_ .:.i..
t 2. V 3... 1....
3.5235... . A.

‘ r

93,
5 an... ..
. trans... a; V .. __ A
5‘ g a i ‘ ‘ . .l
x .mubuwnau‘ w .. «v .u
, . V .Qm‘. : C in» .3 .
. r an...” gr! .uralaad : a 3.
1:5... :53... 9N1}. “*1...le ’5‘
, 33.07. 41% 5‘
2:35 15‘3“.“ c3. *1.
.§.§..w2é.

.
. .~PI\€
52
.21.:1...‘
ll .uVQ-dn , “Ll?
6:1..WET .. . .
. a. H.

. a!
a. . 53...:
s .
3.11133... 3.. y W .it
...r . ﬁnialéjitiﬂ
.33. . ‘9 J.
.? 17L
s
, . It:
5...}...5 s

.d 1:531...‘

:35... .2 .
i . 17:11.
.

. I:
5.93.: .a

.

hunk?) ‘
, , . 3.».
. ﬂash... r:

, ‘ y Runny...

:11 J‘
t a ‘
iii?“
«(zit-5.356.. :55...
3...: «2.55113.-

. a
2. i5 .391 .
. .155. Iago...

a!)
0!

V 23.15":
. - sill): .. I 1:
L3)! szdmilleu-ﬁ. MVAWHFV
J: n. 5.. . s
. .a «1-15.:
A .9 I...
, .. f «a. p.
.. .2.- x‘. 2.1}! “mam".

« . “V“"lg‘ﬁb‘ . ,0 p‘
’33:. “3-5: 1|». 11.nrw..d.ﬂuwau.twwhu,
. hi. 45 ﬁx but.
Q tr!) ~
.

.11..
‘1‘...
. ..

t.

3....
.4314“.
a ~ .
1
.52..
x... p 3;...
355-‘is‘5‘l ‘
a. 1 1:315}.
:1 x. .t I: .
. nah y.d....§nr .935: ~.
6.31., 3. .30
, .x

v(\ .5 v.9.
t! «83:5..‘41;
‘1T

91 r. 2
ottﬁ‘rlli. v0...A.

 

 

3%;1 g. g «aﬁﬁ ~

i3 ,wif?

c.-

 

T112858

«J (1..

l"\

' LIBRARY
Michigan State
University

This is to certify that the

thesis entitled

Using Developed Laboratory Procedures for Discriminating
Potential of Selected Michigan-Grown Soybean
Varieties for Soymilk and Toﬁi Production

presented by

Dianne Thuy Trinh

has been accepted towards fulﬁllment
of the requirements for

Master of Sciencuegree in Food Science

“W 13

Major professor

 

 

Dacha. 7, 200]

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

DATE DUE , DATE DUE DATE DUE
1 01 q '1 ﬁ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIFiC/DateDuepeS-p. 15

USING DEVELOPED LABORATORY PROCEDURES FOR
DISCRIMINATING POTENTIAL OF SELECTED
MICHIGAN-GROWN SOYBEAN VARIETIES FOR SOYMILK
AND TOFU PRODUCTION

By

Dianne Thuy Trinh

A Thesis
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2001

ABSTRACT
USING DEVELOPED LABORATORY PROCEDURES FOR DISCRIMINATING
POTENTIAL OF SELECTED MICHIGAN-GROWN SOYBEAN VARIETIES FOR
SOYMILK AND TOFU PRODUCTION
By

Dianne Thuy Trinh

Seven soybean seed varieties from two Michigan locations were analyzed for
physical properties including seed weight, shape, split seeds, and mixed color of seed
coats, and for chemical properties such as moisture and protein contents. All soybeans
exhibited signiﬁcant differences in seed size, moisture content, and protein content.
Soybeans from Sanilac County in general had higher seed weight, more split seeds, and
were lower in moisture content than those of Allegan County. Using a developed
procedure, soymilk was produced ﬁ'om all seven varieties ﬁ'om both locations. Its quality
was investigated for temperature, total volume, color, pH, density, total solids, and
protein content. Signiﬁcant differences were detected in soymilk volume and protein
content among soybean varieties. Color, pH, density, and total solids were similar among
all soybean varieties studied. Soybeans from Sanilac County produced higher soymilk
volume than those of Allegan County. Toﬁi, produced with a developed procedure from
all seven varieties from both counties, was evaluated for quality. Yield, moisture and
protein contents, and texture were used as parameters to discriminate soybean quality
among varieties. Signiﬁcant differences were exhibited for tofu yield and protein
content, but not for tofu moisture and texture. Location and soybean seed size did not
affect tofu yield, but there were positive correlations between soybean protein and each

of the following: soymilk protein, tOﬁJ yield, and toﬁi protein.

ACKNOWLEDGMENTS

I would like to express my deepest gratitude to my advisor Dr. P.K.W. Ng for his
endless guidance, scientiﬁc insight, encouragement, and friendship. Grateful
acknowledgment is also extended to Dr. J.N. Cash and Dr. J.A. Partridge for serving as
members of my guidance committee, for their comments, technical advice, and support
throughout my graduate program.

Appreciation is also expressed for my ﬁnancial support received ﬁ'om Dr. P.KW.
Ng, the Department of Food Science and Human Nutrition, Michigan State University,
and the Michigan Soybean Promotion Committee. Without these assistantships, I would
have had a dimcult time completing my studies. Thanks are also given to fellow group
members in the cereal science laboratory: E. Tanhehco, R. Mujoo, J.Y. Cha, and D. Ngo
for their help and ﬁ’iendships; and to R1. Wolthuis of Agricultural Engineering
Department for constructing the tofu molding boxes.

I would like to give special thanks to my parents and my parents-in-law for
providing me with opportunities to learn; for their love, patience, unending
encouragement, and, especially, for their part in taking care of my daughter and newborn
son while completing my thesis. Above all, deepest love goes to the most loving person in
my life, my husband, Hao The Trinh, for everything he has done dming my undergraduate
and graduate years. I also would like to dedicate this work to him since it could not have

been accomplished without his support and encouragement.

iii

TABLE OF CONTENTS

Chapter Page
LIST OF TABLES ................................................................................ ix
LIST OF FIGURES ............................................................................... xi
1. INTRODUCTION .............................................................................. 1
II. LITERATURE REVIEW ......... ........................................................... 3
2.1 Soybeans .................................................................................... 3
2.1.1 Seed Morphology ................................................................ 7
2.1.2 Seed Composition ................................................................ 8
2.1.3 Proteins ........................................................................... 10
2.1.4 Oil .................................................................................. 11
2.1.5 Nutritive Value and Health Beneﬁts .......................................... 13
2.1.6 USDA Soybean Grades and Quality Evaluation ............................. 15
2.2 Soymilk Production ....................................................................... 17
2.2. 1 Introduction ....................................................................... 1 7
2.2.2 Traditional Soymilk Preparation ................................................ 20
2.2.3 Modern Soymilk Preparation ................................................... 20
2.2.4 Soymilk Composition ............................................................ 22
2.2.5 Soymilk Quality ................................................................... 22
2.2.5.1 Soybean Variety ........................................................ 22
2.2.5.2 Beany Flavor ............................................................ 23
2.2.5.3 WaterzBean Ratio ...................................................... 23

2.2.5.4 Soaking Time ........................................................... 25

2.2.5.5 Effect of Chemical Composition ..................................... 25

2.2.6 Packaging and Quality Evaluation ............................................. 26

2.3 Toﬁr Production ........................................................................... 28
2.3.1 Introduction ....................................................................... 28
2.3.2 Tofu Processing .................................................................. 29
2.3.3 ToﬁJ Composition ................................................................ 30
2.3.4 Coagulants ........................................................................ 32
2.3.4.1 Types of Coagulants ................................................... 32

2.3.4.2 Concentration of Coagulants ......................................... 34

2.3.4.3 Coagulation time ....................................................... 34

2.3.5 Toﬁi Quality ....................................................................... 34
2.3.5.1 Sources of Water ...................................................... 35

2.3.5.2 WaterzBean Ratio ...................................................... 35

2.3.5.3 118 and 7S Fractions .................................................. 36

2.3.6 Quality Evaluation ............................................................... 36

III. RESEARCH OBJECTIVES ................................................................. 38
IV. MATERIALS AND METHODS ............................................................ 38
4.1 Materials .................................................................................... 38
4.2 Methods ..................................................................................... 38
4.2.1 Phase I: Soybean Analysis ....................................................... 40
4.2.1.1 Physical Analysis ....................................................... 40

4.2.1.1.] Seed Weight ............................................ 4O

4.2.1.1.2 Shape .................................................... 4o

4.2.1.1.3 Split seeds ............................................... 40
4.2.1.1.4 Mixed Color of Seed Coats ........................... 42

4.2.1.2 Chemical Analysis ....................................................... 42
4.2.1.2.1 Moisture Content ....................................... 42
4.2.1.2.2 Protein Content ......................................... 43

4.2.2. Phase II: Soymilk Processing and Analysis .................................... 43
4.2.2.1 Soymilk Preparation ..................................................... 43
4.2.2.2 Soymilk Analysis ......................................................... 45
4.2.2.2.] Temperature ............................................. 45
4.2.2.2.2 Total Volume of Soymilk .............................. 45
4.2.2.2.3 Color of Soymilk ....................................... 46
4.2.2.2.4 pH ........................................................ 46
4.2.2.2.5 Density .................................................. 46
4.2.2.2.6 Total Solids Content ................................... 46
4.2.2.2.7 Protein Content ......................................... 47

4.2.3 Phase III: TOﬁJ Processing and Analysis ....................................... 47
4.2.3.1 Tofu Processing ......................................................... 47
4.2.3.2 Tofu Analysis ............................................................ 49
4.2.3.2.] ToﬁJ Yield ............................................... 49
4.2.3.2.2 Texture ................................................... 49
4.2.3.2.3 Moisture Content ....................................... 51
4.2.3.2.4 Protein Content ......................................... 53

4.2.4 Phase IV: Statistical Analysis .................................................... 53

V. RESULTS AND DISCUSSION ................................. ' .............................. 54
5.1 Soybeans .................................................................................... 54
5.1.1 Physical Analysis .................................................................... 54
5.1.2 Chemical Analysis .................................................................. 57
5.1.3 Relationships among Quality Parameters and Quality Evaluation ............ 59
5.2 Soymilk ..................................................................................... 61
5.2.1 Analyses of Soymilk Temperature, Volume, Protein Content and
Total Solids ......................................................................... 61
5.2.2 Soymilk Color, pH, and Density ................................................. 66
5.2.3 Relationships among Soymilk Volume, Total Solids, Protein Content
and Color ........................................................................... 68
5.2.4 Soymilk Quality Evaluation ....................................................... 68
5.3 Toﬁl ......................................................................................... 72
5.3.1 Toﬁi Yield .......................................... ’ ................................. 72
5.3.2 Chemical Analysis .................................................................. 72
5.3.3 Tofu Texture Analysis ............................................................. 75
5.3.4 Relationships among Tofu Yield, Moisture Content, Protein Content,
and Texture ......................................................................... 78
5.3.5 Toﬁt Quality Evaluation ........................................................... 81
5.4 Correlation Relationships among Quality Parameters of Soybean, Soymilk,
and Tom .................................................................................... 81
VI. SUMMARY AND CONCLUSIONS ...................................................... 85
VII. SUGGESTIONS FOR FURTHER RESEARCH ........................................ 87

vii

APPENDICES ................................................................................. 88

SOYMILK:

APPENDIX 1: Comparison between Hot and Cold Grinding Procedures

for Making Soymilk and Tofu Using Vinton Soybeans
of 1999 ................................................................ 89

APPENDIX 11: Data for Different Soybean Grinding Times to Produce
Soymilk Using Vinton Soybeans of 1999 ....................... 90

APPENDIX III: Mean Values of Soymilk Volume, Total Solids,
Protein Content, and Color of Seven Soybean Varieties
(2000) from Two Locations ..................................... 91

APPENDIX IV: Mean Values of Color Analysis from Commercial
Soymilk and Cow’s Milk .......................................... 92

TOFU :

APPENDIX V: Tofu-Making Formulae Used for Diﬁ'erent Soymilk
Volumes, and Resultant Toﬁl Yield Using Vinton
Soybeans of 1999 .................................................. 93

APPENDIX VI: Tofu Parameters for Tofu Made with Different Coagulant
(Citric Acid) Concentration Using Vinton Soybeans
of 1999 ............................................................. 94

APPENDIX VII: Comparison between Calcium Sulfate Food Grade
Coagulant and Pure Chemical Calcium Sulfate as

Coagulants for Making Tofu ..................................... 95
APPENDIX VIII: Effect of Curd Forming Time on Tofu Yield Using
Vinton Soybeans of 1999 ....................................... 96
APPENDIX IX: Effect of Curd Forming Time on Tofu Yield Using
Vinton Organic of 1999 ........................................... 97
APPENDIX X: Mean Values of Tofu Texture Analysis of Seven Soybeans
Varieties (2000) in Two Locations ............................... 98
BIBLIOGRAPHY ............................................................................... 99

viii

LIST OF TABLES

Table Page
2.1 World Soybean Production, 1997-1999 ................................................ 4
2.2 International Soybean Trade, 1997-1999 ............................................... 6
2.3 Possible Molecular Forces Involved in Formation and Maintenance

of the Structural Matrix of 11S and 7S Globulin Gels ................................ 12
2.4 Amino Acid Composition of Soybeans, Soymilk, and TOﬁJ ......................... 14
2.5 The U. S. Grades and Grade Requirements for Soybeans ........................... 16
2.6 Composition of Soymilk, Cow's Milk, and Human Milk ............................. 19
2.7 Comparison of Three Basic Categories of Soymilk ................................... 24
2.8 Heat Treatment and packaging Considerations for Commercial Soymilk .......... 27

5.1 Mean Values of Soybean physical Analysis Data Among Seven Soybean
Varieties ﬁom Two Locations ........................................................... 55

5.2 Analysis of Variance for Soybean Seed Weight, Moisture Content, and
Protein Content Among Seven Soybean Varieties from Two Locations .......... 56

5.3 Mean Values of Soybean Chemical Analysis Data Among Seven Soybean
Varieties from Two Locations .......................................................... 58

5.4 Correlation Coefﬁcients for Relationships between Soybean Quality
Parameters ................................................................................. 60

5.5 Analysis of Variance for Soymilk Volume, Total Solids, Protein Content,
and Color Among Seven Soybean Varieties from Two Locations .................. 65

5.6 Mean Values of Soymilk pH and Density from Seven Soybean Varieties ......... 69

5.7 Intercorrelation Relationships Betwwen Soymilk Volume, and Total Solids,
Protein Content, and Color ............................................................... 70

ix

5.8 Mean Values of Tofu Yield, Protein Content, and Moisture Content of

Seven Soybean Varieties ﬁom Two Locations .....................................

5.9 Analysis of Variance for Tofu Yield and Protein Content Among Seven

Soybean Varieties from Two Locations .............................................

5.10 Analysis of Variance for Tofu Texture Among Seven Soybean Varieties

from Two Locations ...................................................................

5.11 Intercorrelation Relationships Among TOﬁl Yield, Moisture Content,

Protein Content, Hardness, Cohesiveness, and Gumminess .......................

5.12 Mean Values for Tofu Skin Tests (N) of Toﬁl Samples from Seven

Soybean Varieties Grown in Two Counties .........................................

5.13 Correlation Relationships Among Evaluative Parameters of Soybeans,

Soymilk, and Tofu ﬁom Seven Soybean Varieties Grown in Two Locations. ..

...73

...74

...77

..80

..82

.83

LIST OF FIGURES

Figure Page
2.1 Soybean Varieties with Different Shapes and Color of Seed Coats ................. 9
2.2 Flow Diagram for Preparation of Soymilk and Tofu .................................. 21
2.3 Flow Diagram for Tofu Production ....................................................... 31
4.1 Soybean Analyses ........................................................................... 41
4.2 Developed Procedure for Making Soymilk ............................................. 44
4.3 Developed Method for Tofu Processing with Calcium Sulfate ....................... 48
4.4 Molding and Pressing Tofu ................................................................ 50
4.5 Texture Proﬁle Analysis ................................................................... 52
5.1 Soymilk Volumes of Seven Soybean Varieties from Two Locations ................ 62
5.2 Soymilk Protein Content of Seven Soybean Varieties ﬁ'om Two Locations ....... 63
5.3 Soymilk Color of Seven Soybean Varieties from Two Locations ................... 67

5.4 Tofu Texture Analysis Among Seven Soybean Varieties from Two Locations....76

5.5 Tofu Texture Curve ﬁom Variety Vinton 81 Grown in Allegan County ............ 79

"Images in this thesis are presented in color."

xi

1. INTRODUCTION

For more than 5,000 years, soybeans have been cultivated in the East and
provided a nutritional protein source for human diets in many Asian countries. They
have been utilized in the preparation of many soyfood products, which were introduced to
the West less than a century ago (Chen, 1989). In recent years, soyfood products have
been attracting consumers because of their ﬁinctional and nutritional values (Lo, 1989).
Among these various soyfood products, soymilk and tofu are non-fermented products that
have attracted much attention. They are also considered good sources of protein and
isoﬂavones (Messina, 1994). The modernization of technology and the understanding of
soybean chemistry have helped improve traditional methods and increase soymilk and
tofu production. These new procedures have made soyfood products more palatable and
more digestible (Liu, 1999).

