an...
my
fit: 3
0., .(Q w
.5. .u :3.
RN” Irv Cr»
«3‘ 'M u‘ 5 ”as.“ e...
4'. ‘ «u.
a ‘ . S . l‘ a: I
“My” (‘0 0 fi.‘ fut-o‘-
R‘u- mu’. .Omfiv“ S a?!»
. or..w 1;: a DJ...“
.9th .r 4%. I. Ewu— ”v...
f'u ‘ .I- O "' -6
3 t t:
unw 3...... a n... V...
II. b \ ‘ .05-
. ..I.— D N. a

. . L
Ir: J .03:

'hu‘ a“... 1‘ —. nth. & .MKJ‘

nun . z & a. . EU a

Sufi. 7 .5 pop.» IV" .575 . l

a .3. a! 9I§J L‘

C Q a 2... ",6 4...

, . so ' ‘Lbl
a. . ha a: «J
3...; r3. . ...\ :.

3. v; .. .. .
c... . u. .u an ..
(«UV A»! ‘ Ow“ ”WI.
Am. 3\.

m3” ...

E “x

f. :1.

ers‘.

3.;

tax

a»...

 

.. 52:23.:31:52:;3:22;5:32;. mm

NIH W11 ml 1W1 WWW“WWWL

3 1293 00701 604045 ,

LIBRARY

Michigan State
University

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

1‘ DATE DUE DATE DUE DATE DUE

MAY 1 0 m3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSUIe AnAffirmativ eotnA Miqon/EualOpportunltylmit Ion
ammo-m

--.-—
- -

SOME FACTORS AFFECTING LUMP FORMATION IN FROZEN
STARCH-THICKENED SAUCES

by
Patricia Diane Cummisford

A THESIS

Submitted to the College of Hene‘Economics of Michigan
State University of Agriculture and Applied Science
in partial fulfillment of the requirements
for the degree of

 

MASTER OF SCIENCE

Department of Foods and nutrition
Year 1957

ABSTRACT PATRICIA DIANE CUMMISFORD

The frozen-storage stability of white sauce containing

a thickening agent, skimmed milk, sodium chloride, and a
hydrogenated vegetable oil was studied. Ten thickening
agents were included in this study: corn starch, sorghum
starch, waxy corn starch, waxy sorghum starch, waxy rice
starch, cross-linked waxy corn starch, phosphate crass-
linked waxy corn starch, phosphate cross-linked waxy
sorghum starch, wheat flour, and waxy rice flour. After
frozen storage for periods of one week to six months at
-19.5°c., samples were thawed at 35°C. and evaluated by

a taste panel and by several physical and chemical tests.

Samples prepared with different thickening agents and
stored for different times showed significant differences
for several of the factors evaluated by the panel. Most
of the sauces were lumpy and were considered unacceptaole
by the panel after a one-week storage period. Only sauces
with waxy rice flour appeared smooth even after a three-

month storage period and were acceptable to the panel.

The results of the physical and chemical tests were
in agreement with results obtained from the panel. Sauces
prepared from all thickening agents but waxy rice flour
gave a high degree of separation after only one week of

frozen storage. Waxy rice flour sauces separated upon

ABSTRACT PATRICIA DIANE CUMMISFORD

oentrifugation only after a three-month storage period.

The amount of supernatant from this sample was less than
that observed in any of the ether samples regardless of
storage time. The turbidity of the supernatant from waxy
rice flour sauces was greater than that obtained with any
of the other sauces stored three months, and was comparable
to values obtained with the supernatants from the three
cross-linked starches after a one-week storage period.

The amount of supernatant from the latter samples was much

larger and the sauces were badly lumped.

Samples prepared from glycogen and from nongranular
smylopectins from sago, tapioca, and waxy maize starches
showed marked variation in stability. Sauces and pastes
prepared from glycogen and from tapioca amylopectin were
stable to freezing. Average number of glucose residues
per end group or average molecular weight of the fractions

was not related to stability behavior.

From the study it was concluded that retrogradation
of starch was the primary source of instability, with
presence of milk proteins being a secondary factor. The
tendency to retrograde appeared to be related to structur-
al details of the starch molecules not entirely elucidated

by present analytical methods.

ACKNOWLEDGEMENTS

The author wishes to express her sincere appreciation
and gratitude to Dr. Elizabeth M. Osman for her help and
encouragement during the course of study for this degree

and the writing of this thesis.

She wishes to thank Dr. T. J. Schoch of the Corn
Products Refining Company for the contribution of non»
granular alylopectin samples used, Dr. S. A.‘Watson of
the same company for his work in.the isolation of the
'waxy rice starch, and Dr.‘w. Z. Hassid of the University
of California for his determination of the average branch
length of the waxy rice starch sample used in this
study.

11

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . . .
REVIEWOFLI‘JERATURE.........
Instability of Starch to Freezing
Starch Retrogradation . . . . . .

StarchGranules.........

StaTChF‘raCtionSeeeeooeeeeeee

Mlose O O O O O O O O O O O O O O O O O

Amylopectin . . .. . . . . . . . . . . . .
Milk and the Reaction of Milk to Freezing

PRWDUE O O O O O O O O O O O O O O O O O O 0

Preliminary Investigation . . . . . . . .

Preparation of Samples in the Viscometer .

Evaluation by the Panel . . . . . . . . .

Evaluation by Objective Tests . . . . . .

Determination of Nitrogen Content of the
Supernatant LiqU1dS e o e e e e e e e s e 0

Preparation and Evaluation of Additional
samplesOOOOOOOOOOOOOOOOO.

RESULTS 0 O O O O O O O O O O O O O O O O O O O O O 0
Observations of Sauces during Preparation . . .

Observations of Sauces after Freezing and
The-wingeeeeeeeeeeseeeeese

Results or the Panel Evaluation . . . . . . . .
Results of Objective Tests . . . . . . . . . . .

iii

Page

\OCDVUJUJH

11
13
19
20
2o
22
27
30

32

3h
38
38

39.

1+0
1+5

iv

Observations of Sauces and Pastes Prepared with
Glycogen and Nongranular Amylopectins
before Freezing o e e e e e o e e e e e e e

Results of Objective Tests on Glycogen and

Nongranular Amylopectin Samples after
Freezing and Subsequent Thawing . . . .-. .

DISCUSSION . . e . . . . . . . . . . . . . . . . . .
SUMMARY . . . . . . . . . . . . . . . . . . . . . . .
CONCLUSIONS . . . . . . . . . . . . . . . . . . . . .
APPENDIX . . . . . . . . e . . . . . . . . . . . . .
LITERATURE CITED . . . . . . . . . . . . . . . . . .

50

60
62
6h
72

Table
I.

II.

III.

V.
VI.

VII.

VIII.

XI.

XII.

XIII.

XIV.

LIST OF TABLES

Page

Amounts of Thickening Agents Used per
Batch of Sauce or Paste . . . . . . . . . 2%

Amounts of Thickening.Agents Used per
Batch of Sauce or Paste . . . . . . . . . 36

Panel Scores for Stored Sauces . . . . . . . . “h
Summary of Statistical Analyses . . . . . . . #6
Comparison of Various Thickening Agents . . . ”7
Macroscopic Appearance of Thawed Sauces

and Pastes Prepared with Several

Thickening Agents after One week

Storage at -1905 Cs e e e e e e e e e e e 51

Comparison of Amylopectins with Other
ThiCkening Agents e e e e e e e e e e e e 53

Analysis of Variance of Panel Scores
for smOOthneSS e e e o e e e e e e e e e 65

Analysis of Variance of Panel Scores
for Separation e e e e e e e e e e e o e 66

Analysis of Variance of Panel Scores
for Separated Liquid . . . . . . . . . . 67

Analysis of Variance of Panel Scores
for Mouth Fee]- 0 O O O O O O O O O O O 0 68

Analysis of Variance of Panel Scores
for Flavor 0 e e e e o e e o e e e e e e 69

Analysis of Variance of Panel Scores
for PESte CharaCter e e e e e e e e e e o 70

Analysis of Variance of Panel Scores
for General Acceptability . . . . . . . . 71

LIST OF FIGURES

Figure Page

1. Score Card Used by Panel in Evaluating
Samples . . . . . . . . . ..... . . . . . . 28

2. Appearance of Sauces Prepared from
Several Thickening Agents upon
Thawing at 35°C. after One Month
Storage at “19.500e o e e e e e e e o e e e 1"]-

vi

 

INTRODUCTION

A common binding agent for many food mixtures and
the sauces used with vegetables and meats frequently
employ a white sauce formula as a base. In recent years
there has been much interest in such combinations in the
form of frozen meat pies, frozen tray dinners, and frozen
specialties such as chop suey, macaroni and cheese, and
other gravy-containing mixtures, both for home freezing
and for commercial production. The stability of white
sauces which were held in frozen storage has been in»
vestigated by others who studied effects of viscosity,
homogenization, and various stabilizers and emulsifiers
on stability (10). Fluctuating temperatures during
storage were also studied to determine the effect on the
product (11).

me possibilities of future uses for a thickening
agent stable to freezing are numerous. As noted in the
literature (10), preparation of salad dressings and
desserts as well as sauces and gravies which would be
stable to freezing would be possible if the right
thickening agent or combination of thickening agents
could be found. The increased demand for precooked
frozen foods seemed to warrant further investigation as
to the factors involved in this type of instability.

1

’7

,4

 

This investigation was undertaken to study the
stability behavior of sauces prepared from ten starches
and flours under carefully controlled conditions of
mixing, heating, and stirring. Later the study was
extended to include the behavior of several pure branched

fractions from different starches.

It is hoped that the work described here will
provide some useful information on the factors important
in the instability of frozen starch—containing sauces i
and also indicate the need for further information on
the molecular structure of starch fractions before
complete explanations for the observed behavior can

be made. ‘i

REVIEW OF LITERATURE

A scientific explanation of the behavior of starch
when subjected to various treatments of cooking, aging,
and storing typically used with food products has been
greatly aided by basic studies of the chemical and
physical properties of starches and the constitution of
starch molecules and granules. This brief review
concerns the observed behavior of starches in foods with
respect to various aspects of freezing stability and
retrogradation and information on the structure of starch -
both granular and molecular - and also on factors which
may cause variable results. There is also included a
brief section on the constitution of milk and its
behavior after freezing.

Instability of Starch to Freezing

Observations on the retrogradation of starch by
freezing date back to IBMM, according to the review of
‘work in this area presented by weedruff and MacMasters
in 1938 (5“). They noted several other accounts of
observed retrogradation but pointed out that little
information is given in these reports as to the exact
temperature, time, or description of appearance of
resultant product. Neodruff and Hayden (53) presented

photomicrographs showing retrogradation in corn starch

3

I.

\<

h

and wheat starch gels. They found freezing at -2°C. gave
greater changes than freezing on solid carbon dioxide.

Under the conditions used (-200.) they found a frozen 5%
paste, which had been gelatinized between 75°C. and 95°C.,
gave a fibrous sponge fro-Awhich most of the water could be
pressed. veined areas also appeared in the photomicrographs.
They suggested that the changes might be related to associa-
tion of micelles and aggregates formed by dehydration of
swollen granules during ice crystal formation. This

'would allow molecules to draw together through secondary
valence forces. Ice crystals which formed in gels frozen
in liquid air or solid carbon dioxide would be smaller and

_ would produce less injury to granules. Woodruff and
MacMasters (5h) stated that gels frozen at -70°C. regained
most of the original consistency upon thawing, indicating
less chance for orientation at low temperatures. At low
temperatures, small ice crystals are formed, thus producing
relatively few areas where starch would be free of surround-
ing ice, (dehydrated areas) affording little opportunity
for molecules to become oriented. At higher temperatures

of freezing, large ice crystals are formed, causing many
areas where starch would be free of surrounding ice, thus
providing considerable opportunity for molecular orienta-
tion. weedruff also noted the resemblance of frozen starch
gels and cellulose when their fibrous strands were

observed with a Spierer lens (52).

\

r!

5

Schoch (#2) found that the speed of retrogradation
depended on the concentration of the paste, 1% pastes
retrograding more slowly than 5% pastes. He also found
that some modified starches (dextrinized, oxidized, or
ethylated) did not retrograde. These observations were

not made on frozen samples, however.

