectivity and isomer formation
when determining the best hydrogenation conditions for the various basestocks.
Control of the operating variables that affect the hydrogenation of fats and oils is
necessary to produce the desired product functionality.
Supercritical fluid technology is becoming important in the lipid and food

industry in a variety of fields. So far the emphasis, however, has been on
extraction processes [8]. Reaction and catalytic applications lag behind other
processes [9, 10]. In general, the benefits of SCF in heterogeneous catalysis have
been emphasized in certain isomerisation reactions involving coke deposition and
for enhancing intraparticle diffusion [11-13]. In catalytic reactions, hydrogenation
stands among the important reactions in petroleum processes and in the food and
fine chemical industries [3].
In contrast to the conventional process, the continuous processes using Pd as

catalyst and supercritical fluids (propane or CO2) as solvent, introduced in the
1990s by Tacke et al. [14-17] and Härröd et al. [18-23], showed that the reaction
is extremely fast and the formation of trans C18:1 isomers is reduced. Later, King
et al. [24], using SC carbon dioxide on supported Ni in batch found that the
reaction conditions had a strong influence on the characteristics of the final
product; Ramírez et al. [25] summarized the data of several authors regarding
hydrogenation rates.
CHAPTER 8
Mass Transfer Around a Single Soluble Solid with Different
Shapes Buried in a Packed Bed and Exposed to Fluid Flow
J.M.P.Q. Delgado*
LFC-Laboratório de Física das Construções, Departamento de Engenharia Civil,

Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto Frias, s/n,
4200-465 Porto, Portugal
Abstract: The present work describes the mass transfer process between a moving fluid
and a soluble solid mass (a sphere, a cylinder or a plane surface aligned with the flow, a
cylinder in cross-flow, a prolate spheroid and a oblate spheroid) buried in a packed bed
of small inert particles with uniform voidage. Numerical solutions of the partial
differential equations describing solute mass conservation were undertaken (for solute
transport by both advection and diffusion) to obtain the concentration field in the
vicinity of the soluble surface and the mass transfer flux was integrated to give the
Sherwood number as a function of the relevant parameters (e.g., Peclet number,
Schmidt number, aspect ratio of the soluble mass). Mathematical expressions are
proposed that describe accurately the dependencies found. The solutions of these
problems are useful in the analysis of a variety of physical situations, as in the analytical
models of continuous injection of solute at a point source, in a uniform stream, to
estimate the distance from the “contaminant source” beyond which the levels of
contaminant are expected to fall below some safe limit.
Keywords: Concentration profiles, numerical simulation, diffusion, dispersion,

peclet number, Schmidt number, fluid flow, flat slab, cylinder, sphere, cylinder in
cross flow, prolate spheroid, oblate spheroid.
INTRODUCTION
There are several situations of practical interest, both in nature and in man made
processes, in which there is fluid flow through a bed of inert particles, packed
around a solid mass that contacts with the moving fluid. Examples may be found
in diverse fields, such as dilute catalyst fixed bed reactors, fluidised bed
combustion, ore leaching and water contamination by buried waste. In such
*Address correspondence to J.M.P.Q. Delgado: LFC-Laboratório de Física das Construções,

Departamento de Engenharia Civil, Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto
Frias, s/n, 4200-465 Porto, Portugal; E-mail: jdelgado@fe.up.pt
Mass Transfer Around a Single Soluble STP Flows on Chemical and Biomedical Engineering 197
processes there is an interplay between diffusion, convection and dispersion and a

detailed systematic study of the problem has been given by Coelho and Guedes de
Carvalho [1, 2] for transfer from buried flat surfaces and for transfer from buried
spheres. Both these references present accurate solutions for certain limiting
situations, namely those of low and high fluid velocity (more precisely, low and high
Peclet numbers).
When the mass transfer process occur in a porous media with a fluid flowing
around the soluble particle, and at low fluid velocities (as typically observed in
underground flow), the assumption of thin boundary layer is not legitimate, and
the theoretical analysis developed by Coelho and Guedes de Carvalho [1] is not
applicable. Therefore, it is necessary to employ numerical methods for a correct
analysis of the mass transfer process in more general situations.
Flow around a buried sphere is an important model situation in many processes and
in a recent work Guedes de Carvalho et al. [3] treated the problem numerically, so as
to cover the entire range of values of Peclet and Schmidt numbers.
Flow along buried cylindrical surfaces and flat surfaces are also important model
situations, and were investigated theoretically and numerically by Alves et al. [4],
yielding results for a wide range of values of Peclet number, aspect ratio of
soluble solid mass and Schmidt number.
Over the last years, our focus has been driven preferentially to the analysis of the
mass transfer of particles with spherical geometry. This work is a contribution for
the study of mass transfer of soluble particles with different geometries buried in
inert particles with smaller diameter. Additionally, a simple approximate method
is presented to obtain concentration contours plots for solute distribution around
and downstream of the buried surfaces (with different geometries).
CASE STUDIES
The following sections present a detailed description of the mass transfer and
dispersion process around a soluble solid particle with different shapes (sphere,
cylinder or a plane surface aligned with the flow, cylinder in cross-flow, prolate
spheroid and a oblate spheroid) buried in a packed bed of smaller inert particles
with uniform voidage, with a moving fluid with constant interstitial velocity.
198 STP Flows on Chemical and Biomedical Engineering J.M.P.Q. Delgado
Mass Transfer and Concentration Profiles from a Soluble Flat Slab

Fig. 1 sketches a section through a packed bed along which liquid is flowing,
close to a flat wall, part of which (0 z L) is soluble. Liquid flow is assumed
steady, with uniform average interstitial velocity u. If the concentration of solute
in the liquid fed to the bed is c0 and the solubility of the solid wall is c*, a mass
transfer boundary layer will develop, across which the solute concentration drops
from c=c*, at y=0, to c c0 for large y.
Figure 1: Flow through packed bed near soluble flat surface.
If we restrict the analysis to those situations for which the mass transfer boundary
layer extends over several particle diameters and if a small control volume is
considered, inside this boundary layer, with side lengths z, y and unity, a steady
state material balance on the solute leads to
c  2c  2c
u  D 'm 2  D 'm 2 (1)
z y z
CHAPTER 9
Advances on Viscoelastic Fluid Flow Simulation
Jovani L. Favero1, Argimiro R. Secchi1,*, Nilo S.M. Cardozo2 and Hrvoje
Jasak3
1
PEQ/COPPE-Programa de Engenharia Química, Universidade Federal do Rio
de Janeiro, CT, Rio de Janeiro, RJ, Brazil; 2Departamento de Engenharia
Química, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil
and 3FSB/Wikki-Faculty of Mechanical Engineering and Naval Architecture,
University of Zagreb, Croatia and Wikki Ltd, London, United Kingdom
Abstract: The increasing necessity for use of synthetic polymer products, such as for
production of packaging, parts of appliances, electronics, cars, biomaterials for medical
applications, etc., has led to the polymer industry to seek reduction of waste and increase
product quality and productivity. Consequently, a better understanding of how the
rheological properties of polymers affect their processing and final product quality is of
great importance. In order to obtain faster results with less cost, the studies of modeling and
simulation of polymer processing are increasing every day. This chapter aims to describe
new tools for CFD simulations of viscoelastic fluids, implemented in the OpenFOAM CFD
package due to advantages offered by this software, such as the possibility to use multigrid
techniques and data processing parallelization, besides being a free software and open
source. The constitutive models of Maxwell, UCM, Oldroyd-B, White-Metzner, Giesekus,
Leonov, FENE-P, FENE-CR, linear and exponential PTT, Pom-Pom, XPP and DCPP were
included in the general multimode form. In order to validate the developed solver,
comparisons with numerical and experimental results from literature were carried out. The
results were satisfactory giving credibility to the implemented solver and ensure the
availability of a powerful tool for the study of viscoelastic fluids to be used both in
academia and in industry. An extension of this tool, used for analysis of free-surface
viscoelastic fluid flows using the VOF methodology, is also presented. The die swell
experiment, a classical flow phenomenon used in the rheology literature to present the
concept of viscoelastic effects, was also simulated. The results obtained using Giesekus
model showed the great potential of the developed formulation, once all phenomena
observed experimentally were reproduced in the simulations.
Keywords: Viscoelastic fluids, free surface flow, two-phase flow, finite volumes,
multimode simulation, die swell simulation, differential constitutive equations,
high weissenberg numbers, VOF, CFD, DEVSS, OpenFOAMTM.
*Address correspondence to Argimiro R. Secchi: PEQ/COPPE-Programa de Engenharia Química,

Universidade Federal do Rio de Janeiro, CT, Rio de Janeiro, RJ, Brazil; E-mail: arge@peq.coppe.ufrj.br
234 STP Flows on Chemical and Biomedical Engineering Favero et al.
INTRODUCTION
Synthetic polymer products are of great importance in several industrial sectors,

such as for production of packaging, parts of appliances, electronics, cars,
biomaterials for medical applications and food processing industries. Due to the
increasing demand for this kind of material, reduction of waste and increase of
quality and productivity have become key issues in polymer industry. Processing
operations involving viscoelastic melts, solutions or suspensions are important
steps in the definition of the characteristics and quality of the finished products. In
this sense, modeling and simulation of processing operations appears as a
fundamental tool, leading to better understanding of how the rheological
properties of polymers affect their processability and final product quality, and
reducing time and costs related to the development of processes and products.
The rheological response of viscoelastic fluids is quite complex, including

combination of non-linear viscous and elastic effects, such as strain rate
dependent viscosity, presence of normal stress differences in shear flows,
relaxation phenomena, and memory effects, including die swell [1, 2]. In addition
to the basic mathematical and computational difficulties associated with the
solution of the system of equations resulting from mass, momentum and energy
conservation laws, the analysis of viscoelastic flows requires dealing with an
additional set of variables and equations related to the constitutive equation used
to describe the rheological behavior of the viscoelastic fluid [3].
The first constitutive equation developed to describe viscoelasticity was the

Maxwell model [4]. This model is linear and, consequently, do not pose major
difficulties for the numerical solution of the flow problem. However, like all
linear viscoelastic fluid models (LVF), it is valid only for small deformation rates.
To get a better representation of real viscoelastic fluids at higher deformation rates
more sophisticated and complex non-linear viscoelastic models should be used. In
this class we may mention models derived from the kinetic theory (such as UCM
and Oldroyd-B [5], Giesekus [6], White-Metzner and Leonov [1, 7], FENE-P [8],
and FENE-CR [9]), from network theory of concentrated solutions and melts [10]
(linear and exponential Phan-Thien-Tanner-LPTT and EPTT), and from the
reptation theory [11, 12] (Pom-Pom, XPP, and DCPP models). A more complete
Advances on Viscoelastic Fluid Flow STP Flows on Chemical and Biomedical Engineering 235
list of models for viscoelastic fluid and some detailed comparisons among them
can be found in [1, 2, 7]. Additionally, the multimode versions of these models
provide better agreement with experimental data by using a discrete spectrum of
relaxation time instead of a single relaxation time. The improvement in predictive
ability resulting from the use of multiple modes is generally accompanied by a
computational cost due to the increase of the problem size, with six new variables
for each additional mode [13], in case of 3D simulations.
An important subject that is receiving a special attention, mainly in the last two
decades, is the use of these viscoelastic models to simulate free surface flow. It is
particularly useful for industries that work with processes as extrusion and
injection molding of plastics, metallic leagues or the complex fluids used in food
industry, where the final quality of their products depends strongly on how much
is known about the fluid rheology and its relation with the optimal processing
conditions [14]. The modeling of free surface flow presents the same bases of
internal flows plus an appropriate methodology for treatment of free surface
motion, such as the MAC (Marker-and-Cell) [15] and VOF (Volume-of-Fluid)
[16]. An overview of literature works about viscoelastic free surface simulation
can be found in [17]. Most of works treating viscoelastic fluid flow with free
surface use finite element method to solve simplified cases.
A relevant aspect regarding the analysis of viscoelastic flows is the development of

softwares specifically designed for this purpose. Despite the extensive literature on
viscoelastic behavior modeling and viscoelastic flow simulation, the commercial
packages meant to be applied to analysis of polymeric flows and polymer processing
operations, such as extrusion and injection molding, are actually limited to the
description of purely viscous non-Newtonian phenomena. Development and use of
softwares with viscoelastic flow analysis capability is still carried out almost
exclusively in academic environment, for specific applications.
According to this scenario the insertion of a viscoelastic fluid flow solver in a

widely recognized CFD (Computational Fluid Dynamics) package is a relevant
matter, since it would bring to viscoelastic fluid flow analysis field the main
features of a good CFD package, which include ability of dealing with complex
geometries, unstructured and non-orthogonal meshes, moving meshes, large
CHAPTER 10
Tree-Shaped Flow Structures Viewed from the Constructal
Theory Perspective
António F. Miguel*
Department of Physics and Geophysics Center CGE, University of Évora, Rua

