Vous êtes sur la page 1sur 146

ADDIS ABABA UNIVERSITY

SCHOOL OF GRADUATE STUDIES


INSTITUTE OF TECHNOLOGY
DEPARTMENT OF CHEMICAL ENGINEERING

EFFECT OF PROCESSING ON SOME QUALITY


ATTRIBUTES OF MANGO (Mangifera indica) FRUIT
LEATHER

By

BINYAM TESHOME

A Thesis submitted to the school of Graduate Studies of Addis Ababa


University in partial fulfillment of the Requirements for the Degree of
Master of Science in Chemical Engineering (Food Engineering)

Advisor: Mr. Adamu Zegeye

May, 2010
Addis Ababa
Ethiopia

ADDIS ABABA UNIVERSITY


SCHOOL OF GRADUATE STUDIES
INSTITUTE OF TECHNOLOGY
DEPARTMENT OF CHEMICAL ENGINEERING

EFFECT OF PROCESSING ON SOME QUALITY


ATTRIBUTES OF MANGO (Mangifera indica) FRUIT
LEATHER

By

Binyam Teshome
Approved by the Examining Board:

_____________________

___________________

Chairman, Departments Graduate Committee

Mr. Adamu Zegeye

___________________

Advisor

Dr. Cherinet Abuye

___________________

External Examiner

Dr. Eng. Shimelis Admassu


Internal Examiner

___________________

Acknowledgment
I would like to forward my deepest gratitude to my advisor Mr. Adamu Zegeye for his keen interest
in my thesis work, follow up of my progress, encouragement and support.

I acknowledge Addis Ababa Institute of Technology for the financial support. I am also grateful to
the academic staff of the Department of Chemical Engineering for imparting tremendous
knowledge to me. I appreciate Dr. Eng. Shimelis Admassu for his constant supervision and
recommendation during my project work. Thanks to the Ethiopian Health and Nutrition Research
Institute (EHNRI) for letting me use their food analysis laboratory and facilitating my research,
especially Dr. Cherinet, Ato Adamu and Israel. Ethiopian Et-Fruit Company is also appreciated for
providing the mango varieties used in this research. I thank all the technical staff of my
Departments laboratory, particularly, Ato Hintsasilase Seifu and also Yeshihareg Nesibu for
providing me all the necessary support during the research.

Equally and importantly, I would like to acknowledge all family members particularly Girum
Teshome and my dearest wife Hayley Teshome, who contributed towards my success with their
financial support and encouragement in the course of this research, and also honor my friends who
shared my idea when I was in need.

ii

Table of Contents

CHAPTER Title

Page

Title Page

Acknowledgment

ii

Table of Contents

iii

List of Tables

vi

List of Figures

viii

List of Abbreviations

ix

Abstract

INTRODUCTION

1.1. Background

1.2. Statement of the problem

1.3. Objectives

1.4. Structure of the thesis

LITERATURE REVIEW

2.1. Production and marketing of Mango fruits in Ethiopia

2.1.1 Exporting Mango fruits

2.1.2 Mango value chain analysis in Ethiopia

2.1.3 Asossa market

2.1.4 Addis Ababa market

2.2 Processing of Mango fruits in Ethiopia

12

2.3 Selected Mango varieties for processing

13

2.4 Medicinal uses and by-products of Mango

14

2.5 Mango processing technologies

14

2.5.1 Ripe Mango processing

16

2.6 Fruit leather processing

17

2.6.1 Preparation of fruits

17

2.6.2 Heating, drying and packaging

18

2.7 Mango fruit leather recipes and processing procedures


iii

19

2.7.1 Adding sweeteners and flavoring to fruit leather

2.8 Quality control

20

2.9 Effect of processing on food quality attributes

22

2.9.1 Physicochemical properties

23

2.9.2 Changes on Vitamins

24

2.9.3 Flavor and pigment components

26

2.9.4 Sensory attributes

29

2.9.5 Influence of drying process

30

2.10 Food safety

31

MATERIALS AND METHODS

32

3.1 Raw material source and equipment

32

3.2 Approach for selection and preparation of Mango fruits

32

3.3 Development of Mango fruit leather

33

3.3.1 Raw material preparation and formulation of the puree mix

33

3.3.2 Heating and drying the puree mix

34

3.4 Methods for studying the effect of processing


3.4.1 Effect of processing on drying time
3.5 Quality parameters for studying the effect of processing

20

37
37
38

3.5.1 Proximate analysis methods for the puree and leather

38

3.5.2 Physicochemical analysis

41

3.5.3 Texture analysis of the Mango leather

42

3.5.4 Microbiological analysis

43

3.5.5 Sensory evaluation

44

3.6 Experimental design and data analysis

44

RESULTS AND DISCUSSION

45

4.1 Physicochemical properties of Mango puree

45

4.2 Proximate analysis results of the Mango puree and puree mix

46

4.2.1 Keitt Mango variety

47

4.2.2 Tommy Atkins Mango variety

47

iv

4.3 Effect of heating temperature on viscosity of Mango puree mix

48

4.3.1 Keitt Mango variety

49

4.3.2 Tommy Atkins Mango variety

50

4.4 pH of the puree mix

51

4.5 Effect of drying on the proximate and Vitamin C content of fruit

51

leather
4.6 Effect of temperature and puree load on drying time

52

4.6.1 Drying characteristics of Mango puree mix

52

4.6.2 Analysis of moisture loss during drying process

53

4.7 Texture analysis of the Mango leather

54

4.8 Proximate analysis of Mango fruit leather

55

4.8.1 Moisture content

56

4.8.2 Protein content

57

4.8.3 Fat content

57

4.8.4 Crude fiber content

58

4.8.5 Ash content

58

4.8.6 Carbohydrate content

59

4.8.7 Effect of processing on Vitamin C content of Mango leather

59

4.9 Microbiological analysis of Mango fruit leather

60

4.10 Sensory analysis of Mango fruit leather

61

4.10.1 Effect of drying temperature, puree load and fruit varieties

61

on the sensory qualities


5

SUGGESTED TYECHNOLOGY FOR MANGO LEATHER

66

PROCESSING

5.1 Process description

66

CONCLUSIONS AND RECOMMENDATION

94

6.1 Conclusion

94

6.2 Recommendation

96

REFERENCES

97

ANNEXES

102

List of Tables

Chapter Table

Title

Page

2.1

Estimate of area, production and yield of Mango fruits

4.1

Physico-chemical properties of fruit pulp for the mango varieties

45

4.2

4.3
4.4
4.5

4.6
4.7
4.8

Proximate analysis result for Keitt and Tommy Atkins varieties of


Mango puree and puree mix
Results of viscosities for Keitt and Tommy Atkins variety mango
puree mixes measured at different heating temperature
Proximate analysis result for Mango fruit leathers
Designed sample codes for both varieties of mangos with drying
temperature and puree load
Effect of drying air temperature and puree load on drying time of
Keitt Mango fruit leather
Weights and moisture loss of Keitt variety mango leather
Results of compression test using Texture Analyzer to different
number of Mango leather sheets (layers)

46

48
51
52

53
53
54

4.9

Effect of temperature on proximate composition of mango leather

55

4.10

Effect of puree load on proximate composition of mango leather

55

4.11

Effect of fruit variety on proximate composition of mango leather

56

4.12

Effect of drying temperature, puree load and fruit variety on


vitamin C content

59

4.13

Result of microbiological analysis of mango fruit leather

60

4.14

Effect of drying temperature on sensory qualities

61

4.15

Effect of puree load on sensory qualities

61

4.16

Effect of fruit variety on sensory qualities

62

4.17

Proximate and physicochemical composition of Mango and puree

68

4.18

Specific heat relationships for food product components

73

4.19
4.20

Recipes, calculations and amount of ingredients for making


Mango leather
Typical losses during processing of fruits and vegetables
vi

85
86

4.21

Machinery and equipment required for mango leather production

88

4.22

Manpower requirement for Mango leather production

89

4.23

Raw material costs for Mango leather production

89

4.24

Cost of utilities for Mango leather production

90

4.25

Fixed capital cost estimation for Mango leather production

91

4.26

Estimation of total product cost for Mango leather

92

vii

List of Figures

Chapter Figure

Title

Page

1.1

Structure of the thesis

2.1

Wholesale Mango market in Addis Ababa Market share by region

10

3.1

Process flowchart for Mango leather processing

36

4.1

Viscosity verses temperature of Keitt variety puree mix

49

4.2

Viscosity verses temperature of Tommy Atkins variety puree mix

50

4.3

Qualitative flow diagram for Mango leather processing

67

4.4

Quantitative flow diagram for mango leather processing

78

4.5

Quantitative flow diagram for daily production of mango leather

80

4.6

Equipment layout for mango leather processing

81

viii

List of Abbreviations

AAU

Addis Ababa University

AMARI

Awash Melkasa Agricultural Research Institute

AOAC

Association of Official Analytical Chemists

BAM

Bacteriological Analytical Manual

CIA

Central Intelligence Agency

CSA

Central Statistical Authority

EHNRI

Ethiopian Health and Nutrition Research Institute

ETB

Ethiopian Birr

Et-Fruit

A state owned Ethiopian Fruit Marketing Agency

FAO

Food and Agriculture Organization

FDA

Food and Drugs Administration

GTZ

Gesellschaft fr Technische Zusammenarbeit

LSD

List Significance Difference

MC

Moisture Content

ppm

parts per million

RH

Relative Humidity

SPSS

Statistical Package for Social Scientists

TCA

Trichloro Acetic Acid

TSS

Total Soluble Solid

UAE

United Arab Emirates

USD

United States Dollar

USDA

United States Drug Administration


ix

Abstract
The seasonal production of mango fruit in Ethiopia has to be considered as an opportunity for the
utilization of the fruit. The objective of this research was to study the influence of processing on
some quality attributes of mango fruit leather developed from two fruit varieties namely, Keitt
(local mango) and Tommy Atkins (export standard mango). First, 3.2 kg of mango fruit from both
varieties were allocated for the process to be peeled, cut, sliced into pieces and stones removed.
Mango puree was made using a food processor to obtain 1.65 kg puree. It was put in a beaker and
covered with aluminum foil and then inserted into a water bath fitted with thermostat to control the
temperature. Additional ingredients of Honey, Ginger and Lemon Juice were added and mixed. In
order to cook and shorten the drying time, the mixture was heated at three different temperatures,
600C, 700C and 800C whilst being continuously stirred. The puree mixture was poured onto the
trays to an approximate thickness of 0.64 cm. The trays containing the puree were placed in a
drying oven. Oven drying was conducted for 8 h and finally 0.52 kg of mango leather was obtained.
Drying experiment was also undertaken using convective hot air dryer to minimize the drying time
of the fruit leather using a similar procedure. A minimum drying time of 4 h was achieved in a
convective hot air dryer for Keitt mango fruit leather at 800C with 0.4 g/cm2 puree load. The major
factors considered to have an effect on the leather quality were drying temperatures of 600C, 700C
and 800C, puree load of 0.4 g/cm2 and 0.6 g/cm2 and fruit variety. The developed leather underwent
physico-chemical, textural, microbiological and sensory analysis. The data obtained was analyzed
using SPSS version 17 statistical software. The result indicated that 70.3 % of moisture loss
resulted in the drying process. The viscosity of the mango puree was found to be dependent on
heating temperature. As the temperature increased, the viscosity of the puree first decreased and
then increased within the range of temperature 25.1 to 70.0 0C. The texture analysis result of the
final mango leather showed that 4 sheets and 5 sheets of leather with 5mm and 6mm thickness,
respectively, were found to be suitable for a single bite. The results of the proximate analysis for
both varieties of mango fruit leather indicated that the processing affected the nutritional
composition of the fruit leather. The vitamin C content was also found to be dependent on all
drying temperature, puree load and fruit variety. The vitamin C content of the Keitt mango leather
(26.93%) is greater than that of Tommy Atkins mango leather (22.71%). When compared to the
fresh puree mix, the Keitt mango leather is decreased by 39.66% and that of the Tommy Atkins
mango leather is decreased by 57.82%. The result of microbiological analysis for yeast, coliform,
x

fecal coliform, E.coli, and Shigella species was found to be safe (<1X104) and S.coccus and
Salmonella species were not isolated for both varieties of mango leathers processed at 600C and
0.6 g/cm2 puree load. The Tommy Atkins mango leathers dried at 60 and 700C with 0.4 and 0.6
g/cm2 puree loads were preferred by panelists (P<0.05). The project generally covered the process
and analysis of mango fruit leather and its development at a laboratory scale. Accordingly, an
economically feasible production technology has been suggested.

Key words: Leather, Mango, Processing, Quality

xi

CHAPTER 1
INTRODUCTION
1.1 Background
Mango (Mangifera indica L) is a highly seasonal tropical fruit, very popular among millions of
people in the tropics. It also occupies a prominent place among the best fruits of the world.
However, it is in constant demand, there is a pre-harvest scarcity and at times a post-harvest glut
for this fruit. To increase the availability of this fruit throughout the year, the surplus production
must be processed into a variety of value-added products (Saxena and Arora, 1997; Srinivasan et
al., 2000; Singh et al., 2005). Dried mango products could successfully serve this purpose.
Mangos can be processed into a number of unique products such as dried mango pieces, chutney
and mango leathers (Azeredo, et al., 2006). Processing of mangos enables exporters to serve their
markets even during off season periods for fresh fruits. Mangos are a highly nutritious fruits
containing carbohydrates, proteins, fats, minerals, and vitamins, in particular vitamin A (beta
carotene), vitamin B1, vitamin B2, and vitamin C (ascorbic acid). As the fruit ripens, concentrations
of vitamin C decrease and glucose, fructose, and sucrose concentrations increase (Bally, 2006).
Drying of agricultural products is the oldest and widely used preservation method. It involves
reduction as much water as possible from foods to arrest enzyme and microbial activities hence
stopping deterioration. Moisture left in the dried foods varies between 2-30% depending on the
type of food. In tropical countries, solar dryers can be used to dry fresh produce when average
relative humidity is below 50% during drying period. Drying lowers weights and volume of the
product hence lowers costs in transportation and storage. However, drying allows some lowering in
nutritional value of the product e.g. loss of vitamin C, and changes of color and appearance that
might not be desirable (GTZ, 2009).
Fruit leathers are dried sheets of fruit pulp which have a soft, rubbery texture and a sweet taste.
They can be made from most fruits, although mango, apricot, banana and tamarind leathers are
amongst the most popular. Leathers can also be made from a mixture of fruits. Fruit leathers are
made by drying a very thin layer of fruit puree and other ingredients to produce a leathery sheet of
dried fruit with a texture similar to soft leather (Andress & Harrison, 1999). They may be eaten as
snack foods as a healthy alternative to boiled sweets and also used as ingredients in the
1

manufacture of cookies, cakes and ice cream. Fruit leathers are often targeted at the health food
market, using marketing images such as pure, sun dried, and rich in vitamins.
Mango leather is a traditional product prepared from sound ripe mango. Traditionally, sun drying is
the process employed for preparing mango leather from ripe fruit pulp. However the sun-dried
product is discolored and the process is unhygienic and lengthy. Cabinet drying has been carried
out for making mango leather (Heikal et al., 1972; Mir and Nath, 1995) resulting into a product
with improved color and flavor.
The preservation of fruit leathers depends on their low moisture content (15-25%), the natural
acidity of the fruit and the high sugar content. When properly dried and packaged, fruit leathers
have a shelf life of up to 9 months (FPT, 2009). Although fruit leather is a relatively well
established product overseas, few studies have been published about this kind of product. Most of
the studies utilize not only fruit purees in the fruit leather, but also other ingredients (especially
sugars) and additives. For instance, Chan and Cavaletto (1978) prepared papaya leathers with
sucrose and SO2. They observed that SO2 reduced changes in the colour of papaya leathers during
both processing and storage. Che Man, et al., (1992) prepared sapota leathers from sapota puree,
sucrose, rice flour, sorbic acid, and sodium metabisulphite; the leathers were shelf-stable for 3
months. Jackfruit leathers with added sucrose and sorbic acid were produced by Che Man and
Taufik (1995); the product remained stable for 2 months. Irwandi, et al., (1998) produced 12-week
stable durian leathers from a formulation including sucrose and sorbic acid. Vijayanand, et al.,
(2000) produced 3-month shelf-stable guava leathers with sucrose and metabisulphite. Mango
leather products have very low protein content (12%). Therefore several studies have increased
protein content in the mango leather by adding shrimp flour and rice flour, whey protein isolate and
soy protein isolate (Exama & Lacroix, 1989; Payumo, et al., 1981; Chauhan, et al., 1998). All the
above-mentioned studies reported good consumer acceptance of the fruit leather product.

1.2 Statement of the problem


Mango puree and leather products are largely unknown in Ethiopia, but may have good potential
for a number of reasons. According to Kadir (2009) the current postharvest loss of mango fruits in
Ethiopia is more than 26.3%. However, there is a growing international demand for dried mango
fruits and mango fruit leathers. Consequently the abundant supply of mango fruit in Ethiopia could
be utilized to create new products (including organic fruit leathers) for this overseas market.
The expansion of the bakery industry has created a demand for new ingredients, for cake
production, which are imported and are relatively expensive. Mango fruit leather, produced locally,
could be utilized as a cheaper alternative ingredient. There is also a growing consciousness over the
negative effects of sugar confectionery on dental health and at present there are few alternatives for
concerned parents to give to their children. Therefore, fruit leather products would be a more
acceptable healthy alternative. The current drying method used widely for making fruit leathers is
use of electrical oven and convective hot air drying. The drawback of hot air drying method is a
long drying time and the controlling of drying conditions (including increasing drying temperature,
decreasing initial moisture content of the puree and so on) will result in the quality of the mango
fruit leather. The technology of electrical oven drying method had long been employed to extend
the shelf life of foods. However, there is an urgent need to conduct basic studies to investigate the
effects of processing on the total drying time and some qualities of the mango fruit leather to
optimize the process and set up the processing industry.
The quality of mango fruit leather and the total drying time is directly affected by the drying
method, type of dryer, oven design and operating parameters. These parameters must be well
established and controlled for each type of fruit leathers. In the production of mango fruit leathers,
the puree making process and mixing with other ingredients, the drying temperature, and the puree
load on each tray during drying, the total drying time, packaging of the product and appropriate
storage conditions should be well established for high quality product. In this study, the effects of
processing on some qualities of mango fruit leathers and the total drying time taken are
investigated.

1.3 Objectives
The general objective of the thesis work is to study the influence of processing on some quality
attributes of mango fruit leathers.
The specific objectives of the thesis are to:
Produce fruit leather from locally grown mangos
Conduct physico-chemical analysis of both raw and processed product
Study the effect of processing on the qualities of the fruit leather
Suggest a manufacturing process for mango leather production
Significance:
The study is believed to be significant in that it will:
Reduce postharvest loss of mango fruit in Ethiopia
Produce the best quality mango leather products
Reassure the consumer that the mango leather product is safe for consumption
Optimize the processes for production of good quality mango leather products
Scope:
The study generally covers:
The processing method for production of mango fruit leather
Development of mango leather
Laboratory analysis majorly on: Physico-Chemical Analysis, Proximate Analysis
(Nutritional Composition Analysis), Microbiological Analysis
Sensory Evaluation
Suggestion of technology and economic analysis

1.4 Structure of the thesis

Fig. 1.1 Structure of the thesis

CHAPTER 2
LITERATURE REVIEW
2.1 Production and marketing of Mango fruits in Ethiopia
The Ethiopian government has a plan to expand mango production by distributing high yielding
varieties for small scale farmers, especially in the Southern and Oromia region, by grafting mangos
of known and high yielding varieties. In July 2006, it was announced that the Oromia Government
distributed 14,000 improved seeds of mango. The production of mango fruits for the past six years
in Ethiopia was considered for the study which was found from CSA (2009), and is summarized
and presented below in Table 2.1. With an increase in Ethiopian mango crop production and
considering the current postharvest loss of mango fruits is at 26.3%, there is not only a need but
also a potential for the fruit to be processed into various product types, consequently increasing the
market potential of the mango fruit (Kader and Truneh, 2009). Industrial processing opportunities,
to increase the market value of the initial fruit, may lead to the potential development of the
following products:- Food (mango juice and fizzy drinks, canned fruits and pulp, fruit leather, dried
pieces, jam and chutney), domestic (mango detergent and cleaning agents), beauty (as an applied
product in skin creams products).
Table 2.1 Estimate of area, production and yield of Mango fruits, Meher season
Year

Number of

Area in

Production in

Yield

holders

hectare

quintal

(qt/ha)

2003/04 (1996) E.C

350,067

4,964.00

292,283.00

58.88

2004/05 (1997) E.C

414,574

5,814.00

301,715.00

51.89

2005/06 (1998) E.C

463,868

5,400.31

547,291.24

104.06

2006/07 (1999) E.C

558,976

6,796.10

626,111.83

94.08

2007/08 (2000) E.C

695,030

6,730.83

484,360.97

71.96

2008/09 (2001) E.C

716,447

6,051.00

441,582.00

72.97

2.1.1 Exporting Mango fruits


At present, very little mango is exported from Ethiopia with only 4 tonnes exported in 2006 at a
value of less than US$1000 according to FAO. This represents a significant decline since 2002
when 811 tonnes were exported at a value of US$675,000 (US$832 per tonne). This appears to
have been a particularly high value year however, as the longer term average price for mango
exports has been approximately US$323 per tonne. Anecdotal information from key informants
suggested that one of the main reasons for the drop in mango exports has been the variable quality
of Ethiopian mango exports on arrival in overseas countries. It was reported that Et-Fruit (the state
owned Ethiopian Fruit marketing agency) had been exporting mangoes to countries such as
Djibouti, Saudi Arabia and UAE but had lost some of those contracts due to the poor quality of the
shipments on arrival. This situation highlights the key challenges faced in trying to develop the
export market for Ethiopian mangoes: Under-developed packaging and cold chain for exporting,
High cost of freight to overseas countries, Competing product from Egypt and South Africa and
Minimal production of commercial varieties
In order to begin considering export markets as a viable opportunity, it is essential to consider the
nature of demand in export markets. It is very clear that overseas markets are increasingly
demanding higher quality, commercial varieties of mangoes, and are also becoming more
sophisticated in their preferences for products that are organic. In Asossa then, the first step to even
consider export markets would be to begin growing more commercial varieties such as Kent, Keitt
and Tommy Atkins. Even if the export market was not a viable option in the short term, these
commercial varieties would present a better alternative to the domestic market with less fiber and
higher levels of sweetness than existing hybrid varieties (FAO, 2009).
In Ethiopia, the domestic market, consumption is largely in its fresh form due to the fact that the
cost increment for processing and packaging would make it beyond the purchasing power of the
vast majority of the Ethiopian consumer group (low-income). However, since 1997 the demand for
canned fruits in Ethiopia has increased by 7% suggesting there is a sufficient domestic market for
canned mangoes to be produced. The mango export markets are where the greatest growth
potentials exist for mango producers. The global mango markets are supplied by countries that are
strategically positioned to be preferred suppliers. For example, West Africa is a key supplier to
Europe due to its proximity to Europe and direct sea-links (Truneh, 2009).

