melt transformation extrusion of soy protein

/

MELT TRANSFORMATION EXTRUSION OF SOY PROTEIN; /'

A Thesis Presented to The Faculty of the College of Engineering and Technology Ohio University

In Partial Fulfillment of the Requirement for the Degree Master of Science

by

Corry S. Hendrowarsito ;;:;'

November, 1984

OHIO UNIVERSITY LIBRARY

Acknowledgements

The

author

Professor counsel

John

wishes

R.

to

express

Collier,

her

without

appreciation

whose

to

and

guidance

this study could have not been possible. Thanks is

al so extended to the facul ty of the Department of Cherni cal Engineering also

due

to

for

their

Indro

advice

Subowo,

and

help.

Special

whose

help

and

thanks

patience

is

were

invaluable. Finally, thanks is due to my parents and brothers whose support were unlimited.

i

ABSTRACT

Hendrowarsito, Corry Suzannadevi. M.S. November 1984. Chemical Engineering Director of Thesis: Dr. John R. Collier Title: Melt Transfor.ation Extrusion of Soy Protein (pp. 104, 42 figures, 12 tables)

The

purpose

Transformation As a

protein.

of

thi s

Extrusion result,

research Process

an

was to

improved

to

the

apply

the

extrusion

layered

fibrous

~1el

of

t

soy

texture

occurs in soy protein extrudates. Commercially, fibrous soy protein products are used as meat extenders and substitutes. The premoist soy protein was 1t

in a system consisting of 3/4 - d i a me t e r

extruded single

screw extruder,

an

eighteen

inch

Brabender

conditioner

zone,

and a uniaxial die having a deformation ratio of 24:1, 1/16 11 x

1/2 11

ribbons

process RPt4)

and

were

temperature moisture

produced. profile

content

Variables studied included oC), screw speed (40-80 (160-90

(30-40%).

The

effect

of

these

variables on die pressure, absorption, bulk density, product temperature, and extruder throughput was investigated using response

surface

Studies

analysis.

scanning electron microscopy were product structure.

i i

using

conducted

optical

and

to examine the

The

(300-1500 compared

r~TE

process

psi)

and

to

the

produced

longer

more

higher

residence

conventional

pressure

times low

(5-15

pressure

drops

minutes) extrusion

(less than 500 psi). Control

of both shear rate or stress, and temperature

profile were found to be the most important factors. Product temperature affected

by

and

operating

pressure

were

significantly

screw speed. Shear rate or stress, and pressure

decreased with increasing moisture. The best operating conditions for maximum texturization were a temperature

pr o f t. l e of 160 0 -135 0 -110 0 -50 0 C, 80 RPM,

and

Differential

40%

moisture.

employed to determine

the

scanning

calorimetry

was

crystallinity of the dough.

The

result indicated that DSC was not an appropriate method. Scanning

electron

microscopy

displayed

clearly

the

physical changes which occurred due to process conditions.

iii

Table of Contents

Page List of Figures

vi

List of Tables . .

i x

Chapter 1.

Introduction

1

2.

Background of Study . .

4

2.1

Protein

6

2.2

Soy Protein and Its Commercial Use

8

2• 3

Mechanism of Fiber Formation.

3.

Theory

11 16

Melt Transformation Extrusion Process (MTE) • • • • • • •

16

3.1.1 Shear Stress and Flow Induced Crystallization . . . . . .

17

3.1.2 Pressure Effect on Crystallization

21

3.2

Extrusion Cooking

24

3.3

Characteristics of Textured Protein Prod uc ts . . . • . . . • . . .

28

Response Surface Analysis (RSA)

29

3.1

3.4 4.

Description of Equipment and Material

31

5•

Experimental Procedure

40

5.1

Preliminary Experimentation

41

5.2

Experimentation

45

5.3

Specimen Testings

47

iv

Page

Chapter

6.

Results

7.

Discussion

...

74

8.

Conclusion

....•.•..

87

9.

Recommendation

Bi bl t ography

.......•.

...

•......•.

50

89 91

Appendixes

A. Experimental Data

•.••••••

97

B. Response Surface Analysis Program

101

C. Response Surface Analysis Results

102

v

list of Figures

Figure

Page

1.

Mechanism of Protein Denaturation.

12

2.

Structure of Spherulite .

18

3.

S u 99est e d

18

4.

Elongational Flow in a Converging Die . .

20

5.

Nematic Liquid Crystalline Form.

22

6.

Cross Section of a Typical Food Extruder.

25

7.

S c he (0 a tic Di a g ram for the Un i a x i a 1 - rib bon Die

33

8.

Schematic Diagram for the Fiber Die

34

9.

Photograph of Uniaxial Die Halves.

35

10.

Photograph of Fiber Die Pieces . . •

35

11.

Schematic Diagram for the Extrusion Process with a Melt Conditioner Zone . . .

38

12 .

Front View of the Extrusion Set-up

39

13 .

Simplified Extrusion Flow Sheet

45

14.

Extrusion Rate versus Screw Speed at different Moisture Contents ...

52

Extrusion Rate versus Screw Speed at different Process Temperature.

53

The Effect of Screw Speed and Moisture on the Extrusion Rate at 8 Constant Processing Temperature of 150 C (zone II). • . . •

54

The Effect of Processing Temperature and Moisture on Extrusion Rate at a Constant Screw Speed of 70 RPM ••..••....

55

15.

16.

17.

i~ 0

del for Fib e r For mat ion •

vi

Figure 18.

Page Die Pressure versus Screw Speed at different Processing Temperatures . . . . .

.

56

Die Pressure versus Screw Speed at different Noisture Contents . . . · . .

57

· · ·

19 .

· ·

20.

22.

·

Die Pressure versus Screw Speed for the Fiber Die Runs . . . . . . . . .

59

The Effect of Screw Speed and Moisture on the Die Pressure at a COBstant Processing Temperature of 152.8 C (zone II). . . .

60

· · ·

21.

·

·

The Effect of Temperature and Moisture on the Die Pressure at a Constant Screw Speed of 45 RPM

• •

•

• ••

••••••

61

The Effect of Temperature and Screw Speed on the Die Pressure at a Constant Moisture Content of 40 w/o • • • • • • • . • • •

62

24.

DSC Endotherm for Indium

63

25.

Typical DSC Endotherm of Texturized Soy Protein Product ... ...•..

64

The Effect of Temperature and Moisture on the Product Absorption at a Constant Moisture Content of 35 wlo . . . . . .

65

The Effect of Temperature and Moisture on the Product Absorption at a Constant Screw Speed of 70 RPM ••• ••••••

66

The Effect of Screw Speed and Moisture on the Product Absorptionoat a Constant Processing Temperature of 150 C (zone II) . • . •

67

The Effect of Temperature and Screw Speed on the Product Bulk Density at a Constant Moisture Content of 35 w/o • • • • • •

69

The Effect of Temperature and Moisture on the Product Bulk Density at a Constant Screw Speed of 76 RPM •...•..•.•••.

70

23.

26.

27.

28.

29.

30.

v; i

Figure

Page The Effect of Temperature and Screw Speed on the Product Temperature at a Constant Moisture Content of 27 wlo . . . . . .

71

The Effect of Temperature and Moisture on the Product Temperature at a Constant Screw Speed of 135 RPM . . .

72

The Effect of Temperature and Moisture on the Product Temperature at a Constant Processing Temperature of 140°C (zone II) . . . .

73

34.

Optical Micrograph of Fiber Die Runs, 12X . .