Romão Ramalho 59, 7000-671 Évora, Portugal
Abstract: The generation of flow configuration (design, shape, structure, patterns,

rhythms) is a phenomenon that occurs across the board, in animate and inanimate flow
systems. Scientists have struggled to understand the origins of this phenomenon.
Among the configurations, tree-shaped structures dominate the design of natural and
man-made flow systems. Why are they so important? Is there a physics principle from
which their configuration can be deduced?
Constructal theory, conceived by Adrian Bejan, is the view that the generation of configuration in
nature is a universal phenomenon, which is covered by a law of physics-the constructal law. This
chapter addresses the generation of tree-shaped design in the light of constructal theory.
First, we briefly review the constructal theory that is applicable to animate and
inanimate flow systems. Next, we review the studies that are focused on the constructal
view of tree-shaped flow structures. The constructal theory presented in this chapter
introduces a new paradigm that is universally applicable in nature, engineering and
social sciences.
Keywords: Constructal theory, design in nature, topology optimization, lungs,

tree networks, dendritic, flow geometry, constructal design, Murray law,
tributaries streams, rivers basins, oscillatory flow.
TREE-SHAPED FLOW STRUCTURES
The study of tree-shaped flow structures dates back to a long time ago. Galen, for
example, who lived in the second century A.D. and spent part of his lifetime in
observation of the human body and its functioning, described the arteries as
"trunks divided into many branches and twigs" that nourished the body [1]. He
also wrote that the lung configuration “has all the properties which make for easy
*Address correspondence to António F. Miguel: Department of Physics and Geophysics Center CGE,
University of Évora, Rua Romão Ramalho 59, 7000-671 Évora, Portugal; E-mail: afm@uevora.pt
Tree-Shaped Flow Structures Viewed STP Flows on Chemical and Biomedical Engineering 267
evacuation" [1]. In the 15th century, Giacomo Berengario da Capri described the
circulatory system as "a tree of veins" and Leonardo da Vinci drawn images of the
anatomy of the lungs based on dissect human corpses [2, 3]. Leonard was also
convinced that, as the blood in the human body, water also traverses the earth in
channels and veins, branching out in all directions to vitalize and sustain the earth
and its creatures.
In 1808-1809 Thomas Young attempted to estimate the resistance of a tree-shaped

flow system-the arterial system. According to his “The Croonian Lecture on the
Functions of the Heart and Arteries” [4] he concluded that in order to obtain the
resistance the dimension of the system and velocity is required. Therefore, Young
assumed a symmetrical dichotomous system and presented a geometric ratio
between the diameters of daughter and parent vessels of 1.26 (~21/3). However, he
does not explain why he chose this ratio if empirical or theoretical. In 1917, H. R.
Hess [5] attempted to establish the ratio of 1.26 based on theoretical arguments.
Few years later, C.D. Murray [6] derived based on theoretical arguments the
relationship between the diameters of daughter and parent vessels for
asymmetrical and symmetrical branching systems. Murray obtained this
relationship using biological considerations. According to him, vascular systems
subjected to natural selection, must have an arrangement such that in every
segment of vessel, flow is attained with the least possible energy. Murray assumed
that there are two energy conditions that contribute to the total cost of maintaining
blood flow: the energy required to overcome viscous drag in a fluid obeying
Hagen-Poiseuille’s law, and the energy metabolically required to maintain the
volume of blood and vessel tissue involved in the flow. Therefore, the larger the
diameter the smaller the power required, but also the larger is the power required
for metabolic maintenance of the blood and vessel wall tissue. According to
Murray, a compromise must be established between the two factors to minimize
the total power. Total power as a function of diameter of vessel is minimized by
the value of diameter where the derivative of total power divided by the derivative
of the diameter of vessel equals zero and where the 2nd derivative of total power
divided by 2nd derivative of diameter is greater than zero. For symmetrical and
dichotomous systems, with a vessel of diameter D1 that branches into 2 vessels of
diameter D2, the relationship between parent and the daughter vessels is given by
268 STP Flows on Chemical and Biomedical Engineering António F. Miguel
D1=21/3D2. An assumption should be imposed on this derivation. The power

required for Hagen-Poiseuille flow depends on the internal diameter, whereas the
metabolic requirement depends on the external diameter (i.e., it depends on the
volume of the vessel as well as the volume of the contained fluid). In the
derivatives of formulas the same diameter is used in both terms, which suggests
that the law only appropriates for thin vessels with the same internal and external
diameters. Despite to derive the Young rule by theoretical arguments, Murray’s
work went almost unnoticed for nearly five decades [7].
In summary, symmetrical and dichotomous systems such as those of Young and

Hess obey to the following rule: the cube of the diameter of a parent duct should
equal the sum of the cubes of the diameters of the daughter ducts. This
relationship is usually termed as Murray or Hess-Murray law but perhaps should
be named Young-Hess-Murray law.
Tree-shaped flow structures are not exclusive of living organisms. They play a
basic role in both animate and inanimate systems [8-10]. Tree-shaped flow
structures are of common appearance in leaves, roots, vascular tissues, respiratory
tree, neural dendrites, and bacterial colonies but also in river drainage basins, river
deltas, dendritic crystals, heat exchangers, water distribution networks, urban
growth, etc. The level of complexity which a tree structure exhibits is not simply
of a more advanced degree but of an overall different degree. In fact, scientists
such as Horton [11], Schumm [12] or Shreve [13] took an interest in analyzing the
complex ordering of these networks. They performed topological and
morphometric analyses [14] which can be applied to all branching networks
organized into a hierarchy. Latter, for the description of many of the irregular and
fragmental patterns in nature Mandelbrot [15] has introduced the concept of
fractal (or Hausdorft) dimension. Many of the natural branched structures are
fractal structures. Therefore, they can be approximated by repeating a finite
number that follows a postulated properly designed algorithm, which results in an
increasing number of channels with smaller diameter and an increase in the total
cross-sectional flow area. Fractals do not account for dynamics, hence are
descriptive rather than predictive. Fractal analysis has been applied to a large
variety of animate and inanimate systems whose fractal dimensions have been
sought [16-21]. However, the fractal properties in different systems have different
CHAPTER 11
Determination of Effective Transport Properties of Metallic
Foams: Morphology and Flow Laws
J. Vicente1,*, E. Brun2, J.M. Hugo1, J.P. Bonnet3 and F. Topin1
1
Université de Provence, Laboratoire IUSTI CNRS UMR 6695, Technopole de
Château-Gombert, Marseille, France; 2Mechanical Engineering Department,
University of California, Santa Barbara, CA 93106, United States and 3M2P2
UMR CNRS 6181, Université Paul Cézanne, Europôle de l’Arbois, Bâtiment
Laennec, Hall C. 13545 Aix en Provence Cedex 4, France
Abstract: This paper presents a combined approach to study the impact of the structure of
cellular materials on their transport properties. This work is based on the use of the
morphological 3D analysis software (iMorph), to precisely characterize the geometry of
cellular solids, extensive experimental characterization at sample scale and numerical
simulation of heat and mass transfer at pore scale. With these approaches, and the study
on a wide range of metallic foams, we identify the geometrical relevant criteria for the
single and two-phase flow properties understanding.
Keywords: Single and two phase flow laws, effective transport properties, metal
foams, X-ray tomography, 3d images, morphology, representative elementary
volume (rev), numerical simulations, finite volume, lattice boltzmann.
INTRODUCTION
Solid foams are a relatively new class of multifunctional materials that present
attractive thermal, mechanical, electrical and acoustic properties. Moreover, they
also promote mixing and have excellent specific mechanical properties. They are
widely quoted to present a random topology, high open porosity, low relative,
large accessible surface area per unit volume [1]. They are nowadays proposed for
use in numerous applications such as compact heat exchangers, reformers,
biphasic cooling systems and spreaders. They are also used in high-power
batteries for lightweight cordless electronics, and catalytic field application such
*Address correspondence to J. Vicente: Université de Provence, Laboratoire IUSTI CNRS UMR 6695,
Technopole de Château-Gombert, Marseille, France; E-mail: jerome.vicente@polytech.univ-mrs.fr
Determination of Effective Transport STP Flows on Chemical and Biomedical Engineering 293
as fuel cells systems [2-5]. Usually, geometric properties such as high tortuosity
and high specific surface are proposed to explain their interesting properties.
A synthesis of correlations for heat exchanger design is made by Mahjoob and

Vafai [3]. They gathered heat transfer coefficient and friction factor laws from
experimental, analytical and numerical studies. Thermo-physical and flow
properties depend strongly on local morphology of both pore and solid matrix.
Local change in the structure could govern the properties (e.g., constriction, strut
cross section, surface roughness…). Accurate evaluation of these properties
becomes critical for various uses. Yet, solid foam structure and properties are still
incompletely characterized.
Literature models of effective thermal properties widely used for low porosity
media are more difficult to apply to these high porosity materials [6, 7]. Most of
the modeling approaches are based on arbitrary periodic structures. Thus,
modeling approaches remains yet only qualitative [8-11]. Very few works uses
real geometry (usually obtained from 3D X-ray tomography) to determine
physical properties [12].
In the present paper, the single and two-phase flow behavior is studied. Fluid flow
parameters are experimentally measured at a macroscopic scale. For single phase
flow, numerical simulations are performed at the pore scale on the real matrix
using foam real geometry obtained from tomographic images. Numerical
simulations allow observing locally macroscopic properties as well as velocity,
temperature or pressure fields. Effective properties are, then, deduced to
determine flow laws and heat transfer performance that can be directly compared
to experimental results. Once the numerical approach is validated, Representative
volume element (RVE) size of fluid flow properties is calculated. Then, the
precise knowledge of the foam geometry allows quantifying the impact of the
morphology on the fluid flow parameters.
MATERIALS
A dozen metallic foams samples, chosen in order to investigate a wide range of

pore size, were tested. In our case, the pore diameter varies from 500µm up to
5000µm. The specimens (50mm*150mm) are cut in foam sheet of variable
294 STP Flows on Chemical and Biomedical Engineering Vicente et al.
thickness (from 1.5 up to 10.3 mm depending on pore size). Recemat® metallic

foams (RMF) are produced by the metallization (Nickel) using electro-deposition
of open cell polyurethane foams. The other samples, Ni 100-45, Cu 45-50 and Cu
10-100 that come from other companies (fibernide, SCPS) are manufactured in a
similar way.
In order to ascertain the structural characteristics of these media, we have used the
morphological data obtained using X-Ray tomography and the open source
morphological analysis software iMorph, developed in our laboratory [7].
X-ray microtomography is a technique which allows imaging a sample in 3

dimensions with micron resolutions and even, submicron ones. High resolution
tomographic experiments were performed using a synchrotron X-ray source on
beam line ID19 at the European Synchrotron Radiation Facility (ESRF) in
Grenoble. The 6 MeV electrons produce hard X rays in a wiggler. In our case, the
X-ray beam coming out of the ring was monochromated using two vertical silicon
single crystals and the beam energy was set to 60 keV. The absorption images are
collected using a fast read out frelon 2048*2048 CCD camera developed at ESRF.
The detector is set directly behind the sample and the contrast obtained on the 2D
projections results from the difference in X-ray attenuation by the phases/features
encountered by X-rays in the sample. 1500 2D radiographs, while sample is
rotating from 0 to 180° are recorded. From these radiographs, a last process uses
an appropriate algorithm (filtered back projection) to obtain a series of 2D slices
which build the 3D images.
The structure of cellular open cells materials presents different pertinent scales,
the cell size and the strut thickness for hollow struts, and the strut diameter for
filled struts. The choice of the resolution and of the sample size results then in a
trade off between these scales. Depending on the tomographied samples, we used
two different optics, and fixed the pixel size of the Frelon CCD camera to 7.46µm
(optics 10µm), 29.47µm (optics 40µm), and 0.7µm.
The 3D starting volume, derived from one scan, is divided into a set of grey tone
slices. In the case of a cellular material and due to the large difference between the
linear attenuation coefficient of air and metal, the high-resolution tomographic
Part II: Two Phase Flows on Chemical Enginnering
CHAPTER 12
Flow Visualization in Gas-Solid Packed Beds by Spatially
Resolved Near-Infrared Imaging
Aiouache Farid1,*, Nic An tSaoir Méabh1, and Luis Abreu Fernandes Daniel2
1
Queen’s University Belfast, School of Chemistry and Chemical Engineering, BT9
5AG, Belfast, Northern Ireland and 2Department of Chemistry, University of
Aveiro, Campus de Santiago, 3810-193 Aveiro, Portugal
Abstract: Conventional ‘single point’ spectroscopic techniques have been very

convenient in helping to understand underlying phenomena in gas-solid processes,
leading to concepts that rely on pseudo-homogeneous descriptions. These concepts,
however, are either not sufficiently valid at different spatial scales, or mature enough to
be able to describe local events since they use averaged profiles of concentration,
temperature and packing structure. Nevertheless, spatially-resolved optical techniques
are increasingly cited due to developments in tunable lasers, 2D array detectors and
communication technology. Optical techniques allow experiments to be performed non-
intrusively at high spatiotemporal resolution. The present review presents two
experimental procedures based on spatially-resolved near-infrared (NIR) imaging, in
order to observe temperature and concentration maps in gas-solid packed beds subjected
to effects of the entrance aspect ratio and non isothermal conditions. The first technique
was applied to a gas-solid fluidized bed reactor with a low aspect ratio of tube to
particle diameters (Dt/dp). The technique used NIR broadband light, interference optical
filters centred on absorbing and non-absorbing wavelengths of water vapour, a Vidicon
NIR camera and simple back-projection of collected images. The second technique was
applied to water vapour flow in a packed bed filled with a hydrophobic resin, using a
tunable diode laser, focal planar array detector and tomographic reconstruction
procedure. By “looking into” a thin fluidized bed, the proposed technique allowed
existing models of fixed bed reactors to be extended to pseudo-static bed operations.
The technique was applied to ceria-silica reduction in a fluidized bed reactor, where
radial profiles of water concentration allowed the distinction between surface and bulk
reduction regimes of ceria-silica packing. The tomography technique however, which
observes 3D spatially resolved imaging of temperature and water vapour concentration
in packed beds, revealed cold and hot spots, concentration maps and flow dynamics in
the core packed bed and in the vicinity of the wall. In addition, heat uptake from the
wall and mass transfer between and inside resin particles were found to be strongly
affected by local concentration, temperature and packing structure profiles.
Keywords: Near-infrared, visualization, adsorption, porous media, diffuse
*Address correspondence to Aiouache Farid: Queen’s University Belfast, School of Chemistry and
Chemical Engineering, BT9 5AG, Belfast, Northern Ireland.; E-mail: f.aiouache@qub.ac.uk
Flow Visualization in Gas-Solid Packed STP Flows on Chemical and Biomedical Engineering 333
tomography, packed beds, reactors, 2D modeling, transient processes, water