2.1.2 Mango value chain analysis in Ethiopia


Ethiopia is a country that is often associated with famine and food shortage. Whilst this perception
is the reality for much of the country at certain times, there are also regions within Ethiopia that are
well suited to producing a surplus for particular agricultural commodities. One such location is the
Asossa Homosha region in western Ethiopia, which is particularly suitable to the production of
mangoes. In a 2006 study, it was estimated that as much as 28% of the mangoes sold in the capital
Addis Ababa, were grown in the Asossa region (WAFC, 2006).
Though the immediate market is small, there is a short and established value chain to Addis Ababa
via a 700 km trip. Typically, Asossa exports between 150-300 (10-ton) truckloads of fruit per
season to various regions, principally Addis Ababa. In spite of the high cost of imported fruit,
nationally the volume has risen by 100% in the last seven years, while that of imported juices has
more than doubled in the last four years, evidence of an up-scaling market. The farm-gate price of
mangoes in Asossa ranges from 0.25 to 1.15 Birr/kg while the retail price is approximately 5 Birr
per kg in Addis, compared to 10 Birr per kg of other imported whole fruits. Farmers typically
achieve approximately 5 to 8% of the final retail price (at lower levels) giving them about
1,400Birr in annual income, while wholesalers get about 30% of it but meet the high transport
costs. Analysis reveals a reasonably resilient subsector with a favored market position of the
Assosa mango regional brand. Also, the region appears to have a comparative advantage with
ideal growing conditions for mangoes and high yielding trees. At the production level however, the
value chain is quite rudimentary with mainly subsistence level cultivation, harvesting and posthandling techniques that limit the quality of the fruit. Upstream there are also issues with most
grading and packaging being undertaken following a long road journey to the capital, undermining
not only the quality of fruit but also the potential value generated at the farmer level. At the
wholesale level in Addis Ababa, market traders dominate the landscape and operate in ways that
make it difficult for new entrants to enter the market. Addis wholesalers have strong relationships
with the traders based in Asossa and these two levels of the value chain account for most of the
final retail price. Given the roles they play, it appears that there is not a proportionate addition of
value in the chain, and that is where opportunities lie for improving farmer level value capture in
the chain (James et al., 2009).

2.1.3 Asossa market


Within the market in Asossa, there are three main channels for selling mangoes: Farm gate that
mainly sell to traders (who sell on to Addis), consumers and small retailers; Asossa Town Market
that mainly sell to local consumers and Small Retailers that mainly sell to local consumers. The
market is dominated by the open town market where many farmers bring their produce for sale.
During peak mango season, this market is saturated with mangoes to the point that prices fall
dramatically and many farmers do not even bother to carry their produce home at the end of the
day. As well as the open Asossa market, there are a number of kiosk-style small retailers in Asossa
which suffer from similar problems at peak supply times. A significant point to note is that much of
the mango being supplied at the regional market is not the top grade being produced, but is actually
the remaining produce from what has been already sold to traders at farm gate. In this way, traders
are siphoning the best quality mangoes to sell in the Addis market, but also provide very little
choice to disorganized farmers about where to sell. This study found that farmers on average sell
between 0.4 and 1.15 birr depending on quality, and also customer. The lower price tended to be
achieved in the Asossa market, with the higher price being received at farm gate from traders
purchasing only the higher quality mangoes (Aithal and Wangila, 2006).
2.1.4 Addis Ababa market
The market is dominated by two large wholesale markets, being the Mercato and the Piazza. These
markets are the main destination for agricultural produce arriving from around the country. These
markets serve not only consumers, but are also where supermarkets, large retailers, hotels and
thousands of small kiosk-like retailers source their mangoes. Actual data on the volume of mangoes
sold in these markets is very unreliable, however a number of interview sources have identified that
in the past five years there has been significant growth in market size and increasing consumer
demand. This is largely to do with a steadily increasing population (3.2% p.a.), more sophisticated
consumer preferences for exotic and tropical fruits, and growing incomes amongst emerging
middle and upper class consumers. The average price on the wholesale markets in Addis Ababa
was approximately 3.5 birr per kg, or 3,500 birr per tonne, whereas the final retail price in Addis
could reach as high as 5 birr per kg (5,000 birr per tonne). By analyzing the price differentials
throughout the value chain, it is clear to see that farmers only capture a small portion (8%) of the
final retail price. There is no reliable data on the number of small retailers in Addis, but it is
estimated there are thousands of street level small retailers/kiosks - some of them dealing
9

exclusively in fruit, whilst others sell a range of consumable goods. These are a major channel for
selling mangoes on to final consumers and are particularly convenient as a street level channel in
places of high traffic, with high turnover of goods (CIA, 2008).

Fig. 2.1 Wholesale Mango market in Addis Ababa Market share by region
Source: (Aithal and Wangila, 2006)
The Mercato and Piazza markets are largely controlled by approximately ten major wholesalers,
who deal in fruits and vegetables from all over the country. Industry sources mentioned that the
Addis wholesalers are organised under groups that have strong ethnic ties and tend to operate in
ways that have been described as cartels. Similar to the upstream nature of the value chain, the
wholesalers in Addis operate in a heavily crowded, poorly structured, and underdeveloped market
infrastructure. There is no known refrigerated storage facilities in the market as the cost of this
investment has always been seen as too high. The typical wholesale price in Addis is approximately
3,500 birr per tonne and means that Addis wholesalers achieve between 20-40% of the final retail
price. This is a considerable portion of the final price and reflects the risk that Addis wholesalers
take in ordering mangoes from Asossa. Other value add activities that wholesalers undertake is repackaging and grading the fruit on arrival in Addis, as produce most often arrives in bulk having
10

not yet undergone any systematic grading or packaging. These are functions that are possible at the
farm level and may be areas where farmers can extract a greater price and generate more value in
the chain (Aithal and Wangila, 2006).
According to a recent update from the mango value chain analysis by James et al., (2009), it has
been indicated that within the first six months of the project implementation, the following have
been achieved:

19 farmers cooperatives have been set up and linked to an umbrella cooperative union. Out
of these, 100 Cooperative members have been trained on Mango Processing especially in
producing jam, juice, compote, vinegar, wine etc. This processing is under taken on site in
Asossa.

Farm gate Price of Mango increased from 25 Birr/100kg to 175 and the farmers have started
using weighing scales to measure quantities rather than counting pieces or heaps.

Between March and June 2009, 357 tonnes of mangoes were sold at the price of 569,084
Birr (roughly USD 46,000) to the most reputable fruit dealer- Et-Fruit for the first time.
With World Vision supervision, the income was equitably shared among the farmers.

4,000 bottles of various processed mango products like jam, juice, wine and compote have
been packed and sold to a number of super markets in Addis Ababa.

The farmers have entered into partnership with the Ecological Products of Ethiopia,
(Ecopia) to process and market mango products.

11

2.2 Processing of Mango fruits in Ethiopia


The mango processing industry in Ethiopia is in its infant stage. However, mango is grown in many
parts, especially in the Rift Valley, western and south-western parts of the country. The national
research system has developed a number of varieties but they are not widely spread. Experiences
from other countries in growing this crop will therefore contribute to the success and distribution of
this fruit. The mango industry in Kenya has expanded considerably over recent years, not only in
size but also in the geographical location of commercial and homestead plantings. No longer is
commercial mango cultivation restricted to the Coast Province, as significant plantings of improved
cultivars now also exist in the Eastern and Central provinces, among other regions (EAP, 2009).
The fruit processing industry in Ethiopia is very weak, considering the substantial amount of fruit
that is grown in the country. No doubt, one of the reasons for this is the highly developed
processing industries in other countries which are able to export into countries like Ethiopia and
sell the final product at low cost. Indeed, there were a number of imported, long-life mango juice
brands available throughout Ethiopia and is certain to act as a competitive entry barrier for
domestically produced juice. Investigations of local processors found only one significant player,
who actually imported frozen mango flesh from India for processing juice in Ethiopia. The main
considerations for purchasing Indian imports were the variety, quality, consistency, and price of the
imports. When asked about replacing imports with Ethiopian produced mango, the informant
indicated that would be his preference, however so far, Ethiopian fruit was not able to compare on
the key criteria identified, particularly on price. The informant did however predict that juice
processing would begin to emerge as a more viable sector, as mango juice is clearly the most
favored juice product by consumers. He indicated that demand for the juice as a category was
seeing strong growth, with mango leading this growth. The other key challenges for developing a
fruit processing sector in Ethiopia include: lack of technical knowledge in processing, low level of
technical support for maintenance, low capital base from which to invest, and many low priced
mango juice imports (James et al., 2009).

12

2.3 Selected Mango varieties for processing


Tommy Atkins Mango variety
It is one of the most popular mangos in the world and cultivated in Florida early 1920's. The
Mango cultivar was developed and grown for commercial export. The fruit is a regular oval,
medium to large sized, 12 to 24 ounces, yellowish-orange with deep red to purple blush, thicker
skinned, juicy but firm with medium fiber.

Pic. 2.1 Tommy Atkins (LFI, 2003)


Keitt Mango variety
Indian strain thought to have originated, like the Haden, from a seedling of Mulgoba 1945,
Homestead, Florida. The fruit is a large (20-26 oz.) ovate tapering with slight nose-like
protuberance above its tip. Green to orange-yellow as it ripens; firm flesh with a piney sweetness
and minimal fiber surrounding the seed area. It is a late fruiting mango, often available into fall.

Pic. 2.2 Keitt (LFI, 2003)


Kent Mango variety
Kent mango was first cultivated in Florida in 1944. It is a direct descendant of the Brooks cultivar,
derived from the Sandersha seedling. The fruit is a regular oval shape, large 20 - 26 ounces, with
plump cheeks, greenish-yellow color with red shoulder. Very rich and sweet with fiber-free flesh
(slices clean to the pit - like butter when ripe!) It is a softer mango that really should not be put to
the squeeze test.

13

Pic. 2.3 Kent (LFI, 2003)


2.4 Medicinal uses and by-products of Mango
Mangos are an excellent source of Vitamins A and C, as well as Potassium, Beta-carotene,
enzymes and anti-oxidants. Mangoes are high in fibre, but low in calories (approximately 110 per
average sized mango) fat (only 1 g) and sodium. Mangoes are a good staple for your daily diet. It
has a reputation (not necessarily scientifically proven) as an alternative or complementary medicine
for a wide range of illnesses including beriberi, bronchial diseases, anxiety, insomnia, fatigue,
depression, digestive problems heartburn, constipation, kidney stones.
Mango kernel contains high amounts of fat and starch. The oil extracted from kernel is of good
quality and could be used in cosmetic and soap industries. The kernel flour (starch) after mixing
with wheat or maize flour is used in chapattis in India. About ten percent alcohol could be obtained
from mango kernel by co-culture fermentation (Truneh, 2009).
2.5 Mango processing technologies
The processing of fruits has two objectives. Firstly, to preserve by slowing down the natural
processes of decay caused by microorganisms, enzymes in the food, or other factors such as heat,
moisture and sunlight. The second objective is to convert them into different foods which are
attractive and in demand by consumers Food Processors should utilize their skills to develop
recipes and create attractive products that consumers want to eat. Thereby successful Food
Processors increase product sales and generate profits. Food Processors must select their products
with caution. It is not enough to assume that processing can be a successful business simply
because there a large quantities of cheap fruit available in the marketplace. There must be a good
demand for the end product and this must be clearly identified before designing and investing in the
business. The best types of products for small-scale production are those that have a high addedvalue as well as a good demand. A high added value means that cheap raw materials can be
processed into relatively expensive products. It also means that this can be done at a small scale of
processing, using equipment that is affordable (Fellows and Quaouich, 2004).
14

Fruits like mangos, pawpaw, guavas and bananas, can easily be dried. However, they should be
harvested at the right stage and ripeness. Hard ripe stage in mangoes, pawpaw and bananas gives
best results. Avoid overripe, under mature fruits in order to obtain good products. To prepare the
fruits for drying, wash them thoroughly with clean water. Scrubbing with a brush might be
necessary like in case of mango fruit with a lot of latex cover. The fruits are peeled if necessary and
cut into smaller uniform pieces to ensure faster drying. Stainless steel knives are recommended for
peeling and cutting of the slices or pieces. To avoid discoloration and excessive vitamin losses,
treatment with anti-oxidants like citrus (lemon) juice is done. Fruits like pineapples may require
pre-cooking to soften fibrous tissue hence hasten drying. Drying is done on trays, which should be
made of wood, fabric, plastic or sisal material. This is because metal materials may affect the
drying product negatively e.g. copper destroys vitamin C, iron rusts, aluminum discolors fruits and
corrodes. Most fruits have natural acids and sugars which are preservatives therefore moisture
contents of about 20% i.e. leathery and springy dry (not brittle) is good for storage. This is however
dependent on the fruit or vegetable. After the correct stage of dryness is achieved the product
should be removed from the dryer parked, and stored in a dry, dark store to avoid loss of vitamin A
(GTZ, 2009).
Mangos are processed at two stages of maturity. Green fruit is used to make chutney, pickles,
curries and dehydrated products. The green fruit should be freshly picked from the tree. Fruit that is
bruised, damaged, or that has prematurely fallen to the ground should not be used. Ripe mangoes
are processed as canned and frozen slices, puree, juices, nectar and various dried products. Mango
processing within the home and cottage industries converts the fruit into many other products.
Mango processing presents many problems as far as industrialization and market expansion is
concerned. The trees are alternate bearing and the fruit has a short storage life; these factors make it
difficult to process the crop in a continuous and regular way. The large number of varieties with
their various attributes and deficiencies affects the quality and uniformity of processed products.
Additionally, the lack of simple, reliable methods for determining the stage of maturity of varieties
for processing also affects the quality of the finished products. Many of the processed products
require peeled or peeled and sliced fruit. Within Ethiopia the lack of mechanized equipment for the
peeling of ripe mangoes is a serious bottleneck for increasing the production of these products. A
significant problem in developing mechanized equipment is the large number of varieties available
and their different sizes and shapes. The cost of processed mango products is also too expensive for
the general population in the areas where most mangoes are grown. However, there is a
15

considerable export potential to developed countries but in these countries the processed mango
products must compete with established processed fruits of high quality and relatively low cost
(Dauthy, 1995).
2.5.1 Ripe Mango processing
Mango Puree
Mangoes are processed into puree for re-manufacturing into products such as nectar, juice, squash,
jam, jelly, dehydrated products such as fruit leathers. The puree can be preserved by chemical
means, or frozen, or canned and stored in barrels. This allows a supply of raw materials during the
remainder of the year when fresh mangoes are not available. It also provides a more economical
means of storage compared with the cost of storing the finished products, except for those which
are dehydrated, and provides for more orderly processing during peak availability of fresh
mangoes. Mangoes can be processed into puree from whole or peeled fruit. Because of the time and
cost of peeling, this step is best avoided but with some varieties it may be necessary to avoid offflavors which may be present in the skin. The most common way of removing the skin is handpeeling with knives but this is time-consuming and expensive. Steam and lye peeling have been
accomplished for some varieties. Several methods have been devised to remove the pulp from the
fresh ripe mangoes without hand-peeling. A simplified method is as follows: the whole mangoes
were exposed to atmospheric steam for 2 to 2 and 1/2 min in a loosely covered chamber, and then
transferred to a stainless steel tank. The steam-softened skins allowed the fruit to be pulped by a
power stirrer fitted with a saw-toothed propeller blade mounted 12.7 to 15.2 cm below a regular
propeller blade. The pulp is removed from the seeds by a continuous centrifuge designed for use in
passion fruit extraction. The pulp material is then passed through a paddle pulper fitted with a
0.084 cm screen to remove fiber and small pieces of pulp.
Mango puree can be frozen, canned or stored in barrels for later processing. In all these cases,
heating is necessary to preserve the quality of the mango puree. In one process, puree is pumped
through a plate heat exchanger and heated to 90C for 1 min and cooled to 35 C before being
filled into 30 lb tins with polyethylene liners and frozen at -23.50 C. In another process, pulp is
acidified to pH 3.5, pasteurized at 90C, and hot-filled into 6 kg high-density bulk polyethylene
containers that have been previously sterilized with boiling water. The containers are then sealed
and cooled in water. This makes it possible to avoid the high cost of cans. Wooden barrels may be
used to store mango pulp in the manufacture of jams and squashes. The pulp is acidified with 0.5 to
16

1.0% citric acid, heated to boiling, cooled, and SO2 is added at a level of 1000 to 1500 ppm in the
pulp. The pulp is then filled into barrels for future use.
Dried/dehydrated Mango
Ripe mangoes are dried in the form of pieces, powders and flakes. Drying procedures such as sundrying, tunnel dehydration, vacuum-drying, osmotic dehydration may be used. Packaged and stored
properly, dried mango products are stable and nutritious. One described process involves as pretreatment dipping mango slices for 18 hr (ratio 1:1) in a solution containing 40 Brix sugar, 3000
ppm SO2, 0.2% ascorbic acid and 1% citric acid; this method is described as producing the best
dehydrated product. Drying is described using an electric cabinet through flow dryer operated at
60 C. The product showed no browning after 1 year of storage (Dauthy, 1995). Mango fruits can
also be dried in the form of leathers or bars.
2.6 Fruit leather processing
Fruit leathers are dried sheets of fruit pulp which have a soft, rubbery texture and a sweet taste.
They can be made from most fruits, although mango, apricot, banana and tamarind leathers are
amongst the most popular. Leathers can also be made from a mixture of fruits. Fruit leathers are
eaten as snack foods instead of boiled sweets. They are also used as ingredients in the manufacture
of cookies, cakes and ice cream. The preservation of fruit leathers depends on their low moisture
content (15-25%), the natural acidity of the fruit and the high sugar content. When properly dried
and packaged, fruit leathers have a shelf life of up to 9 months (FPT, 2009).
2.6.1 Preparation of fruits
Fruit should be washed in clean water, peeled and the stones removed. Washing water can be
chlorinated by adding 1 teaspoon of bleach to 4.5 liters of water. All fruit should be ripe and free
from bruising. Any rotten or bruised fruit should be thrown away as this will spoil the color and
flavor of the leather. The puree must be heated to a higher temperature for a longer time to destroy
the enzyme (it must be boiled for 20 minutes). Only stainless steel knives should be used to chop
the fruit. Other metals will discolor the fruit flesh. At the simplest level, fruit is made into a puree
by hand using a food mill. If electricity is available, a liquidizer or blender can be used to increase
the production output. The liquidized fruit is strained or sieved to remove the fibers, seeds, etc to
make a smooth puree. Fruit puree can be semi-processed and stored in sealed drums for further
processing later in the season. Sulphur dioxide (SO2) (600ppm) is added to the drums to prevent the
17

growth of micro-organisms. The semi-processed fruit can be stored for several months. Chemical
preservatives may be added to the fruit puree to maintain a bright color in the leather. Preservatives
are also added if the puree is to be stored before processing. A variety of ingredients can be added
to the fruit puree - sugar to increase the sweetness, citric acid to increase the acidity and chopped
nuts, coconut or spices to vary the taste and flavor.
2.6.2 Heating, drying and packaging
The puree must be heated to 900 C to inactivate the enzymes and reduce the level of
microbiological contamination. A double pan boiler is recommended for heating to avoid burning
the puree. The fruit puree is poured in a thin layer (3-6mm thick) on plastic trays or wooden trays
lined with greaseproof paper. The puree can be poured into a square which is later cut into small
pieces, or into small circles which are rolled up when dry. The leathers should not be dried in direct
sunlight as this will cause the color to fade and reduce the levels of vitamins A and C. Indirect solar
dryers or mechanical dryers should be used. The leather should be dried overnight in a solar dryer
or for about 5 hours in a mechanical dryer. After this time it is turned over and dried on the other
side. The leather is dried until it has a final moisture content of 15-25%. After drying, the leather
pieces should be dusted lightly with starch to prevent them sticking together. All equipment must
be thoroughly cleaned each day to prevent contamination by insects and micro-organisms.
In developing countries fruit leather is usually packaged cheaply with easily sourced materials. The
fruit leather is sold as a roll that is interleaved with greaseproof paper to prevent it from sticking
together. Strips of the leather are weighed, laid on a piece of greaseproof paper and rolled with the
paper. The rolls or discs of leather are packed in polythene or polypropylene heat-sealed bags. The
bags should be placed in boxes to protect them from the light. Fruit leather products in Europe are
packaged as a bar or as mini sweets, within a sealed plastic/foil case. The fruit leather bars within
their plastic/foil case may be packaged and sold as a multipack within a cardboard box made from
recycled paper products. The attractive, modern design of the packaging is specifically aimed at the
health conscious, luxury product niche within the consumer market (FPT, 2009).

18

2.7 Mango fruit leather recipes and processing procedures


The following basic recipes are only guidelines since they depend on the composition of fruit
(which varies between different types) and the different consumer tastes for sweetness. The recipes
needed for mango leather preparation are: fully ripe mango, Sugar (10-15% the pulp weight
according to the variety used and consumer taste), lemon juice or citric acid 2 spoons per kg pulp),
Sodium of potassium metabisulphite (2g per kg pulp), and Glycerin for foods.
The general procedure for making mango fruit leather is as follows:
1. Wash the mangoes in clean water. Drain, Sort and remove any unripe or over-ripe fruit.
2. Peel the fruit with a stainless steel knife and cut the flesh into small pieces.
3. Extract the pulp using a pulper.
4. Weigh the pulp and mix with the sugar, lemon juice and metabisulphite in the ratios
above.
5. Heat at 70-80 0C.
6. Remove the foam from the top of the mixture. Grease the surface of trays with glycerin
to prevent the leather from sticking.
7. Pour the hot puree onto the trays at a ratio of 15kg per square meter of tray area.
8. Place the trays in a dryer. Leave to dry until a final moisture content of 15%. The
product will have a soft, leather-like consistency.
9. Place three sheets of leather on top of each other and cut into small 4x4cm squares.
Wrap each square in cellophane. It may be necessary to dust the squares with corn flour
to prevent excess stickiness.
10. Pack in plastic bags, label and store in a cool dry place (FPT, 2009).
Mango fruit leather can also be made using electric dehydrator or drying oven. Andress (2004)
formulated mango fruit leather with different recipes and methods as described below.
Recipes: include; 4 cups mango puree (from about 4 large, unripe mangoes), 1 cup clover
honey, teaspoon ground cinnamon, teaspoon ground nutmeg, and teaspoon ground cloves.
19

Yield: about 2 dryer trays (14 inches in diameter); 8 fruit rolls.