77

35.

Scanning Electron Micrograph of Run F4 shows porous structure, 700X . . . . . . . . . .

77

Processing Temperature Profile at different Heating Zones. . . . . . . . . . . . . . .

81

37.

Residence Time versus Screw Speed

83

38.

Scanning Electron Micrograph of Untexturized Soy Protein with Strands of Fibers, lOOOX.

84

Optical Microscope of Fibrous Structure of Run F3, 150X . . . . . . . . . • . . . . .

84

Scanning Electron Micrograph of Isolated Fiber of Run F3, 4000x . . . ...

86

41.

Scanning Electron Micrograph of Run 13, lOOOX

86

42.

SAS Program .

31.

32.

33.

36.

39. 40.

• .

101

vii i

list of Tables

Page

Table 1.

Extruded versus Spun Texturizing Ingredients

5

2•

Typical Composition of Soy Flours, Concentrates and Isolates

9

3.

Amino Acid Composition

9

4.

Changes in Characteristics of Soybean Protein at High Temperature Heating. . . . . . . .

15

Experimental Pattern of Processing Condition Code s . . . . . . . . . . . . . . . . . . .

44

Effects of Variables on Extrudate Characteristics . . . . . . . .

75

7.

Die Temperatures

97

8.

Flow Rates . . . . .

98

9•

Pressure Profiles.

99

5. 6.

10.

Extrusion characteristics

100

11 .

Regression Coefficients.

102

12.

Analysis of Variance

103

13.

Levels of Variables Significance on Extrudate Characteri sti cs • . . . • . • . . • • • ••

104

ix

Chapter 1

INTRODUCTION

The

texturization

simulate in

the

been

meat

food

has

and

extenders or total

because in

the

been

rehydrated

at

use

an

as

are

they

interested

texture,

and

and

price.

Those

to

products

have

used

meat

as

They

two

basic

processes,

wet

can

be

favorable of

the

changes finished

which

integrity

produced

spinning

of

products,

restructured

ti ssue.

topic

these

products

histologically

of

The

in

composition

that

[1].

be

imparts

structures and

extrusion

these

can

processing have fibrous muscle

products

significant developments Once

ingredients

attractive

texturally

protein

meat substitutes.

structure,

foods,

the

industry.

manufacturers

their

vegetable

been one of

engineering

texturized

Food

of

this

by

and

work

have

through

similar one

of

to the

thermoplastic

is

the

extruded

material. Theoretically, the temperature of extrusion varies from 80

of

0

to 17SoC (180 o-350 oF). There is the

protein,

resulting

which

pressure

Al though

the

plastic

extrusion,

ranges

process

characteristics

of

contains from

has

rel ied

food its

14

very little degradation

20

to

to

All

60 atm

heavily

extrusion

own.

40%

on

moisture.

(200-900 psi). the

cooking

aspects

The

of

theory has

of

some

production,

2

storage,

handling and

environment should be considered a-

long with economic considerations. At (MTE)

Ohio

University,

process

polymers

has

to

a

produce

been

melt

transformation

highly

investigated

oriented by

extrusion

semi-crystalline

Collier

[3J.

In

this

process a plasticating extruder supplies molten polymer to specifically

designed

dies

through

a

melt

conditioner

(medium pressure pipe). The molecules of the molten polymer are partially oriented by passing the material conditioner

zone

(2000-8000

psi)

through this

immediately

before

ex-

trusion through a converging die. As shown in this research, MTE

was

at

a

lower

process was useful

pressure

range

(500-2000

psi).

This

in enhancing the fibrous texture of soy

protein. Reports have dealt with the extrusion texturization of soy protei n. The product characteri sti cs are thought to be dependent speed, time,

upon

the

following

feed rate, moisture, and

protein

content

independent

variables:

product temperature, [3-5].

investigation was to apply the MTE

The

objective

screw

residence of

this

process to soy protein

and to observe a definite layered structure of fibers in the soy protein extrudate under the predetermined conditions.

can

Texture was used as the basic tool

of observation.

be

of microstructure,

which

viewed

as

originates

a

direct

from

consequence

chemical

composition

and

It

physical

forces acting upon it. Scanning electron micrographs of the

3

inner layers will soy

protein

electron

be

extrudate.

micrographs

soybean researchers

used to reveal

and

Advantages in

studying

the morphology of the of the

soy

protein

have

been

[6-8].

Optical

microscopic

using

the

scanning

ultrastructure shown

by

of

previous

observations

also been used in support of the textural observations.

have

Chapter 2

BACKGROUND OF STUDY

In for

recent years

the

meat-l ike

texturized

transformation texture

of

has

vegetable

powdered

received

soy some

protein

process

protein

into

acceptance

a and

popularity. The simulation of meat depends on such textural characteristics shear and

as

thickness,

friction

forces

smoothness,

[9].

It has

cohesiveness,

been

thought that

this kind of texture develops with the formation of fibers. Fiber

formation

can

be

obtained

through

several

process

which can be either chemical or physical. Many textured

new

processes

protein

have

products.

The

been two

developed most

basic

to

yield

industrial

processes for generating texture from proteins are spinning and

extrusion.

Spinning

of

protein

fibers

involves

modification of the isolated protein through solubilization in

alkali

protein

[10].

During

unwinds

and

the

deaggregates

dispersed flexible chains. spinning,

it

is

membrane.

Protein

alkali

forced fibers

treatment to

form

a

When the material

into

alignment

the

globular

series

of

ready

for

is

through

a

porous

(about 0.003 in diameter),

which

are partially oriented, are coagulated in an acid bath. The fibers are then stretched to a desirable strength and cut into

a

desirable

size.

The

stretching

causes

further

5

orientation of the protein fibers. On the other hand,' the is

a simpler

process.

of

the

thermoplastic

basic

thermoplastic extrusion process

Researchers

have

extrusion

detailed

process

variations

[11-14].

The

process involves plasticizing flour and water in an extruder to

high

temperatures

flashes

off

textured study (table

steam

and

product.

because 1),

and

and

This

of

its

its

pressure. expands,

technique advantages

similarities

The

emerging

resulting has

been

over

the

to

the

in

extrudate a

chosen

dry for

spinning

MTE

and this

process

process.

Other

processes that are less popular are gelation [15J and direct steam

texturization

method

of

[16].

Thus,

extrusion

texturization

[17].

Further

is

not

the

only

details

on

the

extrusion process and soy protein will follow in chapter 3.

Table 1.

Extruded

versus Spun Texturizing Ingredients [14J Advantages

Disadvantages

Thermoplastic extrusion

*Inexpensive *Simple process *Good protein quality *Can absorb water and fat *Thermodynamically effecti ve

*Limited use *Poor structured analogue texture *Flavor, color

Fi ber spinning

*Versatile *Good structured analogue texture

*Expensive *Technically difficult *Low protein quality *Flavor, color

6

2.1

Protein

Native protein molecules are known to be folded well-defined,

unique three dimensional

structures.

with

Princi-

pally the molecules of proteins are made up of carbon, hydrogen, oxygen, nitrogen, sulfur and some traces of phosporuse

The

acids.

protein

These

consists

of

acids

play

amino

p ol yme r t za t t on to

small

a

units,

very

called

important

amino

role

in

form a long chained molecule. They have

toe following chemical formulas typified by [18]:

lysine

leucine

CH 3 CH

CH 3

>CH~HCOOH NH

3

>CHyHCOOH

CH 3

2

valine

isoleucine

The

amino

(-NH 2)

NH 2

and

carboxyl

(-COOH)

groups

are

chemically active, basic and acidic, respectively. Thus the

7

amino

group

of

one

amino

acid

readily

combines

with

the

carboxyl group of another and forms a peptide bond at the center (eq. 1).

o

R

I

II

NH 2-R '-CH 2-COOH + NH 2-R-CH 2-COOH -- H2-y-C-j-I-COOH + R'

H

20

H H

(1)

dipeptide

The remaining free amino and carboxyl groups at the end can react with independent amino acids to form polypeptides. The

possibility

of

enormous.