Although the U. S. has become the world’s largest soybean producer, most of the
soybeans grown are used for animal feed or oil production, and this is true of Michigan-
grown soybeans as well. If more soybeans were consumed by humans, the crop value
would substantially increase. The per capita consumption of soymilk and tofu has
increased rapidly, and a large number of other soyfood products have entered commercial
production over the past few years (Liu, 1999). Thus, evaluation of different soybean
varieties for soymilk and/or toﬁJ production potential is important, and would provide
information to breeders for selecting new varieties with soymilk and/or tofu production

potential. However, currently no standard procedures for the preparation of soymilk and

tofu are available. Therefore, the need for developing standard procedures for preparing

both soymilk and tofu is great.

II LITERATURE REVIEW

2.1 Soybeans

Soybeans are considered to be one of the oldest crops of the East and an important
agricultural commodity in many Asian countries. They emerged as a domesticated crop
around the eleventh century BC. in Northern China, probably by the early Shang dynasty.
The English word “soybean” can be traced back to the beginning of the Christian Era,
when the word “shu” was reported to refer to soybean. With traditional processing
methods and small-scale production, soybean products such as soymilk, tofu, soy paste,
soy sauce, and soy sprouts as well as soybean cultivation in China were introduced to
Japan, Korea, and many other countries in Asia around 1,100 years ago. The Japanese
modiﬁed the traditional Chinese methods and developed a large number of new soyfood
products; for example, two of those products are natto (a bacteria-fermented soyfood) and
tempeh (a yeast-fermented soyfood) (Liu, 1999).

In 1712, Engelbert Kaempfer, a German botanist, ﬁrst introduced the soybean to
Europe. It was later given its genetical name, Glycine max, by a Swedish botanist, Carl
Von Linne. Around the mid-eighteenth century, soybeans were introduced to the United
States where they were used as a forage crop rather than as food for human consumption.
This changed in the early 1900s, when the ﬁrst crushing plant was built and operated in
1911 to extract soybean oil, and later in 1922, when the ﬁrst soybean processing plant was
developed in Illinois (APV, 1998; Liu, 1999).

In a short period of time, the US. became the world’s largest soybean producer,

responsible for almost half of the total world soybean production by 1995 (Table 2.1)

Table 2.1. World Soybean Production 1997-99

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Country 1997-98 1998-99
1,000 metric tons 1,000 metric tons
Canada 2,738 2,737
United States 73,176 74,598
Argentina 19,500 19,900
Bolivia 1,071 620
Brazil 32,500 31,000
Paraguay 2,988 3,000
Italy 1,243 1,192
China 14,728 15,000
India 5,350 6,000
Indonesia 1,306 1,300
Korea, Democratic People’s 420 420
Thailand 338 335
World total 158,072 158,93 1

 

USDA, 2000.

 

(Liu, 1999). Almost one-third of the US. soybean production was exported to the
Netherlands, Japan, China, and other countries during 1997-1999 (Table 2.2); and of the
non-exported soybean crop, the majority was pressed for oil, or used for oil products and
animal feed. Only 1% of recent total soybean production was processed and used as food
for human consumption in the U. S. (Liu, 2000).

Recently, the use of modern technology has improved the quality of soyfood
products to suit Western tastes, resulting in subsequent increases in their consumption.
Modern technology has also reduced processing costs, and improved shelf life of soyfood
products. Recent medical research has delineated the health beneﬁts of soyfoods such as
reducing cholesterol level and heart diseases, inhibiting the growth of breast, colon and
prostate cancer cells, and preventing osteoporosis and menopausal symptoms (Messina
1994, 1999; Riaz, 1999; Alekel and Germain, 2000). Therefore, soybeans are now
considered to be functional foods (Riaz, 1999). The US market has classiﬁed soyfood
products into six groups: soy oil, traditional soyfoods, soy protein products, new-
generation soyfoods, soy-emiched foods, and functional soy ingredients/dietary
supplements (Liu, 2000).

In general, there are two types of traditional soyfood products: nonfermented and
fermented. Soymilk, toﬁt, soy sprouts, and yuba (soy ﬁlm) are the most common
nonfermented soyfoods, whereas soy sauce, miso, tempeh, and fermented tofu are the
most common fermented soyfood products. Among the nonfermented soyfoods,
traditional soymilk and tofu were developed by Lord Liu An in China around 164 BC.

(Shurtleﬂ‘ and Aoyagi, 1990).

Table 2.2. International Soybean Trade, 1997-1999

Principal exporting countries

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Country 1997-98 1998-99
1,000 metric tons 1,000 metric tons
United States 23,761 21,813
Brazil 9,325 9,300
Argentina 3,231 3,300
Paraguay 2,390 2,400
Netherlands 560 1,100
Canada 769 850
Others 1,049 826
World total 41,085 39,589
Principle importing countries
Country 1997-98 1998-99
1,000 metric tons 1,000 metric tons
Netherlands 5,000 4,800
Japan 4,873 4,650
Chim 2,940 3,850
Germany 3,406 3,650
Mexico 3,479 3,600
Spain 3,044 3,050
Taiwan 2,387 2,200
Korea, Republic of 1,340 1,450
Belgium 1,279 1,200
Indonesia 810 1,150
United Kingdom 1,005 950
Italy 768 820
Thailand 600 750
Others 7,628 7,444
World Total 38,559 39,564

 

 

 

 

USDA, 2000.

 

 

2.1.1 Seed morphology

The soybean belongs to the family of leguminosae, subfamily Papilionoideae.
There are three species: Glycine ussurensis (wild), G. max (cultivated), and G. gracilis
(intermediate) (Vaidehi and Kadam, 1989). Glycine max is the most commonly grown
and belongs to the subgenus Soja which has 2n = 40 chromosomes (V errna and
Shoemaker, 1996).

Food legumes are classiﬁed into two types: pulses and oilseeds. Soybeans are
oilseeds (Shurtleﬂ‘ and Aoyagi, 1990). Most legume seeds are made up of three parts: the
seed coat, embryo, and food storage structures. The soybean seed, however, has only two
parts: the seed coat or hull, and the food storage structures. The seed coat accounts for
8-10% ofthe bean’s weight and halfofthe seed's ﬁber; it protects the seed ﬁ'om fungi and
bacteria. The hilum or seed scar is linear-elliptical in shape. There is a small groove or
chalaza of the hilum where the swd coat is joined to the body. Another point at the other
end, called the micropyle, is the opening in the seed coat where the primary root of the
young seedling comes out and through which the bean respirates. Next to the micropyle is
the outline of the hypocotyl or germ, which accounts for 3% of the bean weight. This is
the place where the bean sprouts or grows (Williams, 1950; Liu, 1999).

The food storage structures are comprised of two cotyledons, of which the embryo
is a part. The cotyledons constitute most of the seed bulk, contain mainly protein and oil,
and account for 60-89% of the seed weight. They supply food to the seedling plant during
germination and initial grth (Williams, 1950; Liu, 1999).

Different soybean varieties have different shapes, sizes, color of hila and seed

coats. Soybean seeds vary in shape, from almost spherical, to oval, to elongated, or ﬂat

(Fig. 2.1). On average, the weight of a single seed is 120-180 mg, but larger seeds may
weigh about 260 mg (Disoy, Magna, and Prize varieties), or even more than 500 mg (with
certain genotypes in germplasm collections). There is no identiﬁed standard color for the
hilum: its color is generally reported as black, imperfect black, brown, buﬁ', gray, or
yellow. The color of the seed coat appears to be yellow, green, brown, black, or a
combination of these colors (Fig. 2.1). Soybeans with dark color of hilum or seed coats
lend soymilk and tofu a grayish color; in contrast, soybeans with a light yellow or clear
hilum and seed coats give the whitest color for the end products (Shurtleﬁ‘ and Aoyagi,
1990)

2.1.2 Seed composition

Although soybean seeds of different varieties have similar morphology, they may
vary in chemical composition The differences in physical and chemical compositions of
soybean seeds depend not only on the variety but also on environmental inﬂuences such as
climatic conditions, fertilizer level of the soil, locality where the seeds are grown, etc.
(Vaidehi and Kadam, 1989).

On a dry basis, the four major chemical components of a soybean seed are
proteins, carbohydrates, oil, and ash, constituting about 40%, 35%, 20%, and 5%,
respectively, of the total composition (Liu, 1999). Proteins are stored in protein bodies
(2-20 pm in diameter). Oil is located in small structures called spherosomes (0.2-0.5 pm)
which are located in between the protein bodies (V aidehi and Kadam, 1989). With a
distinct lack of starch, the carbohydrate component consists mainly of sugars (soluble
carbohydrates including sucrose, raﬂinose, stachyose, and verbascose) and ﬁber (insoluble

carbohydrates including cellulose, hemicellulose, and pectin). The major minerals are

 

_»;r .5: I

Figure 2.1. Soybean Varieties with Different Shapes and Color of Seed Coats.

(William, 1950).

Chalky-white

. Yellow seed coat

. Green seed coat

. Maternal green-cotyledon
Genetic green-cotyledon

. Colorless-hilum

Colored-hilum

"Saddle"-pattern
"Saddle"-pattern, black eyebrown
10. Black seed coat

11. Concentric marking of black or brown
12. Flecking of brown or black

13. Noninherited molting

l4. Smudging caused by black pods
15. Bloom on soybean seed

16. Black soybean

17. Brown soybean

18. Red-Brown soybean

19. Imperfect-black soybean

20. Buff soybean

pwsmwewwr

21. Reddish-buff soybean

22. Netted cracking of outer layer of seed
coat

23. Abnormal hilum and defective seed coat

24. Cracking of seed on buff-colored seed

25. Cracking of seed coat in pigment area

26. Seed of wild soybean

27. Small, ﬂat seed

28. Medium-size round seed

29. Large, round seed

30. Large ﬂat seed

potassium, iron, copper, manganese, calcium, magnesium, zinc, and cobalt (Garcia et al.,
1997). Soybean seed also contains water-soluble vitamins (thiarnin, riboﬂavin, niacin,
pantothenic acid, and folic acid), oil-soluble vitamins (vitamin A and E) (Liu, 1999), and
other minor components (protease inhibitors, phenolic compounds, lectin, saponins, and
phytates) that have both beneﬁcial and deleterious effects on the nutritional quality of
soybean seeds (Garcia et al., 1997).

2.1.3 Proteins

With proteins constituting about 40% of the total seed composition, Liu (1999)
has suggested that the soybean should be termed a protein seed rather than an oilseed.
Soybean protein molecules are compactly folded in their native states. The hydrophobic
amino acid side chains of polypeptides are buried inside the molecular subunits, exposing
the hydrophilic amino acids on the surface. Therefore, native soybean proteins are soluble
in water, and about 90% of these proteins can be extracted with water. Additionally,
about 90% of soybean proteins can be precipitated at a low pH of 4.5-4.8, and these are
referred to as acid-precipitable proteins and play an important role in food processing
(Fukushima, 1991).

There are three types of soybean proteins: the proteins involved in metabolism, the
structural proteins, and the storage proteins. Storage proteins are found in high
concentration, and comprise about 80-90% of the total protein content. They can be
fractionated into ﬁve groups: 28 (or-conglycinin), 7S ([3- and y-conglycinin), 9S, 11S
(glycinin), and 153 globulins (Garcia et al., 1997). The main storage proteins are glycinin
and B-conglycinin. Glycinin (11S) and B-conglycinin (7S) constitute about 70% of the

total seed proteins (Maruyama and Sato, 1999). Upon heating, these 7S and 1 IS fractions

10

each form a speciﬁc gel that is dependent on the number of ﬁ'ee -SH groups and S-S
bonds present in amino acid residues that form their respective peptide chains. The 11S
globulins have 2 ﬁee -SH groups and 20 S-S bonds per molecule. These groups and
bonds maintain the molecule's quaternary structure which undergoes only minor change
during denaturation On the other hand, the 7S globulins have no ﬁ'ee -SH groups and
only 2 S-S bonds per molecule. These globulins interact more upon heating and give a
uniform gel with a network-like structure.

Gel formation and the stabilization of network structure also depend on the
molecular forces involved during heating such as hydrogen bonding, hydrophobic
interactions, ionic interactions and disulﬁde linkages (Utsumi, 1985). These forces are
different within 118 and 7S fractions (Table 2.3). Molecules from the 118 fraction have
more afﬁnity to each other than do molecules from the 7S fraction since each fraction
retains its respective number of S-S bonds during heating and 118 molecules have
substantially more S-S bonds. In contrast, 7S molecules have only hydrogen bonds with
which to form gel structures. As a result, gel formed from the 11S ﬁ'action is ﬁrmer tl'mn
gel ﬁ'om the 7S ﬁaction (Utsumi and Kinsella, 1985; Fukushima, 1991).

2.1.4 Oil

Oil constitutes about 20% of the total soybean composition. It is composed of
88.1% neutral lipids, 9.8% phospholipids, and 1.6% glycolipids (Garcia et al., 1997;
Vaidehi and Kadam, 1989). The neutral lipids primarily consist of triglycerides (99%),
small portions of ﬁve free fatty acids [palmitate (16:0), stearate (18:0), oleate (18:1),
linoleate (18:2), and linolenate (1823)], sterols, and sterol esters (Yadav, 1996). In

addition to some minor phospholipid compounds, there are four major phospholipids

ll

Table 2.3. Possible Molecular Forces Involved in Formation and Maintenance of the
Structural Matrix of 118 and 7S Globulin Gels

 

 

 

 

 

 

 

Possible molecular forces involved in
Primary Globulin Formation Maintenance
Component of gel
1 1 S Hydroph
obic interactions
Electrostatic interaction Disulﬁde bonds
Disulﬁde bonds Hydrogen bonds
7S Hydrophobic interactions
Hydrogen bonds Hydrogen bonds

 

 

 

 

 

Utsumi and Kinsella, 1985.

12

present: 35% phosphatidyl choline, 25% phosphatidyl ethanolamine, 15% phosphatidyl
inositol, and 5-10% phosphatidic acid (Liu 1999).

When the seed is cracked during storage or crushed during processing at ambient
moisture and temperature, enzymes and substrates interact in the presence of oxygen and
moisture resulting in rancidity. Polyunsaturated fatty acids and the enzyme lipoxygenase
are accountable for off-ﬂavors and off-aromas of soy and soyfood products during storage
and processing. They are also responsrble for the beany ﬂavor. The ﬂavor compounds
produced are carbonyl compounds, phenolic acids, volatile fatty acids and amines, volatile
neutral compounds, alcohols, and phosphatidyl choline (V aidehi and Kadam, 1989). The
undesired ﬂavor limits wide acceptance and consumption of soyfood products by
Westerners.

2.1.5 Nutritive value and health beneﬁts

Soybeans are a high source of protein, dietary ﬁber, micronutrients (folate, iron,
zinc, and calcium), and phytochemicals (isoﬂavones). Its low cholesterol level gives
soybean unique nutritional value (Messina, 1999). Soybean protein is highly digestible
(92-100%) (Riaz, 1999) and contains most essential amino acids (Table 2.4) in amounts
that exceed the amino acid requirements for children and adults; therefore, soybeans are
considered a good source of protein among legumes (Garcia et al., 1997). Soymilk rmde
from soybeans also is lactose-free, thus soymilk can be used as a substitute for cow’s milk
for lactose intolerant people.

Many researchers have demonstrated that consumption of soybeans prevents
certain heart diseases and improves kidney function for kidney disease patients (Riaz,

1999; Messina, 1999). On October 20, 1999, the Food and Drug Administration (FDA)

13

Table 2.4. Amino Acid Composition (g/l6g Nitrogen) of Soybeans, Soymilk, and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tofu
Amino Acid Soybeans Soymilk Tofu
Aspartic Acid 12.61 11.91 11.70
Threonine 4.1 1 4.01 4.00
Serine 5.74 5.19 5.32
Glutarnic Acid 19.76 19.61 19.26
Proline 5.53 5.33 5.47
Glycine 4.46 4.16 4.14
Alanine 4.49 4.36 4.1 l
Valine 3.73 4.88 4.99
Cystine 0.78 0.03 trace
Methionine 1 .34 1 .59 1.43
Isoleucine 3.46 4.66 4.85
Leucine 7.90 7.94 8.32
Tyrosine 3.90 3.91 3.99
Phenylalanine 4.85 5.15 5.41
Lysine 6. 19 6.08 6.41
Histidine 2.60 2.64 2.64
Arginine 8.64 8.65 8.52

 

Wand and Calvins, 1989.

14

 

approved a health claim that consuming 25g of soy protein per day, along with a diet low
in saturated fat and cholesterol, will help reduce the risk of coronary heart disease
(Ranhotra et al., 2000). Isoﬂavone compounds in soybeans (1-3 mg/g protein) are found
to inhibit the growth of cancer cells, and appear to be associated with lower incidences of
breast, colon, and prostate cancers. Soybean intake also lowers osteoporosis and
menopausal symptoms (Messina 1994, 1999; Riaz, 1999; Alekel and Germain, 2000).
Moreover, soy-based diets can help to control weight because they provide high-quality
protein and high nutrient meals, yet are relatively low in calories (Riaz, 1999). Consuming
soyfoods has also been reported to reduce gallstone formation, aging, Alzheimer’s disease,
and to strengthen the irmnune system, building resistance against cold, ﬂu, and HIV (Riaz,
1999; Arditi et al., 2000); however, scientiﬁc studies are needed to conﬁrm these claims.
All the above reports promote and attract consumers and processors to soyfood and soy-
enriched food products at higher rates than ever.
2.1.6 USDA soybean grades and quality evaluation

In the U. S., according to 1999 regulations, USDA Grade No. l soybeans must
weigh a minimum of 56 pounds per bushel, not exceed 10% splits (the two cotyledons
separated or more than one-quarter of the seed removed), 2% total damaged beans, 0.2%
heat-damaged bean, 1% foreign material, nor exceed 1% black, brown, or bicolor beans
(Table 2.5). In addition, soybean seeds must have about 13% moisture (Shurtleﬁ' and
Aoyagi, 1990). Grades No. 2-4 soybeans have progressively lower minimum weight
requirements per bushel and are lower in moisture content. They have higher limits for
damaged beans, and the presence of foreign material, splits, and soybeans of other color.

