.x?

5554. run“. 7H

m .
5. a:

:-

_. 1.4.. .
.45. .u

www.cumy—F
83
g. L

...
; zv
«.1.

1
d! 3.92.. :-

.. f
6. .9?

Fin :

.3
. .1

I..zfl.Ws.Wui..w.t
5
Away km

1.2.,
._
#1.ow

:

 

 

 

:9 (1.3 v‘ V? [4| VF

 

._ 1‘ .'¥‘¢.b\. :1! . :2

 

" ‘ - I, '1 '1’ w'! I! \ i‘l l’

T}‘o;.'fl I

 

LIBRARY
MIChIgan State
Universaty

 

 

 

This is to certify that the
thesis entitled

THE EFFECTS OF EXTRUSION COOKING AND MILLING
ON THE INSTANT PROPERTIES OF WHEAT AND SOY
POWDERS

presented by

Edmund John Tanhehco

has been accepted towards fulfillment
of the requirements for the

 

 

 

 

Master of degree in Food Science
Science
Mfior Prdfessor’s Signature
M . I“, 100 L
Date

MSU 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

 

 

 

THE EFFECTS OF EXTRUSION COOKING AND MILLING ON THE INSTANT
PROPERTIES OF WHEAT AND SOY POWDERS

By

Edmund John Tanhehco

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

2002

ABSTRACT

THE EFFECTS OF EXTRUSION COOKING AND MILLING ON THE INSTANT
PROPERTIES OF WHEAT AND SOY POWDERS

By

Edmund John Tanhehco

Instant powders, which require only reconstitution in liquid for use in products such as
beverages, can be produced by extrusion cooking. Powders with good instant properties
should disperse easily, without lumps. Instant cereal and soy powders are often difficult
to disperse because of swelling and fine grinding, respectively. In this study, extrusion
was done using a twin-screw extruder and varying the dough moisture content, barrel
temperature, and screw speed. The extrudates were then milled to produce powders with
different particle size ranges. The specific mechanical energy (SME), water absorption
index (WAI), water solubility index (WSI), viscosity, ease of dispersibility, and stability
to sedimentation were measured. The WAI and suspension viscosity of instant wheat
powders were negatively correlated with the SME, and decreased with particle size. The
WSI of wheat powders increased with the SME. The WAI and W8] of instant soy
powders were not strongly influenced by the extrusion conditions used. Wheat powders
with larger particle sizes were easily dispersed, however below 93 um, dispersibility was
inversely correlated with the SME. The addition of lecithin improved dispersibility. Soy
powders displayed poor dispersibility when the particle size was reduced to
approximately 15 um. Stability to sedimentation was greater as particle size decreased

for both wheat and soy powders.

ACKNOWLEDGEMENTS

I would like to express thanks to my advisor, Dr. P.K.W. Ng for his guidance and
patience throughout the completion of my program. His providing of resources and time
are greatly appreciated. Dr. K.D. Dolan and Dr. J.N. Cash are also to be acknowledged
for providing advice and direction while serving on my committee. Special thanks are
given to R.J. Wolthuis for his knowledge and skill regarding much of the equipment used
in the studies. Without the aid of D. Ngo during sample preparation, and the laser
particle size analysis results from Dr. G.A. Hareland of the USDA/ARS Wheat Quality
Lab at North Dakota State University, this work would have been much more difficult to
accomplish. The help of EB. Smith with statistical analysis is also deeply appreciated.
Most importantly I would like to thank those mentioned above along with the other
members of the cereal science lab for providing friendship and making my time much

more enjoyable.

III

TABLE OF CONTENTS

Chapter I Page
LIST OF TABLES ............................................................................................................ IX
LIST OF FIGURES ......................................................................................................... XII
I INTRODUCTION ........................................................................................................... 1
II LITERATURE REVIEW ............................................................................................... 3
2.1. EXTRUSION ............................................................................................................. 3
2.1.1. Single- and Twin-Screw Extruders ................................................................. 4
2.1.2. Independent Variables .................................................................................... 4
2.1.3. Dependent Variables ....................................................................................... 5
2.2. WHEAT FLOUR ....................................................................................................... 6
2.2.1. Pregelatinization ............................................................................................. 7
2.2.2. Extruded Instant Cereal Powders .................................................................... 9
2.3. SOYBEAN UTILIZATION ........................................................................................ 10
2.3.1. Nutritional Value of Soy Protein .................................................................. 11
2.3.2. Soy Drinks .................................................................................................... 13
2.3.3. Mouthfeel ...................................................................................................... 14
2.3.4. Extrusion Of Soy ........................................................................................... 15
2.4. EXTRUSION OF SOY AND CEREAL BLENDs ........................................................... 15
2.5. INSTANT POWDER PROPERTIES ............................................................................. 16
2.5.1. Wettability ..................................................................................................... 17

IV

2.5.2. Sinkability ..................................................................................................... 17

2.5.3. Solubility ....................................................................................................... 18
2.5.4. Dispersibility ................................................................................................. 18
2.6. INSTANTIZATION ................................................................................................... 19
2.7. EVALUATION OF INSTANT PROPERTIES ................................................................. 21
III MATERIALS AND METHODS ................................................................................ 23
3.1. RAW MATERIALS .................................................................................................. 23
3.2. EXTRUSION ........................................................................................................... 23
3.3. POST EXTRUSION PROCESSING ............................................................................. 24
3.3.1. Milling: Wheat Flour (WF) and Wheat Flour + Lecithin (WFL) ................ 24

3.3.2. Milling: Defatted Soy Flour (SF) and Defatted Soy Flour + Lecithin (SF L)
24

3.3.3. Milling: Blend of Raw Defatted Soy Flour with Full Fat Soy Flakes (S-
Blend) 27

3.4. ANALYSIS ............................................................................................................. 27
3.4.1. Laser Particle Size Analysis .......................................................................... 27
3.4.2. Specific Mechanical Energy (SME) ............................................................. 27
3.4.3. Moisture ........................................................................................................ 28
3.4.4. Water Absorption Index (WAI) and Water Solubility Index (WSI) ............. 28
3.4.5. Rapid Visco Analyzer Viscosity of WP and WFL ....................................... 28
3.4.6. Dispersibility ................................................................................................. 30
3.4.7. Stability ......................................................................................................... 31
3.4.8. Trypsin Inhibitor Inactivation of SF, SF L, and S-Blend .............................. 31

3.5. BLENDING OF DIFFERENT POWDERS ..................................................................... 32

V

3.6. STATISTICS ........................................................................................................... 32

IV RESULTS AND DISCUSSION ................................................................................. 33

4.1. EXTRUSION OF WHEAT FLOUR (WF) AND WHEAT FLOUR + LECITHIN (WFL) 33

4.1.1.

4.1.2.

4.1.3.

4.1.4.

4.1.5.

4.1.6.

4.1.7.

4.1.8.

4.1.9.

4.1.10.

4.1.11.

4.1.12.

4.1.13.

4.1.14.

4.1.15

Particle Size Analysis ................................................................................... 33
Specific Mechanical Energy (SME) During Extrusion of Wheat Flour ....... 33
SME During Extrusion of Wheat F lour with Lecithin .................................. 37
Water Absorption Index (WAI) of Extruded Wheat Flour ........................... 37
WAI of Extruded Wheat Flour with Lecithin ............................................... 42
Water Solubility Index (WSI) of Extruded Wheat Flour .............................. 44
WSI of Extruded Wheat Flour with Lecithin ................................................ 47
Rapid Visco Analyzer Viscosity of Extruded Wheat Flour .......................... 48

Rapid Visco Analyzer Viscosity of Extruded Wheat Flour with Lecithin 5]

Wettability ..................................................................................................... 58
Development of Dispersibility Procedure ..................................................... 58
Dispersibility of Extruded Wheat Flour ........................................................ 6O
Dispersibility of Extruded Wheat Flour with Lecithin ................................. 64
Stability of Extruded Wheat Flour ................................................................ 64
Stability of Extruded Wheat Flour with Lecithin ......................................... 65

4.2. EXTRUSION OF DEFATTED SOY FLOUR (SF), DEFATTED SOY FLOUR WITH LECITHIN
(SFL), AND BLEND OF DEFATTED SOY FLOUR WITH FULL FAT SOY FLAKES (S-BLEND)7O

4.2.1.

4.2.2.

4.2.3.

Particle Size Analysis of Extruded and Milled Soy ...................................... 70
Specific Mechanical Energy of Soy During Extrusion ................................. 70
Trypsin Inhibitor Inactivation of Extruded Soy ............................................ 73

VI

4.2.4. Water Absorption Index (WAI) of Extruded Soy ......................................... 76

4.2.5. Water Solubility Index (WSI) of Extruded Soy ............................................ 76
4.2.6. Dispersibility of Extruded Soy ...................................................................... 80
4.2.7. Stability of Extruded Soy .............................................................................. 84
4.3. BLENDS OF WHEAT AND SOY POWDERS FOR INSTANT DRINKS ............................ 89
V CONCLUSIONS .......................................................................................................... 93
VI SUGGESTIONS FOR FURTHER RESEARCH ....................................................... 95

APPENDIX 1: MOISTURE CONTENT OF EXTRUDED AND MILLED POWDERS

........................................................................................................................................... 96
APPENDIX IA. MOISTURE CONTENT OF EXTRUDED AND MILLED WHEAT FLOUR
POWDERS ........................................................................................................................ 97
APPENDIX 13. MOISTURE CONTENT OF EXTRUDED AND MILLED WHEAT FLOUR +
LECITHIN POWDERS ........................................................................................................ 98

APPENDIX IC. MOISTURE CONTENT OF EXTRUDED AND MILLED DEFATTED SOY FLOUR
POWDERS ........................................................................................................................ 99

APPENDIX ID. MOISTURE CONTENT OF EXTRUDED AND MILLED SOY FLOUR + LECITHIN
POWDERS ...................................................................................................................... 100

APPENDIX IE. MOISTURE CONTENT OF EXTRUDED AND MILLED BLEND OF DEFATTED
SOY FLOUR WITH FULL FAT SOY FLAKES (S-BLEND) POWDERS .................................. 101

APPENDIX 11: TORQUE, DIE PRESSURE, AND SPECIFIC MECHANICAL
ENERGY (SME) DURING EXTRUSION .................................................................... 102

APPENDIX IIA. TORQUE, DIE PRESSURE, AND SPECIFIC MECHANICAL ENERGY (SME)
DURING EXTRUSION OF WHEAT FLOUR ........................................................................ 103

APPENDIX IIB. TORQUE, DIE PRESSURE, AND SPECIFIC MECHANICAL ENERGY (SME)
DURING EXTRUSION OF WHEAT FLOUR + LECITHIN .................................................... 104

APPENDIX IIC. TORQUE, DIE PRESSURE, AND SPECIFIC MECHANICAL ENERGY (SME)
DURING EXTRUSION OF DEFATTED SOY FLOUR ........................................................... 105

VII

APPENDIX IID. TORQUE, DIE PRESSURE, AND SPECIFIC MECHANICAL ENERGY (SME)
DURING EXTRUSION OF DEFATTED SOY FLOUR + LECITHIN ........................................ 106

APPENDIX IIE. TORQUE, DIE PRESSURE, AND SPECIFIC MECHANICAL ENERGY (SME)

DURING EXTRUSION OF BLEND OF DEFATTED SOY FLOUR WITH FULL FAT SOY FLAKES
(S-BLEND) .................................................................................................................... 107

APPENDIX III: RAPID VISCO ANALYZER ANALYSIS OF EXTRUDED AND
MILLED POWDERS ..................................................................................................... 108

APPENDIX IIIA. RAPID VISCO ANALYZER ANALYSIS OF EXTRUDED AND MILLED WHEAT
FLOUR POWDERS .......................................................................................................... 109

APPENDIX IIIB. RAPID VISCO ANALYZER ANALYSIS OF EXTRUDED AND MILLED WHEAT
FLOUR + LECITHIN POWDERS ....................................................................................... 1 10

BIBLIOGRAPHY ........................................................................................................... 1 1 1

VIII

LIST OF TABLES

Table Page

Table 3.1. Extrusion Conditions for Wheat Flour and Wheat Flour + Lecithin .............. 25

Table 3.2. Extrusion Conditions for Defatted Soy Flour, Defatted Soy Flour + Lecithin,
and Blend of Defatted Soy Flour with Full Fat Soy Flakes ....................................... 26

Table 3.3. Rapid Visco Analyzer Temperature Profile and Analysis Definitions ........... 29

Table 4.1. Laser Particle Size Analysis of Extruded and Milled Wheat Flour (WP) and
Wheat Flour + Lecithin (WFL) Powders ................................................................... 34

Table 4.2. Specific Mechanical Energy (SME) and Statistical Relationships During
Extrusion of Wheat Flour (WF) and Wheat Flour + Lecithin (WFL) ........................ 35

Table 4.3. Comparison8 Between Properties of Extruded and Milled Wheat Flour (WF)
and Wheat Flour + Lecithin (WFL) Powders ............................................................ 38

Table 4.4. Water Absorption Index (WAI) and Statistical Relationships of Extruded and
Milled Wheat Flour Powders ..................................................................................... 39

Table 4.5. Water Absorption Index (WAI),Water Solubility Index (WSI), and Statistical
Relationships of Extruded and Milled Wheat Flour + Lecithin Powders (<93 um
Particle Size F raction)a ............................................................................................... 43

Table 4.6. Water Solubility Index (WSI) and Statistical Relationships of Extruded and
Milled Wheat Flour Powders ..................................................................................... 45

Table 4.7. Cold Viscosity and Statistical Relationships of Extruded and Milled Wheat
Flour Powders ............................................................................................................ 49

IX

Table 4.8. Hot Paste Viscosity and Statistical Relationships of Extruded and Milled
Wheat Flour Powders ................................................................................................. 53

Table 4.9. Cold Viscosity and Statistical Relationships of Extruded and Milled Wheat
Flour + Lecithin Powders ........................................................................................... 55

Table 4.10. Hot Paste Viscosity and Statistical Relationships of Extruded and Milled
Wheat Flour + Lecithin Powders ............................................................................... 57

Table 4.11. Dispersibilitya and Statistical Relationships of Extruded and Milled Wheat
Flour (WP) and Wheat Flour + Lecithin (WFL) Powders ......................................... 61

Table 4.12. Stability and Statistical Relationships of Extruded and Milled Wheat Flour
Powders ...................................................................................................................... 66

Table 4.13. Stability and Statistical Relationships of Extruded and Milled Wheat Flour +
Lecithin Powders ........................................................................................................ 68

Table 4.14. Laser Particle Size Analysis of Extruded and Milled Soy Powders ............. 71

Table 4.15. Specific Mechanical Energy (SME) and Statistical Relationships During
Extrusion of Soy ......................................................................................................... 72

Table 4.16. Comparison of Properties Between Soy Powdera Formulations .................. 74

Table 4.17. Trypsin Inhibitor Inactivation (TH) and Statistical Relationships of Extruded
and Milled Soy Powders ............................................................................................ 75

Table 4.18. Water Absorption Index (WAI) and Statistical Relationships of Defatted Soy
Flour Powders ............................................................................................................ 77

Table 4.19. Water Absorption Index (WAI) and Statistical Relationships of Defatted Soy
Flour + Lecithin Powders ........................................................................................... 78

Table 4.20. Water Absorption Index (WAI), Water Solubility Index (WSI), and
Statistical Relationships of Blend of Defatted Soy Flour with Full Fat Soy Flakes (S-
Blend) Powders .......................................................................................................... 79

Table 4.21. Water Solubility Index (WSI) and Statistical Relationships of Extruded and
Milled Defatted Soy Flour Powders ........................................................................... 81

Table 4.22. Water Solubility Index (WSI) and Statistical Relationships of Extruded and
Milled Defatted Soy Flour + Lecithin Powders ......................................................... 82

Table 4.23. Dispersibility and Statistical Relationships of Extruded and Milled Defatted
Soy Flour Powders ..................................................................................................... 83

Table 4.24. Dispersibilitya and Statistical Relationships of Extruded and Milled Defatted
Soy Flour + Lecithin Powders .................................................................................... 85

Table 4.25. Stability and Statistical Relationships of Extruded and Milled Defatted Soy
Flour Powders ............................................................................................................ 86

Table 4.26. Stability and Statistical Relationships of Extruded and Milled Defatted Soy
Flour + Lecithin Powders ........................................................................................... 87

Table 4.27. Stability and Statistical Relationships of Extruded and Milled Blend of

Defatted Soy Flour with Full Fat Soy Flakes (S—Blend) Powders ............................. 88
Table 4.28. Dispersibility of 50:50 Blends of Wheat and Soy Powders .......................... 90
Table 4.29. Stability of 50:50 Blends of Wheat and Soy Powders .................................. 91
Table 4.30. Cold Viscosity of 50:50 Blends of Wheat and Soy Powders ....................... 92

XI

LIST OF FIGURES

Figure Page

Figure 2.1. Rapid Visco Analyzer pasting curve of soft wheat flour. Viscosity in Rapid
Visco Units (RVU). ...................................................................................................... 8

Figure 4.1. Relationship between the specific mechanical energy (SME) and the water
absorption index (WAI) for extruded and milled wheat flour powders. .................... 40

Figure 4.2. Relationship between the specific mechanical energy (SME) and the water
solubility index (WSI) for extruded and milled wheat flour powders. ...................... 46

Figure 4.3. Relationship between the specific mechanical energy (SME) and cold
viscosity for extruded and milled wheat flour powders. ............................................ 50

Figure 4.4. Representative Rapid Visco Analyzer pasting curve of extruded and milled
wheat flour powders. Viscosity in Rapid Visco Units (RVU). ................................. 52

Figure 4.5. Relationship between the specific mechanical energy (SME) and hot paste
viscosity for extruded and milled wheat flour powders. ............................................ 54

Figure 4.6. Relationship between the specific mechanical energy (SME) and
dispersibility for extruded and milled wheat flour powders with (WF L) and without
(WF) lecithin. ............................................................................................................. 63

Figure 4.7. Relationship between the specific mechanical energy (SME) and stability for
extruded and milled wheat flour powders. ................................................................. 67

Figure 4.8. Relationship between cold viscosity and stability for extruded and milled
wheat flour + lecithin powders. .................................................................................. 69

XII

I Introduction

Extrusion processing is commonly used in the food industry because of its versatility.
Processing conditions can be altered depending on the raw materials and on the desired
qualities of the end product. Examples of extruded products include puffed cereals for
breakfast or snacks, pet foods, and texturized meat analogs made from soy.

Extruders can also be used as cookers to produce instant food powders that need only
reconstitution in liquid. The starches in cereals, such as wheat and corn, can be
pregelatinized to improve their cold water properties. The advantage of extrusion
cooking is that it can modify the characteristics of starch, such as solubility and viscosity.
The instant powders can then be used for products such as gruels or beverages. Extrusion
has also been used to make low cost, nutritious soy powders for aid programs. However,
soy is not widely accepted in the United States due to objectionable characteristics such
as poor flavor and mouthfeel.

In addition to being ready to use without further cooking, instant powders should also
be easily reconstituted. This is especially important for powders that are meant to be
consumed at home, like instant beverage bases. A powder with good instant properties
should disperse easily into liquid and should not sediment too quickly. Some factors
affecting instant properties include the type of powder and liquid, along with the particle
size of the powder.

Cereal powders can be difficult to reconstitute because of swelling, which can lead to
the formation of lumps. Soy powders are another case where problems can occur

because of the difficulty in mixing fine powders. Various processes are used to instantize

or improve the instant properties of powders, including agglomeration to make larger
particles, and the addition of lecithin to improve wetting.

The Objectives Of these studies were tO determine the effects Of different extrusion
conditions and powder particle Sizes on the instant properties of wheat and soy powders,
to delineate the effects of lecithin on instant properties, and to develop blends of wheat

and soy powders that are suitable for an instant beverage base.

II Literature Review

2.1. Extrusion

Extruders work by mixing materials into a viscous dough-like mass and forcing it
through a restricted Opening. The ingredients are fed into a cylindrical barrel, which
contains either one or two rows of flighted screws (Harper, 1981). As the screw(s) turn,
the materials are mixed and conveyed down the barrel. The lengths of barrels are
described by the ratio of the barrel length to its diameter (L/D), and typically range from
1:1 to 20:1 (Harper, 1981). Changes in the geometry of the screws, along with the
restricted die opening at the end of the barrel, cause the material to be compressed and
the pressure to increase. AS the material makes its way down the barrel, these forces
cause the material to become a plasticized dough, which iS subject to the mixing and
shearing forces of the turning screw(s). The mechanical energy imparted by the screws iS
dissipated in the form of heat. Additional heat can be applied through heating of the
extruder barrel or direct injection of steam. Extruders can be considered high
temperature/Short time cookers because temperatures can reach as high as 200°C with
residence times ranging from only 5-10 seconds to over 1 minute. The shear rate and
temperature are typically highest at the end of the barrel near the exit die, and this is
where most of the cooking and product transformation takes place. Upon exiting the die,
moisture in the extrudate flashes off causing expansion and cooling of the product
(Harper, 1981). During extrusion, some of the changes that take place in the raw material
include starch gelatinization and transformation, protein denaturation and alignment,
killing of microbes, development of colors and flavors, inactivation of antinutritional

factors, and the destruction of vitamins and other nutrients (Harper, 1981).

Important advantages of extruders include their ability to produce a wide range of
products and their high production capacity. Producing foods with extruders is also low
cost due to efficiencies in labor, floor Space and energy consumption. Processing can be
done at lower moistures, reducing the amount of heat needed during processing and

drying of the product (Harper. 1981).

2.1.1. Single- and Twin-Screw Extruders

The two main types of extruders are the single- and twin-screw extruders. Single-
screw extruders rely on one set of screws to convey materials down the barrel by drag
flow (Frame, 1994). For this extruder to function, the material must have enough friction
with the barrel wall so that the material does not rotate with the screws.

Twin-screw extruders consist of two sets of interrneshing screws which can either
rotate in the same (co-rotating) or Opposite (counter-rotating) direction. The more
common co-rotating twin screw extruders convey materials by both drag and positive
displacement flow (Frame, 1994). Materials are transferred from one screw to the other,
which enhances mixing and allows the twin-screw extruder to handle a wider variety of

materials than a single-screw extruder.