This

variation

different

amino

acids,

variations depends

different

among

on

a

sequences

proteins

is

combination

of

of

amino

acid

wi thi n a cha in and di fferent shapes the cha in assumes. The chain can be coiled, are

responsible

proteins.

This

folded or straight. These differences

for complex

the

differences

configuration

in

texture

of a

protein

of can

the be

modified to form fibrous texture by subjecting the material to

external

properties

forces

utilizing

(dough forming,

emulsifying,

film forming,

thickening, gelling,

and others [19]).

protein

psychochemical

moisture holding,

stabilizing, cohesiveness

8

Soy Protein and Its Co.mercial Use

2.2

The

utilization

functional

and

economical

soy

physical

values.

The

including

its physical

reported

[20-22].

emulsification,

of

protein

properties,

functional

value

and chemical

Some

of

viscocity

on

nutritional of

soy

properties,

these

and

depends

holding

important in meat formulation. These functional

and

protein, have been

properties,

water

its

such

as

capacity

are

properties,

which contribute performance aspects in affecting structure and texture formation, outweigh their nutritive contribution [23]. There are protei n:

soy

concentrate

three flour

(65

types (1 ess

of

commercially available

than

65~

to 89% protein),

(90% and higher protein) [23-24].

protei n ) ,

and

soy

soy

soy

protei n

protein

isolate

All three types of these

products can be used to yield a range of textured vegetable protein; the cost increases with the protein concentration. A typical analysis of soy protein concentration is tabulated in

table

2.

Soy

concentrates

(70%)

is

used

as

the

raw

material in this study. There are

three dietary uses of texturized

protein (TVP) [26]:

vegetable

9

Table 2.

Typical Composition of Soy Flours, Concentrates and Isolates [25J Per cent (moisture-free basis) Soy flours Concentrates Isolates

Protein Fat Fi bre Ash Carbohydrates (soluble) carbohydrates (insoluble)

56.0 1.0 3.5 6.0 14.0 19.5

72.0 1.0 4.5 5.0 2.5 15.0

96.0 0.1 0.1 3.5 0 0.3

Table 3. Amino Acid Compo s t t ion'' [26J

Amino Acid Arginine Histidine Isoleucine Leucine Lysine Methionine Methionine + cystine Phenylalanine Threonine Tryptophan Valine

Soy flour 7.0 2.4 4.2 7.7 6.4 1.0 2.2 4.7 3.6 1.7 4.4

a I n, grams per 16 g N. bFood and Agticulture Organization

FAO b reference protein 2.0 2.4 4.2 4.8 4.2 2.2 4.2 2.8 2.6 1.4 4.2

10

(1)

Analogues:

products which are made to resemble another

product.

(2) Supplements:

products

which

are

to

made

a

meet

deficiency. They are not added for textural purposes but for their functional

properties, especially to bind fat

and moisture. (3)

Extenders: to stretch out food which is available. This is the most common use for extruded textured proteins. They can be used with meat to reduce prices and, in some cases, to improve quality.

It can

seen

the TVP

is

comparable to meat. Soy protein is known to contain all

of

be

that

in

nutritional

value

the essential amino acids needed by the human body, except it has a lower than desirable content of sulfur-containing methionine nutritive

3).

(table value

in

Hegarty

soybeans

and

by

Ahn

[27J

comparing

proved

soy-based

the meat

analog with ground beef. Finally, available and

soybean

protein

inexpensive.

is

It is

abundant,

commercially

the largest cash crop in

the United States, exceeding corn, wheat and cotton. used

extensively

in

the

food

industry.

The

price

It is of

the

texturized materials range from 27-45 cents per pound on a dry

basis,

which

after

hydration

translates

cents per pound meat replacement [25].

into

a

9-15

11

2.3

Mechanis. of Fiber For.ation

The mechanism of protein texturization during extrusion cooking

is

not

clearly

understood.

Many

researchers

have

reported that the extruder environment enhances the transformation

of amorphous

soy

protein

to

fibrous

microstruc-

tures [14,31,59-63]. A fiber is defined as a body of matter having a high ratio of length to lateral dimension and which is

principally

molecules

composed of

[28].

Fiber

can

longitudinally be

thought

oriented

of

as

a

linear

result

of

realignment of protein subunits that are disassambled due to pressure and

heat of the extruder environment.

alignment

done

is

by the

This

re-

shearing action of the extruder

[29,30]. Smith emphasizes that the cooking extruder has the ability

to

work

dough

to

restructure

and

retexture

the

proteins [31]. Thermal

denaturation,

texturization,

involves

which

is

gelation

the

and

key

parameter

restructuring.

of The

process is irreversible and is described through a sequence of steps. Figure 1 shows the formation of hydrogen bonds and amide bonds between aligned molecules in a denatured state. During heating, the ionic, disulfide, hydrogen bonds and van der

Waals'

forces

organizing

and

holding

the

native

globular proteins are interrupted and the hydrated proteins begin to unfold.

The relatively linear protein chains are

12

Native state

unfolding

Associating Amide bond

Figure 1.

Mechanism

of Protein Denaturation [23J

13

oriented through a shear environment, sites on adjacent

molecules

can

so that the reactive

cross-link

the

protein

to

achieve a fibrous texture [29,30,33]. Previous involves

reports

the

suggest

formation

that

certain

of

formation

types

of

of

fibers

intermolecular

peptide bonds (see section 2.1). The work of Cumming et al. [34]

describes

the

dissociation

subsequently weight fibers, of

the

pressure of

become

aggregate.

soy

temperature

protein

insolubilized studying

On

extruded

soy

protein

the

can

sulfur-containing

molecular

changes

molecules

by

lateral

{NH - CH- C0 2 2H)2

peptide

chains,

overlapping.

In ll

the

amino

which

are

Burgess

1976,

that lIisopeptide

It

of

is

the

acid

adding

believed curled

groups

protein bonds

between

[36]

an that

the

parallel

Stanley

soy

texture

cystine

generally and

spun

fibrous

elongated of

which

molecular

high

by

on

and

suggested

crosslinking may playa role. They assumed

crosslinking

bonds between

by

form

improved

reaction

formed

subunits

that the

be

influences

formation

adjunct.

in

occur

into

and

Jenkins [35] demonstrated

elemental

that

and

of

free

protein

carboxyl

chains

occur

and amino acid

through

amide

side groups of

the protein chains. The

energy

for

consisting

of

determined

using

endothermic

the

breaking

endothermic and

differential

(90-100

KJ/KG)

forming

denaturation of

scanning [37].

new

bonds

calorimetry

Sensible

process

heat

was to

be

changes

14

occuring

because

of

temperature

rise

in

the

product must

also be considered [37]. Qualitative

changes

in

soy

protein

temperature heating are shown in table 4.

during

high

texture

expans i on property hard fragile ..,..

increase

rapi d i ncr-ease-s--e low decrease

120

binding force (degree of aggregate)

110

rapid decr-ease-s--s l ow increase

intact

105

solubility

cross-strlicture of subuni ts

100 140

.1-

.. I-

. 1*

150

170

rapi d decrease--

rapid increase--

degraded----

160

like s o l - - - - - - -

rap; d decrease-------soft elastic • to.