These retrogradation changes in starches, especially
the behavior when frozen, have been considered serious
problems by workers in both food research and food indus-
tries. Several suggestions have been made that homogenn
ization or the addition of stabilizers such as gelatin
might be used to increase the stability of starchnthickened
frozen foods. Hanson, Campbell, and Lineweaver (10) found
that the type of thickening agent used was a primary
factor in the stability of frozen sauces. They interpreted
their results as indicating that amylopectin starches and
flours minimized the separation and curdling, although
their data did not support this conclusion in all cases.
Beating the frozen sauces was found to improve the
appearance considerably. This group found that sauces
prepared with waxy rice flour were stable after eleven
months storage at ~18°C. In studies on frozen puddings
Hanson, Nishita, and Lineweaver (12) found that puddings
prepared with waxy rice flour were stable for six.to
nine months when stored at 0°F. In the same study, a ‘
soft custard-type pudding which contained waxy rice flour

was stable for two to four months, while a baked custard
preparation of a similar formula was unstable to freezing.
A recent paper by Hansen, Fletcher, and Campbell (11)
reported that stability of starchncontaining mixtures

was improved by the addition of pectin, gelatin, certain
vegetable gums, and Irish moss extractives. A decrease in
egg and in liquid in the formulas and an increase in waxy
rice flour and in sugar improved stability. They found
temperatures fluctuating between.-10°F. and 10°F. more
damaging than storage at a constant temperature of 00F.
In.white sauces prepared with waxy rice flour, 10°F.
storage produced instability after three weeksgtstorage at
fluctuating temperatures of ulOOF. to 10°F. produced in-
stability after two months; 0°F. storage produced inn
stability after twelve months, and in samples stored at
-lO°F., no instability was reported even after thirty-six
months frozen storage. They defined stability as less
than.3% separation. In samples stored at 10°F., no
separation was obtained after a one-month storage period
if 1% pectin, 0.h% to 1.5% gelatin, or 0.hfl gum tragacanth
was added to a waxy rice flour sauce prior to freezing.
They indicated that a 1.5% gelatin addition produced a
wheat flour sauce stable one month at the 10°F. storage

temperature.

7

Starch Retrogradation

Schoch (#2) found an insolubilization of starch when
dilute starch pastes were allowed to stand. Betrogradation
was suggested as designating “all tendencies of starch to
revert to less soluble forms." Schoch and French (#7)
credit Lindet (19) with the origin of the term retrograda-
tion in reference to the spontaneous formation of crystalline
aggregates present in stale bread. Maquenne and Roux (20)
later used the same term in referring to insolubilization
of starch. Noznick, Merritt and Geddes (36) attributed
the staling of bread to branched chains, a concept which
‘was not present in early literature, which attributed
staling to the straight chain components. Schoch and
0 French (#7) found that the soluble starch in fresh bread
was predominately amylopectin and suggested the amylose
fraction was insolubilized and retrograded during baking
and could not be involved in staling. 0n standing, the
solubles of bread decrease, probably by aggregation of
amylopectins into tight lattices involving secondary
bonding between branch ends of different molecules. This
condition is in contrast to the postulation by Heyer (21)
that in strongly swollen starch granules, gigantic
amylopectin molecules form a loose three dimensional
lattice of interlocked branches free of the secondary
bonds that form only slowly during retrogradation. Amylase
was not considered to be involved in this behavior. Schoch

and French (#7) found that stale bread, when.1eached with
water at 50°C., gave a.high amount of solubles. The
authors noted that this was in agreement with the apparent
freshening of stale bread by warming. The retrograded
amylose fraction is not solubilized at 50°C. but retro-
graded amylopectin is. Hhen.50% aqueous wheat starch
pastes were prepared and heated so that the temperature
rise of the paste over hot water resembled that observed
in a loaf of bread during baking, and then allowed to age,
similar behavior was noted under these conditions of
subsequent extraction. The authors suggested that other
bread components may also influence the staling reaction.
It was suggested that aggregation of the amylopectin
molecules involved an oriented association between terminal
branches of these molecules. No aggregation was noted in
the absence of free branches as in betanamylase limit

dextrin, starch ethers, or oxidized starches.
Starch Granules

K. H. Heyer and cosworkers reported, in a series of
articles, their interpretation of the structure and
organization of the starch granule. The granule was thought
of as a series of concentric layers about a central nucleus
(2%). The layers were composed ofmolecules of both
branched and linear fractions which were radially oriented.
The oriented layers were held together by associations of

straight chain molecules and/or the outer branches of the

’.

branched molecules such that microcrystalline areas were
formed. These areas were Spoken of as "micelles" by
Meyer, who used the name given.them by Nageli. within
the granule there were areas of crystalline nature inter-
spersed with areas of a lesser degree of orientation,

forming a loosely bound network.

In spite of a granule structure which is common for
starches, there is wide variation in the behavior of starch
isolated from different species, varieties, and types of
plants (k8). These properties are also influenced by
growing conditions and isolation operations (17). Thus,
in working with starches it is desirable to correlate
several properties determined with one specific sample
rather than to determine one property with one sample
and another property with a different sample.

Starch Fractions

For many years it was suspected that the granules
contained more than one main type of component. In a
recent review of the starch fractions (”6), it was
reported that van Leeuwenhoek (55) in 1716 found, by
microscopic studies, that a portion of the granule was
insoluble even'uhen.the granule was heated in water.

Other early investigators, including Guerinpvarry (7)
described three components, based on their solubilities in

hot and cold water; this observation was not substantiated

’0

 

 

10
by later findings (3h).

It was not until late in the nineteenth century that
an understanding of the nature of starch was furthered by
c. w. Nageli (35). Although his concept of the similarity
of the insoluble component to cellulose, with the soluble
fraction being merely a physical modification of the
insoluble fraction, has more recently been proved in»
correct, he did contribute to our present basic knowledge,
especially with respect to the micellar structure of

granules.

Maquenne and Roux (20) made the first serious attempts
to fractionate starch; a fraction which they called amylo-
cellulose, actually the one which ultimately became known
asamylose, was precipitated by allowing the centrifugate
from starch pastes or autoclaved sols to stand at room
temperature. Malt conversion of the precipitate gave
96-98% yield of maltose. They believed malt ineffective
with amylopectin, and concluded starch consisted of 18-20%
amylopectin since starch gave only 80-82% conversion to
maltose. Amylopectin was not isolated by these workers,
but they were the first to recognize the existence of two
chemical fractions of starch.

Indeed, much of the early work on fractionation ‘
resulted in fractions with varying degrees of purity (#8),
leading to confusion as to interpretation of results.

11

It was not until 19#l and 19#2 that Schoch (#2, #3)
reported the use of butyl or amyl alcohol to selectively
precipitate the component soluble only in.hot water, which
gave an irreversible (retrograded) gel on cooling. The
easily soluble fraction could then.be precipitated by

treatment of the solution with methanol.

The naming of starch fractions has varied considerably,
being especially confusing in the early literature. I; H.
Heyer (26) suggested amylose and amylopectin for the linear
and branched components respectively. Later the terms A-
and B- fraction were introduced by Schoch for these
respective components (##). He later suggested the
descriptive terms ”linear starch fraction" and "branched
starch fractionfl (#5).

Amylose

The structure of starch has been.investigated by
several methods. For amylase the bulk of evidence points
to a linear polymer of glucose units linked by 1,# -
glucosidic bonds of the alpha configuration. Haworth,
Hirst, and webb (l#) found that potato starch, when
completely methylated and then.hydrolysed, gave a large
amount of 2,3,6 - trimethylglucose and a small amount of
2,3,#,6 - tetramethylglucose. The 2,3,#,6 - tetramethyl-
glucose is derived from.the terminal nonreducing end of
the molecule, while all other glucose residues give the

12

2,3,6 7- trimethylglucose. This may be shown for amylase:

 

 

 

 

 

CHZOH HOCHZOH '7 CH20H
H O H H O \
CHOH
OH OH 04 OH H
H OH H OH
. -x
(CH3)2504
NaOH
CIHZOCH3 CHZOCH3 CHZOCH3
_ ' CHOCH3
CH3O H3 H o - OCH3 - .OCH3 H
H OCH3 OCH3 H OCH3
“X
, HQ 4- H20
1
CHZOCH3 CHZOCH3
H o
\
. HOH + (x+l) . OHCHOH + CH3OH
c5430 0CH3 H
H OCH3 3ecu-13

2,3,4,6-TETRAMETHYLCLUCOSE 2,3,6 TRIMETHYLGLUCOSE
Such methylation studies by Meyer ,_ and co—workers (32)

indicated ‘am'lose was a linear polymer of three hundred
glucose residues. The alpha configuration of linkages in
amylose was substantiated by enzyme studies with beta-
amlase. It was found (26) that mlose was completely
converted to maltose by this enzyme, indicating l,# -
glucosidic linkage of the alphaconfiguration. Meyer and
Bernfeld (25) found that alwlose gave a pure dark blue

color with iodine solutions. The intense blue color was

13

not found to be characteristic in general of the branched

fraction of starch and of glycogen.

Amylopectin

Methylation, periodate oxidation, and enzymatic
methods have been used in the characterization of the two
branched polysaccharides amylopectin and glycogen.
Freudenberg and Boppel (6) found 2,6 - and 2,3 - dimethyl-
glucose present after hydrolysis of completely methylated
potato starch, in addition to the tri- and tetramethyl-
glucose components observed with amylose. They thought the
2,6 - dimethylglucose arose from hydrolysis of the other
compounds but suggested that 2,3 - dimethylglucose resulted
from.methylation of residues linked at points of branching.
From.this it is concluded that branching occurs at position
6. Montgomery, weakley, and Hilbert (33) isolated isomaltose
from enzymic hydrolgeates of starch. The structure of the
disaccharide was proved to be 6-alpha- D-glucopyranosyl-
D-glucose from.a comparison of its octaacetyl derivative
with the same derivative of 6-alpha-D-glucopyranosyl—
betanD-glucose isolated from dextran hydrolysates. These
octaacetyl derivatives were identical. Further evidence
for the existence of 1,6 - glucosidic linkages in amylopectin,
and for the alpha configuration of these linkages, is
presented by Thompson and wolfrom.(#9). They isolated

panose, as a derivative, upon the acid hydrolysis of

’1

’l

 

1#

amylopectin. Panose,

CHZOH CHZOH
H O H v H O H
OH H O H O
H OH H OH .
CH2
H 0\
CHOH
OH OH H
H OH

4" {' ISOMALTOPYRANOSYL "' D' GLUCOSE
contains both l,# and 1,6 bonds of the alpha configuration.

That amylopectin differed from the straight chain
structure of amylose was also shown by enzymic studies.
Meyer, Bretano, and Bernfeld (26) found that anvlopectin
could not be completely saccharified by beta-amylase
treatment but rather formed dextrins which could be further
degraded by beta-amylase only after treatment by alpha-
glucosidase (23), specific for alpha configuration at
1,6 - glucosidic linkages. These results did not agree
‘with the schematic formula for amylopectin suggested by
Standinger. Later evidence also suggested that amylopectin
‘was a multiply branched molecule. Peat, Whelan, and
Thomas (38) and Lerner and cosworkers (18) concluded
that of three schematic representations presented, Meyer’s
representation was the only one which could give the ‘
yields of maltose and maltotriose obtained.

"

I.

 

15

The three proposed structures were:

EN __

W __
_' \

HAwoa'rH’s LAMINAR smuomoea’s MEYER’S ABORESCENT

 

 

 

 

 

STRUCTURE . COMB ‘ STRUCTURE
STRUCTURE

Lerner, Illingworth, Cori, and Cori (18) found that
amylopectin samples they analyzed contained at least five
tiers of branches, and that the number of branches per
tier decreased as the reducing and of the molecule was
approached. In earlier studies, Cori and Lerner (#) had
found that the ratio of free to phosphorylated glucose
was characteristic of the polysaccharide studied by
successive phosphorylase and amylo-1,6-g1ucosidase action.
Peat and coaworkers (39) concluded that Meyer's representa-
tion was correct since successive beta-amylase and de-
branching enzyme treatments were necessary to degrade

amylopectin completely.

The third method commonly employed in studying the
constitution of amylopectins was periodate oxidation.
Jackson and Hudson (15, 16) reported their basic work on
the topic in 1936 and 1937. In 19#5, Brown and couworkers

"

16

(2) reported the agreement of periodate data and methylation

values as to average chain length.