vapour, fluid flow, spatial distribution, spectroscopy, catalyst reduction,
dispersion.
INTRODUCTION
Until recently, chemical gas-solid processes were described using ‘single point’
spectroscopic techniques. These techniques have helped to understand fundamental
phenomena by using simplified concepts relying on pseudo-homogeneous
descriptions of underlying events, which are often not valid at different spatial scales
and use averaged concentration, temperature and packing structure data. [1, 2] Our
previous work on fluid flow and chemical kinetics in monolith reactors using spatial
mass spectroscopy (SpaciMS), has demonstrated that local flow fields have a
significant impact on concentration and temperature distributions; leading to
oscillating concentrations, changes in temperature and partial breakthrough during
carbon monoxide (CO) oxidation. [3] The SpaciMS technique however, was found
to be relatively intrusive since it was associated with an increase of 5 % in flow
velocity and 19 % in pressure drops inside the monolith channel. Therefore non
intrusive methods, such as spatially-resolved optical spectroscopy techniques have
attracted more interest in recent years. These techniques have revealed that the
averaged information commonly obtained concerning the performance of gas-solid
processes in terms of activity, selectivity, and stability, are in fact the consequence of
complex interplays between different physicochemical gradients of concentration,
temperature and velocity across different length scales. [2] A well-known example is
the catalytic oxidation of carbon monoxide (CO) on a Pt/Al2O3 catalyst [4].
Spatiotemporal oscillation patterns and moving fronts originating from the
segregated diffusion of CO, were well observed by infrared thermography at a
spatial resolution of millimetres, while spiral waves on an enhanced defect density of
Pt surfaces were revealed by two dimensional (2D) photoemission electron
microscopy at a spatial resolution of nanometres. Further examples have been cited
in recent reviews on spatiotemporal techniques and catalysis by Urakawa & Baiker
[1] and Weckhuysen [2].
Local dispersion profiles can quickly show the effectiveness of heat and mass
transfers inside and around solid beads of a packing bed. Local dispersions are
334 STP Flows on Chemical and Biomedical Engineering Farid et al.
commonly measured by multiple experiments in which invasive probes are placed

at different locations, leading to different residence time distributions. [5-8]
However, recent developments in spatially-resolved techniques have allowed
magnetic resonance imaging (MRI), [9] x-ray, [10] neutron, [11] and laser
spectroscopy [12] to measure anisotropic dispersion coefficients in liquid-solid
reactors, achieving rapid access to scalar data of chemical kinetics and associated
mass transfers in a single experiment. Unfortunately, these techniques were
revealed to be inadaptable or not sufficiently mature to be applied to gas-solid
processes, due to the inherently weak signals retrieved from the gaseous phase.
Nevertheless, optical techniques are increasingly cited in both open and industrial
literature, being boosted by rapid developments in tunable lasers and single or 2D
array detectors and leading to experiments being performed at high spatiotemporal
resolutions. Nadeau et al. [13] measured a residence time distribution in the order
of milliseconds, by directing an infrared laser light at different locations of a
reactor using methane as a tracer probe. The sub-second response time resolution
allowed rapid observation of local maldistribution and radial dispersions. Hindle
et al. [14] used optical tomography to observe spatial distributions of
hydrocarbons in a combustion chamber; Reinke et al. [15] coupled Raman and
laser-induced fluorescence techniques to observe the spatial distribution of stable
and radical species between two catalytic plates accessible by lateral transparent
windows, during the catalytic partial oxidation of methane. Mewes et al. [16] used
near-infrared (NIR) transmittance tomography to observe the spatial breakthrough
of water vapour in a packed bed adsorber. This technique relied on an NIR light
from a distributed feedback diode laser that was split into three lights, collimated
to a packed bed adsorber and detected by three InGaAs photodiode arrays. The
wall effect on adsorption breakthrough was demonstrated in a packed bed
adsorber with a low aspect ratio of tube to particle diameters.
FLOW VISUALIZATION BY SPATIALLY-RESOLVED NEAR

INFRARED IMAGINGTURBULENCE
Even less energetic than MRI tomography, X-ray tomography and neutron
tomography, NIR diffuse and transmittance tomography is becoming increasingly
acceptable for use in biomedical applications such as quantifying organ function.
The transport of NIR light in scattering environments such as tissue materials is
CHAPTER 13
Interfacial Area Modelling in Two-Phase Flow Studies
Christophe Morel*
CEA GRENOBLE, DEN/DER/SSTH/LDLD, 17, Rue Des Martyrs, 38054

Grenoble, France
Abstract: This chapter summarizes various theoretical approaches for interfacial area
modelling in two-phase flow studies. The interfacial area concentration is defined as the
contact area between the two phases per unit volume of the two-phase mixture. As the
surface available for mass, momentum and energy exchanges between the two phases,
this quantity is of the utmost importance. Various approaches were developed for its
modelling in the few last decades and we can classify these approaches mainly in two
groups. The first group is based on the restrictive assumption of spherical particles
(bubbles or droplets) and the interfacial area modelling is often based on an evolution
equation for this quantity. This evolution equation can be derived from a Liouville-
Boltzmann equation written for the distribution function of a quantity characterizing the
size of the particle (currently its diameter or its volume). The second group is
completely different and is based on local instantaneous evolution equations for pieces
of interfaces and their orientation in the flow, defined by the unit normal vector. Here,
the major assumption is the one of closed interfaces but the fluid particle shapes can be
arbitrary. The two kinds of approaches involve an interfacial area balance equation, but
they are different in their theoretical basis. The link between these two approaches is
given in the context of spherical particles. In the more general context of anisotropic
interfaces (i.e., arbitrary shapes), the interfacial area evolution equation is supplemented
by additional evolution equations characterizing the interface anisotropy. Two possible
mathematical formalisms are presented here: the first one is based on the concept of the
interface anisotropy tensor introduced a long time ago in the study of liquid-liquid
dispersions (e.g., polymer blends) and the second one is based on two equations for the
averaged mean and Gaussian curvatures.
Keywords: Interfacial area concentration, two-phase flow, interface anisotropy,

mean and gaussian curvatures, evolution equations
INTRODUCTION
Two-phase flows are encountered in many power and process industries, as well
*Address correspondence to Christophe Morel: CEA GRENOBLE, DEN/DER/SSTH/LDLD, 17, Rue Des
Martyrs, 38054 Grenoble, France. E-mail: christophe.morel@cea.fr
362 STP Flows on Chemical and Biomedical Engineering Christophe Morel
as in environmental studies, biomedical studies.. Here we are essentially

interested in gas-liquid two-phase flows. In this kind of flows, liquid and gas
phases share the flow domain, being separated by many interfaces which can be
considered as surfaces without thickness. The two phases do not evolve
independently since they constantly exchange mass, momentum and energy
through the interfaces. The amount of interfaces in a unit volume of two-phase
mixture is therefore a very valuable information to quantify all the exchanges
between phases. Unfortunately, in the great major part of the flows of interest, all
the flow details are not accessible to the numerical simulation due to the large
number of scales often involved and the associated prohibitive number of freedom
degrees. Due to this complexity, statistical models are often developed to predict
only the main features of the flow evolution. Amongst these statistical models, the
most popular seems to be the so-called two-fluid model [1, 2] which offers the
greatest generality in terms of mechanical and thermal imbalances between
phases. In this approach, two sets of mass, momentum and energy balance
equations are written for the two phases. These equations are averaged ones, they
result from the averaging of the Navier-Stokes equations written for the two
phases, together with interfacial balances. The averaging operator can be a spatial,
a temporal or an ensemble average. Here, for the sake of generality, we use an
ensemble averaging operator.
The equations of the two phases are coupled through interfacial interaction terms
which represent the mass, momentum and energy exchanges between phases on
the average. It is generally admitted that these exchanges are proportional to the
available contact area between phases per unit volume of mixture [1]:
int erfacial transfers  aI  driving force (1)
In this equation, aI denotes the interfacial area concentration (IAC) which is

defined as the interfacial area per unit volume of mixture. It is therefore a
fundamental quantity since it is involved in all kinds of transfers between phases,
and hence partly determines the flow evolution. Several complex physical
phenomena affect the IAC: we can cite bubbles or droplets coalescence and break-
up, phase change like vaporization or condensation, interfacial deformations (like
bubbles or droplets shape changes) [2]. All these phenomena must be taken into
Interfacial Area Modelling in Two-Phase STP Flows on Chemical and Biomedical Engineering 363
account and several recent reviews attest of the vivacity of the subject [1, 3-5].
Here, we will restrict our attention to the case of disperse two-phase flows
characterized by the existence of one continuous phase (carrier phase) and a
second phase embedded in the first as a great number of discrete (dispersed) fluid
particles, each particle being bounded by a close surface. The two examples in
gas-liquid flows are bubbles in a liquid and droplets in a gas. Thus we exclude a
priori separated two-phase flows characterized by a single large and open
interface, even if some of the concepts developed here also apply to this kind of
flows.
THE TWO DESCRIPTIONS OF DISPERSE TWO-PHASE FLOWS AND

THEIR CONSEQUENCES ON INTERFACIAL AREA MODELLING
An intuitive idea of the IAC has been given in the introduction. Now we must
define this quantity more precisely. From a general viewpoint, essentially two
descriptions exist to model a disperse two-phase flow. The first one is the classical
two-fluid model [1, 2] where the two phases are considered in a symmetrical way
and the obtained equations are very general in the sense that no assumption is
made about the interfacial configuration, i.e., the flow regime. The second
description [6] is based on an analogy with the kinetic theory of gases, where the
gas molecules are replaced by macroscopic particles (like bubbles or droplets)
surrounded by the carrier fluid. In the first approach, the IAC is defined as the
sum of all the pieces of interfaces contained in a unit volume. In the second
approach, restricted to disperse mixtures, the IAC involves a sum over the
surfaces of all particles having their centres inside the unit volume.
We first give the definitions of the IAC in the two contexts mentioned above [6,
7]. In the most general context of the two-fluid model, the IAC aI is defined as the
average of a Dirac distribution having the different interfaces as a support:
aI   I (2)
For the particular case of a suspension of N spherical particles having their centres
located at the set of position vectors R(t) and characterized by their radii a(t) (
= 1, …, N), the Dirac distribution appearing in (2) is given by:
CHAPTER 14
Mass Transfer Models for Oxygen-Water Co-Current Flow in
Vertical Bubble Columns
Valdemar Garcia1 and João Sobrinho Teixeira2,*
1
Instituto Politécnico de Bragança, Escola Superior de Tecnologia e de Gestão,
Quinta de Santa Apolónia, 5300-999 Bragança, Portugal and 2Instituto
Politécnico de Bragança, Escola Superior Agrária, Quinta de Santa Apolónia,
5301-857 Bragança, Portugal
Abstract: The present work reports a theoretical and experimental study of mass
transfer for oxygen-water co-current flow in vertical bubble columns. The axial dispersion
of liquid phase was also studied. Experiments were carried out in a 32 mm internal
diameter and 5.35 and 5.37 m height columns. The superficial liquid velocity ranged
from 0.3 to 0.8 m/s and volumetric flow rate ratio of gas to liquid ranged from 0.015 to
0.25. Mathematical models were developed to predict concentration of gas dissolved in
the liquid as function of different physical and dynamic variables for two-phase co-
current downflow and upflow. We obtained for the ratio of the liquid side mass transfer
coefficient to initial bubbles radius, kL/r0=0.12 s-1.
Keywords: Vertical bubble columns, gas-liquid bubble flow, oxygen-water flow,

oxygen-water mass transfer, mass transfer models, gas-liquid co-current
downflow, gas-liquid co-current upflow, mass transfer coefficient, axial
dispersion, liquid axial dispersion coefficient, U tube.
INTRODUCTION
The simultaneous gas and liquid flow frequently occurs in several situations and
in many industrial applications such as distillation columns, chemical and nuclear
reactors, pipelines for hydrocarbon mixtures transport, solar collectors, mass
transfer equipments like bubble columns, packed columns, air lifts pumps, among
others. Bubble columns, equipment where the gas phase is dispersed in small
bubbles in continuous liquid phase, have been used as gas-liquid contactors
*Address correspondence to João Sobrinho Teixeira: Instituto Politécnico de Bragança, Escola Superior
Agrária, Quinta de Santa Apolónia, 5301-857 Bragança, Portugal. E-mail: sobrinho@ipb.pt.
Mass Transfer Models for Oxygen-Water STP Flows on Chemical and Biomedical Engineering 387
devices, like as absorbers and/or strippers, chemical reactors for hydrogenation,

oxidation or chlorination of organic liquids such as fermenters in wastewater
biological treatment units.
Particularly, the air/oxygen-water flow can be found in areas where the water
oxygenation is very important, like the aquaculture and the water treatment units
(drinking, waste and river). The simultaneous flow of oxygen and water in co-
current bubbly regime, first in downflow and then in upflow, occurs in a U tube,
device with applications on referred areas [1-6].
The liquid mixing or axial dispersion is one of the parameters that can influence
the process of mass transfer of gas to liquid and therefore must be known. The
literature on liquid axial dispersion for bubble columns, especially for the air-
water system, is extensive. Many empirical [7, 8] and theoretical [9-11] equations
have been developed in order to predict the axial dispersion coefficient.
In the present work, the liquid phase mass transfer coefficients and the axial
mixing (axial dispersion coefficients) of the liquid phase of vertical bubble
columns operating with co-current downflow and upflow were determined for the
oxygen-water system.
EXPERIMENTAL
The experiments were performed on two vertical bubble columns, linked in the
bottom in U tube form, operating with co-current downflow and upflow of gas
and liquid. The columns are respectively, 5.37 and 5.35 m height and 32 mm
inside diameter.
Gas Absorption
The absorption experiments were performed in acrylic glass tubes. The liquid, tap
water, was circulated in the tubes by a centrifugal pump and was introduced at the
top of test section. The water flow rate was measured by a calibrated rotameter
and was regulated by a ball valve. The gas used for all experiments was pure
oxygen supplied by a high pressure bottle. The oxygen was introduced into the
test column through a porous gas distributor installed at the top of the test section.
The porous gas distributor allowed the formation of small diameter bubbles (about
388 STP Flows on Chemical and Biomedical Engineering Garcia and Teixeira
2-4 mm). The volumetric oxygen flow rate was determined with an electronic
flow meter.
The pressure at entrance of the column was measured by a pressure sensor.