....

1 i ttl e degraded

130

Changes in Characteristics of Soybean Protein at High Temperature Heating [29J

Temperature of heating (OC)

Table 4.

01

......

Chapter 3

THEORY

When a bulk external

polymer

is

there

is

forces,

crystallites or molecules. the

degree

is

greatly

no

polymer

the

mechanical

and

orientation

of

which is defined as

chains

influenced

ternperature gradients in the system. oriented,

in the absence of

preferred

Orientation,

of alignment of

direction,

crystallized

in

a

particular

deformation

by

and

As the polymer becomes

physical

properties

improve

[28,38].

3.1 Melt Transfor.ation Extrusion Process (MTE)

The MTE is a thermoforming process. The objective is to deform a

polymer melt,

direction

or

and to align the chains in a common

directions.

This

process

has

advantages

over

other orientation processes, since orientation is induced in the molten state. quite can

be

In the molten state the deformation can be

influential; affected

molecules,

the

organization

either

directly

at or

all

dimensional

indirectly:

aggregate-crystallite,

the

the

levels basic

crystalline

amorphous entity, the single crystal lamella, and the larger

17

aggregation, called spherulite [39]. The orientation due to the

deformation

polymers,

as

may

well

be

as

developed in

in

glassy

crystalline

or

polymers.

amorphous Since

the

amorphous chains have not experienced crystallization, they do not gain appreciable strength by orientation because they fail

by

separation

crystalline

rather

polymers,

than

the

by

chain

crystallization

scission. is

In

enhanced

the by

chain alignment.

3.1.1 Shear Stress and Flow Induced Crystallization

In

the

molten

unoriented

state,

the

linear

molecules

are randomly coiled. Upon supercooling, the polymer tends to crystallize

and

form

spherulitic

structures

with

no

macroscopic orientation (figure 2). The stacking of parallel lamellae of the

substructures

produces

a

high

local

order

among the amorphous or disordered regions. Flow

induced

produce

a

forming

lamellae

preferred a l i q ne d

in

high

crystallization

deformation. to

alignment, the

a

Mechanically,

begin

to

such

that

orienting

and

slip the

direction.

from

shear

field

can

this

causes

the

their

polymer The

originally

axis

extension

becomes due

to

flow of the folded chains forms stacks of parallel lamellae that can be either along or against the lamellae axis [40]. Figure 3 shows this behavior in a crystalline polymer.

Sphtrut,'ic choins folded at riQht anQ~ to main alis

18

Amorphous Inter - spherulj,ic

material Defect, in fibritt

Amorphous intff- fibril tor material Single .> crystal nucleus

Figure 2.

(a)

Spheruli'e

Structure of Spherulite [28J

~~---_.

(b)

Figure 3.

Suggested Model for Fiber Formation (a) By unfolding of molecules from more than one lamella, (b) By gradual chain-tilting, slip, breaking off blocks of folded chains

19

Upon approaching

the entrance region

of

the

die,

the

velocity distribution of a molten polymeric material changes to

a

"wine

glass

stem

shape"

(figure

4).

The

flow

streamlines converge rapidly inducing an elongational effect of the previously random coiled polymer chain and giving a higher degree of orientation [41-42]. The rate of uncoiling of the polymer chains at the converging section depends on the

deformation

ratio

and

the

type

of

polymer

on

the

processed

[43,44,48]. Previous describes

the

contribute

to

work

using

four

important

the

amount

plastics

of

processing orientation

process

MTE

conditions in

which

the

extrudate

having

reduction

[43,44,46-52]: (1) die design and deformation ratio (2)

screw or line speed

(3)

operating pressure

(4) temperature profile

The

MTE

process

has

been

used

with

dies

ratios from 2:1 to 16:1, with half angles ranging from 10° to 26°, and die geometries uniaxial

or biaxial

that deform the melt in either

directions.

Furthermore,

fiber,

ribbon

and more complex dies have been used along with this process [44,46-52].

Extrusion rates,

controlled partially by

speed, govern the level of deformation

as well as the orientation.

screw

on the polymer melt,

20

!

,

:

1

n

(= 0

n f 0 rm d t ion

z Flow streamlines

Crystal growth tr o nt and i so t he rma 1 line

Figure 4.

Elongational Flow in the "Wine Glass Stem" Region of a Converging Die

21

3.1.2

Pressure Effect on Crystallization

As

proposed

by

Brown

[53],

the

development

of

the

extended chain crystals may be related to the formation of the nematic Illiquid crystals." A nematic structure consists of

a

parallel

stacking

of

rods

with

relatively

perfect

internal structure, but not necessarily matched from end to end (figure 5). Collier postulated that a liquid crystalline form could occur in the materials studied under temperature,

pressure,

and

field

conditions

critical [2].

This

behavior of different crystalline structures (polymorphism) is not limited to the simpler polymers but is also observed in proteins and synthetic polypeptides [54,55].

In the case

of synthetic polymers, the working pressure for MTE ranges from

2000

to

8000

psi,

which

is 1/4-1/5 that of a solid

state extrusion [56].

Thermal properties Earlier

observations

of

oriented

(extended)

polymer

have shown a higher melting point than that of a random melt (quiescent) [43.44,46-52]. In terms of entropy change, (2)

The melting points are, (3)

22

y

)-'2 X

(

(

D

Figure 5.

Nematic Liquid Crystalline Form

23

(4)

where subscript f stands for fusion, q for quiescent, and ex for extended. As, (5)

then, T

m

ex

/ Tm

q

= 6.5 f

q

/ ~Sf

(6) ex

from ( 2 ) ,

Tm

ex

/ T

mq

>

1

or

T

me x

>

Tm = Tm - Tm ex q Hence,

the melting point of polymeric material

related to its degree of orientation.

T

m

( 7)

q

(8 )

is directly

3.2

Extrusion Cooking

Extrusion,

in

general,

refers

to

the

shaping

of

the

products to the desired size and consistency by forcing the material through a die under a high pressure. Extrusion has long been used in the food industry in the making of special shapes

of

food

products

(e.g.

macaroni,

bacon

bits).

Previous researchers have shown that the extrusion process ca n produce

mea t 1i ke

fi ber s

These

[14,31,57-61].

repor ts

provide the "s t a t e of the a r t " of protein texturization by using the extruder. The basic patents of soy protein extrusion are those of Atkinson

[62]

and

Jenkins

[35].

The

newer

patents,

which

were an improvement over the prior patent, did not use a die on

the extruder

and

therefore

had

a

lower

pressure

drop

(below 200 - 500 psi); the resultant product's characteristics

were

less

spongy,

less

hydrated,

and

more

fibrous

[57-59]. Food extrusion owes much of its design and plastic science [37,42]. there

are

artificial

differences polymers

and

However, between protein

theory to

it should be noted that plastics as

as

'chemical',

"b t opol yme r s !

,

natural

polymers. Zuilichem [63] explained these differences as:

1.

Biopolymers shows no spontaneous melting-temperature or trajectory but simply need a certain amount of shear to

25

plasticize the protein-water mix. 2.

The biopolymer is highly sensitive for a long time span of exposure to heat and pressure. It

3.

is

important

that

some

water

be

present

during

extrusion to assure a continuous working condition of an extruder.