In the past, inspection was based on visual and subjective evaluation, which left room for

15

Table 2.5. The U. S. Grades and Grade Requirements for Soybeans

 

Numerical Grades

 

Grading factors

2

3

 

Minimum limits of
Test wgght (lbs/Bu)

56.0

54.0

52.0

49.0

 

Maximum percent limits of
Damaged kernels
Heat damaged
Total damaged
Foreign material
Splits
Soybeans of other color

0.2

2.0

1.0

10.0
1.0

0.5
3.0
2.0
20.0
2.0

1.0
5.0
3.0
30.0
5.0

3.0
8.0
5.0
40.0
10.0

 

Maximum count limits of
other materials

Animal ﬁlth

Castor beans

Crotalasia seeds

Glass

Stones

Unknown foreign
substance

Total

 

 

SuwoN—c

 

ﬂ
cwWONt—O

 

ﬂ
cwwoN—xo

 

smch—o

 

Liu, 1999.

16

 

human variability and error. Recently, grain inspection has become operated with
electronic imaging coupled with computer technology (Liu, 1999).

Soybean quality is determined based on moisture content using a moisture meter,
protein content using the Kjeldahl method, and crude fat content using the Soxlet or
Goldﬁsh method (Moizuddin et al., 1999). The ratio of 11S to 78 protein components is
calculated by quantifying these ﬁactions using a densitometer aﬁer sodium dodecyl
polyacrylamide gel electrophoresis (Iwabuchi and Yamauchi, 1987). The amount of
phosphorus is measured following the method of Lim et a1. (1990). Other quality
parameters include assessing seed size by weighing 100 seeds, and visually determining the
color of hila and seed coats. Splits and cracked seeds are counted and the level of swd

cleanliness is checked.

2.2 Soymilk production
2.2.1. Introduction

Soymilk is a milk-like product, extracted from soybeans that have been soaked,
followed by cold or hot grinding, ﬁltering, and cooking. This traditional method was
developed by Lord Liu An of Huai-nan around 164 B.C.. Soymilk has been used in China
and some parts of Asia for a long time; but it was only in 1910 that Yu-Ying-Li, a Chinese
who lived in Paris, designed a patent for soy “dairy” production. It was accepted as the
ﬁrst patent for soymilk in Britain In 1923, a soymilk factory in Changsa, China, ﬁrst
produced soymilk in bottles. Later, in 1936, an American medical doctor, Harry W1llis
Miller, started the ﬁrst soymilk plant in Shanghai, China. In 1940, a large-scale

commercial soymilk production plant was established by K. S. Lo in Hong Kong, and in

17

1945, this company was renamed the Hong Kong Soya Bean Products Co. Ltd. and
became the producer of the world's ﬁrst soymilk drink. Now it is named Vitasoy (APV,
1998; Chen, 1989).

Over the past several decades, Chinese and other Asian immigrants have inﬂuenced
and brought the art of preparing soyfoods, including soymilk, to the West. Now, soymilk
is widely consumed as a nutritious drink and in soy-based infant formulas. It can be a
substitute for cow’s milk for lactose-intolerant people, and is similar to human milk in
gross composition (Table 2.6) (Chen, 1989).

Chen (1989) described three factors contributing to the increase in consumption of
soymilk: (1) technological improvement, (2) nutrition, and (3) health conscientiousness.
To adapt to Western demand, modern technology has modiﬁed the traditional Chinese
method, and researchers have been developing various procedures to enhance soymilk
quality as well as decrease its beany ﬂavor. There are two methods for soymilk
preparation: the traditional method, wherein soybeans are soaked and ground in cold
water, the slurry is ﬁltered, and the resultant soymilk is cooked (Liu, 1999); and the
modern method (hot grinding method), wherein the unsoaked beans are ground in hot
water, followed by cooking the slurry and ﬁltering out the soymilk (Wilken et al., 1967).
This method reduces the beany ﬂavor by inactivating the enzyme lipoxygenase. Today,
many soymilk manufactures apply modiﬁed methods described by Schroder and Jackson
(1972), Nelson et a1. (1976), Johnson et a1. (1981), and others, to produce soymilk. These
methods involve soaking soybeans that are then blanched and ground in hot water, or even

the use of defatted soy or soy ﬂour as starting materials.

18

 

Table 2.6. Composition of Soymilk, Cow's Milk, and Human Milk

 

 

 

 

 

 

 

 

Item/100g Soymilk Cow’s milk Human milk

Calorie (Kcal) 44 59 62
Water (g) 90.8 88.6 88.2
Protein (g) 3.6 2.9 1.4
Fat (g) 2.0 3.3 3.1
Carbohydrates (g) 2.9 4.5 7.1
Ash (g) 0.5 0.7 0.2
Minerals (mg)

Calcium 15 100 35

Phosphorus 49 90 25

Sodium 2 36 15

Iron 1.2 0.1 0.2
Vitamins (mg)

Thiamine (B1) 0.03 0.04 0.02

Riboﬂavin (312) 0.02 0.15 0.03

Niacin 0.50 0.20 0.20
Saturated fatty acids (%) 40-48 60-70 55.3
Unsaturated fatty acids (%) 52-60 30-40 44.7
Cholesterol (mg) 0 924-99 9.3-18.6

 

Chen, 1989.

19

 

2.2.2. Traditional soymilk preparation

The traditional Chinese method for preparing soymilk and tofu (Fig. 2.2) is still
widely utilized by many soyfood manufacturers. Basically, whole soybeans are soaked in
tap water overnight (8-12 hours), drained, washed, and ground with tap water at a
waterzbean ratio of around 8:1 or 10:1. The slurry is ﬁltered through cheesecloth, the
residue called okara is removed, and the ﬁltrate is boiled and stored. Asians have
accepted and enjoyed soymilk produced by this method, but Westerners do not like its
beany ﬂavor. Many soymilk producers have tried to incorporate other ﬂavors such as
chocolate, strawberry, etc., to mask the beany ﬂavor.

2.2.3. Modern soymilk preparation

In 1967, Wilkens and colleagues developed a new method at Cornell University in
the U. S.. The unsoaked, dehulled soybeans are ground in hot water, after which the
slurry is heated in a steam-jacketed kettle at 80-100°C for 10 minutes to imctivate
enzymes. The slurry is centrifuged to separate okara and soymilk. The soymilk is then
formulated (adding sugar and ﬂavors), bottled, and sterilized at 121°C for 12 minutes.
This is known as the hot-grinding method or the Cornell method. This method has been
successful in reducing the beany ﬂavor of the ﬁnal product.

Commercial soymilk producers have used sodium bicarbonate to improve soymilk
ﬂavor. They have also used defoamers, sweeteners, ﬂavorants, vitamins, minerals,
stabilizers, soy oil, lecithin, and emulsiﬁers to increase its creaminess, mouth feel, and to
improve its quality and taste. Today, soymilk is variously treated with pasteurization,

sterilization, ultra high temperature treatment, and homogenization to increase shelf life

20

WHOLE SOYBEAN S

 

 

Soaking ‘ Water
Rinsing e Water
Grinding 4 Water

 

l

Okara 4— Filtering

1

Cooking

1

SOYMILK

l

Coagulating ‘— Coagulant

l

Whey ‘— Pressing

l

TOFU

Figure 2.2. Flow Diagram for Preparation of Soymilk and Tofu. (Liu, 1999)

21

and reduce chalkiness of the soymilk (Chen, 1989). Soymilk is packaged and sold in
aseptic cartons, reﬁigerated containers, or in powdered form.
2.2.4. Soymilk composition
Soymilk contains approximately 90% water, 3.5-4.0% protein, 2.0% fat (about 40-
48% saturated fatty acids and 52-60% unsaturated fatty acids), 3.0% carbohydrates, 0.5%
ash (calcium, phosphorus, sodium, and iron), and vitamins (thiamin, riboﬂavin, niacin)
(Table 2.6). The total solids and protein contents of soymilk are affected by the
waterzbean ratio (the total weight of water added during soaking and processing per
weight of dry beans) (Chen, 1989).
2.2.5 Soymilk quality
There are many factors that affect soymilk quality: soybean variety, seed size, the
percentage of split seeds, the color of hilum and seed coats, total solids content
(waterzbean ratio), fat, protein, and carbohydrate contents, water source, soaking time,
and method of soymilk preparation. The following sections brieﬂy describe some of these
factors.
2.2.5.1 Soybean variety
Not all soybean varieties are bred for the same end-use purpose; some are for
animal feed and others for oil, soymilk and/or tofu production. In general, high quality
varieties for soymilk and tofu have large seeds, a colorless hilum, thin clear and light
yellow seed coats, yellow cotyledons, and are higher in protein content and lower in oil
content (Liu, 1999). The light colored hilum beans yield the whitest soymilk. The

amount of split seeds play an important role in the degree of off-ﬂavor of the soymilk: if

22

the seed lot has a large percentage of split and cracked seeds, natural oils inside the seeds
become rancid and affect the ﬂavor of the end products.
2.2.5.2 Beany ﬂavor

The beany ﬂavor of soymilk made with traditional processing has been a main
drawback to Westerners. This ﬂavor can be removed by modern processing methods,
such as blanching the soybeans prior to grinding (Nelson et al. 1976), or using a hot
grinding method with unsoaked soybeans (Wilkens et al. 1967). The enzyme lipoxygenase
is responsible for the beany ﬂavor. Activation of this enzyme is initiated by splitting or
cracking of the seed, which releases the enzyme, but it requires the presence of oxygen
and moisture. If the seeds are soaked overnight (8-12 hours), the enzymes are still
inactive; the beany ﬂavor only occurs when the tissues of the cotyledons are damaged by
grinding (Liu, 1999). The undesired volatile compounds include ketones, aldehydes, and
alcohols (Wilkens et al., 1967; Liu 1999).

With increasing temperature of the slurry (ground beans and water) above 80°C,
there is a decrease in both the number and volume of the volatile compounds produced
(Liu, 1999). Wilkens et al. (1967) reported that a temperature of at least 80°C is required
to prevent the development of volatile compounds. The lipoxygenase enzyme is also
inactivated by altering the pH of soymilk to below pH 3 or above pH 10 (Chen, 1989),
though products in these pH ranges are not suitable for human consumption.
2.2.5.3 Water:bean ratio

Soymilk is classiﬁed as "rich", "dairy-like", and "economy" depending on the
amount of water added during soaking and processing (Table 2.7). When less water is

added (e.g., the ratio of waterzbeans is only 5:1 or 6:1 (v/W)), soymilk has a higher solids

23

Table 2.7. Comparison of Three Basic Categories of Soymilk

 

 

 

 

 

 

 

 

 

 

Soymilk Water:bean ratio Solids Protein Fat

Type (liter of water/ (%) (%) (%)

kg soybean)

Rich 5:1 or 6:1 10-11.5 4.5-5.2 2.8-3.2
Dairy-like 8:1 or 8.5:] 7.4-8 3.3-3.6 2.1-2.3
Economy 10:1 6 2.7-3.3 1.2-1.6

Chen, 1989.

24

 

content (protein and ﬂat), and is considered "rich" soymilk (Chen, 1989). This soymilk can
be used to prepare silken toﬁr or yuba. The "dairy-like" soymilk can be used as a
substitute for cow’s milk because of their similarities in fat and protein contents. When
more water is used during processing, the solids content is lower; the ﬁnal product is
referred to as "economy" soymilk and is consumed as a drink (Chen, 1989).

Water comprises a signiﬁcant portion of the soymilk produced, and as a major
volume contributor, can inﬂuence the ﬂavor of the end product. To yield best-ﬂavored
soymilk, added water needs to be cold and not contain any chemical or other
contaminants. Well water or spring water is preferred for soymilk and tofu production as
it is cold and does not contain ﬂuorine or chlorine (Shurtleﬂ‘ and Aoyagi, 1990).
2.2.5.4 Soaking time

Soybeans are soaked prior to processing to soften their cellular structures, increase
protein and nutrient solubility, reduce the amount of energy required for grinding,
decrease cooking time, and increase product yield (Shurtleff and Aoyagi, 1990; Liu,
1999). The soaking time depends on the temperature of the water used since water
temperature affects the rate of bean hydration. The higher the temperature, the faster the
rate of soybean hydration (Wilkens and Hackler, 1969). In general, soybeans are soaked
in ambient temperatures for around 8-12 hours in order to achieve an increase of 2.2 times
their initial weight, or an increase of around 2.4 times in volume (Shurtleff and Aoyagi,
1990). Longer soaking time will increase the loss in total solids content (Liu, 1999).
2.2.5.5 Effect of chemical composition

Soymilk has a lower fat content (2.0%) than cow’s milk or human milk (3.3 and

3.1%, respectively) (Table 2.6), and is rich in polyunsaturated fatty acids and lecithin.

25

Lecithins are pho spholipids, which are essential components of cell membranes, and
essential for the growth, maturation, and functioning of all body cells. Fat in soymilk is
highly digestible and has a caloric value of 16.74 cal per 200 ml of soymilk (Ang and
Kwik, 1985). Higher fat content also affects the aftertaste of soymilk (sour, bitter, and
astringent) resulting ﬁ'om the presence of higher levels of phenolic acids, oxidized
phospholipids, and oxidized fatty acids (Liu, 1999).

Final protein content of soymilk (2.7-5.2%) (Table 2.7) depends mainly on the
waterzbean (v/w) ratio. It inﬂuences the color of soymilk; high protein soymilk usually is
more opaque in appearance.

Soymilk contains only 2.9% carbohydrates which gives 2.02 cal per 200 ml of
soymilk (Ang and Kwik, 1985). To increase the energy and/or the palatability of the
product, sweeteners are added (in the form of sucrose) to around 9-10% (weight of
sucrose/weight of soymilk) (Chen, 1989). Vitamins and minerals are also added to
increase nutritional values and total solids, from around 10% (prior to additives) to 12.5-
15.7% (Ang and vaik, 1985).

2.2.6 Packaging and quality evaluation

Soymilk, with over 90% water, is an ideal medium for bacterial growth. The use
of raw materials, the speciﬁc processing conditions, and the types of packaging utilized all
determine the shelf life of the ﬁnal product. Commercial soymilk produced using different
techniques of pasteurization, sterilization, and ultra high temperature treatment keep

soymilk fresh ﬁom one week to 6-8 months (Table 2.8).

26

Table 2.8. Heat Treatment and Packaging Considerations for Commercial Soymilk

 

 

 

 

 

 

 

 

 

 

Treatment Temperature Time Packaging Shelf life
(°C)
Pasteurization 75 15 sec Plastic bag 1 week
Gable top reﬁigerated
Glass bottle
Sterilization 121 20 min Can 2 years
Glass bottle nonreﬁigerated
Retort pouch
Ultra-high 140 2 sec Aseptic 6-8 mo.
temperature nonreﬁigerated
Liu, 1999.

27

 

Commercial soymilk with a protein content of less than 2% is identiﬁed as
soydrink (Ang and Kwik, 1985). In the West, soymilk with milk-like color, bland ﬂavor,
minimal aftertaste (sour, bitter, and astringent), and less chalkiness is considered to be
high quality soymilk (Liu,1999; Chen, 1989). In general, more dilute soymilk is less
preferable by consumers (Liu, 1999).

Soymilk is tested for quality by its manufacturers and in laboratories using many
parameters: color with the Hunter Lab method, protein content with the Kjeldahl method,
crude fat content with the Goldﬁsh method (Moizuddin et al., 1999), bacteria with a plate
count method, total solids with an oven method or following Johnson’s and Wilson's
method (1984), pH with a pH meter, speciﬁc gravity with a Quevenne Lactometer or
hydrometer (Chen, 1989). The yield of soymilk is reported as the total volume of soymilk
received alter ﬁltering. Volatile components in soymilk are determined by gas

chromatography (erkens et al., 1967).

2.3 Tofu production
2.3.1 Introduction

Toﬁr (Japanese), Dau hu (Vietnamese), Teou fu or Tou fu ho (Chinese) is a
cheese-like product with a mild ﬂavor, a white or creamy color, and a bland taste. The
soybean curd is precipitated ﬁom soymilk by the addition of coagulants such as calcium
and magnesium salts, or an acid (Schaefer and Love, 1992).

The making of tofu is believed to have developed around the same time as that of
soymilk, which was around 164 BC, by Lord Liu An. It was made using nigari or

seawater as a coagulant and this traditional method was followed for years. Then around

28

900 years later, toﬁr-making spread to Japan and other Asian countries and became
popular as a meatless food. The word “tofu” started appearing in writing in the 1500s.
The Japanese modiﬁed the tofu-making procedure with new coagulants (calcium sulfate,
magnesium salts, etc.) and new techniques of cooking (pasteurization, steaming, etc.)
(Liu, 1999). For centuries, toﬁr has been considered a nutritious, inexpensive, and
versatile food in the East, but it was introduced to the West only less than a century ago
(Liu, 1999). In the U. S., tofu started to appear on the market around 1980 and the ﬁrst
commercial toﬁr factory in the U. S. went into production around 1990 (Golbitz, 1995).
Tofu manufacturers and production have kept increasing rapidly and because tofu has a
blandtaste,itwasmoreeasilyacceptedintheWestthansoymilkandhasbeenconsumed
directly fresh, cooked, fried, or mixed with other ingredients.