2.7.1 Adding sweeteners and flavoring to fruit leather
Once you have the basics of making fruit leathers it is fun to try some new flavoring, toppings, or
fillings. Below are some new ideas.
Sweetening - If the puree needs some sweetening, add up to cup sugar for each 2 cups of fruit.
Sugar substitutes may be used. Aspartame sweeteners however may lose sweetness during drying.
Spices Add spices to the puree. Add until the taste is acceptable. Begin by adding 1/8-teaspoon
for each 2 cups of puree. Try: cinnamon, cloves, ginger, mint, nutmeg, allspice.
Flavorings Add flavorings to the puree using only 1/8-teaspoon person 2 cups of puree to start.
Try almond extract, lemon peel, orange extract, vanilla or peppermint.
Toppings After spreading the puree on the drying sheet and before drying, sprinkle a topping
over the puree. Try not to cover entire puree, but just lightly sprinkle the topping. Try coconut,
dried fruits, granola, and sunflower seeds.
Fillings After the fruit leather is dried and cool, spread a thin layer of these fillings. Then roll, cut
and serve. If not served immediately, store in the refrigerator or wait to spread filling until just
before serving. Try melted chocolate, softened cream cheese, peanut butter, marshmallow cream,
jam or jelly. Here is an idea for adding dairy foods and calcium to your diet in a fun tasty way.
Yogurt Drops - 18-ounce vanilla yogurt, 13-oz package of sugar free gelatin powder (any flavor)
and mix the gelatin powder with the yogurt. Using a spoon drop the mixture onto a fruit leather
drying tray. They can be done in small rounds or as leather. Dry until sticky and store in the
refrigerator or freezer. This makes great healthy snack and provides another way to get dairy
products and calcium in the diet (Brown, 2009).
2.8 Quality control
Quality control begins with the acquisition of high-quality fruit concentrate. Many purees are
supplied by well-known fruit processors. Other quality control methods include careful calibration
of all additives, particularly of those additives that affect hardening/malleability (malto-dextrin in
particular). Also, cooking and drying temperatures are monitored closely to ensure moisture
content. Scales are carefully calibrated so that each roll contains just the right amount of extruded
20

product; similarly, the packaging machine is checked and re-checked so that each cardboard
package includes the correct number of fruit leathers. Sample testing is performed periodically as
well (Nancy, 2009).
One of the most important concerns of the food manufacturer is to produce a final product that
consistently has the same overall properties, i.e. appearance, texture, flavor and shelf life. When we
purchase a particular food product we expect its properties to be the same (or very similar) to
previous times, and not to vary from purchase-to-purchase. Ideally, a food manufacture wants to
take the raw ingredients, process them in a certain way and produce a product with specific
desirable properties. Unfortunately, the properties of the raw ingredients and the processing
conditions vary from time to time which causes the properties of the final product to vary, often in
an unpredictable way. How can food manufacturers control these variations? Firstly, they can
understand the role that different food ingredients and processing operations play in determining
the final properties of foods, so that they can rationally control the manufacturing process to
produce a final product with consistent properties. This type of information can be established
through research and development work (see later). Secondly, they can monitor the properties of
foods during production to ensure that they are meeting the specified requirements, and if a
problem is detected during the production process, appropriate actions can be taken to maintain
final product quality (McClements, 1999).
Quality control points:
Use only ripe fruits without bruising or damage. Over-ripe ones can easily become
damaged and bruised. Under-ripe fruits will not have the full flavor.
Use a double boiling pan to avoid burning which can occur if direct heating is used.
Weigh all ingredients to the correct formulation.
Do not dry the leather in direct sunlight as there will be loss of color and vitamins A and
C.
Dust the leather lightly with starch before packing to reduce their stickiness.
Seal the leather packed in the form of a roll interleaved with greaseproof paper to avoid it
sticking together.
21

Check the correct fill-weight before sealing the bags.


If available, use 400 gauge polypropylene bags as they provide greater protection against
moisture (Nancy, 2009).
Manufacturers measure the properties of incoming raw materials to ensure that they meet certain
minimum standards of quality that have previously been defined by the manufacturer. If these
standards are not met the manufacturer rejects the material. Even when a batch of raw materials has
been accepted, variations in its properties might lead to changes in the properties of the final
product. By analyzing the raw materials it is often possible to predict their subsequent behavior
during processing so that the processing conditions can be altered to produce a final product with
the desired properties. Monitoring of food properties during processing is advantageous for food
manufacturers to be able to measure the properties of foods during processing. Thus, if any
problem develops, then it can be quickly detected, and the process adjusted to compensate for it.
This helps to improve the overall quality of a food and to reduce the amount of material and time
wasted. Traditionally, samples are removed from the process and tested in a quality assurance
laboratory. This procedure is often fairly time-consuming and means that some of the product is
usually wasted before a particular problem becomes apparent. For this reason, there is an increasing
tendency in the food industry to use analytical techniques which are capable of rapidly measuring
the properties of foods on-line, without having to remove a sample from the process. These
techniques allow problems to be determined much more quickly and therefore lead to improved
product quality and less waste. The ideal criteria for an on-line technique is that it be capable of
rapid and precise measurements, it is non-intrusive, it is nondestructive and that it can be
automated. Once the product has been made it is important to analyze its properties to ensure that it
meets the appropriate legal and labeling requirements, that it is safe, and that it is of high quality. It
is also important to ensure that it retains its desirable properties up to the time when it is consumed
(McClements, 1999).
2.9 Effect of processing on food quality attributes
Foods undergo changes as a result of processing; such changes may be physical, chemical,
enzymatic, or microbiological (Singh & Heldman, 2001). Food processing is any and all processes
to which food is subjected after harvesting for the purposes of improving its appearance, texture,
palatability, nutritive value, keeping properties and ease of preparation, and for eliminating
microorganisms, toxins and other undesirable constituents (David & Arnold,1999).
22

The composition of a food largely determines its safety, nutrition, physicochemical properties,
quality attributes and sensory characteristics. Most foods are compositionally complex materials
made up of a wide variety of different chemical constituents. Their composition can be specified in
a number of different ways depending on the property that is of interest to the analyst and the type
of analytical procedure used: specific atoms (e.g., Carbon, Hydrogen, Oxygen, Nitrogen, Sulfur,
Sodium, etc.); specific molecules (e.g., water, sucrose, tristearin, b-lactoglobulin), types of
molecules (e.g., fats, proteins, carbohydrates, fiber, minerals), or specific substances (e.g., peas,
flour, milk, peanuts, butter). Government regulations state that the concentration of certain food
components must be stipulated on the nutritional label of most food products, and are usually
reported as specific molecules (e.g., vitamin A) or types of molecules (e.g., proteins) (McClements,
1999).
Many unit operations, especially those that do not involve heat, have little or no effect on the
nutritional quality of foods. Examples include mixing, cleaning, sorting, freeze drying and
pasteurization. Unit operations that intentionally separate the components of foods alter the
nutritional quality of each fraction compared with the raw material. Unintentional separation of
water-soluble nutrients (minerals, water-soluble vitamins and sugars) also occurs in some unit
operations (for example blanching, and in drip losses from roast or frozen foods.
Processing using heat is a major cause of changes to nutritional properties of foods. For example
gelatinization of starches and coagulation of proteins improve their digestibility, and antinutritional compounds (for example a trypsin inhibitor in legumes) are destroyed. However, heat
also destroys some types of heat-labile vitamin, reduces the biological value of proteins (owing to
destruction of amino acids or Maillard browning reactions) and promotes lipid oxidation. Therefore
a continuing aim of food manufacturers should be to find improvements in processing technology
which retain or create desirable sensory qualities and nutritional properties or reduce the damage to
food caused by processing (Fellows, 2000).
2.9.1 Physicochemical properties
The physiochemical properties of foods (rheological, optical, stability, flavor) ultimately
determine their perceived quality, sensory attributes and behavior during production, storage and
consumption. The optical properties of foods are determined by the way that they interact with
electromagnetic radiation in the visible region of the spectrum, e.g., absorption, scattering,
transmission and reflection of light. For example, full fat milk has a whiter appearance than skim
23

milk because a greater fraction of the light incident upon the surface of full fat milk is scattered due
to the presence of the fat droplets.
The rheological properties of foods are determined by the way that the shape of the food changes,
or the way that the food flows, in response to some applied force. For example, margarine should
be spread able when it comes out of a refrigerator, but it must not be so soft that it collapses under
its own weight when it is left on a table. The stability of a food is a measure of its ability to resist
changes in its properties over time. These changes may be chemical, physical or biological in
origin. Chemical stability refers to the change in the type of molecules present in a food with time
due to chemical or biochemical reactions, e.g., fat rancidity or non-enzymatic browning. Physical
stability refers to the change in the spatial distribution of the molecules present in a food with time
due to movement of molecules from one location to another, e.g., droplet creaming in milk.
Biological stability refers to the change in the number of microorganisms present in a food with
time, e.g., bacterial or fungal growth. Foods must therefore be carefully designed so that they have
the required physicochemical properties over the range of environmental conditions that they will
experience during processing, storage and consumption, e.g., variations in temperature or
mechanical stress. Consequently, analytical techniques are needed to test foods to ensure that they
have the appropriate physicochemical properties (McClements, 1999).
2.9.2 Changes on Vitamins
Vitamins are minor components of foods which play an essential role in human nutrition. They are
organic compounds that are necessary in small amounts for proper growth. In general human
beings and animals can not be in a healthy state without vitamins, carbohydrates, fats, proteins,
minerals and water. Very small quantities of vitamins are necessary for health, but a lack of them
may upset the normal metabolism, resulting in deficiency diseases. Many of the vitamins are
unstable under certain conditions of processing and storage and their levels in processed foods,
therefore, may be considerably reduced. Most of the vitamins are also heat sensitive. The
occurrence of the vitamins in the various food groups is related to their water or fat solubility.
Vitamins are classified into two main groups: water soluble vitamins and Fat soluble vitamins
(Deman, 1980).

24

2.9.2.1 Effects of preliminary treatments: trimming and washing on Vitamins


The peeling and trimming of fruits and vegetables can cause losses of vitamins to the extent that
they are concentrated in the discarded stem, skin, or peel fractions. Although this can be a source of
significant loss relative to the intact fruit or vegetable, in most cases this must be considered to be
an inevitable loss regardless of whether it occurs in industrial processing or home preparation.
Alkaline treatments to enhance peeling can cause increased losses of labile vitamins such as float,
ascorbic acid, and thiamin at the surface of the product. However, losses of this kind tend to be
small compared to the total vitamin content of the product. Any exposure of cut or otherwise
damaged tissues of plant or animal products to water or aqueous solutions causes the loss of watersoluble vitamins by extraction (leaching). This can occur during washing, transportation via
flumes, and exposures to brines during cooking. The extent of such losses depends on factors that
influence the diffusion and solubility of the vitamin, including pH (can affect solubility and
dissociation of vitamins from binding sites within the tissue), ionic strength of the extract,
temperature, the volume ratio of food to aqueous solution, and the surface-to-volume ratio of the
food particles (Fennema, 1996).
2.9.2.2 Effects of blanching and thermal processing on Vitamins
Blanching, a mild heat treatment is an essential step in the processing of fruits and vegetables. The
primary purposes are to inactivate potentially deleterious enzymes, reduce microbial loads, and
decrease interstitial gasses prior to further processing. Inactivation of enzymes often has a
beneficial effect on the stability of many vitamins during subsequent food storage. Blanching can
be accomplished in hot water, flowing steam, hot air, or with microwaves. Losses of vitamins occur
primarily by oxidation and aqueous extraction (leaching), with heat being a factor of secondary
importance. Blanching in hot water can cause large losses of water-soluble vitamins by leaching. It
has been well documented that high-temperature, short-time treatments improve retention of labile
nutrients during blanching and other thermal processes. The elevated temperature of thermal
processing accelerates reactions that would otherwise occur more slowly at ambient temperature.
Thermally induced losses of vitamins depend on the chemical nature of the food, its chemical
environment (pH, relative humidity, transition metals, other reactive compounds, concentration of
dissolved oxygen, etc.), the stabilities of the individual forms of vitamins present, and the
opportunity for leaching. The nutritional significance of such losses depends on the degree of loss

25

and the importance of the food as a source of the vitamin in typical diets (Da-Silva and Williams,
1991).
2.9.2.3 Effect of processing on Vitamin C
Vitamin C (L-ascorbic acid) is the least stable of all vitamins and will easily be destroyed during
processing and storage. The rate of destruction of vitamin C is increased by the action of metals,
especially copper and iron, and also by the action of enzymes. Exposure to oxygen and light and
prolonged heating in the presence of oxygen during processing will decrease the vitamin C content
of foods. Factors that affect vitamin C destruction during processing include heat treatment and
leaching. The severity of processing conditions can often be judged by the percentage of ascorbic
acid that has been lost. The extent of loss depends on the amount of water used. Vegetables during
blanching covered with water may lose 80% half covered 40% and quarter covered 40% of the
ascorbic acid. Particle size affects the loss, for example in blanching of small pieces of carrots,
losses may range from 32-50% and losses from large pieces only 22-33%. Blanching of cabbage
may result in 20% loss of ascorbic acid and subsequent dehydration may increase this to a total of
50%. In the processing of milk losses may occur at various stages. From an initial level of about
22mg/l in raw milk the content in the product reaching the consumer may be well below 10mg per
liter. Ascorbic acid is oxidized in the presence of air under neutral and alkaline conditions. At acid
pH the vitamin is more stable for example in citrus juice. Since oxygen is required for the
breakdown, removal of oxygen should have a stabilizing effect. For the production of fruit drinks
it is recommended to de-aerate the water to minimize the vitamin C loss. The type of container may
also affect the extent of ascorbic acid destruction. Use of tin cans for fruit juices result in rapid
depletion of oxygen by the electrochemical process of corrosion. In bottles all of the residual
oxygen is available for ascorbic acid oxidation. To account for processing and storage losses it is
common to allow for a loss of 7-14mg of ascorbic acid per 100ml of fruit juice (Fennema, 1996).
2.9.3 Flavor and pigment components
The flavor of a food is determined by the way that certain molecules in the food interact with
receptors in the mouth (taste) and nose (smell) of human beings. The perceived flavor of a food
product depends on the type and concentration of flavor constituents within it, the nature of the
food matrix, as well as how quickly the flavor molecules can move from the food to the sensors in
the mouth and nose. Analytically, the flavor of a food is often characterized by measuring the
26

concentration, type and release of flavor molecules within a food or in the headspace above the
food (McClements, 1999).
Fresh foods contain complex mixtures of volatile compounds, which give characteristic flavors and
aromas, some of which are detectable at extremely low concentrations (Fellows, 2000). These
compounds may be lost during processing, which reduces the intensity of flavor or reveals other
flavor/aroma compounds. Volatile aroma compounds are also produced by the action of heat,
ionizing radiation, and oxidation or enzyme activity on proteins, fats and carbohydrates. Examples
include the Maillard reaction between amino acids and reducing sugars or carbonyl groups and the
products of lipid degradation or hydrolysis of lipids to fatty acids and subsequent conversion to
aldehydes, esters and alcohols. The perceived aroma of foods arises from complex combinations of
many hundreds of compounds, some of which act synergistically (Maruniak and MacKay-Sim,
1984). In addition, the perceived flavor of foods is influenced by the rate at which flavor
compounds are released during chewing, and hence is closely associated with the texture of foods
and the rate of breakdown of food structure during mastication (Clark, 1990).
The colors of foods are the result of the presence of natural pigments or of added dyes. Pigments
are a group of natural colorants found in animal and vegetable products (Deman, 1980). These
pigments are organic in their nature and generally considered to embrace the pigments already
formed in the foods as well as those which can be formed on heating, storage or processing. Many
naturally occurring pigments are destroyed by heat processing, chemically altered by changes in pH
or oxidized during storage. As a result the processed food may lose its characteristic color and
hence its value. Maillard browning is an important cause of both desirable changes in food color
(for example in baking or frying and in the development of off-colors (for example during canning
and drying. Major food processing activities such as ambient temperature processing, processing by
the application of heat and processing by the removal of heat will affect the flavor, aroma and
pigment of food stuffs (Fellows, 2000).
2.9.3.1 Heat induced processing effects on flavor and color
Most foods have no significant changes to flavor or aroma when correctly blanched, but underblanching can lead to the development of off-flavors during storage of dried or frozen foods. In
fruit juices the main cause of color deterioration is enzymatic browning by polyphenoloxidase. This
is promoted by the presence of oxygen, and fruit juices are therefore routinely de-aerated prior to
pasteurization. In fruits and vegetables, chlorophyll is converted to pheophytin, carotenoids are
27

isomerized from 5, 6-epoxides to less intensely colored 5, 8-epoxides, and anthocyanins are
degraded to brown pigments. Changes are due to complex reactions which involve the degradation,
recombination and volatilization of aldehydes, ketones, sugars, lactones, amino acids and organic
acids. In aseptically sterilized foods the changes are again less severe, and the natural flavors of
milk, fruit juices and vegetables are better retained. Aroma compounds that are more volatile than
water can be lost during evaporation. This reduces the sensory characteristics of most concentrates;
in fruit juices this results in loss of flavor, although in some foods the loss of unpleasant volatiles
improves the product quality, for example in cocoa (Anon, 1981).
Heat not only vaporizes water during drying but also causes loss of volatile components from the
food and as a result most dried foods have less flavour than the original material. The extent of
volatile loss depends on the temperature and moisture content of the food and on the vapor pressure
of the volatiles and their solubility in water vapor. Volatiles which have a high relative volatility
and diffusivity are lost at an early stage in drying. Foods that have a high economic value due to
their characteristic flavors (for example herbs and spices) are dried at low temperatures (Mazza and
LeMaguer, 1980).
The open porous structure of dried food allows access of oxygen, which is a second important
cause of aroma loss due to oxidation of volatile components and lipids during storage. Most fruits
and vegetables contain only small quantities of lipid, but oxidation of unsaturated fatty acids to
produce hydro peroxides, which react further by polymerization, dehydration or oxidation to
produce aldehydes, ketones and acids, causes rancid and objectionable odours. Some foods (for
example carrot) may develop an odour of violets produced by the oxidation of carotenes to ionone (Rolls and Porter, 1973). Evaporation darkens the color of foods, partly because of the
increase in concentration of solids, but also because the reduction in water activity promotes
chemical changes, (for example Maillard browning). As these changes are time and temperature
dependent, short residence times and low boiling temperatures produce concentrates which have a
good retention of sensory and nutritional qualities (Anon, 1981). Blanching brightens the color of
some foods by removing air and dust on the surface and thus altering the wavelength of reflected
light. The time and temperature of blanching also influence the change in food pigments according
to their D value (Fellows, 2000).

28

2.9.4 Sensory attributes


The quality and desirability of a food product is determined by its interaction with the sensory
organs of human beings, e.g., vision, taste, smell, feel and hearing. For this reason the sensory
properties of new or improved foods are usually tested by human beings to ensure that they have
acceptable and desirable properties before they are launched onto the market. Even so, individuals'
perceptions of sensory attributes are often fairly subjective, being influenced by such factors as
current trends, nutritional education, climate, age, health, and social, cultural and religious patterns.
To minimize the effects of such factors a number of procedures have been developed to obtain
statistically relevant information. For example, foods are often tested on statistically large groups
of untrained consumers to determine their reaction to a new or improved product before full-scale
marketing or further development. Alternatively, selected individuals may be trained so that they
can reliably detect small differences in specific qualities of particular food products, e.g., the mint
flavor of a chewing gum Although sensory analysis is often the ultimate test for the acceptance or
rejection of a particular food product, there are a number of disadvantages: it is time consuming
and expensive to carry out, tests are not objective, it cannot be used on materials that contain
poisons or toxins, and it cannot be used to provide information about the safety, composition or
nutritional value of a food. For these reasons objective analytical tests, which can be performed in
laboratory using standardized equipment and procedures, are often preferred for testing food
product properties that are related to specific sensory attributes. For this reason, many attempts
have been made to correlate sensory attributes (such as chewiness, tenderness, or stickiness) to
quantities that can be measured using objective analytical techniques, with varying degrees of
success (McClements, 1999).
Many types of food processing techniques have been employed throughout human history, mainly
to ensure microbiological and chemical safety of foods and to improve palatability. Growing
consumer demand for healthy, nutritious and convenient food is a key driver for improvements and
new developments in food processing. New processes or newly recognized compounds, often
identified due to improved analytical capabilities, require careful evaluation of potential human
health impact. The most important quality attributes of a food to the consumer, are its sensory
characteristics (texture, flavor, aroma, shape and color). These determine an individuals preference
for specific products, and small differences between brands of similar products can have a
substantial influence on acceptability. So during processing great care must be taken to retain or
enhance these properties (Fellows, 2000).
29

2.9.5 Influence of drying process


Drying of agricultural products is the oldest and widely used preservation method. It involves
reduction as much water as possible from foods to arrest enzyme and microbial activities hence
stopping deterioration. Moisture left in the dried foods varies between 2-30% depending on the
type of food. In tropical countries, solar dryers can be used to dry fresh produce when average
relative humidity is below 50% during drying period. Drying lowers weights and volume of the
product hence lowers costs in transportation and storage. However, drying allows some lowering in
nutritional value of the product e.g. loss of vitamin C, and changes of color and appearance that
might not be desirable (GTZ, 2009).
Fruits like mangoes, pawpaw, guavas and bananas, can easily be dried. However, they should be
harvested at the right stage and ripeness. Hard ripe stage in mangoes, pawpaw and bananas gives
best results. Avoid overripe, under mature fruits in order to obtain good products. To prepare the
fruits for drying, wash them thoroughly with clean water. Scrubbing with a brush might be
necessary like in case of mango fruit with a lot of latex cover. The fruits are peeled if necessary and
cut into smaller uniform pieces to ensure faster drying. Stainless steel knives are recommended for
peeling and cutting of the slices or pieces. Drum-drying of mango puree is described as an efficient,
economical process for producing dried mango powder and flakes. Its major drawback is that the
severity of heat preprocessing can produce undesirable cooked flavors and aromas in the dried
product. The drum-dried products are also extremely hydroscopic and the use of in-package
desiccant is recommended during storage (Dauthy, 1995).
To avoid discoloration and excessive vitamin losses, treatment with anti-oxidants like citrus
(lemon) juice is done. Fruits like pineapples may require pre-cooking to soften fibrous tissue hence
hasten drying. Drying is done on trays, which should be made of wood, fabric, plastic or sisal
material. This is because metal materials may affect the drying product negatively e.g. copper
destroys vitamin C, iron rusts, aluminum discolors fruits and corrodes. Most fruits have natural
acids and sugars which are preservatives therefore moisture contents of about 20% i.e. leathery and
springy dry (not brittle) is good for storage. This is however dependent on the fruit or vegetable.
After the correct stage of dryness is achieved the product should be removed from the dryer parked,
and stored in a dry, dark store (GTZ, 2009).

30

2.10 Food safety


One of the most important reasons for analyzing foods from both the consumers and the
manufacturers standpoint is to ensure that they are safe. It would be economically disastrous, as
well as being rather unpleasant to consumers, if a food manufacturer sold a product that was
harmful or toxic. A food may be considered to be unsafe because it contains harmful
microorganisms (e.g., Listeria, Salmonella), toxic chemicals (e.g., pesticides, herbicides) or
extraneous matter (e.g., glass, wood, metal, insect matter). It is therefore important that food
manufacturers do everything they can to ensure that these harmful substances are not present, or
that they are effectively eliminated before the food is consumed. This can be achieved by following
good manufacturing practice regulations specified by the government for specific food products
and by having analytical techniques that are capable of detecting harmful substances. In many
situations it is important to use analytical techniques that have a high sensitivity, i.e., that can
reliably detect low levels of harmful material. Food manufacturers and government laboratories
routinely analyze food products to ensure that they do not contain harmful substances and that the
food production facility is operating correctly (McClements, 1999).