2.1.2. Independent Variables

Independent variables are the operating conditions that can be controlled to produce
the desired extrudate characteristics. One variable that can be changed is the screw
geometry at different positions along the barrel. The flight on a screw refers to the

portion that forms the helical pattern that conveys material when the screw is turned

(Harper, 1981). The number, depth, and angle of the flights affects the conveying
capacity of the screw. For example, at the feed section, screws typically have less
Shallow flights than screws near the die. These deeper flights have greater conveying
capacity while the more Shallow flights have more pumping ability, the latter needed to
force material through the die (Frame, 1994). Mixing elements such as paddles, screws
with cut flights, and reverse pitch screws that convey materials in the opposite direction
can also be used (Harper, 1986). In addition to mixing, these types of elements provide
additional Shear and alter the residence time of the material in the extruder. Other
independent variables include the raw materials, moisture content of the ingredients, die

design, barrel temperature, screw Speed, and feed rate (Harper, 1986).

2.1.3. Dependent Variables

Dependent variables are affected by changes in the independent variables. One of the
most important dependent variables is the specific mechanical energy (SME). It is a
measure of the amount of work being put into the material during extrusion and is a
function of the feed rate, screw Speed, and rheological properties of the extruded material
(Frame, 1994). The SME has been correlated with many product characteristics and is
one of the most important variables to consider for scale up (Levine, 1989; Schuchmann
and Danner, 2000). The shear rate is another important variable and it is affected by the
screw speed, screw geometry, die geometry, and the rheological properties of the
extruded material (Harper, 1981). Other dependent variables include pressure, which can
reach over 1000 psi at the die, viscosity of the material, and residence time (Harper,

1981). The most important dependent variables are the actual product characteristics.

These include product expansion, color, texture, water solubility, and many other

measurable parameters.

2.2. Wheat Flour

Wheat flour can be extruded to make a variety of products including breakfast cereals,
snacks, and instant powders. It is primarily composed of protein and starch. Depending
on the wheat variety, protein content in flour can range from 7-15% at a 14% moisture
basis (Atwell, 2001). Wheat flour is comprised of approximately 63-72% starch on a
14% moisture basis (Atwell, 2001 ), and is regarded as the main functional component in
extruded flour products (Colonna et al., 1989; Guy and Horne, 1988). It is composed of
amylose and amylopectin, which are 0t-1,4 and a-I,6 linked polymers of glucose,
respectively (Thomas and Atwell, 1999). Amylose is a straight chain polymer which can
take the form of a helix with a hydrophobic core. This allows it to bind free fatty acids
and some alcohols. Amylopectin is a branched chain polymer and accounts for
approximately 75% of the starch in wheat. These components of starch are arranged in
discrete, semicrystalline granules, which are approximately 1-45 microns in size. Starch
granules are highly ordered and show birefringence when viewed under polarized light
(Hoseney, 1994).

One of the main functional properties of starch is its ability to absorb water and
increase the viscosity of suspensions. However, native starch in cold water has very low
water solubility and water absorption. It is not until heating takes place that starch begins
to increase the viscosity of suspensions. The irreversible changes that starch undergoes

during heating in water are referred to as gelatinization and pasting (Thomas and Atwell,

1999). During gelatinization, starch granules lose their ordered structure and
birefringence. They also begin to swell and increase in solubility (Atwell et al., 1988).
AS this occurs, the viscosity of the starch/water suspension increases. Gelatinization
occurs over a range of temperatures [52-85°C for wheat starch (Thomas and Atwell,
1999)], leading to a steady increase in viscosity. A peak viscosity is reached when the
majority of starch granules are swollen and still intact. The swollen granules eventually
begin to break down and exude their components, a process referred to as pasting. Once
allowed to cool, the starch components, primarily amylose, can re-associate and form a
gel. The Amylograph and Rapid Visco Analyzer (Figure 2.1) are two instruments that
produce pasting curves of starch by measuring the relative viscosity of starch suspensions

dun'ng programmed heating and cooling cycles.

2.2.1. Pregelatinization

Instantization or pregelatinization is processing to make starches with higher cold
water viscosity and solubility. These instant powders are often used in products that
undergo little or no heating prior to use. Some functions of pregelatinized starch are
binding and the addition of viscosity and texture. Production of instant starches is
commonly done by drum drying a 30-40% solids slurry of starch in water (Thomas and
Atwell, I999). The dried product is then ground into a powder.

Extrusion can also be used to pregelatinize starches and has advantages over drum
drying because starch properties such as water absorption, water solubility, and cold
viscosity can be modified (Colonna et al., 1984). Experiments comparing drum dried and

extruded wheat starches have Shown that extrusion cooking produces starch with lower

 

 

 

 

500
Temperature —
400 -
S _
> 300 a
93
.é‘ -
m
8
.2 20° ‘ _
>
100 — Viscosity
0 w I I I
0 5 10 15 20

Figure 2.1. Rapid Visco Analyzer pasting curve of soft wheat flour. Viscosity in
Rapid Visco Units (RVU).

Time (minutes)

100

90

80

70

60

50

40

3O

20

Temperature (°C)

viscosities and water absorptions, and higher water solubilities (Colonna et al., 1984;
Doublier et al., 1986). Colonna et al. (1984) found that as the water absorption
decreased, the cold water solubility of the extruded starch increased to 60-80% compared
to a solubility of only 14-25% for the drum dried starch. Extrusion led to the disruption
of starch granules and the macromolecular degradation of amylose and amylopectin.
Anderson et al. (1969) Observed that the water solubility of extruded corn grits was
positively correlated with extrusion temperature and negatively correlated with the feed
moisture during extrusion. Higher Shear conditions were found to lower the viscosity of

starch/water slurries.

2.2.2. Extruded Instant Cereal Powders

Schuchmann and Danner (2000) identified extrusion conditions that produced instant
corn powders for use as thickeners and beverages. High barrel temperature and extrusion
moisture, low screw speed, and medium to low shear conditions produced powders with
high suspension viscosity and high water binding capacity. These powders were judged
to be good for use as instant thickeners. To make powders suitable for products such as
infant formulas, conditions were adjusted to produce powders with low suspension
viscosities and high water solubilities. Low extrusion moisture, temperature, and screw
speed, and a high-shear screw configuration produced powders with these characteristics.
Lower suspension viscosity and higher water solubility was correlated with higher SME.
The SME was regarded by Schuchmann and Danner (2000) as the most important

parameter to control.

Anderson et a1. (1970) produced extruded corn grit powders with a range of
characteristics. Extrusion at 25% moisture and 188°C produced a powder suitable for
gruel, while extrusion at 14% moisture and 188°C produced a powder suitable for a
beverage. The latter powder had a lower viscosity and higher solubility in water.
Grinding the extrudate to pass through a 40 mesh screen gave a powder with good
dispersibility and a smooth mouthfeel. Although grinding to a smaller particle size gave

a smoother drink, it also made dispersibility more difficult.

2.3. Soybean Utilization

The United States is the largest producer of soybeans in the world with 95% of its
domestic utilization going towards oil production. Soybeans have become a major
oilseed due to their low cost and high oil content, which is of good quality (Liu, I997a).
After oil extraction, the remaining defatted meal is used primarily for animal feed. Less
than 5% of the meal is further processed for use in human food (Traina and Breene,
1994). One of the reasons for the limited acceptance of soyfoods in the US. is flavor
(Liu, 2000). Products like soymilk are often referred to as having a beany or grassy
flavor, which is objectionable to people unaccustomed to it. Another factor is the
relatively high costs of soy due to limited sales volumes. Additionally, sales are mostly
through specialty health food stores. However, as more people consume soy products,
prices should decrease along with increased availability (Liu, 2000).

Soybeans are commonly processed into soy flour, soy protein concentrates, and
isolates for use as food ingredients. Soy flour is available in either full fat or defatted

forms. Soy protein concentrates and isolates are produced from defatted soy flakes, and

10

are defined as containing no less than 70% protein (N x 6.25) or no less than 90% protein
(N x 6.25), respectively, on a dry basis (Ohren, 1981). Soy powders with a range of
functionalities can be produced by controlling the degree of heat treatment. The protein
dispersibility index (PDI) is one measure of heat treatment that soy products have
undergone. AS heating is prolonged or done at higher temperatures and moistures, the
PDI decreases. Cooking of soy also serves to kill microorganisms, destroy antinutritional
factors and remove bitter or beany flavors (Fulmer, 1989; Heiser and Trentelman, 1989;
Johnson and Kikuchi, 1989). Some functional uses of soy include emulsification,
foaming, fat and water absorption, gelation, flavor binding, and color control. Baked
goods, baby foods, protein beverages, meat products, and hydrolyzed vegetable proteins

are examples of foods to which soy is added (Fulmer, 1989; Wolf, 1970).

2.3.1. Nutritional Value of Soy Protein
The high content and quality of protein in soybeans makes it a good source to meet

dietary needs. Although soy protein contains all of the essential amino acids, it is low in
sulfur-containing amino acids such as methionine (Scrimshaw, 1981; Torun et al., 1981).
However, soy proteins are high in lysine relative to cereal proteins (Horan, 1973; Graham
et al., 1972). Thus, these two plant sources of protein are complementary and, when
blended, can provide a source of complete protein.

To obtain the full nutritional value of soy, heat treatment must be done to inactivate
antinutritional compounds. These include trypsin inhibitors, goitrogens, antivitamins,
and phytates (Liener, 1981). Of these, trypsin inhibitors have been the most widely

studied. An early study by Rackis (1978) found that the inactivation of trypsin inhibitor

11

(TI) activity correlated with an increase in the protein efficiency ratio (PER), measured as
the weight gain of young rats relative to the amount of protein consumed (Liener, 1981;
Liu, 1997b). Although heat treatment can increase protein quality, excess heat should be
avoided to minimize the destruction and biological inactivation of amino acids (DelValle,
1981).

The consumption of soy foods is increasing and this has been due largely to awareness
of the soybean’s health benefits (Liu, 2000). In 1999, the US. Food and Drug
Administration (FDA) approved labeling for a health claim regarding the role of soy
protein in reducing the risk of coronary heart disease (FDA, I999). The ruling came in
response to studies showing that the consumption of 25 grams of soy protein per day
significantly reduced blood cholesterol levels. Based on this, foods containing at least
6.25 grams of soy protein per serving were ruled eligible to make the claim.

Isoflavones are another component of soybeans that have received attention, due to
studies showing that they may inhibit the grth of cancer cells and reduce blood
cholesterol levels (Messina, 1997). Isoflavones have a chemical structure similar to
estrogen and have also been used to treat post-menopausal women (Barnes, 1998;
Setchell and Cassidy, 1999).

Soy can also be a good source of fiber, as the crude fiber contents of defatted soybean
meal and defatted, dehulled soybean meal are 8.2% and 4.3%, respectively (F ulmer,
1989). Potential health benefits of soy fiber include increased bowel function, lowering
of serum cholesterol, and influence on blood glucose levels and insulin response (Riaz,

2001; Slavin, 1991).

12

2.3.2. Soy Drinks

Soymilk is one of the most common soy foods and consists of the water-extractable
portion of soybeans (Chen, 1989). Many different procedures exist to make soymilk, but
the basic steps involve the grinding of soybeans in water and subsequent filtration to
remove okara, the non-water-extractable portion. There must also be a cooking step to
destroy microbes and antinutritional factors like trypsin inhibitors. Some other
processing steps may involve dehulling, soaking before grinding, adding of flavoring, and
homogenization (Chen, 1989). The main goals are to produce a high yield of soymilk
with good flavor. The yield of soybean solids in soymilk typically is 55-60% of the
starting material with 70-80% of the protein being extracted (Chen, 1989). The
remaining unextracted solids comprise what is termed okara. The Illinois process for
soymilk production is a method which incorporates the whole soybean by using grinding
and homogenization to reduce the particle Size enough so that there is an acceptable
mouthfeel and good stability of the beverage (Nelson et al., 1976).

Another process utilizing the whole soybean to produce a beverage was developed by
Mustakas et a1. (1971), in which the soybeans were cooked by extrusion. The resulting
full fat soy flour was then wet milled and spray dried to make a beverage base.

Other soy—based beverages that can commonly be found in health food stores are dry
powder mixes. Many of these powders are marketed as protein supplement beverages or
shakes and utilize soy protein concentrate or isolate. These powders can be expensive
because the production cost per pound of isolate is greater than 10 times the cost to make

one pound of soy flour (Langsdorf, 1981).

13

A disadvantage of some of these beverages and powders, aside from their cost, is that
the whole soybean is not utilized. Okara, the byproduct from soymilk processing, has
found limited use as a human food due to its high fiber content (O’Toole, 1999). Its
disposal is also difficult Since for every kilogram of soybeans processed, about 1.1
kilograms of okara is produced (Khare et al., 1995). Dry powders based on soy protein
concentrates and isolates also do not utilize the whole soybean and contain lower
amounts of some nutrients like fiber. These powders also are often inconvenient to use

because they may require a blender for reconstitution.

2.3.3. Mouthfeel

Incorporating the whole soybean into a beverage is difficult because it can result in a
grainy mouthfeel. Chalkiness, graininess, or a “catch throat” sensation, each refer to an
unfavorable mouthfeel characteristic that can occur in soy-based drinks. This has been
found to be especially true for soymilk produced by the Illinois method, which
incorporates the whole soybean (Kuntz et al., 1978). To reduce the chalkiness of soymilk
produced by this method, Kuntz et al. (1978) tried different pretreatments for the
soybeans, different homogenization pressures, and adjustments of the soybean soaking
pH. They found that soaking the beans in 0.25% NaHC03 and homogenization of the
soymilk at higher pressures resulted in greater particle size reduction and less chalkiness.

Mustakas et a1. (1971) also found that soy flour produced by extrusion had a coarse
mouthfeel when suspended in water. After colloid milling and homogenization, the
mouthfeel was judged to be good. Scanning electron micrograph results showed that the

particle size of the soy powder was reduced from about 75 microns to 20 microns.

14

Traina and Breene (1994) evaluated eight commercially available full fat soy flours
for graininess when suspended in water. They found all eight to be grainy even though
one claimed to be milled superfine to eliminate the chalkiness. Laser particle Size
analysis showed that this flour did not meet the manufacturer’s claim that 80% of the

particles were smaller than 30 microns.

2.3.4. Extrusion of Soy

Extrusion has been investigated as an alternative to cooking with steam for the
production of full fat soy flour (Bookwalter et al., 1971a; Mustakas et al., 1964; Mustakas
et al., 1970). These early works demonstrated that extrusion cooking could be used to
economically process dehulled flakes or grits into high quality, full fat soy flour. By
adjusting extrusion conditions such as moisture and temperature, trypsin inhibitors (TI)
could be inactivated and beany and bitter flavors removed, leaving a bland tasting flour.
Of the extrusion conditions tested by Mustakas et a1. (1964), extrusion of the feed
materials at 20% moisture and 135°C resulted in 89% TI inactivation and the highest
PER. A study by Lorenz et al. (1980) found that extrusion at 143°C resulted in the
highest protein quality as measured by the PER, but only 57% TI inactivation. Extrusion
at 149°C resulted in 74% TI inactivation, but a lower PER. They also found that TI

inactivation increased with the moisture of the extruded raw material.

2.4. Extrusion of Soy and Cereal Blends
The extrusion of blends of soy and cereals has been studied as a low cost method of

producing instant foods for international aid programs. The combination of cereal and

15

soy protein is complementary, and iS a good source of complete protein from a nutritional
and economic standpoint. Numerous studies have shown that the nutritional quality of
these blends are comparable to that of casein (Bookwalter et al., 1971b; Bressani et al.,
1978; Horan, 1973; Jansen et al., 1978; Molina et al., 1983; Torun et al., 1981). Calories
from carbohydrates and oil provide additional nutritive value (Bressani and Elias, 1966).
The starch from cereals also has the functional effect of adding viscosity and texture.
Gruels can be made by choosing extrusion conditions which make starch with high
suspension viscosities while starch with low suspension viscosities are suited for
beverages (Anderson et al., 1970; Bookwalter et al., 1971b; Jansen et al., 1978;

Konstance et al., 1998; Molina et al., 1983).

2.5. Instant Powder Properties

In addition to being ready to use without further cooking, the term instant also refers
to the ease of reconstitution of a powder. However, defining what instant should mean in
this respect is difficult due to the wide range of products possible (Jensen, 1973). A
definition would need to take into account the intended use of the product and the
reconstitution process. For example, a powder that is mixed in a processing plant with
strong agitation would not need instant properties as good as those of a powder intended
for home use and mixing by hand. In general, there are four characteristics to consider
when evaluating instant properties of powders. They are wettability, sinkability,

dispersibility, and solubility (Jensen, 1973; Schubert, 1979; Schubert, 1993).

16

2.5.1. Wettability

Wettability refers to the ability of water to penetrate into the bulk powder (Schubert,
1993). It is affected by particle Size, porosity of the powder, and the contact angle
between the particle and the liquid. In general, faster wetting is favored by larger
particles. Powders with smaller particle sizes have narrower pore structures in the bulk
material, making the penetration of water more difficult, Slowing wetting. Conversely, if
particles are too large, penetration of liquid will take too long due to the formation of gas
inclusions in the pores (Schubert, 1993).

The contact angle between the liquid and the particle also affects wetting. For a
powder Sitting on top of a liquid, the contact angle is dependent on the interfacial tensions
between the solid, liquid, and the surrounding air (Walstra, 1996). If the contact angle is
too large, little or no wetting will occur. This is the case for powders that contain fats
such as cocoa powder. The contact angle also generally increases as particle size

decreases, making wetting more difficult (Schubert, 1979).

2.5.2. Sinkability

Sinkability describes how well particles Sink below the surface of the liquid. Schubert
(1993) has found that particles greater than 100 microns and having a density of 1.5
g/cm3 have good Sinkability. Powders with poor Sinkability may float on the surface,
resulting in a poor product appearance. Sinking is also important to allow for water to

more easily penetrate into a powder.

17

2.5.3. Solubility
Powders that are soluble should dissolve quickly and with a minimum of undissolved
particles. There are also powders such as cocoa powder that are not soluble, but instead

form suspensions, and this property is therefore not applicable (Jensen, 1973).

2.5.4. Dispersibility

A powder with good dispersibility should mix easily and without lumps in the liquid.
It also describes how well the particles remain suspended in the liquid. Dispersibility can
be considered the most important property because it encompasses the other instant
properties and has the most practical importance.

One factor that affects dispersion is particle size. AS particles become smaller,
interparticle adhesion becomes greater, causing the formation of lumps (Schubert, 1979).
The ratio of the force of particle adhesion (F), to the weight of the particle (W), is
inversely proportional to the square of the particle size (Schubert, 1987). Thus, 1 pm
particles have 106 greater F/W ratios than 1 mm particles. According to Schubert, most
food powders are non-cohesive only at particle sizes above 100 um. Additionally,
interparticle forces are much stronger in wet powders.

Instant cereal powders present additional difficulties and often have poor instant
properties because of starch swelling. Lump formation often occurs in fine powders,
which have large surface areas, due to quick hydration and gelling of starch on the outer
surface of a lump (Thomas and-Atwell, 1999). Swelling also occurs, clogging pores and
further impeding wetting (Schubert, 1993). This prevents water from penetrating, leaving

a lump of particles with a dry core, known as a “fish eye” (Thomas and Atwell, 1999).

18

Larger particles having less surface area hydrate more slowly, resulting in easier
dispersion. Dispersibility is usually controlled in drum dried starches through the degree
of milling. Powders with larger particle Sizes are easier to disperse, but do not yield
products with as smooth texture (Thomas and Atwell, 1999). In addition to controlling
the degree of milling, agglomeration is another method used to produce starch with good
instant properties (Schubert, 1993).

Settling of particles is related to particle size and the difference in density between the
particle and liquid (Schubert, 1993). The velocity of sedimentation (w) is described by
the Stokes’ law equation w = Ap g 752/18 1], where Ap is the difference in density between
the solid and liquid, g is acceleration due to gravity, xis the particle diameter, and n is the
viscosity of the liquid (Schubert, 1993). Thus, larger particles with higher density,
sediment more quickly. One way to decrease sedimentation is by increasing the viscosity
of the liquid. The viscosity of water can be increased 10-100 fold by stabilizers such as

starch and colloids without adversely affecting sensory attributes (Schubert, 1993).

2.6. Instantization

The process of instantization describes the enhancing of the instant properties of a
powder. This is often necessary because of the difficulty in dispersing fine or
hydrophobic powders in water. Instantization generally falls into either agglomerative or
non-agglomerative processes (Schubert, 1993).

Agglomeration improves instant properties by loosely bonding together individual
particles to make a larger, granular structure. Once the agglomerate is mixed with liquid,

the bonds should then disintegrate so the powder can disperse. This improves wettability

19

and reduces lump formation (Schubert, 1993). Wet agglomeration is done with a
condensing vapor or liquid spray. This results in the formation of solid bridges holding
particles together after drying. Spray drying is another commonly used method for
agglomeration (Schubert, 1993).

Non-agglomerative processes describe methods that do not involve aggregation of
individual particles. These types of processes usually involve either reducing the contact
angle or increasing the solubility of particles (Schubert, I993).

The addition of lecithin is a commonly used method to reduce the contact angle.
Lecithin consists of a mixture of phospholipids, mainly refined from vegetable oils. Plant
sources include cottonseed, corn, sunflower, and rapeseed. However, the majority of
commercially available lecithin comes from the soybean (List, 1989). Lecithin contains
both hydrophilic and hydrophobic portions, allowing it to be surface active (Schmitt,
1995). Thus, lecithin can bind to hydrophobic powders and also improve their wetting by
liquids. Modification of lecithin by hydroxylation is done to further enhance its water
dispersing and emulsifying properties (List, 1989).

Both agglomerative and non-agglomerative processes are often used to produce
instant drink mixes. A typical formula for a protein-based drink may include whey
protein concentrate, fructose and other flavorings, and 2% lecithin (Sander, 1989). The
production of whole powdered milk with the addition of lecithin combined with Spray
drying is an example where both types of instantization processes are used (Pisecky,

1978).

20

2.7. Evaluation of Instant Properties

How instant powders are produced and evaluated are dependent upon many factors
related to its end use. Considerations include the ease of dispersal required along with
the type of liquid that the powder is to be dispersed in (Sander, 1989). The stability of
particles to sedimentation is another factor that depends on the type of product.
Therefore, one set of standard procedures will not be acceptable for all powders.
Evaluation methods need to be practical, simple, and fast to allow for routine analysis.