Commercial tofu is classiﬁed into four types: soft, regular, ﬁrm, and extra ﬁrm
(Shurtleff and Aoyagi, 1990; Liu, 2000), depending on the type of coagulant used,
concentration of coagulant used, and pressure applied to the curd as well as pressing time.
Tofu is sold in different packages and in several processing types: fresh reﬁigerated (silken
or kinogoshi), pasteurized and reﬁ'igerated, aseptic, ﬁ'eeze-dried (kori), deep-ﬂied
(narnage), etc.. In Japan, reﬁigerated, pasteurized, and deep-ﬂied forms of toﬁr are found
commonly on the market; on the other hand, only reﬁigerated and aseptic types of toﬁr are
found in America and Europe (Mm'phy et al., 1997).

2.3.2 Tofu processing
The traditional Chinese method for preparing tofu was developed many centuries
ago, but in 1964, Dr. Tokuji Watanabe and his co-workers at Tokyo’s prestigious

National Food Research Institute published the ﬁrst scientiﬁc tofu processing method

29

(Shurtleﬁ and Aoyagi, 1990). There are two main steps for preparing tofu: (1) the
preparation of soymilk, and (2) the coagulation of soymilk (Fig. 2.3).

Basically, clean soybeans are ﬁrst soaked overnight (8-12 hrs.), with a volume of
water three times the soybean weight, to obtain double the initial weight of the soybeans.
The soaked beans are drained and ground with fresh water (hot or cold). During this step,
the beany ﬂavor develops. The amount of water used for grinding is adjusted to a
waterzbean ratio (water is the total water uptake by the beans during soaking plus the
water added for grinding) of 10:1 (Shurtleﬂ‘ and Aoyagi, 1990), 6.5:] (Prosoya Inc,
1998), or 6:1 (Liu, 1999). The slurry either is ﬁltered to collect soymilk and okara, and
the soymilk then cooked; or the slurry is cooked, and then ﬁltered to separate soymilk and
okara. Next, soymilk is heated to 70-85°C, and coagulant is added to the soymilk with
constant stirring for 30 sec. Soymilk is allowed to coagulate in a covered container for 5
min (Schaefer and Love, 1992), 8 min (Cai and Chang, 1999), 10 min (Ji et al., 1999; Cao
and Chang, 1998, 1999), 15-20 min (Shrutleﬂ' and Aoyagi, 1990), 30 min (Liu, 1999), or
60 min (Shen et al., 1991; Escueta et al., 1986). The curd is transferred to a tofu box for
molding and pressing for about 15 min (Lim et al., 1990), 30-60 min (Shurtleﬁ‘ and
Aoyagi, 1990), or overnight (Schroder and Jackson, 1972). Toﬁr is covered with ﬁ‘esh
water and stored under refrigerated conditions.

2.3.3 Tofu composition

The composition of tofu varies depending on the type: 88-90% water and 6%
protein for soft tofu; 84.9% water and 7.8% protein for regular tofu; 79.3% water and
10.6% protein for ﬁrm tofu; and extra ﬁrm toﬁr has less than 76% water and more than

10% protein (Saio, 1979; Shurtleﬁ‘ and Aoyagi, 1990). Tofu, in general, also contains

30

Clean, Dry Soybeans

 

 

 

 

Wash and soak ¢ Water
(8-12 hrs.)
Drain > Water
Grind 4 Water
j (hot or cold)
Cook Slurry « Defoarner
(10:1 water:beans (optional)

100-110°c, 15 psi.)

l

Extract Soymilk ; Okara
(6-8% solids)

l

 

 

 

Coagulate « Coagulant
(75-85°C) (salt or acid:
2-3% by initial
l wt of dry bean)
Remove Supernatant , Whey
Press Curds to Form Toﬁr , Whey

 

(0.05-0.2 psi. 15-20 min.)

i

Cut Tofu
(cool inwater 5°C
60-90 min.)

i

Package & Reﬁigerate
(2-4°C)

Figure 2.3. Flow Diagram for Tofu Production.

31

about 3.4-4.3% fat, 1.9-2.3% carbohydrates (including ﬁber), 0.6-0.7% ash, and vitamins
(Smith et al., 1960; Shurtleﬂ‘ and Aoyagi 1990).

Tofu contains 17 out of 20 essential amino acids, the exceptions being asparagine,
glutamine, and tryptophan Glutamic acid and aspartic acid are found in highest
concentrations of 19.53% and 11.62%, respectively. In contrast, there is very little cystine
in tofu (Schroder and Jackson, 1972; Wang and Calvins, 1989).

2.3.4 Coagulants

Formation of tofu is mainly caused by the coagulation of soymilk proteins by salts,
acids, or enzymes (Liu, 1999). For good tofu yield, producers need to choose an
appropriate coagulant for each desired type of tofu product.
2.3.4.1 Types of coagulants

Toﬁr is prepared using many types of coagulants: natural nigari (salt that is

extracted ﬁom sea water), sea water, magnesium chloride, calcium chloride, calcium
sulfate, magnesium sulfate, glucono-delta-lactone (GDL), citrus juices, vinegar, and lactic
acid (Shurtleﬁ and Aoyagi 1990; Liu, 1999). In general, commercial 10111 is mainly
produced with calcium sulfate, GDL, or a mixture of calcium sulfate and GDL, since
these coagulants are inexpensive and give higher toﬁr yield than other coagulants (Tsai et
al., 1981).

In the U. S., natural nigari, magnesium chloride, and GDL are not considered
GRAS (Generally Recognized as Safe), but this does not mean that they are unsafe to use
(Shurtletf and Aoyagi, 1990). On the other hand, food grade calcium sulfate produces the
smoothest textured tofu (Smith et al., 1960), increases nutritional value of tofu by

increasing its calcium level, gives a bland taste to the soybean curd (Schroder and Jackson,

32

1972), and increases tofu yield (Shurtleﬁ and Aoyagi, 1990; Metussin et al., 1992).
During coagulation, calcium sulfate lowers the pH of soymilk, and calcium ions combine
with phytic acid, water and protein molecules to form a larger network structure,
increasing the tofu yield (Lim et al., 1990; Cai and Chang, 1998). However, calcium
sulfate is very sensitive to the environment and easily absorbs moisture, which lowers its
concentration and the subsequent toﬁr yield; therefore, it should be kept in an airtight
container and measured to prepare the coagulant solution just prior to pouring into the hot
soymilk (Shurtleff and Aoyagi, 1990).

Glucono-delta-lactone is commonly used to make silken toﬁr. A mixture of cold
soymilk and GDL is poured into sterilized packages, sealed, and heated to 70-90°C for
30-50 min to give silken tofu. There is no pressing step and whey is not separated from
this tofu. The action of GDL on curd formation is different than that of other coagulants.
1n aqueous solution, heat dissociates the GDL to produce gluconic acid, which lowers the
soymilk pH, coagulates soymilk proteins to form curd, and gives a mild-tasting ﬁnal
product. Toﬁr ﬁ'om this aseptic technique is protected ﬁom spoilage microorganisms and
is more shelf-stable than tofu made with the regular pressing method (Shurtleﬂ‘ and
Aoyagi, 1990; Murphy et al., 1997).

In Japan, commercial toﬁr is prepared with a mixture of GDL and calcium sulfate.
Packaged silken tofu uses a coagulant ratio of 4:1 (GDchalcium sulfate), and unpackaged
silken tofu a ratio of 7:3 or 1:1 (GDchalcium sulfate) (Shurtleff and Aoyagi, 1990). This

mixture of coagulants gives better quality tofu and higher tOﬁl yield (Liu, 1999).

33

2.3.4.2 Concentration of coagulants

Coagulant concentration also plays an important role in preparing tofu. Different
concentrations will produce toﬁr with varying yield, protein content, and texture.
According to Tsai et al. (1981), calcium sulfate concentration from 0.025-0.030M gives
the best yield of tofu from soymilk and a higher protein content for the toﬁr. Increasing
the concentration ﬁom 0.03-0.04M, will decrease the tofu yield, but increase the ﬁrmness
and hardness (Metussin et al., 1992). Overall, 0.03M of calcium sulfate gives the highest
yield and higher protein content of tofu (Metussin et al., 1992).
2.3.4.3 Ceagulation time

Shurtleﬂ‘ and Aoyagi (1990) identiﬁed that good coagulation was when the
soybean curd separated and moved away ﬁom the edges of the container. The color of
whey is also an indicator of the completeness of coagulation (Moizuddin et al., 1999).
The whey needs to be clear, without any white particles of soymilk. Recent studies have
used different coagulation times to form soybean curd (see section 2.3.2), and reported
tofus with different textures and yields.
2.3.5 Tofu quality

The art of making toﬁr ﬁ'om soymilk is similar to the making of cheese from cow’s
or goat’s milk. The technique lms improved with technology and the understanding of
protein chemistry, but making tofu of consistent yield and quality is not easy since there
are many variables affecting tofu yield and quality. These variables include the protein
content of soybeans, chemical composition of soybeans and soymilk (ratio of 11S to 7S

proteins), source of water, total solids content of soymilk, degree of heat-denaturation of

proteins, type of coagulants, temperature of coagulant solution, coagulant concentration,

34

stirring speed when adding coagulant to soymilk, weight for pressing tofu, and pressing
time (Saio, 1979).
2.3.5.1 Sources of water

Making tofu utilizes a large amount of water, 30 pounds of water per pound of dry
soybeans (washing beans 4 lbs, soaking 3 lbs, rinsing 2 lbs, grinding and cooking 8 lbs,
cooling tofu 5 lbs, and cleaning equipment 8 lbs) (Shurtleﬂ‘ and Aoyagi, 1990). Therefore,
choosing an inexpensive source of water is very important. Tap water containing ﬂuoride
and chloride may cause tofu to soften; but if well or spring water is used, minerals such as
calcium or magnesium contained in the water will be the main concern. Hard water (with
the presence of calcium and magnesium salts, especially calcium carbonate) is more
difﬁcult for soybeans to absorb (Jackson and Shin, 1979) and lowers the ﬁnal tofu yield.
2.3.5.2 Water:bean ratio

The water:bean ratio for soymilk production has been studied as it relates to the
ﬁnal tofu product, with different ranges of water added ﬁ'om 5-10:1 (Tsai et al., 1981),
6:1 (Cai and Chang, 1998), 6.5:1 (Prosoya Inc., 1998), 7:1 (Shen et al., 1991), and 9-14:1
(Beddows and Wong, 1987). Each level of water used gave a different tofu yield and
texture. Total solids relates to the protein content of both beans and soymilk. Total solids
content of soymilk also relates signiﬁcantly to toﬁr yield and is controlled by the amount of
water added during grinding. The amount of water that is held in the curd relates to tofu
yield. Using different water:bean ratios for tofu processing, Beddows and Wong (1987)
reported that a water:bean ratio of 10:1 gave the best results for the soybean varieties they

tested.

35

2.3.5.3 118 and 78 fractions

Upon heating, IIS and 7S polypeptides interact with coagulants and form a gel.
Within the 11S ﬁaction, disulﬁde bonds, and hydrophobic and electrostatic interactions
form the protein networks; whereas, hydrogen bonds and hydrophobic interactions are
important in the formation of gel from the 7S fraction proteins. The dissociation of
proteins ﬁ'om both 11S and 7S fractions upon heating helps them interact with each other
and form macrocomplexes that act as a matrix for holding water (German et al., 1982;
Damodaran and Kinsella, 1982; Utsumi and Kinsella, 1985). In the study of Saio and
Watanabe (1978) and Utsumi and Kinsella (1985), pure 118 gels were found to have a
higher water-holding capacity and were harder than 7S gels. Based on these observations,
it has been hypothesized that increasing the ratio of 11S to 78 globulins (1.6 to 3.2)
increases tofu ﬁrmness because more covalent bonds are produced through disulﬁde
bonding resulting in stronger overall molecular forces and a harder tofu (Ji et al., 1999).
2.3.6 Quality evaluation

To control the quality of tofu, various evaluating methods are used by
manufacturers and laboratories. These include measuring protein content with the
Kjeldahl method, moisture content with an oven- or vacuum-drying method, crude lipids
content with the Goldﬁsh method (Schaefer and Love, 1992; Moizuddin et al., 1999) or
Soxhlet method (Metussin et al., 1992), and ash content using H2804/H202 digestion or a
mufﬂe furnace (Metussin et al., 1992). Calcium and magnesium contents are determined
by the atomic absorption spectrophotometric method (Cai and Chang, 1998), or with
EDTA titration (Hach Inc., 1982). Phytic acid analysis is performed with the Schaefer and

Love (1992) method. The amino acid composition is analyzed using the method of

36

Hackler and Stillings (1967) or with a Beckrnan Spinco machine (Schroder and Jackson
1972)

The texture of tofu is tested based on seven parameters: hardness, brittleness,
chewiness, gumminess, elasticity, cohesiveness, and adhesiveness, all of which can be
evaluated with a Texturometer (Bourne et al., 1978). Many researchers have also used
Instron proﬁling analysis to examine the rheological properties of tofu, using the Instron
Universal testing machine (Lee and Rha, 1978; Saio, 1979; Lee et al., 1983; Cai and
Chang, 1999) and the Rheometer (Tsai et al., 1981).

Yield of tofu is expressed by weight in grams of ﬂesh tofu made from 100g of dry
soybean (Wang, 1993; Cai and Chang, 1999), or kg of ﬂesh toﬁr per kg of soybean (Lim
et al., 1990). Color of tofu is determined by the Hunter Lab method (Tsai et al., 1981).
Sensory evaluation is performed by panelists with visual examination, ﬁnger touching, and
a grading scale method (Stone and Sidel, 1993).

Whey is also analyzed for volume, pH with a pH meter, color with Hunter Lab
techniques, light transmittance with a spectrophotometer, and conductivity with a

conductance meter (Moizuddin et al., 1999).

37

III RESEARCH OBJECTIVES

The objectives of this proposal are to test the laboratory procedures developed in
the preliminary studies for producing (l) soymilk and (2) tofu for the purpose of
discriminating among Michigan-grown soybeans for potential soymilk- and/or toﬁr-

making qualities.

IV MATERIALS AND METHODS

4.1 Materials

Different Michigan-grown soybean varieties from 1999 and 2000 crops were used
in this study. For developing soymilk- and tofu-making procedures in the preliminary
studies, seven soybean varieties (Vinton, Vinton organic, oil feed, GMO, IA 2034, IA
3006, and NK20-20) of the year 1999 were used (Data obtained ﬁom these premilinary
studies were listed in Appendices I, II, V, VI, VII, VH1, and IX). To test the procedures
and discriminate among Michigan grown soybean varieties, seven soybean varieties of
the year 2000 (IA 2020, IA 2034, Steyer, HP 204, Vinton 81, Novartis SZOF8, and

Novartis $2020) from two counties (Allegan and Sanilac) were selected.

4.2 Methods
The harvested seed samples were stored in labeled containers at 4°C in a
reﬁ'igerator until use. Four phases were planned for testing the developed laboratory

procedures for making soymilk and tofu.

38

M91

The physical (shape, seed weight, mixed color of seed coats, and percent of split
seeds) and chemical (moisture content and protein content) characteristics of each variety
ﬁom the 2000 Michigan-grown soybean crops were evaluated.
PM

The analyzed soybean varieties were used to prepare soymilk using the VS 40
Prosoya Soycow System (Prosoya, Ottawa, Canada). Three replicates for each variety at
each location were done. Total volume, temperature, density, color, pH, total solids, and
protein content of each soymilk sample were determined.
Phaselﬂ

Soymilk from each of the soybean varieties was used to prepare tofu using
calcium sulfate as the coagulant. Two replicates from each variety at each location were
used ﬁom which to collect quality data. Yield of tofu was determined as kilograms of
ﬂesh tofu per 1 kg of dry soybeans. Moisture and protein contents of the toﬁr samples
were determined and texture analysis was also carried out to determine the ﬁacturability,
hardness, cohesiveness, and gumminess of each tofu sample.
Phase IV

Statistical analyses were reported as mean values with standard deviations; the
coefﬁcients of variation were calculated, and correlation relationships between
parameters were carried out. The Analysis of Variance (ANOVA) was used to examine
for signiﬁcant differences among variables. All of the above parameters listed for Phases
1, II, and HI were used for comparing the quality of the soybean samples used, and of the

soymilk and the tofu prepared from each of the soybean varieties.

39

4.2.1 Phase I: Soybean analysis

After receiving raw materials, soybeans were kept in airtight containers and stored
in a refrigerator (4°C) to maintain quality. Soybean seeds were analyzed for quality
based on physical and chemical characteristics (Fig. 4.1).
4.2.1.1 Physical analysis

Soybeans of each variety were tested visually for physical characteristics at room
temperature. Three aliquots of one hundred seeds (or part seeds) were taken randomly
from each of the soybean bags and subjected to the following amlyses. Each analysis
was carried out on each aliquot (i.e., three replicates), and results were recorded as
percentage mean for each characteristic.
4.2.1.1.] Seed weight

One hundred soybean seeds were transferred into a pre-weighed aluminum dish
(diameter 6 cm, height 1.5 cm). Weight (g/ 100 seeds) of each variety was recorded and
compared.
4.2.1.1.2 Shape

The same lOO-soybean seed aliquots from the above analysis were visually
inspected to determine whether individual seeds were spherical, oval, elongated, or ﬂat.
4.2.1.1.3 Split seeds

When examining the same 100 soybean seeds, any split seeds (the two cotyledons
separated or more than one-quarter of the seed removed) were counted individually and

the results were compared with the US. Grading Chart (Table 2.5, P. 16).