31

CHAPTER 3
MATERIALS AND METHODS
3.1 Raw material source and equipment
The raw materials needed for the research are ripened mango fruits. Mango fruits were collected
from Et-fruit, the Keitt variety mangos were brought from Assossa region and the Tommy Atkins
variety of mangos were brought from Awara Melka in Afar region. The honey used as a sweetener
was collected from Gojam area, a typical place called Debre Markos, and traditionally purified at
home. The lemon used for making juice as a source of ascorbic acid for preserving the mango
puree color and as flavoring agent was bought from Atkilt Tera market in Addis Ababa. The spice
ingredient as a flavoring agent was ground ginger obtained from Mercato.
The equipments required for developing Mango leather were scales, plastic containers to wash
fruit, stainless steel knives, spoons, chopping boards, water bath fitted with thermostat, fruit pulper,
large sealable food grade bins for intermediate storage of pulps, convective hot air dryer or drying
oven, and heat sealer.
3.2 Approach for selection and preparation of Mango fruits
The mango market in Addis Ababa is mainly dominated with the local mango varieties. The
mangos brought by Et-Fruit to the market are generally classified as local and export standards.
Normally the export standard mangos are from Tommy Atkins and Kent varieties which are being
cultivated by the Upper Awash Agro-industry farm and are directly exported most of the time.
These varieties are also cultivated by the Awash Melkasa Agricultural Research Institute
(AMARI). However, this research was not conducted at the peak harvest season of these varieties
of mangos. As a result the experimental work has been initiated with the available Keitt varieties.
Preliminary studies were carried out to determine the appropriate tray load of the Mango leather
and drying time to produce satisfactory product using Tommy Atkins and Keitt varieties of mango
fruits. The pH of the puree was measured by pH meter, titerable acidity and total soluble solid
content of the fruit pulps were determined by AOAC methods (1984).

32

3.3 Development of Mango fruit leather


Both Keitt and Tommy Atkins varieties of Mango fruits were brought to Addis Ababa University,
Department of Chemical Engineering Laboratory, and processed individually with the same
procedure. Immediately after the fruits arrival, the ripened and unspoiled fruits were selected,
sorted and washed properly. Some of the mango fruits with minor blemishes and bruises that
couldnt be suitable for other processes like canning or freezing were also used for this leather
product development as the process is based on drying principle. Accordingly, the mangos were
weighed after the blemishes were cut away and were manually peeled with a stainless steel knife.
They were cut and sliced in to pieces, and the stones and peels were also weighed. These were
removed to make fruit pulp ready for making the puree. The pulp was pureed in blender till
becoming smooth. The pure mango puree was covered and put into a glass jar and refrigerated at
40C. The puree mix was prepared with the proportion of 1kg of Mango puree, 39.4g of honey,
18.2g of lemon juice and 3.1g of ground ginger. This was done based on preliminary test as to the
material balance.
3.3.1 Raw material preparation and formulation of the puree mix
3.3.1.1 Keitt Mango variety
First, 20 kg of Keitt variety mango fruits were bought. The mangos were then brought to AAU,
Food Engineering laboratory, washed thoroughly and pureed with a food processor. Thereafter
700ml of mango puree was taken for a nutritional composition analysis at Ethiopian Health and
Nutrition Research Institute (EHNRI) in the Food Science and Nutrition Directorate Laboratory.
The sample was tested for the contents of the protein, fat, moisture, fiber, ash, carbohydrate and
Vitamin C which were believed to be significant for the research.
Two kilograms of honey was bought from Gojam area and processed traditionally at home for the
purpose of mixing and development of the puree mixture as a sweetener. After the mango was
pureed and mixed with lemon juice, honey and ground ginger in duplicate samples for proximate
analysis so as to compare the effect of the ingredients on the nutritional composition and
physicochemical qualities.

33

3.3.1.2 Tommy Atkins variety


Tommy Atkins variety of 20kg Mango fruits were washed thoroughly and pureed with a food
processor. 700ml of mango puree was taken for a nutritional composition analysis at Ethiopian
Health and Nutrition Research Institute (EHNRI) in the Food Science and Nutrition Directorate
Laboratory. The sample was tested for the contents of protein, fat, moisture, fiber, ash,
carbohydrate and Vitamin C with the same procedure used for the Keitt variety.
3.3.1.3 Making fresh Mango puree
The puree was made using a food grinder immediately after the fruits are peeled and de-stoned to
avoid excessive browning. This food processor with appropriate speed was chosen because the
normal fruit blending machine (mixer) was not able to mix the puree properly since water was not
added during the process. The puree was weighed to be 1.65 kg. Then the puree was placed and
covered in a water bath fitted with thermostat to control the temperature and blanched at low
temperature at 600C over boiling water for 10 minutes. This helped for preserving the natural fruit
flavor and light color and also to inactivate the enzymes. The pulp was then strained through a
screen to remove any peels and other unnecessary materials which were not properly handled
before, to obtain a pure puree.
3.3.1.4 Preserving the Mango color
Since light-color fruit leathers tend to darken during drying, the color of the mango leather was
preserved by adding citric acid. Lemon juice was used in the laboratory instead of crystal or tablet
form of ascorbic acid. This was done by addition of 0.03kg of lemon juice per 1.65 kg of mango
puree. The puree was sweetened by adding 0.065kg of pure honey. Honey was chosen as sweetener
instead of sugar to prevent formation of crystals at the end of drying the leather. Honey resulted in
relatively stickier mango leather.
3.3.2 Heating and drying the puree mix
The juicy puree was heated to cook and shorten the drying time and also save energy. This was
done by placing the ground or pureed mango puree in a hot water bath fitted with thermostat
(Model: GP 200, Grant) and a stirrer at a water temperature of 84 0C until the temperature of the
puree reaches 750C while continuously stirred. The puree was cooked over low heat by constantly
stirring until the mixture thickens. The total time required for the cooking process was two hours.
34

The puree was poured and leveled with approximately 0.4 and 0.6g/cm2 puree load on three
aluminum shield trays which were greased with edible corn oil to prevent sticking on the final
leather. Thereafter, 750mg of ground ginger spice was added on the surface of puree in each tray
for flavoring. The trays were placed in the oven with 7.32cm clearance from the top and the bottom
of the trays. The temperature was adjusted between 60 and 800C and drying continued for 10 h.
The dryness of the leather had been regularly inspected during the drying period. The trays had
been turned and rotated in every hour during the drying time. The dryness of the mango leather was
checked by slightly touching with fingers without putting any finger print until the stickiness of the
leather stopped. It took totally 10 h to get the final leather product by using drying oven. Test for
doneness of the process was also carried out by front teeth, and the final mango leather was felt
tacky being slightly sticky to the mouth with very low moisture content. Finally, the weight of the
mango leathers in each tray was determined and the total weight of the product was calculated. The
general process flowchart for making fruit leather is indicated in figure 3.1 below.

35

Fully Ripened Mangos


Sorting
Washing
Peeling
Cutting
Pulping
Crushing
Mixing
Heating (60, 70 and 800C) for 2 h
Removal of foam
Pour the hot puree on to the trays
Convective drying (60, 70, and 800C) for 4-10 h
Packaging with plastic bags
Labeling
Storage in a cool and dry place
Figure 3.1 Process flowchart for Mango leather processing

36

3.4 Methods for studying the effect of processing


Drying was conducted according to a central composite design with two independent variables:
drying temperature (60, 70, and 800C) and puree load (puree mass per dish area, 0.40.6 g/cm2)
according to Azeredo (2006). The ranges of the variables were determined from preliminary tests.
The experimental design was carried out to compare the effects of processing on the qualities of
both varieties of mango fruit leather and suggest the optimum conditions and the minimum time
required for the leathers to achieve a moisture content of 1518%. Triplicate samples were taken at
15 min intervals for 2 h and submitted for moisture content determination. The product obtained
under the optimum conditions was finally submitted to evaluation of the quality parameters
including sensory analyses.
3.4.1 Effect of processing on drying time
3.4.1.1 Drying experiment using convective dryer
The experiment was conducted using a convective dryer in AAU Food Engineering Laboratory to
determine the influence of drying air temperature and puree load on drying time of mango fruit
leather. The dryer consists of a drying chamber, electrical heater, blower and temperature controller
(20 to 80C, dry bulb temperature). During the experiment, the drying air velocity was set at 0.5
m/s and the relative humidity of the air was 651%. The puree load of 0.4, 0.5, 0.6g/cm2 were
uniformly spread in square stainless steel meshed tray sheeted with aluminum foil.
The convective dryer was switched on at the beginning of each test and the required temperature
was set. When the dryer reached the steady state condition, the mango puree samples were
uniformly spread in square stainless steel meshed trays size (10cm X 10cm) covered with
aluminum foil and kept in the drying chamber of the dryer. Air flow was parallel to the surface of
the Mango puree. The drying was continued until the moisture content of the samples dropped to
15 to 18% (Azeredo et al., 2006).
The moisture content of the dried samples was measured by placing the samples at 105C in MB45
Moisture Analyzer. More than 1g of sample was taken for each analysis of the product from both
mango varieties treated with different temperatures and puree load. Determination of the moisture
content was done in triplicate.
37

3.5 Quality parameters for studying the effect of processing


The effect of processing during development of mango fruit leathers was studied by analyzing the
moisture loss and the texture of the final product, conducting proximate analysis, physico-chemical
analysis and sensory evaluation. The influence of temperature and the puree load on the drying
dishes was also studied to find out how the drying process affects the qualities of the mango fruit
leather and how long will it take to dry the fruit leather.
3.5.1 Proximate analysis methods for the puree and leather
The proximate analysis was conducted at Ethiopian Health and Nutrition Research Institute
(EHNRI), and all the moisture content, protein, fat, crude fiber, carbohydrate and ash were
determined according to AOAC methods. All of the chemicals and reagents used for the analysis of
the mango puree, puree mix and the final fruit leather were supplied by the Food Laboratory of
EHNRI.
3.5.1.1 Moisture determination
The amount of water present in the mango sample was considered to be equal to the loss of weight
after drying the sample to constant weight at a temperature of 100oC according to Method AOAC
(2003).
Formula:

% Moisture

W2 * 100
W1

%Totalsolids

W3 * 100
W1

Where: W1 , W2 and W3 are in wet base.

W1 = weight of wet sample


W2 = loss of weight
W3 = weight of dry sample
3.5.1.2 Crude fiber
38

The crude fiber content of the Mango sample was analyzed according to Method AOAC (2003); a
fat-free or low fat content sample is treated with boiling sulfuric acid and subsequently with boiling
potassium hydroxide or sodium hydroxide, the residue after subtraction of the ash is regarded as
fiber.
Formula:
Crude fiber g / 100 g

W1 W2 100 M
W3

Where: W1 = Crucible weight before drying (g)

W2 = Crucible weight after drying (g)


W3 = Sample dry weigh (g)
M = Moisture content of the sample (%)

3.5.1.3 Crude protein


The crude protein content of the Mango sample was analyzed according to Method AOAC (1984);
all nitrogen is converted to ammonia by digestion with a mixture of concentrated sulfuric acid and
concentrated orthophosphoric acid containing potassium

sulfate as a boiling point raising agent

and selenium as a catalyst. The ammonia released after alkalinization with sodium hydroxide is
steam distilled into boric acid and titrated with sulfuric acid.
Formula:

mg nitrogen T B * N *14
g nitrogen / 100 g sample

Total nitrogen %

mg ofnitrogen *100
mg sample

T B * N *1.4 *100
W

Crude Protein = total nitrogen (%) * 5.78


Where: T = volume of sulfuric acid solution used in the titration of test materials.
39

B=Volume of sulfuric acid solution used in the titration for the blank.
N = normality of the acid
14 = Eq. wt of nitrogen
W = wt. of the sample
3.5.1.4 Fat content
The fat content of the Mango sample was analyzed according to Method AOAC (1984); fat is
extracted with ether (peroxide free) from dried samples in a soxhlet apparatus, and the other is
evaporated from the extraction flask. The amount of fat is calculated from the difference in weight
of the extraction flask before and after extraction.
Formula:

W W2 W1
Fat g / 100 g fresh sample

W * 100 % moisture
WD

Where: W= weight of fat (g)

W2 = weight of extraction flask after extraction (g)


W1 = weight of extraction flask before extraction (g)
WD = weight of dried sample (g)
3.5.1.5 Ash content
The organic matter is burned off at low temperature and the inorganic materials remaining are
cooled and weighed. Heating is carried out in stages, first to derive the water, then to char the
product thoroughly and finally to ash at 550oC in a muffle furnace (AOAC, 1984).

Formula:

40

% Ash

W2 *100
W1

Where: W1 = weight of sample

W2 = weight of ash

3.5.1.6 Carbohydrates
The total carbohydrate contents of the mango sample (C %) by mass including crude fiber can be
obtained as follows:
Formula:

C % 100 P F A M
Where: P The mass percent of protein
F The mass percent of fat
A The mass percent of ash
M Moisture content (%)
Therefore, the Utilizable Carbohydrate (CHO) = Total CHO Crude fiber
3.5.2 Physicochemical analysis
The physico-chemical analysis for the mango puree (including: pH, moisture content, and
viscosity) was undertaken at AAU Chemical Engineering laboratory. The pH was measured using
pH meter. The TSS was measured by hand refractometer, titrable acidity was determined by
titration method, Brix /acidity ratio was calculated and the vitamin C content was also measured
using Spectrophotometric method at EHNRI.
3.5.2.1 Determination of viscosity of the Mango puree
The viscosity of the puree was measured by Vibro viscometer (SV-10, JAPAN), (2001) model
which works between 0.3 and 10,000 mPa.s at different temperatures during the heating process on
continuous mixing. Mango puree sample was poured into the cup until viscosity surface reaches
between the level gauges. The level gauge indicates between 35 and 45ml. The cup was then

41

attached on the table along the guides, gently lowering the sensor plates above the sample surface
and the viscosity was measured.
3.5.2.2 Determination of Vitamin C (Total Ascorbic acid) content
The vitamin C content of the mango fruit puree and leather were determined as to the laboratory
manual of EHNRI Food Chemistry Laboratory using Spectrophotometric method. The reagents
used were deionized water, 9NH2SO4, 85% H2SO4, 5% meta phosphoric acid, HPO3, H3PO4, 2%
2.4-DHPH in 100 ml of 9N H2SO4, 2% Thiourea, saturated bromine solution, 6% Trichloroacetic
acid and ascorbic acid standard. The apparatus used were: Grinder, Centrifuge, Filtration apparatus
(funnel, vacuum pump, and filter paper), pipette, test tubes, volumetric flasks, Erlenmeyer flasks
(250 ml), water bath apparatus and ice bath, and UV-Vis Spectrophotometer.
The procedure used for the determination of the Vitamin C content of the mango fruit leather
includes: extraction, oxidation to Dehydroascorbic acid, formation of Osazone, formation of
soluble pigments (Treatment with 85% H2SO4), measurement of Color, and development of
standard curve.
The calculation to o
btain the vitamin C content of the fruit leather was done using the following formula:
Mg ascorbic acid/100g

A(sample) - A(blank)
A(10 g std) - A(blank)

Where, A(sample) = Absorbance of the sample


A(blank) = Absorbance of the blank
A(10 g std) = Absorbance of 10 g standard
3.5.3 Texture analysis of the Mango leather
Texture analysis was done in the Food Engineering laboratory using Texture Analyzer (Model:
TA-Plus, LLOYD instrument, England). The software used for analysis of the texture was
NEXYGEN Plus. For this laboratory experiment, Volodkevitch bite test was used. This test was
designed to imitate incisor teeth sharing through a food sample. The set comprises upper and lower
teeth which during the test, are brought together until nearly touching. The samples were
42

positioned on the lower tooth and the results of the peak force required to bite through the
samples was measured. A preliminary test was conducted for analysis of the textural characteristics
of the Keitt mango fruit leather. The texture analyzer setting was kept at preload stress 7.3N,
preload stress speed of 21mm/min and a test speed of 100 mm/min. The height of the mango
leather sample was set at 5mm, the width was 10mm and the area was 50mm2 for each analysis.
The experiment was conducted for different layers of mango leathers to find out how much layers
of mango leathers could be suitable for biting. This compression test simulates the evaluation of
toughness by consumer for a single bite of the mango leather. The absolute peak force from the
resulting curve was considered as the amount of force required to compress the mango leather with
front teeth. The maximum shear force required for cutting the mango leather was taken as the
texture index of the dried product. The force-displacement and energy used for cutting the leather
was recorded (Pushpa, 2006).
3.5.4 Microbiological analysis
The microbiological analysis was conducted at EHNRI for two samples selected from the final
products of both mango varieties which were treated at 600C temperature and 0.6g/cm2 puree load.
They were selected on the basis that microorganisms grow as the drying temperature was lower and
the puree load was greater in relation to the other developed products.
Determination of Mold and Yeast was conducted using NMKL, No. 98, 1997 method (Annex 1).
Aerobic Plate Count (APC) was determined as to NMKL, No. 86, 2006 (Annex 2). Coliform count
was carried out according to NMKL, No. 44, 2004 (Annex 3). Fecal coliform count and E. coli was
determined by FDA/BAM, 2006 (Annex 4). Determination of S. aureus count and B. cereus was
done using NMKL, No. 66, 2003 (Annex 5). Isolation of Salmonella, Shigella, and Stapylococcus
spp. was carried out by NMKL, No. 71, 2005 (Annex 6).

43

3.5.5 Sensory evaluation


The sensory tests were conducted with 20 panelists. The panelists were asked to give their
individual ratings on all the leather characteristics including color, aroma, taste, flavor, toughness
and overall acceptability of the Mango leather. A 9-point structured hedonic scale (1 extremely
disliked to 9 extremely liked) was used to conduct the preference test.
3.6 Experimental design and data analysis
A factorial arrangement in a completely randomized design was set to see the effects of processing
on the qualities of both varieties of mango fruit leather and suggest the optimum conditions and the
minimum time required for the leathers to achieve a moisture content of 1518%. The experiment
was undertaken to establish the optimum variables required to achieve the best product
characteristics.
The experiment was repeated two times in duplicate samples. Two independent variables were
selected for both varieties of mango fruit:
1. Drying temperature (60 to 80 0C)
2.

Puree load (puree mass per dish area, which was in the range of 0.4 to 0.6g/cm2,
(Azeredo et al., 2006).

Analysis of variance was done and level of significance was set at 5%. The data obtained was
statistically analyzed by using Statistical Package for Social Scientists (SPSS 17th Version).

44

CHAPTER 4
RESULTS AND DISCUSSION
4.1 Physicochemical properties of Mango puree
The preliminary test results of the physico-chemical analysis of the mango pulp for both varieties
of fruits are presented in Table 4.1. The selected physico-chemical parameters which may influence
the quality of mango fruit leather were pulp moisture content, TSS, pH, acidity, Brix /acidity and
the vitamin C content.
Table 4.1 Physico-chemical properties of fruit pulp for the mango varieties
Variety

Moisture
(%)

TSS
0

pH

Brix

Acidity

Brix/Acid

Vitamin C

g/100g

Ratio

(mg/100g)

Keitt

83.42

16.34

4.10

0.25

65.36

47.77

Tommy

85.33

17.23

4.60

0.19

90.68

29.91

Atkins

The results measured in the preliminary study indicated in the Table shows that the TSS, pH,
titrable acidity are within the range for chemical composition of common ripe mango cultivars
grown in different countries (Doreyappa & Ramannjaneya, 1995). The report indicates that the
composition and properties of ripe mango range: TSS (14.60-22.27Brix), pH (3.80-4.90), acidity
(0.11-0.55 %) and total sugars (9.28-20.90%). They stated that the physico-chemical characteristics
of fruits and the technological qualities of the products processed vary with the variety of the fruit.
Some varieties are most suitable for specific application. The TSS of Keitt mango puree was less
than that of Tommy Atkins, whereas the titrable acidity and vitamin C content of the Keitt mango
puree were greater than that of the Tommy Atkins.
The physiochemical properties of foods (rheological, optical, stability, flavor) ultimately
determine their perceived quality, sensory attributes and behavior during production, storage and
consumption (McClements, 1999). Physico-chemical characteristic of four mango varieties
45

(palmer, keitt, Ameliorel and mango) were studied (Germain et al., 2003). It was mentioned that a
pulp extracted from ripe fruit of the stated cultivars has a pH range of 3.91-4.35. Rameshwar
(1993) stated that mango is considered to be matured when the total soluble solids, a measure of
sweetness, reach 8 Brix. Kalra et al. (1995) also pointed out that harvesting is usually initiated
when a few mango fruits on the tree begin to ripen and fall. Optimum maturity stage lead to
uniform ripening and better storage life of about 1 week at 35 0.1C and 65 1% RH ( Jha et al.,
2004).
4.2 Proximate analysis results of the Mango puree and puree mix
The proximate analysis result for both varieties of mango puree and puree mix is presented on table
4.2 below.
Table 4.2 Proximate analysis result for Keitt and Tommy Atkins varieties of Mango puree
and puree mix
Sample Type
Mixed Mango Puree + Honey +

Mango Puree
Component

Lemon Juice
Keitt

Tommy Atkins

Keitt

Tommy Atkins

Moisture (%)

83.42

85.33

78.92

78.81

Fat (%)

0.65

0.16

0.83

0.22

Protein (%)

0.50

0.44

0.56

0.44

Ash (%)

0.30

0.36

0.37

0.37

Crude Fiber (%)

0.54

1.68

0.92

1.68

Carbohydrate (%)

14.59

12.03

18.4

18.48

Vitamin C (mg/100g)

47.77

29.91

53.85

44.64

46

4.2.1 Keitt Mango variety


Generally the proximate analysis result shows that there is an increase in nutritional content of the
mango puree after enriched with the honey and lemon juice. Besides a decrease in moisture content
has also observed which is suitable for drying purpose with less removal of moisture from the
puree mix.
4.2.2 Tommy Atkins variety
The result shows that there is a decrease in moisture content of the puree mix and an increase in fat,
protein and ash, and carbohydrate contents. The ash remained constant. The vitamin C content has
also increased possibly due to the effect of the addition of the lemon juice. Therefore, mixing of the
Tommy atkins mango puree with honey, lemon juice and ground ginger generally increased the
nutritional content.

47

4.3 Effect of heating temperature on the viscosity of Mango puree mix


According to the study, the effect of temperature on the viscosity of the puree mix for both verities
of mangoes was determined in 1.5h period and the results are discussed below.
Table 4.3 Results of viscosities for Keitt and Tommy Atkins variety mango puree mixes
measured at different heating temperature
Keitt

Tommy Atkins

Temperature (0C) Viscosity (m.pa.s) Temperature (0C) Viscosity (m.pa.s)


25.1

964

25.1

510

30.0

955

30.3

485

35.3

948

35.0

443

40.2

913

40.1

426

45.0

869

45.2

414

50.1

818

50.0

401

55.3

797

55.3

393

60.7

791

60.2

405

63.9

804

65.1

555

65.0

820

67.0

625

68.1

841

68.1

676

69.0

845

70.0

915

48

4.3.1 Keitt Mango variety


These results can easily be graphically illustrated below in Figure 4.1.

Figure 4.1 Viscosity verses temperature of Keitt variety puree mix


The result generally showed the decrease in viscosity at the beginning from 964m.pa.s to 791m.pa.s
at different treatment temperatures. Then at 63.90C, the viscosity of the puree mix started to rise to
804m.pa.s and increased afterwards up to 915m.pa.s at 70.00C. At the final measured viscosity, the
puree mix was found cooked and the odor was a bit changed. So, the measured puree was finally
decided to be out of the water bath and became ready for the next drying process. From the result
of the laboratory experiment it can be concluded that the viscosity of mango puree is temperature
dependent. As the temperature increases, the viscosity of the puree first decreases and then it
increases within the temperature range of 25.1 to 70.0 0C.

49

4.3.2 Tommy Atkins variety


These results can easily be graphically illustrated below in Figure 4.2.