In 19#8, Potter and Hassid (#0) determined end-groups
in amylose and amylopectin samples which were carefully
purified and defined as to source (17). The reaction
involved a splitting occurring between adjacent carbon

atoms bemarégg free hydroxyl groups:

 

 

 

 

\
c)
(W1I10c3
H 1
Cltfni (x, 5*Effli
H. ‘0 t1 (3 H
(ma c»1 *' <)« t1 CH1
H <DH. 0&1 _J r1 CH1
x.
lHm
c
d” 9-H
0 '0
F—
? (xi CH2 'CHiCH '1 ‘CHéoH
,C OH H 0+1 H H 0\
c >L )— :fL -
H/ \\ O .1 O C H
o q C-H 0H CH -H o
01/ H '0' a 6 | O
+ [/0 . L .4 +lp
HCOH HCOH

Usually, free formic acid is determined by titration and
the number of nonreducing end groups is calculated from
this value. The method has been modified by several

people, and recently Hamilton and Smith (8) used a sodium

borohydride reduction to form polyalcohols from the

polyaldehydes present after treatment with periodate.

17

After hydrolysis, the mixture of glycerol (derived from
terminal residues) and erythritol (derived from nonterminal
residues) could be examined for relative quantities of
each. From the results they obtained, it was concluded

(9) that free glucose should be obtained after amylopectin
is subjected to the described treatment only if bondings
other than alpha l,M- and alpha 1,6- glucosidic linkage
were present. Occasional observations have suggested 1,5
(l) and 1,3 linkages (51), but there is no general agreement
on this matter, the bulk of evidence pointing to the other
types of linkage as the principal ones.

The structure of glycogens has been studied in many
cases concurrently with amylopectin (30, 18). The results
indicate that glycogen.resembles amylopectin generally, but
the former is more bushy and more highly branched with
shorter branches, both inner and outer. Cori (3) has

found glycogen to possess at least ten tiers of branches.

Meyer and Heinrich (28, 29) could find no amylase
present in certain starches, namely waxy rice and waxy
maize starches. The former was described as having
properties approaching those of glycogen. Later it was
noted (31) that much variation occurred in.fractions
obtained from.tapioca and from'wazy maize amylopectins with
respect to percent end groups and susceptibility to
enzyme degradation.

18

In enzymatic studies with amylopectin and glycogen,
Meyer and Fuld (27) and Meyer (22) reported that amylopectin
gave 15 to 18 glucose units per outer branch, 8 to 9 per
inner branch, and h% branching. Glycogen gave 6 t0 7
glucose units per outer branch, 3 units per inner branch,
and 9% branching. From.the results it was concluded that
glycogen was unable to form insoluble crystalline regions

because of shortness of outer branches.

Several Specific amylopectins have been studied in
detail. Potter and Hassid (#0) using purified amylopectins
which had been prepared and studied by Lansky, Schoch,
and K001 (17), found these fractions had between 22 and 27
residues per end group according to data obtained from
periodate oxidation. Hassid (13) later found waxy rice
starch gave a value of 20 residues when the same technique
0 was employed. Molecular weights of these fractions were
also determined by osmotic pressure measurements of
acetylated amylopectins (#1). They obtained the
following results:

Source Code Residues/end group Molecular Weight
Sago S-l-B 22 1,000,000
Corn C-109-B 25 5,000,000
Corn C-lhl-B 26 6,000,000
Wheat W-Z—B 23 l”000,000
Easter Lily 1.3-3 27 3 ,000,000
Tapioca T-3-B 23 3,000,000
Potato Fh3/M-B 27

 

Waxy Rice

 

 

19
Milk and the Reaction of Mfllk to Freezing

webb and Hall (50) noted that after freezing a
hydrophilic sol will repeptize upon thawing while a
hydrophobic s01 precipitates when thawed. Casein, a
weakly hydrated protein, may‘disperse. The destruction
of the caseinate colloid involves long storage times and
there is a progressive decrease in dispersion of casein
'with frozen storage. Denaturation of casein proceeds in
a limited but progressive manner. Skinned Iilk when
frozen at ~18°C. showed distinct casein.separation at
12 weeks and at 17 weeks a clear-serum separated by

thawing at r00m.temperature.

The protein content of skimmed milk is generally
given as between 33 and 3.8% and the protein content of
whey is about 0.85% (5).

PROCEDURE
Preliminary Investigations

A preliminary investigation of several factors was
made before an attempt was made to set up the experiment
in.fina1 form. Since carefully controlled cooking con_
ditions were desired, the effects of preparing the sauces
in the Corn Industries Viscometer were determined by
comparing standard white sauces prepared using this
instrument with standard white sauces prepared over a
boiling water bath with manual stirring. Not only were
the temperature increments similar for the sauces prepared
by the two methods, but also the viscosities of the sauces
prepared by the two methods were comparable. In determdnp
ing viscosity of the prepared sauces, a MacMichael Viscosi-
meter was used. Under conditions of the test (number 28
wire, sauce at 68°C., and water bath at 68°C.), the range
of readings for sauces prepared in the viscometer was 1% to
17 scale units and the range for sauces prepared over
boiling water was 13 to 19 units. The readings indicate
the ranges obtained on duplicate determinations with each
of two sauces for each method. The sauces stirred manually
showed readings of 13 to 16 on two determinations from one
batch of sauce and 15 to 19 on two determinations of the
other batch. Sauces prepared in the viscometer showed the
same reading (1% to 17) for the four determinations. Upon

20

21

thawing after freezing, sauces prepared by both methods
showed lumpiness. Microscopic examination of the sauces
also indicated similarity, both before and after freezing.
0n the basis of these results, it was decided to prepare

the experimental samples in the viscometer.

The ratio of ingredients used by Hanson (10) was
originally considered for use in this experiment. It was
found, however, that with the amounts of thickening agents
necessary to give final hot paste viscosities comparable to
that obtained with wheat flour the fat was not emulsified
in all instances. The sauces were cooked five minutes after
maximum viscosity was reached, or, with those thickening
agents showing no maximum, ten minutes after the start of
the increase in viscosity was observed. Because of the
difficulty in fat emulsification, the amount of fat was
decreased to the smallest proportion.which would still
allow easy blending of the large amount of flour which was
used. The sauces were prepared so that the amount of
thickening agent used would give a final viscosity of

about 30 gram-centimeters.

In order to measure the amount of liquid separated
from the thawed samples by an objective method, it was
necessary to find a method which would precipitate the
lumps present in the thawed samples but would not bring
about separation of freshly prepared sauces. 'In tests on

frozen samples prepared with various thickening agents, it

22

was found the flocculent material of the frozen samples was
precipitated by a Sorvall Superspeed Angle Centrifuge run
at 6100 r.p.m. (12,200 G. force) for 15 minutes. This same
treatment caused no separation in either the unfrozen sauces

or the skimmed milk used in preparing the sauces.

It was also noted that the supernatant obtained on
centrifugation of the thawed samples varied greatly in
turbidity. Investigation showed that differences between
the supernatants could be distinguished by reading percent
transmission of light using a distilled water blank as
100% transmission. The test samples were diluted with
distilled water to obtain samples in the readable range.

A Bausch and Lomb Spectronic 20 Colorimeter was used. The
wavelength of 625 millimicrons was chosen for the tests
since this wavelength most effectively eliminated the
absorption of light due to the presence of yellow-green

components of whey and flour pigments.
Preparation of Samples in the Viscometer

Sauces were prepared using ten thickening agents. The
preparation pattern consisted of seven randomized blocks.
The seven blocks were prepared over an eleven-day period,

with each block being completed within an eighteenphour time.

The ingredients used in the sauces were each obtained
from a single source or lot with the exception of the

skimmed milk, which was obtained from the Michigan State

23

Creamery in four lots. The processing of the milk was
controlled so that the four lots would be as nearly alike
as is possible in a natural product such as milk. The
first three blocks were prepared from the first two lots
of milk and of the remaining four blocks two were prepared
with each of the two remaining lots of milk.

The fat used was a hydrogenated vegetable oil
(Crisco, lot 076A6).

Chemically pure crystalline sodium chloride'
(Malinkrodt Analyzed Reagent) was used for the salt
ingredient.

The thickening agents used in the experiment were
each obtained from a single lot. The sources of the

various thickening agents are given in Table I.

The "dry" ingredients were weighed onto smooth,
lightly waxed paper. A stainless steel spatula was used
in transferring the materials to the paper from the cone
tainers in which the ingredients were stored. A.Harvard
trip balance was used, and the weighings were made to

one-tenth of a gram.

The milk was weighed in a flexible polyethylene
bowl. Due to the large amount used, a larger direct-

reading balance was used.

The amounts of thickening agents used per batch of

21+

TABLE I

.AMOUNTS OF THICKENING AGENTS USED FER BATCH OF SAUCEa 0R PLSIEb

W

 

Thickening Agent Grams

‘Waxy Corn Starch 21.1
American Maize-Products Company

Waxy Rice Starch 21.5

Isolated from waxy rice flour by
Dr. 80 ‘0 W‘t’on

'Uaxy Sorghum.Starch 22.0
Corn Products Refining Company
Phosphate Crossulinked‘waxy Corn Starch 36.1

American Maize-Products Company

Phosphate Cross-linked‘Waxy Sorghum Starch 37.0
Corn Products Refining Company

waxy Rice Flour 37.5
Rice Products Company

Sorghum Starch 38.0
Corn Products Refining Company

Corn.Starch No.0
Corn Products Refining Company

Cross-linked waxy Corn Starch kl.5
Rational Starch Company

Wheat Flour 58.0

Gold Medal Brand, purchased locally

3‘ 1082 grams ski-ed milk
7.2 grams sodium.chloride
50.0 grams hydrogenated vegetable 011

b 1082 grams distilled water

.—.-- 5*

—-.—e

<- v—una
.. —. —- .
.
—-—-—-*' e -. r- - .e—Q
. .
I‘ ,
_.
. , -
a I v
. / I (
.
I w
.
.n r ’
.
.
,.
f "
r ' ' ’
-
V .
4.
l
M.,,_..,.-.- -.-..-..._u

 

25

sauce are given in Table I. The amounts of fat, sodium
chloride, and skimmed milk remained constant per batch for
all thickening agents: fat, 50.0 grams; sodium chloride,
7.2 grams; and, milk, 1082 grams.

The fat was liquified by warming slightly in an
aluminum sauce pan and the thickening agent blended with
it to form a roux. Approximately three-fourths of the
skimmed milk, which had been heated to 60°C., was added.
The sodium chloride was then added and the mixture was
stirred to dissolve the crystals. The mixture was then
transferred to a one-liter Florence flask for pouring
into the cooking beaker of the viscometer, the stirrer
of which was started before adding the mixture. The
mixing utensils were washed with the remaining warm milk
which was then added to the mixture in the viscometer. The
viscosity recorder was started and the temperature of the

mixture was recorded each minute.

Throughout the cooking, the water-glycerol bath
surrounding the stainless steel cooking beaker was main-
tained at 100°C. plus or minus 0.5°C. It was occasionally
necessary to add additional water-glycerol mixture to
maintain the level of the bath such that the beaker would
be immersed t0 the pr0per level. The stirring speed was
maintained at level two, with the scraper turning 2M r.p.m.
clockwise and the agitator turning 60 r.p.m. counterclockwise.

26

The cooking was considered completed five minutes
after maximum viscosity, with those thickening agents
showing a maximum viscosity, or, in those showing no
maximum viscosity, at a point where the increase in
viscosity per time unit was small, about ten minutes

after the start of the rise in viscosity.

The recorder and the stirrer were then st0pped, the
stirrer was detached, and the beaker was removed from the
cooking bath. The contents of the beaker were transferred
to a one-quart Pyrex glass measuring cup to cool the sauce
and thus prevent further cooking, and to facilitate pour-
ing. Any fat which was not emulsified was skimmed off
and, in the case of slight lumping which occurred con-
sistently in.sauces prepared from one of the thickening

agents, the sauce was strained through a kitchen strainer.

Lacquered S-Z short tins were tared and filled with
120 grams of sauce. Immediately after filling, each can
was sealed using a hand operated can sealer. The cans
were initially cooled in an ice bath, being transferred to
a chest-type freezer after the eight cans from a batch were
sealed. The cans were identified with code numbers indicat-
ing the block and the thickening agent. The cans were
held in the quick-freeze portion of the freezer overnight
and then transferred to the storage portion of the freezer.

The freezer was maintained at -l9.5°C. plus or minus 0.5°C.

27

throughout the three-month storage period during which
subjective and objective evaluations of samples were made
periodically. Samples remaining at the conclusion of the
three-month storage period were packed in dryice after
four months and transported from East Lansing, Michigan, to
Urbana, Illinois, where they were held in frozen storage
for two months after which objective examinations were

carried out.