4o

Soybeans

Physical analysis Chemical analysis
_ Seed weight Moisture content
_ Shape Protein content
—- Split seeds
— Mixed color of seed coats

 

Figure 4.1. Soybean Analyses.

41

4.2.1.1.4 Mixed color of seed coats

The same 100-soybean seed aliquots were visually examined for mixed color of
seed coats: clear, light yellow, yellow, green, brown, black, or a combination of these
colors. The percentage of mixed color seed coats was reported.
4.2.1.2 Chemical analysis

Each chemical analysis was done with three replicates for each variety at each
location and the results reported in percentage means. Additionally, just prior to protein
content analysis, a 250-g soybean seed sample of each variety was ground with a coffee
grinder (Model KSM4, Braun Inc., Woburn, MA) until very ﬁne soy ﬂour was obtained.
4.2.1.2.] Moisture content

Moisture content analysis of whole soybean seeds was performed using a
Motomco moisture meter (Model No. 919, Safe-Grain, Inc., Mason, Ohio) according to
AACC 44-11 (AACC, 2000). Soybeans were taken out of the reﬁigerator, still stored in
their airtight containers, and left at room temperature overnight. A 250-g sample of
soybean seeds was poured into the chamber of the moisture meter after calibration and
the seed temperature was recorded. Results were collected, compared with the provided
soybean chart ﬁom Safe-Grain, Inc., and the moisture content was calculated based on
the individual temperature of each seed sample.

The moisture content of each soy ﬂour sample was analyzed following AACC
Method 44-31 (AACC, 2000) with some modiﬁcations. A S-g soy ﬂour sample was
weighed in a pre-weighed aluminum weighing dish, dried in an air oven (Model 737F,

Fisher Scientiﬁc Co., Pittsburgh, PA) for 2 hours at 130°C, cooled in a dessicator, and the

42

ﬁnal dried weight recorded. The percent moisture was determined based on the weight
loss of the sample during drying.
4.2.1.2.2 Protein content

The protein content of soybeans was determined by a micro-Kjeldahl procedure
following AACC Method 46-13 (AACC, 2000) with some modiﬁcations. A 100-mg
sample of soy ﬂour was weighed on weighing paper, and then digested for 3 hours with
digestion block DS-20 (Tecator, Hoganas, Sweden), followed by distillation with a 1003
Distilling Unit (Tecator, Hoganas, Sweden), and the solution was titrated with HCl for
nitrogen content determination. The protein content was calculated with the nitrogen

factor of 6.25.

4.2.2 Phase II: Soymilk processing and analysis

Each soybean variety was used to make soymilk with three replicates on three
different days. From each batch of soymilk, an aliquot of fresh soymilk was saved and
analyzed with three replicates to determine color, pH, density, total solids, and protein
content.
4.2.2.1 Soymilk preparation

A modiﬁed laboratory procedure for making soymilk was developed in
preliminary studies (Fig. 4.2). Soymilk was prepared using a Prosoya Soycow System
(Model VS 40, Prosoya Inc., Ottawa, Ontario, Canada). A l-kg sample of soybeans was
washed and soaked in 3 liters of distilled water at 22°C for 16-18 hours. Distilled water
was chosen to help eliminate contaminants (such as chlorine and ﬂuorine ﬁom tap water),

and to control the mineral content of the water and other factors that could affect soymilk

43

Dry Soybeans

 

 

 

(1 kg)
Wash and Soak 4 Distilled Water
(16-18 hrs) (3 liter, 22°C)
Drain } Water
(5 mins)
Grind 4 Distilled Water
(3mins) (22°C, water:beans = 6.5:1, liters/kg)
Cook Slurry

(110°C, 15 psi., hold for 3 mins)

1

Extract Soymilk , Okara
(30 foot-pound)

l

Soymilk

Figure 4.2. Developed Procedure for Making Soymilk.

quality. The soaked beans were drained in a colander for 5 minutes. The amount of
water collected from draining was recorded. The drained beans were transferred to the
Prosoya Soycow System, as outlined in the manual. The amount of water added to the
grinder was adjusted to establish a ﬁnal water:bean ratio of 6.5:1 (liters/kg of original dry
bean weight) (Prosoya Inc., 1998). The soaked beans were then ground for 3 minutes.
The slurry was cooked at 110°C until the pressure reached 15 psi; this pressure was held
for 3 minutes after which the slurry was transferred to a press (Prosoya Soycow System)
that was covered with a sterilized ﬁlter bag. The slurry was pressed until 30 foot-lbs of
pressure was reached by a torque bar and the soymilk was collected separate from the
okara.
4.2.2.2 Soymilk analysis

After ﬁltering, 500 ml of fresh soymilk was saved for analysis (275 ml for color
and pH, 200 ml for density, 20 ml for total solids, and 5 ml for protein content). All
results were reported as mean values of nine replicates ﬁom three batches of soymilk.
4.2.2.2.] Temperature

Soymilk temperature was measured directly ﬁom the ﬂow of soymilk coming
from the press. To increase the accuracy, a digital thermometer (Model 15-077-14;
Fisher Scientific Co., Pittsburgh, PA) was used. It had a separate probe and results were
shown within two seconds.
4.2.2.2.2 Total volume of soymilk

Total soymilk volume was reported as the total volume of milk-like product,

extracted from soybeans after ﬁltering.

45

4.2.2.2.3 Color of soymilk

Color of soymilk was determined with a colorimeter (Model D25-PC2A, Hunter
Associates Laboratory, Inc., Reston, VA). An 80-ml soymilk sample at room
temperature was poured into a testing dish and examined. The results were reported in L-
, a—, and b- values, where L=100 indicated white, L=0 was black, negative "a" meant
color was towards green, positive "a" meant color was towards red, negative "b"
indicated color was towards blue, and positive "b" meant the color was towards yellow.
4.2.2.2.4 pH

Soymilk pH was measured ﬁ'om ﬂesh soymilk at room temperature, the same day
as soymilk processing. It was measured with a digital pH meter (Accumet portable
Model AP62, Fisher Scientiﬁc Co., Pittsburgh, PA), standardized with buffer solutions of
4.0, 7.0, and 10.0.
4.2.2.2.5 Density

The density of soymilk was determined for room temperature fresh soymilk
samples on the same day as processing. Soymilk was poured into a pre-weighed 50 ml
volumetric ﬂask to measure a 50-ml soymilk sample. This was then weighed and the
density of the soymilk in g/ml was calculated.
4.2.2.2.6 Total solids content

The total solids content was determined according to AOAC Method 925.23
(AOAC, 1984) with some modiﬁcations. A 5-g ﬁ'esh soymilk sample (the same day of
soymilk processing) was weighed in a pre-weighed aluminum dish, dried in an air oven at
98-100°C for three hours, and cooled in a dessicator. The sample was weighed again and

soymilk total solids was reported as percentage of initial weight.

46

4.2.2.2.7 Protein content

Protein content of soymilk was determined by micro-Kjeldahl Method 46-13
(AACC, 2000) with some modiﬁcations. A l-ml sample of soymilk was pipetted into a
Kjeldahl tube and digested for two hours, followed by distillation and titration, and the

soymilk protein content was calculated using a Nitrogen factor of 6.25.

4.2.3 Phase III: Tofu processing and analysis

Tofu was produced by coagulating ﬁesh soymilk. Alter saving an aliquot (about
500 ml) of soymilk for chemical analyses, the remaining soymilk was used for tofu
production. Each soybean variety was processed ﬁ'om soybeans to soymilk to tofu two
times to evaluate the reproducibility of the process.
4.2.3.1 Tofu processing

A modiﬁed laboratory procedure for making tofu was developed in preliminary
studies (Fig. 4.3). From each batch of soymilk, two bricks of tofu were made (i.e., two
replicates). Tofu was processed with calcium sulfate, following the methods of Shurtleff
and Aoyagi (1990) and Prosoya Inc. (1998), with some modiﬁcations. After receiving
soymilk from the Soycow, one liter of hot soymilk (85-88°C) was transferred to a plastic
container and maintained at 80°C (Schroder and Jackson, 1972) in a water bath for 15
minutes (Fig. 4.3). A calcium sulfate solution (0.03M) was prepared at 65°C and the
heated plastic container was removed ﬁom the water bath. Hot soymilk was stirred ﬁve
times in a clockwise direction and coagulant solution was immediately added and mixed
into the soymilk sample. This mixture was stirred slowly ﬁve times in an anti-clockwise

direction. The container was covered and the soymilk was allowed to coagulate

47

Soymilk
(1 liter)

Boil
(15 mins in water bath, 80°C)

l

Coagulate 4 Coagulant
(30 mins) (CaSO4.2H20, 0.03M, at 65°C)

1

Mold
(break curds, 10 times)

1

Press Curds g Whey
(1 hr, 27 g/cmz)

l

Tofu

 

Figure 4.3. Developed Method for Tofu Processing with Calcium Sulfate.

48

undisturbed for 30 minutes (Liu, 1999) at about 77°C. The curd was then cut into small
pieces by stirring 10 times with a regular large table fork to release whey; the curd pieces
were transferred to a laboratory-designed tofu box (11.43 x 11.43 x 10.16 cm) lined
with cheesecloth for molding (Fig. 4.4). Whey was separated from the curd by pressing
for 1 hour with 27 g/cmz. The tofu was unwrapped from the cheesecloth and the weight
of the ﬂesh tofu brick was recorded. The brick of tofu was placed in a labeled container,
covered with tap water, and stored in the reﬁ'igerator at 4°C for texture, moisture, and
protein analyses the next day.
4.2.3.2 Tofu analysis
4.2.3.2.] Tofu yield

Toﬁr yield was expressed by weight in kilograms of fresh tofu produced per
kilogram of dry soybeans (Wang, 1993; Cai and Chang, 1999). The weight in kilograms
of each whole brick of ﬂesh tofu sample was determined; this weight was used to
calculate the tofu yield per kilogram of soybeans as follows:

Wt of tofu (kg) Total Volume of soymilk (liter)

Yield = x
1 liter soymilk 1 kg dry soybeans

 

4.2.3.2.2 Texture

One-day-old tofu brick samples were used for texture analysis using the TA. HDi
Texture Analyzer (Texture Technologies, Scarsdale, NY) and examined with two tests:
Skin test: The tofu brick was placed underneath the TA 52 probe (2 mm diameter); this
test was done three times for each toﬁr brick sample (two tests randomly on two corners,

and one test in the center of the tofu brick). The probe traveled at the speed of 10 mm/sec

49

 

Figure 4.4. Molding and Pressing Tofu.

50

during the pre- and post-test. When the descending probe touched the sample, the probe
speed decreased to 0.5 mm/sec for a distance of 15 mm downward and back up to
constitute the test itself, after which the probe released fully back upwards. The time for
the experiment was set at 60 sec. Force needed to go through the toﬁr skin was reported
as an average of the three readings from each tofu brick in Newtons (N).
Compression test: The tofu brick was placed underneath the TA 4 probe (40 mm in
diameter and 20 mm tall). The test speed was set at 0.5mm/sec. This test was done with
two bite cycles: the probe touched the sample and compressed downwards to 60% of the
brick height (approximately 21 m), then released upwards (again, at 0.5 mm/sec), and
then performed a second compression bite in the same location using the same
parameters. The time for the test was set at 150 seconds and the test was performed
randomly on two locations on each brick. All the peaks recorded were reported as forces
in N and the program calculated the areas under the curves in Ns (N ewton-second) (Fig.
4.5). The ﬁrst peak (if there was one) was considered as the fracturability of tofu; this
reﬂects the force in N to break the curd. The second peak of the curve was expressed as
the hardness of tofu; the force (N) needed to compress the curd. The cohesiveness was
determined by the ratio of Area 2/Area 1 (Fig. 4.5). The gumminess (N) was deﬁned as
the product of hardness and cohesiveness (Bourne et al., 1978).
4.2.3.2.3 Moisture content

After texture analysis, four pieces of one-day-old toﬁr were cut from the four
comers of a tofu brick, with an area of 2 cm x 2 cm for each piece. The skins from those
pieces were discarded, and the interior toﬁr cut and mixed together to have a uniform

curd. Moisture content of the unifome mixed tofu curd was determined following

51

FORCE -’

 

 

4__ FIRST 8le —> 4 SECOND 8le ———-r

 

 

 

 

 

 

 

 

 

4—DOWNSTROKE-——> 4— ursraoxs —-> a—oownsnoxE—> ouvsnoxs >
HARDNESS l
rucrusasum

I nanourss 2
1
l
l
O
I l
' I
' I
' I
' I
' 1
' I

Area 1 : :
' l
' I
' Area 2 .
: +smNcmess-vl 1 L

A,” 3 |+sm~cmess+| L/
I
ADHESIVE fORCE

TIME -9

Figure 4.5. Texture Proﬁle Analysis. (Bomne et al., 1978)

52

AACC Method 44-31 (AACC, 2000) with some modiﬁcations. A 5-gram sample of
mixed tofu curd was weighed in a pre-weighed aluminum dish. It was further out into
small pieces to increase the surface area and dried in an air oven at 130°C for 2 hours.
The sample was cooled in a dessicator and weighed again. Total solids content of each
tofu brick was calculated, and the moisture content was determined and reported as a
percentage.
4.2.3.2.4 Protein content

One-day-old toﬁr samples were used for protein determination following AACC
Method 46-13 (AACC, 2000) with some modiﬁcations. A 100-mg sample of uniformly
mixed tofu curd (see 4.2.3.2.3 for sample preparation) was weighed on a weighing paper
and transferred to a Kjeldahl tube. The sample was digested for 2 hours, followed by
distillation and titration as per the AACC Method. A Nitrogen factor of 6.25 was used to

determine protein content.

4.2.4 Phase IV: Statistical analysis

All data were reported in mean values with standard deviations; the coefﬁcients of
variation were calculated to check reproducibility. The P-values and AN OVA
parameters were calculated using the 5% signiﬁcant difference level with the SAS
program, version 8.01 (SAS Institute, Cary, NC). The Quattro Pro version 9 (Corel
Corporation, Ottawa, Ontario, Canada, 1999) was used to assist with data computation
for intercorrelation and correlation relationships between parameters of soybeans,
soymilk, and tofu. The results were used to discriminate soybean varieties for soymilk

and toﬁr production potentials.

53

V RESULTS AND DISCUSSION

The laboratory procedures developed in the preliminary studies were used for
making soymilk and toﬁr, and for analyzing parameters of soybeans, soymilk, and tofu.
These methods gave good results and discriminated among seven soybean varieties from
two locations. Proxirnate analysis data for all of the soybean (physical and chemical
analyses), soymilk (temperature, volume, total solids, protein content, and color) and tofu
(yield, moisture content, protein content, and texture) samples among seven varieties of
soybeans from both locations (Allegan and Sanilac counties) are reported in the following

sections.

5.1 Soybeans
5.1.] Physical analysis

Mean values and standard deviations of seed weight and percent of split seeds are
listed in Table 5.1. The coefﬁcients of variation of these parameters were less than 5%.
Analysis of variance among seed weight data is presented in Table 5.2. Signiﬁcant
differences for seed weights were found among varieties from the same location as well as
within the same variety grown in different locations (p<0.05). In each location, variety IA
2020 showed the highest weight (21 .63g/ 100 seeds in Sanilac Co. and 18.77g/ 100 seeds in
Allegan Co.). Overall, Allegan county soybean seeds were smaller in size than the Sanilac
county seeds (except for IA 2030 and Novartis 82020). In the present study, soybean

seed weight ranged from 14.69-21.63 g/ 100 seeds, which was relatively low compared to

54

Table 5.1 Mean Values of Soybean Physical Analysis Data Among Seven Soybean
Varieties from Two Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variety Seed Weight Split Seeds
(g[100 seedsL (%)
Sanilac Co. Allegan Co. Sanilac Co. Allegan Co.
IA 2034 17.891053 18.291024 0.00 0.00
Steyer 16.771009 15.061031 0.33 0.00
HP 204 19.531050 17.001042 0.00 0.00
Vinton 81 19.221025 16.941027 0.33 0.00
IA 2020 21.631005 18.771016 0.00 0.00
Novartis 16.561026 14.691005 0.00 0.00
820F8
Novartis 14.951038 15.901038 0.33 0.00
S2020

 

55

 

Table 5.2. Analysis of Variance for Soybean Seed Weight, Moisture Content, and
Protein Content‘ Among Seven Soybean Varieties from Two Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of Degree of Seed Weight Moisture Protein
Variation Freedom
Main Effects F-value
County 1 189.61” 14281.8” 40.76"
Variety 6 182.49b 242.14” 54.91”
County & 6 33.16” 311.13” 5.01b
Variety
' Dry basis.
b Signiﬁcant at p<0.05.

56

 

the results ﬁ'om Wang et al. (1983) of 15.42-35.51 g/100 seeds, DeMan et al. (1989) of
17.20-23.30 g/100 seeds, and Lim et a1. (1990) of 871-4127 g/ 100 seeds.

Soybean samples examined of all varieties grown in Allegan County contained no
split seeds. But among those grown in Sanilac Co., soybean samples ﬁom Steyer, Vinton
81, and Novartis 82020 varieties had means of about 0.33% split seeds (Table 5.1).
Soybeans of the seven varieties ﬁom both counties demonstrated no mixed color of seed
coats: they all had clear, light yellow seed coats. Based on US. Standard Grades (Table
2.5), all seven varieties from both locations are Grade 1 since all seed samples contained
no foreign materials, had less than 10% split seeds, and less than 1% soybeans of another
seed coat color.