Oxygen concentration in the water (liquid phase) was measured by means of five
calibrated electrodes at five levels. The temperature was measured in 4 points by
thermocouples. The operating temperature was 17±2ºC. The recorded data was
stored by an acquisition system connected to a PC unit.
Experiments were performed at superficial liquid velocities, U, ranging from 0.3

to 0.8 m/s and at gas and liquid volumetric flow rates ratio, χ=G0/L, ranging from
0.015 to 0.25. The experimental set-up is shown in Fig. 1.
Figure 1: Schematic diagram of the experimental set-up. OD1-OD5-oxygen probes; PM-pressure

meter.
Axial Dispersion
Determination of the axial dispersion parameters was based on the typical method
of tracer injection. A small amount of tracer (1x105 g/m3 NaCl aqueous solution)
was manually injected as fast as possible at the top of the downflow column and
at the bottom of the upflow column.
CHAPTER 15
Characterization of Turbulence and Flow Regimes in Bubble
Columns Based on Nonlinear Chaos Analysis of Various Data
Stoyan Nedeltchev*
Institute of Technical Chemistry, TU Braunschweig, Hans-Sommer-Strasse 10,
38106 Braunschweig, Germany
Abstract: The hydrodynamics of gas-liquid bubble columns operated in the churn-

turbulent regime are studied by examining the liquid turbulence along the height of the
bubble bed (based on chaotic and statistical methods). In the case of air-water and
nitrogen-PSS8 systems based on comparison of the KE values along the height of the
bubble bed it is shown that the degree of liquid turbulence (chaos) is stronger in the
vicinity of the gas distributor and it increases again in the upper part of the reactor.
These results help to understand better the occurrence of the circulation zones in the
bubble bed. The work includes results from five different bubble columns (0.1, 0.16,
0.19, 0.38 and 0.8 m in inner diameter) equipped with porous or perforated plate
distributor or cross sparger. The boundaries of the various flow regimes in different gas-
liquid systems are also identified. Five different liquids (tap water, Therminol LT,
polyalphaolefin liquid (PSS8), 1-butanol and gasoline) and two different gases (nitrogen
and air) are used. Different measurement techniques (Computer-Automated Radioactive
Particle Tracking, Computed Tomography, Nuclear Gauge Densitometry, differential
pressure and absolute pressure transducers) are employed. It is found that by means of
the Kolmogorov entropy (KE) algorithm, average cycle time and the average absolute
deviation can be identified the boundaries of chain-bubbling regime, bubbly flow
regime, first and second transition regimes and coalesced bubble regimes (consisting of
4, 3, 2 and 1 regions). The results can be used for the preparation of flow regime maps.
Keywords: Bubble columns, hydrodynamics, flow regime identification, churn-

turbulent flow regime, liquid turbulence, nonlinear chaos analysis, time series,
kolmogorov entropy, information amount, degree of chaos, nonintrusive
radioactive techniques, pressure transducers, gas holdup measurements, zones of
enhanced turbulence
INTRODUCTION
The bubble column reactors are commonly used in the process industries for
*Address correspondence to Stoyan Nedeltchev: Institute of Technical Chemistry, TU Braunschweig,

Hans-Sommer-Strasse 10, 38106 Braunschweig, Germany; E-mail: snn13@gmx.net
Characterization of Turbulence and Flow STP Flows on Chemical and Biomedical Engineering 413
carrying out a variety of liquid phase reactions [1]. The simple construction and
absence of moving parts make bubble columns cheap and easy for maintenance.
Usually these reactors operate in two hydrodynamic regimes: at low superficial
gas velocities uG the homogeneous (bubbly flow) regime occurs in which the
dispersion consists of bubbles that are roughly uniform in size. At high uG values
(beyond the transitional gas velocity) the heterogeneous (churn-turbulent) regime
occurs. In this flow regime there is a wide bubble size distribution depending on
the system properties [2]. Usually at high uG values a wide gas phase residence
time distribution is observed which is often detrimental to reactor conversion and
selectivity.
The main parameters that characterize the bubble column hydrodynamics are the
gas holdup, mean bubble diameter, mixing and the volumetric mass transfer rate.
These parameters can be affected by the column geometry and the existing
operating conditions, of which the gas and liquid flow rates are the most
important. Other key variables are the properties of the continuous media used
inside the reactor and the type of gas sparger used.
Some authors [3] reported that gas holdup increases with axial height to a value at
the top of the column over 1.6 times higher than the one at the bottom of the
column. According to them, this axial dependence of the gas holdup occurs
regardless of the sparger type and column height. These observations were made
in a large bubble column (23 m in height, 1.0 m in diameter). The variation in gas
holdup between the top and the bottom of the reactor increases as larger columns
are used [3]. It is worth noting that near the bottom of the column there is an
uneven (non-uniform) distribution of the bubbles over the cross-section, creating
columns of bubble-water jets/plumes.
A large part of the volume of the smaller columns is taken up by the sparger zone,
which is located above the sparger. The sparger zone is where bubble coalescence
and breakup are not in equilibrium and, as a result, where bubble size changes
more rapidly with height than throughout the rest of the column. This affects
particularly the measured bubble size in the smaller columns. Giovannettone et al.
[3] reported that the Sauter mean bubble diameter does not vary substantially with
column diameter. Their results show that the Sauter mean bubble diameter does
414 STP Flows on Chemical and Biomedical Engineering Stoyan Nedeltchev
not depend on the total unaerated height of the column while the distribution of
bubble sizes exhibits some influence. According to the authors the bubble
diameter does not vary with depth or gas flow rate under all operating conditions
except near the sparger. Colella et al. [4] measured bubble diameter in two bubble
columns of varying heights and diameters and found that bubble size decreased
with height in the shorter bubble column, however it did not change in the larger
bubble column above a height of 0.375 m from the gas sparger. Polli et al. [5]
found similar results and estimated the sparger zone to extend one column
diameter above the sparger. Magaud et al. [6] found that bubble diameter changes
little with height in columns of comparable size.
TURBULENCE
Turbulence in bubble columns is associated with the formation of liquid

circulation patterns and it is considered to be a nonlinear stochastic phenomenon.
Hinze [7] developed the fundamentals of the turbulence in dispersion processes.
As the bubbles rise in a bubble column they carry liquid upwards with them in
their wake and at higher uG values in between them (in swarms). At certain axial
position the liquid flows down again, thus giving rise to overall liquid circulation
patterns. This effect was first reported by De Nevers [8] who stated that the liquid
circulation pattern is induced by density differences caused by the maldistribution
of bubbles over the cross-section of the bubble column. In the core of the column
the gas holdup is higher than close to the wall, resulting in liquid upflow in the
central region and liquid downflow along the wall. It is worth noting that the
maldistribution of gas holdup is generally modeled with a power law. Various
models describe the flow field in terms of a large-scale circulation pattern [9, 10].
Joshi and Sharma [11] assumed the formation of a stack of donut-shaped
circulation cells. Zehner [12] developed a model based on circulation cells that
span the total column diameter. Devanathan et al. [13] demonstrated the presence
of one large liquid circulation pattern.
It is most relevant to speak about turbulence in the churn-turbulent regime. It is

characterized by vigorous mixing, wide distribution of bubble sizes and the
existence of a radial gas holdup profile which causes liquid circulation [14]. It is
worth noting that in this flow regime, bubble coalescence and break-up occur.
CHAPTER 16
Characterization and Modeling of Flotation Processes
G.G. Kagramanov* and V.A. Kolesnikov
D.I. Mendeleyev University of Chemical Technology of Russia, 125047,

Miusskaya Sq., 9, Moscow, Russia
Abstract: Flotation is successfully applied in various processes, such as galvanic

tailings purification, washing and lubricating liquids treatment, oil/water emulsions
separation, etc. An interesting implementation appears to be using flotation as
pretreatment in membrane separations. The prospective engineering approach for
contaminated wastewaters treatment is combined membrane-electroflotation
process.Membrane-electroflotation is a complex physico-chemical process taking place
in “liquid-gas-solid” system, based on the hydrodynamic phenomena of gas-liquid
streams. In case of membrane flotation the dependence of gas content on
bubbling rate has been determined. There are two areas of dependence—at low
gas content (5-17%), area I and at higher gas content (17-29%), area II. Based on
the value of single bubble emersion rate calculated for every measured bubble
diameter, an average volumetric single and group bubble emersion rate in
membrane flotation was determined. In electroflotation the average radius of
hydrogen bubbles increases from 20 microns to 90 microns depending on module’
height; and stationary distribution of bubbles is achieved upon 1-4 min. Hydrogen
bubbles content versus the process time is characterized by the time of retard. At the
same time the oxygen appears in electroflotation unit at the very beginning of the
process without any retard time. Increase of hydrogen bubbles residence time in the
upper sections of electroflotation module leads to rise of average radius and decrease of
bubbles containing from 0.00525 to 0.00109.
Keywords: Wastewater treatment, flotation, membrane-electroflotation, gas

content, flotocomplex, bubbles flow rate, experimental characterization
INTRODUCTION
Various wastewater treatment methods have been developed in recent years

because of the high environmental risks of the wastewater discharge and rising
water scarcity in the world. Flotation is one of the most cost-effective separation
methods, applied with increasing frequency in wastewater treatment practice.
*Address correspondence to G.G. Kagramanov: D.I. Mendeleyev University of Chemical Technology of

Russia, 125047, Miusskaya Sq., 9, Moscow, Russia; Email: kadri@muctr.ru
Characterization and Modeling of Flotation STP Flows on Chemical and Biomedical Engineering 441
Flotation is a hetero-phase systems separation method, using gas bubble

attachment. The attached particles (droplets) are floated and separated from the
liquid dispersing medium [1-4]. Water and wastewater applications of flotation
are metal coating wastes purification, washing and lubricating liquids treatment,
surface water clarification and thickening of activated and coagulated sludge, etc.
The flotation methods are traditionally classified based on the method of gas (air)
bubbles production [1, 5]. The methods can be differentiated in three groups:
1) dispersed-air flotation, air bubbles are formed by flow through a

diffuser (pneumatic flotation) or by mechanical shearing action
(mechanical flotation);
2) dissolved-air flotation: pressure flotation (dissolving air in water at

overpressure and flotation at atmospheric pressure) and vacuum
flotation (dissolving air in water at atmospheric pressure and then
applying a vacuum for flotation);
3) other flotation methods: electroflotation using controlled mild

electrolysis for production of hydrogen and oxygen bubbles,
biological flotation using gas of biological origin (fermentation), etc.
The most prospective approach for contaminated wastewaters treatment is

combined membrane-electroflotation process.
MEMBRANE FLOTATION
Microporous membranes are frequently used as air diffusers [6, 7]; these devices
have some price benefits and can produce air bubbles, which are small enough to
achieve a large interfacial surface area between air and water. This process is
similar to the well-known membrane emulsification process, which uses
microporous membranes for a stable emulsion production [8-11].
The Theory of Bubbling Bed Energy Minimum in Flotation

The bubbling rate w is related with gas content and single bubble emersion rate w0
with the equation:
442 STP Flows on Chemical and Biomedical Engineering Kagramanov and Kolesnikov
w   f   w0 (1)
Where f   is a function of gas content, expressing bonding effect of emerging

bubbles.
Finding of function f   , that is the ratio of group emersion rate w  and single
bubble emersion rate, appears to be a challenge, as the literature on this problem is
contradictory. There is a number of the empirical formulas, most frequently
presented as [12, 13]:
f    1   
n
(2)
If the height of bubbling bed is big enough (about 1 m and more) the gas (air) is
distributed along the height according to the energy minimum of gas-liquid bed. A
function of minimum of potential, kinetic, dissipation and surface tension energies
could be presented as follows [14]:
H  w2 3  w2 6 
J     g 1    x   x     1    dx (3)
0 
 2 1    4  1   
2n
d 
d-pore size (diameter) (m);
g-gravitational acceleration (9.81 m/s2);
H-height of bubbling bed (m);
J-the function of minimum bubbling bed energy;
w-bubbling rate (superficial velocity of gas flow), m/s;
w0-single bubble emersion rate (m/s);
w1-group emersion rate of bubbles (m/s);
x-vertical coordinate (m);
ρ-liquid (water) density (kg/m3);