In order that the involvement of the thickening
agents in the instability problem could be ascertained,
pastes were prepared using 1082 grams distilled water with
the amounts of thickening agents used in the sauces. In
preparing these, the procedure resembled that used in
preparation of sauces. However, the thickening agent was
blended with the water and the mixture cooked to 95°C.

(#0 minutes) since the viscosity was not sufficient to
activate the recorder. Also, only four instead of eight
cans were frozen for each batch. After one week these

samples were thawed for use in objective tests.
Evaluation by the Panel

A five member panel composed of faculty and staff
members and graduate students from the Department of Foods
and Nutrition evaluated the sauces using the score card

illustrated in Figure 1.

The storage times included in the study were one week,

28

.eoanmdu unduesneve.nd Hanan he one: when enoom

.H shaman

 

 

 

 

 

 

 

 

1 - 343
I :1 H? t E
"madam” .c
mundane .o _
halo» .AHA mopeds-no
no a .dr epMMH
b h
hudma .o
nuance may .9 nopwdh
as .e
- d a
human .u
human .0
human .9 Hook name:
81 Bi): mm c, 1 heel“;
keeper .pi eouensnom
anon .e n kw, +F
«a a one
nofipsnsnoa each .n nowvensnom
as nomvenenoe on .e
w , .Hmmduldm,
enema emnsa .0 .
modem» Hanan .n unennuoomm
umoanemoao. .e
nape: son...“ - N N‘
I! Pug 0 0H 0 e. 1.. sew.
‘zDum FHrst I -

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

' a
‘I' l . I h‘ I -i-’lll‘ .1. ,i‘Ic l'. l III.».|‘!.|III.I V ’l' 4‘". Ic.‘ I- I..'.’ I." 3%] u. ‘I. It:
a . p . u .
. . ~ _
. . . e
. a - w . — i . a

. . .

w I. u ‘ fi . 0
‘-.ol1rlbv.li s .l : 6"”... I .7.‘ x . a . {'1 3|... as \ cl),l.c.l..(Q|.I'eI-.v‘l.’|1‘lvll4 '01.}!11. (I; 3'.) n'-, a! tilli IIKIII. a...“ bull

.

y 4 v a

O o c . .
. w ” . m m . a

. .. . s

- o-
‘ b-" um‘m -
..
n
a.

n
. l
.. .
m _
l l _
'lll'll'io" lite- tlell . . a IV'I, It‘s! in by} I u I u‘{.‘. 1." .t'rtll‘lcllli .b. O ..'{vln.!.‘u1.c lull
\ — . .o . .
I a . w
.3 p . N
e __ . .
q u

.
.
. D ,
‘7' I . I'.‘ .7 .I .eIO .. a}! l .I' i. all!" oil ..- CIT..- c'OII n|4l‘ 1" a .bi \l+‘t!$.l‘."* IIIII...» .Isl... ‘ t I." . c II.
— o a I

. . u
a . w w .
p . w
s
e a
e . .. .
u . . .
c
O . . .. .
I.‘ .I. I. -‘.I.a‘ "( l._e‘.c‘|{.-l .tslll' - ll... T.‘ 'Olo OII . - I- D ' s..‘1‘-‘ iii. tel
. .. .. u
C . . o .
m . .
. o ” p
. .
m . , .. . .
. n
q . _ _ .
O .. . 4 u _ . .
. . — u p
., a s . w
0 o ., * .
n . 1 .
c . .
o i _ . .
‘ I cl... t- 0.]! Cult . ‘31:; cl .. A’l ....I|I.l. . $3.15.- .. ‘9‘ .c cell'- . It Is I. {I}.
0 \ ' l . .. 4 ,I I. I . 11-1 ‘ l
0 ~ . e . .
u . N T
1 I
. . n c a
a o n
s m . .
O o n . s .
o D . I u
u . . .
a . C .
h .\ t . s . _ r
I .Il-Ilv. v I 49.. .’-l 'lil' .. ., DY . -b'l. on. c. :’ Otlll. 1 «III! '- ll .’.,4I‘t“s?.’u’l O ' 0 Us .‘+ ’4‘! a... is). O I It
. . . - .
. n
. . w _ _
. .
. o o . ~
. n v .
. . . .
— . .
0 s a . .
y , . a s
a
. .4 u . a
O _ is a . .
ell‘ ' Q . I . |I ,- o A i c.. I . 3.0-0.. 1 L c . . O u s C. I .. ‘.’ 1!- T. .‘thl‘I-il‘ .7. ‘* a l. .. as I'SI. lwa’, "t.*. .e: s, I vs .. c .71 I ..I . 7
. u . e
a . ,
l i . .. n
. u .
_ u . . ~ . i I
‘6‘l.’ll"l'-!h .cl- (0].. 0.5.0 . . Y‘ 0‘ ‘l..,.!|'. t I... .I’wl .. .3 Iv, vitallil‘fa. l.’ ‘0‘.’\'.s'h. .elsee I I‘v'll. I )1 .81.. n! (.l

29

one month, and three months. For each storage time, the
panel was scheduled on two consecutive days. Four
replications were used. Samples were presented in
duplicate from one block on the first day and on the
second day were presented in duplicate from another
block. Samples were presented in two groups for each
replication, five frozen samples being presented with a
wheat flour sauce freshly prepared to serve as a com-
parison. During the four replications, frozen sauces
were grouped so that each thickening agent was presented
with every one of the others at least once, so that all

possible comparisons were made.

In preparing the samples for panel evaluation,
duplicate samples were removed from the freezer and the
cans immersed in a 35°C. water bath for one hour. The
cans were then opened and the contents of one of the cans
was placed in an open Pyrex Petri plate (15 centimeters
diameter) while the contents of the duplicate can was
divided among five Pyrex watch glasses (10 centimeters
diameter). The control sauces were similarly prepared.
The samples in Petri plates were placed on a black back-
ground to provide contrast to aid in evaluation of smooth-
ness, amount of separated liquid,and characteristics of
the separated liquid. The samples on watch glasses were
arranged on white enameled trays. The panel members were

asked to evaluate the remaining factors from these

*1.

'0

 

30
individual samples.

All samples including the freshly prepared sauce
were identified by code numbers only. All scoring was

done in a room provided for that purpose.

The data obtained from the panel evaluation of the
frozen samples was analyzed using an analysis of variance.
In the method of analysis employed, the error terms used
in calculating F values consisted of pooled significant
interaction errors. Thus, in calculating the F value for
"Thickening Agents," all significant second-order inter-
action sums of squares involving thickening agents were

combined into a new error term.
Evaluation by Objective Tests

The samples on.which the physical tests were performed
were thawed in the same manner as those used in the sub-
jective evaluation. Two samples were centrifuged simul-
taneously using three stainless steel centrifuge tubes per
sample. The tubes were closed with stainless steel pressure
caps, placed in a Servall Superspeed Angle Centrifuge, and
the top of the centrifuge securely closed. The centrifuge
'was then gradually brought to a speed of approximately
6100 r.p.m. (a force equivalent to 12,200 6.). The speed
was controlled by means of a rheostat which had been

calibrated so that certain points on the rheostat control

’0

 

31

allowed a known maximum speed. The centrifuge was maintained
at the desired speed fifteen minutes, after which it was
gradually stopped. The caps were removed from the tubes

and a one-milliliter aliquot of the supernatant from one

tube of each sample was transferred to a nine-milliliter
distilled water blank in a stoppered test tube. The
remainder of the supernatant was decanted into a hundred-

milliliter graduated cylinder and the volume recorded.

The diluted aliquot of the supernatant was used in
the determination of turbidity. A Bausch and Lomb
Spectronic 20 Calorimeter set at a wavelength of 625
millimicrons was used in these determinations. The
instrument was adjusted so that distilled water gave 100%
transmission. Optically matched glass tubes were used for
both blank and samples. The instrument was frequently
checked using the distilled water blank. In certain cases
a further dilution was needed. One-milliliter aliquots of
the appropriate samples previously diluted 1 to 10 were
added to nine-milliliter distilled water blanks in.the
necessary instances. All aliquots and blanks were measured

using calibrated pipettes of appropriate sizes.

The objective tests were run on two consecutive days
and coincided with the panel evaluation. One sample of
each thickening agent was centrifuged each day and
turbidity of the supernatant was determined. Checks were

32

run if values obtained on duplicate samples varied more
than five milliliters in the amount of supernatant, and
whenever time permitted, the values for turbidity were
checked if they varied by 10%.

The objective tests were also run on samples stored
six months which were thawed in a water bath at 100°C. for
#5 minutes and then allowed to equilibrate at room
temperature for 15 minutes before centrifugation.

Determination of Nitrogen Content of the
Supernatant Liquids

In order to determine whether the differences in
turbidities were due to the presence of starch or of
protein, Kdeldahl nitrogen determinations were made on
aliquots of the supernatants from samples of one of the
blocks which had been stored one month. At the same time
the nitrogen in the supernatant from a sample prepared
fromnwheat flour and distilled water and held frozen.three
weeks was determined in order to correct the wheat flour
supernatant nitrogen.value of sauces for the soluble

nitrogenous material from the flour.

The nitrogen determinations were carried out by a
modified.Kjeldahl method currently used in the Chemistry
Department of Michigan State University. Duplicate
aliquots were used from supernatants of sauces prepared

from all thickening agents with the exception of waxy

33

rice flour which gave no separation on centrifugation

after a one-month period of frozen storage. Duplicate
aliquots of a solution of ammonium sulfate, which con-
tained a known quantity of the compound, were used as
standards. Five-milliliter aliquots of sample were
pipetted into dry five hundred-milliliter Kjeldahl

flasks. To each flask 0.3 gram cupric sulfate, 10 grams
potassium sulfate, and 25 milliliters concentrated

sulfuric acid were added. The flasks were placed in
digestion racks and heated for one-half hour after the
contents became a clear blue-green color. The flasks

were allowed to cool in the racks and then 175 milliliters
of cool tap water was added to each flask. Granulated

zinc was added to the cool flasks and 75 milliliters of
W sodium hydroxide was added by pouring it down the side
of the flask so that the sodium hydroxide would collect in
a layer at the bottom of the flask and would not mix with
the contents of the flask. The flasks were then connected
with the distillation columns and the contents of the flask
mixed by swirling. The flasks were heated and the dis-
tillate was collected in 50 milliliters of h% boric acid
solution contained in a 500-milliliter Erlenmeyer flask.
The boric acid solution contained ten drops methyl red
indicator and two drops 0.05% methylene blue solution. the
distillation was st0pped when the boric acid plus distillate
totalled approximately 225 milliliters.

3h

The contents of the receiving flasks were then
titrated with standard 0.1052-N. hydrochloric acid to the
end point as judged by comparison with a reference solution
prepared by placing 50 milliliters of h% boric acid solution,
175 milliliters distilled water, ten drops methyl red
indicator, and two drops 0.05% methylene blue solution into
a 500-milliliter Erlenmeyer flask. This solution is violet
colored and has a pH of approximately 5.0.

The amount of ammonium sulfate in the standards and
the protein equivalent of the samples were calculated. The
factor used in calculating percent protein content was that
given for milk proteins by the Association of Official
Agricultural Chemists (37): 6.38 times percent nitrogen
in the sample.

Preparation and Evaluation of Additional Samples

Since all of the thickening agents with the exception
of waxy rice flour produced sauces exhibiting marked
instability with as little as a oneaweek period of frozen
storage, a further investigation of possible contributing
factors was made. In order to determine the effect of a
highly ramified structure on stability in frozen storage,
glycogen was selected for investigation. Several none
granular samples of amylopectins were also included to
determine whether granular structure was a contributing
factor and whether all amylopectins gave similar results.

35

Several of the thickening agents used previously were also

included as checks.

Five percent of the amounts of ingredients used in the
viscometer preparations was used in preparing the correspond-
ing small samples. Two grams was the weight arbitrarily
chosen for the amylopectins and glycogen. Amounts of
thickening agents used are recorded in Table II. In the
preparation of these samples, 2.5 grams of fat was melted
in a large test tube in a boiling water bath. A roux was
made with the thickening agent, and 5h.1 grams skimmed milk
was added and mixed, using a thermometer to stir the
mixture. After 0.36 gram sodium chloride was added, the
mixture was cooked to 95°C. in a boiling water bath.

Twenty grams of the sauce was weighed into each of two

size one porcelain crucibles and the remaining portion

was poured into a third crucible. The crucibles were
covered with porcelain lids and wrapped in polyethylene
film before freezing to prevent excessive dehydration.