The International Dairy Federation (IDF) has written a set of standard procedures for
the evaluation of powdered milk instant properties. According to IDF-Standard 87:1979,
wettability is measured by dropping powdered milk on the surface of water in a beaker
and measuring the time it takes for all of the particles to sink below the surface. This
type of method requires that the powder be Spread evenly over the surface of the water.
Various methods of dropping the powder onto the liquid surface have been proposed and
it is recognized that spreading the powder evenly is very difficult (Schubert, 1979). This
method may also be more representative of Sinkability than wettability.

Another method to measure wettability, proposed by Schubert (1993), calls for the
placement of powder into a container with a sieve on the bottom. The container is then
attached to a balance from above and lowered onto the surface of a liquid. The change in
force measures how much liquid is imbibed.

According to IDF Standard 87: 1979, dispersibility is measured by stirring powdered
milk in water for a standard time before allowing it to stand still. A portion of the milk is

then poured through a 150 micron Sieve and the total solids of the filtrate are measured.

21

Schubert (1993) developed a device to remove the middle portion of a drink, which has
been stirred and allowed to stand. The solids content of this portion is then measured
along with the remaining portion, which consists of the top and bottom of the drink. The
degree of dispersion is measured by dividing the solids concentration of the middle
portion by the solids concentration of the remaining portions.

According to Sander (1989), dispersibility can be measured by mixing the powder in a
beaker with a magnetic stirring bar. The mixing is standardized by measuring the height
of the liquid caused by the vortex. The time to disperse the powder is then measured.
Further characterization can include pouring the drink over a sieve and counting the

number of lumps that remain.

22

III Materials and Methods

3.1. Raw Materials

Five different raw materials were extruded:

WE: Soft white wheat flour (Star of the West, Frankenmuth, MI).

E: 96% sofi white wheat flour (dry basis) and 4% deoiled hydroxylated soy lecithin
(Precept 8120, Central Soya, Fort Wayne, IN).

_F: Raw defatted soy flour (N utrisoy 7B, Archer Daniels Midland, Decatur, IL).

S_FL: 96% raw defatted soy flour (dry basis) and 4% deoiled hydroxylated soy lecithin.
_S_-I_3_l§r_1_c_l: 50:50 blend of raw defatted soy flour with raw dehulled full fat soy flakes

(Thumb Oil Seed Producers’ Cooperative, Bad Axe, MI).

3.2. Extrusion

Extrusion was done with an APV co-rotating, twin screw extruder, with a barrel
diameter of 19 mm and barrel length to diameter (LD) ratio of 25:1 (Model MP19T2-25,
APV Baker, Grand Rapids, MI). Heating of the barrel was controlled by electric heating
elements jacketing the barrel. Heating was divided into five zones along the length of the
barrel, with zone one nearest the feed section and zone five nearest the exit die. Moisture
was adjusted to the dough moisture content during extrusion by injection of water into the
barrel using a Brook Crompton E2 Metripump (Hudders Field, England). Dry materials
were fed using a K-TRON K2M twin-screw volumetric feeder (K-TRON, Pittman, NJ).
Die pressure was measured by a pressure transducer (Model EPR3-3M, Dynisco, Sharon,
MA) placed before the die. Torque was measured as % of the total motor power (2000
W).

23

Moisture, screw speed, and temperature were each varied at two levels, for a total of 8
different combinations. The dry material feed rate was kept constant along with the
screw configuration (listed in Table 3.1). The extrusion conditions for WF and WFL are
listed in Table 3.1 and conditions for SF, SF L, and S-Blend are listed in Table 3.2.

Extrusion of each raw material was repeated on a different day.

3.3. Post Extrusion Processing

The extrudates were dried overnight in a food dehydrator (Proctor and Schwartz Inc.,
Philadelphia, PA) at 45-50°C. Powders with different particle size ranges were produced
from the extrudates by milling and sieving. After milling, the powders were stored at

5°C until analysis.

3.3.1. Milling: Wheat Flour (W F) and Wheat Flour + Lecithin (WFL)

The WP and WFL extrudates were progressively ground and sieved to produce
particle sizes of 145-249 um, 93-145 um, and <93 um. Grinding was done using a coffee
grinder (Braun Inc., Woburn, MA) and an impact mill with 1.0 mm and 0.50 mm screens
(Udy Cyclone Mill, Udy Corp., Fort Collins, CO). Sieving was done with a rotary Sifter

(Sampl-Sifter, Great Western Mfg. Co, Leavenworth, KS).

3.3.2. Milling: Defatted Soy Flour (SF) and Defatted Soy Flour + Lecithin (SF L)
The SF and SF L extrudates were progressively ground and sieved to produce particle
sizes of 93-145 um and <93 um. Grinding was done using a coffee grinder and an

impact mill with 1.0 mm and 0.50 mm screens. Sieving was done with a rotary Sifter. A

24

Table 3.1. Extrusion Conditions for Wheat Flour and Wheat Flour + Lecithin

 

 

 

 

 

 

 

 

 

Feed Rate Total Material Screw Speed Barrel Temp. Zones
(kg/hr)‘ Moisture (%) (rpm) I-SQC)”
1.8 25 300 40/50/80/110/120
1.8 25 200 40/50/80/ 1 10/ 120
1.8 35 200 40/50/80/1 10/120
1.8 35 300 40/50/80/1 10/120
1.8 35 300 40/60/90/130/145
1.8 35 200 40/60/90/130/145
1.8 25 200 40/60/90/130/145
1.8 25 300 40/60/90/130/145

 

 

 

 

 

Screw Configuration (feed inlet —) die): 8DT, 7x30°FKP, 7DT, 4x60°FKP, 3x30°RKP, 2DS, 6x60°F KP,
4x30°RKP, 2DS. D = Diameter, T = Twin Lead Screw, S = Single Lead Screw, FKP = Forward Kneading
Paddle, RKP = Reverse Kneading Paddle; 1 Kneading Paddle = 0.2SD; ID = 19 mm

 

' Dry basis.
b Zone 5 = nearest to exit die.

25

 

Table 3.2. Extrusion Conditions for Defatted Soy Flour, Defatted Soy Flour +
Lecithin, and Blend of Defatted Soy Flour with Full Fat Soy Flakes

 

 

 

 

 

 

 

 

 

Feed Rate Total Material Screw Speed Barrel Temp. Zones
(kg/hr)ll Moisture (°/o) (rpm) 1-5 (°C)”
1.8 30 200 40/50/80/1 10/120
1.8 30 300 40/50/80/110/120
1.8 40 300 40/50/80/1 10/120
1.8 40 200 40/50/80/1 10/120
1.8 30 300 40/60/90/130/145
1.8 30 200 40/60/90/130/145
1.8 40 200 40/60/90/130/145
1.8 40 300 40/60/90/130/145

 

 

 

 

 

Screw Configuration (feed inlet —) die): 8DT, 7x30°F KP, 7DT, 4x60°F KP, 3x30°RKP, 2DS, 6x60°FKP,
4x30°RKP, 2DS. D = Diameter, T = Twin Lead Screw, S = Single Lead Screw, FKP = Forward Kneading
Paddle, RKP = Reverse Kneading Paddle; l Kneading Paddle = 0.25D; ID = 19 mm

 

‘ Dry basis.
b Zone 5 = nearest to exit die.

26

 

portion of the <93 um fraction was further ground using a jet mill (Model 00 Jet-O-Mizer

Mill, Fluid Energy Aljet, Telford, PA) to produce a third, finer particle size.

3.3.3. Milling: Blend of Raw Defatted Soy Flour with Full Fat Soy Flakes (S-Blend)

The S-Blend extrudates were ground with an impact mill through a 1.0 mm screen.
They were then sieved through a 93 um screen. Only one particle Size was produced due

to difficulties grinding and Sieving because of the high fat content.

3.4. Analysis
Similar analyses (see below) were performed in duplicate for all of the extruded

powders unless indicated otherwise.

3.4.1. Laser Particle Size Analysis
Mean particle sizes of the milled fractions were determined by laser particle size
analysis with a Beckman-Coulter LS 130 laser diffraction instrument (Beckman, Hialeah,

FL). Samples were suspended in isopropanol and measured for a 60 second time interval.

3.4.2. Specific Mechanical Energy (SME)

The SME in during extrusion in Watt hours/kg was calculated according to the
equation of Brent et al. (I 997):
SME = [actual screw rpm / rated screw rpm] x [% torque / 100] x [motor power in W /
feed rate in kg/hr] [1]

rated screw Speed = 500 rpm, motor power = 2000 W

27

3.4.3. Moisture
Moisture content of the milled extrudates was determined by drying 1 g of powder in

an air oven (Model 737F, Fisher Scientific Co., Pittsburgh, PA) at 130°C for 1 hour.

3.4.4. Water Absorption Index (WAI) and Water Solubility Index (WSI)

The water absorption and water solubility indexes were measured according to the
procedure of J in et al. (1995) with some modifications. Two grams of powder were
mixed with 20 ml of distilled water in 30 ml round bottom centrifuge tubes. The tubes
were allowed to Sit for 10 minutes and were inverted 3 times each at 5 and 10 minutes.
After 10 minutes, the suspensions were centrifuged (Model J2-21M, Beckman
Instruments Inc., Fullerton, CA) at 1000 x g for 15 minutes. The supernatant was then
collected and the WAI was calculated as:

WAI = weight of pellet / sample weight (dry basis) [2]
The water solubility index was measured by drying the supernatant in an air oven
overnight at 60°C. WSI was calculated as:

WSI = weight of dried supernatant / sample weight (dry basis) [3]

3.4.5. Rapid Visco Analyzer Viscosity of WF and WFL

Pasting properties were determined using a Rapid Visco Analyzer (RVA Model 4,
Newport Scientific Pty. Ltd., NSW, Australia). A 3.5 g sample of powder (14% m.b.)
was mixed with 25 ml of distilled water by shaking to ensure that no lumps were present

before analysis. The temperature profile and analysis definitions are listed in Table 3.3.

28

Table 3.3. Rapid Visco Analyzer Temperature Profile and Analysis Definitions

 

 

 

 

 

 

 

Temperature (°C) Time (minutes)
25 0
25 2
95 7
95 10
25 15
25 20

 

Paddle Speed (rpm): 160

 

Analysis Definitions‘

 

 

 

 

 

 

 

 

 

Cold Viscosity Peak Viscosity at 0.2 — 2 Minutes
Hot Paste Viscosity Peak Viscosity at 2 — 10 Minutes
Hold Trough Viscosity at 10 — 19 Minutes
Breakdown Hot Paste Viscosity - Trough Viscosity
Final Viscosity Viscosity at End of Test
Setback Final Viscosity — Trough Viscosity
Peak Time Time of Peak Viscosity (Minutes)

 

‘ Viscosity in Rapid Visco Units (RVU).

29

 

3.4.6. Dispersibility

Dispersibility measured the % total solids that passed through a Sieve after mixing the
powder in distilled water. Mixing was done with an overhead mixer (Mixer Head Model
50003-30, Cole-Farmer, Vernon Hills, VA) by a 4 x 2 x 0.1 cm paddle set at 200 rpm.
The paddle was centered with its length horizontal, 1 cm below the liquid surface in a
250 ml beaker containing 100 ml of distilled water at 21°C. Mixing time was set at 1
minute and 10 seconds. Five grams of powder were poured into the beaker during the
first 10 seconds of mixing. After mixing, the suspension was immediately poured
through a wire mesh sieve. The WF and WFL 145-249 um and 93-145 um powders and
the 93-145 um SF and SFL powders were poured through a sieve with 1.59 mm
openings. The <93 um particle Size fractions for WF, WFL, SF, and SF L were poured
through a sieve with 471 um openings. The SF and SFL jet milled powders were also
poured through a sieve with 471 um openings. The S-Blend powders were poured
through a 1.59 mm-mesh sieve. For each sample tested, the suspensions were poured
through the indicated sieve and into a 250 ml separatory funnel with an enlarged stopcock
to aid the passage of the suspension. Twenty ml of the suspension were immediately
drained and 4 g were measured for solids content by drying in an air oven (Model 73 7F,
Fisher Scientific Co., Pittsburgh, PA) for 2 hours at 110°C. Dispersibility was calculated
as the solids concentration of the suspension passing through the sieve, divided by the
solids concentration possible if all of the particles were evenly dispersed in the
suspension.
% Dispersibility = [bzo / (g sample dry basis) / (g sample d.b. + 100 g water)] x 100 [4]

bzo = solids concentration (w/w) of drained 20 ml

30

3.4.7. Stability

Stability measured the amount of sedimentation in the remaining approximately 80 ml
of suspension in the separatory funnel. This remaining suspension was allowed to settle
undisturbed for 2 minutes. Fifty ml was then drained and discarded. The remaining
portion, consisting of the upper layer of the suspension, was then drained and a 4 g
aliquot was measured for solids content. This was then divided by the solids
concentration of the respective 20 ml aliquot initially drained off to measure
dispersibility.
% Stability = [t30 / 1320] x 100 [5]

I30 = solids concentration (w/w) of drained upper layer

3.4.8. Trypsin Inhibitor Inactivation of SF, SFL, and S-Blend

Trypsin inhibitor inactivation (TII) was measured according to Hamerstrand et al. (1981).
This assay is based on the increase in absorbance at 410 nm when trypsin (Type 1, from
Bovine Pancreas, Sigma Chemical Co., St. Louis, MO) is incubated with the substrate
benzoyl-DL arginine (BAPA) (Sigma Chemical Co., St. Louis, MO). The presence of TI
results in a decrease in absorbance. The <93 um particle Size fraction was used for the
assay. S-Blend powders were first defatted with 10 ml of hexane per 1 g of powder. A
dilution factor of 1/2500 was used for the raw samples and a factor of 1/156.3 was used
for the extruded samples. The percent inactivation of trypsin inhibitor for the extruded

samples was calculated as:

% TII = {1- [(mg TI / g extruded sample) / (mg TI / g raw sample)]} x 100 [6]

31

3.5. Blending of Different Powders

After analyses (Section 2.4) of the powders, 50:50 blends were made between various
wheat and soy powders. The powders chosen for blending represented a range of instant
properties related to their use for instant beverage bases. These blends were measured for

viscosity, dispersibility, and stability.

3.6. Statistics

Statistical differences between extrusion conditions, particle sizes, and raw materials
were calculated by the mixed method with SAS version 8.01 (SAS Institute, Cary, NC).
Correlations between variables were calculated using Microsoft Excel 97 (Microsoft,

Redmond, WA)

32

IV Results and Discussion

4.1. Extrusion of Wheat Flour (WF) and Wheat Flour + Lecithin (WFL)

After extrusion of the WP and WFL, the extrudates were dried overnight and ground
to produce powders with three particle sizes: 145—249 mm, 93-145 um, <93 um. The
following are results along with analysis and discussion regarding the instant properties

measured for all extruded wheat flour samples.

4.1.1. Particle Size Analysis

The mean particle sizes of the 145-249 um fraction produced by milling and sieving of
the WF and WFL extrudates ranged from approximately 210 um to 230 pm as measured
by laser particle Size analysis (Table 4.1). The measured Size of the 93-145 um Sieve
fraction ranged from 130-140 um and the <93 um Sieve fraction ranged from 50-60 um.
It should be noted that if any swelling occurred when the powders were dispersed in
isopropanol for the laser analysis, the particle size measurements would be inflated.
Based on appearance and feel, powders of the two largest particle Sizes were still

somewhat granular, while the powder below 93 um was finer.

4.1.2. Specific Mechanical Energy (SME) During Extrusion of Wheat Flour

The SME was Significantly affected by changes in moisture, screw speed, and
temperature during extrusion (p<0.05, Table 4.2). The different combinations of these
variables resulted in a range of SME from 83i1 Wh/kg to 190i7 Wh/kg. At lower
moisture content and higher screw speed, there was a significant increase in the

mechanical energy input into the extruded materials. The effect of temperature on the

33

Table 4.1. Laser Particle Size Analysis of Extruded and Milled Wheat Flour (W F)
and Wheat Flour + Lecithin (WFL) Powders

 

 

 

 

 

 

 

 

145 - 249 um 93 - 145 pm <93 pm
Sieve Fraction Sieve Fraction Sieve Fraction
31.73;? :2: Mean Particle Size (um)
25/300Iii5 rep I 2'0 I33 59
25/300Ii55 rep 2 2 '4 129 51
2530371216 rep I 2'9 I37 52
2530371216 rep 2 231 I42 57

 

 

 

 

 

Raw Wheat Flour = 38 um

 

' M = moisture (%), S = screw Speed (rpm), T = zone 5 temperature (°C). Each repetition extruded on a
different day.

34

 

la

.;

TEL Li L;

L

>\1~F\.\KT\\

Table 4.2. Specific Mechanical Energy (SME) and Statistical Relationships During

Extrusion of Wheat Flour (WF) and Wheat Flour + Lecithin (W F L)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions WF [ WFL
M/S/T SME (Wh/kg)
25/300/120 190r7 l65i7
25/200/120 144i] 13345
35/200/120 7&8 753:1
35/300/120 I 18il 1 121:5
25/300/145 173r4 1503.0
25/200/145 120r1 104il
35/200/ I 45 83:1 6940
35/300/145 1 18i3 102r3
WF I WFL
F value
M 608.94* 1371.74*
3 349.00* 782.09*
T 1527* 127.39*
M*S 633* 2.34
M*T 2586* 30.51 *
S*T 0.07 3.47
M*S*T 1.52 1246*

 

 

 

Least Squares Means

 

M: 250/0 - 350/0

Estimate of Difference (Wh/kg)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

200 rpm, 120°C 67* 51*
200 rpm, 145°C 37* 35*
300 rpm, 120°C 73* 53*
300 rpm, 145°C 55* 48*
S: 200 rpm - 300 rpm
25%, 120°C -46* -32*
25%, 145°C -53* -46*
35%, 120°C -40* -37*
35%, 145°C -36* -33
T: 120°C - 145°C
25%, 200 rpm 25* 29*
25%, 300 rpm 18 15*
35%, 200 rpm -5 5
35%, 300 rpm -1 10

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured beteween repetitions of extrusion.

* Significant at p<0.05.

35

 

SME was not as great as that of moisture and screw speed, but an increase in barrel
temperature at 25% moisture and a screw speed of 200 rpm led to a significant decrease
in the SME that was not seen with an increase in temperature at 35% moisture with the
same screw speed.

Numerous studies have found similar effects of extrusion conditions on SME. Whalen
et al. (1997) reported a decrease in torque as temperature increased, especially at 15-25%
moisture, the lower range used in their experiment. Increasing the barrel temperature has
also been reported by others to reduce the SME (Bhattacharya and Hanna, 1987; Bruin et
al., 1978). Brummer et al. (2002) reported SME values that increased from below 100
Wh/kg to over 300 Wh/kg as the moisture content of extruded corn starch decreased from
32% to 13%. Bhattacharya et al. (1999) and Della Valle et al. (1989) also found that the
SME decreased as moisture content increased.

These results can be explained by the effects of moisture and temperature on dough
rheology. As either moisture or temperature increase, dough viscosity decreases (Harper,
1981). A dough with lower viscosity requires leSS motor power to extrude, resulting in a
lower SME.

Increasing the screw speed has been reported to both increase (Bhattacharya et al.,
1999; Della Valle et al., 1989) and decrease (Van Zuilichem et al., 1983) the SME. As
the screw speed increases, the degree of fill in the barrel decreases and the viscosity of
the dough may also decrease, leading to a lowering of the SME (Frame, 1994). However,
in the present experiments the increase in screw Speed was large enough to overcome the

slight reduction in torque from the motor, resulting in a higher SME (Appendix Ila).

36

4.1.3. SME During Extrusion of Wheat Flour with Lecithin

The different extrusion variables affected the SME of the wheat flour + lecithin
formulation in a similar manner to their effect on SME of wheat flour alone. Lower
temperature, lower moisture, and higher screw speed all had significant effects of
increasing the SME (p<0.05, Table 4.2). However, adding lecithin to the wheat flour
resulted in an overall decrease in the SME (p<0.05, Table 4.3). This most likely occurred
due to lubrication provided by the lipids in the WFL sample during extrusion. The

replacement of 4% of the wheat flour may have also added to the reduction of the SME.

4.1.4. Water Absorption Index (WAI) of Extruded Wheat Flour

Raw wheat flour had a WAI of 1.91 g/ g while the extruded WF powders had WAI
values that ranged from 6.3:t0.21 g/g to 8.48i0.14 g/g (Table 4.4). The WAI was
dependent on both the extrusion conditions and powder particle size. Extrusion moisture,
screw speed, and barrel temperature all had Significant effects on the WAI (p<0.05, Table
4.4). Extrusion at higher moisture and temperature led to an increase in the WAI, but
increasing the screw speed led to a decrease in the WAI. There was also a significant
decrease in the WAI as the powder particle size decreased (p<0.05, Table 4.4, Figure
4.1).

These effects of different extrusion conditions on the WAI are generally consistent
with other published results (Anderson et al., 1969; Gomez and Aguilera, 1983; J in et al.,
1995; Wen et al., 1990). During extrusion, the WAI increases up to a maximum,
corresponding to an increase in the degree of starch damage. As greater dextrinization

occurs, the WAI then decreases (Colonna et al., 1989). According to Gomez and

37

Table 4.3. Comparison”l Between Properties of Extruded and Milled Wheat Flour
(WF) and Wheat Flour + Lecithin (WFL) Powders

Estimate of Difference

WF - WFL
SME 14*

WAI 043*
WSI 0095*

Cold 16*
Hot Paste V -33*

0/o -l3.9*
Sta °/o 9.9*

 

‘ Comparison done for <93 um particle size.

SME = specific mechanical energy, WAI = water absorption index, WSI = water solubility index, RVU =
Rapid Visco Units.
* Significant at p<0.05.