The main components of a food extruder are the same as those

a

of

compression

thermoplastic screw,

barrel,

extruder. d t e Ls ) ,

are:

They

and

heating

feeder,

system.

In

this process, moistened products are plasticized in a tube by

a

combination of heat,

Figure

8

shows which

barrel,

the i s

basic divided

pressure, process into

and

in 3

mechanical the

food

stages:

shear.

extruder

mixing

and

compressing, heating and cooking.

DRIVE, GEAR

a

RE DUCER THRUST BEARING

FEED HOPPER

COOLING

WATER JACKET

\

PRESSURE TRANSDUCER

THERMOCOUPLES /

DIE

DISCHARGE THERMOCOUPLE

BREAKER PLATE FEED SECTION

COMPRESSION SECTION

METERING

SECTION

BARREL WITH HARDENED LINER

SCREW WITH INCREASING ROOT DIAMETER

Figure 6.

Cross Section of a typical food extruder [39]

26

(1) Mixing and Co.pressing (feed zone) The moi stened material enters the extruder through the feed zone. The relatively free-flowing granular particles of the

meal

cause

a

turbulent

like

pattern

in

the

intake

section of the extruder. This flow insures intimate contact of protein with water with food.

Then

the

screw

very

further

little

internal

compresses

and

shear of mixes

the

action,

and

product. No cooking is desired in this zone [64].

(2) Heating (transition zone) The

second

zone

continues

the

mixing

concomitantly imparts heat into the mixture due to shearing action of the screw. This heat is used by the proteinaceous material solid

to

to coagulate and polymerize. This transition from a

fluid

is

associated

with

a

set

of

chemical

reaction called 'cooking'.

(3) Cooking (.etering section) The meaning of cooking

here

is

the conversion and/or

reaction of the major food constituents - carbohydrate, fat protein, occur

and

with

water. food

Two

types

biopolymers

of cooking are

reactions which

protein

denaturation

(section 2.2) and starch gelatinization. In these reactions, water and food materials themselves interact to create new, altered forms which have a distinctly different rheological behavior.

This

cooking

process

is

time

and

temperature

27

dependent, which probably changes with the concentration and quantities

of

the

chemical

environment.

Food

alteration

the

of

texturizing

species

extrusion feed

process

present

results

ingredients

and

in

this

and

in

through respect

the

the the

is

shear

chemical

cooking and

significantly

different from the melting processes, which occur during the extrusion of the words

thermoplastic

'melt' or

resin.

Thus,

the

application

'melting' is a misuse [39J.

Most of the cooking is done in this critical cooking

is

mainly

done

by

shear heat as the material the

die.

The

of

highly

externally

supplied

zone.

and

The

viscous

is conveyed through the barrel to

turbulent

flow

pattern

is

transformed

into a laminar flow to minimize back flow across the protein strands.

this

At

stage

the

materials

are

simultaneously

oriented and coagulated in the direction of the chamber. During

this

whole

process,

the

viscocity

properties of the dough can differ drastically. about

this

Hermansson

is

very

[66],

the

limited.

According

viscosity

of

the

to

and

Information

Briskey

system

physical

[65]

changes

and with

the degree of protein hydration.

Therefore,

the

variables

of

conditions are,

1. temperature profile

2. screw speed/line speed 3. design of die(s)

extrusion

processing

28

4. moisture 5. pressure profile, and 6. residence time

3.3 Characteristics of textured protein products

A variety

of

texture and other These

tests

extrusion maintain of

are

tests

properties used

conditions quality

results

have

of

used

determine

on

product for

different

to characterize the

textured

to

standards

from

been

the

protein effects

products. of

varying

characteristics,

production

runs.

investigations

is

and

to

Comparison difficult,

because no standard set of tests is used [37J. Only

two

laboratory: value). extruded degree

bulk

Bulk

absorption

of

p o r o s i ty

and

gives

products,

is an

tests

density

density

dried of

types

water

the

while

of

important

were

the

possible

absorption

degree water

of

functional

of

the

gives

the

textures.

property

our

(hydration

expansion

absorption

products'

in

of

Water

textured

protein products as the products are used after rehydration. This

value

gives

absorption and

an

indication

of

the

retention capabilities.

vary in different laboratories.

extrudate

maximum

Hydration conditions

29

3.4 Response Surface Analysis (RSA)

It

is

convenient

to

visualize

geometrically

the

relation between response and the various factor levels. RSA method

represents ( or

factors

the

response

independent

assuming

by

that exist

variables,

when in

k

an

experiment, the response (or dependent variables) will be a function of the levels at which these factors are combined [61].

( 10 )

The function ~ is called the response function. The response

surface

is

represented by a polynomial.

For the case of three variables, a quadratic polynomial was proven adequate to fit the data [3-5]. The model is,

( 11 )

The

above

first

equation

and

second

takes

into

degree

as

account well

variations as

those

due due

to to

i nt e'r act ion s • Response region are

is

of

the

represented

obtained

by

independent variables by

making

contours. one

a

in

These surface

variable

equal

to

a

certain contours constant

30 value

and

then

solving

the

fitted

equation

as

a

quadratic

equation in the other two. The

application

of

quite popular [3,45,68].

this

method

in

food

industry

is

Chapter 4

DESCRIPTION OF EQUIPMENT AND MATERIAL

A.

Extruder A

laboratory

Brabender,

single-screw

Model

extruder,

had

200,

the

C.W.

following

specifications: barrel diameter - 0.75 11 ; LID 20:1; feed

hopper

gravity

electric heaters, watt

heaters

on-off

by

monitored

speed

heating

2

zone

independently controlled by 800

proportional

variable

feed;

two

West

controllers;

motor

assembly,

JPC

Model

drive

unit

equipped

with

a

tachometer, and capable of controlling screw speed from

0-200

RPM.

The

discharge

indicated on a West Model with a range of

was

controlled

Controller,

a by

pressure

was

1586 pressure indicator

to 10,000 psi. The motor speed a

Fincor

manufactured

by

2400

MKII

INCOM,

Motor

DC

International

Inc.

B. Extruder Screw The screw used was made of 4140 chrome alloy with a

standard

flights

with

compression increasing

ratio

of

2:1.

It

had

screw root diameter

20

from

0.475" to 0.605 11 with a 0.608" axial channel width

32

and

a

0.007

11

flight

clearance.

Angle

of

the

helical screw was 25°.

c.

A Melt Conditioning Pipe A 15 11 with

length 111

diameter

pipe

outside was

(medium

pressure-lO,OOO

diameter

used

as

a

and

0.687"

connection

psi) inside

between

the

barrel and the die •

. ;

D.

Extruder Dies The dies were made of 304 stainless steel: (a) A split die with a uniaxial deformation ratio of 6:1 was used. The die opening consisted of a

slit

1/16 11

thick

and

1/2 11

wide

which

produced a tape or ribbon like extrudate. (b) A fiber die with a circular opening of 0.020" diameter and was

fitted

111

to

length was also used. This die a

holder

and

produced

a

string-like extrudate. Each of the above dies produced a vertical, downward extrusion and were heated with fitted 600 watt

heaters

control

board.

controlled

Figures

automatically

7 -

from

the

10 show the designs,

dimensions, and views of the dies.

33

o •

-1/16" 1/2"

J

L---

-2 1/4"-----3/16"

- - - - - - - - - 3 1/2 ..- - - - - - -......

Figure 7 .

Schematic Diagram for the Uniaxial-ribbon Die (A) Side view, and (B) Top view

34

4.