All three samples were frozen. The last poured, containing
less than twenty grams of sauce was thawed after only one

day of frozen storage and examined macroscopically.

Skimmed milk was also heated to 95°C. and samples
were frozen. Skimmed milk, fat, and sodium chloride
mixtures were similarly prepared. _The various thickening
agents were also prepared with distilled water by making a

”J

 

36

TABEE II

means or THICKENING sans uses an BATCH or sum“ or PASEb

 

 

Thickening Agent Grams
Waxy Corn Starch 1.06
Uhxy Rice Starch 1.08
Biosphate Cross-linked

waxy Corn Starch 1.80
Waxy Rice Flour 1.81
Corn Starch 2.00
Glycogen (Pfanstiehl) 2.00
Tapioca Starch Amylopectin c

Batch T-8-B (T. J. Schoch) 2.00
Sago Starch Amylopectin c

Batch 8-2-3 (1'. J. Schoch) 2.00

Waxy Haize Branched Material

 

I 9+.1 grams skimed milk
0.36 gram sodium chloride
2.5 grams hydrogenated vegetable oil

b 9+.1 grams distilled water

c For fractionation procedure see S. Lansky M. K001,
and T. J. Schoch, Properties of the Fractions and
Linear Subfractions from Various Starches. J. Am.
Chem. Soc. 71: I+066-‘+075 (l9‘+9).

' _ A ..l - ‘
-0. ~- .— fl . ' . ~-'- 7 . .
-- w- ‘H’ — ’ ‘ - . I I -"
I
.
I
I
e a ~~ee\- -I ~~' ' ~ . ‘ ~

‘—.7.

Ad

 

37

slurry and heating it to 95°C. In all instances duplicate
preparations were made with two twenty-gram samples being

frozen from each. The remaining paste from each batch was
poured into a third crucible. This sample was then stored
frozen for two days, at which time macroscopic evaluations

were made.

Half of the twenty-gram samples were stored frozen
for one week and then.examined and evaluated by the same
objective tests that had been used on the former samples.
The crucibles were removed from the freezer and allowed
to thaw at room temperature (27°C.) for two hours. The
contents of each crucible was transferred to a stainless
steel centrifuge tube. Subsequent centrifugation and
turbidity measurements were the same as previously de-
scribed. The remaining samples were stored at -l9.5 plus
or minus 0.5OC. for four weeks after which they were
packed in dry ice for transporting from East Lansing,
hflchigan, to Urbana, Illinois, for two months additional

storage.

RESULTS
Observations of Sauces during Preparation

Although precautions were taken to make the preparation
of all the sauces as nearly the same as possible, differences
in the ingredients themselves introduced slight variations
in procedure. There was some variation in the time required
to combine the ingredients, since certain thickening agents
‘were more readily blended with the fat and milk than others.
However, the total time was consistent for all batches
made with one thickening agent.

In general, each thickening agent followed a consistent
pattern during cooking with respect to temperature increments
with time, total time required to reach the end point of
cooking, final temperature of the mixture, and viscosity
pattern. There was some variation in the last mentioned
property occurring primarily with one batch of milk. The
temperature of the mixture in the Cooking beaker rose
steadily to a maximum value with all thickening agents with
the exception of wheat flour white sauces. These showed a
rise in temperature initially, followed by a fall and
subsequent rapid rise to the final value. The fall in
temperature occurred simultaneously with the start of the
rise in viscosity. The drop and subsequent rise to the
former high value occurred within a two-minute interval.

38

4'

 

RESULTS
Observations of Sauces during Preparation

Although precautions were taken to make the preparation
of all the sauces as nearly the same as possible, differences
in the ingredients themselves introduced slight variations
in procedure. There was some variation in the time required
to combine the ingredients, since certain.thickening agents
were more readily blended with the fat and milk than others.
However, the total time was consistent for all batches
made with one thickening agent.

In general, each thickening agent followed a consistent
pattern during cooking with respect to temperature increments
‘with time, total time required to reach the end point of
cooking, final temperature of the mixture, and viscosity
pattern. There was some variation in the last mentioned
property occurring primarily with one batch of milk. The
temperature of the mdxture in the cooking beaker rose
steadily to a maximum.value with all thickening agents with
the exception of wheat flour white sauces. These showed a
rise in temperature initially, followed by a fall and
subsequent rapid rise to the final value. The fall in
temperature occurred simultaneously with the start of the
rise in viscosity. The drop and subsequent rise to the

former high value occurred within a two-minute interval.

38

39

There was no comparable occurrence in the waxy rice flour
samples or in.the wheat flour paste prepared with distilled

water 0

Samples prepared from phosphate cross-linked waxy corn
starch consistently showed a slight lumpiness which

necessitated straining.

In Spite of the relatively lean formula used, the
various sorghum starChes formed sauces which showed conp
siderable unemulsified fat. The unemulsified fat was
greater in amount with sorghum starches than with the
corresponding corn starches. Sorghum starch sauce had a
layer of fat on top; corn starch had a lesser amount. Waxy
sorghum starch sauces had considerable unemulsified fat;
waxy corn starch sauces had a slight amount. Phosphate
cross-linked waxy sorghum starch sauces had a slight
amount of unemulsified fat; modified waxy corn starch

sauces in general showed no unemulsified fat.
Observations of Sauces after Freezing and Thawing

At the onedweek storage time the samples prepared from
wheat flour showed considerable lumping, as did waxy corn
starch and waxy sorghum starch sauces. The corn starch and
sorghum starch sauces became rigid and had a spongy texture
which exuded a watery liquid when cut. Increased storage

times merely intensified the changes in these samples.

1+0

After a one-week period of frozen storage, sauces
prepared from certain of the thickening agents were relative-
ly free from lumping and had very little if any watery
liquid separated. The samples which behaved in this.
manner included sauces prepared with waxy rice flour, waxy
rice starch, and the three derivatized starches. Longer
periods of frozen storage caused marked deterioration in
all of these samples. At the end of the one-month storage
period, waxy rice flour and waxy rice starch sauces appeared
superior to the others, which were lumpy and sandy but
showed no considerablechange in the apparent liquid
separation. After a storage period of three months, all
samples were definitely changed. Waxy rice flour sauces
showed a slight sandiness but no visible separation, while
the others were badly lumped and separated. Figure 2
illustrates differences apparent after the one-month storage
period.

Results of the Panel Evaluation

Panel scores are summarized in Table III. The
descriptive terms most frequently used in describing the
various sauces may be summarized as follows:

Control - homogeneous no separation, velvety,
bland, short, acceptable

Wheat flour sauces - large lumps much separation,
watery liquid, sandy, bland, friable,
not acceptable

Cross-linked Waxy
Corn Starch

 

Unfrozen Wheat Flour

 

Waxy Rice Flour

Figure 2. Appearance of
sauces prepared from several
thickening agents upon
thawing at 35°C. after one
month storage at 919.5°C.

 

“— “-—.

 

 

Corn Starch

 

Wax Sorzhum Starch

 

Phosphate Cross-linked Phosphate Cross-linked
Waxy Sorghum Starch Waxy Corn Starch

‘+3

Waxy rice flour sauces - homogeneous, no separation,
velvety, short, bland, acceptable

Waxy rice starch sauces:
1 week: homogeneous, no separation, milky
liquid slippery, bland, not acceptable
later: large umps, much separation, watery
liquid, slippery, off-flavor, not
acceptable

Waxy corn starch sauces - large lumps, much separation,
watery liquid, slippery, bland, stringy or
gummy, not acceptable

Waxy sorghum starch sauces - large lumps, much
separation, watery liquid, slippery, bland,
stringy or gummy, not acceptable

Sorghum starch sauces - large lumps much separation,
watery liquid, sandy, blan , friable, not
acceptable

Corn starch sauces - large lumps much separation,
watery liquid, sandy, bland, friable,
not acceptable

Cross-linked waxy corn starch sauces - large lumps,
some to much separation 'watery liquid
sandy, raw cereal and off-flavor, short,
not acceptable

Phosphate cross-linked waxy corn starch sauces:

1 week: large lumps, some separation milky
liquid, raw cereal and off-flavor,
short, not acceptable

later: large lumps, much separation. watery
liquid, raw cereal and off-flavor,
short, not acceptable

Ph03phate cross-linked waxy sorghum starch sauces:
1 week: homogeneous, no separation, sandy,
raw cereal and off-flavor, friable,
gummy, acceptable
later: large lumps much separation watery
liquid, sandy raw cereal and off-
flavor, friable, gummy, not acceptable

Analysis of variance of the panel scores showed no
significant differences in replications for any of the seven

factors evaluated. Thickening agents showed very significant

’1

’.

/3

./'

.o-‘m . .

.-

. I. v. 0-8.11.1.Iu‘ilt.‘ :11. .0 .)\\u s! .‘asl... stir I: l .10. c‘ ‘ .l..a1.(l.|!lts1 a.- .3-
.o
.
.
. o f s
.\ w n
I Q C C O O C O C c
f I IO. I ’1 ”
§
5

.
0
'
t
I
w o

I a o 9 I a C e e

I
u
1 0
I
O O O 9 I O I O O
c
I x.
a
Q
.
.
l I, i. .I vs! sol-is. . I’l’ '- e .|.. 1“)! I I. If I II. ' its...) "¢III.III i.‘
u
v
I. v . s v
- "

 

 

.~-r-

I'IL-

I.

‘l

I"):

r C
a p
O O O
-
I a e
u _
. r
I Q 0

I It; .‘Dil. '1‘ (9 I.,

.....— _ a.

I 0 O .-
I
.
a
. , .
O O C I n
I
u
I I O O .
.
.

l+5

differences in all factors except flavor, in which
differences were significant. Storage times showed very
significant differences in smoothness, separation, separated
liquid, and acceptance, significant differences in mouth
feel and paste character, and no differences in flavor.
There was a very significant difference between scores of
different judges on all factors except flavor in which there
were no differences. It should be noted that each judge

was relatively consistent in scoring and probably statisti-
cal differences were due to the range in scores of the panel
members, for certain judges tended to give scores con»
sistently higher than those of other judges. The
statistical significance of various factors is summarized
in Table IV. The details of the analysis may be found in
the Appendix, Tables VIII through XIV.

Results of Objective Tests

The results summarized in Table V show the behavior
of sauces and pastes prepared with the various thickening
agents when subjected to centrifugation and other physical
and chemical tests after frozen storage. In most cases, the
initial change (after a oneaweek storage period) was great,
and further storage caused only a slow continuation of the
deterioration. However, with several thickening agents,
notably waxy rice starch, waxy rice flour, phosphate
cross-linked waxy corn starch, and phosphate cross-linked

1+6

TABLE 17

SUMMARY OF STATISTICAL ANALYSES

_ A l—A‘tl

Factor Thickening Storage Judges Replications
Agents Times

 

 

 

Smoothness In: as an: /
Separation In an: an: /
Separated Liquid u an an: /
Mouth Feel at s an: /
Flavor a / / /
Paste Character an s an /
Acceptability an ass as /
*Tignificant at 51 level 7 Not significant _—

" significant at 1% level

See the Appendix, Tables VIII through XIV for details
of the analysis.

».—-

s

.“—-

OW—~ —-

 

L+7

"
9 a n O a Q a e e a e
A I
e u
.
. p
.
a
.
. ., .
O O O Q o O O 9 0 I
_ l. .
A .
l .
.
.. .
. I .
1
0 o o e a O o 0 O Q .
I
n
.
a n o m e e o e a e
.
.
.A a! Y I s» n .1 'la ‘- O I bu" ‘1'- Alt-. ’0) I 10.. win?! 0 3|
.
.
.
. e e a o o e a e o e e
m C. .,
. ,
I ‘ I a I
a .
.. .
..
.

 

-
_- D
c...-

 

 

4 s ‘-
. m—.,. - -
u

'.
»_ a ~‘- -' «s-‘u. - .

 

----

-- ... ,,__

 

.. _..-.. o—A-o ._ . .
o w “— -u ~ .
o
o -— .4'< Q. -~.-- - '—
I
e
‘n

 

‘+7

.
Q I O
_
O O
O O

'—.-o-

1...,

—~
c '—
adv-‘0,

u on

-...,. ,-.,. ‘ 4-“...
Q

- .A’. .
n
I

0'--

“ p.4- .4 ..

CL
‘

c - .
o . g .
.
n - . .v" w—-.
u
... -.d. . . _
‘ 'v . ,._,- a“- n.
u u

“+—

 

1+8

waxy sorghum starch, the change during the first week of
storage was small but an increase in storage time caused

relatively large changes in the properties of the sauces.