38

Table 4.4. Water Absorption Index (W AI) and Statistical Relationships of Extruded
and Milled Wheat Flour Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
EXtms'g/S’rnd'mns 145-249 um 93-145 pm <93 pm
WAflg/g)
25/300/120 7.48:t0.26 7.09:t0. I 6 6.30:I:O.2I
25/200/120 8. I 53:0.02 7.79i0.03 7.1 liO.25
35/200/ 120 NT NT 7.05:1:002
35/300/ 120 NT NT 6.94:t0.11
25/300/145 7.64:1:014 7.32i0.20 7.07:0.13
25/200/145 8.235006 7.87:t0.07 7.67i0.06
35/200/145 8.48i0.14 8.34i0.26 7.66i0.ll
35/300/145 7.76i0.07 7.60i0.08 7.10:1:008
Raw Wheat Flour = 1.91 g/g
Particle Size
145—249pm ] 93-145 pm I <93 pm
F value
M 9.42* 26.91 * 4.5
S 6913* 61.34* 54.73*
T NT NT 5579*
M*S 0.33 0.38 687*
M*T NT NT 3.92
S*T NT NT 0.76
M*S*T NT NT 5.45*
Least Squares Means
M: 25% - 35% Estimate of Difference (g!
200 rpm, 120°C 0.06
200 rpm, 145°C -0.30* -0.51* 0.01
300 rpm, 120°C -0.64*
300 rpm, 145°C -0-20 -0.40* -0.03
S: 200 rpm - 300 rpm
25%, 120°C 0.81*
25%, 145°C 0.63”“ 0.63* 0.60*
35%, 120°C 0.11
35%, 145°C 0.72”“ 074" 0.56*
T: 120°C - 145°C
25%, 200 rpm NT NT -0.56*
25%, 300 rpm NT NT -0.77*
35%, 200 rpm NT NT -0.61*
35%, 300 rpm NT NT -0.16
ParticleSize C-B C-A B-A
Estimate (g/g) -0.56* -0.85* -O.29

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
NT = not tested due to the inability to form a firm pellet after centrifugation.

A = I45-249um particle size, B = 93-145um particle size, C = <93um particle size.

* Significant at p<0.05.

39

 

 

 

 

 

 

l r: 145-249 = -0.81, 93145 = -0.90, <93 = -0.62 0 145-249
’x93-145
x
8'5 - :52:
x : o
; 8 )fi 1
a I W o I
I § .
r g 7 5 I x! k
3 I .II I X I X
7 I
‘ I I I X
6.5 I
I
5 .
50 70 90 110 130 150 170 190

site (Wit/kg)

Figure 4.1. Relationship between the specific mechanical energy (SME) and the
water absorption index (WAI) for extruded and milled wheat flour powders.

40

Aguilera (1983), the WAI is dependent on the binding of water by hydrophilic groups
and on the ability of macromolecules to form gels. More highly dextrinized starches
having lower water absorption capabilities. Decreased WAI measurements were also
noted by Wen et al. (1990) to be consistent with an increase in starch fragmentation.
Colonna et al. (1984) compared the effects of drum drying versus extrusion on wheat
starch. They found that extrusion led to greater depolymerization of amylose, and
especially amylopectin, than did drum drying. In a subsequent study this group
compared the water swelling capability of drum-dried and extruded wheat starch
(Doublier et al., 1986). They found that drum-dried starches had higher cold water
swelling capacities than the extruded starches. This was attributed to low water
accessibility of extruded starches due to their compact structure.

The SME is an important variable to consider since its increase has been correlated
with a greater degree of starch fragmentation and conversion (Brummer et al., 2002;
Della Valle et al., 1989; Klingler et al., 1986; Kokini, 1993; Schuchmann and Danner,
2000). Brummer et al. (2002) measured the molecular weight of corn starch after
extrusion by Size exclusion chromatography and found an exponential decrease in weight
as the SME increased. The role of mechanical energy input is supported by the negative
correlation between the SME and WAI for all three particle size fractions (Figure 4.1).
The conditions of moisture, screw Speed, and temperature that increased the WAI,
decreased the SME. That starch fragmentation is reduced at lower SME values is also

supported by the water solubility index data (see section 4.1.6).

41

4.1.5. WAI of Extruded Wheat Flour with Lecithin

The WA] and water solubility index were not measured for some milled and sieved
extrudates in the 145-249 um and 93-145 um particle size fractions because a firm pellet
could not be formed. At <93 um, there were only small differences in the WAIS of
samples produced under different extrusion conditions (Table 4.5). There was a
Significant interaction between moisture and temperature (p<0.05). At 145°C, the WAI

increased as moisture increased, but at 120°C, an increase in moisture led to a decrease in
the WAI.

These results may represent conditions which span the increase of the WAI up to a
maximum and then its decrease. At 145°C the combination of thermal and mechanical
energy may have fully cooked the starch. The extrusion conditions would have then been
severe enough to where the WAI was beyond its maximum. Thus, increasing the
moisture content and the resultant decrease in the SME would result in less degradation
of the starch, and a higher WAI. At 120°C. the starch may not have been fully cooked
due to the presence of lecithin and its lowering effect on the SME. Under these
conditions, the maximal WAI may not have been reached. Conditions which reduced the
amount of starch damage would then have the opposite effect on the WAI as they would
for a fully cooked starch. This would explain why an increase in feed moisture in this
case led to a decrease in the WAI. This hypothesis is supported by RVA data presented
in section 4.1.9.

The WAI values of Wheat flour + lecithin were generally lower than that of wheat

alone, although there was not a large difference (Table 4.3).

42

Table 4.5. Water Absorption Index (WAI),Water Solubility Index (W SI), and
Statistical Relationships of Extruded and Milled Wheat Flour + Lecithin Powders
(<93 um Particle Size Fraction)a

 

Extrusion Conditions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

M/S/T WAI (g/g) WSI (g/g)
25/300/120 6.63:0.07 0122:0001
25/200/120 6.61 i004 0086:0006
35/200/120 6.48:0.24 0047i0001
35/300/120 6.41:019 00661-0006
25/300/145 6.44:0.08 0106i0.004
25/200/145 6.6li0.21 0066:0007
35/200/145 7.19:026 00521-0002
35/300/145 7.12:0.03 0079i0.001

Raw Wheat Flour + Lecithin Raw Wheat Flour + Lecithin
: 205 gig = 0.073 gig

WAI WSI

F value F value

M 7.69* 421.28*

S 0.75 333.52*
T 1378* 1.67

M*S 0 1045*

M*T 2395* 48.29*

S*T 0.31 l 1.52*
M*S*T 0.39 0.94

 

 

Least Simares Means

 

M: 25% - 35%

Estimate of Difference (g/g)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

200 rpm, 120°C 0.13 0039*
200 rpm, 145°C -O.58 0015*
300 rpm, 120°C 0.23 0046*
300 rpm, 145°C -0.69* 0028*
S: 200 rpm - 300 rpm
25%, 120°C -0.03 -0.027*
25%, 145°C 0.17 -0.040*
35%, 120°C 0.08 -0.020*
35%, 145°C 0.07 -0.027*
T: 120°C - 145°C
25%, 200 rpm 0.00 0020*
25%, 300 rpm 0.20 0.006
35%, 200 rpm -071* -0.005
35%, 300 rpm 072* -0.0l3

 

 

’ 145-249 um and 93-145 um fractions did not form firm pellets after centrifugation and were not

measured.

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.

* Significant at p<0.05.

43

 

4.1.6. Water Solubility Index (WSI) of Extruded Wheat Flour

The WSI of the extruded powders ranged from about 0.10 g/g to 0.30 g/g (Table 4.6).
Extrusion feed moisture had the most Significant effect on the WSI. The WSI increased
when the moisture was reduced from 35% to 25% at the 93-145 um and <93 um particle
size ranges (p<0.05). Increasing the screw Speed also led to an increase in the WSI for
the 93-145 and <93um particle sizes (p<0.05). Temperature did not have a significant
effect.

Like water absorption, water solubility is also related to conditions which favor the
degradation and fragmentation of starch (J in et al., 1995; Gomez and Aguilera, 1983).
However, the inverse relationship exists to water absorption, as increased fragmentation
leads to greater water solubility. According to Diosady et al. (1985) and Davidson et al.
(I 984), the most important factor regarding the fragmentation of starch is shear stress.
Various studies have shown that increases in temperature or moisture during extrusion
lead to decreases in starch fragmentation, while increases in screw speed lead to
increasing amounts of fragmentation (Davidson etal., 1984; Wen et al., 1990). Davidson
et al. (1984) states that for polymers, the main effect of increasing temperature is a
decrease in viscosity, which results in a decrease in Shear stress. An increase in feed
moisture would also have the effect of lowering viscosity, while increasing the screw
speed increases the shear rate (Harper, 1981).

The conditions of moisture, temperature and screw speed that led to an increase in
SME, also resulted in an increased WSI value. The SME was positively correlated with
the WSI for all three particle size fractions (Figure 4.2). There were not significant

differences in WSI among the particle sizes. This may be due in part to missing data; at

44

Table 4.6. Water Solubility Index (W SI) and Statistical Relationships of Extruded
and Milled Wheat Flour Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
Ex‘ms'g‘lgflrndmms 145-249 pm 93-145 pm <93 pm
WSI (g/g)
25/300/120 0.173i0.008 0.233i0004 0.3022t0.028
25/200/120 0.141:t0.010 0175:0018 022330016
35/200/120 NT NT 0.083i0. 161
35/300/120 NT NT 0.1 152t0.021
25/300/145 0.157i0020 019650009 0.225i0.016
25/200/145 0.119:t0.006 0.157i0.012 0177i0.031
35/200/145 012510.008 0.101i0.020 0120210024
35/300/145 0.128zt0.033 01131-0004 0131:1:0004
Raw Wheat Flour = 0.020 g/g
Particle Size
145249 pm 1 93-145 pm ] <93 pm
F value
M 0.98 5469* 145.69*
S 0.99 706* 1835*
T NT NT 3.02
M*S 1.24 2.67 4.48
M*T NT NT 1989*
S*T NT NT 1.72
M*S*T NT NT 0.06
Least Squares Means
M: 25% - 35% Estimate of Difference (g/g
200 rpm, 120°C 0141*
200 ijm, 145°C -0.620 0065* 0,057
300 rpm, 120°C 0188*
300 rpm, 145°C 0-037 0102* 0.094
S: 200 rpm - 300 rpm
25%, 120°C -0079
25%, 145°C -0-035 -0.048* -0.048
35%, 120°C -0.032
35%, 145°C 0.622 -0.012 .001 |
T: 120°C - 145°C
25%, 200 rpm NT NT 0.046
25%, 300 rpm NT NT 0.077
35%, 200 rpm NT NT 0038
35%, 300 mm NT NT 0017
Particle Size C - B C - A B - A
Estimate (g/g) 0.010 0.031 0.023

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.

NT = not tested due to the inability to form a firm pellet after centrifugation.

A = 145-249um particle size, B = 93-145um particle size, C = <93um particle Size.

* Significant at p<0.05.

45

 

 

 

 

 

0 145-249 r: 145-249 = 0.77, 93-145 = 0.93, <93 = 0.92 l
03 x93-145 'l
I [E433 I I
0.25
I X
. :5 I . X
I ‘3', 0.2 I X
m X X
i 3 o .
x C
0.15 of o .
I ' .l O
I X X:
0.1 I X
I
l 0.05 4.
50 70 90 110 130 150 170 190

SME (Whlkg)

Figure 4.2. Relationship between the specific mechanical energy (SME) and the
water solubility index (W SI) for extruded and milled wheat flour powders.

46

the particle sizes of 145-249 um and 93-145 um, solid pellets would not form during the
measurement of the WSI of powders from some of the extrusion treatments, resulting in
unobtainable data. However, it appeared that water solubility was higher as particle size

decreased at the upper range of the WSI measured.

4.1.7. WSI of Extruded Wheat Flour with Lecithin

The WSI values of powders made from extruded WF L were mainly affected by the
feed moisture and extruder screw speed. Lower moisture content and higher screw speed
resulted in a higher WSI (Table 4.5). Although the general effects of the extrusion
conditions were similar to the results for wheat flour extruded alone, the WSI was
significantly lower when WF was extruded with lecithin (p<0.05, Table 4.3). The WSI
for the WFL only reached a maximum of 0122:0001 g/g compared to 030210028 g/g
for the WF (Tables 4.5, 4.6).

A decrease in the water solubility of extruded cereals when lipids are added has been
attributed to two different effects. Colonna et al. (1983) found that extrusion of cassava
starch with added monoglycerides and triglycerides resulted in less starch degradation.
This was attributed to the lubricant effect of lipids inside of the extruder. Experiments by
Schweizer et al. (1986) and Mercier et al. (1980) found that lipid complexation inside of
the amylose helix decreased the water solubility of starches. However, Mercier et al.
(1980) also found that only free fatty acids and monoglycerides could form the complex.

Fats such as soy lecithin were not able to complex because they were too large to fit into

the helix.

47

4.1.8. Rapid Visco Analyzer Viscosity of Extruded Wheat Flour

Depending on extrusion conditions and particle size, cold viscosity values of WF
powder samples ranged from approximately 55 RVU to over 200 RVU (Table 4.7).
These viscosities were much greater than that of raw wheat flour at 25°C. Decreasing the
extrusion moisture and temperature led to decreases in cold viscosity (p<0.05). However,
at the lower screw speed, the cold viscosity was Slightly higher. These results were also
inversely correlated with the SME, as greater mechanical energy input was related to a
decrease in the cold viscosity (Figure 4.3).

According to Whalen et al. (1997), cold viscosity is due to the presence of swollen
granules and non-granular starch in the form of high molecular weight polymers and
dextrins. Drum drying, which involves much less shear than extrusion cooking, was
found to produce much higher viscosity starch suspensions than extruded ones (Doublier
et al., 1986). The decrease in viscosity after extrusion was attributed to lower swelling
capacity and lower molecular size of amylose and amylopectin in extruded starches.
Additional studies have also related increases in the severity of extrusion conditions and
decreases in cold viscosity (Whalen et al., 1997; Gomez and Aguilera, 1983).
Bhattaycharya et al. (1999) was able to correlate the SME and cold viscosity for an
extruded potato and wheat flour blend. A negative correlation was found in their study as
well as in the current study. The positive correlation of the WSI and negative correlation
of the cold viscosity with the SME provides additional evidence of the relationship
between increased starch breakdown and lower cold viscosity (Figures 4.2 and 4.3

Powder particle Size of the extruded WF samples was also a significant factor

affecting the RVA viscosity (p<0.05, Table 4.7, Figure 4.3). The <93 urn fraction had

48

Table 4.7. Cold Viscosity and Statistical Relationships of Extruded and Milled

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Wheat Flour Powders
Particle Size
145249 pm F 93445 pm <93 pm
Extrusion Conditions Cold Viscosity (RVU)
M/S/T
25/300/120 108:5 I 10:17 54:4
25/200/120 121:10 172:0 65:6
35/200/120 148:7 208:3 143:17
35/300/120 153:2 196:79 133:28
25/300/145 135:20 161:3 94:2
25/200/145 152:29 229:25 1 1 1:2
35/200/145 213:23 244:37 156:2
35/300/145 206:9 218237 132128
Raw Wheat Flour = 3 RVU
Particle Size
145249 pm [ 93-145 pm <93 pm
F value
M 6693* 762* 9122*
S 1.63* 5.76* 605*
T 5043* 5.67* 1506*
M*S 1.18 1.68 0.06
M*T 584* 0.52 8.45*
S*T 0.40 0.08 0.61
M*S*T 0.06 0.01 0.10
Least Squares Means
M: 25% - 35% Estimate of Difference (RVU)
200 rpm, 120°C -27 -37 -78*
200 rpm, 145°C -61* -15 -45
300 rpm, 120°C -44 -86 -79*
300 rpm, 145°C -71 * -57 -33
S: 200 rpm - 300 rpm
25%, 120°C 12 62 II
25%, 145°C 18 68 17
35%, 120°C -4 I3 10
35%, 145°C 7 26 24
T: 120°C - 145°C
25%, 200 rpm -32 -58 -46
25%, 300 rpm -27 -51 —40
35%, 200 rpm -65* -36 -13
35%, 300 rpm -54* -22 1
Particle Size C - B C - A B - A
Estimate (RVU) -81* -43* 38*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

 

Viscosity measured in Rapid Visco Units (RVU).

Standard deviation measured between powders produced from repetition of extrusion.
A = 145-249um particle size, B = 93-145um particle size, C = <93um particle size.

* Significant at p<0.05.

49

 

“300 4----.. WW

 

 

 

x 6145-2491
. , ll<93
A x .. W -
3 X OX
! E 200 X . q
I 9? ' >n< x
1 § 150 a ' .I 6
| .2 .
. > . o x .
g 100 I I . X
‘ O
I 50 f ' I
I 0 r: 145-249 = 065. 93-145 = -0.75. “-193 = -0.87
‘ 50 70 90 1 10 130 150 170 190

SME (Whlkg)

Figure 4.3. Relationship between the specific mechanical energy (SME) and cold
viscosity for extruded and milled wheat flour powders.

50

significantly lower cold viscosity than the two larger particle Size fractions. Similar
results were reported in a study by Becker et al. (2001), who examined the effects of
milling on RVA viscosity profiles. Smaller particle sizes produced by grinding with
different mills and sieving resulted in lower RVA viscosities for extruded corn and wheat
pellets.

In addition to the peak viscosity, particle size also had an effect on the hydration of the
powders. As shown in Figure 4.4, the time to reach peak viscosity increased as the
particle size increased. The peak viscosity of the <93 um particle size WF powder
typically occurred at the start of analysis. while the 145-249 um fraction took more than
2 minutes to reach peak viscosity. This would explain why the 93-145 um fraction had a
higher cold viscosity than the 145-249 um fraction (Figure 4.3). After 2 minutes, the
heating cycle of the RVA began and the peak viscosity at this time was recorded as the
hot paste viscosity. However, this was not a true gelatinization peak that would occur
when heating raw starch. Once the 145-249 um fraction was fully hydrated, its viscosity
increased and was similar to the 93-145 um fraction (Figure 4.5). The lack of a

gelatinization peak when comparing the cold and hot paste viscosities, also indicates that

the powders were fully cooked (Tables 4.7, 4.8).

4.1.9. Rapid Visco Analyzer Viscosity of Extruded Wheat Flour with Lecithin
When examining RVA viscosities of powders from extruded WFL, a significant

interaction was noted between feed moisture and extruder temperature (p<0.05, Table

4.9). At 35% moisture, increasing the temperature to 145°C caused the cold viscosity to

increase while at 25% moisture, the effect of temperature was not significant. Increasing

51

 

180

160 _. \93-145 um
I \
140 - l \.-“-_
120 . I, \ 145-249 IJm

 

100 - h, \

Viscosity (RVU)

 

 

 

 

 

Time (minutes)

Figure 4.4. Representative Rapid Visco Analyzer pasting curve of extruded and
milled wheat flour powders. Viscosity in Rapid Visco Units (RVU).

52

Table 4.8. Hot Paste Viscosity and Statistical Relationships of Extruded and Milled

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Wheat Flour Powders
Particle Size
145249 pm | 93-145 pm | <93 pm
Extrusm“ commons Hot Paste Viscosity (RVU)
M/S/T
25/300/120 126:1 [02:12 47:2
25/200/120 147:9 148:4 55:2
35/200/120 261:1 210:3 139:3
35/300/120 210:16 196:79 121:8
25/300/145 152:30 151:2 84:5
25/200/145 180:31 213:19 97:7
35/200/145 251:21 243:36 141:7
35/300/145 234:6 213:33 121:22
Raw Wheat Flour = 226 RVU
Particle Size
145249 pm ] 93-145 pm 1 <93 pm
F value
M 105.24* 13.44* 168.28*
S 11.73* 5.10 1035*
T 4.43 5.85 19.77*
M*S 0.34 0.89 0.79
M*T 1.74 0.84 17.47*
S*T 0.67 0.23 0.15
M*S*T 1.40 0 0.04
Least Squares Means
M: 25% - 35% Estimate of Difference (RVU)
200 rpm, 120°C -1 14* -61 -84*
200 rpm, 145°C ~71* -31 -44*
300 rpm, 120°C -84* -94 -74*
300 rpm, 145°C -81* -62 -37*
S: 200 rpm - 300 rpm
25%, 120°C 21 46 8
25%, 145°C 27 62 13
35%,120°C 51 14 18
35%, 145°C 17 30 20
T: 120°C - 145°C
25%, 200 rpm -32 -64 -42*
25%, 300 rpm -26 -49 -37*
35%, 200 rpm 10 -34 -2
35%, 300 rpm -24 -17 0
Particle Size C - B C - A B - A
Estimate (RVU) -84* -95* -11

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Viscosity measured in Rapid Visco Units (RVU).
Standard deviation measured between powders produced from repetition of extrusion.

A = 145-249um particle size, B = 93-145um particle size, C = <93um particle size.

* Significant at p<0.05.

53

 

300 ......_....i 7----2: _,_.., . ..H......:.............. 2.2--..

 

 

 

 

 

0 145153;:
e 9 X -
z 250 o X .X93 145
:5 - I... BE? _
I E 200 — x xx
. E ‘8
I 3 ° ’
1 8 150+ . ' 1‘ xx
. '- I
. E " F ' .
1 (D
i a 100 1 x x
I ‘6 .I
1 :1:
a 50 I‘ I .
0 r: l45-249=-0.90, 93-145=-080. <93=-0.88
50 70 90 110 130 150 170 190

SME (Whlkg)

 

Figure 4.5. Relationship between the specific mechanical energy (SME) and hot
paste viscosity for extruded and milled wheat flour powders.

54

Table 4.9. Cold Viscosity and Statistical Relationships of Extruded and Milled
Wheat Flour + Lecithin Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
145-249 pm 1 93-145 um <93 pm
E“'“s"l"'1'/§;’rnd"'°"s Cold Viscosity (RVU)
25/300/120 136:7 241:8 106:3
25/200/ 120 128:6 239:0 1 18:19
35/200/120 71:1 160:12 52:4
35/300/ 120 90:1 168: 11 62:7
25/300/145 127:6 235:3 101:1
25/200/145 122:6 233:21 85:9
35/200/145 207:2 216:8 1 17:27
35/300/145 177:11 166:5 121:16
Raw Wheat Flour + Lecithin = 4 RVU
Particle Size
145-249 pm L 93-145 pm <93 um
F value
M 729* 13428” 4.38
S 0.01 3.58 0.41
T 32121“ 4.15 9.35
M*S 4.39 5.01 0.10
M*T 41642" 1050* 3462*
S*T 1990* 789* 0.59
M*S*T 15.92" 7.46* 1.50
Least Squares Means
M: 25% - 35% Estimate of Difference LRVU)
200 rpm, 120°C 57* 79* 66*
200 rpm, 145°C -85* l7 -33
30mm, 120°C 46* 73* 45
300 rpm, 145°C -50* 68* -20
S: 200 rpm - 300 rpm
25%, 120°C -8 -2 12
25%, 145°C -5 '1 '16
35%, 120°C ~19 -7 -10
35%, 145°C 30* 50* -3
T: 120°C - 145°C
25%, 200 rpm 6 6 33
25%, 300 rpm 7 7 6
35%, 200 rpm -137* -55* -65*
35%, 300 rpm -87* 1 -59*
Particle Size C - B C - A B - A
Estimate (RVU) -1 12* -37* 75*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Viscosity measured in Rapid Visco Units (RVU).
Standard deviation measured between powders produced from repetition of extrusion.
A = 145-249nm particle size, B = 93-145um particle size, C = <93um particle size.