All dry soybean seeds were spherical in shape (Fig. 2.1 - picture 2). After soaking,
they became ﬂat in shape (Figure 2.1 — picture 30), and increased to almost three times
their original size.

5.1.2 Chemical analysis

Soybean moisture content and protein content were reported in mean values and
standard deviations in Table 5.3. The coeﬂicients of variation of both parameters were
less than 5%. In Sanilac County, all seven soybean varieties were shown to have a lower
moisture content (range from 8.32-8.90%) than those of Allegan County (from 9.50-
11.15%) which also had a wider range. Of the soybean varieties studied, Novartis SZOF 8
had the highest moisture content in Sanilac County and IA 2020 had the highest moisture
content in Allegan County (Table 5.3). The mean values for moisture content, ranging
from 8.32-11.15% in this study, were somewhat higher than those reported by Lim et al.

(1990) of 6.62-1 1.02%.

57

Table 5.3. Mean Values of Soybean Chemical Analysis Data Among Seven Soybean
Varieties from Two Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variety Moisture' Protein Content'"c
1%) (%)

Sanilac Co. Allegan Co. Sanilac Co. Allegan Co.

IA 2034 8.321008 9.601002 423411.12 44.231020

Steyer 8.361007 1 1.061008 42.191062 42.541014

HP 204 8.361003 10.931003 44.0411 . 12 43.571053

Vinton 81 8.351006 10.731004 43.501028 44.181049

IA 2020 8.821004 11.151004 42.621043 43.671068

Novartis S20F8 8.901003 9.501008 38.251091 41.041018

Novartis S2020 8.631000 105010.07 38.401033 41.041022
'n=3
bn=6.

cDry basis.

58

 

On a dry basis, varieties Novartis S20F8 and Novartis S2020 Ind the lowest
protein contents at each of the locations (Table 5.3). Variety HP 204 ﬁ'om Sanilac County
and variety IA 2034 item Allegan County had the highest protein contents in their
respective location: 44.04% and 44.23%, respectively.

Signiﬁcant differences were detected in soybean moisture content (p<0.05) and
protein content (p<0.05) among all soybean varieties and between locations, they are
listed in the ANOVA table of data (Table 5.2). Depending on conditions of the
environment in which soybeans were grown (e.g., climate, soil, etc.), soybean varieties
ﬁom different locations might have different moisture contents and protein contents. In
addition to the differences in soybean moisture and protein contents, differences in the
ratio of 118/7S proteins present in soybean varieties genetical differences may also affect
soymilk and toﬁr qualities.

5.1.3 Relationships among quality parameters and quality evaluation

The correlation coeﬂicients among physical and chemical characteristics of
soybeans are presented in Table 5.4. There was no correlation between soybean moisture
content and protein content. A positive correlation between seed weight and protein
content (dry basis) was found in both locations. This indicated that the larger the soybean
seed, the higher its protein content. This ﬁnding is in general agreement with Smith and
Circle (1978). On the other hand, DeMan et al. (1989) and Lim et a1. (1990) both
reported that there was no correlation between seed weight and soybean protein content,
and did not observe any correlation between soybean moisture content and protein

content.

59

Table 5.4. Correlation Coefficients for Relationships between Soybean Quality

 

 

 

Parameters
Parameter Sanilac Co. Allegan Co.
Seed weight vs. Protein content’ 0.750b 08105
Seed moisture vs. Protein content“I -0112 0.700

 

 

 

 

 

‘ Dry basis.
b Signiﬁcant at p<0.05.

60

In general, soybeans ﬁom Sanilac County had higher seed weight and lower
moisture content than those from Allegan County. Among all varieties from both
locations, varieties Novartis S20F8 and Novartis $2020 from Sanilac County had smaller
seed size and the lowest protein contents (16.56g/ 100 seeds and 14.95 g/ 100 seeds for
seed weight, and 38.25% and 38.40% for protein content on a dry basis, respectively).
Soybeans from Sanilac County also were shown to have more split seeds (up to 0.33% for
varieties Steyer, Vinton 81, and Novartis S2020) than those ﬁ'om Allegan County (0%),
which would affect soymilk quality because of the off ﬂavor produced during the soaking

and grinding procedures (Fig. 5.1).

5.2 Soymilk
5.2.1 Analyses of soymilk temperature, volume, protein content and total solids
Soymilk temperature upon exiting the soycow machine was in the range of 86-
90°C for all varieties. Soynrilk temperature was measured to ensure its temperature was
above 80°C, because this is the minimum temperature to inactivate trypsin inhibitors,
denature soybean proteins, kill microbes for extending product shelf-life, increase the
extraction of soymilk from the slurry, and prevent its beany ﬂavor (Liu, 1999). Mean
values and standard deviations for soymilk volume and protein content are shown in Figs.
5.1 and 5.2, respectively. The coeﬂicients of variation for soymilk volume and protein
content were less than 5%. Among seven varieties from both locations, variety Steyer
grown in Allegan County gave the highest soymilk volume of 6.63 liters per kg
soybeans. On the other hand, variety IA 2020 grown in the same Allegan County

produced the lowest amount of soymilk: 6.27 liters per kg soybeans (Fig. 5.1).

61

 

 

8.00

7.00 -

6.00 .

5.00 -

4.00 -

3.00 -

Soymilk Volume (Ilters)

2.00 <

1.00‘

 

 

 

0.00‘ IA2034 Steya' HP204 leon81 M2020 Novatis Novatis

saws 32020
Variety

 

 

 

Ll Sanilac Co. IAllegan CO.

 

 

Figure 5.1. Soymilk Volumes of Seven Soybean Varieties from Two Locations.

62

 

 

   
  

Protein (%)
w &

N

 

 

‘ |A2034 Steyer HP204 Vinton 81 16.2020 Novart's Novartis
_ szora 32020
Vanety

lSanilac Co. lAllegan 00. l

 

 

 

 

Figure 5.2. Soymilk Protein Content of Seven Soybean Varieties from Two
Locations.

63

Signiﬁcant differences were detected among soymilk volumes and protein contents
(p<0.05) of varieties grown in the same county and within the same variety grown in both
locations (Table 5.5). Although the same water:bean ratio of 6.5:1 (v/w) was used during
soymilk processing, soybeans had different oil contents, and protein contents as well as
containing different types of proteins, and absorbing water at a different rate during
soaking. In addition, depending on the amount of soluble proteins in soybeans and their
water retention capacity, the amount of soymilk released and remaining in okara differed
among varieties, and thus resulted in different soymilk yields. Because of volume
differences among soymilk yields and differences in soybean protein contents among
varieties, the concentration of soymilk proteins was also different, ranging ﬁ'om 3.92-
4.51% in Sanilac County varieties and from 3.84—4.57% in Allegan County varieties.
Among all soybean varieties, variety IA 2034 grown in Allegan County produced soymilk
with the highest protein content (Fig. 5.2). Soymilk protein contents in this study were
higher (from 3.66-4.57%) than those previously reported by Chen (1989) of 3.5-4.0%.

Soymilk total solids (ranging from 8.18-8.66% in Sanilac Co. and 8.11-9.02% in
Allegan Co.) were similar for all varieties in both locations (Table 5.5). Soymilk total
solids is related to the water:bean ratio (Liu, 1999); therefore, in the present study, the
same water:bean ratio of 6.5:1 (v/w) was used among varieties during soaking, grinding
and cooking, and resulted in similar soymilk total solids in all varieties. Even though all
soybean varieties produced soymilk with similar total solids, soymilk protein contents
were quite different due to the differences in protein and carbohydrate contents among the

soybeans samples, and the differences in extractability of soluble proteins content in

Table 5.5. Analysis of Variance for Soymilk Volume, Total Solids, Protein Content,
and Color Among Seven Soybean Varieties from Two Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of Degrees Volume Total Protein Color
Variation of Solids Content
Freedom
Main F-value
Effects
L a b

County 1 12.341' 0.54 7.23' 4.50‘ 1.95 1.53

Variety 6 1.72 3.62‘l 15.82‘ 0.83 6.86' 6.97‘l
County & 6 5.29’ 2.16 1000' 1.61 2.27 6.55

Variety

‘ Signiﬁcant at p<0.05.

65

soybeans after pressing the slurry (Lim et al., 1990). In the present study, soymilk total
solids were relatively lower (8.11-9.02%) compared to the total solids values of 9.09-
9.90% reported by Lim et al. (1990). This was likely due to the different water:bean ratio
used by Lim et al. in this study (4.7 :1, v/w). A higher level of water used would result in a
higher soymilk vohrrne, but lower total solids, and lower protein content (Table 2.7).
5.2.2 Soymilk color, pH, and density

Among the varieties studied, signiﬁcant differences were detected for the
whiteness of soymilk (assessed by the L-value, black 0, white 100) between locations
(p<0.05) (Table 5.5). For each soybean variety, the a-value (red +/green -) was similar for
both locations (Fig. 5.3), but signiﬁcant differences were detected for a-values among
soybean varieties within a location. In the present study, the means of the b-value (yellow
+/ blue -) were different among varieties in the same county. However, the whiteness of
soymilk (the L-value) is more important than the other two factors that measure the
degree of red, green, yellow or blue color present (the a- and b-values). When comparing
soynrilk from the varieties studied with commercially prepared soymilk, it was noted that
the mean L-values (range 76.27-81.76, Appendix III) for whiteness of soymilk from all
seven varieties at both locations were higher than those of soymilk from the Vitasoy
Company (L = 72.4) and ﬁ'om the Edensoy Company (L = 66.7) (Appendix IV).
Generally, cormnercial soymilk has higher total solids because other ingredients such as
sucrose, vitamins, minerals, and gums are added; these additives also affect soymilk color.
When compared with cow's milk, the soymilk produced in the present study had L-values
just a little less white than 2% fat cow's milk (L = 81.2), and 3.5% fat cow's milk (L =

88.7) (Appendix IV).

66

 

 

90.00

80.00 -

70.00 1

60.00 —

50.00 -

40.00 —

30.00 a

Hunter Color Values

20.00 4

10.00 4

 

 

 

 

 

0.1!) -

 

 

-1 0.00

 

 

|A2034 Steyer l'PZM ““0181 |A2020 Natalie m
820:8 $2020

Variety

I L—vdue for Sanilac Co. I L-vdue for Allegar Co. I a-value for Sanilac Co.
I a-value for AlIegan Co. I b-value for Sanilac Co. I b-vdue for Allegar Co.

 

 

 

 

 

 

 

Figure 5.3. Soymilk Color of Seven Soybean Varieties from Two Locations.

67

Soymilk pH and density among varieties and ﬁom both locations were similar and
ranged ﬁom 6.56-6.66 for pH and ﬁ'om 1.01-1.02 g/ml for density (Table 5.6). The pH
values were relatively higher than those reported by Lim et al. (1990) of 6.42-6.55. In the
present study, the soymilk pH was in an optimum range (pH between 3 and 7) for Ca2+
ions ﬁ'om calcium sulfate coagulant to bind to soybean proteins. Liu (1999) reported that
at soymilk pHs below 3 and above 7, binding between Ca2+ ions and soy proteins did not
occur. Soymilk densities were similar among all the studied soybean varieties. This was
expected since soymilk total solids had also been similar among these varieties.

5.2.3 Relationships among soymilk volume, total solids, protein content, and color

There was a negative relationship between soymilk vohrrne and total solids (Table
5.7); this irrrplied that adding more water during grinding would reduce soymilk total
solids (Table 2.7). Soymilk protein content does not necessarily relate to soymilk yield; it
depends on the types of soybean proteins present, especially soluble proteins. The
presence of more soluble proteins in soybeans resulted in high soymilk volume (Liu,
1999). For soymilk volume and color (L-value for whiteness), the correlation was a
negative relationship for soynrilk made from soybean varieties grown in Sanilac County (r
= -0.945) and Allegan County (r = 0322).

5.2.4 Soymilk quality evaluation

Variety Steyer from Allegan County produced the highest soymilk volume (Fig.
5.1). Overall, soynrilk produced from Sanilac County varieties were greater in volume
than those nrade from Allegan County varieties. The inverse relationship between soynrilk

volume and total solids indicated that variety Steyer in Allegan County produced the

68

Table 5.6. Mean Values of Soymilk pH and Density from Seven Soybean Varieties

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variety pH' Density"
lg/ml)
Sanilac Co. Allegan Co. Sanilac Co. Allegan Co.
IA 2034 6.611002 6.371008 1.011033 1.021011
Steyer 6.661001 6.611005 1.011034 1.021004
HP 204 6.611001 6.641003 1.021022 1.021010
Vinton 81 6.641001 6.551005 1.021019 1.021034
IA 2020 6.661004 6.631002 1.021022 1.021028
Novartis 820F8 6.561003 6.591001 1.011023 1.021015
Novartis 82020 6.561005 6.561001 1.011018 1.021012

 

:11:
Il
>01"

69

 

Table 5.7. Intercorrelation Relationships between Soymilk Volume, and Total

Solids, Protein Content, and Color

 

 

 

 

 

 

 

 

Parameter Sanilac Co. Allegan Co.
Soymilk volume vs. Total solids -0307 -0.688
Soymilk volume vs. Protein 0815‘ 0.187
Soymilk volume vs. Color (L- -0.945‘ -0322
value)
' Signiﬁcant at p<0.05.

70

 

highest soymilk volume (6.63 liters) with the lowest total solids (8.11%). In contrast,
variety IA 2020 ﬁom Allegan County produced the lowest soymilk volume (6.27 liters)
with the highest total solids (9.02%). With its highest soymilk protein content among the
varieties studied, variety IA 2034 from Allegan County (Fig. 5 .2) would be expected to
produce toﬁr with higher yield and protein content in the following experiments of the
present study.

To produce good quality soymilk, the most important factor that the producer
needs to consider is the water:bean ratio (v/w). This will affect the soymilk volume
produced, total solids, protein content, and the whiteness of the soymilk (Chen, 1989).
Besides the main factor of water:bean ratio, a producer needs to select a proper variety of
soybean that can produce high soynrilk yield; choose a location for the processing plant
that has a source of water without contaminants; select equipment for soybean grinding,
soymilk extraction, heat treatment, and packaging; and formulate the end products (Liu,
1999). Soynrilk needs to have a white appearance, be a ﬂuid that is not too thin but not
too thick, and have nutritional value. Depending on the culture and the demand of
customers (for example, Western customers prefer soymilk without a beany ﬂavor, but
Eastern people favor it), the producer needs to choose the proper method of processing
(hot or cold grinding, or unsoaked beans method) to produce high quality soymilk that

suits consumer tastes.

71

53 T01

5.3.1 1

coellic
range
reports
l‘arianc.
exhibits
both 10(
overall 2
mainly d
depends
Walerzbt

5.3.1 Cl

5.3 Tofu
5.3.1 Tofu yield

Mean values and standard deviations of toﬁr yield are presented in Table 5.8. The
coefﬁcients of variation of the tofu yield were less than 5%. In the present study, the
range of 2.67-3.43 kg tofu produced per kg soybean was relatively lower than that
reported by Lim et al. (1990) of 4.45-5.26 kg tofu/kg soybean Results from analyses of
variance on this data are summarized in Table 5.9. Signiﬁcant differences in yield were
exhibited (p<0.05) among varieties within each location and within the same variety from
both locations. Variety IA 2020 grown in Sanilac County gave the highest tofu yields
overall and variety Novartis 820F8 from both counties gave the lowest yield. Toﬁr yield
mainly depended on the amount of water retained in the curd, and the retention probably
depends on factors such as soymilk total solids, the amount of soluble proteins, the
water:bean ratio, etc. (Beddows and Wong, 1987).
5.3.2 Chemical analysis

Toﬁr moisture and protein data are reported in Table 5.8 along with their mean
values and standard deviations. The coefﬁcients of variation were less than 5%. For
protein content, on a dry basis, there were signiﬁcant diﬁeremes in tofu protein content
(p<0.05) among all varieties grown in both locations (Table 5.9). Schaefer and Love
(1992) reported that based on variation in soybean protein content, tofu protein content
might be expected to be different among varieties. Their ﬁndings were in agreement with
the present study. The mean values for protein content in the present study ranged ﬁom

48.23-58.68%, and were similar to those reported by Lim et al. (1990) of 46.03-52.50%.

72

3uce

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

éuee
asp be .
3.32.3 3.33.3 3.33.3 3.3g ﬂow 8.333 2.333 :83 awe—gel
3.33.3 3.33.3 3.39.3 3.33.? 8.333 8.333 3.33 var—gel
«0.33.3 3.32.3 3.33.3 3.33.3 3.333 3.333 33 5
2.33.3 3.33.3 3.33.3 2.33.3 S.33.~ 30.333 3 53:;
3.33.3 3.33.3 3.33.3 3.33.3 3.323 3.333 van mm
3.32.3 3.33.3 3.33.3 3.35.3 3.3~ mm 2.333 .3.»on
3.33.3 3.33.3 3.33.3 3.3393 2.3“ mm 3.33.». 3.3 5
do :32? .00 83m do 533 60 83m .00 ammo—Z .00 83:3
3.1 ex; :32. gas a:
69.53:: 65895 ‘20; bota>

 

 

23:39— 959 SP... 833...; 52.33 :33

no 2.3.30 9.53:2 as £32.30 £82.— .29; such. he 82.5 532 .36 035.

73

Table 5.9. Analysis of Variance for Tofu Yield and Protein Content‘ Among Seven
Soybean Varieties from Two Locations

 

 

 

 

 

 

 

 

 

 

 

Source of Variation Degrees Yield Protein
of
Freedom
Main Effects F -value
County 1 9.12” 20.56"
Variety 6 30.22b 29.35F
County & Variety 6 8.89“ 3.76“
' Dry basis.
b Signiﬁcant at p<0.05.