Part III: Single and Two Phase Flows on Biomedical
Enginnering
CHAPTER 17
Pulse Wave Propagation in Large Blood Vessels Based on Fluid-
Solid Interactions Methods
Tomohiro Fukui1,*, Kim H. Parker2 and Takami Yamaguchi3
1
Department of Mechanical and System Engineering, Kyoto Institute of
Technology, Gosyokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan;
2
Physiolosical Flow Studies Group, Department of Bioengineering, Imperial
College, London, SW7 2AZ, United Kingdom and 3Department of Biomedical
Engineering, Graduate School of Biomedical Engineering, Tohoku University, 6-
6-01 Aoba-yama, Sendai 980-8579, Japan
Abstract: Pulse Wave Velocity (PWV) is recognized by clinicians as an index of

mechanical properties of human blood vessels. This concept is based on the Moens-
Korteweg equation, which describes the PWV in ideal elastic tubes. However, measured
PWV of real human blood vessels cannot be always interpreted by the Moens-Korteweg
equation because this formula is not precisely applicable to living blood vessels. It is
important to understand the wave propagation in blood vessels for a more reliable
diagnosis of vascular disease. In this study, we modeled uniform arteries in a three-
dimensional coupled fluid-solid interaction computational scheme, and analyzed the
pulse wave propagation. A commercial code (Radioss, Altair Engineering) was used to
solve the fluid-solid interactions. We compared the regional PWV values obtained from
various computational models with those from the Moens-Korteweg equation, and
discuss the accuracy of our computation. The PWV values from the thick-walled artery
model are lower than those from the Moens-Korteweg equation. Nevertheless, the
differences are less than 7% up to 12 m/s of the PWV, indicating these computational
methods for the PWV analysis are accurate enough to evaluate its value quantitatively.
Keywords: Pulse wave propagation, fluid-solid interactions, PWV, large blood

vessels, blood flow, Moens-Korteweg equation, arterial wall stiffness, arbitrary
lagrangian eulerian, wave reflection, sound speed.
INTRODUCTION
The diagnosis of cardiovascular disease by measuring pulse wave velocity (PWV)
*Address correspondence to Tomohiro Fukui: Department of Mechanical and System Engineering, Kyoto
Institute of Technology, Gosyokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan; E-mail:
fukui@kit.ac.jp
Pulse Wave Propagation in Large Blood STP Flows on Chemical and Biomedical Engineering 461
is believed to be a promising technique. The PWV is defined as the velocity of an

arterial wall disturbance toward the periphery, which occurs, for example, due to
contraction of the ventricle. In general, the PWV is determined by measuring the
time delay of the waveforms, t, between the two sites with a known distance L.
Therefore,
L
PWV  . (1)
t
The PWV is believed by clinicians to be increased with the severity of vascular

disease such as atherosclerosis [1-3]. The concept for applying the PWV as an
index of vascular disease is based on the Moens-Korteweg equation [4], which
formulates the PWV of a long straight elastic tube. According to the Moens-
Korteweg equation,
Eh
PWV  , (2)
2 ri
where E is the Young’s modulus of the arterial wall, h is the wall thickness,  is
the blood density, and ri is the internal radius of the artery. For thick-walled tubes,
the Moens-Korteweg equation has been modified by computing the strain on the
middle wall of the tube [5].
Eh , (3)
PWV 
2  ri  h 2
In the presence of flow, we assume that the wave will be convected with the
cross-sectional averaged velocity of the blood [6]. For a thick-walled tube with
flow, we therefore use the “modified Moens-Korteweg equation”,
Eh
PWV   U ., (4)
2  ri  h 2
where U is the cross-sectional averaged velocity of the blood. It is recognized that

an increased Young’s modulus E will result in an increased PWV. This
462 STP Flows on Chemical and Biomedical Engineering Fukui et al.
interpretation, however, is not always applicable to living blood vessels because

the Moens-Korteweg equation includes some assumptions that are not valid for
human blood vessels. The Moens-Korteweg equation is valid only when an
infinitely long, straight and mechanically as well as geometrically homogeneous
tube whose wall is very thin is filled with a still, non-viscous fluid. In a real
artery, however, the anatomy and the constitution of the blood vessel differ from
place to place, therefore, the mechanical properties of the arterial wall depend on
its regional position [7]. Moreover, the geometry of the blood vessel is not
infinitely long and straight but distributed complicatedly in a three-dimensional
space, including many branches, curved regions, and tapering toward the
periphery. In addition, the blood is not stationary but flows with its velocity
changing in time and space. Hence, the diagnosis of cardiovascular disease by
measuring PWV, which relies on the Moens-Korteweg equation, is not correct in
the strict sense. The measured PWV of the human blood vessel is a result of
several superimposed factors which influence each other, and are not so simple
that the Moens-Korteweg equation can be applied. It is necessary to understand
the wave propagation in blood vessels so as to make the measurements of the
PWV a more reliable diagnosis of vascular disease.
In this study, we modeled uniform arteries in a three-dimensional fully coupled

fluid-solid interaction computational scheme, and analyzed the pulse wave
propagation. Then we estimated the PWV of long uniform arteries to assess the
accuracy of our computation.
METHODS
Pulse wave propagation can be described as an arterial wall disturbance caused by

the ejection of the blood from the heart that propagates mainly toward the
periphery. In this study, we imposed a steady flow as the basic flow perpendicular
to the cross section of a three-dimensional long straight artery, then, a single pulse
was imposed to produce a compression wave, which induced arterial wall
displacement that propagated toward the periphery, and assessed the speed of the
pulse wave propagation. In the uniform artery study, we compared the PWV
values obtained from the computation with those from the Moens-Korteweg
equation.
CHAPTER 18
An-Harmonic Modeling of the Peripheral Distortion of the
Arterial Pulse
Panagiotes A. Voltairas1,*, D.I. Fotiadis1, A. Charalambopoulos1 and L.K.
Michalis2
1
Department of Materials Science and Engineering, University of Ioannina, GR
451 10, Ioannina, Greece and 2Department of Cardiology, Medical School and
Michaelidion Cardiac Center, University of Ioannina, GR 451 10, Ioannina,
Greece
Abstract: As blood circulates through the arterial tree, the flow and pressure pulse distort.
Principal factors to this distortion are reflections from arterial bifurcations and the viscous
character of the blood flow. Both of them are listed and expounded in the literature. Nevertheless,
apart from direct numerical simulations, based on Navier-Stokes like equations, where the
nonlinearities of inertial effects are taken into account, there isn’t any qualitative, as well as
quantitative, analytical formula that explains their role in the distortion of the pulse. We derive an
analytical quasi-linear formula, which emanates from a generalized Bernoulli’s equation for a
linear viscoelastic flow in a quasi-elastic cylindrical vessel. We report that close to the heart (e.g.,
aortic arc), convection effects related to the change in the magnitude of the velocity of blood
dominate the alteration of the shape of the pressure pulse, while at remote sites of the vascular
tree, convection of vorticity, related to the change in the direction of the velocity of blood with
respect to a mean axial flow, prevails. Comparison between the an-harmonic theory and related
pressure measurements is also performed.
Keywords: Arterial pulse, arterial pressure, flow pulse, pressure pulse, large
blood vessels, blood flow, arterial pulse distortions, an-harmonic model,
haemodynamic model, numerical simulations.
INTRODUCTION
Variations in the form of the arterial pressure and flow pulse are often used as
clinical indicators [1, 2]. Therefore, it is important to gain an understanding of the
underlying mechanisms that constitute the arterial pulse in normal physiology.
Each beat of the heart creates a pressure wave which travels along the arterial
*Address correspondence to Panagiotes A. Voltairas: Department of Materials Science and Engineering,

University of Ioannina, GR 451 10, Ioannina, Greece; Email: pvolter@cs.uoi.gr
An-Harmonic Modeling of the Peripheral STP Flows on Chemical and Biomedical Engineering 473
network, changing shape as it moves away from the heart. The amplitude of the
flow wave, or pulse, decreases monotonically with the distance from the heart.
The same applies to the pressure pulse but only at remote locations of the arterial
network. This is due to viscous dissipation, accompanied by the viscoelasticity of
the vessel wall. Nevertheless, initially, the systolic peak of the pressure pulse
increases, as it travels away from the heart. In the arterial periphery, about half a
period after the systolic peak a secondary peak, the diastolic peak, appears.
Another feature of the waveform distortion is that irregularities of the proximal
waveform are smoothed out peripherally (like the dicrotic notch: a notch in the
pressure pulse that marks the closure of the aortic valve). A long-standing
problem is the interpretation of these flow and pressure pulse distortions in terms
of physiological factors (cardiac output, blood’s viscosity, vessel wall elasticity,
mean flow velocity, etc.) and geometrical characteristics of the arterial network
(distance from the heart, vessel diameter, degree of bifurcation of the arterial
network, etc.). The proposed models span areas from Fourier analysis [2] and
solitary pressure waves in elastic tubes [3], to computational fluid dynamics
simulations of local flows [4], and hybrid models of the arterial network, which
frequently use measurements as inlet conditions [5-7].
In a previous work, we proposed an analytical, an-harmonic, one dimensional,

viscous model to account for the above quoted interdependence [8]. Despite the
lack of the observed periodicity, this time-domain approach predicted the viscous
dissipation of the flow and pressure pulse at remote sites of the arterial network
and achieved a quantitative agreement with the systolic peak of the arterial
pressure pulse, but failed to interpret the initial increase of the systolic peak.
The purpose of this study is to extend our previous model [8], in order to interpret
the initial increase of the systolic peak of the pressure pulse as it propagates away
from the heart (and other arterial pulse distortions), through the inclusion of: an
elaborate estimation of all classes of convection effects, a physiologically sound
constitutive model for the rheology of blood and a three dimensional (3D),
analytical, an-harmonic, time-domain, un-lumped [9] model.
METHODS
The complexity of the fluid-structure (blood-arterial wall) interaction is usually

modeled with viscoelastic constitutive laws, either for the blood [4], or for the
474 STP Flows on Chemical and Biomedical Engineering Voltairas et al.
arterial wall [10], under normal [11], or pathological conditions [12, 13]. In this
work we neglect the viscoelasticity of the arterial wall and introduce it indirectly,
through a small radial displacement that meets the analytical needs of the model and
some effective material parameters. Although we study the flow in an infinite
cylindrical tube, reflections from arterial bifurcations are taken into account by
means of the eigenfunction expansion of the derived quasi-linear boundary value
problem.
The 3D Model
Hence, we assume that blood is a non-Newtonian, linear, Maxwell fluid, that
flows in an infinite, quasi-elastic tube of circular cross section. By quasi-elastic
we mean a tube that undergoes small but non-vanishing, radial displacements of
the arterial wall. Thus, we solve the problem of flow in a rigid tube, and introduce
appropriately modified no-slip boundary conditions. The conservation of linear
momentum, the conservation of mass for an incompressible fluid and the
constitutive law for a linear viscoelastic Maxwell fluid are:
 u 
  u  u        P  f , (1)
 t 
  u  0, (2)

    u     u   ,
T
     Tw (3)
t  
respectively.  denotes tensor product,  is the stress tensor,  is the operator

of linear viscoelasticity and Tw is the relaxation time. u, P,  and  are the
blood’s velocity vector, pressure, shear viscosity and density, respectively. We
assume that the blood flow is driven apart from the pressure gradient and from an
additional body force f in equation (1) which originates from the heart impulse
during systole. It is a non-conservative force, of the form:
f   A, A  G  r    z, t  e . (4)
We use cylindrical coordinates  r , , z  .  is a dimensionless, non-separable,

spatiotemporal function. f has the typical form for a polar fluid with rotational
viscosity [14]. The blood flow is also rotational, collinear to f :
CHAPTER 19
Numerical Analysis of Blood Flow in Stenosed Channels
Stéphanie Ferreira1, Ricardo P. Dias2,3,*, Carlos Balsa4 and Carla S.
Fernandes4
1
Escola Superior de Tecnologia e Gestão de Bragança, Campus de Santa
Apolónia, 5301-857 Bragança, Portugal; 2Departamento de Tecnologia Química
e Biológica, Escola Superior de Tecnologia e Gestão de Bragança, Campus de
Santa Apolónia, 5301-857 Bragança, Portugal; 3CEFT-Centro de Estudos de
Fenómenos de Transporte, Faculdade de Engenharia da Universidade do Porto,
4200-465 Porto, Portugal and 4Departamento de Matemática, Escola Superior de
Tecnologia e Gestão de Bragança, Campus de Santa Apolónia, 5301-857
Bragança, Portugal
Abstract: Wall shear rates and pressure developed in blood vessels play an important role on the
development of some clinical problems such as atherosclerosis and thrombosis. In the present
work, blood flow behaviour was numerically studied in simplified domains and several
relevant local properties were determined. We believe that the obtained results will be
useful in the interpretation of some phenomena associated to some clinical problems. To
describe the rheological behaviour of blood, three constitutive equations were used-
constant viscosity, power-law and Carreau model. Numerical predictions for the blood
flow in stenosed channels were in good agreement with analytical results, indicating
that the computational model used to describe the studied problem is reliable. Pressure
attains maximum values close to the top of the atheroma and shear rates achieved
maximum values at the walls located in the nearby of the atheroma. It was also
observed that, with the studied flows, the impact of the non-Newtonian behaviour of the
blood on the velocity profiles was not significant. This observation can be explained by
the magnitude of the obtained shear rates.
Keywords: Blood, atheroma, power-law model, carreau model, newtonian fluid,

computational fluid dynamics, velocity, shear rate, pressure, fanning friction factor
INTRODUCTION
The survival of a complex organism with large dimensions, like the human
*Address correspondence to Ricardo P. Dias: Departamento de Tecnologia Química e Biológica, Escola