* Significant at p<0.05.

55

 

the moisture from 25% to 35% at 120°C resulted in a decrease in the cold viscosity for
the 145-249 um and <93 pm particle sizes. At 145°C, there was an increase in the cold
viscosity as feed moisture increased.

These trends were similar to those observed for the WAI and changes in cold viscosity
may be explained by similar reasons. The decrease in viscosity at 120°C when moisture
was raised from 25% to 35% may occur because the starch was not fully cooked at this
point. The decrease in SME as moisture increased would result in less mechanical
breakdown of the starch. If mechanical energy input was more important than moisture
to the cooking of starch, a decrease in the SME would lead to a decrease in the WAI. A
gelatinization peak, which is evident when comparing the cold and hot paste viscosities at
these extrusion conditions, supports this (Tables 4.9, 4.10). Additional gelatinization and
starch damage would lead to an increase in cold viscosity up to a maximum. After being
fully cooked, starch damage would result in decreasing viscosity. This may have been
the case for the powders at 145°C. Since there was no gelatinization peak for these
samples, it can be inferred that increasing the moisture would decrease the amount of
starch degradation, and result in higher cold viscosity values.

The effects of particle size on cold viscosity for WFL were similar to those for WF,
where the smallest particle size had the lowest cold viscosity (Table 4.9). The cold
viscosity of the 93-145 mm fraction was higher than that of the 145-249 um fraction.

This was probably again due to a hydration effect (section 4.1.8) since the hot paste

viscosities were similar for the two particle size fractions (Table 4.10).

56

Table 4.10. Hot Paste Viscosity and Statistical Relationships of Extruded and
Milled Wheat Flour + Lecithin Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
145-249 pm 1 93-145 pm 1 <93 um
Extrusrgtll/g/orndltions Hot Paste Viscosity (RVU)
25/300/120 225:7 242:8 106:4
25/200/120 238:6 244:0 123:9
35/200/120 342:1 353:12 179:1
35/300/120 306:1 281:11 151:1
25/300/145 228:6 237:3 103:4
25/200/145 288:6 258:21 122:12
35/200/145 314:2 221:5 145:6
35/300/145 260:1 l 169:7 137:7
Raw Wheat Flour + Lecithin = 213 RVU
Particle Size
145-24911111 1 93-145 pm 1 <93 um
F value
M 5712* 4.55 149.50*
S 25.45““ 5621* 31.35*
T 0.44 144.16* 15.98*
M*S 0.3 2630* 0
M*T 15.72* 167.84* 11.21*
S*T 4.30 0 2.04
M*S*T 0.84 3.81 2.97
Least Squares Means
M: 25% - 35% Estimate of Difference (RVU)
200 rpm, 120°C ~104* -108* -56*
200 rpm, 145°C -26 37 -23
300 rpm, 120°C -81* -39 —45*
300 rpm, 145°C -32 68* -34*
S: 200 rpm - 300 rpm
25%, 120°C 12 2 17
25%, 145°C 60 21 19
35%, 120°C 36 71* 28*
35%, 145°C 54 52* 8
T: 120°C - 145°C
25%, 200 mm -51 -14 2
25%, 300 rpm -3 5 3
35%, 200mm 28 132* 34*
35%, 300rpm 46 112* 14
Particle Size C - B C - A B - A
Estimate (RVU) -1 18* - 142* -24

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Viscosity measured in Rapid Visco Units (RVU).
Standard deviation measured between powders produced from repetition of extrusion.

A = l45—249um particle size, B = 93-145um particle size, C = <93um particle size.

"' Significant at p<0.05.

57

 

4.1.10. Wettability

Various procedures were attempted in preliminary experiments. These included
brushing powder samples through a sieve, as is done with cocoa powder (Anonymous),
and dropping powder through a funnel as described by the International Dairy Federation
Standard Method 87:1979. However, for any method to be meaningful and reproducible,
the powder must be spread evenly on the liquid surface because even small areas of
unevenness can lead to large differences in wetting time. None of the methods gave
acceptable evenness of powder spreads, and due to the poor reproducibility using the

attempted methods, wettability was not measured in this study.

4.1.11. Development of Dispersibility Procedure

Preliminary experiments found that the few methods published describing procedures
to measure dispersibility (Sander, 1989; IDF 87:1979) did not suit the products being
measured in the current study. Since the main concerns regarding the powders produced
in this study were the formation of lumps during mixing with water and sedimentation of

particles, the method that was designed needed to take these properties into account.

MM: An overhead stirrer was chosen because it could provide consistent mixing at a
defined speed. The paddle was placed with one half beneath the surface of the water so
that as it turned, the powder would be more evenly spread on the surface. With the entire
paddle below the surface, the action of the vortex would be to pull all of the powder into

one lump in the center of the liquid.

58

Mixing time: One minute was chosen as the mixing time after determining the time

 

needed for some commercial products that had acceptable instant properties. Although
this method is not able to discriminate among powders that may completely disperse
prior to one minute, all powders that did disperse before this time would probably be

considered acceptable.

Separation of undispersed solids: After mixing, the suspensions were poured through
sieves to remove any lumps. A powder with many lumps that did not disintegrate would
have fewer dispersed solids making it through the sieve and there would be a lower solids
concentration in the sieved suspension. A larger sieve mesh size was needed for some of
the larger particle size powders because of swelling. The smallest sieve size that would
allow for free passage of the suspension without blockage of the sieve was chosen. The

use of a separatory funnel was chosen to allow for undisturbed drainage.

Measurement of dispersibilig: Twenty ml of the sieved suspension were immediately
drained so that sedimentation did not occur and an aliquot was measured for solids
concentration on a g/ g basis. An aliquot of approximately 4 g was taken to reduce the

drying time.

Measurement of stability: Two minutes was chosen as the time to allow the remaining
approximately 80 m1 of the sieved suspension to remain undisturbed because this time
was found to be enough to discriminate among samples in preliminary experiments. It

also allowed for a faster assay and was within reasonable expectations for a product.

59

Fifty ml were drained to remove many of the solids that had sedimented. The remaining
top portion was chosen to be measured for solids content because a solution with a large

amount of sedimentation would have a lower solids content near the top.

4.1.12. Dispersibility of Extruded Wheat Flour

Dispersibility was 100% for the two largest particle sizes, independent of extrusion
conditions. Below 93 microns, extrusion moisture and temperature had significant
effects (p<0.05, Table 4.11). Dispersibility was greater than 90% at the higher
temperature and moisture levels. At low moisture and temperature, many large lumps
formed and powder dispersibility was less than 60%.

The effects of particle size reduction on mixing and dispersibility were noted by
Anderson et al. (1969). In their study, extruded corn grits ground to 20 to 40 mesh were
judged to be good for instant gruels. Particles smaller than 40 mesh had poor
dispersibility because of lumping and balling and were deemed not suitable for use in
products such as instant drinks. The particle size for which dispersibility was found to
become difficult for corn grits was much larger than that determined for wheat flour in
this experiment. Even at the 93-145 um particle range, the wheat flour powders
dispersed very easily. It was not until a much finer powder (<93 um) was produced that
dispersibility became poor.

The effect of particle size may be due to many of the factors listed in section 2.5.
These include an increase in the strength of interparticle forces, an increase in the contact
angle, and a decrease in the bulk powder pore size as powder particles become smaller

(Schubert, 1987; Schubert, 1993).

60

Table 4.11. Dispersibilitya and Statistical Relationships of Extruded and Milled

Wheat Flour (WF) and Wheat Flour + Lecithin (WFL) Powders

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions WF WFL
M/S/T Dispersibility (%) Dispersibility (%)
25/300/120 58.7:18 89.7:12
25/200/120 57.4:4.5 91.1:0.9
35/200/ 120 93.5:0.1 948:2.1
35/300/120 93.6:O.4 93.7:0.5
25/300/145 74.0:5.4 92.5:0.4
25/200/145 73.9: 1 .1 97.3:08
35/200/145 87.5:0.4 91.0:20
35/300/145 892:2.9 89.1:1.3

 

 

 

Raw Wheat Flour = 83.8%

1 Raw Wheat Flour + Lecithin = 92.2 %

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

WF WFL
F value F value
M 32206“ 0.51
S 0.32 1286*
T 1486* 0.06
M*S 0 1.49
M*T 5793* 4525*
S*T O 2.59
M*S*T 0.24 1.06
Least Squares Means
M: 25% - 35% Estimate of Difference (%)
200 rpm, 120°C ~36.2* -3.8
200 rpm, 145°C -l3.6* 6.3*
300 rpm, 120°C -35.0* -4.0
300 rpm, 145°C -15.2* 3.4
S: 200 rpm - 300 rpm -1.3 1.4
25%, 120°C -0.1 4.8
25%, 145°C -0.1 1.2
35%, 120°C -1.7 1.9
35%, 145°C
T: 120°C - 145°C -16.6* -6.2*
25%, 200 rpm -15.4* -2.8
25%, 300 rpm 6.0 3.8
35%, 200 rpm 4.5 4.6
35%, 300 rpm 4.5 4.6

 

 

" Results for <93 um fraction. 145-249 um and 93-145 um fractions had dispersibility of 100%.

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
A = 145-249nm particle size, B = 93-145um particle size, C = <93um particle size.

* Significant at p<0.05.

61

 

The rate of hydration may also be an important factor to consider in relation to particle
size. As shown by the RVA data (Table 4.7, Figure 4.4), the <93um size particles
hydrated and reached peak viscosity much more quickly than the particles of larger sizes.
Swelling reduces dispersibility by clogging pores in the powder, limiting the penetration
of water (Schubert, 1993). A delay in the hydration of larger sized particles may allow
for dispersal to take place before it is limited by swelling. Thomas and Atwell (1999)
also noted that formation of lumps often occurs in fine powders, which have large surface
areas, due to quick hydration and gelling of starch on the outer surface of a lump.

Although it is known that dispersibility of starch-containing powders is more difficult
because of swelling, it was not known what the effects of the starch modifications that
take place during extrusion would be on dispersibility. In general, conditions which led
to an increase in the SME, made powder dispersibility poorer. There was an inverse
correlation between dispersibility and the SME for the <93 um powder fraction (Figure
4.6). It is also important to note that the SME was inversely correlated to the WAI and
cold viscosity (Figures 4.1 and 4.3), but positively correlated to the WSI (Figure 4.2).
Dispersibility was higher in powders with high WAI and high cold viscosity, and low
WSI.

These results run counter to what might be expected since dispersibility is more
difficult in viscous liquids (Schubert, 1993). Increasing water solubility of powders
might be thought to improve dispersibility by reducing the amount of particles that could
cause lumps. One hypothesis for the differences noted in dispersibility is that the rate of

hydration may be quicker for poorly dispersible powders. If conditions that increase

62

 

‘ o d? "E'WF
0'0 I ° C. o oWFL‘
, I O o o . __
90 o l . 00 I 0
Ԥ 80 .
E; ' i
‘ E l ,
1:2 70 u
' fl
1 8
In
.. 60 I I
a I
’ I
50
WFr=-0.76
4O

 

 

 

 

50 7o 90 110 130 150 170 190 1
SME (Whlkg)

Figure 4.6. Relationship between the specific mechanical energy (SME) and
dispersibility for extruded and milled wheat flour powders with (WF L) and without

(W F) lecithin.

63

water solubility and starch granule breakdown also increase the rate of hydration,
swelling and gelling could occur before dispersion of the particles is able to take place.
This would then lead to the formation of lumps. It is also interesting to note that
stickiness of extruded products has been related to increased water solubility (Mercier et
al., 1980). Dextrinized starches could stick together more, possibly resulting in more
lumps. This could explain why dispersibility was poorer at higher SME and higher water

solubility.

4.1.13. Dispersibility of Extruded Wheat Flour with Lecithin

As with the wheat flour, the 145-249 um and 93-145 mm particle sizes were 100%
dispersible for the WFL powders. Dispersibility for the <93 pm samples ranged from
89.1% to 97.3%, significantly greater for the WFL than for the WP (p<0.05, Tables 4.3,
4.11). However it is not clear if the properties specific to lecithin or properties attributed
to extrusion with lipids are responsible for the improved dispersibility. Lecithin-specific
effects may have been the improvement of wetting or the addition of electrostatic
repulsive forces between the particles (Schubert, 1993). The changes in water
absorption, water solubility, and viscosity when lecithin was added to the wheat flour
may also be important. However, these changes may occur with the addition of lipids in
general and not just with the addition of lecithin. Further work in this area to clarify the

observation is needed.

4.1.14. Stability of Extruded Wheat Flour

Although the two largest particle size fractions were easily dispersed, they were not
as stable to sedimentation. The 145-249 um particle size fraction had the poorest

64

stability and all of its values were below 40% (Table 4.12, Figure 4.7). Moisture was the
only extrusion variable to have a significant effect, as a decrease was associated with
increased stability (p<0.05, Table 4.12). Stability at 93-145 um covered a wide range
from very poor to very good. Moisture was the only significant variable and stability was
greater after extrusion at 25% than at 35% (p<0.05). At <93 um, there was a smaller
range measured of stabilities, with most being around 80%.

According to Stokes’ law (Schubert, 1993), larger particles would be expected to have
poorer stability, as was the case for the 145—249 um fraction. Stability to sedimentation
would be expected to be greater at higher cold viscosity, but there was not a correlation
between the two. This may occur because as particles sediment, viscosity would then
also decrease. It is also important to keep in mind that the RVA measures viscosity with
a paddle, which keeps particles suspended. Changes in particle size and density due to
water uptake and swelling should also be considered. The increase in stability at lower

moisture extrusion may also be related to increasing solubility at these conditions.

4.1.15. Stability of Extruded Wheat Flour with Lecithin
There were again large differences in stability among the different particle size
fractions (Table 4.13). Stability was highest for the <93 um fraction and in this case, it

was correlated with cold viscosity (Figure 4.8). As the cold viscosity increased, stability
improved from 60% to over 75%. Overall, WFL had lower stability than WF (p<0.05,

Table 4.3). This may be due to the lower water solubility and cold viscosity of the WFL

65

Table 4.12. Stability and Statistical Relationships of Extruded and Milled Wheat
Flour Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
E“'“s'l°"1'/§;’rnd't'°“s 145-249 pm 93-145 mm <93 pm
Stability (%)
25/300/120 37.5:28 898:0.1 80.9:0.9
25/200/120 34.4:9.7 91.4:1.4 77.7: 1 .5
35/200/120 7.7:0.7 44.1:3.9 74.7:1.6
35/300/120 72:0.4 51.8:207 788:3.1
25/300/145 33.4125 86.0:32 82.0:5.5
25/200/145 36.5:0.4 92.3:2.0 84.5:1.4
35/200/145 12.3:1.5 69.8:14.1 87.9:1.9
35/300/145 99:23 60.5:32 83.1:0.3
Raw Wheat Flour = 6.3%
Particle Size
145-249 pm 1 93-145 m 1 <93 pm
F value
M 193.03* 5328* 0.02
S 0.06 0.27 0
T 0.49 2.96 3538*
M*S 0.06 0.12 0.09
M*T 1.51 4.17 4.91
S*T 1.45 1.42 11.60*
M*S*T 0.20 0.45 0.56
Least Squares Means
M: 25% - 35% Estimate of Difference (%)
200 pm, 120°C 272* 47.4* 3.0
200 rpm, 145°C 242* 22.6 -3.4
300 rpm, 120°C 298* 38.1* 2.1
300 rpm, 145°C 23.5* 25.5 -1.1
S: 200 rpm - 300 rpm
25%, 120°C -3.1 1.6 -3.2
25%, 145°C 3.1 6.4 2.5
35%, 120°C -0.5 -7.7 -42
350%, 145°C 2.4 9.3 4.8
T: 120°C - 145°C
25%, 200 rpm -2.1 -0.9 -6.9
25%, 300 rpm 4.1 3.9 -1.2
35%, 200 rpm -5.1 -25.7 -13.2*
35%, 300 rpm -22 -8.7 -4.3
Particle Size C - B C - A B - A
Estimate (%) 8.0 588* 509*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
A = 145-249nm particle size, B = 93-145um particle size, C = <93pm particle size.

* Significant at p<0.05.

66

 

 

 

 

 

 

 

90 . if 25,: x
. ‘I xx 1
7 O
80 o . o 0. o ’
70 x
l g 60 A X X
Q —1 I
E 50 x 1
r 40 x . .5
m * I I I i
30 I __-.__ ;
2° .n145-249 i
' ' 93445 i
10 - I I _x
I I I .<93
0 :: .....
50 7o 90 110 130 150 170 190

SME (Whlkg)

Figure 4.7. Relationship between the specific mechanical energy (SME) and
stability for extruded and milled wheat flour powders.

67

Table 4.13. Stability and Statistical Relationships of Extruded and Milled Wheat

Flour + Lecithin Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
E""“"‘I’Jl'/gf°rnd"'°"s 145-93 ttm 93445 1.1111 <93 um
Stability (%)
25/300/120 62:0.3 49.6:08 73.5: 1 .3
25/200/120 5.4:0.l 56.6:6.9 71 6:00
35/200/120 6.3:0.1 28.5:19.1 61.0:0.9
35/300/120 6.5:0.1 418:3.5 65.0:2.5
25/300/145 6.3:0.1 432:9.5 74.3:0.1
25/200/145 5.3:0.1 51.5:5.9 688:0.8
35/200/145 7.1:0.1 45.4:1.6 798:2.3
35/300/145 7.4:0.0 46.5:00 762:0.4
Raw Wheat Flour + Lecithin = 7.6%
Particle Size
145-249 pm 1 93-145 um [ <93 pm
F value
M 262.44* 5.73* 5.50
S 8836* 0 842*
T 4900* 0.39 108.57*
M*S 2500* 3.41 6.98*
M*T 4900* 4.23 142.69*
S*T 1.00 0.70 2.22
M*S*T 0.04 0.46 17.68*
Least Squares Means
M: 25% - 35% Estimate of Difference (%)
200 rpm, 120°C -0.9* 28.1 10.7*
200 rpm, 145°C -1.8* 6.1 -11.1*
300 rpm, 120°C 03 7.8 8.6*
300 rpm, 145°C -1.2* -3.4 -l.9
S: 200 rpm - 300 rpm
25%, 120°C -0.9* 7.1 -1.9
25%, 145°C -1.0* 8.3 -5.6*
35%, 120°C 02* -13.3 -4.0
35%, 145°C -0.4 -l.1 3.7
T: 120°C - 145°C
25%, 200 rpm 0.1 5.2 2.9
25%, 300mm -0.1 6.4 -0.8
35%, 200 rpm 08* -16.9 -18.9*
35%, 300 rpm -1.0* ~4.7 -11.2*
Particle Size C - B C - A B - A
Estimate (%) 25.9* 65.0* 39.1*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

 

Standard deviation measured between powders produced from repetition of extrusion.
A = 145-249um particle size, B = 93-145um particle size. C = <93um particle size.
* Significant at p<0.05.

68

100 - »—-~

 

 

 

 

r=082
90
1 cf. 80 ° 0
E mo 0 o
I 5 7 0 ° 0
8 0 o o
"’ o
o
60 ° 0
50 1
4O 60 80 100 120 140
Cold Viscosity (RVU)

Figure 4.8. Relationship between cold viscosity and stability for extruded and
milled wheat flour + lecithin powders.

69

4.2. Extrusion of Defatted Soy Flour (SF), Defatted Soy Flour with Lecithin (SFL),
and Blend of Defatted Soy Flour with Full Fat Soy Flakes (S-Blend)

After extrusion of SF, SF L, and S-Blend samples, the extrudates were dried overnight
and ground to produce soy powders of three particle size fractions: 93-145 um, <93 um,
and jet milled. The following are results along with analysis and discussion regarding the

instant properties of extruded soy.

4.2.1. Particle Size Analysis of Extruded and Milled Soy

The mean particle sizes. measured by laser diffraction ofthe 93-145 um SF and SFL
powder fractions were approximately 150 um (Table 4.14). The particle size measured
by laser light scattering may be larger than the sieve size due to differences in the way
that particle sizes are measured. Particles that are not spherical or which scatter light in
different ways can affect the results of laser particle size analysis and give different
results than physically sieving with a mesh screen (Webb, 2001). The mean particle sizes
of the <93 um SF and SFL fractions produced by sieving were approximately 50 um
(Table 4.14). The S-Blend powders had slightly larger average particle sizes. Jet milling .
of the SF and SFL extrudates reduced them to powders with mean particle sizes of :1 5

pm (Table 4.14).

4.2.2. Specific Mechanical Energy of Soy During Extrusion

The specific mechanical energy (SME) during extrusion of the SF ranged from 120:8
thg to 362:16 Wh/kg depending on the conditions (Table 4.15). Moisture had the
most significant effect on the SME, followed by screw speed, and temperature (Table

4.15). Higher moisture and temperature resulted in a lower SME while increasing the

70

Table 4.14. Laser Particle Size Analysis of Extruded and Milled Soy Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

93 - 145 pm <93 um Jet Milled
Sieve Fraction Sieve Fraction Fraction
Treatment‘ . .
M /S/T rep Mean Particle Size (um)
SF
30/300/120 rep 1 ‘55 58 19
SF
30/300/120 rep 2 ‘53 56 17
SF
40/200/120 rep 1 58 19
SF
40/200/120 rep 2 51 16
SF
40/200/145 rep 1 54 14
SF
40/200/145 rep 2 57 15
SFL
30/300/120 rep 1 ‘50 57 15
SFL
30/300/120 rep 2 146 56 13
SFL
40/300/120 rep 1 54 12
SFL
40/300/120 rep 2 54 11
SFL
40/200/145 rep 1 50 15
SFL
40/200/1455p 2 56 9
S-Blend _ 68
30/300/ 120 rep 1 -
S-Blend - 67
30/300/120 rep 2 -
S-Blend - 64 -
30/200/145 rep 1
S-Blend - 73 _
30/200/145 rep 2

 

 

 

 

 

Raw Defatted Soy Flour = 25 pm

 

' M = moisture (%); S = screw speed (rpm); T = zone 5 temperature (°C). Each repetition extruded on a
different day.