.. ........ t:-:-:-:·:-:-:-:· ......... ......... .......... t:::::::::::::::: ~:

\9

.

I

-CD

~,

4

II

8 /~ 1 --.

-

~

f

=

It)

",...

•

...'".

"-

::::::::::::::::~

" J

10

QJ

..c: -4-> S-

o

4-

E rtj

SO)

9 /1.--+ It

-+--

rtj

0 U

r

.."

~::::::::::::~

N

,... "' r-

r'i"••••••, ••••••••

=f

....

PI)

1

:::::::::::::::::Il

.~

=r ,....

~:::::::::::::::i

L

-4->0

~

~:::::::::::::::

-

QJ

.,.... .,....

..

rtj

E

S-

QJ

Q)

..c: ..0

u·,....

U1 u,

"•

co

I

QJ

S~

0)

..

lJ..

35

Figure 9.

Figure 10.

Photograph of Uniaxial Die Halves

Photograph of Fiber Die Pieces

36

E.

Controllers Two of the three Gardsman temperature control units manufactured by West Instrument Corporation were used to control They

had

range

a

peratures

in

controlled

with

temperatures Q-800oF

of

the

piping

two

Love

in the barrel. The

(425°C).

and

the

Model

52

dies

tem-

were

controllers

mounted on the control board. They had a range up to 400 oC. Temperatures and pressures in the barrel and the die were sensed by Dynisco strain gauges, r~1 0 del

TPT 43 2 A-I QM- 6 / 18 ,

and

measur e d

b yaW est

Model 15-86 and a Dynisco Model ER 478Al pressure gauges.

F.

Optical Microscope A Wild

M5A

Stereomicroscope

observations

overall

magnification

depending

on

Photomicrographs

1.4X

optical

the

used

for

1 abora tory.

was

range

the of

the

i n

texture

was

structure

to

The

20QX,

combination. were

taken

by

MPS15/11 Semiphotomat (632.8 mm) assembled on the

M5A

Stereomicroscope

using

a

35

mm

film

(ASA

400/DIN 27).

G.

Scanning Electron Microscope (SEM) A Hitachi

Model

HHS-2R

Scanning

Electron

37

Microscope

was

used

to

positive/negative black type

665

viewing

(ASA

75/DIN

the

photograph and

The

20).

three-dimensional

white

sample

Polaroid

SEM

is

on

film,

capable of

structures over a range

of 20-280,000 magnification.

H.

Sputter Coater Prior

to

SEM

examination,

the

were

samples

coated with gold or gold/palladium deposition in a Hummer V sputter coater, manufactured by Technics.

I.

Differential Scanning Calorimeter A

Perkin-Elmer,

scanning

calorimeter

melting point of

to

connected

recorder

to

a plot

DSC-1B

Model

soy

was

used

protein.

Perkin-Elmer,

the

rate

of

differential

to

detect

the

This equipment was Model

heat

56

chart

input

versus

temperature.

J.

Soybean Defatted Soybean protein concentrate, PROCON 2000,

was

Decatur,

obtained 11.

It

from

contained

A.E. 70%

solid basis and 5-7% moisture.

Staley, protein

Mfg. on

Co.,

a dry

o

Figure 11.

I

Tacla•••t.r

8arr •I

1

Zone 2

Z0 ne

Zonl 4

l e o n d I t Ion. r

Zone 3

1- Die ~

Schematic Diagram for the Extrusion Process with a Conditioner Zone

Zonf

~

Gauae

r - - - - - - - - Pr••• ur. - T. . p.raturt

,

w

cc

39

c:

o .,...

OJ

..s::

+-> 4-

o

.,... :>

Q)

S-

::s 0')

u,

Chapter 5

EXPERIMENTAL PROCEDURE

All

the

prepared

by

Brabender

samples

tested

extruding

in

premoist

laboratory extruder,

this

investigation

soybean Model

flour

200.

in

a

were C.W.

To complete the

screw assembly, a conditioner zone (18" spacer) was placed between

the

additional

die

plate

and

the

barrel,

which

provided

volume after extruder screw discharge.

Previous

workers proved that the conditioning zone improved pressure uniformity time

of

behind the

the

die

material

plate,

in

the

increased

the

residence

extruder,

and

improved

crystallization [51,52]. All

compression

torqued to 75 ft-lb up,

the

die

fittings

in

the

assembly

(figures 11 and 12).

channels

were

cleaned

from

line

were

Prior to setting old

polymer

by

sanding with 600 grit sandpaper. The ribbon die halves were assembled with six 0.25" X 2.5 screws,

which

t he r mo c o upl e s ,

then

were

transducers

11

grade eight socket head cap

torqued were

to

tightened

90 to

ft-lb.

All

prevent

any

leaking during operation. Independent

variables

selected

for

the

process

were

temperature, feed moisture and screw speed. The selection of these critical variables were based on findings reported by previous researchers and through preliminary experimentation

41

[14,31,59-63]. The dependent variables are pressure profile and line speed.

6.1

Preliminary experi.entation

The profile

first objective was and

processing

to

find

conditions.

the

The

best temperature

extrusion

assembly

was divided into 5 temperature zones: I and II - the barrel, III and IV - the conditioner zone, and V - the die (figure 11).

Through

preliminary experimentation,

it

was

necessary

to force feed the material through the hopper. The premoist soybean feed

was

ground

hopper.

generated, clogged

Because

the

the

the

by

of

soybean

feed

screw and

the

considerable

developed

inlet.

pushed

This

a

tacky

effect

was

heating the section nearest to the hopper.

back

amount

into the of

steam

consistency

and

reduced

not

by

If this section

were heated, the steam would be absorbed by the incoming soy material. The steam caused caking and made smooth operation impossible. The

temperature

determined

by

decreasing

temperature

better

choice

careful

than

settings

for

observation

of

an

assembly

of extrudate

distribution that

the

toward

the

increasing

distribution. The former case had two advantages:

were

quality.

A

die

a

was

temperature

42

(a) Most of the cooking was done in the barrel zone. A decreasing

temperature

distribution

prevented de-

gradation of the material. (b) The

material

did

not

extrude

at

too

high

a

temperature in the die. Excessive expansion caused by flashing steam could destroy or seriously limit the formation of the fibrous structure, however, a certain amount of expansion of the product was also important in order to obtain a fibrous structure.

In the past, a steep temperature gradi ent was appl ied to

enhance

[48,51,52]. the die

and

freeze

the

highly

oriented

This was usually done by immersing

in a water bath as

a cooling

caused the pressure to build up.

medium,

In this study,

extrudate the tip of which

also

the effects

of the die land temperature gradient were not observed to occur. The die temperature was heated to SOOC, temperature caused the material

since a lower

to stop flowing out of the

die passage. Too high a temperature (100°C) at the die made the

product

emit

separated

bursts

of

burnt

individual

pieces. It appeared that some pieces would stick in the die nozzle until the pressure built up sufficiently to dislodge them. The material near the end of the die expanded rapidly, producing a rapid outflowing of material

which

fragmented

into individual pieces. This product was unassayable.