Visco.
(m.pa.s)

TemperatureVsViscosity

800
700
600
500
400
300
200
100
0

Figure 4.2 Viscosity verses temperature of Tommy Atkins variety puree mix
The result indicated that a decrease in viscosity at the beginning from 510m.pa.s to 393m.pa.s at
different treatment temperatures. At 60.20C, the viscosity of the puree mix started to rise to
405m.pa.s and increased afterwards up to 676m.pa.s at 68.10C. Finally, the puree mix was found
cooked and the odor was changed. So, the measured puree was finally decided to be out of the
water bath and became ready for the next drying process. This shows that the viscosity of mango
puree is also temperature dependent. As the temperature increases, the viscosity of the puree first
decreases and increases within the range of 25.10C to 68.1 0C. From the experiment, it has been
found that the viscosity of the Tommy Atkins mango puree mix is generally lower than that of the
Keitt variety mango puree mix.

50

4.4 pH of the puree mix


The pH of the Keitt variety puree mix on average was 4.1 and that of the Tommy Atkins puree mix
was 4.6. With the addition of citric acid, lemon juice was added to bring the pH of the mango puree
mix down to 3.9 to make it safe for microbial effects.
4.5 Effect of drying on the proximate and Vitamin C content of fruit leather
After the mango puree was dried with the combination of different drying temperature and puree
load for both mango varieties, the fruit leather was subjected to nutritional composition analysis in
Ethiopian Health and Nutrition Research institute (EHNRI). The laboratory analysis result is
presented in Table 4.4 below.
Table 4.4 Proximate analysis result for Mango fruit leathers
Sample Code
Component

Moisture (%)

16.25

19.36

15.59

17.12

15.64

18.76

14.86

15.96

14.42

13.21

14.57

15.13

13.69

14.89

Fat (%)

1.98

2.27

4.68

1.61

2.23

1.99

3.55

1.76

2.47

2.85

4.12

0.95

0.33

0.19

Protein (%)

2.21

1.75

2.20

2.27

2.81

2.47

2.59

2.44

2.46

2.76

2.31

2.09

2.30

2.26

Ash (%)

1.84

1.41

2.04

1.89

2.19

1.93

2.29

2.10

1.66

2.16

2.29

2.08

1.52

1.94

Crude Fiber
(%)

6.52

5.14

5.77

7.41

7.11

7.97

7.37

6.70

5.57

6.87

5.00

4.59

4.02

5.34

Carbohydrate
(%)

71.2

70.07

69.72

69.70

70.02

66.88

69.34

71.04

73.42

72.15

71.71

75.16

78.14

74.66

Vitamin
(mg/100g)

44.44

30.88

23.57

22.91

31.91

23.76

23.08

15.59

15.70

14.98

33.95

17.19

23.61

33.92

Where, the sample codes were designed for both varieties of mango with drying temperature and
puree load as:

51

Table 4.5 Designed sample codes for both varieties of mangos with drying temperature and
puree load
Keitt

Tommy Atkins

Temperature (0C)

0.4g/cm2

0.6 g/cm2

0.4 g/cm2

0.6 g/cm2

60

70

80

Code M and N is product samples experimented with convective hot air dryer for comparison
of the effect of processing using drying oven. Product M was developed at 700C with 0.6 g/cm2
puree load from Keitt mango variety where as product N was developed from Tommy Atkins
Mango with the same condition.
4.6 Effect of temperature and puree load on drying time
4.6.1 Drying characteristics of Mango puree mix
The effect of drying air temperature and puree load on drying time taken to reach the final moisture
content (15 to 18%) is presented in Table 4.6. The drying time was generally shorter at higher
temperatures due to quick removal of moisture. The puree load also affected the drying time at all
drying temperatures. Drying time was considerably prolonged for more than 7 hrs for 0.6g/cm2
puree at all drying air temperatures. Drying at higher temperature, 80C reduced the required
drying time by 42.9 %, 25% and 30 % for 0.4 g/cm2, 0.5 g/cm2, and 0.6 g/cm2 puree loads
respectively.

52

Table 4.6. Effect of drying air temperature and puree load on drying time of Keitt Mango
fruit leather
Drying Time of Mango Fruit Leather (h)
Drying Temperature (0C)

0.4 g/cm2

0.5 g/cm2

0.6g/cm2

60

10

70

80

The moisture content of the mango fruit leather was measured in every hour during the drying
period until the final moisture content of the leather reaches 15% to 18%. More than 1g of sample
was being taken for each analysis of the products from both mango varieties treated with different
temperature and puree loads.
4.6.2 Analysis of moisture loss during drying process
The analysis of moisture loss of the leather after drying was determined and reported in Table 4.7.
The result of the analysis showed 70.3% moisture loss after drying process is completed.
Table 4.7 Weights and moisture loss of Keitt variety mango leather
Mango and Mango Products

Weight (kg)

Mangos

3.2

Puree before drying

1.75

Leather after drying

0.52

% Moisture Loss

70.3%

53

4.7 Texture analysis of the Mango leather


A preliminary test was conducted for analysis of the textural characteristics of the Keitt mango fruit
leather. The measurement of texture of the mango leather product was believed to be significant in
that it would be useful for correlating the toughness of the final mango leather product with that of
the sensory analysis results. This could also be useful as a reference and for selecting the best
product characteristics in terms of textural quality. The results are given in Table 4.8.
Table 4.8 Results of compression test using Texture Analyzer to different number of Mango
leather sheets (layers)
Mango

leather

trials Thickness

Extension

Width

with Different number of (mm)

Limits

(mm)

sheets

(mm)

Load (N)

Time (s)

Batch 1 (4 sheets)

10

2.4

2.9

Batch 2 (5 sheets)

10

22.9

3.5

Batch 3 (6 sheets)

10

Exceeds Limit

3.2

Batch 4 (6 sheets)

10

Exceeds Limit

3.0

Batch 5 (6 sheets)

10

3.4

3.5

It was observed that the load needed to compress (or to bite as a consumers teeth) for 6 sheets of
mango fruit leather with 7mm thickness and extension limits of the texture analyzers probe set at
6mm and 5mm, exceeded the limit. Therefore, it is recommended that fruit leathers with 4 sheets
and 5 sheets having 5mm and 6mm thickness respectively could be suitable for a single bite.

54

4.8 Proximate analysis of Mango fruit leather


Table 4.9 Effect of drying temperature on proximate composition of mango leather
Proximate composition

Temperature

Moisture

Protein

Fat

Fiber

Ash

Carbohydrate

(0C)

(%)

(%)

(%)

(%)

(%)

(%)

60

17.080a

2.108a

2.637a

6.231a

1.728a

70.217a

70

16.305b

2.580b

2.381a

7.309b

2.127b

69.284b

80

14.333c

2.407c

2.597a

6.454c

2.048b

72.161c

a-c

Means bearing the same letters in the same column are not significantly different at P < 0.05

Table 4.10 Effect of puree load on proximate composition of mango leather


Proximate composition
Moisture

Protein

Fat

Fiber

Ash

Carbohydrate

Puree Load (g/cm2)

(%)

(%)

(%)

(%)

(%)

(%)

0.4

15.222a

2.432a

3.171a

6.510a

2.052a

70.518a

0.6

16.590b

2.298b

1.906b

6.819b

1.884b

70.590a

a-b

Means bearing the same letters in the same column are not significantly different at P < 0.05

55

Table 4.11 Effect of fruit variety on proximate composition of mango leather


Proximate composition
Moisture

Protein

Fat

Fiber

Ash

Carbohydrate

Mango Variety

(%)

(%)

(%)

(%)

(%)

(%)

Keitt

16.273a

2.411a

2.298a

6.626a

1.864a

70.518a

Tommy Atkins

15.538b

2.318b

2.778b

6.703b

2.071b

70.590a

a-b

Means bearing the same letters in the same column are not significantly different at P < 0.05

4.8.1 Moisture content


The analysis shows that there is a significant difference (P < 0.05) between moisture content means
of the mango leathers due to temperature, puree load and fruit variety. The average moisture
content of the fruit leather is dependent on the drying temperature. As the drying temperature
increases from 600C to 700C and then 800C, the average moisture content decreases to 17.08%,
16.31%, and 14.33% respectively (Table 4.9).
The results in Table 4.10 also indicate that the moisture content is dependent on the puree load.
When the puree load was 0.4g/cm2, the moisture content was 15.22% and when the puree load
increased to 0.6g/cm2, the moisture content of the leather has also increased to 16.59%.
On Table 4.11, it was indicated that the moisture content of the leather is also dependent on the
mango varieties. The moisture content of the Keitt variety mango leather 16.27% was greater than
the Tommy Atkins mango leather 15.53%. When compared to the fresh puree mix, the moisture
content of the Keitt variety mango leather has decreased from 78.92 to 16.27 with 79.38% and that
of Tommy Atkins from 78.81% to 15.53% with 79.24% because of the drying effect. According to
Azeredo, et al., (2006) the average moisture content of mango fruit leather is in the range of 15% to
18%. Therefore, except for 14.33% average moisture content of the mango leather which was dried
at 800C, the other fruit leathers have an average moisture contents within the mentioned range.

56

4.8.2 Protein content


From the analysis, a significant difference (P < 0.05) has been observed on the average protein
content with all drying temperature, puree load and fruit variety. As indicated in Table 4.9, the
average protein content of the fruit leather 2.58% was relatively higher at 700C drying temperature
than that of at 600C and 800C which are 2.108% and 2.40% respectively. Table 4.10 also shows
that there is an increase in average protein content with an increase in puree load. At 0.4g/cm2
puree load, the average protein content was 2.43% and at 0.6 g/cm2, it was 2.29%. The fruit variety
has also an effect on the average protein content of the fruit leather. The mango leather from Keitt
variety has an average protein of 2.41% and 2.31% for Tommy Atkins (Table 4.11). It has been
observed that there is a difference in the average protein content with the mango variety of the
puree mix, 0.56% for Keitt and 0.44% for Tommy Atkins in Table 4.2 and 4.3 respectively.
However, the results show that the average protein content of the Keitt variety mango was greater
than that of the Tommy Atkins for both puree mix and final leather.
4.8.3 Fat content
The mean fat contents of the fruit leathers were significantly different (P < 0.05) as they are
affected by the fruit variety and the puree load. However, drying temperature didnt have a
significant effect on the fat content indicated on Table 4.9. In Table 4.10, it can be observed that
when the puree load increases from 0.4 g/cm2 to 0.6g/cm2, the average fat content of the fruit
leather decreases from 3.17% to 1.90% respectively. As indicated in Table 4.11, the Keitt variety
fruit leather has an average fat content of 2.29% and that of Tommy Atkins is 2.77%.

57

4.8.4 Crude fiber content


The mean crude fibers of the mango fruit leathers were also significantly different from each other
(P < 0.05). All drying temperature, puree load and fruit variety affected the composition. It was
observed that the mean difference of the crude fiber for all drying temperature, was significantly
different. As indicated in Table 4.9, at 700C drying temperature, the average fiber content 7.30%
was relatively higher than that of the fiber contents 6.23% and 6.45% treated with 600C and 800C
drying temperature respectively. In Table 4.10, it is indicated that as the puree load increases 0.4
gcm2 to 0.6gcm2 the average fiber content also increases 6.51% to 6.81%. It was observed that
(Table 4.11), the Keitt mango leather has lower 6.62% crude fiber than the Tommy Atkins mango
leather 6.70%. As indicated in Table 4.2 and 4.3, the Keitt mango puree mix has 0.92% which is
lower than that of the Tommy Atkins (1.68%).
4.8.5 Ash content
With the exception of temperature, the ash content of the fruit leather was significantly affected (P
< 0.05) by the puree load and the fruit variety. It was observed in Table 4.10 that the average ash
content of the fruit leather decreased from 2.05% to 1.88% when the puree load increased 0.4 g/cm2
to 0.6g/cm2. As indicated in Table 4.11, the Keitt mango leather has an average ash content of
1.86% while Tommy Atkins mango leather has 2.07%.

58

4.8.6 Carbohydrate content


The analysis showed that drying temperature has a significant effect on the average carbohydrate
content of the mango leather (P < 0.05) From Table 4.9, higher average carbohydrate content
72.161% was observed at 800C drying temperature in relation to that of 600C and 700C with
70.21% and 69.28% respectively. Both the puree load and the fruit variety did not have a
significant effect on the carbohydrate content.
4.8.7 Effect of processing on Vitamin C content of Mango leather
Vitamin C (L-ascorbic acid) is the least stable of all vitamins and will easily be destroyed during
processing and storage. Exposure to oxygen and light and prolonged heating in the presence of
oxygen during processing will decrease the vitamin C content of foods (Fennema, 1996).
Table 4.12 Effect of drying temperature, puree load and fruit variety on vitamin C content
Factors

Temp. (0C)

Puree Load (g/cm2)

Variety

a-g

Vitamin C (%)
60

30.446a

70

23.585b

80

20.445c

0.4

28.773d

0.6

20.877e

Keitt

26.937f

Tommy Atkins

22.714g

Means bearing different letters in the same column are significantly different at P < 0.05

The average vitamin C content of the fruit leather was significantly affected (P < 0.05) by all the
drying temperature, puree load and fruit variety. A significant effect on the average vitamin C
content of the fruit leather was observed for all drying temperatures 600C, 700C, and 800C. When
the drying temperature and puree load increased, the average vitamin C content decreased. The
average vitamin C content of the fruit leather was also dependent on the fruit variety. The vitamin
C content of the Keitt mango leather (26.93%) is greater than that of Tommy Atkins mango leather
59

(22.71%). When compared to the fresh puree mix, the Keitt mango leather is decreased by 39.66%
from 44.64% to 26.93% and the Tommy Atkins mango leather is decreased by 57.82% from
53.85% to 22.71%.
4.9 Microbiological analysis of Mango fruit leather
The result of the microbiological analysis for the two fruit leather is given in Table 4.13
Table 4.13 Result of microbiological analysis of mango fruit leather
Result for Samples (cfu/g)

B2 (0.6g/cm2) at 600C

D2 (0.6g/cm2) at 600C

Mold count at 250C/5-7 days

6X104

5X102

Yeast count at 220C/5-7 days

1X104

<1X101

APC at 300C/72 h

1X105

1.02X105

Coliform count

<1X101

<1X101

Fecal coliform count

<1X101

<1X101

E.coli count

<1X101

<1X101

S.coccus spp

Not isolated/25g

Not isolated/25g

Salmonella spp

Not isolated/25g

Not isolated/25g

Shigella spp

<1X104

5.2X103

Parameters (Tests)

i.e. APC = Aerobic bacteria plate count


Product B2 is Keitt mango leather developed at 600C and 0.6g/cm2 puree load where as product D2
is Tommy Atkins mango leather developed at 600C and 0.6g/cm2 puree load.
In the counts < 1 X 101 is the standard reporting format for plates from all dilution of the sample
has no colonies. The microbiological result for yeast, coliform, fecal coliform, E. coli counts and
Shigella species for both varieties of mango leathers was found to be safe < 1 X 104. S.coccus and
60

Salmonella species were not isolated for both products. The mold and aerobic bacterial plate counts
were found to be higher. This could be as a result of unhygienic condition during the process of the
product development. However, the product had been subjected to sensory analysis and there was
no harm in any of the panelists which indicates that there is no significant effect on health.
4.10 Sensory analysis of Mango fruit leather
4.10.1 Effect of drying temperature, puree load and fruit varieties on the sensory qualities
Table 4.14 Effect of drying temperature on sensory qualities
Sensory qualities
Overall
Temperature

acceptability

Color

Aroma

Taste

Flavor Toughness

60

7.388a

6.988a

7.012a

7.013a

6.300a

7.175a

70

7.075a

6.763a

7.088a

6.763a

6.125a

6.838a

80

6.788a

6.687a

6.813a

6.825a

5.462b

6.475b

(0C)

a-b

Means bearing the same letters in the same column are not significantly different at P < 0.05

Table 4.15 Effect of puree load on sensory qualities


Sensory qualities
Overall
Puree Load

acceptability

Color

Aroma

Taste

Flavor

Toughness

0.4

6.983a

6.775a

6.817a

6.742a

5.525a

6.575a

0.6

7.183a

6.850a

7.125a

6.992a

6.400b

7.083b

(g/cm2)

a-b

Means bearing the same letters in the same column are not significantly different at P < 0.05
61

Table 4. 16 Effect of fruit variety on sensory qualities


Sensory qualities
Mango Variety

Overall acceptability
Color

Aroma

Taste

Flavor Toughness

Keitt

6.083a

6.475a

6.600a

6.483a

5.508a

6.250a

Tommy Atkins

8.083b

7.150b

7.342b

7.250b

6.417b

7.408b

a-b

Means bearing the same letters in the same column are not significantly different at P < 0.05

4.9.1.1 Color
Color is one of the quality parameters of fruit leathers. The color of the fruit leather was
significantly affected (P < 0.05) only by the fruit variety. On average, the color of the fruit leather
dried at 600C and 700Cwas liked moderately by the respondents and the other which was dried at
800C was preferred slightly (Table 4.14). Moreover, the color of the fruit leather dried at 0.4 g/cm2
and 0.6 g/cm2 was approximately liked moderately by the respondents (Table 4.16). The
respondents also liked very much the Tommy Atkins fruit leather, whereas for Keitt mango fruit
leather, it was only slightly liked. Therefore, the Tommy Atkins Variety mango fruit leather was
relatively of good quality based on the panelists preference. The color difference between the two
mango varieties is shown in the pictures below.

62

Pic. 4 Tommy Atkins Mango fruit leather

Pic. 5 Keitt Mango fruit leather


4.9.1.2 Aroma
The aroma of the mango fruit leather is another quality parameter to grade the final products. From
the sensory analysis result, it was observed that the aroma of the fruit leather was significantly
affected (P < 0.05) by the mango variety. The drying temperature and the puree load did not have
significance effect on the aroma. Table 4.14 and 4.15 indicate that the color of the fruit leather
dried at all temperatures 600C, 700C, and 800C and puree load respectively was moderately liked by
the respondents having a value of 7. Table 4.16 shows that the aroma of the Tommy Atkins mango
fruit leather was liked moderately and that of the Keitt variety was slightly liked. Therefore, based
on the panelists preference, the Tommy Atkins mango fruit leather had relatively a better aroma
than that of the Keitt variety.
63

4.9.1.3 Taste
The taste of the mango fruit leather was not significantly affected (P < 0.05) by the drying
temperature and the puree load. Table 4.14 and 4.15 show that the taste of the fruit leather was
liked moderately by the respondents on average. However, the taste of the fruit leather was
significantly affected by the fruit variety. Table 4.16 indicates that the Keitt variety mango fruit
leather was liked slightly by the respondents, where as the Tommy Atkins fruit leather was liked
very much. Therefore, it can be concluded that the taste of the Tommy Atkins mango leather had a
better quality in terms of taste.
4.9.1.4 Flavor
The flavor of the mango fruit leather was not significantly affected (P < 0.05) by the drying
temperature and the puree load. In Table 4.14 and 4.15, it is indicated that the flavor of the fruit
leather was moderately liked by the respondents approximately. However, the flavor of the fruit
leather was significantly affected by the variety of the mango fruit. Table 4.16 indicates that the
flavor of the Keitt fruit leather was slightly liked, where as the flavor of the Tommy Atkins fruit
leather was moderately liked by the respondents. Therefore, based on the panelists preference, the
flavor of the Tommy Atkins fruit leather was relatively of good quality. Heat not only vaporizes
water during drying but also causes loss of volatile components from the food and as a result most
dried foods have fewer flavors than the original material. The extent of volatile loss depends on the
temperature and moisture content of the food and on the vapor pressure of the volatiles and their
solubility in water vapor. Volatiles which have a high relative volatility and diffusivity are lost at
an early stage in drying. Foods that have a high economic value due to their characteristic flavors
(for example herbs and spices) are dried at low temperatures (Mazza and LeMaguer, 1980).
4.9.1.5 Toughness
The toughness of the fruit leather is another quality parameter for grading fruit leathers. From the
sensory analysis result, it was observed that the toughness was significantly affected (P < 0.05) by
all the drying temperature, puree load and the fruit variety. The drying temperature has a significant
effect on the toughness of the fruit leather especially at 600C and 800C and at 700C and 800C and
their interactions. Table 4.14 indicates that the toughness of the fruit leather dried at 600C and 700C
was slightly liked, whereas the toughness of the fruit leather dried at 800C was neither liked nor
disliked by the respondents. A decrease in toughness was also observed with an increase of the
64

drying temperature. The toughness of the fruit leather dried with 0.4g/cm2 puree load as in Table
4.15 was neither liked nor disliked where as the one with 0.6 g/cm2 was slightly liked by the
respondents. An increase in toughness with an increase of the puree load was also observed.
As indicated in Table 4.16 on average, the Keitt mango leather was neither liked nor disliked
whereas that of the Tommy Atkins mango leather was slightly liked by the respondents.
Additionally, 25% of the panelists suggested that the toughness of the fruit leather shall be softer.
Since fruit leathers are not common in Ethiopia, the respondents find the product harder than an
ideal or their expectations. Therefore, according to the panelists result, the Tommy Atkins mango
leather dried at 60 0C or 700C with 0.6 g/cm2 is considered to be of better quality mango fruit
leather in terms of toughness.
4.9.1.6 Overall acceptability
The overall acceptability of mango fruit leather was significantly affected (P < 0.05) by all the
drying temperature, puree load and fruit variety. There is a significant effect on overall
acceptability due to drying temperature at 600C and 800C. The overall acceptability of the mango
leathers decreases with an increase of temperature Table 4.14. It has been also observed that the
overall acceptability of the fruit leather dried at 600C and 700C was moderately liked by the
respondents, whereas the fruit leather dried at 800C was only slightly liked.
The overall acceptability of the fruit leather with 0.4 g/cm2 and 0.6 g/cm2 puree load was
moderately liked by the respondents. Table 4.16 indicates that the overall acceptability of the Keitt
mango leather was liked slightly, whereas the Tommy Atkins mango leather was moderately liked.
Therefore, according to the panelists preference, the Tommy Atkins mango fruit leather dried at
600C and 700C with both puree loads 0.4 g/cm2 and 0.6 g/cm2 is considered to be the best leather in
terms of overall acceptability.

65

CHAPTER 5
SUGGESTED TYECHNOLOGY FOR MANGO LEATHER PROCESSING
5.1 Process description
Mango fruits can be processed and preserved in leather form. This can easily be achieved at a small
scale industry through the following process procedures. First, the preliminary operations
including: washing, sorting, peeling, cutting (slicing) and blanching will be done and then mixing,
concentrating and drying will follow and finally, packaging, labeling and storage will make the
process complete. Processing the mango fruits is intended to preserve them by slowing down the
natural processes of decay caused by microorganisms, enzymes in the food, or other factors such as
heat, moisture and sunlight and to change them into mango leather form, which are attractive and in
demand by consumers. The processors should use their skills to develop attractive recipes and
make mango leathers that consumers want to eat. By doing this successfully, they can increase
sales and earn an income.
Preliminary plant design is suggested for small scale local mango fruit (Keitt Variety) leather
processing industry in Ethiopia. This was done with the consideration of the availability of the
mango fruit with in the country throughout the year as well as the low price of the incoming raw
mango fruits to the processing industry. However, from the sensory analysis, it was observed that
most of the consumers preference is towards Tommy Atkins fruit leather having a higher sensory
quality. As a beginning, the researcher believes that a small scale plant design shall be suggested
for processing locally grown mango fruits as the product could be relatively expensive. Moreover,
with the same plant design if implemented with a lower capital investment, it would also be
suitable for processing of the export standard Tommy Atkins mango fruits whenever there is plenty
supply because the fruit is very seasonal. The following flow sheet (Figure. 4.3) clearly describes
the process for making mango leathers for small scale processing industry.