74

 

Tofu moisture contents obtained in the present study (84.82-86.47%) were similar
to those observed by Lim et al. (1990) (86.58-88-58%). Water retained in the tofu was
similar among all varieties and resulted in tofu samples with almost the same moisture
contents. There is no standard moisture content for commercial tofu. Hard toﬁi contains
75.00-79.00% moisture and soft toﬁ1 ranges ﬁom 84.70-85.90% (Gebre and Summer,
1983), or 84.07% moisture for hard tofu and 87.08% for soft tofu (Tsai et al., 1981).
Therefore, based on moisture content, tofu from the present study could be classiﬁed as
soft tofu. Variety Novartis S20F8 produced tofu with the lowest moisture and protein
contents (dry basis). Variety IA 2034 from both locations gave tofu with the highest
protein contents (Table 5.8). This was due to the fact that IA 2034 produced soymilk
with a high protein content, ﬁom which the tofu was made.

5.3.3 Tofu texture analysis

Mean values and standard deviations of tofu texture are presented in Fig. 5.4. The
texture parameters included fracturability, hardness, and gumminess, for which values
ranged ﬁ‘om 6.46-11.19 N, 7.84-11.67 N, and 3.01-5.17 N, respectively. The
cohesiveness values ranged ﬁom 0.38 to 0.50. Results of the four parameters of texture
of fresh tofu samples from the seven varieties in the same location were signiﬁcantly
diﬂ‘erent (p<0.05) (Table 5.10). Similar ﬁndings were reported by Wang (1993).
However, within the same variety, results from the texture evaluations were similar
(p>0.05) between locations; except for cohesiveness, tofu samples ﬁ'om the seven varieties

were similar in both locations.

75

 

Fractu rability

lllll

M2034 Steyer HP204 Vinton 81 |A2020 Novartb Novartb
SZOF8 S2020

Force (N)

 

Varieties

ISanlac Co. .Allegan Co. I

 

 

Hardness

Force (N)

 

IA 2034 Steyer HP 204 Vinton 81 IA 2020 Novartis Novartis
SZOFB $2020
Va rleties

ISenilnc Co. lAllegen Co.

 

 

Gumminess

11min

IA2034 Steyer HP 204 Vinton a1 IA2020 Novartis
_ _ $20F8 stozo
_____ l
Esgriwugc»Co. IAllegan Co.

Figure 5.4. Tofu Texture Analysis Among Seven Soybean Varieties from Two
Locations.

Force (N)

 

 

 

 

76

Table 5.10. Analysis of Variance for Tofu Texture Among Seven Soybean Varieties
from Two Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of Degrees Fracturability Hardness Cohesivenese Gumminess
Variation of
Freedom
Main F-Value

Eﬁ‘ects

County 1 29.261‘ 20. 14' 0.35 14.433

Variety 6 8.041' 4.58' 6.75‘I 6.50“
County & 6 2.26 0.77 0.57 0.45

Variety

' Signiﬁcant at p<0.05.

77

 

In the present study, the coefﬁcients of variation were less than 20% for all texture
parameters. These results were similar to those reported by Wang (1993), but less than
the coefﬁcients of variation of 33% reported by Schaefer and Love (1992) for toﬁi
hardness. Variety Vinton 81 from Allegan County produced the hardest toﬁl. Toﬁi made
from variety Novartis SZOF8 from both locations had high values of hardness,
cohesiveness and gumminess (Fig. 5.4). As an example, Fig. 5.5 shows a computer
readout and the calculation of fracturability, peaks of hardness, and areas 1 and 2 for the
tofu texture analysis of Vinton 81 grown in Allegan County.

Texture is an important factor that affects consumer acceptance of tofu and tofu
products. This property plays a main role in the selection of type of coagulant and
methods for making tOfll. Improvements in technology, especially the development of the
Texture Proﬁle Analysis and computerized instruments (e. g., Texture Analyzer TA-DHi)
have been found very helpful in toﬁi quality evaluation and assurance.

5.3.4 Relationships among tofu yield, moisture content, protein content, and texture

No correlation was obtained between toﬁJ yield and protein content, and tofu yield
and moisture content (Table 5.11) at p>0.05, but a negative correlation was found
between toﬁr yield and cohesiveness (p<0.05), and toﬁi yield and gumminess (signiﬁcantly
for Sanilac Co. at p<0.05, but not for Allegan Co.). A negative correlation was also found
between tofu moisture and hardness for Sanilac county. This indicates that high water
retention in toﬁi is associated with an increase in product moisture content, and with a
product that is softer. On the other hand, increasing toﬁi yield is associated with less

cohesiveness and less gumminess of the ﬁnal product.

78

Force [N]
14.00.

12.11}
1003‘
8.1.114
6H}
41]}
2.00

one
0 50.0 11110 150.0
Time [sec]

 

-mﬂ

Force (II) 2 3 4 5 B
14.0}

 
 
 
 
 
 
 
 
 
 
 

121]}

101]]

8.0}

6.0]

4110

 

 

 

 

Figure 5.5. Tofu Texture Curve from Variety Vinton 81 Grown in Allegan County.

79

Table 5.11. Intercorrelation Relationships Among Tofu Yield, Moisture Content,
Protein Content', Hardness, Cohesiveness, and Gumminess

 

 

 

 

 

 

 

 

 

 

Parameter Sanilac Co. Allegan Co.
Tofu yield vs. Proteina 0.618 0.613
Tofu yield vs. Moisture 0.684 0.532
Tofu yield vs. Hardness —0.483 -0.41 8
Tofu yield vs. Cohesiveness -0.93 lb -0.7451’
Tofu yield vs. Gumminess -0.73 8" -0.61 1
Toﬁl hardness vs. Moisture -0.875b -o.501

 

' Dry basis.
b Signiﬁcant at p<0.05.

80

 

5.3.5 Tofu quality evaluation

Among all seven soybean varieties from both locations, variety Novartis S20F8
produced toﬁl with the lowest yield, moisture content, and protein content, but which was
high in all the texture parameters studied (hardness, cohesiveness, and gumminess). Based
on results from the present study, soybeans from variety Novartis SZOF8 would be a good
choice for a producer wishing to make hard tofu, or from variety Novartis 82020
harvested in Sanilac County to produce soﬁ tOfll. In general, varieties IA 2034, Steyer,
and HP 204 gave high tofu yield, protein content, and medium hardness. These properties
are considered good for the production of regular tofu (section 2.3.1). Based on
ﬁ'acturability results, variety IA 2034 ﬁ'om Allegan County produced tofu with the highest
regained force to break the curd. The results from the tofu skin test (Table 5.12), with
coefﬁcients of variation less than 20%, conﬁrmed that tofu produced from variety IA 2034
of Allegan County (the hardest to fracture) also had the hardest skin. These tofu
characteristics exhibited by IA 2034 of Allegan County are suitable for the production of

certain tofu products (e. g., ﬁied tofu, toﬁl pouches, etc.).

5.4 Correlation relationships among quality parameters of soybeans, soymilk, and
tofu

Positive correlation relationships were detected between soybean protein content
and soymilk total solids among varieties from Sanilac county (p<0.05) (Table 5.13).
Positive relationships were also found between soybean protein and soymilk protein
contents (p<0.05), and between soymilk protein and toﬁr protein contents, especially for

varieties ﬁ'om Allegan County. Therefore, soybean protein content and soymilk protein

81

Table 5.12. Mean Values for Tofu Skin Tests (N) of Tofu Samples from Seven
Soybean Varieties Grown in Two Counties

 

 

 

 

 

 

 

 

 

 

 

Variety Sanilac Co. Allegan Co.
[A 2034 0.19i0.03 0.29i0.04
Steyer 0.16:0.02 0.19:1:0.01
HP 204 0.20i0.04 0.24:0.03
Vinton 81 0.27i0.04 0.22:0.04
IA‘ 2020 0.191003 0.211002
Novartis SZOF8 0.19:0.03 0.19:1:0.03
Novartis 82020 0.15:0.02 0.19i0.04

 

82

 

Table 5.13. Correlation Relationships Among Evaluative Parameters of Soybeans,

Soymilk, and Tofu from Seven Soybean Varieties Grown in Two Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sanilac Co. Allegan Co.
Soybean and Soymilk Correlation

Soybean protein‘ vs. Soymilk total solids 0.847b 0.426
Soybean proteina vs. Soymilk protein 0.852b 0.777b

Soymilk and Tofu Correlation
Soymilk total solids vs. Toﬁt yield 0.541 -0.094
Soymilk total solids vs. Tofu protein“ 0.185 0.217
Soymilk total solids vs. Tofu hardness 0.213 0.444
Soymilk volume vs. Tofu protein‘I 0.768F -0.403
Soymilk protein vs. Tofu yield 0.618 0.613
Soymilk protein vs. Tofu proteina 0.349 0.8035

Soybean and Tofu Correlation
Soybean proteinll vs. Toﬁr protein' 0.634 0.947b
Soybean proteinII vs. Tofu yield 0.688 0.505
Soybean protein' vs. Tofu hardness 0.105 0.317
Soybean seed weight vs. Tofu yield 0.693 0.494
Soybean seed weight vs. Tofu hardness 0.183 0.105

 

‘ Content on a dry basis.
b Signiﬁcant at p<0.05.

83

 

content played important roles in producing tofu with high protein content. These
ﬁndings are similar to those reported by Wang et al. (1983), DeMan et al. (1989), Lim et
al. (1990), and Schaefer and Love (1992). In the present study, the correlation between
soybean protein and soymik protein (r = 0.852 for Sanilac County and 0.777 for Allegan
County) was lower than that reported by Schaefer and Love (1992) of r = 0.96, but higher
than the result reported by Liu (1999) of r = 0.593. No relationship was found between
soymilk total solids and tofu yield (p>0.05). This ﬁnding was similar to that reported by
Johnson and Wilson (1984). This lack of relationship is likely due mainly to the
differences in soybean protein content among varieties, and the different types of protein
as well as differences in the protein ﬁactions present in soybean varieties, causing proteins
to coagulate in different ways during 1011.! making. There was no correlation between
soybean protein and tofu yield (p>0.05), in contrast to the report ﬁ'om Wang et al. (1983),
but similar to the results of Lim et al. (1990) and Wang (1993).

No relationship between soybean seed weight and toﬁi yield (p>0.05) was found in
the present study (Table 5.13). This indicated that larger soybean seed weight would not
be expected to give higher tofu yield. Thus, in comparison to soybean protein
characteristics, soybean seed weight probably is not a key factor for selecting soybean

variety to produce tofu (Wang, 1993).

84

VI. SUMMARY AND CONCLUSIONS

In the present study, the laboratory procedures developed in the preliminary
studies for making soymilk and tofu were used to discriminate among seven varieties of
soybeans ﬁ'om two locations. All replicated data were similar with acceptable coefﬁcients
of variation. This indicates that the developed procedures could be used, in a reproducible
way, to evaluate soybean varieties for soymilk- and tofu-making potentials.

The two main factors inﬂuencing the differences in soybean protein content among
and within varieties are: (1) the location where the seeds were grown and affected by soil
and other environmental conditions, and (2) the genetics of each of the varieties planted
(Smith and Circle, 1978). To produce good soymilk and a high quality toﬁt products,
Smith and Circle (1978) suggested selecting larger soybean seeds with higher protein
content, lower oil content, and those that had reached maturity.

For soymilk processing in the current study, soybean varieties ﬁ‘om Sanilac County
gave the highest soymilk volume--especially variety Steyer. Soybean varieties that yield
high soymilk volumes can be more proﬁtable to producers. For making tofu, soybeans
from variety IA 2034 ﬁom Allegan County had a high protein content and produced tofu
with a high protein content. Greater force was required to break the curd, and tofu ﬁ'om
this variety had a harder skin, hence, variety IA 2034 grown in Allegan County would be
preferred for fried tofu products. Varieties Novartis 820F8 grown in both counties had
low seed weights, low protein contents, and produced hard curd which might be selected
to make hard tofu. With medium tofu hardness and high yield, soybeans from IA 2034,

Steyer, and HP 204 grown in both counties might be selected by manufacturers to produce

85

regular tofu, another marketable form of tofu. And variety Novartis S2020 grown in

Sanilac County would be a good choice for making soﬁ tofu.

86

VII. SUGGESTIONS FOR FURTHER RESEARCH

In the present study, all seven varieties of soybeans grown in two locations from

one crop year (2000) showed signiﬁcant differences in soybean physical and chemical

properties. To further improve and understand soymilk and tom qualities, the following

suggested research could be conducted:

More crop years (e.g., two or three) are needed with the same varieties grown in the
same locations, to compare all properties among the varieties and evaluate for
possible differences among harvest years.

The same harvest year comparisons could be applied with the making of soymilk and
tofu with the new crops of soybean varieties.

The ratio of 1 IS/7S proteins in the soybean varieties studied needs to be analyzed and
manipulated to achieve high tofu yield and desirable toﬁl texture.

All soymilk samples among varieties studied need to be analyzed for their beany ﬂavor
via sensory analyses.

The inﬂuence of processing conditions, including the technique of adding coagulant

and the stirring rate of all soymilk during coagulation, should be investigated further.

87

APPENDICES

88

APPENDIX I

Comparison between Hot and Cold Grinding Procedures for Making Soymilk Using

 

 

 

 

 

 

 

1999 Vinton Soybeans
Method Cold Grinding Hot Grinding
Soybean l 1
('58)
Soaking Time 15.5 15.5
(hm)
Waer:Bean Ratio 6.8:] 6.821
Soymilk Volume 6.12 6.00
(D
Soymilk pH 6.7 6.7
Soymilk Total Solids 8.5 5.8
(72

 

 

 

 

Comparison Between Hot and Cold Grinding Procedures for Making Tofu Using

 

 

 

 

 

 

 

 

 

 

 

 

1999 Vinton Soybeans
Method Cold Grinding Hot Grinding
Soybean 2 2
(k8)
Soaking Time 15.5 15.5
(hm)
Water:Bean Ratio 4.6:] 4.6:]
Soymilk Temperature 65 55
(°C)
Soymilk used 3 3
(I)
Coagulant' 0.03 0.03
Concentration
(M)
Average Tofu Weight 122.20 very few and small curds;
(g tofu/liter soymilk) could not press for tofu
' Calcium sulfate.

89

 

 

APPENDIX II

Data for Different Soybean Grinding Times to Produce Soymilk

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Using 1999 Vinton Soybeans
Grinding Time 3 5
(min)
Dry Soybean 1 1
(5g)
Water:Bean Ratio 68:1 6.8:]
Soymilk Volume 7 7.4
(l)
Soymilk Temperature 79.2 78.0
(°C)
Soymilk Total Solids 9.5 9.5
(%)
Soymilk Color
L 80.5 81.7
a -3.0 -2.8
b 15.5 15.6
Coagulant‘ Concentration 0.03 0.03
(M)
Coagulant' Temperature 65 65
(°C)
Soymilk Volume to Make 1 l
Tofu (l)
Soymilk Temperature 80 80
(°C)
Curd Forming Time 5 _ 5
(min)
Pressing Weight 3.4 3.4
(kg)
Pressing Time 1 1
(hr)
Average Tofu Weight 367.1 443.6
(Ltofu/liter soymilk)

 

 

 

 

' Calcium sulfate.

90

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.ou:
.cnce
.mue.
8.3 8.3 3.3 8.3 8.3 2.8 2.3 2.3 8.3 8.3 8.3 8.3 :88
8.2 8.2 5%- 8.2 8.8 8.8 2.2. 88 8.» 2.8 8.8 28 8.82
8.3 8.3 8.3 8.3 8.3 8.3 3.3 2.3 8.3 2.3 2.3 8.3 888
8.2 8.2 23. 88- 8.8 8.8 88 8.». 2..» 8.» 8.8 8.8 5:32
8.3 2.3 2.3 3.3 2.3 8.3 8.3 2.3 2.3 3.3 2.3 8.3
:2 3.3 88- 88- 3.8 8.8 2.... 83 8.8 8.8 8.8 28 88 S
8.3 2.3 8.3 2.3 8.3 8.3 2.3 2.3 8.3 2. 3 2.3 2.3 3
2.2 3.2 :8- :2- 8.; 8.8 8+ 3.4 8.x 8.8 8.8 :8 55>
8.3 8. 3 8.3 8.3 8. 3 8.3 2.3 8.3 8.3 8. 3 2.3 8.3
8.2 8.2 88. 8.4- 8.8 8.8 ems 3.4 8.x 8.” 98 28 ..8 .E
8.3 8.3 8.3 8.3 8.3 8.3 8.3 2.3 8.3 8.3 8.3 8.3
8.2 8.2 2a- 2a- 8.8 8.; 8.». 8.4 :8 8.» 8.8 8.8 .388
8.3 8.3 8.3 2.3 8.3 8.3 8.3 :.3 3.3 8.3 8.3 8.3
8. z 8.: 8. N- 8. a- 8. 8 R. 8 8.4 8.4 :8 8.» 8.8 8.8 3.85
.8 .8 .8 do .8 .8 .8 .8 .8 .8 .8 .8
:80? 888 e822 23% 583 325m Ewe? 085m 582? 825m 883 085m
8 e a
3: ex; e35
.330 9389:— enE—em .33. $83.5 but!»
233531—

9.; SP: 825 833.35 5.2—mew 55m .3 3:5 23 .3830 £82.— .nE—ew .33. .9823» van—mom he 82.3 :32

E 592934

91

APPENDIX IV

Mean Values of Color Analysis from Commercial Soymilk and Cow's Milk

 

 

 

 

 

 

Sample Cow’s Milk Commercial Soymilk
Low F at (2%) Whole Milk Vitasoy Edensoy

L 81.2 88.7 72.4 66.7

a -0.9 -4.2 -2.7 0.2

b 3.7 7.7 5.8 19.2

 

 

 

 

 

92

 

APPENDIX V

Tofu-Making‘ Formulae Used for Different Soymilk Volumes, and Resultant Tofu
Yield Using 1999 Vinton Soybeans

 

 

 

 

 

 

 

 

 

 

 

 

Soymilk Soybean Water: Soymilk Coagulant" Coagulantb Average

Volume used Bean Temperature Concentration Temperature Tofu
(0 (kg) Ratio (°C) (M) (M) Weight

(g)
2 2 6.821 75 0.38 65 421.20
1.5 2 6.8:] 73 0.30 65 318.65
1 2 6.8: l 73 0.38 65 211.65
' Using cold grinding method.