Superior de Tecnologia e Gestão de Bragança, Campus de Santa Apolónia, 5301-857 Bragança, Portugal;
E-mail:ricardod@ipb.pt
490 STP Flows on Chemical and Biomedical Engineering Ferreira et al.
organism, requires a system prepared to transport several substances for and from
every organism’s cells [1-3]. This is the principal function of the circulatory
system. In this system, the blood is an aqueous vehicle responsible for the
transportation of nutrients, oxygen, chemical mediators, antibodies, defense cells
and toxic substances [1], maintaining the ideal environment for the survival and
function of the entire organism cells [3].
Blood
The blood is an opaque and heterogeneous fluid composed by a yellowish fluid-
the plasma-and a series of cellular elements [1]. The blood is contained in a closed
compartment-the circulatory system-that keeps it in motion in a unidirectional
stream [4]. In an adult, the volume of blood is about 7% of the body weight or
about 5 liters (in an adult with about 70 Kg of body weight) of which almost 60%
is plasma [5].
The blood cells have a short lifetime in the blood flow. Therefore, they are
constantly replaced for new cells [4, 5]. The blood cell formation process is
designated as hemocitopoiesis or hematopoiesis [4].
The steam cell is rare, less than one in every 10000 nucleated cells of the bone
marrow. The mitotic division of the steam cell originates two daughter cells, one
remains as steam cell and the other will pursue for one of two differentiation
ways: the lymphoid way, which gives origin to lymphocytes, and the myeloid way
that gives origin to the rest of the blood cells [3, 4, 6].
Blood Rheology
Inside de normal circulatory flow, blood behaves as a non-Newtonian fluid. As a
first approach, blood viscosity can be determined by the viscosity of plasma and
internal friction between plasma and the cells in suspension (solution-gel) [7].
Four factors influence the viscosity: hematocrit, temperature, velocity and

diameter of the blood vessel, as it will be described below.
The hematocrit increase leads to the increase of the viscosity (Fig. 1). When the
hematocrit overcomes 80%, blood cannot be considered a fluid but a gel. For the
Numerical Analysis of Blood Flow STP Flows on Chemical and Biomedical Engineering 491
referred levels of hematocrit, the erythrocytes become aggregated in a way that

the blood flows with difficulty and, consequently, the cardiac work is huge [8].
Figure 1: Schematic representation of the relation blood viscosity vs. hematocrit.
Under normal conditions, temperature (Fig. 2) does not play an important role in
the blood viscosity. However, in in vitro experiments, surgeries requiring
circulation outside the body and hypothermic conditions, there is an increase of
the blood viscosity [8].
Figure 2: Schematic representation of the relation blood viscosity vs. temperature.
Since blood is not a Newtonian fluid, blood viscosity varies with the flow velocity,
as presented in Fig. 3. The blood only flows when a certain tangential force is
exceeded and, after this force, presents a pseudoplastic behaviour, i.e., its viscosity
decrease with the increase of the velocity, until a certain limit is reached [8].
Johnston et al. [9] explored the rheology of blood during the flow in arteries and
concluded that the non-Newtonian behaviour of the blood could be well described
CHAPTER 20
Blood Flow Behavior in Microchannels: Past, Current and Future
Trends
R. Lima1,2,*, T. Ishikawa3, Y. Imai3 and T. Yamaguchi4
1
DTM/ESTiG, Braganca Polytechnic, C. Sta. Apolonia, 5301-857 Bragança,
Portugal; 2CEFT, Engineering Faculty, Porto University, Rua Dr. Roberto Frias,
4200-465 Porto, Portugal; 3Department of Bioengineering & Robotics, Graduate.
School of Engineering, Tohoku University, 6-6-01 Aoba, 980-8579 Sendai, Japan
and 4Department of Biomedical Engineering, Graduate School of Biomedical
Engineering, Tohoku University, 6-6-01 Aoba, 980-8579 Sendai, Japan
Abstract: Over the years, various experimental methods have been applied in an effort
to understand the blood flow behavior in microcirculation. Most of our current
knowledge in microcirculation is based on macroscopic flow phenomena such as
Fahraeus effect and Fahraeus-Linqvist effect. The development of optical experimental
techniques has contributed to obtain possible explanations on the way the blood flows
through microvessels. Although the past results have been encouraging, detailed studies
on blood flow behavior at a microscopic level have been limited by several factors such
as poor spatial resolution, difficulty to obtain accurate measurements at such small
scales, optical errors arisen from walls of the microvessels, high concentration of blood
cells, and difficulty in visualization of results due to insufficient computing power and
absence of reliable image analysis techniques. However, in recent years, due to
advances in computers, optics, and digital image processing techniques, it has become
possible to combine a conventional particle image velocimetry (PIV) system with an
inverted microscope and consequently improve both spatial and temporal resolution.
The present review outlines the most relevant studies on the flow properties of blood at
a microscale level by using past video-based methods and current micro-PIV and
confocal micro-PIV techniques. Additionally the most recent computational fluid
dynamics studies on microscale hemodynamics are also reviewed.
Keywords: In vitro blood, microcirculation, blood flow, red blood cells,

microchannel, confocal micro-PIV/PTV, blood rheology, cell-free layer,
computational fluid dynamics, multiscale blood flow, multiphase flow, numerical
model, microscale blood flow, boundary element method, particle method.
*Address correspondence to R. Lima: DTM/ESTiG, Braganca Polytechnic, C. Sta. Apolonia, 5301-857

Bragança, Portugal; E-mail: ruimec@ipb.pt
514 STP Flows on Chemical and Biomedical Engineering Lima et al.
BLOOD FLOW BEHAVIOR IN MICROCHANNELS
Blood in large arteries may be treated as a homogenous fluid from a macroscopic

perspective. However, in reality blood is a suspension of deformable cells in a
viscous fluid plasma. In microcirculation, which comprises the smallest arteries
and veins, the flow behavior of individual blood cells and their interactions
provide the microrheological basis of flow properties of blood at a macroscopic
level. As a result, in microcirculation it is fundamental to study the flow behavior
of blood at cellular level.
Haemodynamics of the Circulatory System

The primary function of the circulating blood is to transport oxygen, nutrients,
waste products and heat throughout the body. The oxygenated blood leaves the
heart, with high pressure and velocity (Reynolds number (Re) > 2000), through
the aorta and moves into other large arteries, then into arterioles and, finally, into
the capillaries. On its way back from the capillaries the blood moves into venules
and then into veins in order to return to the heart and to begin the same journey all
over again (see Fig. 1a).
Figure 1: a) Schematic diagram of the circulatory system. b) Schematic diagram showing the
changes in blood pressure and Re on the large arteries, capillaries and veins [1, 4, 15].
Blood Flow Behavior in Microchannels STP Flows on Chemical and Biomedical Engineering 515
In large arteries, where the diameter of the blood vessels is large enough
compared to individual cells, it has been proved adequate to consider blood as a
single-phase fluid [1-3]. Accordingly, blood in large arteries may be treated as a
homogeneous fluid (continuous viscous fluid) where its particulate nature is
ignored. Moreover, due to the large Reynolds number (Re) in arteries, blood flow
is governed by inertial forces. However, arteries divide into successive smaller
arteries and consequently the cross-sectional area of the vascular bed increases.
As a result both pressure and velocity decrease as the blood flows into the smaller
vessels. When the blood reaches the arterioles and capillaries the Re became less
than 1, where viscous force dominates over inertial forces (see diagram in Fig.
1b). At this microscale it is fundamental to take into account the effects of the
multiphase properties of the blood on its flow behavior [1, 2, 4]. A clear example
of the multiphase nature of the blood, illustrated in Fig. 1a, is the formation of a
plasma layer at microvessels less than 300 μm, known as Fahraeus-Lindqvist
effect [5]. It is based on the Fahraeus-Lindqvist effect we define microcirculation
as the one including all microvessels or microchannels with a diameter smaller
than 300 μm.
Blood Composition
Blood is not a homogeneous fluid, but one composed of a suspension of cells,
proteins and ions in plasma. In normal blood, three types of cells comprise about
46% of its volume (see Table 1). These cells are the red blood cells (also known
as erythrocytes), white blood cells (also known as leukocytes) and platelets (also
known as thrombocytes) (see Fig. 2).
Figure 2: Scanning electron microscope image of human blood cells (adapted from [6]).
CHAPTER 21
A Survey of Microchannel Geometries for Mixing of Species in
Biomicrofluidics
Francesco Pennella, Francesco Mastrangelo, Diego Gallo, Diana Massai,
Marco A. Deriu, Giuseppe Falvo D'Urso Labate, Cristina Bignardi, Franco
Montevecchi and Umberto Morbiducci*
Department of Mechanics, Politecnico di Torino, 24 Corso Duca degli Abruzzi,
Turin, 10129, Italy
Abstract: Microfluidics is changing the way modern biology is performed and is becoming
a key technology in the field of micro arrays, DNA sequencing, and Lab on a Chip
applications. Microsystems, being compact in size, disposable, and ensuring high speed of
analysis using decreased sample volumes, allow to replace large-scale conventional
laboratory instrumentation with miniaturized devices, reducing hardware costs, and
assuring low reagent consumption and faster analysis. At the microscale mixing of species
becomes crucial to i) improve the effectiveness of and ii) speed up chemical reactions, but
it is often critical to be achieved, since microfluidics is characterized mainly by very low
Reynolds flows, and cannot take advantage of turbulence in order to enhance mixing.
Hence, given that diffusion-driven mixing in very low Reynolds number flow regimes is
characterized by long time scales, methods for enhancing the rate of the mixing process are
essential in microfluidics. In order to enhance mixing, several techniques have been
developed. In general, mixing strategies can be classified as either active or passive,
according to the operational mechanism. Active mixers employ external forces in order to
perform mixing, so that actuation system must be embedded into the microchips. On the
contrary, passive mixers avoid resorting to external electrical or mechanical sources by
exploiting characteristics of specific flow fields in microchannel geometries to mix species,
offering the advantage to be easy to be produced and integrated. In this work, a survey of
the passive micromixing solutions currently adopted is presented. In detail, the most widely
used microchannel geometries and the metrics used to quantify mixing effectiveness in
microfluidic applications are discussed.
Keywords: Microfluidics, microchannel, mixing, active micromixers, passive,

micromixers, lamination, split-and-recombine, chaotic advection, dean, vortices,
mixing metrics, coloured particle tracking.
INTRODUCTION
Microfluidics is made of “devices and methods for controlling and manipulating
*Address correspondence to Umberto Morbiducci: Department of Mechanics, Politecnico di Torino, 24

Corso Duca degli Abruzzi, Turin, 10129, Italy; E-mail: umberto.morbiducci@polito.it
A Survey of Microchannel Geometries STP Flows on Chemical and Biomedical Engineering 549
fluid flows with length scales less than a millimetre” [1]. The essence of
microfluidics is to use streams of fluids (e.g., channel-based biochips) or droplets
(e.g., digital microfluidics) as carriers of biochemical samples to carry out pre-
determined sample manipulation procedures. As microfluidics is massively
applied as operating principle in biochips, and a large amount of studies dealing
with microfluidics aim at improving biochip design and development, the term
"microfluidics" is regarded as interchangeable with biochip, Lab-on-a-Chip
(LoC), or Micro-Total-Analysis Systems (µTAS) [2]. Microfluidics is achieving a
position of significance as an interdisciplinary technology with applications in
many important areas such as biosensing, diagnostics and drug discovery. In fact,
microfluidics is becoming an effective and reliable tool in cell biology [3],
biochemistry [4], neuroscience [5], drug testing [6], and other areas of biology
and biomedical engineering. The enormous attraction towards microfluidics
results from its capability to achieve significant reduction in reagent volumes, in
performance time, and in power consumption while allowing massive parallelism.
In a microfluidic system, one of the most important components is the
micromixer. In fact, microfluidic mixing is the central process of most
microfluidic devices for medical diagnostics, genetic sequencing, chemistry
production, drug discovery, and proteomics. Biological processes such as cell
activation, enzyme reactions and protein folding often involve mixing of reactants
for initiation [1]. Mixing is also necessary in LoC platforms for complex chemical
reactions [7]. Micromixers can be integrated in a microfluidic system or work as
stand-alone devices. Therefore, the investigation of micromixers is fundamental
for comprehending transport phenomena on the microscale.
Here we present a review of the micromixing solutions currently adopted. After a

brief introduction, mixing principles and the key mass transport phenomena will
be shown. Then, different concepts and designs for active and mainly passive
micromixers are analyzed. In the end, the metrics used to quantify mixing
effectiveness in microfluidic applications are discussed.
MIXING PRINCIPLES AND MIXING STRATEGIES
In microfluidics, the small size of the conduits, the low flow regimes, and
viscosity of fluids are typically predominant factors in establishing local flow
550 STP Flows on Chemical and Biomedical Engineering Pennella et al.
fields where convective mixing is difficult to be achieved. The relative influence