66

Mangos
ReceptionandStorage
PreliminaryWashing
SortingandGrading
SecondaryWashing
Weighing
HotCaustic
Soda

Peeling
Destoning(Pulping)

Peel+Caustic
SodaSlurry
Stones

Pulp
Blanching
Honey+
Citric Acid

Mixing
Puree
Concentrating
Drying
Packaging
MangoLeather

Fig. 4.3 Qualitative flow diagram for Mango leather processing

67

Spices

Material and Energy Balances


Material balance as per laboratory work:
The material balance will enable us to quantify the substances going in and out of the entire process
and piece of equipments. Emissions can be calculated as the difference between the input and
output of each listed substances. If there is any accumulation within the equipment during process,
it should be accounted for the calculation.
Literature Data:
A survey of the literature will often reveal general information and specific data pertinent to the
development of a design project. The literature data for physicochemical and composition of
mango and mango puree is summarized in Table 4.17.
Table 4.17 Proximate and physicochemical composition of Mango and puree
Value Ranges (% Fresh Weight)
Composition

Mango Fruit

Mango Puree

Moisture

72.185.5

6786

Crude protein

0.305.42

0.31.5

Crude fiber

0.302.38

0.050.6

Ash

0.291.13

0.130.61

Total soluble solids

1724

0.280.64

Total sugars

10.518.5

12.2817.54

pH

4.05.6

11.2715.38

TA (as citric)

0.327

0.11.1

Source: Hulme (1971), Caygill et al. (1976), Wu et al. (1993), and Narain et al. (1998).

68

The literature data for average compositions of Honey is summarized below:


Moisture Content (M.C) = 17.1%
Carbohydrate = 82.4%
Proteins, amino acids, vitamins, and minerals = 0.5%
Source: National Honey Board (2009).
The literature data for average composition of lemon juice is:
Moisture content = 92%
Solids = 8%
For Lemon Fruit, the average values per 100g of edible portion are:
Water = 89g
Protein = 1.10g
Fat = 0.30g
Carbohydrate = 9.32g
Fiber = 2.8g
Source: USDA National Nutrient Database (2004), within the Handbook of Fruits and Fruit
Processing, Hui (2006).
Laboratory test result:
The laboratory data for the process of mango leather production provides information useful for a
rapid material balance. This is summarized as follows:
Mass of Mango fruits = 3,195g = 3.2 kg
Mass of peel = 629.6g = 0.631kg
Mass of Stones = 824.8g = 0.82kg
Mass of Pulp = 1652.6g = 1.65kg
69

Mass of mango leather = 520g = 0.52kg


Mass of Honey = 65g
Mass of Spices = 5g
Mass of Lemon juice = 30g
Drying time = 8 h
Water temperature of the water bath = 800C
Temperature of the puree = 750C
Assumptions:
Plant production capacity of 200 days per year
Production of 200kg of mango leather per day
Therefore, based on all the above information, including the laboratory records and the literature
reviews, the following detailed material balance is for the laboratory scale is done.
Overall mass balance on the peeling unit:

Mangos
M 1= 3.2kg

Peeled Mangos
Peeling

M2 = 2.56kg

Peel M3 = 0.64kg
M1 = M 2 + M3
3.2 = 2.56 + 0.64

70

Overall mass balance on the pulping unit:

Pulping

Peeled Mango

Pulp

M2 = 2.56kg

M4 = 1.65kg

Stones, M5 = 0.91kg
M2 = M 4 + M 5
2.56 = 1.65 + 0.91
Overall mass balance on mixing unit
Honey, M6 = 0.065kg, y6 = 0.17

Pulp

(Blending)

M4 = 1.65kg
Y4= 0.79

Puree Mix,

Mixing

M9 = 1.75kg

y9 = ?
Lemon Juice,

Spices, M8 = 0.005kg, y8 = 0.11

M7 = 0.03kg, y7 = 0.92
Mass balance:
M4 + M6 + M7 + M8 = M9
1.65 + 0.065 + 0.03 + 0.005 = 1.75kg
Water balance:
W4 + W6 + W7 + W8 = W9
=> M4y4 + M6y6 + M7y7 + M8y8 = M9y9
71

=> (1.65*0.79) + (0.065*0.17) + (0.03*0.92) + (0.005*0.11) = 1.75y9


=> 1.75y9 = 1.3427
=> y9 = 0.767
Therefore, the moisture content of the mixed puree is 76.7%.
Mass balance on the heating unit:

Ti= 200C

T = 75oC

Heating

Mixed Puree

Hot Puree

M10 = 1.75kg

M11 = 1.75kg

y10 = 0.767

y11 = 0.767

Here, the input is equal to the output.


M10 = M11, and y10 = y11
Energy balance on the heating unit
The heating rate (heat load) of a heat exchanger (q, kW), required to heat the product (m, kg/s) by a
temperature difference (T, K or C), is estimated from the equation:
q = mCpT
Where, Cp (kJ/kg K) is the specific heat of the product.
To determine the specific heat capacity of the mixed puree, we use:
Cp, mixed puree = (CpiMi)
Where, Cpi = The specific heat capacities of each component in the mixture, and
Mi = Mass fractions of each components
The specific heat of the mixed puree is estimated using the Choi and Okos model for determination
of specific heat of foods based on Table 4.18 summarized below.
72

Table 4.18 Specific heat relationships for food product components

Therefore, based on the calculations within Table 4.18 the specific heat of the mixed puree was
found.
Cp of Mixed Puree at 75oC= 3.45KJ/kg.K . Thus, the heat can be determined by,
Q = MCpT
= 1.75kg*3.45KJ/kg.K* (75-20)
= 332.06KJ
Therefore, the heat required to heat the puree is 332.06KJ.

73

Overall mass balance on the drying unit


Moist Air out, M12

Tray Drying

Hot Puree

Mango Leather
M11 = 1.75kg

Dry Air in

M13 = 0.52kg
Y11 = 0.767

y13 =?

Mass balance:
M11 = M12 + M13
1.75kg = M12 + 0.52kg
M12 = 1.23kg of water
Therefore, 1.23kg of water is removed from the drying process.
Water balance:
M11y11 = M12 + M13y13
1.75*0.767 = 1.23 + 0.52y13
Y13 = 0.2156
Therefore, the final moisture content of the product was 21.6%.
Energy balance for drying:
The parameters that are critical for the production of mango fruit leather were selected and fixed to
calculate the air speed of the dryer.

74

T4 = 41oC
Moist Air out, G2, y2

Drying

Mango Puree

Mango Leather

M1 = 1.75kg

M2 = 0.52kg

x1 = 0.767

Dry Air in,

x2 =0.2156

T1 =40oC

G1, y1

T2 =

T3 = 50oC
The fixed parameters were as follows:
The relative humidity of the inlet hot air is set at 10% and its temperature at 50c.
The following assumptions were made so that the air follows an adiabatic saturation curve
(constant wet bulb temperature line) on the Psychrometric chart.
(Cp)air, (Cp)water, (Hv)water are independent of temperature at the prevailing process conditions.
The enthalpy changes undergone by the un-evaporated liquid water and the solid components in
going from T1 to T2 are negligible as compared to the changes undergone by the entering hot air
and the evaporated water.
Even though, the air at the interface could have a relative humidity of closer to 100%, it is the
nature of engineers to assume the worst case and benefit. Hence the relative humidity of the out let
air is assumed to be 40%.
Then from material balance calculations, Perrys handbook of chemical engineers and
psychrometry the following data is generated:
y1 = 0.00794
y2 = 0.01185
= 2382kj/kg
75

Cp(product) = 3.452KJ/kgoK
Cp(air) = 1.007
Cp(water liquid) = 4.179
Cp(water vapor) = 1.88
Where, is the latent heat of vaporization of water at 50c. Therefore, the total mass balance and
component balance on the drier solves for the two unknowns, G1 the inlet amount of the hot air and
G2 the outlet amount of wet air in kilo grams.
Total mass balance:
M1 + G1 = M2 + G2
=> M1 - M2 = G2 - G1
=>G2 - G1 = 1.23 . (eq.1)
Component balance on water:
M1x1+G1y1 = M2x2 + G2y2
=> G2y2 - G1y1 = M1x1 - M2x2
=> G2y2 - G1y1 = 1.23 ... (eq. 2)
We got two equations and two unknowns, G1 and G2.
Then, solve for G1 by first multiplying (eq. 1) by y2:
(G2 - G1 = 1.23) y2; where, y2 = 0.01185
=> G2y2 - G1y2 = 0.0146 (eq. 3)
Then solve (eq. 2) and (eq. 3) simultaneously;
G2y2 - G1y1 = 1.23
(G2y2 - G1y2 = 0.0146) multiply with (-1)
G1 (y2 - y1) = 1.2154
76

G1 = 1.2154
(y2 y1)
Where y2 = 0.01185
y1 = 0.00794
=> G1 = 310.85kg air and from (eq.1),
G2 - G1 = 1.23
=> G2 = 312.08kg air
Energy balance on the convective drier
Based on the above assumptions the simplified energy balance equation becomes;
Q = (m, evaporated water) () + (m, evaporated water) (cp, water liquid) (T3 T1)
Q = (1.23*2382.7) + (1.23*4.179)* (10)
=> Q = 2982.13 kj
Therefore, the amount of energy required by the dryer is 2982.13kj.

77

(eq.4)

The quantitative flow diagram for the whole process as per laboratory scale for material balance is
presented below in Figure 4.4.
Mangos, 3.2kg

Peels, 0.64kg
Peeling

Peeled Mangos, 2.56kg


Pulping

Stones, 0.91kg

Pulp, 1.65kg
Honey, 0.065kg

Mixing
(Blending)

Lemon Juice, 0.03kg, M.C=92%

M.C=17%

Spice, 0.005kg, M.C=11%

Mixed Puree, 1.75kg, M.C=76.7%


Heating
(800C)

Hot Puree, 1.75kg, M.C=76.7%


Convective
Drying

Water, 1.23kg

Mango
Leather

0.52kg, M.C= 21.6%

Figure 4.4 Quantitative flow diagram for mango leather process

78

Product commercialization (Scale up the laboratory scale mango leather process in to


small scale processing industry)
The process selected for the manufacture of the mango leathers is semi-continuous even though the
drying process is batch. Provisions for possible shutdowns for repairs and maintenance, and
seasonal production of mango fruits are incorporated into the design of the process by specifying
plant operation for 200 calendar days per year.
The laboratory result shows that one can get an output of 0.52kg of mango fruit leather from 3.2kg
of mango fruits. To scale up this in to a small scale processing plant with the overall assumptions
of 200kg production of mango leather per day, we need 1230.8kg mango fruits supply per day:
i.e. 3.2kg mango fruit => 0.52kg mango leather
X => 200kg mango leather
Therefore, X = 1230.8kg of mango fruit
The plant operates 200 days per year, so the annual production of mango leather will be: 200kg X
200 = 40,000kg.
Since the intended daily production of mango leather is 200kg, we can take out a scale up factor for
multiplying all the laboratory scale results to obtain the daily needs of all the inputs and outputs of
the whole process. This can be done by dividing 200kg of leather/day with 0.52kg of leather.
Therefore, we get a scale up factor of 384.62 for daily production. Therefore, based on the factor,
the quantitative flow diagram for the whole process as per daily basis for material balance is
presented below.
On average 5 liters of water is required to wash 1 kilogram of mango fruit. So to wash 1230.8 kg of
mangos we need 6154L of water.

79

Ripened Mangoes
1230.8kg/day
Water in, 6154L/day

Waste water out, 6154L/day

Washing

Mangos, 1230.8kg/day

Peels, 246.2kg/day

Peeling

Peeled Mangos, 984.6kg/day


Pulping

Stones, 350kg/day

Pulp, 634.6kg/day, M.C=79%


Honey, 25kg/day
M.C=17%

Lemon Juice, 11.5kg/day, M.C=92%


Mixing
(Blending)

Spice, 1.9kg/day, M.C=11%

Mixed Puree, 673.1kg/day, M.C=76.7%


Heating

Hot Puree, 673.1kg/day, M.C=76.7%


Drying

Water, 473.62kg/day

Mango Leather

200kg/day, M.C= 21.6%

Figure 4.5 Quantitative flow diagram for daily production of mango leather

80

Fig. 4.6 Equipment layout for mango leather processing

81

Equipment sizing and selection


The equipment design for this preliminary process evaluation involves determining the size of the
equipments in terms of the volume, flow per unit time, or surface area. The determination of the
equipments capacity per hour basis is done with the general assumption of 200 working days per
year and 8 working hours per day with two shifts. Therefore, the production can be further reduced
in to an hour basis by taking 200kg of mango leather production per day and assuming the plant
operates 8hrs; we can get 25kg/h production of mango leather.
Washing unit equipment selection:
Fruit Washing Machine: Washing is intended to be completed maximum within two hours, so;
Washer Capacity:
1230.8kg = 615 kg/h
2h
Therefore, select a fruit washer with a capacity of washing 700kg of mango fruits per hour.
Peeling unit equipment selection:
Peeler Capacity: peeling is also intended to be completed within maximum of two hours; so,
1230.8kg = 615 kg/h
2h
Therefore, select a fruit peeling machine with a capacity of peeling 700kg of mangos per hour.
Pulping unit equipment selection:
Pulper Capacity: the pulping process is intended to be completed with a maximum of one hour; so,
984.6kg = 984.6kg/h
1h
Therefore, select a pulper with a capacity of pulping 1,000kg of mangos per hour
82

Mixing unit equipment selection:


Mixer (Blending Machine) Capacity: the blending and mixing process together should also be
completed within an hour; so,
673kg = 673kg/h
1h
Therefore, select a mixer with a capacity of mixing 700kg of ingredients per hour.
Heating unit equipment selection:
Heat Exchanger Capacity: the heating process can also be finished within an hour, so;
673kg = 673kg/h
1h
Therefore, select a heat exchanger with a capacity of heating 700kg of mixed mango puree per
hour.
Drying equipment selection:
Convective hot air dryer:
Drying of mango leather was experimented in laboratory and it was found that, to dry 1.75kg of
mango puree with a M.C of 76.7%, 3 trays were used each trays were 25 X 35Cm. Therefore, the
total area for drying was 0.265m2. From this result, it can be scaled up to dry 673kg of mango
puree with total area of 101.165m2. Therefore, select a convective hot air dryer with a surface area
of 105m2 to dry the hot mango puree.
Economical analysis
Production planning:
Production can be planned using the calculations below and answering the following questions.
The information required to do this includes:
200days per year/12months per year = 16 working days on average per month.
83

How much product (kg) is sold each month?


200kg leather per day * 16days = 3,200kg/month
Answer: 3,200kg/month assuming all the produced mango leathers are sold
How many hours are worked per day?
Answer: 8 hours
How many days are worked per month?
Answer: 16 days on average per month
This information is first used to calculate the daily production rate, so that ingredients and
packaging can be ordered. Then the average amount of production per hour (termed the product
throughput) can be calculated to find the size of equipment and numbers of workers required.
Daily production rate:
The daily production rate is calculated as follows:
Production rate (kg/day) = Amount of product sold/month (kg)
Number of days production/month
= 3200kg = 200kg/day
16days
This figure is used to decide how much raw material, ingredients and packaging to buy.
Raw materials and ingredients
Having decided how much product to make, a processor needs to calculate how much fruit to buy.
This is based first on the recipe for the product and secondly on the likely levels of wastage and
losses during the process.

84

Calculating weights of ingredients from a recipe for making mango leather


A recipe for mango leather (Table 4.19) shows in the left column of the table below and the
amounts of ingredients needed to make 0.52 kg are shown in the right column, with the calculation
in the centre.
Table 4.19 Recipes, calculations and amount of ingredients for making Mango leather

Recipe

Calculations

Amount needed to make 0.52kg of


Mango Leather

Mango = 3200g

(3200/3300) X 0.52

0.504kg

Honey = 65g

(65/3300) X 0.52

0.0102kg

Lemon Juice = 30g

(30/3300) X 0.52

0.0047kg

Ground Ginger = 5g

(5/3300) X 0.52

0.00079kg

Total = 3,300g

0.52kg

However, the amounts of raw material and ingredients that are calculated from the recipe are not
the amounts that are used. Losses arise from peeling, from spoiled raw materials that are thrown
away during sorting, from spillage during filling into packs or from food that sticks to equipment
and then lost during cleaning. The average typical losses during processing have to be considered
during the production of mango leather. These losses can be referred to in Table 4.20 presented
below.

85

Table 4.20 Typical losses during processing of fruits and vegetables


Stages in a Process

Typical Losses (%)

Washing Fruits and Vegetables

0-10

Sorting

5-50

Peeling

5-60

Slicing/dicing

5-10

Batch preparation/weighing

2-5

Boiling

5-10

Drying

10-20

Packaging

5-10

Machine Washing

5-20

Accidental Spillage

5-10

Rejected packs

2-5

(does not include evaporation losses)


(Source: Fellows, P., Midway Technology Ltd, Bonsall, UK)
The amount of usable food after raw materials are prepared for processing is known as the yield
and is calculated as follows:
Yield (%) = Weight of raw material actually used in the process X 100
Weight of raw material that is bought
On average mangos cost 0.6 Birr each and a single fruit weighs 200g (i.e. 3Birr/kg). 3.2kg of
mango fruits are bought for 9.6 Birr and after peeling and coring there is 1.65kg available for
processing.
Therefore, Yield = 1.65 *100 = 51.56% (i.e. 48.44% is waste)
86

3.2
The UNIDO Technology Manual (2004) has indicated that the typical losses during the preparation
of mango fruits during peeling and de-stoning to be 45%. Therefore, the result of wastage in this
case 48.44% is relatively comparable.
The true cost of raw materials depends on the yield and can be calculated as below:
True raw material cost = Supplier cost * 100
% Yield
The true cost of the usable part of a single fruit = 0.6 * 100 = 1.16Birr
51.56
This result also shows the value addition of the product from the raw material after preliminary
process.
From the laboratory result theoretically, 0.52kg of mango leather is available for sale. Ignoring
other production costs (labor, depreciation etc.) the value of the product is therefore:
Cost of raw materials = Mango + Honey + Lemon juice + Ginger (or Spice)
= 9.6Birr + 2.275Birr + 0.3Birr + 0.5Birr
= 13.15Birr
Cost of raw materials = 13.15Birr = 25.28 Birr/Kg
Weight of product

0.52kg

Therefore, processing has increased the value of the raw materials 13.15/kg to Birr 25.28/kg.

87

Cost benefit analysis of Mango leather production


Plant parameters for Mango leather production
Capacity

40,000kg per year

Number of shifts /day

Working days/year

200

The amount of total product per year = 40,000kg/yr


The amount of raw materials needed:
- Mango fruits = 246,160kg/yr

- Lemon Juice = 2,300kg/yr

- Honey = 5,000kg/yr

- Spice (Ground Ginger) = 380kg/yr

Machinery and equipment

Table 4.21 Machinery and equipment requirements for mango leather production
Item

Quantity

Capacity

Unit price, Birr

Total Price, Birr

(kg/h)
Fruit washer

700

23,520.00

23,520.00

Fruit Peeler

700

25,000.00

25,000.00

Fruit pulper

500

23,640.00

23,640.00

Mixer (Blender)

700

100,000.00

100,000.00

Heat Exchanger

700

100,000.00

100,000.00

Tray Dryer

200kg/24h

60,000.00

60,000.00

Total

88

332,160.00

Manpower requirement

Table 4.22 Manpower required for Mango leather production


Human Resource

Number

Monthly average Total


salary, birr

monthly Total yearly

salary, birr

salary, Birr

Manager

3,000.00

3,000.00

36,000

Skilled

1,800.00

7,200.00

86,400

Unskilled

10

800.00

8,000.00

96,000

Total

15

5,600.00

18,200.00

218,400.00

Cost of raw materials

Table 4.23 Raw material costs for Mango leather production


Unit price,
Item

Quantity (kg/yr)
Birr/kg

Total price,
birr

Mango fruits

246,160

3.00

738,480.00

Honey

5000

35.00

175,000.00

Lemon Juice

2300

10.00

23,000.00

Spices (Ground Ginger)

380

30.00

11,400.00

Total

947,880.00

89

Cost of utilities

Table 4.24 Cost of utilities for Mango leather production


Unit price, Total price, Birr

Item

Quantity

Electric Power

70,000kwh

0.90

65,000.00

Water

1,230,800 l

0.005

6,154.00

birr

Total

71,154.00

90

Fixed capital cost estimation

Table 4.25 Fixed capital cost estimation for Mango leather production
Item

Description / factor
A. a. Equipment

Total Cost, Birr


332,160.00

b. Installation

0.47* 332,160.00

156,115.20

c. Instrumentation

0.18* 332,160.00

59,788.80

d. Piping

0.66* 332,160.00

219,225.60

e. Electrical

0.11* 332,160.00

36,537.60

B. Building +auxiliary

0.70* 332,160.00

232,512.00

C. Service facilities

0.70* 332,160.00

232,512.00

D. Land

0.06* 332,160.00

19,929.60

Total direct cost

A+B+C+D

1,288,780.80

A. Engineering & supervision

0.1*1,288,780.80

128,878.08

B. Construction + contractor fee

0.1*1,288,780.80

128,878.08

C. Contingency

0.06*1,288,780.80

77326.848

Total indirect cost

A+B+C

335,083.008

III. Fixed capital investment

Direct +Indirect cost

1,623,863.808

IV. Working capital

0.15*1,623,863.808

243579.57

V. Total capital investment

III +IV

1,867,443.379

I Direct Costs

II.
Costs

Indirect

91

Estimation of Total Product Cost (TPC)

Table 4.26 Estimation of total product cost for Mango leather


Item

Description/factor

Total

cost,

Birr
A. a. Fixed Charges
a. Depreciation

0.1*mach. + 0. 02*building 37,866.24


cost

I.
cost

Manufacturing

b. Local taxes

0.02*FCI

c. Insurance

0.006* FCI

Total of A

32477.28
9743.18
80,086.70

B. Direct production cost


Total product cost (tpc)

Total fixed charge/0. 15

533911.33

a. Raw material

Calculated

947,880.00

b. Utilities

Calculated

71,154.00

c. Operating labor (ol)

0.1*tpc

53,391.13

d. Supervisory

0.1*ol

5,339.11

e. Maintenance

0.05*FCI

f. Lab charges

0.12*ol

Total of B

II.

81,193.19
6406.94
1,699,275.69

C. Plant overheads

0.1*tpc

53,391.13

Total manufacturing cost

A+B+C

1,832,753.52

0.05*tpc

26,695.57

General a. Administrative cost


92

Expenses

b. Distribution
c.