Separated liquid in waxy rice flour sauces and pastes
was not watery but was viscous and was difficultly separable
from the sediment. The solid phase from sorghum starch and
corn starch sauces and pastes tended to entrain liquid,
making accurate, reproducible measurements of the volume
of the liquid difficult. There was wide variation in
turbidities of supernatant liquids with certain thickens

ing agents, primarily in the group of derivatized starches.

A comparison of results obtained for samples stored
six months and thawed at 35°C. and those thawed at 100°C.
is possible from Table V. The contents of cans subjected
to thawing at 100°C. reached temperatures of approximately
90°C. With the elevated thawing temperature there were
visible differences as compared to samples thawed at
35°C. The waxy starch sauces - corn, sorghum, and rice -
showed some visible watery liquid separation and had
large, soft, slippery lumps in samples thawed at 100° c.
These same starches produced sauces with considerably more
‘watery liquid separation.when thawed at 35°C. Corn starch
and sorghum starch sauces showed considerable visible
separation of a watery fluid and had a grainy appearance
with many small lumps. This latter appearance is in

'15

#9

contrast to that of similar sauces thawed at 35°C., which
consisted of a single Spongy mass with exuded water surround-
ing it. Wheat flour sauces appeared to be very grainy and
had visible separated liquid if thawed at 100°C., while
samples thawed at 35° C. were very lumpy with more liquid
visible. ‘waxy rice flour sauces thawed at 100°C. appeared
to be smooth and showed no visible separation. Those
thawed at 35°C. showed a slight amount of separation but
were smooth. The three derivatized starches gave sauces
with soft lumps but no evidence of watery liquid separation
was visible in samples thawed at 100°C. This is in sharp
contrast to samples thawed at 35°C., which showed graini-
ness and visible watery liquid separation.

The nitrogen content of supernatant liquids corresponded
to turbidity readings of the corresponding samples which
were stored one monflh.

Observations of Sauces and Pastes Prepared with

Glycogen and Noanranular Amylopectins
before Freezing '

The tapioca branched fraction gave a stringy paste
when mixed with water. However, on heating, the stringiness
disappeared below 80°C. The sago fraction was more diffi-
cult to disperse in water, complete dispersion being
effected only by heating to 92°C. The waxy maize fraction
formed a tacky mass with water, which was somewhat dis-

persed on.heating to 95°C., but at this temperature some

r‘

50

undispersed starch remained and dispersion occurred on
heating an additional 15 minutes. Glycogen dispersed
readily and the dispersion cleared somewhat at 85° C.

The sauces prepared with glycogen and the sago starch
fraction did not emulsify fat, as well as those prepared
with tapioca and waxy maize fractions. Glycogen and the
sago fraction showed less apparent thickening than did
tapioca and waxy maize fractions. All produced much less
thickening than did the granular starches. The thickening

agents used as checks behaved as noted previously.

Unfrozen samples standing at room temperature for two
days showed no-precipitate. The pastes from the saga and
maize amylopectins and water, after storage for two days
in the freezer, were thawed, giving lumpy products. The
thawed paste from the sago fraction was heated to 60°C.,
whereupon the lumps disappeared and did not reform after
standing at room temperature for two days. The thawed
paste from.the waxy maize fraction was allowed to stand at
room temperature for two days without further heating; no
apparent change in clarity occurred.

Results of Objective Tests on Glycogen and Nongranular
Amylopectin Samples after Freezing
and Subsequent Thawing
The appearances of sauces and pastes prepared with

nongranular amylopectins and glycogen, as well as certain

51

TABLE VI

MCBOSCOPIC APPEARANCE 01" mm SAUGES AND PASES
mums WITH SEVERAL THICKENING A HTS
AFTER ODE WEEK STWGE AT -19. C.

 

_—

 

Thichening Agent Water Pastes Sauces
Corn Starch Much lum ing a Very lumpy a with
watery l qu (1 much watery liquid
separated separated
Phosphate Cross- Some lumping Very flocculent
linked Waxy Slight separation
Corn Starch *
Waxy Corn Starch Marked lumping Several large lumps
. ginsiderable separa-
on

Waxy Rice Starch Some separation Some lumping
Waxy Rice Flour Slight separation lo separation
Tapioca Starch

mlopectin No lumping Ho lumping
Waxy liaise

Branched

Material Slight lumping Slight lumping
Sago Starch Very noticeable Very noticeable

mlopectin lumping lumping

Glycogen No lumping lo lumping

 

 

. .ea-e -. o a -- .._ ‘ .5. ~15..- ~-~.-. .-_..‘. nu-U‘Qe..- -e..-.-v-. O- u o 0.0--- .- -- -. -
-0. o . o‘— - . m e. ----'- sea—av” ...-~a ..” _.--‘~..—D '- . - n-”'--°---“‘¢|o.-'.O

.

u__ ‘pafi—s o—d‘m . -a— o v .c-Io‘-e. . A-a.’ 0" e ‘um --._fi..~~O--
.
I u
.
A
O
-
e
.
-
I
O
,
m
. e
,— , . .
. r
u . . e
' .
v r
O
' I
r’ ' ’
I
U D I I
‘ e
. . — . _
v - w .
D

f ' r

a .. . .— —. --—- -——n e wag-u “cv- .s t, ._v"r-v~ . s - -— «a CI...)- §'~~-.~ e a -C- -—

_~-—

52

thickening agents used as controls, thawed at 27°C. after
a onedweek storage period are given in Table VI. The
control samples appeared as they did in previous
observations. There were, however, marked differences

in appearances of the other samples.

The results of objective evaluation of these samples
are summarized in Table VII. The samples which were
included as checks behaved in the same manner as in
previous tests. In samples stored three months, the
results were more difficult to explain than in samples
stored one week. Skimmed milk which was subjected to the
experimental conditions showed some instability after the
dry-ice packing in transfer and the glycogen and tapioca
fraction sauces gave results comparable to the values
obtained with milk. However, glycogen and tapioca
amylopectin pastes showed no separated material even

after the drybice shipping.

 

. - o.- .- o y - s
. e_ > . n v

-- - u. .— . . s

a can... h“ C..- - ..

— — - _ -

m -—-- —e - a...” .- e

- -_ - i——_ - . .,

- -.
w
. ,
»'-._.
c ‘.

53

- 9:“ a , -
. v , - .. —v
I
. ,_ e _ . A
u
- . ~ 0 « a.
_ .‘ . -

— .. .
o— .
.
m
es - ....—I~ w.
v
.
O
o
. - a. . ,.
n
O C

I.
we
-‘.-—
o - I...
-m-ce‘q-
‘c-
9
9 ».
‘(
I
I
Q
A C
O
U
U
._~v“
-e —---”§

.-‘-~-

--— m
~-‘—~

g... ~~.

bow" D ‘-

0'-

- _-
- c

.. _
.,

I- -

._.-....

—~—. -‘_A

.~'-

. o
'-
—
“an,

a -
O
y...

“a---

 

-— .. _ - —
- -e- .._-
0
fl.._.-.._.—,- -
c . .u - - -
a . . . u-Q-
.
.. —ne~. . -—
. .
e
\
' .
' v I e
e
,-
..
4
e
. -. III--- —
~

' -—
p - n
p...

DISCUSSION

0n the basis of the results presented, it is evident
that the rate and extent of lumping induced by frozen
storage of cooked starch-containing mixtures vary greatly,
according to the type of starch used. Lumping was
apparent after freezing systems containing only starch
or flour and water and was not necessarily dependent on
the presence of milk. Correlation between lumping of
starchdwater pastes and white sauces prepared from the
corresponding starches indicated that starch was the
primary factor in causing the behavior. Marked differences
in the turbidity of the separated liquids were observed.
Kjeldahl analysis of the liquids, separated by centri-
fugation, showed a correlation between nitrogen content
and turbidity. Use of starches such as corn starch and
sorghum starch and wheat flour which contain relatively
high amounts of straight chain fraction (amylose) resulted
in the separation of liquids which had a low protein
content. The waxy starches, which contain primarily
amylopectin, were in an intermediate group while the
derivatized waxy starches showed relatively large amounts
of nitrogenous material in the supernatants following a

one-month period of frozen storage.

When this information is considered in conjunction

5n

55

with the observations of the lumps formed, it appears that
there may be formed a network of starch molecules (either
those released from.broken granules or those at the granule
surface) which entrains and/or binds the protein molecules
into the interior of the aggregate, thus effectively
removing nitrogenous material from the liquid. Since the
size of such aggregates would be quite large, the aggregate
is no longer dispersable and is precipitated. The fact that
the cross-linked starches produce fewer chains free for the
formation of such a network may explain their smaller
tendency to form such lumps, especially in the early storage
period. In such instances, combination between free side
chains is greatly lessened since the surface chains are
partially linked together by a phosphate (or other) bridge
and are not available for starch to starch interaction.

The cross-linking is not entirely satisfactory in.prevent-
ing lumping since such sauces are not free from obvious
lumping, especially with longer storage periods. Possibly
this continued action during storage is due to an opportunn
ity for further orientation of the ”sandy" particles

formed early in storage such that opportunity for inter-
action of the starch is possible. Cross-linking does not
prevent aggregation and lumping, but merely decreases the

rate of aggregation and size of the resulting lumps.

The behavior of a highly ramified structure as found
in glycogen is interesting in this respect. The outer

56

branches of glycogen are relatively short and offer less
chance of intermolecular attraction than do the outer
branches of the waxy starches which were studied. The
outer branches of amylopectins are approximately 15 to 18
glucose units in length while the outer branches of
glycogen are 6 to 7 glucose units in length, according to
Meyer (22). Even though glycogen, being nongranular, is
exposed to the system on all sides, aggregation does not
occur. From the behavior of the nongranular mlopectins
it is evident that possession of granular structure is not
requisite to aggregation. The lumps formed in nongranular
amylopectin sauces tended to be smaller than those in
granular waxy starch sauces. Possibly this difference is
due to the absence of the granules. Thus whole granules
are not involved when intermolecular binding occurs, the

result being a smaller aggregate with the nongranular starches.

The widely varied behavior of nongranular amylopectins
obtained from several starches cannot be related to either
average branch length or average molecular weight, since in
two of the fractions studied (tapioca and sago amylopectins)
which gave sauces with a great difference in stability,
Potter and Hassid (#0, #1) found these values to be nearly
the same. The waxy rice starch gave nearly the same value
for average branch length when the same method of analysis

was “59d (13) e

57

It is apparent from the observations and data
presented that certain fine details of structure of the
amylopectins must vary considerably. Although average.
branch length of the several amylopectins was the same,
the behavior of these amylopectins varied greatly with
respect to stability when pastes or sauces were frozen.

It is possible that two amylopectins with approximately
equal average branch lengths would vary greatly with respect
to distribution of branch length. An amylopectin with
branches of relatively uniform branch length could have the
same average branch length as an amylopectin in.which there
were very long and very short branches as well as some in
the intermediate range. The same situations can be

postulated for the molecular weight distribution.

That retrogradation of starch is a primary factor in
the instability of frozen sauces was further supported by
the results obtained when the sauces were thawed in boiling
water. Since this treatment raised the temperature of the
sauces to approximately 90°C., the retrograded amylopectin
portion would be expected to redisperse (M7). The appearance
of sauCes treated in this manner was considerably more
desirable (smoother with less visible liquid separated) than
correSponding sauces stored the same period of time but
heated to 35°C. Also the sauces thawed in boiling water
gave less separated liquid by centrifugation. Turbidities
of these liquids were also greater. The behavior of the

.llki ‘ ills!

 

waxy starches is especially striking in these respects:
separated liquid was reduced to about 20% to 50% of that
obtained with sauces thawed at 35°C. The cross-linked
starches also showed a reduction in liquid separated by
centrifugation. The ordinary starches were improved less
than the others by the boiling water thaw. The relatively
high amylose content of these starches is undoubtedly
responsible for this behavior since retrograded amylose

is highly insoluble and heating to 90°C. under the
prevailing conditions would not effect dispersion (#7).

0f the samples heated to 90°C. after a sithonth
frozen storage period, only waxy rice flour sauces were
free from visible liquid separation and lumps. A.frozen
food product consisting of a white sauce base would be
expected to behave much like the sauces alone do, especially
if thawed without agitation (as in meat pies, frozen oven
dinners, or items thawed over boiling water). The informa-
tion obtained in this study suggests that, of the thickenp
ing agents used which are available in commercial quantities,
waxy rice flour is the only one showing sufficient stability
to freezing to give a product which is acceptable if heated
to 90°C. in thawing. Stability even-with this thickening
agent was not sufficient if the product was not subjected to
temperatures above 35°C. in thawing. 'Waxy rice starch does
not show this unusual stability, even when thawed at 100°C.