" Citric acid.

93

 

APPENDIX VI

Select Tofu Parameters for Tofu Made with Different Coagulant (Citric Acid)
Concentrations Using 1999 Vinton Soybeans

 

 

 

 

 

Concentration Average Tofu Moisture Texture
(%) Weight (%) (N)
(g tofu/liter soymilk)
0.18 209.53 79.10 4.25
0.19 178.83 77.75 12.17
0.38 187.23 84.03 8.27

 

 

 

 

94

 

APPENDIX VII

Comparison between Food Grade Calcium Sulfate and Pure Chemical Calcium
Sulfate as Coagulants for Making Tofu

 

Variety

 

Formula and Resultant
Data

Mixed Soybeans

Vinton (1999)

 

Weight of Soybean
03L

 

Water:Bean Ratio

6.8:]

6.8:]

 

Soymilk Volume
(0

6.5

 

Soymilk Total Solids
(%)

8.5

10

 

Calcium Sulfate

Food grade

Chemical

Food grade

Chemical

 

Coagulant Concentration
(M)

0.03

0.03

0.03

0.03

 

Coagulant Temperature
(°C)

65

65

65

65

 

Soymilk Volume
(I)

1.5

1.5

1.5

1.5

 

Soymilk Temperature for
Making Tofu (°C)

73.8

74.0

70.0

70.0

 

 

Average Tofu Weight
(g tofu/liter soymilk)

 

373.8

 

404.3

 

439.2

 

487.1

 

95

 

APPENDIX VIII

Effect of Curd-Forming Time on Tofu Yield Using 1999 Vinton Soybeans

 

Curd-Forming Time

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(min)
Formula Used and Measured 15 30 60
Parameters Data
Dry Soybean 1 1 1 1 1
(kg)
Water:Bean Ratio 6.8:] 6.6:1 6.6:1 6.6:1 6.8:1
Soymilk Volume 7 7 7 7 7
(l)
Soymilk Temperature 79.2 86 86 86 79.2
(°C)
Soymilk Total Solids 9.5 6.3 6.3 6.3 9.5
(%)
Coagulant‘ Concentration 0.03 0.03 0.03 0.03 0.03
(M)
Coagulant‘ Temperature 65 65 65 65 65
(°C)
Soymilk Volume to Make Tofu 1 1 1 l 1
(liter)
Soymilk Temperature 80 80 80 80 80
(°C)
Pressing Weight 3.4 3.4 3.4 3.4 3.4
(kg)
Pressing Time 1 1 1 l 1
(hr)
Average Tofu Weight 367.1 313.4 313.4 315.7 409.2
‘ (g tofu/liter soymilk)
Curds were stuck Curds Tofu Tofu had
in cheese cloth, were was higher
very soft, and easy to ﬁrmer yield
separated apart take and had because
out of good soymilk
cheese shape had
cloth higher
total
solids

 

 

' Calcium sulfate.

96

 

APPENDIX IX

Effect of Curd-Forming Time on Tofu Yield Using 1999 Vinton Organic Soybeans

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Curd—Forming Time
(min)
Formula Used and 5 ‘ 15 30
Measured Parameters Data
Dry Soybean 2 1 1
(kg)
Water:Bean Ratio 6.521 6.5:] 6.5:]
Soymilk Volume l3 l3 13
(I)
Soymilk Temperature 88 87 88
1°C)
Soymilk Total Solids 13.5 13.6 14.5
(%)
Coagulant' Concentration 0.03 0.03 0.03
(M)
Coagulant‘ Temperature 65 65 65
(°C)
Soymilk Volume to Make 1 1 1
Tofu (l)
Soymilk Temperature 80 80 80
(°C)
Pressing Weight 3.4 3.4 3.4
£81
Pressing Time 1 1 1
0!")
Mean Tofu Weight 489.65i27.02 5026515406 493.53i41.06
(g tofu/liter soymilk)
Curds were stuck Curds were Tofu was
in cheese cloth, easy to take ﬁrmer and
very soft, and out of cheese good shape
separated apart cloth

 

8 Calcium sulfate

97

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3.333 3.353 33:3 8323 1.3? $333 333.3 3.333 amnmmz
3.323 2.353 3.333 2.333 3.332: 3.333 6.323 8.332 2mm?
3.33.3 3.383 3.333 3.33.3 3.33.3 3.333 3.333 2.333 335
2.333 2.332 3.393 8.393 3.3%.: 3.32.3 3.33.2 3.33.2 s 5...;
2.333 3383 3.333 8.333 3.332 2.33.3 2.32.2 3.333 .3 .E
3333 3.332 33:3 33:3 3323 9.353 3.3.3.3 3.333 3%
3.383 3.333 3.333 33:3 33.33:: 3.323 3.32.: 3.333 .33..
.00 .00 .00 .00 .00 .00 .00
532 33am 5&2? 8&3 $322 8.53 5322 .8 833
3658350 30:038....0 38:65am @Enwwaoﬁm but!»

 

 

2:583— ?th 5 83¢ Buota> cashew 55m no £3.22 9:555 33. .3 82.5 :32

X Nun—Zak:

98

BIBLIOGRAPHY

AACC, 2000. Approved methods, Tenth ed. American Association of Cereal Chemists,
St. Paul, MN.

Alekel, D.L. and Germain, AS, 2000. Isoﬂavone-rich soy protein isolate attenuates
bone loss in the lumbar spine of perirnenopausal women. Am. J. Clin. Nutr.
72(3): 844-852.

Aug, H.G. and Kwik, W.L., 1985. Development of soymilk - a review. Food Chem.
17:235-250.

AOAC, 1990. Ofﬁcial methods of analysis. Fifteenth ed. Association of omcial
analytical chemists Inc. Arlington, VA.

APV, 1998. Soymilk process and application, Technology update.

Arditi, T., Meredith, T., and Flowerman, P., 2000. Renewed interest in soy isoﬂavones
and saponins. Cereal Food World. 45(9):414—417.

Beddows, CG. and Wong, J., 1987. Optimization of yield and properties of silken tofu
ﬁom soybeans. I. The water:bean ratio. Intern. J. Food Sci. Technol. 22:15-21.

Bourne, M.C., Kenny, J.F., and Barnard, J., 1978. Computer-assisted readout of data
ﬁom texture proﬁle analysis curves. J. Texture Studies. 9:481-494.

Cai, TD. and Chang, KC, 1998. Characteristics of production-scale toﬁr as affected by
soymilk coagulation method: propeller blade size, mixing time and coagulant
concentration. Food Research International. 31(4):289-295.

Cai, T. and Chang, KC, 1999. Processing effect on soybean storage proteins and their

relationship with toﬁi quality. J. Agric. Food Chem. 47:720-727.

99

Chen, S., 1989. Preparation of ﬂuid soymilk, Pp. 351, in: Food uses of whole oil and
protein seeds. Lucas, E.W., Erickson, DR, and Nip, W.B. American Oil
Chemistry Society: Campaign, IL.

Damodaran, S. and Kinsella, J.E., 1982. Eﬁ‘ect of conglycinin on the thermal aggregation
of glycinin. J. Agric. Food Chem. 30:812-817.

DeMan, L., DeMan, J.M., and Buzzell, KL, 1989. Composition and properties of
soymilk and toﬁr made from Ontario light hilum soybeans. Can. Inst. Food Sci.
Terchnol. J. 20: 363.

Escueta, E.E., Bourne, M.C., and Hood, LR, 1986. Effect of boiling treatment of
soymilk on the composition, yield, texture and sensory properties of tofu. Can
Inst. Food Sci. TechnoL J. l9(2):53-56.

Fukushima, D., 1991. Recent progress of soybean protein foods: chemistry, technology,
and nutrition. Food Reviews International. 7(3):323-351.

Garcia, M.C., Torre, M., Marina, M.L., and Laborda, F., 1997. Composition and
characterization of soyabean and related products. Critical reviews in food
science and nutrition. 37(4):361-391.

Gebre, EA. and Summer, A.K., 1983. Preparation of high denaturation of soybean
protein by ﬁeezing. Part I. Agr. Biol. Chem. 35(4):449-459.

German, B., Damodaran, S., and Kinsella, J.E., 1982. Thermal dissociation and
association behavior of soy proteins. J. Agric. Food Chem 30:807-811.

Golbitz, P., 1995. Traditional soyfoods: processing and products. Journal of Nutrition.
125:570S-572S.

Hach Inc., 1982. Food and feeds analysis instruction manual. Hach Inc. Loveland, CO.

100

Hackler, LR. and Stillings, BR, 1967. Amino acid composition of heat-processed
soymilk and its correlation with nutritive value. Cereal Chem. 44:70-77

Howell, RW. and Caldwell, BE, 1978. Genetic and other biological characteristics, Pp.
27-60. Chpt. 2 in: Book of soybean: chemistry and technology. Vol.1. The Avi
Publishing Company, Inc. Westport, CN.

Iwabuchi, S. and Yanmuchi, F., 1987. Determination of glycinin and B—conglycinin in
soybean proteins by immunological methods. J. Agr. Food Chem. 35(2):200-
205.

Jackson, J.M. and Shin, B.M., 1979. Fundamentals of food canning technology. Avi
Publishing Company, Inc. Westport, CT.

Ji, M.P., Cai, T.D., and Chang, KC, 1999. Tofu yield and textural properties ﬁ'om three
soybean cultivars as affected by ratios of 7S and 118 proteins. J. Food Sci.
64(5):763-767.

Johnson, L.A., Deyoe, C.W., and Hoover, W.J., 1981. Yield and quality of soymilk
processed by steam-infusion cooking. J. Food Sci. 46:239-243.

Johnson, L.D. and Wilson, L.A., 1984. Inﬂuence of soybean variety and the method of
processing in tofu manufacturing: comparison of methods for measuring soluble
solids in soymilk. J. Food Sci. 49:202-204.

Lee, CH. and Rha, C., 1978. Microstructure of soybean protein aggregates and its
relation to the physical and texture properties of the curds. J. Food Sci. 43:79.

Lee, Y.C., Rosenau, JR, and Peleg, M., 1983. Rheological characterization of tOﬁJ. J.

Texture Studies. 14:143-154.

101

Lim, B.T., DeMan, J.M., DeMan, L., and Buzzell, R.I., 1990. Yield and quality of toﬁr
as affected by soybean and soymilk characteristics. Calcium sulfate coagulant. J.
Food Sci. 55(4):]088-1092, 1111.

Liu, K., 1999. Soybeans, chemistry, technology, and utilization. Aspen Publishers, Inc.,
Gaithersburg, MD.

Liu, K., 2000. Expanding soybean food utilization. Food Technol. 54(7):46-58.

Lo, G., 1989. Nutritional and physical properties of dietary ﬁber ﬁom soybeans. Cereal
Food World. 34:530-534.

Maruyama, N. and Sato, K, 1999. Structure-physicochemical function relationships of
soybean B-conglycinin constituent subunits. J. Agri. Food Chem. 47:5278-5284.

Messina, M.J., 1994. Soyfoods: Their role in disease prevention and treatment. Pp 442-
448. Chpt. 10 in: Soybeans chemistry, technology, and utilization, Liu, K. (1999).

Messina, M.J., 1999. Legumes and soybeans: overview of their nutritional proﬁles and
health effects. Am. J. C1in.Nutr. 70(3):439S-4508.

Metussin, R., Alli, I., and Kermasha, S., 1992. Micronization effects on composition and
properties of tofu. J. Food Sci. 57(2):418-422.

Moizuddin, S., Johnson, L.D., and Wilson, LA, 1999. Rapid method for determining
optimum coagulant concentration in tofu manufacture. J. Food. Sci. 64(4):684-
487.

Murphy, P.A., Chen, H.P., Hauck, CC, and Wilson, L.A., 1997. Soybean protein
composition and toﬁi quality. Food Technol. 51(3):86-88, 110.

Nelson, A.I., Steinberg, M.P., and Wei, LS, 1976. Illinois process for separation of

soymilk. J. Food Sci. 41:57-61.

102

Prosoya Inc., 1998. VS 30/V S 40 Soyacow system manual. Ottawa, Ontario, Canada.

Ranhotra, G., Gelroth, J., and Strouts, B., 2000. Use of soy protein in grain-based foods.
Technical Bulletin, American Institute of Baking. 12 (8).

Riaz, M.N., 1999. Soybeans as functional foods. Cereal Food World. 44(2):88-92.

Saio, K., 1979. Tofu-relationships between texture and ﬁne structure. Cereal Food
World. 24:342-354.

Saio, K. and Watanabe, T., 1978. Differences in functional properties of 7S and 11S
soybean proteins. J. Texture Study. 9:135-157.

Schaefer, M.J. and Love, J ., 1992. Relationships between soybean components and toﬁr
texture. J. Food Quality. 15:53-66.

Schroder, DJ. and Jackson, H., 1972. Preparation and evaluation of soybean curd with
reduced beany ﬂavor. J. Food Sci. 37:450-451.

Shen, C.F., DeMan, L.D., Buzzell, R.I., and DeMan, J.M., 1991. Yield and quality of
tofu as affected by soybean and soymilk characteristics: Glucono-delta lactone
coagulant. J. Food Sci. 56(1): 109-112.

Shurtleff, W. and Aoyagi, A., 1990. Tofu and soymilk production, The book of tofu,
Vol. 2. The Soyfood Center, Lafayette, CA.

Sizer, C.E., 1989. Aseptic packaging of soymilk. Chpt 6. Pp. 98-101, in: Food uses of
whole oil and protein seeds. Lucas, E.W., Erickson, D.R., And Nip, W.K.
American Oil Chemists’ Society, Charnpaign, IL.

Smith, A.K., Watanabe, T., and Nash, A.M. 1960. TOﬁJ from Japanese and United

States Soybeans. Food Technol. 14:332-336.

103

Smith, A. and Circle, S. 1978. Soybeans: Chemistry and technology. Vol.1. The Avi
Publishing Company, Inc. Westport, CN.

Stone, H. and Sidel, J .L., 1993. Sensory evaluation practice, 2nd Ed. Academic Press,
San Diego, CA.

Szczesniak, AS, 1963. Classiﬁcation of texture characteristics. J. Food Sci. 28:385-
389.

Tsai, S.J., Lan, C.Y., Kao, CS, and Chen, S.C., 1981. Studies on the yield and quality
characteristics of toﬁt. J. Food Sci. 46:1734-1737.

United States Department of Agriculture, 2000. Agricultural statistics 2000. United
States Printing Ofﬁce Washington 2000, Washington, DC. pIII-17 & 19.

Utsumi, S. and Kinsella, J.E., 1985. Force involved in soy protein gelation: effects of

various reagents on the formation, hardness and solubility of heat-induced gels made

from 78, 118 and soy isolate. J. Food Sci. 50:1278-1282.

Utsumi, S. and Kinsella, J.E., 1985. Structure-ﬁmction relationships in food proteins:
subunit interactions in heat-induced gelation of 78, 118, and soy isolate proteins.
J. Agric. Food Chem. 33:297-303.

Vaidehi, MP and Kadam, 8.8., 1989. Soybean. Pages 1-31 in: The handbook of world
food legumes: nutrition chemistry, processing technology and utilization, Vol.3.
Salunke, D.K. and Kadam, S.S. CRP Press, Inc. Boca Raton, F 1a.

Verma, D.P.S. and Shoemaker, RC, 1996. Soybean: genetics, molecular biology and
biotechnology. Cab International, Oxon, UK.

Wang, H., 1993. Evaluation of tofu from selected soy cultivars and breeding lines. M. S.

Thesis, Michigan State University, East Lansing, MI.

104

Wang, H.L., Swain, E.W., and Kwolek, W.F., 1983. Effect of soybean varieties on the
yield and quality of tOﬁl. Cereal Chem. 60(3):245-248.

Wang, H.L. and Calvins, JR, 1989. Yield and amino acid composition of ﬁactions
obtained during tOﬁl production. Cereal Chem. 66:359-361.

Wilkens, W.F. and Hackler, LR, 1969. Effect of processing conditions on the
composition of soymilk. Cereal Chem. 46: 391-397.

Wilkens, W.F., Mattick, LR, and Hand, DB, 1967. Effect of processing method on
oxidative off-ﬂavors of soybean milk. Food Technol. 21: 86-89.

Williams, LR, 1950. Structure and genetic characteristics of the soybean. Chpt 3, in:
Soybeans and soybean products, Vol I. Klares Markley, Interscience Publishers,
Inc. New York, NY.

Yadav, NS, 1996. Genetic modification of soybean oil quality. Chpt 8, in: The book of
soybean: genetics, molecular biology and technology. Verma, D.P.S. and

Shoemaker, RC. Cab International, Oxon, UK.

105