Research

0.1*tpc

53,391.13

& 0.05*tpc

26,695.57

0.05*tpc

26,695.57

development
d. Interest
Total general expenses

85,477.83

III. Total Product Cost

I +II

1,918,231.35

1,918231.35
= 47.96 Birr/kg
40,000.00
Total product cost/kg of mango leather

Financial evaluation
Assume selling price of 1kg of mango fruit leather = 65.00 Birr
Expecting all produced product will be sold,
Total income = 65*40,000 = 2,600,000.00 Birr
Gross income = total income-total product cost
= 2,600,000.00 1,918,231.35 = 681,768.65 Birr
Let the tax rate be 35% (income tax of Ethiopia)
Taxes = 0.35*681,768.65 = 238,619.03 Birr
Net profit = gross income tax => 681,768.65- 238,619.03= 443,149.62 Birr
Return on investment =

Payback period =

netprofit
443,149.62
100 =
100 = 23.73%
totalcapitalinvestment
1,867,443.379

FCI
1,623,863. 808
=
= 3.4 years
NP Depre
443,149.62 37,866.24

93

CHAPTER 6
CONCLUSIONS AND RECOMMENDATION
6.1 Conclusion
To be competitive and increase the market share and profits, the mango leather processing industry
must ensure that its products are of higher quality, less expensive, and more desirable than their
competitors, and also safe and nutritious. To meet these standards, the processing industry needs to
analyze the mango fruits, before, during and after the manufacturing process. The most important
concerns of the mango leather manufacturer is to produce a final product that consistently has the
same qualities with overall properties, such as appearance, texture, flavor and also shelf life.
However, the properties of the raw ingredients and the processing conditions vary from time to
time which causes the properties of the final product to vary, often in an unpredictable way.
Therefore, the processing industry has to wisely control the manufacturing process to produce a
final product with consistent properties. This can be achieved by characterizing and measuring the
properties of the incoming raw materials, to ensure that they meet certain minimum standards of
quality that have previously been defined by the manufacturer. If the industries standards are not
met it has to reject the material since variations in its properties might lead to changes in the
properties of the final product.
The nutritional content of the puree, after being mixed with honey and lemon juice, increased for
both varieties while the moisture content decreased. The viscosity of Tommy Atkins variety mango
puree mix was found to be lower than that of Keitt variety. The drying time was generally shorter
at higher temperatures due to quick removal of moisture. The puree load also affected the drying
time at all drying temperatures. Generally, the drying time was considerably prolonged for more
than 7 h for 0.6g/cm2 puree load at all drying air temperatures (600C, 700C and 800C). The result of
the analysis showed 70.3% moisture loss after drying process is completed. The texture analysis
result of the final mango leather showed that 4 sheets and 5 sheets of leather with 5mm and 6mm
thickness respectively were found to be suitable for a single bite. The results of the proximate
analysis for both varieties of mango fruit leather indicated that the processing affected the
nutritional composition of the fruit leather. The vitamin C content was also found to be dependent
on all drying temperature, puree load and fruit variety. The vitamin C content of the Keitt mango
94

leather (26.93%) is greater than that of Tommy Atkins mango leather (22.71%). When compared to
the fresh puree mix, the Keitt mango leather is decreased by 39.66% and that of the Tommy Atkins
mango leather is decreased by 57.82%. The microbiological analysis of the mango fruit leather has
also indicated that the products are generally safe for consumption. The sensory analysis result also
indicated that the Tommy Atkins mango leathers dried at 60 and 700C with 0.4 and 0.6 g/cm2 puree
loads were well accepted by the panelists especially in terms of color, taste, flavor and overall
acceptability.
Currently in Ethiopia there is no fruit processing industry involved in fruit leather development
process. Most fruit processing industries are interested in production of juices and similar types of
products. However, fruit leather production can easily be implemented taking advantage of
utilizing fruit normally not selected for juice making, freezing or other processes. Mangos can be
preserved by production of mango fruit leathers without any addition of preservatives. This meets
consumer demand for health food products, in a form of fruit leathers. In the country, the
production of fruit leathers is almost unknown and if this project were to be implemented, it will be
very attractive and profitable. Besides, the application of this kind of simple processing technology
based on drying principles is one way of minimizing postharvest loss.
The project has designed and developed the process for the production of mango fruit leather at a
small scale level within Ethiopia. It has been demonstrated that the mango fruit leather production
would result in a feasible and strong return on investment. Therefore, investors who are interested
in food processing can take advantage of seasoning and diversifying mango products within the
country and develop a value added product with strong profit.

95

6.2 Recommendation
The mango properties during processing have to be monitored and measured. So that if there is any
problem developed it can be quickly detected, and the process will be adjusted to compensate for it.
This helps to improve the overall quality of the mango leather and reduce the amount of material
and time wasted. The final mango fruit leather product has to be also analyzed and characterized to
ensure that it retains its desirable properties up to the time when it is consumed, meets the
appropriate high quality requirements, and that it is safe for consumption.
By mixing different proportions of various types of fruits, fruit leathers can be developed to
nutritionally enrich the final product and attract customers attention. The fruit leather products are
being targeted at consumers requirement with increased awareness towards health food choices.
Consequently fruit leathers are being marketed as luxury health food products in Europe.
Therefore, if mango fruit leathers can be produced and exported, it will contribute to the countrys
foreign exchange. Once the technology is applied and implemented for mango fruit processing
industry, it can also be used for processing of different types of fruit leathers at different seasons of
the year whenever the fruit types are abundant. So it is recommended that further researches need
to be conducted on the processing of fruit leathers, fruit bars and dried fruits and the process needs
to be designed and implemented within the country in the near future.

96

REFERENCES
Aithal, A. and Wangila, J. (2006). Rapid assessment of fruit markets in Ethiopia, based on research
in Addis Ababa, Mekele and Awassa: Final report for Improving Productivity and Market
Successes (IPMS) for Ethiopian Farmers Project, Addis Ababa.
Andress, E.L. (2004). Mango Leather. Developed at The University of Georgia, Athens, for the
National Center for Home Food Preservation. Released by, Department of Foods and
Nutrition, College of Family and Consumer Sciences.
Andress, E.L., & Harrison, J. A. (1999). So easy to preserve (4th ed.). Athens: University of
Georgia Cooperative Extension Service.
Anon, T. (1981). Special feature: evaporation plant. Food Process Ind., February 2529
Azeredo, M.C., Brito S., Moreira E.G, Farias L & Bruno M. (2006). Effect of drying and storage
time on the physico-chemical properties of mango leathers. International Journal of Food
Science and Technology, 41, 635638
Bally, S.E. (2006). Mangifera indica (mango): Species Profiles for Pacific Island Agro forestry.
Ver.3.1. www.traditionaltree.org. Retrieved on December, 2009
Brown, S. (2009). Fruit Leathers. Food Safety and Nutrition Faculty. Washington State
University Extension programs.
Caygill, JC., Cooke, R.D., Moore, D.J., Read, S.J., Passam, H.C. (1976). The mango (Mangifera
indica L.) harvesting and subsequent handling and processing: an annotated bibliography.
London: Tropical Products Institute.
Chan, HT. and Cavaletto, CG. (1978). Dehydration and storage stability of papaya leather. Journal
of Food Science, 43, 17231724.
Che Man, Y. B., and Taufik, Y. C. M. (1995). Development and stability of jack fruit leather.
Tropical Science, 35, 245250.
Che Man, Y. B., Taufik, Y. C. M., & Karim, M. N. A. (1992). Storage stability of ciku leather.
ASEAN Food Journal, 7, 5355.
97

CIA. (2008), Central Intelligence Agency. World Factbook, accessed on July 2008.
Clark, R.C. (1990). Flavor and texture factors in model gel systems. In: A. Turner (ed.) Food
Technology International Europe. Sterling Publications International, London, pp. 271277.
CSA. (2009). Central Statistical Authority. Addis Ababa , Ethiopia.
Da Silva, J. J. R. F., and Williams R. J. P. (1991). The Biological Chemistry of the Elements: The
Inorganic Chemistry of Life. Oxford: Clarendon Press.
Dauthy, M. (1995). Fruit and Vegetable Processing. FAO Corporate Document Repository.
Originated by Agriculture and Consumer Protection. Series title: FAO Agricultural Services
Bulletin - 119. ISBN: 92-5-103657-8, V5030/E.
David, A. and Arnold, E. (1999). Benders Dictionary of Nutrition and Food Technology (7th Ed.).
Wood head Publishing Ltd and CRC Press LLC.
Doreyappa, ING., and Ramanjaneya KH. (1995). Evaluation of some mango varieties for their
suitability for canned mango juice. Journal of Food Science & Technology. 32: 323 325.
EAP. (2009). (Ethiopian Agriculture Portal).
http://www.eap.gov.et/AgriculturalCommodities/Crops/Fruits/Mango.asp. Retrieved on
November, 2009.
Exama, A. and Lacroix, C. (1989). Development of a high protein paste. I. Influence of processing
parameters. Science Des Aliments, 9, 285305.
FAO. (2009). Food and Agriculture Organization. STAT accessed July 2009.
FDA. (2006). Food and Drugs Administration. Bacteriological Analytical Manual (FDA/BAM).
USA.
Fellows, P. and Quaouich, A. (2004). UNIDO Technology Manual. Small-Scale Fruit and
Vegetable Processing and Products. Production Methods, Equipment and Quality Assurance
Practices. United Nations Industrial Development Organization, Vienna.
Fellows, P. (2000). Food Processing Technology: Principles and Practice. (2nd ed.). Wood head
Publishing Limited and CRC Press LLC.
98

Fennema, R. (1996). Food Chemistry (3rd Ed). New York: Marcel Dekker, Inc.
FPT (2009). Fruit processing toolkit. (FPT) http://www.fao.org/inpho/content/fpt/Fruit/index.htm
Retrieval date, Oct. 2009.
Germew, B. (2005). Food chemistry laboratory manual. Haramaya: Haramaya University. Food
Science and Postharvest Technology Department, Ethiopia.
GTZ (2009). Fruit and vegetable drying. Processing of tree crops products - a farmer trainer
training manual. ITFSP.
Heikal, H. A., El-Sanafiri, N. Y., & Shooman, M. A. (1972). Some factors affecting quality of
dried

mango

sheets.

Agricultural

Research

Review,

50,

185194.

http://trc.dost.gov.ph/trcfile/Technology%20Snapshots/inpho/Fruit/gnleath.htm.

Retrieved

on October 2009.
Hulme, AC. (1971). The mango. In: Hulme AC, editor. The biochemistry of fruits and their
products. New York, New York: Academic Press, Inc. Vol. 2. pp. 238252.
Hui, Y. Jozsef, B, Pilar, M Cano, Todd W., Gusek, Jiwan S. Sidhu, and Nirmal K. Sinha, N
(2006). Handbook of Fruits and Fruit Processing. Blackwell Publishing Professional 2121
State Avenue, Ames, Iowa 50014, USA.
Irwandi, J., Che Man, Y. B., Yusof, S., Jinap, S., & Sugisawa, H. (1998). Effects of type of
packaging materials on physicochemical, microbiological and sensory characteristics of
durian fruit leather during storage. Journal of the Science of Food and Agriculture, 76, 427
434.
James S., Rowlands C., Kamara J.K. and Haddis B. (2009). Analysis and Baseline Status of the
Mango Value Chain from Homosha and Assosa to Addis Ababa, September 2008. The
Ssemwanga Centre for Agriculture and Food, World Vision Australia, World Vision
Ethiopia.
John, M.D. (1980). Principles of Food Chemistry. Connecticut: AVI Publishing. USA.
Kader, A. (2009). Postharvest Losses of Fruits and Vegetables in Developing Countries: A Review
of the Literature. UCDavis. aakader@ucdavis.edu. Retrieved on November, 2009.
99

Kalra, S.K., Tandon, D.K., & Singh, B.P. (1995). Mango ripening guide. Journal of Queensland
Agric. 114, 369-371.In D.K.Salunthe & S.S Kandam (eds).
LFI (2003). London Fruit, Inc. Mango Varieties. Produced by: Stewart & Strauss Advertising, Inc.
http://www.freshmangoes.com/varieties.html. Retreived on January, 2010.
Maruniak, J. A. and Mackay-Sim, A. (1984). In: J. R. Piggott (ed.) Sensory Analysis of Foods.
Elsevier Applied Science, London, pp. 2358.
Mazza, G. and Lemaguer, M. (1980). Flavor retention during dehydration of onion. In: P. Linko, Y.
Malkki, J.Olkku and J. Larinkari (eds.) Food Process Engineering, Vol. 1, Food processing
systems. Applied Science, London, pp. 399406.
McClements, D.J. (1999). Analysis of Food Products. Retrived on janurary 2009.
Nancy, B. (2009). "Fruit Leather." How Products Are Made. Gale Research Inc. 2000.
Encyclopedia.com.

30

Oct.<http://www.encyclopedia.com>.

Online

dictionary

and

encyclopedia of facts, information, and biographies.


NHB (2009). National Honey Board. (NHB) www.honey.com/. Retrieved on Dec. 24, 2009.
http://www.foodreference.com/html/art-honey-health.html.
Pramanik, W. K., and Sengupta, J. P. (1978). A preliminary study of composition of mango sheets.
Institutes of Chemists, 50, 2526.
Pushpa, Rajkumar, PG., Gariepy, Y., and Raghavan, G.S.V. (2006). Microwave Drying of
Enriched Mango Fruit Leather. The Canadian Society for Bioengineering. Edmonton
Alberta.
Rameshwar, A. (1993). In: Advances in horticulture fruit crops, K.L.Chadha & O.Pareek (eds.),
Malhotra Publishing House New Delhi ,Vol 4, pp.1661-1702.
Rolls, B. A. and Porter, J. W. G. (1973). Some effects of processing and storage on the nutritive
value of milk and milk products. Proc. Nutr. Soc. 32, 915.
Saxena, R., and Arora. V. P. (1997). Consumers of processed fruit and vegetables products. An
analysis of buying behaviour. Indian Food Industry, 16(3), 25-36.

100

Singh, G., A. Biswas, and K. Muthukumarappan. (2005). Development of technology for the
calcium fortified mango yoghurt. ASAE Paper No. 056163. St. Joseph, Mich.: ASAE.
Singh, R.P., and Held man. R.D. (2001). Introduction to food engineering (3rd ed). Great Britain,
Glasgow: bath Colourbooks Print.
Srinivasan, N., S. Elangovan, and P. Chinnaiyan. (2000). Consumer perception towards processed
fruit and vegetables products. Indian Economic Panorama. 10(3), 11-12.
Truneh, D. (2009). Value chain development of mango and highland fruits froduction. SNV
Ethiopia.
Vijayanand, P., Yadav, A. R., Balasubramanyam, N., & Narasimham, P. (2000). Storage stability
of guava fruit bar prepared using a new process. Lebensmittel-Wissenschaft &Technologie,
33, 132137.
WAFC. (2006). World Agroforestry Centre. Final report for Improving Productivity and Market
Successes (IPMS) for Ethiopian Farmers Project, Addis Ababa. (ICRAF), Nairobi, July
11.Addis Ababa, Mekele and Awassa.
Wu JS., Chen H., and Fang T. (1993). Mango juice. In: Nagy S, Chen CS, Shaw PE, editors. Fruit
juice processing technology. Auburndale, FL: Agscience, Inc., pp. 621648.

101

ANNEXES
Annex 1 Determination of aerobic colony count for mould and yeast in food (NMKL, No. 98,
1997)
Method principle
The aerobic colony count estimates the number of viable aerobic mould and yeast per g or ml of
product. A portion of the food homogenate is mixed with a specified agar medium and incubated
under specific conditions of time and temperature. It is assumed that each viable aerobic mould/
yeast will multiply under these conditions and give rise to a colony.
Responsibilities: All trained stuff with adequate experience
Positive Control: Any spp. of mold and yeast.
Negative Control: check sterility of PDA medium and diluents used by pouring control plates and
incubate at 37oc and 22 oC for 5-7 days.
Equipment to Calibrate: Incubator 37C and 22oC, Autoclave, and pH-meter.
Media: PDA
Diluents: peptone water
Procedure:
1. Preparation of food homogenate
Transfer 10ml of liquid sample to 90ml of diluents or 25g of sample to 225 ml of diluents in a flask
if shaker used or in sterile plastic bag if stomacher used to make 101 dilution (the first dilution)
2. Dilution
2.1 Mix homogenate by shaking and pipette 1ml into a tube (labled102 containing 9ml of normal
saline. Mix carefully by aspirating 10 times with a pipette
2.2 From the first dilution, transfer with the same pipette 1ml to 2nd dilution tube containing 9ml of
the Ns, Mix with a fresh pipette
2.3 Repeat using 3rd or more until the required numbers of dilutions is made
102

2.4 shake all dilution carefully.


3. Pour plating
3.1 Pipette 1ml of the food homogenate and of each dilution of the homogenate into each of the
appropriately marked duplicate dishes.
3.2 Pour into each petridish 15-20ml of the PDA.
3.3 Mix the sample dilution and agar medium thoroughly and uniformly, allow solidifying.
4. Incubation. Incubate the prepared dishes, inverted, at 370c and 220c for 5-7 days.
5. Counting the colonies. Following incubation, count all colonies on dishes containing 30300 colonies and recorded the results per dilution counted.
Verification: If there is growth on the negative control or if there is no growth on the positive
controls the test should be repeated.
Expressions of results: calculate the average count and multiply by the dilution. And express the
result in cfu per g ml (if a liquid sample)
-

the result at 370c reported as yeast and mold count at 370c

the result at 220c reported as yeast and mold count at 220c

Annex 2 Determination of Aerobic Plate Counts (APC) in food (NMKL, No. 86, 2006)
Method principle
The aerobic colony count estimates the number of viable aerobic bacteria per gm or ml of a
product. A portion of the diluted sample mixed with a specified agar medium and incubated under
specific temperature for 48 hr. It is assumed that each viable aerobic bacterium will multiply under
these conditions and give raise to colonies.
Terms:
Mesophillic bacteria: an organism whose optimum growth lies within a range generally accepted as
20-450C

103

Psychrophilic bacteria: an organism which grows optimally at or below 15 0C, which has an upper
limit for growth at 20 0C , and which has a lower limit of 0 0C or lower.
Termophilic bacterial: an organism whose optimum growth temperature is >45 0C
Responsibilities: All trained stuff with adequate experience.
Positive Control: Reference material with known aerobic plate count.
Negative Control: check sterility of PCA

medium

and

diluents

used

by

pouring

control plates.
0

Equipment to Calibrate: Incubator37 C for Mesophillic bacteria, Incubator 55 C for Termophilic


0

bacterial, Refrigerator 2-8 C for Psychrophilic bacteria, Autoclave, PH meter, Safety


cabinet, Pipettes controllers, Colony counting device, Centrifuge optional, only for some
food type, Stomacher optional, only for none liquid sample, and Digital balance.
Media: PCA
REAGENT: 2%sodium citrate (tempered to 45 0c) (for cheese sample only)
Diluents: peptone water diluents (SOP 4.2.15)
Procedure:
1. Sample preparation
Transfer 10ml of liquid sample to 90ml of diluents or 25g of sample to 225 ml of diluents in a flask
if shaker used or in sterile plastic bag if stomacher used to make 101 dilutions (the first dilution)
Mix well with shaker/stomacher
2. Dilutions
Mix the first dilution by shaking then pipette 1ml into a tube (labled102) containing 9 ml of normal
saline. Mix carefully by aspirating 10 times with a pipette.
From the 102 dilution, transfer with the same pipette 1ml to the tube (labled103) containing 9ml of
the diluent, Mix with a fresh pipette. Repeat until the required numbers of dilutions are made.
3. Pour plating
104

Pipette 1ml of each serial dilution into each of the appropriately marked duplicate dishes. Pour 1520ml of the molten PCA kept at 45 0C into each Petri dish. Mix it thoroughly and allow it to
solidify.
4. Incubation.
Incubate the dishes, inverted, at 35 0C or for dairy products at 320C for 48 hr.
N.B: Avoid excessive humidity in the incubator, to reduce the tendency for spreader formation, but
prevent excessive drying of the medium by controlling ventilation and air circulation. Agar in
plates should not lose weight by more than 15% during 48 hours of incubation.
5. Counting the colonies.
Following incubation, count all colonies within the range of 30-300 colonies and record the results
per dilution counted.
Sample preparation: weigh 10g of the sample in to a sterile 250ml Erlenmeyer flask; marked to
indicate 100ml volume. Add sterile saline peptone to 100ml mark. Dissolve and shake thoroughly.
Dilution:

1:10, 1:100, 1:1000, etc

Dilution factor:

1 x 101, 1102, 1 103 etc

Inoculation: Pipette 1ml of the food homogenate and of each dilution of the homogenate into each
of the appropriately marked duplicate dishes followed by pour plating of PCA.
Incubation: Incubate the prepared dishes, inverted, at 350C for 48 hours, and for dairy products at
320 C for 483 hrs.
Counting colonies: Following incubation, count all colonies on dishes containing 30-300
Colonies, including those of pinpoint size and recorded the results per dilution counted.
Verification: If there is growth on the negative control and /or no growth on the
positive control the test should be repeated with the corrected media
Expression of results: express the result in cfu per g /ml (if a liquid sample)
Calculation formula: Use the best two consecutive dilutions, as n1 and n2 to calculate the results.
105

N =C/V (n1 + 0.1n2) d


Where,

C = is the sum of colonies on all plates counted


V = is the volume applied to each plate
n1= is the number of plates counted at first dilution.
n2= is the number of plates counted at second dilution,
d = is the dilution from which first count was obtained.
N= is the average plate count.