Since the starch was isolated from a portion of waxy rice

59

flour from the same lot as the flour used in the study,
there seems to be a stabilizing factor in the flour. This
stabilizing factor has not been defined or investigated in
this study.

Sauces prepared with the amylopectin from tapioca
starch showed unusual stability, even when thawed at 27°C.
These samples appeared to be stable even under conditions
which caused destabilization of the milk. This behavior
seems explainable only on the basis of the molecular

structure of the tapioca amylopectin.

SUMMARY

The behavior of ten different thickening agents was
studied with reSpect to stability of frozen white sauces.
Wheat flour, corn starch, sorghum starch, waxy corn starch,
waxy sorghum starch, waxy rice starch, waxy rice flour, a
cross-linked waxy corn starch, a phosphate cross-linked
waxy corn starch, and a phosphate cross-linked waxy sorghum
starch were used in amounts which, together with a given
amount of skimmed milk, of sodium chloride, and of
hydrogenated vegetable oil, gave a final viscosity of
about 30 gram-centimeters as measured five minutes after
maximum viscosity or ten minutes after the start in the
increase in viscosity (in samples giving no maximum)
when cooked in the Corn Industries Viscometer. Samples
were frozen in tin cans and held at ~19.5°C. for varying
periods of time .

A taste panel found all sauces except that prepared
with waxy rice flour unacceptable after a one-month storage
period. The panel scores indicated that the changes were
great with a oneaweek storage period and longer storage only

accentuated differences.

The panel scores coincided with centrifugation and
turbidity tests on liquid separated from the frozen sauces
upon thawing. Differences in turbidity paralleled protein

60

61
content of the supernatant.

From results obtained on sauces thawed at 100°C., it
appeared that retrogradation of starch was the primary

factor in instability.

Sauces prepared from nongranular amylopectins showed
widely varying behavior and stability which was not related
to average branch length or average molecular weight.
Sauces and pastes prepared from glycogen and tapioca
branched fractions were stable to freezing while waxy
maize amylopectin and sago starch amylopectin gave
unstable sauces and pastes when thawed at 27°C. after a

onedweek storage period.

CONCLUSIONS

White sauces prepared with wheat flour, corn starch,
sorghum starch, waxy corn starch, waxy sorghum starch, waxy
rice starch, ph05phate cross-linked waxy corn starch,
phosphate cross-linked waxy sorghum starch, and cross-
linked waxy corn starch showed marked instability within a
onedweek period when stored frozen at -l9.5°C. and thawed
at 35°C. Sauces prepared from waxy rice flour showed
instability only after a three-month storage period. The
instability appears to be related to a retrogradation of
the starch components as well as to the secondary de-
stabilization of milk proteins. Storage times and
thickening agents were statistically significant with
respect to their influence on several factors evaluated.
Centrifugation of sauces and turbidity tests of supernatant
liquor further supported evidence of a progressive de-
stabilization. Protein content of the supernatant was
related to the turbidity measurements, indicating either
physical or chemical involvement of protein in the
retrograded starch network. The nature of this involvement
was not determined in this study. Since instability of
starchewater pastes paralleled that of the frozen sauces, it
appeared that the protein was not a major factor in the
separation of the mdxtures.

62

63

Studies on behavior of nongranular amylopectins
showed that granular structure neither prevented nor
caused the instability. Instability could not be related
to either average branch length or average molecular
weight. Certain details of the fine structure of
amylopectins are not known and methods for their
determination are at present inadequate. The need for
further studies on the structure of amylopectins is

indicated by the results of this project.

APPENDIX

APPENDIX

TABLE VIII

“ISIS OF VARIAIG or PAIL SOME FOB SMOOMSS

 

 

 

Source of Variation ‘ 8.8. DJ. 11.8. P Value
Total 189*.52 599
Thickening Agents 861.97 9 95.78 7.97%
Storage Times $2.50 2 121.25 10.13%:
Judges 131.0» 1+ 32.76 9.75‘"
Replications 5.h6 3 1.82 0.22‘
Thickening Agents x

Storage Times 135.30 18 7.52 11.13"“I
Thickening Agents x

Judges 120.92 .36 3.36 ¥.97**
Mihmtgggts I 30. 56 27 lol‘l- 1.68'l'
Storage Times x Judges 7.95 8 0.99 1.57
Storage Times x

Replications 26.7‘t 6 It.“ 6.60»
Judges x.nep1ications 11.92 12. 0.99 l.¥7
Error 320.17 M7N' 0.68

 

1— .: L w -

‘ . .
computed using pooled interaction error

* significant at 5‘ level

" significant at 1’ level

65

-mw

_._.

(—

e-‘n. ‘ .e-u-a-

'- .. ‘

- v
I
\
.
.
_ C
C
r

III—eh.“

. ’09- o.
.. ~.. . ,_ .
. '-
I
.

n

a .
..l c- .e. .h.

a

\e--
. .
-r. 1"-
.
a
f
a
a
a
l
s
. r

66

TABIBII

AIALISIS 01' VW W PANEL some r03 SEPARATIOI

 

 

 

Source of Variation 3.8. D.!’. 11.8. P Value
Total 2589.96 599
Thickening Agents 1hh5.77 9 160.6h 11.32sc.
Storage Times 388.¥2 2 19¥.21 7.58‘**
Judges 73.88 1+ 18.»? 9331‘"
replications 3.83 3 1.28 0.05“
Thickening Agents x

Storage Times 21h.51 18 11.92 20.77""I
Thickening Agents 1

Judges 81.72 36 2.27 3.96M
Thickening Agents x

Replications 17.55 27 0.65 1.13
Storage Times x.Judges 9.68 8 1.21 2,11e
Storage Times x

Replications 75.00 6 12.50 21.78**
Judges x.Replications 7.5h 12 0.63 1.10
Error 272.0h EVH' 0.57

 

‘ computed using pooled int;raction error

* significant at 5: level
*‘significant at 11 level

o-~-

v-

p--. ..

\-"‘v

o-

.

‘7.- ~,

 

67

TABLE I

ANALEIS OF VARIAIQ OF PAIL SCMS rcm SEPARARD LIQUID

 

Source of Variation S.S. DJ. 11.8. P Value
Total 2571.76 599
Thickening Agents 1397. 56 9 155.28 10.78%"!I
Storage Times k16.19 2 208.10 8.00“"
Judges 57.36 h 18.3% 8.29%”
Replications V73 3 1.58 0.12‘
Thickening Agents x

Storage Times 223.01 18 12.39 20.27"
Thickening Agents x

Judges 72.17 . 36 2.00 3.28"
Thickening Agents x

Replications 16.67 27 0.62 1.00
Storage Times x Judges 10.66 8 1.33 2.18"'
Storage Times x '

Replications 73.92 6 12.32 20.16**
Judges x Replications 9.80 12 0.82 1.3'»
Error 289.69 Wk 0.61

 

‘ computed using pooled interaction error
* significant at 51 level
”significant at 1‘ level

O.‘

.-#r

‘V-n».—

68

TABLE II

ANALYSIS OF VARIANCE W PAEL some FOR “(HIS PEEL

 

i —Z.

w

 

Source of Variation S.S. DJ. 14.8. 1" Value
TQM 1768.73 599
Thichnins Aunts 718.27 9 79.81 6.88%:
Storage Times 117.73 2 58.86 5.55“
Judges 290.16 E» 60.0‘t 9,385"
Replications 1.86 3 0.u9 0.11‘
Thickening Agents x

Storage Times 186.38 18 8.13 11.89"
3:13:33? “a“. x 150.61 36 ”.18 6.12"!
Thickening Agents x

Replications 23.73 27 0.88 1.29
Storage Times x Judges 9.57 8 1.20 1.75
Storage Times x

Replications 18.86 6 2.88 3.62»
Judges xneplications 22.00 12 1.83 2.68"
Error 32h.07 87h 0.68

‘ computed using pooled interaction :rror
* significant at 91 level
Msignificant: at 15 level

-‘de's—

.ea...

.-

a...

0-. a.

fi" eat. ‘1:

69

TABLE XII

HALISIS OP VARIANCE OF PAEL some r03 PLAVQ

‘—

 

Souree of Variation 8.8. DJ. 11.8. P Value
Total 1366.39 599
Thickening Agents 208.83 9 23.20 2.19%
Storage Times 80.25 2 20.13 1.72‘
Judges 157.69 I» 39.82 2.32‘
Replications 22:51 3 7.50 1.02‘
“1:332: 4‘12? ‘ 62.61 is 3.1.3 2.3;...
“13333" “'5'“ x 256.71 36 7.13 5,35...
Thickening Agents x

Replications 25.3.5 27 0.9+ 0.77
Storage Times x Judges l+3.00 8 5.37 #31"
Storage Times x '

Replications 17.27 6 2.88 2.36*
Judges x Replications 53.85 12 ‘1».‘t9 3.68"
Error 578.22 #78 1.22

w

‘ computed using pooled interaction error
* significant at 55 level
I""‘significant at 15 level

- l . , a .. ‘ . . a, . . . _ , s .s . .
.- . g, — . s , - O H
O O O O
”"N - - - . . e ~... . -. - 1 . . a ,, . .
II.

“__‘_-.... o»— .a a ..>A-A .- a. cm“- - e v we— . .. - - sag-4

.._ _ . ., _ ...¢ — -. . fl...

5
r
I
.
4
.
~ I
I
1
‘-
nr... . s..- e. sen-.1..- ..
s
u
\
- ~-

~... -.

70

TABLE XIII

ANALYSIS a VARIAIG 0P PAEL SCUBS PCB PASB CHARACTER

 

 

Source of Variation 8.8. DJ. 1&8. P Value
Total 1852.80 599
Thickening Agents 852.11 9 99.68 6.l+9‘«
Storage Times 118.50 2 71.75 5.59%
Judges 111.82 M 27.96 b.1198"
Replications 1.1m 3 0.87 0.05‘ ‘
Thickening Agents x

Storage Times 170.83 18 9.119 13.33"
Thickening Agents x

Judges 129.08 36 3.58 5.0+"
mickening Agents x

Replications “-0.35 27 1.39 2.10'“
Storage Times x Judges 7.50 8 0.9+ 1.32
Storage Times x

Replications 20.11 6 3.35 8.713”
Judges x Replications 38.27 12 - 3.19 8.88"
Error 337.86 Wk 0.71

 

‘ compute} using pooled interaction error
"' significant at 51 level
”significant at 11 level

~ — _11.-..___'- e- .e‘ -- - - ~- .- k . -ma .. ~~--..._ . . o - "-e‘
.o a .. . . - . . m-- - rip _ - -..'.¢.. .Q. . . _‘- , ...,.s -,.._~ . - - -...‘.-- e
a e
C U C U C I .
* -.~- I- - . .7, Ice 9 9 be av4~—-- o a . 4-- e- - »--o . .— , , .. -. I - ' . m ‘-.- - e . -- t -
I
O
I s
, .
.
O C .
. , . . '-
I ,
e . O . ..
. ’ I e .
O O O ‘
I l
C C ' .
s
.
' e
. . K. ‘ O 4 g .
Y
I I
s
C O _ O -
. f , I 3 I
. a 5 I
e e - . ,e .
I . e
' f
.
,O 0 ~ 0 ‘
I e.
V.
.
m ,0 , - I
I I - . . .
1 .
. .
. .
t
m 0 , . .
I I I I
e - e . ,
”u e - . a. - ‘u .- - . .. «m 42- :- e ~<. -~ a " sue—u - ---'m .p..-¢ . a.he .e. . e.- --‘l-- . c .s o- I
o . o . .
, u ' e .

 

 

71

TABLE XIV

ANALYSIS OF VARIAIG OI" PAEI. SCORES FOB mum ACCEPTABILITY

 

‘—

 

" computed using pooled interaction error
"' significant at 55 level

"significant at is level

Source of Variation 8.8. DJ. 11.8. 1" Value
Total 1555.98 599
mickening Agents 5911.29 9 66.03 6.51%:-
Storage Times 152.53 2 76.26 5.97‘**
Judges 228.29 ‘1- 57.06 9.36%:
Replications 3.28 3 1.10 0.18‘
Thickening Agents x
Storage Times 129.07 18 7.17 12.56":
Thickening Agents x
Judges 105. 26 36 2 . 92 5. 12”
Thickening Agents x
Replications 16.33 27 0.60 1.06
Storage Times x Judges 11.311 8 1.82 2.11811'
Storage Times x ‘
Replications 25.11 6 11.18 7.33"
' Judges x Replications 19.89 12 1.66 2.90“
Error 270.70 #79 0.57

.,.r(g.‘ s..-

2.