SF = defatted soy flour, SFL = defatted soy flour + lecithin, S-Blend = blend of defatted soy flour with full
fat soy flakes.

71

 

Table 4.15. Specific Mechanical Energy (SME) and Statistical Relationships During
Extrusion of Soy

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions SF 1 SFL 1 S-Blend
M/S/T SME (Wh/kg)
30/300/120 361:16 227:10 126:0
30/200/120 253: 16 167:6 89:2
40/200/120 146:1 1 106:13 71:9
40/300/120 188:1 1 156:23 93:4
30/300/145 324:30 168:6 115:4
30/200/145 23 8:37 137:4 84:5
40/200/145 120:8 99:8 69:16
40/300/145 164:6 152:3 98:3
SF 1 SFL I S-Blend
F value
M 21800" 212.50* 50.05*
S 5538* 233.51* 10483"
T 7.49* 60.49* 1.42
M*S 823* 0.95 1.87
M*T 0 3670* 2.90
S*T 0.27 4.29 1.01
M*S*T 0.42 645* 1.24
Least Squares Means
M: 30% - 40% Estimate of Difference (Wh/kg)
200 rpm, 120°C 107* 61* 185
200 rpm, 145°C 118* 39* 15
300 rpm, 120°C 174* 71* 33*
3003m, 145°C 160* 16 17
S: 200 rpm - 300 rpm
30%, 120°C -l09* -6|* _3721:
30%, 145°C -87* -31* -31*
40%, 120°C -42 -51* -23
40%, 145°C -45 -54* -30*
T: 120°C - 145°C
30%, 2004mm 16 30* 5
30%, 300 mm 38 59* 12
40%, 200 rpm 27 7 2
40%, 300 rpm 24 4 -5

 

 

 

 

SF = defatted soy flour, SFL = defatted soy flour + lecithin, S-Blend = blend of defatted soy flour with full

fat soy flakes.

M = moisture (%); S = screw speed (rpm); T = zone 5 temperature (°C).
Standard deviation measured beteween repetitions of extrusion.

* Significant at p<0.05.

72

 

screw Speed increased the SME. The effects of changes in these extrusion conditions on
SME were similar to those noted for the extruded wheat flour and can be explained
similarly. Moisture and temperature affect the rheological properties of dough in the
extruder, and increasing the screw speed adds mechanical energy.

The addition of lecithin to the defatted soy flour resulted in a Significant decrease in
the SME during extrusion when compared to that of the soy flour alone (p<0.05 Table
4.16). This may be due to a lubrication effect, which was also seen when lecithin was
added to wheat flour (see section 4.1.3). The low end of the range of SME values for
SFL was similar to the SME of soy flour without lecithin, but the high end of the range of
SME only reached 227:10 Wh/kg (Table 4.15). Changes in moisture, screw speed, and
temperature during extrusion of SFL had similar effects on the direction of change in the
SME as during extrusion of SF, but in the case of SFL extrusion, screw speed had the
most Significant effect.

Blending the defatted soy flour with full fat soy flakes resulted in a further lowering of
the SME. The different extrusion conditions also had less effect on the SME when
extruding S-Blends, resulting in a much smaller range of SME values (69:16 Wh/kg to
126:0 Wh/kg, Table 4.15). Increasing the screw speed had the largest effect of
increasing the SME followed by decreasing the moisture (Table 4.15). Varying the

temperature did not have a significant effect on the SME.

4.2.3. Trypsin Inhibitor Inactivation of Extruded Soy
Trypsin inhibitor inactivation was greater than 90% for all of the extrusion conditions

and among each of the three raw material formulations (Table 4.17). Decreased moisture

73

 

Table 4.16. Comparison of Properties Between Soy Powder“ Formulations

 

 

 

 

 

 

 

 

 

Estimate of Difference
A - B A - C B - C
SME (Wh/kg) 73* 131* 58*
T11 (“/Q 0.1 0.6 0.5
WAI (g/g) 0.043 0.13 0.08
WSI (g/g) -0.019* 0.011* 0008*
Dispersibility (%) 4.9"“ 0 -4.9*
Stability (%) -4.5* -l .7* 29*

 

 

 

 

" Comparison done for <93 um particle size.

A = defatted soy flour, B = defatted soy flour + lecithin, C = blend of defatted soy flour with full fat soy
flakes.

SME = specific mechanical energy, T1] = trypsin inhibitor inactivation, WAI = water absorption index,
WSI = water solubility index.

* Significant at p<0.05.

 

 

74

Table 4.17. Trypsin Inhibitor Inactivation (T11) and Statistical Relationships of
Extruded and Milled Soy Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions SF 1 SFL ] S-Blend
M/S/T Inactivation (%)
30/300/120 94.7: 1 .2 95.0:0.9 92.4:08
30/200/120 94.4: 1 .0 93.4: 1 .8 93.3:0.8
40/200/120 91 5:07 92.2: 1 .0 91 0:09
40/300/120 92.1:12 92.7:1.5 91.1:1.0
30/300/145 95.9:0.4 952:0.0 94.7:08
30/200/145 95.4:08 948:0.1 95.3:0.5
40/200/145 93.6:0.5 92.9: 1 .0 94.1:0.3
40/300/145 93.6:08 938:0.2 94.0:1.7
Raw SF = 30.4 mg/g Raw S-Blend = 23.2 mg/g
SF 1 SFL 1 S-Blend
F value
M 263.18* 22.14* 4629*
S 4.67 5.41 3.29
T 9324* 5.13 160.19*
M*S 0.07 0.07 3.52
M*T 4.38 0.02 3.62
S*T 0.39 0.39 0.02
M*S*T 1.36 1.14 0.26
Least Squares Means
M: 30% - 40% Estimate of Difference (%)
200 rpm, 120°C 28* 1.3 22*
200 rpm, 145°C 19* 2.0 1.3
300 rpm, 120°C 2.6 2.3 1.3
300 rpm, 145°C 23* 1.4 0.7
S: 200 rpm - 300 rpm
30%, 120°C 03 -1.6 0.9
30%, 145°C -0.4 -0.3 0.6
40%, 120°C -0.5* -0.6 -0.1
40%, 145°C -0.0 -0.9 0.1
T: 120°C - 145°C
30%, 200 rpm -l.0 -l .7 -2.1
30%, 300 rpm -1.2* -0.2 -2.3*
40%, 200 rpm -2.0 -0.7 -3.0*
40%, 300 rpm -1 .5* -1.0 -2.9*

 

 

 

SF = defatted soy flour, SFL = defatted soy flour + lecithin, S-Blend = blend of defatted soy flour with full

fat soy flakes.

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.

* Significant at p<0.05.

75

 

and higher temperature had Significant, but small effects of increasing the inactivation.
Greater than 90% destruction of trypsin inhibitors after extrusion of full fat soy flours has
also been reported by various authors (Bookwalter and Mustakas, 1971a; Harper, 1979;
Konstance et al., 1998). Inactivation of trypsin inhibitors iS mainly caused by
temperature. Van den Hout et al. (1998) did not find evidence of the involvement of

Shear forces.

4.2.4. Water Absorption Index (WAI) of Extruded' Soy

Increasing the temperature from 120°C to 145°C resulted in a small decrease in the
WAI of the SF, while changes in the other extrusion conditions had almost no effect
(Table 4.18). The WAI decreased as particle size decreased (p<0.05), however there was
not a large decrease when compared to the raw soy flour. Traina and Breene (1994)
found large differences in water absorption among commercial full fat soy flours
produced by different methods. Of the samples they measured, the water absorption
ranged from 0.3-2.3 ml/ g. They also found an inverse relationship between the protein
dispersibility index (PDI) and water absorption.

Extrusion of the SF L and S-Blend samples also did not produce any large differences
in the WAI values (Tables 4.19 and 4.20). This occurred even though the SME was
much lower with the addition of lipids. The factor with the greatest effect on the WAI of

SF L was again, particle size (Table 4.19).

4.2.5. Water Solubility Index (W SI) of Extruded Soy
Extrusion of the different soy flour formulations led to large decreases in the WSI

values compared to those of the respective raw materials. The raw materials had WSI

76

Table 4.18. Water Absorption Index (W Al) and Statistical Relationships of
Defatted Soy Flour Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
Extrusion Conditions 93-145 um I <93 um [ :15 um (Jet Milled)
M/S/T WAI (g/g)
30/300/120 3 .90:0.071 3 .72:0.00 3 .44:0.07
30/200/120 3.96:0.03 3.85:0.00 3.48:0.01
40/200/120 4.21:0.10 3.88:0.01 3.43:0.01
40/300/120 4.03:0.07 3.78:0.00 3.40:0.09
30/300/145 3.87:0.21 3.69:0.01 3.42:0.15
30/200/145 3.87:0.23 3.67:0.00 3.39:0.18
40/200/145 3.77:0.057 3.52:0.01 3.28:0.10
40/300,145 3.86:0.08 3.63:0.00 3.30:0.08
Raw Defatted SoyFlour = 3.71 g/g
Particle Size
93-145 urn [ <93 pm I 215 pm (Jet Milled)
F value
M 1.38 0.99 6.93*
S 0.48 0.83 0.01
T 1036* 35.75* 829*
M*S 0.02 0.83 0
M*T 4.54 621* 1.33
S*T 1.94 845* 0.82
M*S*T 0.83 0.34 0.03
Least Squares Means
M: 30% - 40% Estimate of Difference (g/
200 rpm, 120°C -0.25 -0.04 0.05
200 rpm, 145°C 0.10 0.15 0.11
300 rpm, 120°C -0.13 0.10 0.04
300 rpm, 145°C 0.01 0.06 0.12
S: 200 rpm - 300 rpm
30%, 120°C 0.06 0.13 0.04
30%, 145°C 0.01 -0.02 -0.03
40%, 120°C 0.18 0.11 0.03
40%, 145°C -0.09 -0.11 -0.02
T: 120°C - 145°C
30%, 200 rpm 0.09 0.18 0.09
30%, 300 rpm 0.04 0.04 0.02
40%, 200 rpm 0.44 0.36 0.15
40%, 300 rpm 0.18 0.15 0.10
Particle Size C - B C - A B - A
Estimate (glg) -0.33* -0.54* -0.22

 

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).
Standard deviation measured between powders produced from repetition of extrusion.
A = 93-145 um particle size, B = <93um particle Size, C = ~15 um (jet milled).

* Significant at p<0.05.

77

Table 4.19. Water Absorption Index (WAI) and Statistical Relationships of

Defatted Soy Flour + Lecithin Powders

 

Particle Size

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions 93-145 pm j <93 pm 1 :15 pm (Jet Milled)
M/S/T WAI (g/g)
30/300/120 3.86:0.03 3.81:0.21 3.32:0.12
30/200/120 4.06:0.07 3.93:0.1 1 3.34:0.08
40/200/120 3.95:0.05 3.61:0.05 3.24:0.04
40/300/120 4.07:0.01 3.71:0.17 3.22:0.06
30/300/145 3.87:0.11 3.57:0.11 3.17:0.08
30/200/145 3.85:0.04 3.71:0.02 3.42:0.17
40/200/145 4.14:0.07 3.61:0.02 3.21:0.01
40/300/145 3.91:0.08 3.45:0.00 3.17:0.05
Raw Defatted Soy Flour + Lecithin = 4.16 g/g
Particle Size
93-145 um [ <93 pm j :15 pm (Jet Milled)
F value
M 10.77* 2038* 5.46
S 4.84 5.09 3.56
T 1.83 25.79* 0.67
M*S 0.25 1.99 1.46
M*T 3.29 1.99 0.01
S*T 1.09 3.90 2.06
M*S*T 20.40 3.36 1.20
Least Squares Means
M: 30% - 40% Estimate of Difference (g/g
200 rpm, 120°C 0.12 0.33 0.11
200 rpm, 145°C -0.30* 0.10 0.21
300 rpm, 120°C -0.21 0.10 0.10
300 rpm, 145°C -0.04 0.13 0.01
S: 200 rpm - 300 rpm
30%, 120°C 0.20 0.13 0.03
30%, 145°C -0.03 0.14 0.25
40%, 120°C -0.13 -0.11 0.02
40%, 145°C 0.24 0.17 0.02
T: 120°C - 145°C
30%, 200 rpm 0.22 0.23 -0.08
30%, 300 rpm 001 0.24 0.15
40%, 200 rpm -0.20 -0.01 0.03
40%, 300 rpm 0.17 0.27 0.06
ParticleSize C-B C-A B-A
Estimate (g/g) -0.41* —0.70* -0.29*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
A = 93-145 um particle size, B = <93um particle size. C = :15 um (jet milled).

* Significant at p<0.05.

78

 

Table 4.20. Water Absorption Index (WAI), Water Solubility Index (W SI), and
Statistical Relationships of Blend of Defatted Soy Flour with Full Fat Soy Flakes (S-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Blend) Powders
ExtruslpvrIr/gfq‘ndrtrons W AI (g/g) WSI (g/g)
30/300/120 3.58:0.08 0235:0002
30/200/120 3.61:0.13 0236:0004
40/200/120 3 50:0. 16 0240:0004
40/300/120 3.57:0.10 0237:0009
30/300/145 3.68:0.01 0236:0001
30/200/145 3.70:0.07 0235:0004
40/200/145 3.53:0.22 0234:0006
40/300/145 3.56:0.12 0233:0004
Raw S-Blend WAI = 3.627 g/g, WSl = 0.408 g/g
WAI WSI
F value F value
M 3.17 0.01
S 0.04 0.27
T 0.81 0.88
M*S 0.46 0.17
M*T 0.66 1.09
S*T 0.02 0.17
M*S*T 0.04 0.01
Least Squares Means
M: 30% - 40% Estimate of Difference (g/g)
200 rpm, 120°C 0.] 1 -0.004
200 rpm, 145°C 0.18 0.001
300 rpm, 120°C 0.01 -0.002
300 rpm, 145°C 0.12 0.004
S: 200 rpm - 300 rpm
30%, 120°C 0.03 0.002
30%, 145°C 0.03 -0.001
40%, 120°C -0.07 0.003
40%, 145°C -0.03 0.002
T: 120°C - 145°C
30%, 200 rpm -010 0.001
30%, 300 rpm -010 -0002
40%, 200 rpm 003 0.005
40%, 300 rpm 0.02 0.004

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).
Standard deviation measured between powders produced from repetition of extrusion.
* Significant at p<0.05.

79

 

values of around 0.4 g/ g while the extruded powders had WSI values ranging from
approximately 0.21-0.26 g/g (Tables 420-422). AS with the WAI, there were only small
differences in the WSI among the extrusion conditions. Changes in particle Size
produced significant differences in the WSI (p<0.05, Tables 420-422). As the particle
size decreased, there was a small decrease in the WSI.

The decrease in the WSI of samples after extrusion is largely due to a decrease in the
solubility of the soy protein since it makes up a little over 50% of the defatted soy flour.
The nitrogen solubility index (N SI), a measure of protein solubility, has been found to
decrease below 20% after extrusion of full-fat soy flour (Harper, 1979). Bookwalter et
al. (1971a) produced extruded full-fat soy flours with NSI values ranging from 50% to
16%. In the present study, although extrusion was able to cook the soy, as shown by
trypsin inhibitor inactivation, it was not able to do so without significantly decreasing the
WSI under the extrusion conditions used. Conditions which increase the water solubility
of the soy carbohydrates and proteins could be advantageous especially for use in

products such as beverages.

4.2.6. Dispersibility of Extruded Soy

In the current study, dispersibility was found to be mainly a function of particle size
and not of the extrusion conditions as the SF particle size fractions 93-145 um and <93
um were 100% dispersible. However, the z15 um (jet milled) SF powders had
dispersibilities from 80% to 90% (Table 4.23). Surprisingly, the SF powders with
lecithin had poorer dispersibility. The 93-145 um SFL powders still had 100%

dispersibility, however the <93 pm fractions had dispersibilities of approximately 95%

80

Table 4.21. Water Solubility Index (WSI) and Statistical Relationships of Extruded
and Milled Defatted Soy Flour Powders

 

Particle Size

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions 93-145 pm 1 <93 pm 1 :15 um (Jet Milled)
M/S/T WSI (g/g)
30/300/ 120 0214:0006 0230:0001 0249:0003
30/200/120 0210:0004 0224:0001 0234:0013
40/200/120 0194:0007 0213:0006 0241:0004
40/300/120 0.21 1:0.015 0222:0001 0239:0007
30/300/145 0226:0011 0232:0013 0251:0013
30/200/145 0218:0000 0232:0002 0245:0015
40/200/145 0207:0001 0224:0008 0238:0016
40/300/145 0210:0001 0224:0003 0241:0009
Raw Defatted Soy Flour = 0.454 g/g
Particle Size
93-145 um <93 pm =15 um (Jet Milled)
F value F value F value
M 1035* 1419* 1.21
S 4.55 2.37 1.46
T 4.55 6.76 0.38
M*S 0.26 O. 14 1.21
M*T 0.33 0.14 0.52
S*T 0.41 2.37 0.06
M*S*T 1.41 0.14 0.45
Least Squares Means
M: 30% - 40% Estimate of Difference
200 rpm, 120°C 0.016 0.012 -0.007
200 rpm, 145°C 0.012 0.008 0.007
300 rpm, 120°C 0.004 0.008 0.010
300 rpm, 145°C 0.165 0.008 0.01 1
S: 200 rpm - 300 rpm
30%, 120°C -0.004 -0.006 -0.015
30%, 145°C -0008 0.000 -0.007
40%, 120°C -0.017 -0.009 0.002
40%, 145°C 0.007 0.000 -0.003
T: 120°C - 145°C
30%. 200 rpm -0.008 -0.008 —001 1
30%, 300 rpm 0012 -0.003 -0.002
40%, 200 rpm 0125 -0.012 0.003
40%, 300 rpm 0.007 -0.003 -0.002
ParticleSize C-B C-A B-A
Estimate (g[g) 0017* 0031* 0014*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
A = 93-145um particle size, B = <93um particle size, C = ~15 um (jet milled).

* Significant at p<0.05.

81

 

Table 4.22. Water Solubility Index (WSI) and Statistical Relationships of Extruded
and Milled Defatted Soy Flour + Lecithin Powders

 

 

 

 

 

 

 

 

 

 

 

Particle Size
Extrusion Conditions 93-145 pm I <93 pm I :15 pm (Jet Milled)
M/S/T wsr (g/g)
30/300/120 0218:0003 0244:0005 0268:0002
30/200/120 0210:0001 0235:0001 0257:0001
40/200/120 0217:0002 0241:0001 0258:0001
40/300/120 0215:0004 0239:0002 0257:0002
3073007145 0221:0003 0252:0001 0271:0001
30/200/145 0218:0001 0250:0001 0268:0001
40/200/145 0213:0001 0241:0004 0262:0003
40/300/145 0221:0001 02490:0000 0265:0001

 

 

 

 

Raw Defatted Soy Flour + Lecithin = 0.415 g/g

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Particle Size
93-145 pm I <93 pm 1 =15 pm (Jet Milled)
F value
M 0.18 1059* 4639*
S 2669* 2229* 21.10*
T 1219* 101.15* 6526*
M*S 2.36 1.40 1568*
M*T 7.04* 1479* 0.23
S*T 1.58 0.79 1.23
M*S*T 1640* 2835* 1327*
Least Squares Means
M: 30% - 40% Estimate of Difference (g/
200 rpm, 120°C -0007 -0.006 -0.001
200 rpm, 145°C 0.005 0010* 0.005
300 rpm, 120°C 0.003 0.005 0.01 1*
300 rpm, 145°C 0.001 0.003 0.006
S: 200 rpm - 300 rpm
30%, 120°C -0.009* -0.009* -0011*
30%, 145°C -0004 -0.002 -0.003
40%, 120°C 0.002 0.002 0.002
40%, 145°C -0008 -0.009* -0.003
T: 120°C - 145°C
30%, 200 rpm 0008* -0016* -0.015*
30%, 300 rpm -0003 -0.008* -0003
40%, 200 rpm 0.004 0.000 -0.004
40%, 300 rpm -0006 0024* -0.008*
Particle Size C - B C - A B - A
Estimate (g/g) 0019* 0047* 0027*

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
A = 93-14Sum particle size, B = <93um particle size, C = :15 um (jet milled).

* Significant at p<0.05.

82

Table 4.23. Dispersibility and Statistical Relationships of Extruded and Milled
Defatted Soy Flour Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions 2=15 pm (Jet Milled)‘I
M/S/T Dispersibility (%)
30/300/120 84.6:1.4
30/200/120 82.3:2.6
40/200/120 92.8: 1 .6
40/300/120 805:0.6
30/300/145 910:06
30/200/145 86.0:22
40/200/145 805:0.3
40/300/145 87.0:3.3
Raw Defatted Soy Flour = 77.8%
F value
M 12.16*
S 9.36*
T 1450*
M*S 0.75
M*T 2.44
S*T 9.36*
M*S*T 2.70
Least Squares Means
M: 30% - 40% Estimate of Difference (%)
200 rpm, 120°C -06
ZOMm, 145°C 5.5
300 rpm, 120°C 4.2
300 mm. 145°C 40
S: 200 rpm - 300 rpm
30%, 120°C -2.4
30%, 145°C -5.0
40%, 120°C 2.4
40%, 145°C -6.5
T: 120°C - 145°C
30%, 200 rpm -3.7
30%, 300 rpm -64
40%, 200 rpm 2.3
40%, 300 rpm -6.5

 

' 93-145 um and <93 um fractions had 100% dispersibility.

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.
* Significant at p<0.05.

83

and the =15 um (jet milled) fractions had dispersibilities below 80% (Table 4.24). The
SFL powders had a tendency to form many small lumps that did not disperse. The blend
of defatted soy flour with full fat soy flakes was 100% dispersible.