of inertial and viscous forces can be quantified by the Reynolds number:
V Lch
Re 
 (1)
where V is the mean velocity in the flow direction, Lch is the characteristic length
of the channel, (typically the diameter of the microchannel, in microfluidics), and
υ is the kinematic viscosity of the fluid. At microscale, seeking to rise Reynolds
number is very energy consuming: in fact, this non-dimensional number at the
microfluidic scale generally varies in a range going from less than unity up to a
few hundred, well within the range of laminar flow, so the efficient mixing
obtained with turbulence is, in most cases, not achievable in the common practice.
Hence, three are the main transport phenomena involved in microfluidics:
molecular diffusion, advection and Taylor dispersion. Molecular diffusion is the
predominant driving force for any mixing to occur, and it is the consequence of
the random motion of molecules. This mechanism of transport can be
synthetically expressed by means of the molecular diffusion coefficient, which is
a physical feature of the species and the liquid (i.e., diffusion coefficient of a
molecule depends on the viscosity of solute). Another nondimensional quantity,
the Péclet number (Pe), is often used to describe flows behaviour at the
microscale:
V Lch
Pe 
D (2)
where V is the mean velocity in the flow direction, Lch is the characteristic length
of the channel and D is the molecular diffusivity. From a physical viewpoint, the
Péclet number is the ratio between mass transport due to convection and the one
due to transverse diffusion. High values of the Péclet number indicate that
diffusive mixing occurs at a much slower rate than the timescales associated with
fluid motion. Typically, in microchannels Péclet numbers are relatively high in
(Pe > 100). This is like to say that diffusive and convective transport phenomena
combine to produce characteristic mixing lengths Δym:
CHAPTER 22
Endothelial Cell Responses to Fluid Shear Stress: From
Methodology to Applications
Toshiro Ohashi1,* and Masaaki Sato2
1
Faculty of Engineering, Hokkaido University, Kita 13, Nishi 8, Kita-ku, Sapporo,
Hokkaido, 0608628, Japan and 2Graduate School of Biomedical Engineering,
Tohoku University, Aramaki Aza, Aoba, 6-6-01, Sendai 980-8579, Japan
Abstract: Vascular endothelial cells are constantly stimulated by blood flow-induced

shear stress throughout the vasculature and respond by changing morphology and
cytoskeletal structures as well as by modulating cell physiological functions. In
particular, since endothelial cell responses to fluid shear stress have been implicated in
the localization of atherosclerosis, the effects of fluid shear stress on endothelial cell
morphology and functions have been exclusively studied. In fact, previous observations
have given preferential localization of lipid accumulation in atherosclerosis in the
arterial tree, such as branching and curved regions where blood flow is unsteady and
spatially and temporally altered. So far, a lot of efforts have been made to study
endothelial mechanotransduction to flow, indicating the fact that after applying fluid
shear stress to endothelial cell monolayer, cells exhibit marked elongation and
orientation in the direction of flow. It is now accepted that morphological changes of
endothelial cells are closely associated with modulation of cell physiology and
pathology. The need for experimental techniques for studying endothelial responses to
flow has lead to development of different types of flow chambers. Conventional flow
chambers include a cone-and-plate flow chamber and a parallel-plate flow chamber,
both of which provide steady, fully-developed laminar flow. Furthermore, in order to
provide non-steady flow and disturbed flow, novel flow chambers have been developed,
such as a tapered channel and an obstacle-included channel. More recently, microfluidic
flow chambers have emerged with a great potential for a high throughput analysis. The
purpose of this review is to first summarize many types of flow chambers that apply
fluid shear stress onto a monolayer of endothelial cells. Next, experimental studies on
endothelial cell responses to fluid shear stress are highlighted, focusing on changes in
cell morphology associated with cytoskeletal remodeling.
Keywords: Vascular endothelial cells, hemodynamic forces, fluid shear stress,

endothelial cell remodeling, mechanotransduction, cytoskeletal structures,
atherosclerosis, fluid flow-imposed systems, parallel-plate flow chambers,
microfluidic technologies, microchannels, lab-on-a-chip.
*Address correspondence to Toshiro Ohashi: Faculty of Engineering, Hokkaido University, Kita 13, Nishi
8, Kita-ku, Sapporo, Hokkaido, 0608628, Japan; E-mail: ohashi@eng.hokudai.ac.jp
580 STP Flows on Chemical and Biomedical Engineering Ohashi and Sato
INTRODUCTION
Vascular endothelium is a monolayer of endothelial cells forming an interface

between blood vessels and blood flow. It is well known that endothelial cells
sense and respond to shear stress generated by blood flow with alterations in
morphology [1-5], cytoskeletal structures [6, 7], and cellular functions including
proliferation [8], migration [9, 10], differentiation [11, 12], secretion [13, 14],
adhesion [15], and permeability [16-19]. In fact, endothelial cells convert external
mechanical forces into biochemical signals through a process termed
mechanotransduction [20-22]. Notably, these endothelial cell responses have
crucial implications for physiology and pathology of blood vessel walls. In
particular, endothelial cell responses to fluid shear stress have long been
implicated in the localization of atherosclerosis.
Atherosclerosis is generally characterized by the formation of atheroma on the

intima, which occurs primarily at branching and bends of the arterial walls. In the
preferential regions of atherosclerosis, endothelial cells would experience
complex blood flow such as boundary layer separation and secondary flow.
Therefore, atherosclerosis researches have been exclusively dedicated to the
effects of fluid shear stress on endothelial cell responses, including changes in
morphology and cytoskeletal structures, over the last 30 years. So far, a lot of
efforts have been made by many researchers to study the correlation between
flow-induced shear stress and morphology of endothelial cells. It is widely
accepted that when endothelial cells are exposed to fluid shear stress, they become
elongated and aligned in the direction of flow [1-3]. As mentioned above,
morphological changes of endothelial cells are known to influence several aspects
of endothelial cell biology that are critical to normal endothelium functions [22].
In order to apply well-controlled fluid flow to monolayer of cells, specially

designed in vitro flow chambers have been developed. Conventional flow
chambers typically include a cone-and-plate flow chamber and a parallel-plate
flow chamber [3, 23]. These chambers are suitable for applying steady, laminar
flow to a monolayer of cells cultured on a bottom plate. Followed by the
development of these conventional systems, novel flow chambers have been
developed to provide flows with temporal and spatial gradients of shear stress and
Endothelial Cell Responses to Fluid Shear STP Flows on Chemical and Biomedical Engineering 581
disturbed flows [24-27] because blood flow in arteries is essentially unsteady,

pulsatile and, in particular, disturbed at the branching and curved sites. More
recently, microfluidic technologies have been exclusively used to miniaturize such
conventional flow systems, where a variety of microchannels are fabricated [28-
32]. The advantages of microchannels are to realize high-throughput analysis, low
consumption of samples, low consumption of expensive reagents and low
fabrication cost.
Studies on endothelial cell mechanotransduction by using flow chambers have

been providing a wide range of new insights into the underlying mechanism of
atherosclerosis, which has drawn a great attention of not only medical researchers
but also biomedical engineering researchers. Furthermore, recent advances in
developing microfluidic devices are expected to attract more micromachining
researchers. In this review, different types of flow chambers that apply flow-
induced shear stress onto a monolayer of endothelial cells are first summarized,
including from conventional flow chambers to recently reported microfluidic flow
chambers. Next, experimental studies on endothelial cell responses to fluid flow
are described, highlighting changes in morphology and cytoskeletal structures.
VASCULAR ENDOTHELIAL CELLS AND MECHANOTRANSDUCTION
Vascular endothelial cells line the interior surface of blood vessels and form a
monolayer serving as a barrier between the vessels and blood flow (Fig. 1(a))
[33]. Endothelial cells regulate multiple physiological functions such as vascular
tone, thrombosis and coagulation, smooth muscle cell growth and migration,
vascular permeability, etc. [34] and therefore endothelial dysfunction may have a
crucial role in the pathogenesis of atherosclerosis. Hemodynamic forces
constantly exerted on endothelial cell monolayer are directly resolved into two
components such as shear stress and hydrostatic pressure, and indirectly induce
circumferential stretch due to wall motion. It has been well recognized that these
hemodynamic forces play a central role in physiology and pathology of
endothelial cells. A typical flow-induced regulation is that shear stress-mediated
nitric oxide (NO) production by endothelial cells induces smooth muscle
relaxation, thereby controlling vascular tone.
CHAPTER 23
Micro-Flow Visualization of Magnetic Nanoparticles for
Biomedical Applications
R. Lima1,2,*, R.J. Joseyphus 3,‡, T. Ishikawa4, Y. Imai4, and T. Yamaguchi5
1
Department of Mechanical Technology, ESTiG, Braganca Polytechnic, C. Sta.
Apolonia, 5301-857 Bragança, Portugal; 2CEFT, Engineering Faculty, Porto
University, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal; 3Department of
Physics, National Institute of Technology, Tiruchirappalli 620 015, India;
4
Department of Bioengineering & Robotics, Grad. Sch. Eng., Tohoku University,
6-6-01 Aoba, 980-8579 Sendai, Japan and 5Department of Biomedical Eng.,
Grad. Biomedical Sch. Eng., Tohoku University, 6-6-01 Aoba, 980-8579 Sendai,
Japan
Abstract: An investigation to measure the flow behavior of magnetic nanoparticles through

a 100m microchannel is conducted. The magnetic field is applied externally by a
permanent magnet and by using a micro-PTV system it was possible to measure the flow
behavior of magnetic nanoparticles at different flow rates and magnetic fields through a
100m glass capillary. A strong dependence on both magnetic and hydrodynamic force is
observed on the nanoparticles fluidic paths. Based on these in vitro studies, important
parameters and issues that require further understanding and investigation are point out.
Keywords: Micro/nano flow, magnetic nanoparticles, superparamagnetism, Fe-

oxide layer, hyperthermia, microchannel, micro-PIV/PTV, glass capillary,
microcirculation, blood, physiological fluid, drug delivery, cancer therapy,
bioseparation, MRI contrast agent.
INTRODUCTION
Magnetic nanoparticles are used in a variety of biomedical applications such as

hyperthermia, drug delivery, bioseparation and MRI contrast agents [1-3]. Drug
delivery using magnetic nanoparticles is a more promising application since it is a
noninvasive technique where high efficiency could be achieved. The advantage
*Address correspondence to R. Lima: Department of Mechanical Technology, ESTiG, Bragança

Polytechnic, C. Sta. Apolonia, 5301-857 Bragança, Portugal; E-mail: ruimec@ipb.pt
‡
E-mail: rjustinj(at)nitt.edu
Micro-Flow Visualization of Magnetic STP Flows on Chemical and Biomedical Engineering 601
using magnetic nanoparticles in drug delivery is that it could be used for localized
drug delivery whereas the conventional drug delivery methods would result in drug
associated side effects. The magnetic nanoparticles for biomedical applications are
usually coated with a biocompatible surfactant layer which prolongs the circulation
time of the nanoparticles in the blood. The size of the magnetic nanoparticles should
be less than 100 nm to prevent its capture by mononuclear phagocyte system [4].
Though magnetic nanoparticles of sizes below 50 nm could be easily synthesized
and dispersed, the polymeric coating would render the particle size to more than 50
nm [5]. Hence magnetic nanoparticles of smaller sizes are preferred. When the size
of the magnetic nanoparticles becomes smaller, thermal relaxation would be
dominant and result in superparamagnetism. Superparamagnetism is a phenomenon
arising due to the dominance of thermal energy over anisotropy energy (kT>KV).
The thermal energy and anisotropy energy are given by kT and KV respectively,
where k is the Boltzmann constant, T is the temperature, K is the magnetocrystalline
anisotropy constant and V is the volume of the particle. The superparamagnetic
particles would exhibit close to zero coercivity and hence could be controlled easily
by external magnetic field. However the superparamagnetic nanoparticles are hard to
saturate and hence it requires sufficiently larger field. Thus it would be advantageous
to use magnetic materials which show larger saturation magnetization even at low
magnetic field. Fig. 1(a) shows the typical hysteresis loop at room temperature for
magnetic materials with high K and Fig. 1(b) shows the hysteresis loop for a
superparamagnetic material. The difference in magnetic parameters such as
coercivity and saturation magnetization is obvious from the Fig. 1.
Although smaller magnetic particles are preferable for biomedical applications,