3.

6.

7.

9.

10.

LITERATURE CITED

Blom, J. and Schwarz B. Bindungstypen in der Starke.
Acta Chem. Scand. 6: 697-708 (1952)

Brown, F. Dunstan, 8., Halsall, T. G., Hirst, E. L.
and Jones, J. K. N. Application of new methods
of end-group determination to structural
problems in the polysaccharides. Nature 156:

785-786 (1995)

Cori, G. T. Enzymatic structure analysis and molecular
weight determination of polysaccharides. Makromol.
Chem. 20: 169-180 (1956)

Cori, G. T. and Lerner, J. Action of amylo-l,6-
glucosidase and phCSphorylase on glycogen and
amylopectin. J. Biol. Chem. 188: 17-29
(1951

Eckles C. H. Combs, W. B. and Macy, H. Milk
Milk Pr ducts. 3rd. edition. pp. 319 and
322-3 . Mc raw Hill Book Company, Inc., New
Ybrk. (1993)

Freudenberg K. and Boppel, H. Die Stelle der
Verzwelgung der Starkeketten. Naturwissenschaften
28: 26% (1980)

Guerin-Varry, R. T. Ann. Chim. Phys. (Paris) 56:
225-252 (183%); ibid., 61: 66-9’+ (1836)
As discussed in Hadley, J. A., Editor. Stgzch
% Its iv t v 8. Chapter 6. John Wiley
an Sons, c., New Ybrk. (195%)

Hamilton J. K. and Smith, F. Reduction of the
pro ucts of periodate oxidation of carbohydrates.
II. A new method for the end groug assay of
amylopectin. J. Am. Chem. Soc. 7 : 5907-
5909 (1956)

Hamilton, J..K. and Smith, F. Reduction of the
products of periodate oxidation of carbohydrates.
III. The constitution of amylopectin. J. Am.
Chem. Soc. 78: 5910-5912 (1956)

Hanson, H. L.i Campbell, A. and Lineweaver, H.
o

Preparat n of stable frozen sauces and gravies.
Food Technol. 5: H32-hh0 (1951)

72

 

e v e m
r _
I q ' U
. _ i _ _ :
e. a a 0
. m
.
_ . e. .. e
. . .
v / s
a- P.
.
‘ O . ‘ I
o v V
. e
_ I J Q
Q
a
‘ \
flrx Q 7 .. y -
\ P1.
a. Q - 3 ~ k \ s .
I . o
a . P.
-
I Q . .
. a o s
.s an
I . j e
O a
F. a.
, v e
o 9
u .
I I
s 0 e
m . fl. .
e a e e
. \
e.\ I g
o P
. 9
-
e a
pa s ..
. .
e
e e
1

"‘41.“

 

 

 

11.

12.

13.
1%.

15e

16.

17.

18.

19.

20.

21.

73

Hanson, H. L., Fletcher, L. R. and Campbell, A. A.
The time-temperature tolerance of frozen foods.
V. Texture stability of thickened precooked
frozen foods as influenced by composition and
staggge conditions. Food Technol. 11: 339b3h3

Hanson, H. L. Nishita, K. D. and Lineweaver, H.
Preparation of stable frozen puddings. Food
Technol. 7: h62-M65 (1953)

Hassid, W. Z. Private communication. (1957)

Haworth ‘W. N., Hirst, E. L. and Wébb, J. I.
Poiysaccharides. Part II. The acetylation and
methylation of starch. J. Chem. Soc. 2681-
2690 (1928)

Jackson, E. L. and Hudson, C. S. The oxidation of
alpha-methyl d-mannopyranoside. J. Am. Chem.

Soc. 58: 378-379 (1936)

Jackson, E. L. and Hudson, 0. 8. Studies on the
cleavage of the carbon chain of glycosides by
oxidation. A new method for determining ring
structures and alpha and beta configurations of
glggosides. J. Am. Chem. Soc. 59: 99h-1003

1 7

Lansky, 8., Kboi, M. and Schoch, T. J. Properties of
the fractions and linear subfractions from various
starches. J. Am. Chem. Soc. 71: 8066-8075 (1989)

Lerner J. Illingworth, B., Cori, G. T. and Cori, C. F.
Struc ure of glycogens and amylopectins. II.
Analysis by stepwise enzymatic de radation. J.
Biol. Chem. 199: 6819651 (19525

Lindet, L. Sur les etats que presente l'amidon dans
le pain tendre et dans le pain rassis. Bull.
Soc. Chim. 27: 638-639 (1902)

Maquenne, L. and Roux, E. Sur la constitution, 1a
saccharification et la retrogradation des empois
de fecule. Bull. Soc. Chim. 33: 723-731 (1905)

Meyer . H. Uber Verzweigungen und‘Vernetzungen der
égglopigfiénmolekule. Naturwissenschaften 28:

 

"."11‘ ‘

1. ,

 

 

 

 

22.

23.

2h.

25.

26.

27.

28.

29.

30e

31.

32.

33.

7%

Meyer, K; H. Chemistry of Glycogen. Ford, F. and
Werkman, C. H., Editors. Advances in Enmologz
Vol. III pp. 109-135 Interscience Publishers,
Inc., New York. (l9‘+3)

Meyer, K. H. and Bernfeld, P. Recherches sur l'amidon.
V. 'L'amylopectine. Helv. Chim. Acta 23:

875-885 (1990)

Meyer K. H. and Bernfeld, P. Recherches sur l'amidon.
1511. Sur la structure fine du grain d'amidon et
sur les phenomenes du gonflement. Helv. Chim.

Acta 23: 890-897 (1980)

’Meyeri K. H. and Bernfeldi P. Recherches sur l'amidon.

IV. La reaction co oree a 1'iode de l'amidon et
dugfilycogene. Helv. Chim. Acta 2%: 389-393
1 1

Meyer, K; H., Bretano 'W. and Bernfeld, P. Recherches
sur l'amidon. II. Sur la nonhomo eneite do
l'amidon. Helv. Chim. Acta 23: 5e853 (1980)

Meyer K; H. and Fuld, M. Recherches sur 1'amidon.
ill. L'arrangement des restes de glucose dans le
glycogene. Helv. Chim. Acta 2%: 375A378 (1981)

Meyer K; H. and Fuld M. Recherches sur l'amidon.
KVII. L'amidon du riz collant. Helv. Chim. Acta
2%: INCH-1HO7 (1981)

Meyer . H. and Heinrich, P. Recherches sur l'amidon.
IXIV. La composition de quelques especes d'amidon.
Helv. Chim. Acta 25: 1639-1650 (1992)

Meyer K; H. and Press, J. Recherches sur l'amidon.
.1. La degradation du glycogene par la beta-amylase.
Helv. Chim. Acta 21:: 58.62 (191:1)

Meyer K; H. and Settele, W. Recherches sur l'amidon.
LII. _Sur l'inhomogeneite des amylopectines de
tapioca et de waxy maize. Helv. Chim. Acta 36:
197.20% (1953)

Meyer, K; H., Wertheim, M. and Bernfeld, P. Recherches
sur l'amidon. IV. Methylation et determination
des groupes terminaux d'amylose et d'amylopectine
de mais. Helv. Chim. Acta 23: 8655875 (1990)

Montgomery’ E. M., Weakley, F. B. and Hilbert ‘G. E.
Isolation of 6-(alpha-D-glucopyranosyl)-D—glucose
(Isomaltose) from enzymic hydrolysates of starch.
J. Am. Chem. Soc. 71: 1682-1687 (1999)

 

w‘"

‘7‘1‘

 

 

3k.

35.

36.

37.

38.

39.

MO.

#1.

M2.

M3.

#1:.

75

Nageli, C. W. 'Die Starkekorner' Zugich (1838) h
As discussed by RadleyéhJ. A., E tor. ggc

,ang,;§§ De ivatives. apter 6. John Wi ey
and Sons, Inc., New Yerk. (195M) .

Nageli C. W. 'Beitrage zur naheren Kenntnis der
Starkegruppe in chemischer und physiologischer
Bezeihung‘ Leipzig (187%) As discussed by
Hadley, J. A., Editor. Starch ggg;1§§_Degivatives.
Chapter 6. John Wiley and Sons, Inc., New Y0rk.

(199+)

Noznick, P. P. Merritt, P. P. and Geddes, W. F.
Staling studies on breads containing w maize
starch. Cereal Chem. 23: 297-305 (1 6)

foicial Methods of Anal sis 2f|t e Association,2;
prriTr-e a Wemis amt . ion
19 . 202

Peat, 8. Whelan, W. J. and Thomas, G. J. Evidence of
multiple branching in.waxy maize starch. J.
Chem. Soc. ‘+5%6-br51+8 (1952)

Peat, S. Whelan, W. J. and Thomas, G. J. The enzymatic
synthesis and degradation of starch. Part XXII.
Evidence of multiple branching in waxy maize starch.
A correction. J. Chem. Soc. 3025-3030 (1956)

Potter, A. L. and Hassid, W. Z. Starch. I. End-group
determination of amylose and amylopectin.by
periodate oxidation. J. Am. Chem. Soc. 70:
3888-3890 (1998)

Potter, A. L. and Hassid, W. Z. Starch. II. Molecular
weights of amyloses and amylopectins from starches
of various plant origins. J. Am. Chem. Soc. 70:
3778-3777 (1998)

Schoch, T. J. Physical aspects of starch behavior.
Cereal Chem. 18: 121-128 (1991)

Schoch, T. J. Fractionation of starch by selective
precipitation with butanol. J. Am. Chem. Soc.
6%: 2957-2961 (1982)

Schoch, T. J. The fractionation of starch. Pigman, W.
W. and Welfrom, M. L., Editors. Advances in
ggpbohzgpgte ggemistrx. Vol. I pp. 2 7.277
Academic Press Inc., New York. (1985)

w.“" e (

 

 

 

.Jl'slllllllllls (It'll-

5—“‘

 

%5.

%6.

%7.

%8.

%9.

50.

51o

52.

53.

55.

76

Schoch T. J. Naming starch fractions. In "Words
a out words." Chem. Eng. News. 30: 320% (1952)

Schoch T. . The starch fractions. Radley, J. A.
Editor. Starch and Its Derivatives. Chapter 6.
John Wiley and Sons, Inc., New York. (195%)

Schoch, T. J. and French, D. Studies on bread staling.
I. The role of starch. Cereal Chem. 2%:

231-289 (1987)

Tanner, L. P. and Englis, D. L. A study of starch from
different varieties and t pes of corn. Food
Research ,5: 563-581 (1 0)

Thompson A. and Wolfrom M. L. Degradation of amylo-
pectin to panose. J. Am. Chem. Soc. 73:
58%9-5850 (1951)

webb, B. H. and Hall 8. A. Some physical effects of
freezin upon milk and cream. Journ. Dairy
Sci. 1 : 275-286 (1935)

Wolfrom, M. L. and Thompson, A. Occurrence of the
(l,3)-linkage in starches. J. Am. Chem. Soc.
78: 9116-9117 (1956)

W00druff, S. Microsco y of starch by the S ierer lens.
Ind. Eng. Chem. 0: l%09-l%13 (19385

Woodruff, S. and Hayden H. The effect of freezing on
the physical and microscopic character of gels of
corn and wheat starches. J0ur. Agr. Research

52: 233-237 (1936)

Woodruff, 8. and MacMasters, M. M. Gelatinization and
retrogradation changes in corn and wheat starches
shown by photomicrografifls. University of Illinois
Agr. Exp. Sta. Bull. 5 (1938)

VanLeeuwenhoek, A. Opera Omnia. 'EpiStolae Physiologicae
super compluribus Naturae Arcanus' Volume IV
pp. 232,(l7l9) As discussed by Hadley J. A.,

Editor. St c and ts ivatives. Chapter 6.
Jehn Wiley EEK Sons, Inc., New YCrE. (195%) p

w-H

v"‘

 

 

 

 

 

’V“"‘. i‘

 

 

 

w-u""

 

 

ROOM USE ONLY.

514""

r use. as»

   

Demco—293

 

 

 

HICHIGQN STQTE UNIV LIBRQRI

llllllllll llllllllllllllll IIIIH llill IIHIIIIIII ll ililill

31293007016045