Round the result to two significant figures and express it as a number between 1.0and 9.9
multiplied by 10 x where X is the appropriate power of 10.
Annex 3 Enumeration of coliform (MPN) (NMKL, No. 44, 2004)
Method principle
Graduated amount of food (diluted) sample are transferred to a series of fermentation tubes
containing lactose or lauryl sulphite tryptose broth of proper strength, it is usual practice to
inoculate to three fermentative tubes. The tubes are incubated at 35+ 0.5 0C for 24 and 48hrs.The
formation of gas in any of the tubes with in 48hr ,regardless of the amount, constitutes as positive
for coliform and the absence of gas formation with in this period considered as negative for coli
form . Confirm the coliform by BGBB
Responsibilities: All trained stuff with adequate experience
0

Positive Control: any Coliform spp. For total coliform test 44.5 C gas positive E.coli for fecal
coliform test.
Negative Control: Uninoculated tubes with the media and inverted tube
0

44.5 C negative E.aerogenes for fecal coliform test.


0

Equipment to Calibrate: Incubator37 C, PH meter, Safety cabinet,


Pipettes controllers, Colony counting device, Centrifuge optional, only for some food type,
0

Stomacher optional, only for none liquid sample, Digital balance, and Water bath 44 C.
106

Media: LSTB, BGBB, EC broth,


Diluents: Peptone water
Procedure:
Presumptive test for coliform group (MPN)
1. Preparation of the first dilution (101)
Transfer 10ml of liquid sample to 90ml of diluents or 25g of sample to 225 ml of diluents in a flask
if shaker used or in sterile plastic bag if stomacher used to make 101 dilution (the first dilution) and
Mix well with shaker/stomacher. Mix homogenate by shaking and pipette 1ml into a tube
containing 9 ml of normal slain. Mix carefully by aspirating 10 times with a pipette. From the first
dilution, transfer with the same pipette 1ml to 2nd dilution tube containing 9ml of the Ns, Mix with
a fresh pipette. Repeat using 3rd or more until the required numbers of dilutions are made. Shake all
dilution carefully.
2. Inoculation
Inoculate each of 3 replicate tubes of LSTB broth per dilution (containing inverted

tubes) with

1ml of the previously prepared 1:10, 1:100 and 1:1000 dilutions using sterile pipette for each
dilution.
3. Incubation: Incubate the LSTB tubes at 35+ 0.5 0C for 48hrs.
4. Reading: Record tubes showing gas production after 48hr
5. Result reading: Record all tubes showing gas within 48+ 2hrs and refer to MPN table for the 3
tube dilution and report results as the presumptive MPN of coliform bacteria per g (or ml of liquid
product).
6. Confirmed test for coliform group (MPN)
Subculture all positive tubes showing gas within 48 + 2 hours 2 hours in to BGB broth by means of
the 3 mm loop. Incubate all BGB tubes at 35+ 0.5 0C for 48 + 2 hours. Record all BGB tubes
showing gas, and refer to the MPN table for 3 tube dilution. Report results as confirmed MPN of
coliform bacteria per g (or ml of liquid product).
107

Annex 4 Determination of coliforms, fecal coliforms and E. coli by using MPN technique
(FDA/BAM, 2006)
50 g of the sample was weighted into sterile high-speed blender jar. Frozen samples can be
softened by storing it for <18 h at 2-5C, but do not thaw. 450 ml of Butterfield's phosphatebuffered and water were blended for 2 min. If <50 g of sample are available, weigh portion that is
equivalent to half of the sample and add sufficient volume of sterile diluents to make a 1:10
dilution. The total volume in the blender jar should completely cover the blades.
Decimal dilutions with sterile Butterfield's phosphate diluents were prepared. Number of dilutions
to be prepared depends on anticipated coliform density. Shake all suspensions 25 times in 30 cm
arc or vortex mix for 7 s. Do not use pipettes to deliver <10% of their total volume. Transfer 1 ml
portions to 3 tubes for each dilution for at least 3 consecutive dilutions. Hold pipette at angle so
that its lower edge rests against the tube. Let pipette drain 2-3 s. Not more than 15 min should
elapse from time the sample is blended until all dilutions are inoculated in appropriate media.
Incubate tubes at 35C. Examine tubes and record reactions at 24 2 h for gas, i.e., displacement
of medium in fermentation vial or effervescence when tubes are gently agitated. Re-incubate gasnegative tubes for an additional 24 h and examine and record reactions again at 48 2 h. Perform
confirmed test on all presumptive positive (gas) tubes.
Annex 5 Enumeration of Staphylococcus aureus (NMKL, No. 66, 2003)
Method principle
Certain staphylococci produce enterotoxins which cause food poisoning. This ability to produce
enterotoxins, with few exceptions, is limited to those strains that are coagulase positive, and /or
produce a heat stable nuclease (TNase). This method determines the presence of S. aureus by
plating known quantities of (dilutions of) food sample onto a selective agar. After incubation
presumptive staphylococcal colonies are selected and subjected to confirmatory tests from the
results of these tests the number of S. aureus per g or ml of the food is calculated .The quantity
that present may indicate a potential for the presence of enterotoxin , or they may also indicate a
lack of adherence to good hygienic practices.
Responsibilities: All trained stuff with adequate experience
Positive Control: S.aures
108

Negative Control: E.coli


Equipment to Calibrate: Incubator 37C, Autoclave, pH-meter
Media: Baird-parker agar and Mannitol salt agar
Reagent: Human plasma or commercial rabbit plasma
Diluents: Normal saline (Ns)/peptone water
Sample preparation:
Weigh 10g of the sample in to a sterile 250ml Erlenmeyer flask marked to indicate 100ml volume.
Add sterile normal saline to 100ml mark. Dissolve and shake thoroughly.
Procedure (1): Using Baird-parker agar media
1. Preparation of food homogenate
Transfer 10g of sample with sterile spoon or other depending on the sample type in to a
sterile250ml Erlenmeyer flask containing 90ml of normal saline.
2. Dilution
2.1 mix homogenate by shaking and pipette 1ml into a tube containing 9ml of normal slain. Mix
carefully by aspirating 10 times with a pipette
2.2 From the first dilution, transfer with the same pipette 1ml to 2nd dilution tube containing 9ml of
the Ns, Mix with a fresh pipette
2.3 repeat using 3rd or more until the required numbers of dilutions are made
2.4 shake all dilution carefully.
3. Pipette 0.25 ml of the material on the plates (two plates for each dilution) and spread with a
sterile bent glass rod. The plates are incubated at 370c for 24- 48 hr.
4. select plates with 30-300 separate colonies which are black and shiny with narrow white
margins and surrounded by the zones extending in the opaque medium .Mark the position of
these colonies and re-incubate for 24-hrs. Count all colonies with the above appearance that
developed in the second 24 hrs incubation and submit these for coagulase test. Then total the
109

colonies which produced clear zones in both periods of incubations. Multiply by 4 and by the
dilution factor to calculate the number of staphylococcus auras per ml of sample.
Procedure (2): Using Mannitol salt agar media
Inoculate 0.1ml

of the sample into the surface of the medium. Incubate as above and count the

typical colonies which form yellow zones and not those surrounded by red or purple zones .This
give the number of suspected staphylococcus.
Dilution: 1:10, 1:100, 1:100, etc
Dilution factor : 1 x 101, 1102, 1103
Inoculation: spread with a sterile bent glass rod
Incubation: Incubate the prepared dishes, inverted, at 370c for 48 h
Counting colonies: following incubation, count all colonies on dishes containing 30-300 colonies
and recorded the results per dilution counted.
Verification:

If there is growth on the negative control or no growth on the positive control the

test should be repeated.


Expression of results: after calculating the average count and multiply by the dilution express the
result in cfu per g or ml (if a liquid sample).
Annex 6 Isolation of Salmonella (NMKL, No. 71, 2005)
Method principle
The procedure consists of six distinct stages. The initial handling of the food and the Non-selective
enrichment stage (pre-enrichment) very according to the type of food examined.
Non- Selective Enrichment (Pre enrichment)
The test sample is initially inoculated into a non- inhibitory liquid medium to favors the repair
and growth of stressed or sub lethally- injured salmonellae arising from exposure to heat,
freezing, desiccation, preservatives, high osmotic pressure or wide temperature fluctuations.
Selective Enrichment
110

Replicate portions of each pre-enrichment culture are inoculated into two enrichment media to
favor the proliferation of salmonellae through a selective repression or inhibition of the growth of
competing microorganism.
Selective Plating
Enrichment cultures are streaked onto selective differential agars for the isolation of salmonellae.
Purification
Presumptive Salmonella isolates are purified on MacConkey agar plates or SS agar plates.
Biochemical Screening
Isolates are screened using determinant biochemical reactions.
Serological Identification
Polyvalent and / or somatic grouping antisera are used to support the tentative identification of
isolates as members of Salmonella spp. If available, complete sero-typing is essential.
Responsibilities: All trained stuff with adequate experience
Positive control: Salmonella spp.
Negative control: Uninoculated media
Equipment to calibrate: Incubator 37oC, Blender, stomacher or other homogenizing device.
Medium: Nutrient Broth (NB), Trypticase ( Tryptic, Tryptone) Soy Broth, Brilliant Green Water.
Buffered peptone Water (BPW), Skim Milk Medium, Tetrathionate Brilliant Green Broth (TBG),
Selenite Cystine Broth (SC), Bismuth Sulfite Agar (BS), Brillian, Green Sulfa Agar (BGS),
MacConkey Agar, SS agar, Nutrient Agar, Triple Sugar Iron Agar (TSL), Lysine Iron Agar (LIA)
and Urea Agar ( Christensen's).
Reagent
1 Commercial biochemical test kits.
2 Polyvalent and single grouping somatic (O) and flagellar (H) antisera.
111

3. Physiological Saline.
Procedure
Handling of Sample Units
Analyze samples as soon as possible. If necessary, store samples under time and temperature
conditions that will prevent the growth or death of native micro flora. If sample units have been
abused in transit, re-sampling of the lot should be carried out.
Frozen Foods; sample units that show no signs of thawing upon receipt may be stored in the freezer
at- 100C to - 20 0C. Dried and shelf stable foods may be stored at room temperature. Refrigerate all
other foods, including those that are received in a partially thawed condition; analyze these
samples as soon as possible preferably within 24 h of receipt. Thaw frozen samples at room
temperature within 60 min, if this is not possible, thaw the samples at refrigerator (4 to 100C)
temperature.
NOTE:
a) Large samples (e.g. whole chicken ) may not readily thaw at refrigerator temperatures. For
greater expediency, enclose the frozen sample in a heavy- walled paper bag and thaw overnight at
room temperature. This technique maintains the product surface cold during the thawing process.
b) Appropriate containers should ensure that the drippings from the product do not contaminate the
laboratory environment.
If the sample unit received for analysis is less than the recommended analytical unit, analyze the
entire amount and record the weight used
Blending of samples should be limited to the minimum time required to produce a homogeneous
suspension. Excessive blending could result in physical damage that would adversely affect the
viability of endogenous micro flora. For products that do not require blending; disperse the
analytical unit into the appropriate pre-enrichment broth.
Use aseptic techniques and sterile equipment at all stages of analysis. Containment during the
handling of powdered products is critical if cross- contamination of the work environment is to be
avoided.
112

Non-selective Enrichment (Pre-enrichment)


Compositing of Analytical Units
To reduce the workload, up to 15x25g (ml) analytical units may be composited into a single test
sample (e.g. 375 g or ml). If a sample unit consists of more than one container, aseptically mix the
contents of the containers prior to withdrawal of the analytical unit. If not possible or practical, the
analytical unit shall then consist of equal portions from each of the containers
Sample Analysis
The required analytical unit is dispersed into a suitable active enrichment broth Nutrient broth (NB)
and buffered peptone water (BPW) are equally reliable and can be used interchangeably as
general purpose pre-enrichment. If the pH of the pre-enrichment mixture lies outside the range of
6.0-7.0, adjust with 1N Na OH or 1N HCl.
NOTE; if the sample unit consists of a container with little food material, thoroughly rinse the
interior of the container with a suitable pre-enrichment broth medium and incubate the rinse in a
sterile flask. This eventuality

is more frequently encountered in situations involving consumer

complaints or food poisoning investigations. Positive Salmonella and a negative medium control
should be set up in parallel with the test samples. Incubate the pre-enrichment mixture and the
positive and negative controls at 350.5 0C for 18 - 24 h.
Selective Enrichment
With a sterile pipette, 1.0 ml of the pre-enrichment culture into each of 9 ml of selenite cystine
(SC) and tetrathionate brilliant green (TGB) broths. Incubate SC and TGB broths for 242 h at
350.50C and 430.50C, respectively.
Selective plating
Streak replicate loopful of each selective enrichment culture onto BS and BGS agar to obtain well
isolated colonies. The enrichment cultures may be streaked onto additional plating media for the
isolation of salmonella. Incubate plates at 35 0.50C for 242 h.
If colonies suggestive of Salmonella have not developed on BS plates, incubate for an
additional

242.

Examine incubated plates for colonies suggestive of Salmonella. Typical

Salmonella usually occur as pink to fuchsia colonies surrounded by red medium on BGS agar and
113

as black colonies on BS agar with or without a metallic sheen, and showing a gradual H2Sdependent blackening of the surrounding medium with increasing incubation time.
NOTE:
Lactose- and/or sucrose- fermenting Salmonella strains develop a coliform- like (greenish)
appearance on BGS agar. A heavy growth

of non- salmonellae may also mask the presence of

Salmonella on this medium.


b. BS agar can retard the growth of Salmonella serovars other than S. typhi unless poured plates
are refrigerated (4 to 100C) for 24 h prior to streaking. The absence of suspect colonies on the
plates indicates that the analytical or composite test samples did not contain Salmonella spp.
Purification
Streak suspect colonies onto MacConkey agar for purification, Incubate plates at 350.50C for
242 h .Typical Salmonella colonies are lactose- negative and will appear as colorless colonies on
this medium. However, lactose- positive biotypes will occur as pink colonies.
Biochemical Screening
With a sterile needle, inoculate suspect colonies into the biochemical media or in commercial
diagnostic kits that would yield equivalent results. Incubate the biochemical media for 18 - 24 h at
35 0.50C.
NOTE: Erroneous biochemical results may be obtained if tubes are not loosely capped during
incubation.
Commercial diagnostic kits may be used to obtain detailed biochemical profiles of bacterial
isolates. If none of the isolates from a particular analytical unit are suggestive of Salmonella, the
analytical unit is considered to be free of salmonellae. If the presence of Salmonella is suspected,
proceed with serological testing. If serological testing is not to be performed within 72 h, inoculate
suspect isolates into nutrient agar slants and incubate at 35 0.50C for 24 2 h. Store the agar
slants at refrigerator (4 to 100C) temperature. Nutrient agar slants that have been stored for more
than 72 h should not be used for serological testing. Prepare fresh agar slants for this purpose.

114

Annex 7 Proximate compositions of the Fruit Leathers


7.1 Tests of Between-Subjects Effects on Moisture Content of the Fruit Leather
Dependent Variable: Moisture content
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

69.598a

11

6.327

11165.420

.000

Intercept

6071.893

6071.893

1.072E7

.000

temp

32.107

16.054

28329.735

.000

Pload

11.234

11.234

19824.735

.000

variety

3.241

3.241

5720.029

.000

temp * Pload

8.646

4.323

7629.029

.000

temp * variety

9.514

4.757

8394.971

.000

Pload * variety

.558

.558

984.971

.000

temp * Pload * variety

4.297

2.148

3791.206

.000

Error

.007

12

.001

Total

6141.497

24

Corrected Total

69.605

23

a. R Squared = 1.000 (Adjusted R Squared = 1.000)

115

7.2 Tests of Between-Subjects Effects on Protein Content of the Fruit Leather


Dependent Variable: Protein
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

1.900a

11

.173

40.506

.000

Intercept

134.213

134.213

31467.881

.000

temp

.913

.456

106.973

.000

Pload

.107

.107

25.087

.000

variety

.052

.052

12.274

.004

temp * Pload

.094

.047

10.973

.002

temp * variety

.442

.221

51.782

.000

Pload * variety

.007

.007

1.533

.239

temp * Pload * variety

.287

.143

33.607

.000

Error

.051

12

.004

Total

136.165

24

Corrected Total

1.952

23

a. R Squared = .974 (Adjusted R Squared = .950)

116

7.3 Tests of Between-Subjects Effects on the Fat Content of the Fruit Leather
Dependent Variable: Fat
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

25.963a

11

2.360

22.445

.000

Intercept

154.642

154.642

1470.536

.000

temp

.304

.152

1.445

.274

Pload

9.604

9.604

91.323

.000

variety

1.383

1.383

13.153

.003

temp * Pload

.192

.096

.911

.428

temp * variety

1.329

.664

6.317

.013

Pload * variety

11.931

11.931

113.453

.000

temp * Pload * variety

1.222

.611

5.810

.017

Error

1.262

12

.105

Total

181.867

24

Corrected Total

27.225

23

a. R Squared = .954 (Adjusted R Squared = .911)

117

7.4 Tests of Between-Subjects Effects on the Crude Fiber of the Fruit Leather
Dependent Variable: Crude Fiber
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

15.868a

11

1.443

324.463

.000

Intercept

1065.973

1065.973

239764.603

.000

temp

5.175

2.587

581.979

.000

Pload

.572

.572

128.718

.000

variety

.036

.036

8.176

.014

temp * Pload

.425

.213

47.824

.000

temp * variety

1.825

.912

205.236

.000

Pload * variety

.026

.026

5.819

.033

temp * Pload * variety

7.808

3.904

878.150

.000

Error

.053

12

.004

Total

1081.895

24

Corrected Total

15.921

23

a. R Squared = .997 (Adjusted R Squared = .994)

118

7.5 Tests of Between-Subjects Effects on the Ash Content of the Fruit Leather
Dependent Variable: Ash Content
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

1.758a

11

.160

4.006

.012

Intercept

92.932

92.932

2329.111

.000

temp

.713

.356

8.934

.004

Pload

.170

.170

4.266

.061

variety

.258

.258

6.458

.026

temp * Pload

.335

.167

4.193

.042

temp * variety

.019

.009

.235

.794

Pload * variety

.067

.067

1.672

.220

temp * Pload * variety

.198

.099

2.475

.126

Error

.479

12

.040

Total

95.169

24

Corrected Total

2.237

23

a. R Squared = .786 (Adjusted R Squared = .590)

119

7.6 Tests of Between-Subjects Effects on the Carbohydrate Content of the Fruit Leather
Dependent Variable: Carbohydrate
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

71.128a

11

6.466

42.520

.000

Intercept

119469.370

119469.370

785595.071

.000

temp

34.480

17.240

113.366

.000

Pload

.073

.073

.477

.503

variety

.031

.031

.203

.661

temp * Pload

2.745

1.373

9.025

.004

temp * variety

9.147

4.574

30.074

.000

Pload * variety

20.387

20.387

134.061

.000

temp * Pload * variety 4.265

2.133

14.024

.001

Error

1.825

12

.152

Total

119542.324

24

Corrected Total

72.953

23

a. R Squared = .975 (Adjusted R Squared = .952)

120

Annex 8 Vitamin C content of the Fruit Leather


8.1 Tests of Between-Subjects Effects on the Vitamin C content of the Fruit Leather
Dependent Variable: Vitamin C
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

1777.290a

11

161.572

543.837

.000

Intercept

14791.232

14791.232

49786.062

.000

temp

418.564

209.282

704.425

.000

Pload

374.065

374.065

1259.072

.000

variety

106.977

106.977

360.076

.000

temp * Pload

2.734

1.367

4.601

.033

temp * variety

663.752

331.876

1117.067

.000

Pload * variety

.988

.988

3.326

.093

temp * Pload * variety

210.210

105.105

353.775

.000

Error

3.565

12

.297

Total

16572.086

24

Corrected Total

1780.855

23

a. R Squared = .998 (Adjusted R Squared = .996)

121

Annex 9 Sensory Characteristics of the Fruit Leather


9.1 Tests of Between-Subjects Effects on the Color of Fruit Leather
Dependent Variable: Color
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

29.383a

11

2.671

8.687

.000

Intercept

1204.167

1204.167

3915.989

.000

temp

1.441

.720

2.343

.138

Pload

.240

.240

.780

.394

variety

24.000

24.000

78.049

.000

temp * Pload

.243

.121

.394

.683

temp * variety

1.852

.926

3.012

.087

Pload * variety

.060

.060

.195

.667

temp * Pload * variety

1.547

.774

2.516

.122

Error

3.690

12

.308

Total

1237.240

24

Corrected Total

33.073

23

a. R Squared = .888 (Adjusted R Squared = .786)

122

9.2 Tests of Between-Subjects Effects on the Aroma of the Fruit Leather


Dependent Variable: Aroma
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

4.331a

11

.394

3.436

.022

Intercept

1113.844

1113.844

9720.818

.000

temp

.390

.195

1.702

.224

Pload

.034

.034

.295

.597

variety

2.734

2.734

23.858

.000

temp * Pload

.210

.105

.916

.426

temp * variety

.190

.095

.829

.460

Pload * variety

.070

.070

.615

.448

temp * Pload * variety

.703

.352

3.069

.084

Error

1.375

12

.115

Total

1119.550

24

Corrected Total

5.706

23

a. R Squared = .759 (Adjusted R Squared = .538)

123

9.3 Tests of Between-Subjects Effects on the Taste of the Fruit Leather


Dependent Variable: Taste
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

8.935a

11

.812

2.291

.085

Intercept

1166.220

1166.220

3288.988

.000

temp

.323

.162

.456

.644

Pload

.570

.570

1.609

.229

variety

3.300

3.300

9.308

.010

temp * Pload

1.903

.952

2.684

.109

temp * variety

1.703

.852

2.402

.133

Pload * variety

.120

.120

.340

.571

temp * Pload * variety

1.013

.507

1.429

.278

Error

4.255

12

.355

Total

1179.410

24

Corrected Total

13.190

23

a. R Squared = .677 (Adjusted R Squared = .382)

124

9.4 Tests of Between-Subjects Effects on the Flavor of the Fruit Leather


Dependent Variable: Flavor
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

7.943a

11

.722

2.541

.062

Intercept

1131.627

1131.627

3982.264

.000

temp

.271

.135

.477

.632

Pload

.375

.375

1.320

.273

variety

3.527

3.527

12.411

.004

temp * Pload

.772

.386

1.359

.294

temp * variety

1.031

.515

1.814

.205

Pload * variety

.042

.042

.147

.708

temp * Pload * variety 1.926

.963

3.389

.068

Error

3.410

12

.284

Total

1142.980

24

Corrected Total

11.353

23

a. R Squared = .700 (Adjusted R Squared = .424)

125

9.5 Tests of Between-Subjects Effects on the Toughness of the Fruit Leather


Dependent Variable: Toughness
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

20.741a

11

1.886

5.220

.004

Intercept

853.234

853.234

2361.893

.000

temp

3.123

1.561

4.322

.039

Pload

4.594

4.594

12.716

.004

variety

4.950

4.950

13.704

.003

temp * Pload

.857

.429

1.187

.339

temp * variety

3.816

1.908

5.281

.023

Pload * variety

1.354

1.354

3.747

.077

temp * Pload * variety

2.047

1.024

2.834

.098

Error

4.335

12

.361

Total

878.310

24

Corrected Total

25.076

23

a. R Squared = .827 (Adjusted R Squared = .669)

126

9.6 Tests of Between-Subjects Effects on the Overall Acceptability of the Fruit Leather
Dependent Variable: Overall Acceptability
Type III Sum of
Source

Squares

df

Mean Square

Sig.

Corrected Model

16.005a

11

1.455

8.455

.000

Intercept

1119.300

1119.300

6504.409

.000

temp

1.961

.980

5.697

.018

Pload

1.550

1.550

9.010

.011

variety

8.050

8.050

46.782

.000

temp * Pload

.881

.440

2.559

.119

temp * variety

2.456

1.228

7.136

.009

Pload * variety

.220

.220

1.281

.280

temp * Pload * variety

.886

.443

2.574

.117

Error

2.065

12

.172

Total

1137.370

24

Corrected Total

18.070

23

a. R Squared = .886 (Adjusted R Squared = .781)

127

Pic. 6 Keitt Mango grown in Assossa region

Pic. 7 Tommy Atkins Mango grown


at Awara Melka in Afar region

Pic. 8 Oven drying of Mango leather

Pic. 9 Tommy Atkins Mango leather in the


middle of two Keitt variety leathers

128

A. Tommy Atkins Mango puree

B. Keitt Mango Puree

Pic. 10 The difference between the Tommy Atkins and Keitt variety Mango puree

Pic. 11 Oven dried Keitt variety Mango fruit leather

129

Pic. 12 Tommy Atkins Mango fruit leather

130

Pic. 13 Keitt variety Mango fruit leather

Pic. 14 Tommy Atkins variety Mango fruit leather

131

Pic. 15 Keitt Mango leather

Pic. 16 Tommy Atkins Mango leather

132

Pic. 17 Mango Fruit Leathers produced from Tommy Atkins and Keitt varieties of mango
fruits

Pic. 18 Storage of Mango fruit leathers in a refrigerator

133

Declaration

I, the undersigned, declare that this thesis is my original work and has not been presented for a
Degree in any other University, and that all sources of materials used for the thesis have been duly
acknowledged.

Name: Binyam Teshome Tesfaye

Signature:

___________________________

Place: Addis Ababa, Ethiopia

Date of submission:

___________________________

This thesis has been submitted for examination with my approval as University advisor.

Name: Mr. Adamu Zegeye

Signature:

______________________

134