for applications such as drug delivery, the particles should be able to respond to
an external magnetic field even at a larger separation. This requires that the
magnetic nanoparticles should exhibit very large saturation magnetization. FeCo
and Fe are the magnetic materials with very large saturation magnetization of 240
and 218 emu/g respectively [6]. Although FeCo has higher saturation
magnetization, the biocompatibility and surface modification are difficult to
achieve. On the other hand Fe particles of very high magnetization could be easily
synthesized through chemical methods [7]. The biocompatibility could be
achieved by a Fe-oxide layer over the Fe particles [8] as the biocompatibility of
602 STP Flows on Chemical and Biomedical Engineering Lima et al.
Fe-oxides is well known. However when the particle size becomes smaller, the Fe
particles naturally undergoes oxidation and the large surface area of smaller
particles with Fe-oxide layer would reduce the saturation magnetization. This is
illustrated in Fig.2 where larger Fe particles would show higher magnetization.
(a)
Magnetization (a.u)
-10 -5 0 5 10
Applied field (kOe)
0.8
0.6
0.4
Magnetization (a.u)
0.2
(b)
0.0
-0.2
-0.4
-0.6
-0.8
-10 -5 0 5 10
Applied Field (kOe)
Figure 1: (a) Typical hysteresis loop at room temperature for a high K magnetic material and (b)
hysteresis loop for a superaparamagnetic material. The superaparamagnetic material shows
negligible coercivity.
CHAPTER 24
A Computational Study on the Possibility of the Initialization and
Development of Intracranial Aneurysms Considering Biofluid and
Biosolid Mechanics
Yixiang Feng1,*, Shigeo Wada2 and Takami Yamaguchi3
1
Mechanical Engineering Research Laboratory, Hitachi Ltd, Hitachinaka, Ibaraki
312-0034, Japan; 2Department of Mechanical Science and Bioengineering,
Graduate School of Engineering Science, Osaka University, Japan and
3
Department of Biomedical Engineering, Graduate School of Biomedical
Engineering Tohoku University, 6-6-01 Aoba-yama, Sendai 980-8579, Japan
Abstract: Intracranial aneurysms are local dilations of the arterial wall which have a
very high morbidity rate if they rupture. Although the mechanism initiation, growth, and
rupture of intracranial aneurysms are still unknown yet, it is believed to be closely
related to both biosolid and biofluid mechanics. Therefore, a multi-physical model is
needed to study the pathophysiology of intracranial aneurysms. In this study, we
introduce a numerical model on the development of intracranial aneurysms considering
the interaction between fluid and structure interaction. The blood flow is considered to
be incompressible, Newtonian, and laminar. The vessel wall is considered to be elastic
and isotropic. The coupling between the structural and fluid domain is performed using
a two-way weak coupling method. Three general shapes are adopted in this study,
namely a straight vessel, a curved vessel, and a vessel with bifurcations. They represent
vessel geometries that are most typical to the cerebral vasculature. The numerical model
is a "rule-base" one in a sense that different kinds of rules can be tested. In our study,
we adopt the high wall shear stress hypothesis as a cause for aneurysm initiation and
development. A threshold value is used for the wall shear stress. It is shown that
aneurysm initiation and development can be realized using our numerical model. And
the influence of WSS threshold, the Reynolds number and some other parameters are
also discussed.
Keywords: Intracranial aneurysms, arterial wall, biosolid and biofluid mechanics,

numerical model, fluid and structure interaction, straight vessel, curved vessel,
vessel with bifurcations, cerebral vasculature, wall shear stress, aneurysm
initiation and development, blood flow.
*Address correspondence to Yixiang Feng: Mechanical Engineering Research Laboratory, Hitachi Ltd,
Hitachinaka, Ibaraki 312-0034, Japan; E-mail: yixiangf@gmail.com
614 STP Flows on Chemical and Biomedical Engineering Feng et al.
INTRODUCTION
Intracranial aneurysms (IAs) are characterized by abnormal ballooning of the cerebral

arteries at weaker parts of the arterial wall. The danger of an intracranial aneurysm is
when it continues to grow and ruptures, causing subarachnoid hemorrhage (SAH),
which has a fatality rate from 32% to 67% [1]. Unfortunately, despite recent progress
in the research of intracranial aneurysms, the mechanism leading to their initiation
and further development remains unclear. Hemodynamic stresses are believed to play
a vital role in the pathophysiology of IAs [2-6]. Many experimental and numerical
studies have been performed mainly to elucidate the flow field and hemodynamics
inside an aneurysm [7-10]. These studies were successful in a sense that the
hemodynamic parameters can now be obtained in most complex geometries at
patient-specific level. However, most of these studies were still based on static
models in which aneurysms have already formed. Very few studies focus on the
dynamic progress of how aneurysms initiate and how they grow up [11].
In contrast to the well-known “image-based” computer simulation, we propose a

“rule-based” computer simulation in this study. In “rule-based” simulations, the
emphasis is not to obtain the hemodynamic parameters in a specific geometry, but
rather to test and verify various hypotheses on how the vascular disease is
developed [12, 13]. Since many researches have investigated hemodynamics in
order to clarify the mechanism leading to the initiation and development of
aneurysms, it should be worthwhile for providing the rule-based computer
simulation as a novel tool to test and verify various hypotheses.
While most previous studies on aneurismal hemodynamics adopted single field

CFD approaches, a coupled fluid-structure interaction (FSI) analysis is necessary
to capture the interactive features during the development of aneurysms. Coupled
FSI algorithms can be classified as either monolithic (implicit) or partitioned
(explicit). With the implicit method the fluid and structure domains are integrated
into a single arbitrary Lagrangian-Eulerian space. With the explicit method the
two domains are modeled and solved separately. Load and displacement are
updated once or several times per time step at the wetted surface. In this study, the
explicit approach is chosen for the design of FSI, because it permits the use and
customization of familiar matured solvers for each domain.
A Computational Study on the Possibility STP Flows on Chemical and Biomedical Engineering 615
Among several theories regarding the genesis and progression of intracranial

aneurysms, we regard that the so-called “vicious cycle”, i.e., the interactive nature
of hemodynamic stresses, structural defects, and morphological changes is most
likely to be responsible for the development of aneurysms [14, 15]. Since the
growth of aneurysm modifies the distribution of the blood flow field and
hemodynamic stresses [16], which may cause more degenerative changes to the
arterial wall and result in further progression of the aneurysm, we argue that this
repeated interactive procedure is a key to understand the hemodynamics-induced
progression of the disease.
Various degenerative changes in the arterial wall structure have been observed
and are supposed to be responsible for the initiation of intracranial aneurysms [17-
19]. These include mechanical fatigue, occlusion of the vasa vasorum [20],
existence of media gaps [21], apoptosis of medial smooth muscle cells (SMCs)
[14], reduced production of collagen [19, 22, 23], and enzymatic digestion [24].
Although the exact cause for the degeneration of the arterial wall has not been
clarified yet, it is accepted that hemodynamic forces alone are able to cause
degeneration of the vascular structure. Among various hemodynamic factors, high
wall shear stress (WSS) has been proposed to be the candidate responsible for
aneurysm initiation [3, 8]. Krings and colleagues [25] have reported a method for
creating saccular aneurysms in rabbit in areas of high WSS. They also observed
loss of internal elastic laminar in the high WSS region. Kim et al. [26] also
reported that sites of endothelial injuries corresponds well to areas of high WSS.
All these experimental studies point to the conclusion that exaggerated WSS can
degenerate the arterial wall and give rise to the initiation of intracranial
aneurysms.
Although we emphasize on the acquired hemodynamic origin of aneurysm

progression in this study, we admit that hemodynamic stresses alone are not likely
to be the only cause for the initiation and development of intracranial aneurysms.
Pre-existing arterial diseases in elderly patients, or genetic factors, particularly in
younger patients, have also been reported to be of importance [27]. Actually, there
is no difficulty in incorporating these pre-existing factors into our model. We did
apply some pre-existing deterioration in the local wall to trigger the initiation of
Index
2D modeling 332-360
2D Momentum Model 341
3D images 292-331
Activated sludge 153-174

Activated Sludge Model (ASM) 155, 156
Activated sludge models 153-174, 154, 164
Active micromixers 548-578
Adsorption 33, 332-360
Ammonia 26, 27
Aneurysm initiation and development 613-633
An-harmonic model 472-488
Anisotropic Interfaces 368
Arbitrary Lagrangian Eulerian 460-471
Arrhenius Law 129
Arterial pressure 472-488
Arterial pulse 472-488
Arterial pulse distortions 472-488
Arterial wall 613-633
Arterial wall stiffness 460-471
Atheroma 489-512
Atherosclerosis 579-599
Autothermal reactor 26-51, 32
Average Chemical Reaction Rate 128, 148
Averaged reynolds navier-stokes equations 153-174, 160
Axial dispersion 386-411, 388, 389, 398, 401
Backmixing model 90
Index STP Flows on Chemical and Biomedical Engineering 635
Biochemical reaction 153

Biological Model 155
Bioseparation 600-612
Biosolid and biofluid mechanics 613-633
Blood 489-512, 600-612
Blood flow 460-471, 472-488, 513-547, 613-633
Blood rheology 513-547
Boundary Conditions in Lattice Boltzmann Simulations 67, 69
Boundary element method 513-547
Brownian Motion 52
Bubble Column Reactor 413
Bubble Columns 386, 412-439
Bubble size 413, 415
Bubbles flow rate 440-459
Bubbling Bed Energy 441
Cancer therapy 600-612

Carreau model 489-512
Catalysis 4, 20, 26, 176, 333
Catalyst Effectiveness 184
Catalyst reduction 332-360
Catalytic Converter 4
Cell-free layer 513-547
Cerebral vasculature 613-633
CFD 153-174, 159
Chaotic advection 548-578
Chaotic Mixing 79
Chemical Mapping 12
Chemical Methods Measurement Degree of Mixing 82
Chemical reaction 117
Chemical Reactor 4, 26, 31, 79, 332-360, 387
636 STP Flows on Chemical and Biomedical Engineering Dias et al.
Chemical Reactor Modeling 158, 79, 179

Churn turbulent 412-439, 415
Coalescence Re-Dispersion Model of Mixing 86
COD 155
Coloured particle tracking 548-578
Compartmental approach 153-174
Complete Segregation 81
Computational Fluid Dynamics (CFD) 10, 79, 174-195, 177, 179, 233-265, 235,
489-512, 513-547
Concentration Measurements 349
Concentration Profile 196-232, 198, 205, 212, 217, 223, 355
Confocal micro-PIV/PTV 513-547
Constructal design 266-291
Constructal theory 266-291, 270, 271, 287
Convection 197
CUBISTA 200
Curved vessel 613-633
Cylinder 196-232, 205
Cylinder in cross flow 196-232, 212
Cytoskeletal structures 579-599
Damköhler number 113-152, 127

Darcy Law 209, 298
Dean 548-578
Deformation rate 237
Degree of Chaos 412-439, 417
Degree of Mixing 82
Dendritic 266-291
Design in nature 266-291
DEVSS 233-265, 238
Die swell 233-265, 234, 258

Two Phase Flows Sample

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Two Phase Flows Sample

Transféré par

Droits d'auteur :

Formats disponibles

Single and Two-Phase Flows on Chemical

and Biomedical Engineering

CEFT - Transport Phenomena Research Center,

CEFT - Transport Phenomena Research Center

Permission for Use of Material and Reproduction

List of Contributors iii

Part I: Single Phase Flows on Chemical Enginnering

1. Spacims-Probing the Internal Behaviour of 3D Structured

2. Modelling of a Monolithic Reverse Flow Reactor for Selective

3. Mesoscopic Simulation of Rarefied Gas Flow in Porous Media 52

4. Mixing Through Half a Century of Chemical Engineering 79

5. Application of the Probability Density Function Method to

6. Activated Sludge Models Coupled to CFD Simulations 153

8. Mass Transfer Around a Single Soluble Solid with Different

9. Advances on Viscoelastic Fluid Flow Simulation 233

10. Tree-Shaped Flow Structures Viewed from the Constructal

11. Determination of Effective Transport Properties of Metallic

Part II: Two Phase Flows on Chemical Enginnering

12. Flow Visualization in Gas-Solid Packed Beds by Spatially

13. Interfacial Area Modelling in Two-Phase Flow Studies 361

14. Mass Transfer Models for Oxygen-Water Co-Current flow in

16. Characterization and Modeling of Flotation Processes 440

Part III: Single and Two Phase Flows on Biomedical Enginnering

17. Pulse Wave Propagation in Large Blood Vessels Based on Fluid-

18. An-Harmonic Modeling of the Peripheral Distortion of the

19. Numerical Analysis of Blood Flow in Stenosed Channels 489

20. Blood Flow Behavior in Microchannels: Past, Current and Future

21. A Survey of Microchannel Geometries for Mixing of Species in

22. Endothelial Cell Responses to Fluid Shear Stress: From

24. A Computational Study on the Possibility of the Initialization And

In the chemical engineering arena, there is an interesting combination of chapters

In the biomedical arena a number of chapters cover many fascinating articles

Dr. Vishwas V. Wadekar

Recent advances either in computational and experimental techniques are

José Carlos B. Lopes

School of Chemistry and Chemical Engineering, David Keir Building, Queen’s

Keywords: SpaciMS, monolith, mass spectroscopy, catalyst, kinetic oscillations,

to drastically reduce the environmental impact of human activities, so that

Figure 1: Sketch of a common catalytic converter.

The SpaciMS is a minimally invasive spatially resolved capillary-inlet mass

thermocouples [1-6], which enables the determination of chemical composition of

The SpaciMS approach compared to the conventional ‘end-pipe’ analysis strategy

THE SPACIMS SET-UP

The equipment is subdivided in 5 parts:

 A mass spectrometer (MS),

Abstract: One of de most important environmental problems nowadays is the abatement

Keywords: Reverse flow reactor (RFR), selective catalytic reduction (SCR),

*Address correspondence to Emilio Muñoz: Department of Chemical Engineering and Environmental

Country NOx [Kilotonnes] NH3 [Kilotonnes] SO2 [Kilotonnes]

The European Environmental Agency, in the status report up to 2008

Figure 1: Comparison of NOx projections and ceilings in 2010.

4 NH3 + 4 NO + O2 → 4 N2 + 6 H2O (1)

Institute of Chemical Engineering and High Temperature Chemical Processes,

Abstract: The accurate description of flow in nano-scale pores or channels is very

Keywords: Rarefied flow, porous media, transition regime, direct simulation

Recent advances in microtechnology and, in particular, in microelectromechanical

*Address correspondence to Vasilis N. Burganos: Institute of Chemical Engineering and High

involve features of small characteristic size, including vacuum processes, high-

Even though it is widely used, such a classification oversimplifies the underlying

Therefore, it is essential to define additional parameters in order to classify the flow

 Boltzmann Equation 

Continuum Flow Slip Flow Transition Flow Free-molecule

Kn 10-3 10-2 10-1 100 10+1

Figure 1: Flow regimes and governing equations.