Wettability was not measured, though it influences dispersibility, and is affected by
surface charges along with the size and shape of proteins (Hagerdal and Ltifqvist, 1978).
Paulsen and Horan (1965) found that wetting was related to heat treatment and the
protein dispersibility index (PDI) of defatted soy flour. The contact angle was lower as
the PDI decreased, indicating better wetting after heat treatment. Dispersibility is also
affected by the ionic composition and pH of the liquid in which the proteins are being
dispersed (Kinsella, 1976). The results of the current experiment indicated that particle
size was the most important factor affecting dispersibility at the extrusion conditions

used.

4.2.7. Stability of Extruded Soy

Stability was also mainly a function of particle Size and not the extrusion conditions.
The 93-145 1.1m particle size fractions of the SF samples had stability levels below 40%
(Table 4.25). The <93 um fractions had a stabilities of approximately 55%, and the =15
um (jet milled) powders had stabilities ranging from 74% to 81%. Stabilities for the SFL
samples were slightly higher, but demonstrated a similar dependence on particle Size
(Table 4.26). The S-Blend powders had stability levels similar to the <93 um fractions

for the SF and SFL powders (Table 4.27).

84

Table 4.24. Dispersibilitya and Statistical Relationships of Extruded and Milled

Defatted Soy Flour + Lecithin Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions Particle Size .
M/S/T <93 um 1 215 pm (Jet Milled)
Dispersibility (%)
30/300/120 960:1.3 79.9:2.1
30/200/120 963:0.4 75.0:02
40/200/120 93 6:24 72.4: 1 .0
40/300/20 960:0.0 71 .7:2.5
30/300/145 94.4:].6 782:4.7
30/200/145 94.6:02 765:3.5
40/200/145 94.3:32 763:2.3
40/300/145 959:1.6 73.9:3.9
Raw Defatted Soy Flour + Lecithin = 74.9%
Particle Size
<93 pm T :15 pm (Jet Milled)
F value
M 0.20 958*
S 1.15 0.51
T 0.69 1.40
M*S 1.79 3.98
M*T 1.35 1.65
S*T 0.03 1.04
M*S*T 0.07 0.09
Least Squares Means
M: 30% - 40% Estimate of Difference (%)
200 rpm, 120°C 2.7 2.6
200 rpm, 145°C 0.3 0.2
300 rpm, 120°C 0.0 8.2
300 rpm, 145°C -1.5 4.3
S: 200 rpm - 300 rpm
30%, 120°C 03 '5-0
30%, 145°C 0.2 -1.7
40%, 120°C -2.4 0~7
40%, 145°C -1.7 2.5
T: 120°C - 145°C
30%, 200 rpm 1.8 -1.5
30%, 300 rpm 1.6 1.8
40%, 200 rpm 07 ~39
40%, 300 rpm 0.1 -2.2
Particle Size A - B
Estimate (%) -19.7*

 

 

 

‘ 93-145 um fraction had 100% dispersibility.

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.

A = jet milled fraction, B = <93 um fraction.
* Significant at p<0.05.

85

 

Table 4.25. Stability and Statistical Relationships of Extruded and Milled Defatted
Soy Flour Powders

 

 

 

 

 

 

 

 

 

 

 

Particle Size
Extrusion Conditions 93-145 pm I <93 pm =15 um (Jet Milled)

M/S/T Stabilipy (%)
30/300/120 298:1.1 54.3:0.4 78.1:0.0
30/200/120 332:2.3 567:0.5 748:2.3
40/200/120 37.5:18 556:1.3 800:1.7
40/300/120 351:4.7 57.9:22 77.9:2.1
30/300/145 30.4:18 53.7:1.3 77.7:1.6
30/200/145 302:1.6 53.3:1 .1 79.3:0.5
40/200/145 29.3:0.7 548:0.6 81.1:12
40/300/145 29.5:0.4 555:0] 81.1:0.4

 

 

 

 

Raw Defatted Soy Flour = 74.6%

 

 

Particle Size

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

93-145 pm 1 <93 pm :15 pm (.let Milled)
F value
M 7.94* 698* 1241*
S 4.31 0.25 0.02
T 3517* 1076* 8.38*
M*S 0.11 5.19 1.85
M*T 17.54* 0.17 0
S*T 5.45 0.30 0.89
M*S*T 0.14 3.84 596*
Least Squares Means
M: 30% ~ 40% Estimate of Difference (%)
200 rpm, 120°C ~4.3 1.] ~52
200 rpm, 145°C 0.9 ~l.5 ~18
300 rpm, 120°C ~52 ~3.6 0.3
300 rpm, 145°C 1.0 ~1.9 ~3.4
S: 200 rpm ~ 300 rpm
30%, 120°C 3.5 2.4 ~3.3
30%, 145°C ~02 ~04 1.6
40%, 120°C 2.5 ~2.3 2.2
40%, 145°C ~02 ~08 0.0
T: 120°C ~ 145°C
30%, 200 rpm 3.0 3.4 ~4.5
30%, 300 rpm ~07 0.7 0.4
40%, 200 rpm 82* 0.8 ~l.l
40%, 300 rpm 5.5 2.4 ~32
Particle Size C ~ B C - A B ~ A
Estimate (%) 23.5* 469* 23.4*

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).
Standard deviation measured between powders produced from repetition of extrusion.
A = 93-145 um particle size, B = <93um particle size, C = =15 um (jet milled).

* Significant at p<0.05.

86

 

Table 4.26. Stability and Statistical Relationships of Extruded and Milled Defatted

Soy Flour + Lecithin Powders

 

Particle Size

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions 93-145 run 1 <93 pm 1 :15 pm (Jet Milled)
M/S/T Stability (%)
30/300/120 351:2.5 602:2.6 856:0.3
30/200/ 120 38.6: 1 .3 62.5:3.3 84.0:] .1
40/200/120 39.8: 1 .1 61 .4:1 .0 865:3.5
40/300/120 378:0.8 61.3:1.1 872:1.2
30/300/145 323:0.8 58.4:2.3 84.9:2.1
30/200/145 32.6: 1 .1 58.0:0.9 84.0: 1 .6
40/200/145 33.5:08 58.7:0.3 88.3:1.3
40/300/145 322:0.8 57.4:1.7 88.0:1.1
Raw Defatted Soy Flour + Lecithin = 73.4%
Particle Size
93-145 pm 1 <93 pm [ =15 um (Jet Milled)
F value
M 641* 0.01 1231*
S 1484* 2.77 0.82
T 129.73* 4146* 0.32
M*S 0.05 0.07 0.43
M*T 2.91 0.02 1.05
S*T 4.61 0.59 0.23
M*S*T 1.89 3.86 0.01
Least Squares Means
M: 30% ~ 40% Estimate of Difference (%)
200 rpm, 120°C ~12 1.1 ~25
200 rpm, 145°C ~09 ~08 ~4.3
300 rpm, 120°C ~27 ~12 ~1.6
300 rpm, 145°C 0.2 1.0 ~3.1
S: 200 rpm ~ 300 rpm
30%, 120°C 3.5 2.4 ~1.6
30%, 145°C 0.3 ~04 ~l.0
40%, 120°C 2.0 0.1 ~07
40%, 145°C 1.3 1.3 0.3
T: 120°C ~ 145°C
30%, 200 rpm 60* 4.6* 0.1
30%, 300 rpm 0.4 1.8 0.7
40%, 200 rpm 63* 2.7 ~18
40%, 300 rpm 56* 3.9 ~09
Particle Size C ~ B C ~ A B ~ A
Estimate (%) 263* 508* 24.5*

 

 

 

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).

Standard deviation measured between powders produced from repetition of extrusion.

A = 93-145 um particle size, B = <93um particle size, C = =15 um (jet milled).

* Significant at p<0.05.

87

 

Table 4.27. Stability and Statistical Relationships of Extruded and Milled Blend of
Defatted Soy Flour with Full Fat Soy Flakes (S-Blend) Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Extrusion Conditions Stability (%)
M/S/T
30/300/120 57.5:0.3
30/200/120 578:1.1
40/200/120 57.7:0.9
40/300/120 57.5:1.1
30/300/145 558:0.4
30/200/145 559:1.6
40/200/145 566:2.1
40/300/145 563:1.1
Raw S-Blend = 75.3%
F value
M 0.20
S 0.13
T 634*
M*S 0
M*T 0.34
S*T 0
M*S*T 0
Least Squares Means
M: 30% ~ 40% Estimate of Difference (%)
200 rpm, 120°C 0.1
200 rpm, 145°C ~07
300 rpm, 120°C 0.1
300 rpm, 145°C ~06
S: 200 rpm ~ 300 rpm
30%, 120°C 0.3
30%, 145°C 0.2
40%, 120°C 0-2
40%, 145°C 0-3
T: 120°C ~ 145°C
30%, 200 rpm 1.9
30%, 300 rpm 1.8
40%, 200 rpm 1.1
40%, 300 rpm 1.2

 

M = moisture (%), S = screw speed (rpm), T = zone 5 temperature (°C).
Standard deviation measured between powders produced from repetition of extrusion.
* Significant at p<0.05.

88

 

4.3. Blends of Wheat and Soy Powders for Instant Drinks

Combining the different wheat and soy powders with different functional properties
produced blends with a range of instant properties (Tables 428-430). In general,
dispersibility was better with WFL-soy blends than with WF-Soy blends. Dispersibility
was poorest in blends that contained SF L powder. The differences in stability were not
great among the various samples. However, formulations containing the S-Blend
powders were poorest.

Suspensions with ranges in viscosity were produced and blends with WF L generally
had lower viscosities than blends with WF. By choosing the right extrusion conditions
and degree of milling, powders with improved viscosity, dispersibility, and stability can
be produced.

The combining of wheat and soy flours prior to extrusion may be more convenient,
however, separate extrusion allows for better control of the product characteristics.
Particle Size requirements for acceptable instant properties are also different for the two
extruded materials. Instant wheat flour powders are much more difficult to disperse and
should be of relatively larger particle sizes for acceptable dispersibility characteristics.
Soy powders, on the other hand, need to be ground to a fine powder to improve the final
product mouthfeel. An additional consideration is that blends which contain
approximately 14 g of soy powder per serving have enough soy protein to meet the Food
and Drug Administration criteria for a health claim regarding lowering the risk of heart

disease - ~ 6.25 g of soy protein per serving (FDA, 1999).

89

Table 4.28. Dispersibility of 50:50 Blends of Wheat and Soy Powders

 

 

 

 

 

 

Dispersibility (%)
WF 25/300/120' ssavovrlirzo 2561/15/11 20 35%5/1120
swig/120 70-6 89-4 90.1 95.3
30/5507120 53-5 86-9 84.5 88.3
3313305320 74-5 ‘00 99.8 100

 

 

 

 

 

' moisture (%), screw speed (rpm), zone 5 temperature (°C).
WF = wheat flour, WF L = wheat flour + lecithin, SF = defatted soy flour, SF L = defatted soy flour +
lecithin, S-Blend = blend of defatted soy flour with full fat soy flakes.

90

 

Table 4.29. Stability of 50:50 Blends of Wheat and Soy Powders

 

 

 

 

 

 

Stability (%)
WF WF WFL WFL
25/300/120' 35/200/120 25/300/120 35/200/120
SF
30/300/120 76-1 68.9 70.2 66.5
SFL
30/300/120 73-4 71.0 71.6 67.4
S-Blend
30/300/120 64-8 60.5 59.2 54.5

 

 

 

 

 

" moisture (%), screw speed (rpm), zone 5 temperature (°C).
WF = wheat flour, WFL = wheat flour + lecithin, SF = defatted soy flour, SFL = defatted soy flour +
lecithin, S-Blend = blend of defatted soy flour with full fat soy flakes.

91

 

Table 4.30. Cold Viscosity of 50:50 Blends of Wheat and Soy Powders

 

Cold Viscosity (RVU)

 

 

 

 

 

 

WF WF WFL WFL
25/300/120' 35/200/120 25/300/120 35/200/120
3.0/3337120 ‘2 7 7 6
30386017120 1 1 8 6 4
33131037120 ‘5 12 5 4

 

 

 

 

" moisture (%), screw speed (rpm), zone 5 temperature (°C).
WF = wheat flour, WFL = wheat flour + lecithin, SF = defatted soy flour, SFL = defatted soy flour +

lecithin, S-Blend = blend of defatted soy flour with full fat soy flakes.

Viscosity measured in Rapid Visco Units (RVU).

92

 

V Conclusions

Through the control of extrusion conditions and particle size during milling, instant
wheat powders with different properties can be produced. Depending on the desired
product, water absorption, solubility and viscosity can be adjusted. The specific
mechanical energy provides a measure of the starch transformations taking place during
extrusion and is correlated with final product characteristics. The addition of lecithin
provides another means to modify instant wheat powders, especially as it relates to the
improvement of dispersibility.

These results can be applied to the case of instant powders for beverages where
dispersibility and viscosity are important. Extrusion of wheat flour with lecithin would
be best suited because viscosity can be controlled while still producing a powder with
good dispersibility. Control of particle size during milling is also important to viscosity,
dispersibility, and stability to sedimentation.

Although extrusion cooking of soy inactivated trypsin inhibitors, the changes in
extrusion conditions in this study were not able to create soy powders with a range of
functional properties as has been done with wheat flour. Therefore extrusion of defatted
soy flours may not be as practical or economical as toasting during the desolventizing
process in the production of defatted soy flours. However, the extrusion of full-fat soy
flours has been shown previously to be an efficient means of processing and may be
practical for blends of full and defatted soy.

The milling of soy flours is crucial to the dispersibility and stability of the instant
powders produced. To improve stability so that soy powders could be used in products
such as beverage mixes, a very fine powder is needed. However, dispersibility can

93

become a problem with fine powders, as was the case with the =15 11m (jet milled) soy
powders. The addition of lecithin to the defatted soy flour prior to extrusion did not

improve dispersibility.

94

VI Suggestions for Further Research

One area for further investigation regarding instant wheat powders is the relationship
between starch modifications and dispersibility. Of interest is the poor dispersibility of
wheat powders found for samples with high water solubility. Another related area of
study is the rate of hydration of the powders. Better control of particle size, such as the
removal of very fine particles from the WP <93 um particle Size fraction, may also be
important. The role of lecithin is another area to investigate as other lipids aside from
lecithin may improve dispersibility. Formulations of lecithin with different hydrophilic-
lipophilic balances, blended with wheat or soy may work better. Sensory studies would
identify which particle Sizes are best suited for products such as gruels or beverages.
They are also necessary to determine if milling of the soy was able to adequately reduce
its graininess. Sensory studies would also be valuable to relate dispersibility and stability

values to end product qualities.

95

Appendix I

Moisture Content of Extruded and Milled Powders

96

r 1 6L1, 1— 1. 1_. 1.1L 1: L

1F

1 1—1

lhlrhl

Appendix Ia. Moisture Content of Extruded and Milled Wheat Flour Powders

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Moisture (%)
Particle Size (um)
Treatment Rep 145-249 93-145 <93
25/300/120 l 7.1 6.9 6.3
2 4.3 4.3 3.6
25/200/120 1 6.4 6.1 5.6
2 5.0 5.0 4.9
35/200/120 1 4.9 4.8 4.7
2 5.7 5.4 5.0
35/300/120 1 4.9 4.9 4.8
2 5.6 5.5 5.2
25/300/145 l 5.0 5.3 4.7
2 3.9 4.3 3.8
25/200/145 1 6.1 6.1 5.8
2 4.4 4.3 4.4
35/200/145 1 4.5 4.7 4.4
2 4.3 4.1 3.6
35/300/145 1 3.8 3.8 3.6
2 6.0 6.0 . 5.6

 

 

 

 

 

 

Raw Wheat Flour = 13.2% moisture

 

97

 

Appendix Ib. Moisture Content of Extruded and Milled Wheat Flour + Lecithin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Powders
Moisture (%)
Particle Size (pin)
Treatment Rep 145-249 93-145 <93
25/300/120 l 6.0 5.6 5.5
2 5.6 5.9 5.8
25/200/120 1 5.8 5.6 5.8
2 5.5 5.6 5.8
35/200/120 1 6.1 5.9 6.0
2 5.7 5.4 5.0
35/300/120 1 6.5 6.7 6.3
2 6.0 6.2 6.1
25/300/145 l 5.6 5.6 5.7
2 5.3 5.2 5.6
25/200/145 1 5.8 5.8 6.2
2 5.6 5.8 5.5
35/200/145 1 5.8 5.8 5.2
2 6.0 5.9 6.1
35/300/145 1 5.6 5.6 5.2
2 5.9 5.8 5.4

 

 

 

 

 

 

Raw Wheat Flour + Lecithin = 12.8% moisture

 

98

 

Appendix Ic. Moisture Content of Extruded and Milled Defatted Soy Flour

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Powders
Moisture (%)
Particle Size (pm)

Treatment Rep 93-145 <93 Jet Milled
30/300/120 1 6.6 6.6 5.3

2 5.7 5.9 5.3
30/200/120 1 7.0 7.0 6.1

2 5.7 5.5 5.0
40/200/120 1 6.7 6.5 5.3

2 5.6 6.0 5.0
40/300/120 1 7.8 7.2 5.7

2 6.1 6.1 5.1
30/300/145 1 7.2 6.9 5.9

2 6.7 6.1 5.6
30/200/145 1 7.1 7.2 5.5

2 6.1 5.8 5.7
40/200/145 1 6.2 6.3 5.2

2 6.3 6.4 6.2
40/300/145 1 6.5 6.6 5.3

2 6.2 6.6 6.2

 

 

 

 

 

 

Raw Defatted Soy Flour = 6.0% moisture

 

99

 

Appendix Id. Moisture Content of Extruded and Milled Soy Flour + Lecithin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Powders
Moisture (%)
Particle Size (pm)

Treatment Rep 93-145 <93 Jet Milled
30/300/120 1 7.6 7.5 7.0

2 5.6 5.6 5.7
30/200/120 1 7.1 7.0 6.3

2 5.6 5.6 5.6
40/200/120 1 5.7 5.8 6.0

2 5.9 5.7 5.7
40/300/120 1 7.1 7.0 6.1

2 5.7 5.7 5.5
30/300/145 1 8.0 8.0 6.8

2 5.9 6.0 6.0
30/200/145 1 7.5 7.3 6.2

2 6.2 6.1 5.9
40/200/145 1 7.7 7.5 6.5

2 6.4 6.1 6.0
40/300/ 145 1 6.9 6.6 6.4

2 6.9 6.7 6.1

 

 

 

 

 

 

Raw Defatted Soy Flour + Lecithin = 60% moisture

 

100

 

Appendix Ie. Moisture Content of Extruded and Milled Blend of Defatted Soy
Flour with Full Fat Soy Flakes (S-Blend) Powders

Treatment Moisture %
30/300/120 6.5
6.4

30/200/120 6.7
6.1

40/200/120 6.5
6.1

40/300/120 6.6
6.3

30/300/145 6.9
6.3

30/200/145 6.9
6.6

40/200/145 7.3
6.4

40/300/145 6.8
6.1

 

Raw S-Blend = 9.3% moisture

101

Appendix II

Torque, Die Pressure, and Specific Mechanical Energy (SME) During Extrusion

102

.5535 we consume. Eat 30355 $3306 5953 @2388 =o=£>ou vaucfim
.Cev obscuomEB w econ u h ASP; 8on 328 n m 65 0:58.: 1 .2

 

 

 

 

 

 

 

 

 

 

 

 

 

3:. ”new 3% ”a 3 5. Q2
:3 $32 9% 3 3 ENE.“
13.: :38 $3 I 2. :ER
9m: 3:: N :3 E a. :53
E“: ca: 38 ”a as: Q2
3: :32 EN 3 cs: Q2
.43.. 51.3: :3. «a 8:893
3.2 $83 1:; 4a 8:252
9.2.5 :55 :9: 2:82.: ea 3.; 2.35 came: 25: coo: Egghkwfiaiim

 

 

 

 

 

ESE 32;? he =22.be war—.5 @sz .3.—gm— ..SEnnoo—z 050on was asmmohh 95 Jar—eh é: «Ex—25¢:

103

 

damabxo mo costume: Set @3355 £2.33 conga; 35808 5:35: EAESm
.Auov 238358 m 28m n H 555 :0on 328. u m 6x: 0.538.: n 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SUNS $2: _va am 3:8:2
38 one: 38 3 3:862
:2: :3; 3; m E 225 :3
3o: Zm _ N 38 «N 2:: :8
$2 _ 83: 38 ME 8:8:2
:8 ”N3: 38 ”a 8:862
32 $5 :8 E 8:888
as ”N32 :2 E 8:8:2

3.23: Si :9: 2.58:. «5 ex: 2.th 9:3 33— 80.,— m=o=_..=.w:wfi_m=§m

 

55604 + ESE 285$ no 53333; uni—5 £sz 3.5.—m 33:23: 95925 "Ea .onammohm 05 6:5th 5: “Ex—25¢:

104

 

damp—38 :0 53332 Soc 3269:— fiogroa 5233 c9582: .8830: 28:85
.Auov 888388 n econ u H Asa: 3on 328 u m .3.: 2:868 n E

 

 

 

 

 

 

 

 

 

 

3%. 83mm E; 3 3:88:

”Hos 88.8 $3 3 3:898

~38 888 N _ 8» 3 2:888

9888 88% £8 98 2:8:8

_ Ex: 8%? 3: 9m 8:8:8

_ :5: 3 3: 3mm 3 8588

£88 :38 $8 0.8 8:898

£88 888 $8 0.8 8:888
3:55 5% 5.: 2.58... «E 3.: SEE. 585 33. c8... seal—mnwnfiaeam

 

 

 

 

 

 

.53.”— xcm 68.39: a: =22.th untam— Am—Emv magma—m 38533—2 «59on cu: .9.—58...— oa ditch .9: “5:89;.