43

Once

particular

processing

temperatures

were

set,

a

series of experiments with the same temperature setting were conducted. This reduced excessive use of raw material during the transition periods to a new temperature settings. Since the extruder was not self emptying,

too little

moisture, too high a temperature, and too high a compression ratio were all avoided because any of these would cause the materials

remaining

in

the barrel

to

harden and

lock

the

screw [69]. A blocked extruder, due to overheating or high frictional drag of the product, costs a considerable amount of maintenance time for dismantling, cleaning and repair. An experimental design was chosen with three levels of temperature,

three

levels

of

moisture and

four

levels

of

screw speed to allow estimation of second order effects in the

empirical

statistical

variables (table 5).

model

for

three

independent

A2

82 C2

Al

B1 C1

30

35

40

C3

83

A3

C4

84

A4

80 100

02

E2 F2

E1 F1

60

01

40

R P

~1

F3

E3

03

F4

E4

04

80 100

150 - 125 - 100

Note: The temperature at zones I and V were unheated and 50°C, respectively

60

RPM

140 - 115 90

H2

HI

12

G2

G1

II

60

40

RPM

13

H3

G3

14

H4

G4

80 100

160 - 135 - 110

Processing temperature profile, Zones II - III - IV (OC)

Experimental Pattern of Processing Condition Codes

40

wlo

M0 i stu r e

Table 5.

~

+::at

45

Experi.entation

6.2

a WET

DRY

...

SOLID - LIQUID

.....

INGREDIENTS

BLENDER

.......

AFTERDRYER .

,.-

EXTRUDER.

Figure 13. Simplified extrusion flow sheet

Figure added

to

residual

13

the

shows

soybean meal

prior

flowsheet. to

Moisture

was

because

the

extrusion

moisture content of the meal

is normally very low the

a simplified

after oil

extraction

(5-7 weight percent or w/o)

[69]. As

present design did not allow direct water addition

the extruder, a food processor was

in

used for moistening the

powder. In order to have a uniform product, a food processor was

used

was

added

to mix

the dry flour with water. Distilled water

slowly

along

with

the

continuous

mixing

and

breaking action of the steel blade, so that it maintained a free fl owi ng movement of powder to prevent the development of large aggregates. Water addition was accomplished in 3-5

46

minutes and

mixing

ceased after

an

additional

3 minutes.

Batch sizes were normally about 300 grams of dry blend. Once the temperature settings on the extrusion system were reached, the motor was turned on and

the screw speed

was adjusted to achieve the desired tachometer setting. Then the hopper was fed with premo;stened soybean meal. In order to achieve a continuous feeding, the mix was hand-fed to the extruder hopper. An excessive amount of mix in the hopper prevented

free

flow

of

the

material

because of caking or bridging of

into

the

ingredients

in

extruder the

feed

hopper. Sufficient time (20-30 minutes) was allowed in order to have a steady state system. Estimation of the steady state was based on the enough material

temperature and pressure readings. After

at each shear rate had been produced

(.!.15

feet), the screw speed was changed to another desired shear rate. Elapsed time was allowed for the transition period (20 minutes). Data collected consisted of the steady state values of temperatures in all zones in degree Celsius, pressure at the exit

of

the

barrel,

pressure

at

the

die

in

psi,

and

extrusion rate in in/min. Table 5 shows the variations of variables selected. Extruded plastic

bags,

refrigerated.

samples labeled

were

collected,

with

the

placed

extruder

run

in

sealed

codes

and

47

The second objective was to analyze the effect of using higher

pressure conditioning.

Higher pressure drop

at the

die was attempted. This was done by replacing the ribbon die with a fiber die. Temperature profile chosen was unheated160-135-10Q-50oC and screw speed of 40, 60, 80 and 100 rpm.

c.

Speci.en Testings

All

the

samples

were

photographed

and

tested

for

moisture absorption capacity, bulk density and thickness. It was necessary to examine the specimens as soon as possible because

the

microbial

extrudates

and

will

not

remain

fresh

due

to

enzyme action. Under refrigerated conditions

the material lasted only for 2-3 weeks.

Water absorption capacity was evaluated by soaking 50 grams of extrudate segments in a beaker filled with 200 ml water. After 15 minutes of rehydration, the excess water was removed by Afterwards,

draining the

with

sample

a was

tea

strainer

reweighed.

for

The

15

seconds.

percent

water

absorption was calculated as the percentage weight increased based on the dry weight.

Bulk

density

was

determined

by

weighing

12-in

long

extrudate. The volume was obtained by multiplying the length by

average width and

was 40.8S;

thickness.

Average degree of puffing

puffing is defined as the degree of extrudate's

48

volume expansion due water vapor. The

to pressure drop and

flashing of the

product density was obtained by dividing

the weight by the calculated volume.

Microscopic stages, f4 i

Examinations

optical

c r 0 9 rap h s (S EM).

were

taken

microscopic Sam p1 e s for

immediately

0

after

were

divided

and

Scanning

into

two

Electron

ptic a 1 microscopic studies

they

were

extruded,

because

they were still moist and easy to layer. Preparing samples for SEM was more complicated than for the optical

microscope. However, only a small

sample

be

can

viewed

at

one

time.

The

area of the

samples

obtained

during the extrusion were frozen in liquid nitrogen. Samples for

SEM

were

placed

onto

specimen

a

stub

covered

with

double-coated cellulose adhesive tape. The area around the specimen was coated with a small streak of silver conductive paint in order to minimize charge build-up from the primary electron beam.

Afterwards,

gold-palladium

(60:40)

specimens microscope,

were

in

examined

Model

the

HHS-2R.

a in

specimens were coated with sputter

a

The

Hitachi

coater.

The

scanning

photographs

were

coated electron

taken

on

positive/negative black and white Polaroid film (ASA 75/DIN 20) •

Differential

Scanning

Calorimeter

(OSe).

Samples

were cut into thin pieces and weighed to the nearest tenth of a milligram. They weighed approximately 5-15 milligrams. Then they were sealed into specially designed aluminum pans

49

supplied by Perkin-Elmer and placed on the Perkin-Elmer DSC unit. The instrument was calibrated with a standard heavy Indium sample (163.S oC melting point) at 20°C/min and a full scale deflection set at a full

of eight millicalories.

The recorder was

scale range of five millivolts and the chart

speed was set at 40 rom/min. Statistical

of

a

stepwise

performed using

Design.

multiple

The data were analyzed by means regression.

The

analyses

were

the extrudate characteristics as dependent

variable versus the processing temperature, screw speed, and moisture.

All

possible

subsets

of

the

regression

were

performed using the SAS package [70]. Then, response surface plots were made from the derived regression equations.

Chapter 6 RESULTS

A series of experiments was conducted according to the above design. The

intent was

to

investigate

the effect of

independent process variables upon dependent variables. The protein

concentrates

used

on

all

runs

were

assumed

to

contain 5% moisture prior to any water addition. The results of response surface analysis are tabulated in

tables

plots

11

and

include

12

all

predicted data.

in

the

These

Appendix

C.

The

experimental plots

response

design

illustrate

surface

data

and

the

the contour of the

dependent variable against two of the independent variables, while

setting

analysis e.g.

one

usually

highest

of

the

predicts

output,

the

highest

variables area

constant.

with

absorption

Response

optimum response, rates,

etc.

The

shape of the optimum, the "center of the s y s t e ra'", can be a maximum, minimum, or a mix of the two, a "saddle point". The results

of

the

dependent

variables

of

this

study

show

a

"saddle po i nt " which implies the existence of two distinct regions of maximum yield a two peak system (figures 23, 29, and 32).

The

area of the two peak system means

that there

are two maximum peaks in the system. Sometimes the center of this

area

is

found

outside

the

experimental

design.

The

surface in this region of the experiments represents either

51

an inclined ridge or an inclined trough. The

effects

of

flow

rate

volumetric speed.