105

.:o_m=bxo we consume Eoc noun—no.5 flotsam 5953 c2388 cousin: Banana
.Aoov BEER—ES m econ n ._. .AEEV 8on 328 n w 6x; 0520:. u 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

M82 388 38 :H 3.588
88 838 $8 in 3:88
$5 Zn; E; 8” 3:. _ :8
8&2 8%? :9 8” 2:25. m
83: 8E: $8 3 8:8: 3
232 338 $8 3. 8: .8: v
9:2 810% 3% 3 8: .88
o _ #8 8% _ o 8% o.” 8: 5:8

3:53 5% cm“: 238... an ex: 2.35 :35 22 as“. Eocefiwfiziuu

5.225

+ 2.2% how :33qu .«o afimeuam— “5.5: £sz kahuna 32:23: 95925 “Ea Juana...— oa .2...th i: x3593

6
O
1

.56:be Lo SEER: Soc :82:th 30:38 50253 3382: c235: 2855
.Gov 8388.88 m econ n H .355 30% 38% u m 65 8.56:. n .2

 

 

 

 

 

 

 

 

 

 

38 833 :8 3 2:82:
8% 8%. _ £8 3” 2:258
83” 3 3: $8. 9.8 2:588
8m: 8%: :8 cm 2.588
3.8 8%.: :8 3 8.588
8: 83: $8 3 8:088
88 Six: :8 cm 8.588
$8. $88 88 cm 8388
3:56 8% cm: 2.38... 05 3.: 2.55 9:3 83. can Scacnuwfiaefim

 

 

 

 

 

:85.ng mow—«E xcm 3,..— ==h 5:5
ESE ham 10:.“qu ac _Eo—m no comm—Eta— wutsn £sz 3.5.—m 1815532 95025 .23 .ousmmohm 05 .9.—PEP .0: “5.59:.

107

 

Appendix III

Rapid Visco Analyzer Analysis of Extruded and Milled Powders

108

.86:be mo cotton": Eat 3260.5 £0“.ch 5253 8532: 5:35“. 23:85

.53: 9:5 8m_> E5 5 8580:. £883

mvtochm

mvtochm

mVCOOQmN

mvtoonN

om_\OCm\mm

ON_\com\mm

ON _ \oonN

HNZ 3”. 3:893

5%:
533m;

€533: 2:5
no.5. 2:;
as; 25 :8; _EE 3%.. 8: 25

53: 53: :23“;
558m; 533m; 25:53:.

52: 53:

SB:
$.33 genie: 2o: Em

«EEK

9.2.3.5 Lao—h 2.3.—>9 62:2 can vac—_hxm— no EPA—.22 L0N%_«=< can; 21¢“ a:— fiunomndw

 

109

 

623.55 Mo .8289: Soc 326th "cacao“. 5033 3532: 2:833 cavcfim
.325 3:5 03.5 Band 5 3532: b303,;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

110

 

 

 

 

 

 

 

 

 

 

o w onm mam 3 m: 2N q 4E5 39;
omfimm 2156 mfiom vamv Emma— on. has? SHR— mov
vamnm oid mfimv _ih wfimfl 13: mag: nib: $2-8 mvtoonm
$32 odflfiv mfiv mfiwm menme 32 vfiocm _ I”:— own??—
Omfim _ m fofifio _Hnm oamm 36m— Nfifl onE 5i: mov
mmaomm NoHWN name oamw Zoom mum—N 92mm mac—m $1-3 mvzoomkm
m _ «mom odflmé name w _ “mo Efioom mfivm a _ “E m mason 03.51
m _ i m. odfiud _me mflmm mix: 33 m _ amm— ofimw mov
mmfiwmm Yofioé afioc $2. Efiomm 3.0— 2%me _mfimmm 373 m3\oom\mm
mam. _ Ndid name wamw minnow mumom Efiwwm oamfl ovmfiv.
$2: fimflmé 73m mflv _«No NH: vino— _H_o_ mav
omfiomm fioflxm 36v mfimm wavmm ma? 35mm mHmmm $78 mESGQmN
meow. :13» mfiov Eng 2 av _ N N“: m _ mem 8me av??—
NIKZ odds mfllm 3mm :2: Nune— 12 2 fume mov
2%ch odfimd mi? 360 nfimom mi: wan—mm :32 32.8 omzoonm
VH3 0.300 vi? 3% _«wwm m i: 303 13a $.93.
$3 odino Snow meow oawfl cum _ N 1&2 vamm mav
mmfimmm odfimd each mag _mawmm _va mmfimmm Bacc— $1-3 om_\oom\mm
my”: odfioé Name oi ¢ ova» _ m NHVN wVHNVm 3 _ n avméi
cmfl. _ m mmfimé _Hmm mfiov $2: E“: 0H3— afinm: mov
oamem odeN vamv ofiwm mawmm mac. Sufism oflomm 37.3 om_\oom\mm
Siem— mdfloé on”? $8 vim—mm m3— onmm Bumm— 9A-?—
mflk: odfi .m 36m 15. 3mg cam. $2: $62 mov
:33 _deN 3mm oficv oflwmm _fl; wfimvm wflvm €73 omtoonm
0H. m. odilu 12mm :3 mm— m mam. efimmm chS «www.m—
EEx 255 E
A «20>5 25h 53: mmvmw» SB: 5%: ”WWW”, bfivww 5:22,.— 25.58;
“Bantam :Bogaonm 2o: «Em
38m 2ch gang .2:— 32?— 8: 2.5 22.25

 

PEPE:— :___=qu + 5::— 23:3 12:2 can con—raga we £3.22 hunk—.22 3a; 293— A== 565%:

 

BIBLIOGRAPHY

111

Bibliography

Anderson, R.A., Conway, H.F., Pfeifer, V.F., and Griffin Jr., EL. 1969. Gelatinization of
corn grits by roll — and extrusion — cooking. Cereal Science Today. l4(1):4-12.

Anderson, R.A., Pfeifer, V.F., Bookwalter, G.N., and Griffin Jr., EL. 1970. Instant CSM
food blends for world — wide feeding. Cereal Science Today. 16(1):5-11.

Anonymous. Technical brochure: SOL Lecithinated Cocoa Powders. Koog aan de Zaan,
Holland.

Atwell, W.A., Hood, L.F., Lineback, D.R., Varriano-Marston, E., and Zobel, HF. 1988.
The terminology and methodology associated with basic starch phenomenon. Cereal
Foods World. 33 2306-31 1.

Atwell, WA. 2001. Wheat Flour American. Association of Cereal Chemists St. Paul, MN

Barnes, S. 1998. Evolution of the health benefits of soy isoflavones. Proceedings of the
Society for Experimental Biology and Medcine. 217:386—392.

Becker, A., Hill, SE, and Mitchell, J .R. 2001. Milling — a further parameter affecting
the rapid visco analyzer (RVA) profile. Cereal Chemistry. 78(2):166-172.

Bhattacharya, M., and Hanna, MA. 1987. Influence of process and product variables on
extrusion energy and pressure requirements. Journal of Food Engineering. 6: 153-163.

Bhattacharya, S., Sudha, M.L., and Rahim, A. 1999. Pasting characteristics of an
extruded blend of potato and wheat flours. Journal of Food Engineering. 40:107-111.

Bookwalter, G.N., Mustakas, G.C., Kwolek, W.F., Mc Ghee, J .E., and Albrecht, W.J.
1971a. Full-fat soy flour extrusion cooked: Properties and food uses. Journal of Food
Science. 36:5-9.

Bookwalter, G.N., Kwolek, W.F., Black LT, and Griffin Jr., E.L. 1971b. Corn meal/soy
flour blends: Characteristics and food applications. Journal of Food Science. 3621026-
1032.

Brent, J .K., Mulvaney, S.J., Cohen, C., Bartsch, J .A. 1997. Thermomechanical glass
transition of extruded cereal melts. Journal of Cereal Science. 26:301-312.

Bressani, R., and Elias, LG. 1966. All — vegetable protein mixtures for human feeding.
The development of of INCAP vegetable mixture 14 based on soybean flour. Journal of
Food Science. 31:626-631.

 

 

Bressani, R., Braham, J .E., Elias, L.G., Cuevas, R., and Molina, MR. 1978. Protein
quality of whole com/whole soybean mixture processed by a simple extrusion cooker.
Journal of Food Science. 43:1563-1565.

Bruin, S., Van Zuilichem, DJ, and Stolp, W. 1978. A review of fundamental and
engineering aspects of extrusion of biopolymers in a single screw extruder. Journal of
Food Process Engeneering. 2:1-3 7.

Brummer, T., Meuser, F., van Lengerich. B.. and Niemann, C. 2002. Effect of extrusion
cooking on molecular parameters of corn starch. Starch/Starke. 5421-8.

Chen, S. 1989. Preparation of fluid soymilk. Pages 341-352 in: Proceedings of the world
congress on vegetable protein utilization in human foods and animal feedstuffs. T.H.
Applewhite, ed. AOCS: Champaign, Illinois.

Colonna, P., and Mercier, C. 1983. Macromolecular modification of manioc starch
components by extrusion cooking with and without lipids. Carbohydrate Polymers. 3:87-
108.

Colonna, P., Doublier, J .L., Melcion, J .P., de Monredon, F ., and Mercier, C. 1984.
Extrusion cooking and drum drying of wheat starch. 1. Physical and macromolecular
modifications. Cereal Chemistry. 61(6):538-543.

Colonna, P., Tayeb, J ., and Mercier, C. 1989. Extrusion cooking of starch and starchy
products. Pages 247-319 in: Extrusion Cooking. C. Mercier, P. Linko, and J .M. Harper,
eds. The American Association of Cereal Chemists: St. Paul, MN.

Davidson, V.J., Paton, D., Diodsady, L.L., and Rubin, L.J. 1984. A model for
mechanical degradation of wheat starch in a single-screw extruder. Journal of Food
Science. 49:1154-1157.

DelValle, FR. 1981. Nutritional qualities of soya protein as affected by processing.
Journal of the American Oil Chemists’ Society. 58:419-429.

Della Valle, G., Kozlowski. A., Colonna. P.. and Tayeb. J. 1989. Starch transformation
estimated by the energy balance on a twin screw extruder. Lebensmsmittel -
Wissenschaft - und - Technolgie. 22:279-286.

Diosady, L.L., Paton, D., Rosen, N., Rubin, L.J., and Athanassoulias, C. 1985.
Degradation of wheat starch in single-screw extruder: Mechano-kinetic breakdown of
cooked starch. Journal of Food Science. 50:1697-1706.

Doublier, J .L., Colonna, P., and Mercier, C. 1986. Extrusion cooking and drum drying of

wheat starch II. Rheological characterization of starch pastes. Cereal Chemistry.
63(3):240-246.

113

FDA. 1999. Food labeling: health claims; soy protein and coronary heart disease. Federal
Register. 64:206.

Frame, ND. 1994. Operational characteristics of the co-rotating twin-screw extruder.
Pages 1-51 in: The technology of extrusion cooking. N.D. Frame, ed. Blackie Academic
and Professional: Glasgow, Scotland.

Fulmer, R.W. 1989. The preparation and properties of defatted soy flours and their
products. Pages 55-61 in: Proceedings of the world congress on vegetable protein
utilization in human foods and animal feedstuffs. TH. Applewhite, ed. AOCS:
Champaign, Illinois.

Gomez, M.H., and Aguilera, J .M. 1983. Changes in the starch fraction during extrusion-
cooking of corn. Journal of Food Science. 48:378-381.

Graham, G.G., Baertl, J .M., Placko, RP, and Cordano, A. 1972. Dietary protein quality
in infants and children. VIII. Wheat — or oat — soy mixtures. The American Journal of
Clinical Nutrition. 25:875-880.

Guy, R.C.E., and Home, A.W., 1988. Extrusion and co-extrusion of cereals. In Food
Structure: Its creation and evaluation. J .M.V. Blanshard and J .R. Mitchell, eds.
Butterworths: London.

Hagerdal, B., and Lofqvist, B. 1978. Wettability and surface pressure of myoglobin
treated with acetone. Journal of Food Science. 43:27-30, 40.

Hamerstrand, GB, Black, L.T., Glover, J .D. 1981. Trypsin inhibitors in soy products:
modification of the standard analytical procedure. Cereal Chemistry. 58(1):42-45.

Harper, J .M. 1979. Food extrusion. Critical Reviews in Food Science and Nutrition.
11(2):]55-215.

Harper, J .M. 1981. Extrusion of foods, volume I. Harper, J .M. ed. CRC Press, Inc., Boca
Raton, Florida.

Harper, J .M. 1986. Extrusion texturization of foods. Food Technology. 40:70-76.
Heiser, J ., and Trentelman, T. 1989. Full - fat soya products - manufacturing and uses in
foodstuffs. Pages 52-54 in: Proceedings of the world congress on vegetable protein
utilization in human foods and animal feedstuffs. T.H. Applewhite, ed. AOCS:
Champaign, Illinois.

Horan, RE. 1973. Wheat — soy blends high — quality protein products. Cereal Science
Today. 18:11-14.

114

l;

Hoseney, RC. 1994. Starch. Pages 29-64 in: Principles of cereal science and technology.
American Association of Cereal Chemists, Inc. St. Paul, MN.

IDF Standard 87:1979. Determination of the dispersibility and wettability of instant dried
milk.

Jansen, G.R., Harper, J .M., and O’Deen, L. 1978. Nutritional evaluation of blended foods
made with a low — cost extruder cooker. Journal of Food Science. 43:912-925.

Jensen, J .D. 1973. Methods of instantizing powders for the preparation of food drinks.
The Manufacturing Confectioner. 53:47-56.

Jin, J ., Hsieh, F ., and Huff, HE 1995. Effects of soy fiber, salt, sugar and screw speed on
physical properties and microstructure of corn meal extrudate. Journal of Cereal Science.
22: 185-194.

Johnson, D.W., and Kikuchi, S. 1989. Processing for producing soy protein isolates.
Pages 66-77 in: Proceedings of the world congress on vegetable protein utilization in
human foods and animal feedstuffs. T.H. Applewhite, ed. AOCS: Champaign, Illinois.

Konstance, R.P., Onwulata, C.I., Smith, P.W., Lu, D., Tunick, M.H., Strange, ED, and
Holsinger, V.H. 1998. Nutrient — based corn and soy products by twin — screw extrusion.
Journal of Food Science. 63:864-868.

Khare, S.K., Krishna, J ., and Gandhi, AP. 1995. Citric acid production from okara by
solid state fermentation. Bioresource Technology. 54:323-325.

Kinsella, J.E.,1979. Functional Properties of Soy Proteins. Journal of the American Oil
Chemists’ Society. 56:242-258.

Klingler, R.W., Meuser, F., Niedek, EA. 1986. Einfluss der art der energieubertragung
auf structurelle und funktionnelle merkmale von starke. Staerke. 38:40-44.

Kokini, J .F . 1993. The effect of processing history on chemical changes in single- and
twin-screw extruders. Trends in Food Science and Technology. 4(10):324-329.

Kuntz, D.A., Nelson, A.I., Steinberg, MP, and Wei, LS. 1978. Control of chalkiness in
soymilk. Journal of Food Science. 43:1279-1283.

Langsdorf, A.J. 1981. Economics of soya protein products and outlook. Journal of the
American Oil Chemists’ Society. 58:338-340.

Levine, L. 1989. Scaleup, experimentation and data evaluation. Pages 57-90 in: Extrusion

Cooking. C. Mercier, P. Linko, and J .M. Harper, eds. The American Association of
Cereal Chemists: St. Paul, MN.

115

Liener, LE. 1981. Factors affecting the nutritional quality of soya products. Journal of the
American Oil Chemists’ Society. 58:406-415.

List, GR. 1989. Commercial manufacture of lecithin. Pages 145-161 in: Lecithins —
sources, manufacture and uses. B.F. Szuhaj ed. American Oil Chemists’ Society:
Champaign, Illinois.

Liu, K. 1997a. Agronomic characteristics, production, and marketing. Pages 1-24 in:
Soybeans chemistry, technology, and utilization. K. Liu, ed. Chapman and Hall: United
States.

Liu, K. 1997b. Chemistry and nutritional value of soybean components. Pages 25-113 in:
Soybeans chemistry, technology, and utilization. K. Liu, ed. Chapman and Hall: United
States.

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

Lorenz, K., Jansen, G.R., Harper, J. 1980. Nutrient stability of full — fat soy flour and
corn — soy blends produced by low — cost extrusion. Cereal Foods World. 25:161-2, 171-
2.

Mason, W.R., and Hoseney, RC. 1986. Factors affecting the viscosity of extrusion-
cooked wheat starch. Cereal Chemistry. 63(5):436-441.

Mercier, C., Charbonniere, R., Grebaut, J ., and de la Gueriviere, J .F. 1980. Formation of
amylose-lipid complexes by twin-screw extrusion cooking of manioc starch. Cereal
Chemistry. 57(1):4-9.

Messina, M.J. 1997. Soyfoods: Their role in disease prevention and treatment. Pages 442-
448 in: Soybeans chemistry, technology, and utilization. K. Liu, ed. Chapman and Hall:
United States.

Molina, M.R., Braham, J .E., and Bressani, R. 1983. Some characteristics of whole
cornzwhole soybean (70:30) and ricezwhole soybean (70:30) mixtures processed by
simple extrusion cooking. Journal of Food Science. 48:434-43 7.

Mustakas, G.C., Griffin, E.L., Allen, LE, and Smith, QB. 1964. Production and
nutritional evaluation of extrusion — cooked full — fat soybean flour. The Journal of the
American Oil Chemists’ Society. 41 :607-614.

Mustakas, G.C., Albrecht, G.N., Bookwalter, McGhee, J .E., Kwolek, W.F., and Griffin,
EL. 1970. Extruder - processing to improve nutritional quality, flavor, and keeping
quality of full — fat soy flour. Food Technology. 24:102-108.

Mustakas, G.C., Albrecht, W.J., Bookwalter, G.N., Sohns, V.E., and Griffin, EL. 1971.
New process for low -— cost , high — protein beverage base. Food Technology. 25:80-86.

116

 

Nelson, A.I., Steinberg, MP, and Wei. LS. 1976. Illinois process for preparation of
soymilk. Journal of Food Science. 41:57-61.

Ohren, J .A. 1981. Process and product characteristics for soya concentrates and isolates.
Journal of the American Oil Chemists’ Society. 58(3):333-335.

O’Toole, D.K. 1999. Characteristics and use of okara, the soybean residue from soy milk
production — a review. Journal of Agricultural and Food Chemistry. 47:363-371.

Paulsen, T.M., and Horan, F .E., 1965. Functional characteristics of edible soya flours.
Cereal Science Today. 10(1):]4-17.

Pisecky, J. 1978. Instant whole milk powder. Dairy Industries International. 43(8):5-6, 8-
10.

Rackis, J .J . 1978. Biologically Active Components. Pages 158-202 in: Soybeans:
Chemistry and technology. A.K. Smith, and SJ. Circle, eds. AVI Publishing Co.:
Westport, Ct.

Riaz, MN. 2001. Uses and benefits of soy fiber. Cereal Foods World. 46(3):98-100.

Sander, EH. 1989. Lecithin in beverage applications. Pages 197-206 in: Lecithins —
sources, manufacture and uses. B.F. Szuhaj, ed. American Oil Chemists’ Society:
Champaign, Illinois.

Schmitt, H. 1995. Agglomeration and lecithination. Food Technology in New Zealand
30(9):6.

Schubert, H. 1987. Food particle technology. Part 1: Properties of particles and
particulate food systems. Journal of Food Engineering. 6:1-32.

Schubert, H. 1979. Processing and properties of instant powdered foods. Pages 675-684
in: Food process engineering volume 1. P. Linko, Y. Malkki, J. Olkku, and J. Larinkari,
eds. Applied Science Publishers LTD: London, England.

Schubert, H. 1993. Instantization of powdered food products. International Chemical
Engineering. 33:28-45.

Schuchmann, HP, and Danner, T. 2000. Product engineering using the example of
extruded instant powders. Chemical Engineering and Technology. 23(4):303-308.

Schweizer, T.F., Reimann, S., Solms, J ., Eliasson, AC, and Asp, N.G., 1986. Influence
of drum-drying and twin-screw extrusion cooking on wheat carbohydrates, II, effect of
lipids on physical properties, degradation and complex formation of starch in wheat flour.
Journal of Cereal Science. 41249-260.

117

Scrimshaw, NS. 1981. World need for protein. Journal of the American Oil Chemists’
Society. 58:389-400.

Setchell, K.D.R. and Cassidy, A. 1999. Dietary isoflavones: Biological effects and
relevance to human health. The Journal of Nutrition. 129:758S-767S.

Slavin, J. 1991. Nutritional benefits of soy protein and soy fiber. Journal of the American

Dietetic Association. 91(7):816-819.

Soy Protein Council. 1987. Soy protein products. Soy Protein Council: Washington, DC.

Thomas, D.J., Atwell, WA. 1999. Starches. American Association of Cereal Chemists:
St. Paul, MN.

Torun, B., Viteri, F.E., andYoung, V.R. 1981. Nutritional role of soya protein for
humans. Journal of the American Oil Chemists’ Society. 58:400-406.

Traina, MS, and Breene, W.M. 1994. Composition, functionality and some chemical
and physical properties of eight commercial full-fat soy flours. Journal of Food
Processing and Preservation. 18:229-252.

Van Zuilichem, D.J., Stolp, W., and Janssen, L.P.B.M. 1983. Engineering aspects of
single and twin screw extrusion cooking of biopolymers. Journal of Food Engineering.
2: 157-175.

Van den Hout, R., Jonkers, J ., Van Vliet, T., Van Zuilichem, D.J., and Van ‘T Riet, K.,
1998. Influence of extrusion shear forces on the inactivation of trypsin inhibitors in soy
flour. Trans IchemE. 76(C):155-161.

Walstra, P. 1996. Dispersed Systems: Basic Considerations. Pages 95-156 in Food
Chemistry. O.R. F ennema, ed. Marcel] Dekker: New York, New York.

Webb, RA. Interpretation of particle size reported by different analytical techniques.
[Online] Available http://\va.micromcritics.com/ps_clzone_particlearticlehtml, April
19, 2001.

Wen, L., Rodis, P., and Wassennan, BR 1990. Starch fragmentation and protein
insolubilization during twin-screw extrusion of corn meal. Cereal Chemistry. 67(3):268-
275.

Whalen, P.J., Bason, M.L., Booth, R.I., Walker, CE, and Williams, P.J. 1997.

Measurement of extrusion effects by viscosity profile using the rapid visco analyzer.
Cereal Foods World. 42(6):469-475.

118

Wolf, W.J. 1970. Soybean proteins: their functional. chemical, and physical properties.
Journal of Agricultural and Food Chemistry. 18(6):969-976.

119