Figures

moisture

14

contents

screw was

and

15

and

speed

on

primarily show

process

extrusion a

the

rate

function trends

of

at

t emp e r a t ur e s ,

or

screw

different

respectively.

Moisture content effects were more significant than that of processing

temperature.

caking

the

of

material

Higher t

moisture

reducing

content

output

produced

rate.

Process

temperature effects were more dramatic at lower screw speed and leveled off at higher screw speed. three

Effects of all

the

processing variables are represented by the response

surface plots in figures 16 and 17. For e xamp l e , in figure 16, the effect of screw speed and moisture on the extrusion rate

at

symbols.

a

constant

The

temperature

darkest

symbols,

figure with a value of 62.18

is

at

represented

the

to 69.53

upper

left

inches

five

by

of

the

per mi nu t e ,

represents the highest value range of extrusion rate shown in this figure. The value occured at a screw speed of 90 to 100 RPM at a moisture content of 20 to 22.5 weight percent. Decreasing

extrusion

rates

are

represented

by

the

other

symbols along contour lines, at roughly 15 inches per minute interval. Figures 18 and

19 depict the pressure profile at the

die versus screw speeds. This pressure was an indication of how much energy was required to force

the material

the die orifice. To overcome high frictional

out of

forces in the

52

RUNS.

41

ZONE II .

0

...!L

'30%

A

H ---

35%

I .........

40%

8

Vl ~

c:

OJ ~

c:

o

u

.

....-..c:

,.....

-

:a: "",

-

"'-

c

c:

OJ

s;

::::s ~

Vl

o

::::

.. . D

-0 Q) Q)

.

0Vl

3:

OJ

s,

z o

u

.

-

( /)

V>

-a:

(J

J

Vl

::::s

Vl

t-

...

Q)

I&J

:2

Q)

a:

LLI

)(

>

::)

~

..J

0:::

>

c: o

o

19

s;

res

Vl

:::::s s;

~

x

w

15 OJ

s;

:::::s

en

o

20

40 SCREW

80 SPEED

(Rpm)

80

L.L..

lX &II

iii J .:

z o

a:

C

11.I

...

....,

'-a:

&:

-2

,...

5

30

...

0

»:

I

/'

2.0

I

15.

..... ..... ..... /

1 ·····F; gure

r

./

/

/

.

/ /

/

/

~

....... .

/

...

/

/

.

..

/'"

/

...

/'

...

/

C:i

o

o

. .. .....

/

... .

...

/'

F

c

RUNS

/'

.... •••••

//

c 160·C

I eo·

140·C

ZONE II

~ ~

.

.

-------

.

",,---

a

"",

.

..... ..

SCREW

40

SPEED

(Rpm)

60

80

100

,...

10

c:

>

o

J ..J

Z

&&I

I-

o

C

&I.

U1 W

~~

o .J

8:

4(

LaJ

-.....

...

IQ

X

,

a-

50

...A

Ext rus i on Rate versus Screw Speed at 0; fferent Process; ng Temperatures .

... .. ..... .

/

/

,/

/

o

/

/

~

A /' ~/

[]

./

/

.....----8I .,..",.""'-

J

-----0

54 Cn~!TOlJR

PLOT

OF

CO~JTOURS

SPE:::" (Pf"'~) "'.Jf' MrtsT'JPF: EX""RU5JQf'~ r-~T~C:: (rN/~rN)

SCREW

AF + -----+------+------+------+------+------.---------~J.0

~~.5

27.0

30.~

~4.J

3~.5

y

++++ ......

••••••

IO.72 L58 18.0 7492

32.775:59

XXXYXX

:12. 7755

~

47.47E?6

PFRQSe

47."-7626

....... Figure 16.

62. 1 7(,C)~

18.074t:;2 (,2.17()q~

(.,Q

.52 7

2 ()

The Effect of Screw Speed and Moisture on the Extrusion Rate at a Constant Processing Temperature of 150-125-100-50°C.

55

TEMPERATURE (e) AND MOISTtfAE. CONTeNT CONTOURS ARe E>eTRlIsrON RATES (IN/MIN)

CONl"OUR PLOT OF

CONTOUR PLOT 1 c.

.)• ;..;

I

+

,

_K.~ ~~_

~~

~

15 7 . 5 +

,. E 1'\

p

I,

lSS.J +

IS?

r;

-

;

R f.

lr;).J

I I

+

r

+

T

u R E

1

t.

7• 5

14r:;.J

14?C-:

1 /-J.

I•

+

FQFEE8E XXYXXXYXX . FP~~8~ XXXXXXXXy

,

•••••

++ .... ++ ...

••••••••

••••••••• P~A~PR XXXX~YXXX. +++++++~+++~+ •••••••••• 8~F:.::.P~· y.xxXy)(XX)(X ++++++ + ++ . AQ~~O .xxxxxxxx~ ~+.+++++++++ •••••••••••• +++++++++++-+ • • • • • • • • • • • • •

~~Q., )<

~-(

XXXXX

+++ .. + •••••• +++++ •••••• ++++ ++ + .... +++ ++++ +++

+ .... ++ +

I

++ + .. + ...

---+-------+------+------+----- .. -----+-------~o.o

?3.5

27.0

30.5

34.J

37.5

'.40ISiURE

...... t- ..

~

.. 4- to

'< X X '< ~~ ~~

.......

EFA":'A8

Fi gure 27 .

y

22.12736

32. 55552

32.55552 Al.41183 50.2fS\.l13 59.12.444

50.

.\. 1 • .:. t 1 83 1~6813

59.12444-

03.55260

The Effect of Temperature and Moisture on the Product Absorption at a Constant Screw Speed of 70 RP~1

67

CONTOUR PLOT OF

SCPE~

S~EED

CONTOUR S

ARE

(Pf"'~)

vs

r~()I$:tJRE

,

~BSOhPTrON~

CONTOlJP nLor

or

CO",JTE~~T

('11/0 )

(~/a)

X2*)(3

110 ....

1 :) ')

9J

I +

I

+

I

5 C R

E

80 ....

c.

I +

w

70

••• +++~++++ ••••••••••••••••••••••••••••• ++++++++ ••••••••••••••••••••••••••• ++ + f++ + •••••••••••••••••••••••••• + + + + + ... • •••••••••••••••••••••••• ..... +++++++++++ • ••••• ••••• •• • • •• • ••• ••• ...... ++++ .... +++ •••••••••••••••••••••••

.+

XX

+++++++++~+.++

~x~xx~

+~+~++++++.++++

+++++++++++++++ + + ....... +........ + .... +........

~x~xxx~ )C X)("( )C)()()(.

~XXX~)(XXX

•••••••••••••• ++++++ •• +++.+~.~ •••••••••••••

xx~~~x~~ xxxx~x~x

... +++++~++++++++~

I

+

I I

)( )( x)(

ReeS

••••••••••••

..........

+++~~~~++++++.+~

x~~~~x~xx

eF8

................ •• •• • • • • • • •• • ••

+.+++++++++++~+~

~~~~~xxxx Q,

60

••••••••••••••••••••••

++++++++++++++ ••••••••••••••••••••• ~xxx +.++++ ... +++++++~ •••• • •• • ••• •••• •• • ,XXXX ++++++++++ .. ++++ ••••••••••••••••••• ~xx

+++++++ ... ++++++++

x)( '(xx. ... + + ..... + + + ...... + .. fa ++ f-

x~~xxx~xx

.++~++++++++++....

+

•••••••••

FEPB X)(XXXXX: