v "If .
I

 

—v~<‘ .n-o—r~ ' A

TF6! EFFEC‘E' OF VAREGUS SUGARS AND
SERWS 0% W15 VESCQSETY AND GEL
$TREF€GTH 0F A FWE PER CENT CMMS?ARCH PASTE

Thai: for flu [309m 94 M. S.
MECHfiGM STATE UNNERSETY
Mme Lyons Egan
1957

jfifibls

I
/

 

f-‘

3 1293 10668 3232

1;; x WWWWWMiami x

This is to certify that the
thesis entitled
THE EFFECT OF VAQIOUS SUGARS AND ‘JI’iUPS ON ”It 55 VISCOSITY
AND GEL S'I‘RiNGTH OF A FIVE PER CENT COPIISTARCH PASTE

presented by
Maura Lyons Bean

has been accepted towards fulfillment
of the requirements for

M. So degree in FOOdS and Nutrition

 

‘/ Major professor

Date W

0-169

 

 

 

MSU

LIBRARIES
~—

\—

 

 

 

RETURNING MATERIALS:
RTace in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped beIow.

 

 

I<~8¥{jfi;tj{§é

15.9%: 1 5 W534
.' '.|‘n\.&; 3.1” ‘ " I

{4 s ‘ .
~ H.‘ u 1' .:':
“a k .' (I *L/

NU¥*A9s“»’td

 

 

 

THE EFFECT OF VARIOUS SUGARS AND SIRUPS ON THE VISCOSITY AND
GEL STRENGTH OF A FIVE PER CENT CORNSTARCH PASTE

by

Maura Lyons Bean

A THESIS

Submitted to the College of Home 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

‘(ear 1957

.ABSTRACT MAURA LYONS BEAN

The purpose of this investigation was to study the
effects of various sugars and sirups on the viscosity and
gel strength of a 5% cornstarch paste. The 5% paste repre-
sents the approximate concentration of cornstarch present hi
cream pie fillings. Fructose, glucose, sorbitol, lactose,
maltose, sucrose, invert sirup and three corn sirups of
different D.E. values were added to the aqueous cornstarch
slurry before cooking. The dry substance of the sweetening
agent was present in concentrations equal to 5, IO, 20, 30,
and 50% of the total water present.

The Corn Industries Viscometer was used for pasting
and also for measuring and recording the viscosity of the
hot paste during cooking. The paste was heated by a glyc-
erol-water bath maintained at 100.0 1 0.20. Pasting was
terminated 5 minutes after maximum viscosity had been
reached.

Gel samples were prepared immediately after cooking
and were aged for 18 hours at approximately SOCL Gel strength,
was determined by an embedded disk method that utilized the
recording device of the Corn Industries Viscometer.

In amounts greater than 20%, there were significant
differences between the effects of the different sugars and
sirups. The monosaccharides were markedly different from

the other sweetening agents.

ABSTRACT MAURA LYONS BEAN

With all the sweetening agents, the maximum hot-paste
viscosity increased as the sugar and sirup solids were
increased up to 10 or 20%. At the 30 and 50% levels, the
viscosity decreased. The pastes containing monosaccharides
had higher viscosities than those containing di- or polysac-
charides at corresponding concentrations with the exception
of the 5% level.

Gel strength was increased above that of the control
when 5% concentrations of monosaccharides or some monosac-
charide-containing sirups were present. Above 10% all of
the sugars and sirups progressively lowered the strength of
the gels. The decreases were greater with the disaccharides
and sirups than with fructose, glucose and sorbitol.

The differences in the gelatinization behavior of the
starch granules seemed to be related to the size of the sugar
and sirup molecules. These differences showed their effects
in the resulting gels. Sufficient swelling of the starch
granules, as evidenced by the viscosity of the paste, was
necessary for gel formation when the paste was cooled. The
stereochemistry of the sugars may also influence starch be-
havior. There seemed to be no direct relationship between
the thickening and gel-forming behavior of the pastes and

the number of sugar hydroxyl groups present.

ACKNOWLEDGEMENTS

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

She wishes to thank Dr. Dena Cederquist for her kind
understanding during the writing of this thesis, Dr. Elaine
Rutherford for her interest and helpful suggestions, and
Dr. William Baten for his assistance in the statistical
analysis of the data.

The author owes a special debt of gratitude to Dr.
Pauline C. Paul, with whom she worked for several years and

who inspired and encouraged her in so many ways.

ii

TABLE OF CONTENTS
Page
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . I
REVIEW OF LITERATURE . . . . . . . . . . . . . . . . . 3
Starch Fractions . . . . . . . . . . . . . . . . . 3
Granular Structure . . . . . . . . . . . . . . . . 7
Swelling . . . . . . . . . . . . . . . . . . . . . 9
Gelatinization . . . . . . . . . . . . . . . . . . IO
Viscosity . . . . . . . . . . . . . . . . . . . . . 12
Gel Properties . . . . . . . . . . . . . . . . . . I5
Effects of Sugars . . . . . . . . . . . . . . . . . 19
PROCEDURE . . . . . . . . . . . . . . . . . . . . . . 26
Equipment . . . . . . . . . . . . . . . . . . . . . 26
Viscometer . . . . . . . . . . . . . . . . . . . 26
Gel tester . . . . . . . . . . . . . . . . . . . 28
Ingredients . . . . . . . . . . . . . . . . . . . . 30
Formulas . . . . . . . . . . . . . . . . . . . . . 32
Preparation . . . . . . . . . . . . . . . . . . . . 3h
Ingredients . . . . . . . . . . . . . . . . . . 3h
Slurries . . . . . . . . . . . . . . . . . . . . 35
Viscosity Tests . . . . . .‘. . . . . . . . . . . . 36
Preparation of Gels . . . . . . . . . . . . . . . . 37

Measurement of Gel Strength . . . . . . . . . . . . 37

iii

iv

Page

RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . DO
Viscosity . . . . . . . . . . . . . . . . . . . . . DO
Temperature . . . . . . . . . . . . . . . . . . . . 55
Gel Strength . . . . . . . . . . . . . . . . . . . 55
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 66
CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . 68
APPENDIX . . . . . . . . . . . . . . . . . . . . . . . 71

LITERATURE CITED . . . . . . . . . . . . . . . . . . . 76

Table

VI.
VII.
VIII.

IX.

XI.

XII.

XIII.

XIV.

LIST OF TABLES

Sources of Sugars and Sirups . . . . . . .
Formulas for Starch Mixtures
Approximate Composition of the Corn Sirups

Effect of Sugars and Sirups on Maximum Hot-
Paste Viscosity

Analysis of Variance for Maximum Hot-Paste
Viscosity

Paste Temperature at Maximum Viscosity . .
Effect of Sugars and Sirups on Gel Strength
Analysis of Variance for Gel Strength

Analysis of Variance for Maximum Hot-Paste Vis-
cosity at the 5% Concentration of Sweetening
Agent

Analysis of Variance for Maximum Hot-Paste Vis-
cosity at the 10% Concentration of Sweetening
Agent . . . . . . . . . . . . . . . . . .

Analysis of Variance for Maximum Hot-Paste Vis-
cosity at the 20% Concentration of Sweetening
Agent . . . . . . . . . . . . . . . . .

Analysis of Variance for Maximum Hot-Paste Vis-
cosity at the 30% Concentration of Sweetening
Agent

Analysis of Variance for Maximum Hot—Paste Vis-
cosity at the 50% Concentration of Sweetening
Agent I I I I I I I I I I I I I I I 0

Analysis of Variance for Gel Strength at the
5% Concentration of Sweetening Agent

V

Page

31

33
as

50

72

72

73

73

71+

Table

XVI.

XVII.

XVIII.

Analysis of Variance for Gel Strength at the
10% Concentration of Sweetening Agent

Analysis of Variance for Gel Strength at the
20% Concentration of Sweetening Agent .

Analysis of Variance for Gel Strength at the
30% Concentration of Sweetening Agent

Analysis of Variance for Gel Strength at the
50% Concentration of Sweetening Agent

Page

7N

7h

75

75

LIST OF FIGURES

Figure

I. Effect of sugars and sirups on the gelatiniza-
tion of cornstarch . . . . . . . . . . . . .

2. Gelatinization curves for cornstarch pastes
containing 50% fructose, glucose, sucrose,
and invert sirup . . . . . . . . . . . . . .

3. Gelatinization curves for cornstarch pastes
containing 50% glucose, maltose, and corn
sirup hydrolyzed to 63.7, 56.3, and
u3 DIEI I I I I I I I I I I I I I I I I I I

h. Gelatinization curves for cornstarch pastes
containing 50% fructose, glucose, sorbitol,
lactose, maltose, and sucrose . . . . . .

5. Effect of per cent concentration of sugars and
sirups on maximum hot-paste viscosity . . .

6. Effect of moles of sugar on maximum hot-paste
viscosity . . . . . . . . . . . . . .

Effect of per cent concentration of fructose,
glucose, sucrose, and invert sirup on gel
Strength I I I I I I I I I I I I I I

8. Effect of per cent concentration of glucose,
maltose, and three corn sirups on gel
strength . . . . . . . . . . . . . . . . .

9. Effect of moles of sugar on gel strength . . .

Page

NI

1&5

MB

51

53

61
63

INTRODUCTION

Cornstarch is used in food preparation for its thicken-
ing and gel-forming properties. In many dessert—type products,
such as puddings and cream pie fillings, it is cooked in the
presence of sucrose, milk, and other ingredients. Methods of
combining and cooking these ingredients are relatively simple
but the behavior of starch sometimes becomes complex in the
presence of other ingredients and failures often occur when
these products are prepared.

A comparison of recipes for cream pie fillings showed
that the amount of starch used varied over a wide range (1.6
to 6.1% in those examined), the amount of sugar also varied
(13.6 to 31.1%), and the other ingredients varied to a lesser
degree.

In recent years, several studies have been reported on
the effect of sucrose on the behavior of cornstarch during
cooking (12, 26, MI), the effect of various aqueous media on
the gelatinization of cornstarch (23), and the effects of
combinations of ingredients as found in cream pie fillings
(ll, 21, 25). Some studies have been reported on the effect
Of sucrose on pastes and gels of various starches (20, 31, 60).
Buchanan and Lloyd (15) studied the effect of corn sirups on

some properties of cooked starch.

I

With the increased use of various sugars and sirups
other than sucrose in food preparation, a better understanding
of their effects on other food ingredients, such as starch,
seems desirable. Generalizations can be made to explain the
behavior of starch when gelatinized in milk but unless the
effect of each of the food constituents present in the milk,
including lactose, is known, the basic reasons for the behav-
ior of the starch cannot be known. With increasing use of dry
milk solids and stress put on doubling or tripling the amount
present in fluid milk, it seems even more important to study
the effect of the various basic constituents.

This investigation was undertaken to study the effects
of various sugars and sirups on the viscosity and gel strength
of a 5% cornstarch paste. Distilled water was the liquid
medium for pasting, and various levels of glucose, fructose,
sucrose, lactose, maltose, sorbitol, corn sirup, or invert
sirup were added to the cornstarch slurry before cooking.

The concentrations of the sweetening agents ranged from 5% to
50% of the weight of the water present and covered the amounts
used in preparing cream pie fillings.

It is hoped that the work described here will provide
some of the basic information needed for further work on the
effects of food constituents on the physical properties of

starch thickening agents.

REVIEW OF LITERATURE

Progress in explaining the behavior of starch during
cooking was hindered until recently by a lack of knowledge
concerning the composition of the starch. Early investigators
reported inconsistent results on pasting different starches
and on pasting the same starch under different conditions.

It is now known that starches vary due to the structure and

the amounts of the constituent molecules within the granules
(38) and that the properties of a single starch can vary de-
pending on the manufacturing process it has undergone (h?) as
well as on the variety and on the conditions during the growing
season (59). Differences in gelatinization procedure are also

known to affect the resulting pastes (3, 29).
Starch Fractions

Many early studies assumed starch to be a single entity.
Some workers mentioned the possibility of more than one compo—
nent but due to inadequate methods of separation their results
varied and in some cases contradicted each other.

In a review of the starch fractions, Schoch (53) pointed
out that as early as 1716 Leeuwenhoek, in his microscopic
studies, recognized the presence of more than one substance

in the starch granule. He described a hull that remained

u

undissolved when starch was heated in water. He termed this
component an "undigestible hull" and called its contents the
"nutritive substance". A century later other European authors
described two fractions on the basis of their relative solu-
bility in water. Further work produced a theory which des-
cribed three components by their degree of solubility in hot
and cold water. Schoch pointed out that this latter theory

is frequently mentioned but has never been widely accepted.

In the latter part of the nineteenth century, C. W.
Naegeli, interested in the botanical nature of starch, con-
tributed much to the basic knowledge of our present-day con-
cepts of starch structure. Schoch named him the father of
starch chemistry. Naegeli considered the insoluble component
as being similar to wood cellulose and thought of the soluble
component as a different modification of the same substance.
Only recently has this concept been proved to be in error.
However, he was among the first to consider the theory of
mlcellar organization within the starch granule.

Early in the twentieth century Maquenne and Roux (35)
by leaching starch with hot water, were able to dissolve part
of it. This dissolved material precipitated on standing.
These workers used the term "retrogradation" to describe the
phenomenon of insolubilization. The deposited material could
be redissolved when heated to 150°C. and the solution gave a

blue color with iodine.

As Schoch (53) pointed out they did not actually iso-
late the other fraction but mentioned it as a minor component,
amylopectin. Actually this fraction is now known to make up
about 70-85% of the cornstarch granule. Such an error was
due to the inadequacy of their method of separation; however,
these studies initiated further study of starch from a chem-
ical point of view.

Later investigators used enzymes to study the compo-
nents. Incorrect assumptions were made about the fractions
due to a lack of knowledge of the behavior of the enzymes
and of the interference by the non-carbohydrate impurities
in the starch.

More recent studies have utilized certain polar organic
substances for the fractionation of starch. In l9hl and 19u2,
Schoch (DB, b9) reported the precipitation of the amylose
component by butyl or amyl alcohol. This selectively pre—
cipitated component is soluble only in hot water, and on
cooling retrogrades or forms an irreversible gel. The super—
natant liquor can be treated with methanol to give a precip-
itate that is easily dissolved and which does not retrograde.

Confusion in naming the fractions of starch has added
to inaccuracies in interpreting various authors' works. Some
investigators have used the terms, amylose and amylopectin,
in a manner opposite to others. K. H. Meyer (37) proposed
naming the linear fraction "amylose" and the branched compo-

nent "amylopectin". Schoch (50) preferred to use the terms

"A—fraction" and "B—fraction" to describe the linear and
branched chains respectively. More recently Schoch (52)
suggested the terms "linear starch fraction" and "branched
starch fraction" to simplify the nomenclature.

Although differing in their naming of the fractions,
these and other starch chemists generally are in agreement
as to the properties of the two carbohydrate components of
starch. The A-fraction or amylose is usually accepted to be
the linear fraction or one containing very few branched
chains. It is a polymer of glucose units joined by ancx-l—9
u-glucosidic linkage. Such long linear chains of a poly-
saccharide contain quantities of hydroxyl groups which allow
for hydrogen bonding forces. Schoch (51) pointed out that
the A-fraction is soluble in hot water but shows a tendency
to retrograde or revert to an insoluble state on cooling due
to the strong associative forces generated by the hydrogen
bonds. In low concentrations (1%) this insolubilization
produces a precipitate of amylose. In greater amounts (5%),
an irreversible gel is produced from the A-fraction.

The B-fraction or amylopectin is considered to be a
relatively branched molecule. Molecular weights from one to
six million have been reported for amylopectin (NS). The
branches have been shown to be 22 to 27 glucose units in
length (uh), joined by the samecx-l—oh-glucosidic linkage
as in the linear chains. .At the points of branching the

units are connected by an a—l—a6-glucosidic linkage. Many

branches are combined in one tree-like molecule containing
several thousand glucose units. This highly branched struc-
ture does not allow for the orderly association between mol-
ecules possible with the linear chains; hence this component
does not have a tendency to retrograde or gel, thereby showing
greater colloidal stability. It is useful as a thickening
agent because of its capacity to hold water within its branches.
Its presence in a granule diminishes the precipitating and
gel-forming tendencies of the linear fraction.

In reactions with iodine, the A-fraction shows a
strong affinity for the halogen giving a deep blue color,
while the B-fraction shows weak affinity with the production

of a red color (6).
Granular Structure

To interpret the mechanism of starch behavior in water,
a knowledge of granular structure Is necessary. Early studies
considered the starch granule to be composed of an outside
membrane or "sac" that was insoluble and the contents of such
a "sac" to be the soluble portion. Caesar and Cushing (18), in
discussing some of T. C. Taylor's unpublished work, pointed
out that he challenged the membrane theory by showing through
electrophoresis that the so-called "shell" was composed of the
same proportions of the A- and B-fractions as the entire granule.

K. H. Meyer (39) supported and modified some of the

earlier work to explain the structure of the starch granule.

He proposed that starch molecules are laid down in concentric
layers about a nucleus to form a granule. Both fractions are
oriented within these layers in a radial direction. At many
points, segments of the linear fraction and the linear outer
branches of the B-fraction associate side by side to form
crystalline bundles to which the term "micelles" was given

by Naegeli and retained by Meyer. These bundles or crystal-

“T“""§

line regions are connected by parts of the chains or branches
which are not Included in the micelles and which form the
looser, amorphous areas that allow some hydration and swelling
without rupture of the granule.

Meyer thought the micelles differed in size and that
some could be destroyed in warm water thus allowing further
swelling of the granules. Meyer felt that possibly larger
micellar regions occurred in the outer layer thus functioning
as a shell to hold the granules intact during swelling.

This latter theory, coupled with the concept that the
branched chain contributes most of the structure to the crys-
talline regions, has given rise to a new hypothesis that a
membrane does exist which is composed mainly of the branched
fraction. Badenhuizen (5), in discussing this, reported that
in studies on various starches, he has found no decrease in
the amount of linear fraction from the center of the granule
to its periphery. He found that the linear fraction was

either localized in the center or distributed regularly

throughout the granule. This latter finding is in agreement

with Taylor's work which was mentioned earlier.

Swelling

Starch is insoluble in water at room temperature. The
granules will become somewhat hydrated with water entering
into the intermicellar spaces but they will not swell appre-
ciably. This hydration is slow and is reversible (2h). Meyer
(39) has stated that the granules can take on 25—30% of their
weight in cold water.

As the temperature of the water is increased the hydro-
gen bonds which have held the molecules together in a compact
granular form are weakened and tend to dissociate. This
loosens many areas in the structural network and more water
is able to permeate the granules, thus causing noticeable
swelling. Some of the small crystallites dissolve and aid
in further loosening of the granule network. The larger mi-
celles function to keep the swollen granules intact. Some
amylose molecules, which are not held as firmly in the net-
work as the branched amylopectin, diffuse out of the granules
(39). This leaching out of the amylose or A-fraction is the
phenomenon which served as a basis in some of the early

fractionation studies on starch.

10
Gelatinization

The term gelatinization is used to refer to the changes
which starch undergoes as it swells in water under the influ-
ence of heat, salts, or other agents (59). The temperature
at which the granules swell to produce a noticeable increase
in the viscosity of the paste has been designated by many as
the gelatinization temperature. Many studies have been under-
taken to determine this temperature for various starches.

Kerr (35), in a review of the various methods used,
felt that the choice of a method for determining the tempera-
ture of gelatinization would depend on the investigator's
concept of which phenomenon in starch behavior defines the
point of gelatinization.

Alsberg and Rask (I), in l92u, observed in viscosity
studies on corn and wheat starches, that gelatinization does
not occur as a sharp transition point but instead is a con—
tinuous process which extends over a temperature range. They
felt that the gelatinization temperature was affected by many
factors both inherent in the starch and due to techniques of
manufacture and of preparation of the pastes.

Other workers have used microscopic techniques in
gelatinization studies. Uncooked starch exhibits a "maltese
cross" effect when viewed under a polarizing microscope. On
swelling the granules lose this polarization cross. This

phenomenon has been associated with gelatinization and there

II

is a temperature range over which it occurs that is character-
istic for each starch (bl, 5h).

Photomicrographs have also been used to study gelatin-
ization of starch. Starches differ in their appearance and
size depending on the plant source of the granules. Behavior
during gelatinization is also typical for each variety of
starch. These identifying characteristics have been shown
by Sjostrom (5b) in a series of photomicrographs on various
plant starches, both before gelatinization and at different
stages of hydration and swelling. Woodruff and MacMasters
(59) have studied the gelatinization of corn and wheat starches
through the use of photomicrographs.

Morgan (DO) used a photoelectric method for following
the swelling of starch pastes. When heated in water, starch
suspensions increase in translucency. By measuring the light
transmitted during heating of a starch slurry and plotting it
against the temperature of the paste, Morgan developed charac-
teristic curves for each type of starch. The curves showed
the progress and completion of gelatinization and from these
curves the gelatinization temperature could be determined.

Anker and Geddes (u) reported gelatinization studies
using a Brabender Amylograph. They preferred to term the
temperature at which the initial rise in viscosity occurred
as the transition temperature and felt that this point des-

ignated the commencement of the gelatinization process.

 

l2
Viscosity

One of the most important uses of starch in food prep-
aration is its ability to thicken an aqueous liquid. The
degree of thickening or viscosity referred to in such a paste
is usually an apparent or anomalous viscosity rather than true
viscosity. Brimhall and Hixon (lb) have shown that starches
exhibit true viscosity only in very dilute pastes, below 2%
for cornstarch. Such low concentrations are usually imprac-
tical in food preparation. Higher concentrations produce a
structural viscosity which is a function of the rate of shear.
Starch investigators usually refer to this apparent or anom-
alous viscosity simply as ”viscosity" or in some cases as
"consistency" (l6).

Katz and coworkers (3h) pointed out that starch pastes
are not colloidal solutions but are suspensions of highly
swollen granules. They concluded that the viscosity of starch
pastes is due to the flow of water which is made more diffi-
cult by the swollen granules and is also due to the rubbing
of the granules against each other.

Anker and Geddes (h) stated that the viscosity exhib-
ited by starch pastes depends on the extent of aggregation
of the granules, the extent of swelling, and the extent of
granule disintegration or rupture. These factors are influ-
enced by the variety of the starch, the method of manufacture,

and the techniques used in preparing the starch paste (36).

13

Many investigators have reported the influence of such
factors as rate and duration of heating, method and rate of
agitation, concentration of starch, pH, and presence of other
substances.

Bechtel (8) studied pasting of starches with the Corn
Industries Viscometer and found that rapid heating lowered
the gelatinization temperature and increased the maximum vis—

cosity of 5% pastes of unmodified cornstarch. These results

 

were in agreement with those observed by Caesar and Moore (17),
using a Caesar Consistometer.

Harris and Jesperson (29) studied the swelling power<sf
some cereal starches and found that beyond the first marked
rise in viscosity there was no increase in the volume of
swollen granules with a long heating period. After a peak
viscosity was reached, there seemed to be a decrease in vis-
cosity with continued cooking. Anker and Geddes mentioned
granule disorganization as the commonly accepted reason for
this decrease in viscosity and also suggested thatzniincrease
in the permeability of the swollen granules might play apart.

Bechtel and Fischer (9) concluded that the rate of
pasting is the variable most affected by the rate of agita-
tion. They pasted one sample in a double boiler, having a
capacity of one liter and fitted with a high speed mechanical
stirrer. Another sample was pasted in the Corn Industries
Viscometer, which also had a capacity of one liter. The

stirring device of the viscometer was operated on slow speed.

ill

Both pastes were heated in a 92°C. water bath. The IO% corn—
starch slurry, pasted in the double boiler with more rapid
agitation, was gelatinized in less than one-half of the time
required for the paste cooked in the viscometer.

Caesar (l6) pointed out that viscosity decreased with
severe agitation and that this was a disturbing factor in
early gelatinization studies where the agitation in the in-
struments used, the Stormer and MacMichael viscosimeters,
influenced the viscosity.

Anker and Geddes (h) found that the concentration
of starch influenced markedly the characteristics of the
paste. They reported that as the starch concentration in-
creased, there was an appreciable decrease in the temperature
at the initial rise in viscosity, a marked increase in maximum
viscosity, and a more rapid decrease in viscosity after maxi-
mum. Bechtel (8), in studying the same variable on the Corn
Industries Viscometer, reported similar results. He also
found that successive equal increases in starch concentration
caused increasingly large differences in viscosity. Both
studies showed that there was a linear relationship between
the logarithm of the starch concentration and the logarithm
of the maximum paste viscosity.

Anker and Geddes in their studies on the effect of
pH on the gelatinization of starch used buffered solutions to
regulate pH values. They found that the maximum viscosity was

decreased as the pH was increased from u to 7. They used

15

bimaleate and citrate buffers and found that the different
buffers affected the viscosity of the pastes to different
degrees. They suggested that the results could be due to
the ions present rather than to the pH.

Bechtel (8) used sodium hydroxide and hydrochloric
acid in his pH studies in order to avoid the possible effects
of the ions contained in the buffers. His results showed no
marked differences in the maximum or final viscosity on un-
modified cornstarch between a pH of u and 7. Higher or lower
pH values did show some effect. At a pH of 9, the highest
reported, the unmodified starch had a higher maximum viscosity

and a greater degree of breakdown than at lower pH values.

Gel Properties

In addition to the thickening ability of a starch, its
gel-forming properties are also important in characterizing
a particular type of starch. As mentioned previously, the A-
fraction or amylose is responsible for gel formation in starch
paste. The highly ramified structure of the B-fraction does
not allow for association of its branches and so It cannot
form a gel.

Some early workers thought that on heating in water,
the swollen granules disintegrated to form colloidal solutions
that set on cooling in a manner analagous to gelatin. Alsberg
(2) did not agree with this analogy. He felt that at lower

temperatures some of the rigidity of the granules was

 

recovered and that possibly the granule surfaces were sticky,
favoring their agglutination, therefore making the pastes
more resistant to flow.

Meyer and coworkers (38) were able to explain gel for-
mation on the basis of granule structure. They stated that
during gelatinization one end of a chain diffused out of a
granule while the other end was retained in the crystalline
regions or micelles. The so-called "dissolved" end became
entangled with chains from other granules and on cooling
formed a turbid, coherent, elastic gel. This reassociation
of chains of molecules has been attributed to the presence
of hydrogen bonds (I9, 56). The number of such connecting
links that formed in a given paste would directly affect its
gel strength.

The gel properties of starch have been measured by
numerous types of instruments which have actually measured
different characteristics of a gel, such as breaking point,
elasticity, rigidity, and resistence to cutting.

Some very early work, cited by Saare and Martens (b6),
attempted to measure the stiffness of a starch by soaking
threads with the starch paste, allowing them to dry in a
vertical position, then observing the position of a bend
when the threads were suspended horizontally. The farther
out on the thread that the bend occurred, the stiffer was
the starch. These workers cited another test whereby metal

spindles were dropped into a starch paste and the depth of

 

l7

penetration was a measure of the solidity of the paste. Their
own method consisted of implanting a disk in a warm paste,
allowing it to cool, and then measuring the force necessary

to pull it out. This method has undergone many modifications
and the latest mechanized refinement of it (32) is the one
used in this study.

Woodruff and Nicoli (60) studied starch gels by obser-
vation of the molded gel when turned out of its container.
Photographs were used as a permanent record of the gel's
characteristics.

Woodruff and MacMasters (59) used a Tarr—Baker jelly
tester which measured the hydrostatic pressure necessary for
a plunger to break through the surface of a gel. Their studies
on viscosity as measured by a Stormer viscosimeter and on gel
strength as described above gave good evidence that these two
properties measured two unrelated characteristics of starch
pastes.

Brimhall and Hixon (13) felt that actual disruption
of the gel structure involved variability in the degree it
stretched and so was not accurate. They devised a rigido-
meter which measured the resistance offered by a paste to
being stretched or its elasticity.

Bechtel (10),in a study on the gel properties of native
and modified cornstarch,compared several gel testing instru—
ments, the rigidometer mentioned above, several plunger-type

instruments, and a modification of the embedded disk

LP. "v.1“ ‘fln—T AV —‘ ‘“
.

18

instrument. The latter instrument gave the best results on
replicate tests and the values obtained correlated well with
those obtained by the other instruments. However, as pointed
out by the author, all the instruments were in the range ac-
ceptable for industrial evaluation.

Various authors point out the need for precision in
preparing and testing starch gels. The variety of starch, as
well as its concentration has a great effect on the strength
of the gel. The temperature to which a paste is cooked di-
rectly affects the gel strength. Woodruff and Nicoli (60)
determined the gelatinization temperature in 5% pastes of
various starches by noting a sharp change in translucency.
They pasted these starches to their respective gelatinization
points, all of which were below 88°C. and found that although
the pastes thickened they did not form gels. By heating sim-
ilar pastes to 90°C. and above, gels were formed on cooling
when cereal starches were used. Effects of cooking tempera-
ture were also noted by Brimhall and Hixon (13) in their
rigidity studies. Bechtel (10) found that unmodified corn-
starch gels increased in breaking strength when the cooking
temperature was increased to 9u°C. A further temperature
increase seemed to decrease the gel strength.

Osman and Mootse (h3)studied gels of many starches
pasted in distilled water in the Corn Industries Viscometer.
They found that the length of the cooking period after the

maximum viscosity was reached markedly influenced gel

19

characteristics. The gel strength, as measured by an embedded
disk method, was greater for all the starches pasted to five
minutes after maximum viscosity as compared to the strength

of those pasted only to maximum viscosity. The gels prepared
from pastes cooked for 20 minutes after maximum showed in-
creased strength for some starches and decreased strength for
others.

The time and temperature of aging also affects the gel
strength and so must be controlled in order to get reproducible
results. Bechtel (10) stated that starch gels increase rapidly
in strength during the first 10 hours and then the rate drops
off. Gel strength decreases with increasing temperatures. The
latter phenomenon has been observed in the writer's laboratory
and steps were taken to standardize the testing conditions for

this study.
Effects of Sugars

Most of the work reporting the influence of sugars on
starch behavior has been connected with studies on cornstarch
as affected by sucrose. Some authors have studied other
starches but very few have used other sugars.

Woodruff and Nicoli (60) studied gels of various
starches as affected by sucrose. Using 100 grams of a 5%
starch paste, they added 10, 30, 50, and 60 grams of sucrose
and gelatinized the pastes in test tubes heated in a water

bath to 90°C. The gels were poured into small crucibles,

20

aged 2h hours, and then turned out of the molds for comparison
by observation. The cereal starches, including cornstarch,
gave well-formed but increasingly more tender gels with 10,
30, or 50 grams of sucrose. With the addition of 60 grams
of sucrose a thick sirup rather than a gel was obtained. The
authors wondered if the 60 grams of sucrose might not have
acted as a diluent since the volume of the paste had been
increased to 132 cc. However, when the control slurry was
diluted with water to 132 cc. a firm gel was obtained. These ‘
authors suggested that large amounts of sucrose prevented the
starch granules from imbibing the water necessary for swelling.
Nevenzal (bl) also varied the concentration of sucrose
in a 5% starch paste. Fifteen grams of sucrose in 100 grams
of starch paste resulted in a gel that was tender yet firm;
30 grams of sucrose produced a gel that was too tender; and
no gel was formed when 50 to 66 grams of sucrose was present
in 100 grams of starch slurry. Comparison of photomicrographs
of starch granules with and without sucrose showed that at
69°C. control pastes without sucrose exhibited very little
anisotropy when viewed under a polarizing microscope, while
pastes gelatinized with 50% sucrose had much anisotropy even
at 80°C. Since the granules lose this "maltese cross" effect
on swelling, it appeared that the sucrose inhibited swelling
of the granules. The photomicrographs in this study also
showed decreased swelling of the starch granules pasted with

S'UCI‘OSB .

Using laboratory prepared starch, Chidester (21) noted
similar effects on gels of a 6% cornstarch paste prepared with
increasing amounts of sucrose. She pasted 5 grams of starch,
78.3 grams of water and amounts of sucrose ranging from 16.6
grams to 37.5 grams. The latter amount yielded a gel that
did not hold the shape of the mold but did have some body.

In the same laboratory, Bowersox (12) pasted various
concentrations of cornstarch in water in the presence of
sucrose. A 10% sucrose addition decreased the viscosity and
the gel strength slightly, 20 and 30% additions caused a more
marked decrease in these properties. The addition of 30%
sucrose resulted in a thick sirup in a u% cornstarch paste,
an extremely tender gel in a 5% paste, and a tender but well-
formed gel in a 6% paste.

In a bakers' trade journal, Trempel (57) stated that
sugar interfered with the swelling of the starch granules
during cooking and advised the addition of not more than
three and one-half times as much sugar as starch by weight
in preparing pie fillings. If more sugar is needed, he sug-
gested it should be added after the starch is gelatinized.

He suggested that corn sirup could be present up to four and
one-half times the weight of the starch without producing too
great an effect.

Whittenberger and Nutting (58) found somewhat different

behavior witit potato starch. On pasting the starch in water,

a temperature of 85°C. was sufficient for complete

22

gelatinization and well-formed gels. Above this temperature
the granules ruptured and the gel strength decreased. When
8% potato starch slurries were pasted with sucrose, the gel-
atinization temperature was raised and the gel strength in-
creased with increasing sucrose concentrations up to 50%.
Sixty-five per cent sucrose caused insufficient granular
swelling, as observed microscopically, even when the pastes
were heated to 95°C. and held for 20 minutes. It had been
erroneously suspected from the results of the increase in

gel properties with sucrose that a mechanism similar to that
in pectin gels might be present. It was thought that the
lack of gel formation in the concentrated sugar pastes might
be caused by persistent granules interfering with possible
hydrogen bonding effects of the sucrose. Subsequently, these
workers prepared starch-sugar pastes in which the granules
had been destroyed by mechanical shearing. No gels resulted,
thereby showing the need for intact granular structure for
gelation and the fact that sucrose does not play a role in
starch gels similar to that in pectin jellies. They reported
similar results using dextrose and glycerol.

By studying x-ray diffraction patterns of potato starch
as affected by sucrose, Nikuni and coworkers (M2) concluded
that sucrose prevented the retrogradation (linear association)
of starch by binding the water.

Hester (30, 31) studied the changes induced by sucrose

on various starches. A control cornstarch paste contained

23

6.5 grams of the dry starch per 100 grams of water. Four
increments of sucrose from 15.9 grams to 39.5 grams per 100
grams of water were used and the mixtures were pasted in the
Brabender Amylograph. Cooking was terminated as soon as the
pastes reached 950C. Gels were prepared and measured by the
Exchange Ridgelometer after aging overnight at room tempera- .
ture. The viscosity attained during gelatinization to 95°C. @
was greater when 15.9 grams of sucrose was added but with
higher concentrations of sucrose the viscosity at this tem- H
perature became progressively lower. The lowest level of
sucrose decreased the strength of the cornstarch gel. The
next increment yielded a very tender gel which collapsed
within a minute after removal from its container. With.
greater concentrations of sucrose, thin pastes were obtained
in place of gels. Chemical determinations of the amount of
amylose in solution showed a decrease with increasing amounts
of sucrose. Hester then hypothesized that less amylose was
available for gel formation. Other effects of sucrose noted
by this worker included a higher temperature at the initial
rise in viscosity, decreased granular swelling, and less dis-
integration of swollen granules with increasing concentrations
of the sweetening agent. Hester reasoned that the dissolved
sugar molecules hydrated, thus leaving less free water for ,
starch hydration.
Hains (26) prepared 11% cornstarch pastes to which were

added amounts of sucrose equal to 20, MO, and 60% of the weight

fill

of the water. She found that 20% sucrose raised the maximum
hot—paste viscosity as measured by the Corn Industries Visco-
meter, while the hO and 60% additions decreased the maximum
viscosity. Increasing amounts of sucrose increased.the tem—
perature and the time of initial rise in viscosity and of the
maximum viscosity. Gel strength as measured by a Fuchs pene—
trometer showed a decrease as sugar was increased.

Halliburton (27) studied the effect of sucrose on the
thickening and gel properties of wheat flour and the flour
fractions. With the starch fraction, 7.22 grams of starch
was pasted to 95°C. in 100 grams of water. Using a Brookfield
viscosimeter, she found that the viscosity was increased by
13.2 and 26.u grams of sucrose and then decreased when 39.6
grams was present. She stated that less disintegration of
the granules might have occurred when the lower amounts of
sucrose were present. This would increase the number of in-
tact swollen granules and thus increase the viscosity. Larger
amounts of sucrose tended to inhibit swelling of the granules.
Gel strength, measured by an Exchange Ridgelometer, decreased
with successive increases of sucrose until no gel was formed
with the greatest addition of sucrose. She stated that su-
crose reduced the hydration capacity and swelling of the
starch, thus decreasing the amount of amylose available for
gel formation.

Ferree L23), using a Corn Industries Viscometer, found

that 15% and 21% sucrose present in an 11% cornstarch paste

increased the maximum hot-paste viscosity significantly as

compared to the control. Twenty-seven per cent sucrose gave

a maximum viscosity value similar to that of the control.
The temperature at the initial viscosity rise and at the

maximum viscosity increased with successive increments of

SUCTOSO .

PROCEDURE

Equipment

Viscometer

 

The instrument used to cook the starch pastes and to
measure the viscosity was the Corn Industries Viscometer,
developed for the cornstarch industry, and described by Kesler
and Bechtel (36). It measures and records continuously the
viscosity of a paste during the heating period and also during
a cooling period if so desired.

.A steel beaker is immersed in a liquid bath for cooking
the paste. The bath may be water if a heating temperature of
less than 100°C. is required. The temperature of the liquid
is maintained by a submerged electric heating element, ther-
mostatically controlled to i O.2°C.

In this study it was necessary to use a mixture of
glycerol and water in order to keep the bath at 100°C. and
have a minimum amount of evaporation. Both the concentration
and the height of the bath were checked and adjusted daily
before a day's run was begun. The specific gravity of the
liquid medium was kept at 1.13 and the height was such that
during cooking it was about one to two cm. below the top of

the hot paste.

26

27

Evaporation losses in the pastes during cooking are
kept at a minimum with an efficient condenser-type cover.
This cover, which comes in two sections, is designed for use
as an air condenser or as a water condenser. In this study
it was used very effectively as an air condenser. There is
an opening in one side of the cover for a thermometer for
recording paste temperature during cooking.

The starch paste is kept suspended during cooking by
an intricate stirring device containing scraper blades which
effectively remove the layer of cooked paste from the sides
and bottom of the cooking beaker. This design aids to mini-
mize any insulating effect caused by an adhering layer of
paste.

The scraper blades are mounted on a hollow shaft
through which is suspended a propeller which serves to stir
the center of the paste and also to measure its viscosity.

As the propeller moves in a counterclockwise direction, it

is subjected to a force, the intensity of which depends on

the viscosity of the paste being cooked. This force or torque
is transmitted through a set of differential gears to a drum
connected by a cable to a dynamometer built into a recorder.
The dynamometer contains a weight arm and has an attached pen
which records the torque on a strip-type chart. The dynamom-
eter arm has interchangeable weights so that the full scale
reading of the chart can be varied from zero to 225, 450,

900, or 1800 g.-cm. depending on the viscosity of the starch

28

paste being studied. In this study no weights were used
since all the pastes were in the lower range of zero to 225
g.-cm.

The stirring mechanism has four speeds. For optimum
agitation of the paste and minimum shearing of the swollen
granules, Bechtel (7) recommends that speed 2 be used for
laboratory testing of starch pastes. This speed, which turns
the scraper at 2h r.p.m. and the propeller at 60 r.p.m., was
used in this study.

The chart recording device moves the chart downward
at the rate of one-half inch per minute. The charts have a
scale of 0 to 100. Bechtel's tables (7) are used for chang—

ing the chart readings to g.-cm. of torque.

Gel Tester

 

The gel strength was measured by a gel tester, de-
signed by Hjermstad (32) and built on this campus for our
laboratory. It utilizes the embedded disk method of meas—
uring gel properties. The principle of implanting a disk
in a hot paste, allowing it to set, and then measuring the
force required to remove the disk was first reported by Saare
and Martens (N6). The force required to pull the disk out of
the set paste is a measure of the resistance or strength of
the gel. This method was later modified by Hamer (28), Kerr
(35), and Bechtel (10). The latter worker claims that this

means of measuring gel strength has a precision of

29

approximately 3%. His study showed that the embedded disk
method was somewhat more sensitive than other instruments
for measuring properties of starch gels. It also correlated
very highly with other instruments in performance.

The apparatus, as designed by Hjermstad, is a device
which lowers a stage at a slow even rate of b.25 mm. per
minute. A jar containing the gel to be tested can be clamped
onto the stage. The gel contains an embedded disk which is
attached to a rod with a hooked top. A cord is attached to
the rod and also to the dynamometer of the Corn Industries
Viscometer.

Since the dynamometer of the viscometer is adjusted
to zero for the viscosity of water, it is necessary to attach
a counter weight to the cord to supplement the weight of the
rod. The combined weight should be such that a zero reading
is obtained on the chart recorder. The cord rides on a low-
friction pulley which is set directly above the center of the
gel.

As the stage is lowered, the embedded disk is pulled
down with the gel. Therefore a continuously increasing force
is exerted on the gel by the dynamometer arm. This force is
recorded on the chart of the viscometer which is started at
the beginning of the test. A curve results which gives a
measure of the deformation of the gel with time and also the

Yield point of the gel. Tender gels show a leveling off of

30

the curve at the yield point; stronger gels produce a curve
that drops suddenly at this point.

The disks made for our instrument were the same as
those proposed by Bechtel (10) and recommended by Hjermstad
(32). In place of the jars described by these investigators,
250-ml. beakers were used. These were etched with a line
to designate the pouring level of the hot paste. When the
paste was poured to this level, the disk was immersed 3 cm.
below the surface of the gel when supported by a slotted
cover designed to fit the beaker and to center the disk in
the gel. It was felt that the use of the 250-ml. beakers in
place of the jars which had a ISO-ml. capacity would lessen

effects due to the side of the container.
Ingredients

The cornstarch for this study was part of a lot ob-
tained from the Corn Products Refining Company of Argo,
Illinois. It was the ordinary cornstarch of commerce. The
entire amount was placed in a polyethylene bag and was stored
at room temperature.

The sugars and sirups were obtained from various
sources. These are listed in Table I along with the water
content of the sirups. The solids in the invert sugar sirup
were 5k.92% inverted according to the analysis supplied by
the manufacturer (33). The three samples of corn sirup were

of different degrees of hydrolysis as denoted by the term

31

dowpmmwfifimum>co do copmz do ofisooHoE oco mcwmpcoo *

 

mwocHHHH .omcd .SchEou mcficfieom mposo0cm dcou $.0H
mfiocHSHH .omc< .kcmafiou mcflcwmom mposooum dcou q.®H
mHocHHHH .omc< .zcmasou mcwcfiwom moUBUOcm dcoo :.@H

.> .z .muoch> ..ocH .mcwmsw 6cm madckm oocwdom om.mm

mocoow ooom msasmu

mHocHHHH .cmmoxst .zchEoo Hmowsocu ficofipmcmem
mHocHSHH aammoxjmzkxcmanu HmoHEocO ficofipmcmdm
oLmZmHoQ .coomcHEHHB .SchEoo coozom mmfip<

mwocfiHHH .omc< .zcmanu mcwcfidom moosUOcm dcou

mHocHHHH .cmmoxsmz..kcmafioo HmoHEocu Hzmfiumcwgm

.m.Q
.m.Q

.m.Q

m: .omxHECD ascfim dcou
m.bm .ooxHEcD ascfiw dcou
m.mb .ooxHEcD ascfim acou

ascww cmmsw pco>cH

machw
omocosw
$.a .o .mmooflaz-o
*Hmwooom .d .0 .mmooomqua
seduceswsto .Soucncow-m

Ammofiocouv omoosflo.

Hwfioodm .m .0 .omoooscmao

 

R
condom Scoocou
coom3

mummsw

 

mdDme QZ< mm<03w LO mmumbom

H mam<H

32

"dextrose equivalent" or D.E. This is the copper reducing
power of the solids expressed as dextrose. The higher the
D.E. value the greater the hydrolysis of the sirup. The
63.7 D.E. sirupxmmsan acid-enzyme converted sirup, and those
with lower D.E. valuesuere straight acid converted sirups.
The term "unmixed" means that the flavorings which are present
in corn sirups for retail sales have not been added.

The sucrose was stored in a covered glass container.
The sorbitol was stored in a polyethylene bag. The other
samples were kept tightly covered in the containers in which
they were received. All samples were stored at room tempera-

ture.
Formulas

A 5% cornstarch slurry was used as the basic mixture.
This consisted of 5 grams of cornstarch per 95 grams of
distilled water. This ratio of starch to water remained
constant throughout the study. The weight of the starch
was calculated on the dry weight basis; the moisture content
was found to be 10.93% as determined by drying in a vacuum
oven at 120°C. to constant weight.

The amounts of sugars and sirups were also calculated
on the dry weight basis. The concentrations of dry sugar and
sirup solids were calculated as 5, 10, 20, 30 and 50% of the
weight of the liquid. A control paste containing no sweeten-

ing agent was also included in the study.

33

The weights of starch, sugar, and water were calculated
to yield approximately one liter of slurry. This gave a paste
level in the cooking beaker which was just below the height
of the scraper blade of the viscometer, thus allowing for the
most efficient stirring of the paste during cooking.

In order to minimize the effect of pH on viscosity and
gel strength, sodium hydroxide was added to the slurry to
adjust the pH of the cooked paste to approximately 6.60. It
would then be close to the pH of milk.

It was found in preliminary work that 3.5 ml. of 0.1 N
sodium hydroxide was needed in the control paste to obtain a
final pH of 6.60 after cooking. A proportional amount was
also added to the sweetened pastes.

The pH of the sugar solutions was adjusted to 6.60
just before using. The sodium hydroxide solution was in-
cluded in the weight of the water.

Table 11 lists the quantities of ingredients at each

concentration of sugar or sirup solids.

TABLE II
FORMULAS FOR STARCH MIXTURES

 

 

 

 

 

Sugar 0.1 N NaOH Included
Concentration Starch* Sugars Water in Water Weight
% of Water Wt. g. g. g. , m1.
0 50.0 -- 950.0 3.50
5 50.0 u7.5 950.0 3.50
10 50.0 95.0 950.0 3.50
20 k7.5 180.5 902.5 3.30
30 u5.0 256.5 855.0 3.15
50 no.0 380.0 760.0 2.80
s Dry weight basis

31;

Preparation

Ingredients

 

The sugars and sirups were weighed into 1000 ml.
beakers on a single beam platform balance. Five hundred
grams of distilled water was added to each dry sugar sample
to dissolve it. The mixture was then heated slightly to aid
solution. The sirups used were diluted with water to the
same concentration. The sirup and water was heated in order
to mix thoroughly. The total weightcfl‘the sugar,\vater,2nui
beaker was noted before heating so that an adjustment could
be made for any evaporation loss when the solution cooled.

All solutions were prepared the day they were used.
They were made sufficiently in advance to allow them to cool
to room temperature before use. By dissolving the dry sug-
ars, the effects of the rate of solution on viscosity of the
starch paste during cooking were eliminated. The dry sugars
were then on a comparable basis with the sirups.

Just before use, the pH of the cooled solutions was
determined with a Beckman pH meter, model H2. If needed,
adjustment of the pH to 6.60 was made by adding 0.1N sodium
hydroxide. The amount of alkali added was noted. After the
pH adjustment the solution was weighed and a correction made
for water which was lost during heating and cooling.

The cornstarch for each sample was weighed into a

tared aluminum dish using a triple beam balance.

35

The distilled water to be used for mixing was weighed
just before it was needed. This water included the sodium

hydroxide necessary to adjust the pH of the cornstarch paste
to 6.60.

Slurries

 

The weighed starch was put through a powder funnel
into a one-liter Florence flask. Approximately 150 ml. of
water was used to rinse out the aluminum dish and the powder
funnel. This water and starch was then swirled in the flask
about 50 times to insure complete mixing free from lumps.

For the control sample, all but 100 m1. of water was
added to the flask and the slurry mixed well before pouring
into the cooking beaker. The remaining 100 m1. of water was
used to rinse the flask.

For the sweetened pastes, the sugar solution was added
to the flask after the starch and 150 m1. of water were thor-
oughly mixed. Part of the remaining water was used to rinse
out the flask after the starch and sugar slurry was poured
into the cooking beaker.

All the ingredients were at room temperature, 28 t
2°C. Once the cornstarch was moistened the mixing of the

slurry proceeded quickly. The total mixing time never ex-
ceeded two minutes and the method was the same for all

samples.

36
Viscosity Tests

The Corn Industries Viscometer, described in the sec-
tion on equipment, was used for cooking and for obtaining a
pasting history of the slurries.

The cooking beaker, stirring device, and one-half the
condenser cover were placed in the viscometer during the
heating of the glycerol-water bath to 100°C. When the bath
was at 100°C. for a few minutes, the stirring device was
started. The slurry was then mixed and poured into the
cooking beaker. Immediately the recording device was turned
on, then the other half of the condenser cover and the ther-
mometer were put in place.

The paste was cooked for five minutes after the maxi-
mum viscosity was reached. In some instances with large
amounts of certain sugars, a decided peak or maximum vis-
cosity was never reached. In these cases the endpoint of
cooking was arbitrarily chosen after a considerable leveling
off of the viscosity rise.

The temperature of the paste was recorded at one min-

ute intervals during cooking.
Preparation of Gels

Gel strength was determined by an embedded disk method
using the Hjermstad apparatus, described in the section on

equipment.

37

The cooked sample of paste was removed from the vis-
cometer and poured immediately into four 250-ml. beakers.
The first sample poured contained any skin or lumps that
might have formed during cooking and was used to measure
the pH of the cooked paste.

The remaining three beakers, used for gel strength
determinations, were filled to a predetermined mark. Imme-
diately the covers were put in place, the disks immersed in
the hot paste to a level 3 cm. below the surface of the paste
and supported by the slotted cover. Ten ml. of mineral oil
was poured on the surface of the gel to minimize skin forma-
tion. The samples were then placed on a level refrigerator
shelf and aged for 18 hours at 3°C.

The low aging temperature was decided upon in order
to have conditions that would be comparable to the normal

handling of puddings containing cornstarch.

Measurement of Gel Strength

After 18 hours the samples were removed from the
refrigerator and held at room temperature for 15 minutes
before being subjected to the gel strength test. The samples
were removed singly so that each one stood just 15 minutes
before the test was started.

This holding period was decided upon when preliminary
tests showed that gel strength measurements varied appreci-

ably when started during the first 10 minutes after removal

38

:from.the refrigerator. .As the temperature of the gel in—
creased, the strength decreased markedly and then leveled
off after about 10 minutes.

Most workers studying gel properties have aged their
samples at room temperature or in 25°C. water baths and so
did not observe this phenomenon.

After the 15 minute period of holding, the slotted "1

cover was removed and the sample was clamped onto the stage

 

of the gel-lowering device. The cord from the dynamometer
was looped around the hook on the embedded disk. The stage
was adjusted until the cord was almost taut and the pen of
the recording device rested on zero. The gel-lowering de-
vice and the pen were started simultaneously and the test
was made automatically.

At the breaking point of the gel, the test was stopped
and the point noted on the chart. This point was used as
the measure of gel strength of the sample. In some samples
this point was at the end of a plateau where the deformation
curve had leveled off. In others there were higher points
on the curve and then a drop in the measured strength of the
gel before it broke. Hjermstad (32) states that this higher
strength just prior to breaking is believed to be due to
surface-skin effects at the oil-starch gel interface.

Since this phenomenon could not be explained as being

due to treatment, order of sample, or temperature of sample

39

or of room, it was decided to use the actual yield point of

the gel as a measure of its strength.

The average of three replications was reported as the

gel strength of a sample.

_. -V an“. .‘_...‘O F

RESULTS AND DISCUSSION
Viscosity

'The qualitative effects of the different sugars and
sirtQDS'were similar in their influence on the hot—paste
\Jisccmsity of the cornstarch pastes. Duplicate samples gave
very/sshnilar gelatinization curves. In order to show the
Itistory of gelatinization as affected by each of the sugars
and sirups at the various concentrations, curves were con-

structed for the average torque values of the duplicate runs
and are reproduced in Figure 1.

Lower concentrations of the sweetening agents tended

to increase the viscosity of the paste while increments
above 20% of the weight of the water used gave decreasing
viscosities. Time of gelatinization increased slowly with
additions of small amounts of most sweetening agents; some
sugars and sirups noticeably increased the time of gelatin-
izatHNIat the 30% level, and all of them markedly increased
the gehfldnization time at the 50% level.
Smw differences in gelatinization behavior of corn-
stanfltdue to the type of sweetening agent present can be

ruded atthe lower concentrations but the most obvious dif—

ferenunsare brought out by comparison of the gelatinization

Ito

 

 

 

bl

160r 30% Fructose

NO.”

203

 

 

O L—__4J _-L.. -_._L..._...__-.S,L.- ._.J.___.._l

10 2O 30

Time (minutes)

Torque
g.-cm.

160'? Su

11,0 L 40‘].

II
1 2 Q P ’ Io‘l.

. 572.

100 ~ I: 07.
I.

0

rose

301

l:

60 ~

' O
. ‘h.
‘ uh. _.

no r

20' /

 

O L_....._...A "has. 1...- _. .s.

O 10 2O 30

Time (minutes)

Figure I.
of cornstarch.

 

0%

 

 

30
Time (minutes)
Torque
g.-cm.
160 Invert Sirup
[ aoz
1u0

120'” [‘7' ”07° 50%
' r I

O I -__- J- ..-.......L_ -.- _--L.- - “I:

0 IO 20 30

Time (minutes)

Effect of sugars and sirups on the gelatinization

 

 

 

 

 

 

 

 

 

 

N2

 

 

 

 

 

 

'Torwque Torque
9--Cm- gucm.
I6C>t 160 r
1 Maltose f CSU 63.7 D.E.
- 4m- , 0;
mo u ;
12C)“ 3°” 120‘?
50% ,
100" 100: -w«5b%
80" 80?
so: I: so?
hot ‘j / uof
5 I
20 " l (I 20 :-
cull 212 . 0L, .. L .
O 10 2O 3O 0 30
Time (minutes) Time (minutes)
Torque Torque
g.-Cm. g._cm.
lbOr csu 56.3 D.E. 160V csu us D.E.
lkOt ao%. 1h0> (0%
apt
30% . L
120" ' 120 30%
5%
100~ 1007 0%
5‘01 -507.
80' 80L
No” hot
20 — 20 r
0 . ,d_ . . “m,~__j~_ 0 ‘1 . I
O 10 20 30 0 10 20 30

Time (minutes)

Fimne 1. Continued.

Time (minutes)

 

 

 

 

N3

 

 

 

 

 

 

Torque
g.-cm.
160 -
Sorbitol
1N0 * act
30%
120 ’- loyo
100 :- . 500/0
80 L
I
60 L
1+0 t
20 r
O L-.-- -44]. -_ -... .L ._ ..-- .L-...,.. I .._-._..J
0 10 20 30
Time (minutes)
Lactose
-—~—- :07.

 

 

J_ J |

O 10 2O 30
Time (minutes)

 

Figure 1. Continued

All

curves of the pastes containing 50% levels of the various
sugars or sirups.

Figure 2 compares the effect of 50% glucose, fructose,
sucrose and the inverted sirup with a control paste. The
paste containing sucrose at the 50% level showed behavior
markedly different from the pastes containing the monosac-
charides. The monosaccharides increased the time of gela-
tinization but allowed enough swelling of the cornstarch
granules so that the maximum viscosities were above the
control paste. Fructose and glucose possibly allow swelling

of more granules without subsequent rupturing, thereby adding

to the overall viscosity. Fifty per cent sucrose retarded
the initial rise in viscosity appreciably longer than the
monosaccharides at this level and inhibited swelling of the
starch to a marked degree. This was shown by the much more
gradual increase and the lower final paste viscosity which
was less than the control.

The invert sirup, which was 5N.92% inverted, gave
results which were probably dueto the ratio of glucose,
fructose, and sucrose present.

The pastes containing the monosaccharides showed some
decrease in viscosity after maximum suggesting the probabil-
ity of some granular rupture, while the sucrose-containing
paste did not show a decrease in viscosity. Preliminary ‘

tests showed that, even with cooking 30 minutes past maximum

W ha; a”: n-Ia-m

 

 

 

 

NS

Torque

90"ch

“+0.1 Fructose
Glucose

   
   
   

l20r Control ‘Invert Sirup

 

100?

f /'F‘
80' Sucrose
60!

NOT I

I r
207 (

01..ij . . t--__.L.--_..4_.______.I

0 IO 20 30 NO

Time (minutes)

Figure 2. Gelatinization curves for cornstarch
pastes containing 50% fructose, glucose,

sucrose, and invert sirup.

Torque
g.-cm.

lNOr Glucose

. , Control
120 Maltose

CSU 63.7 D.E.

CSU 56.3 D.E.
CSU N3 D.E.

100}

80}

so;

20 30 N0

Time (minutes)

C)
1

O 10

Gelatinization curves for cornstarch pastes

containing 50% glucose, maltose, and corn

Figure 3.
sirup hydrolyzed to 63.7, 56.3, and N3 D.E.

N6

viscosity, the pastes containing sucrose did not decrease
in viscosity. The starch granules probably had not swollen
enough to start to rupture.

Figure 3 shows the influence of 50% levels of glucose,
maltose, and the three corn sirups on the gelatinization be—
havior of the starch. A control paste was added for compar—
ison. At the 50% level, the glucose permitted enoughimwzlling
of the starch to give a maximum viscosity above that of the
control paste. There was some granular rupture as evidenced
by a decrease in viscosity after maximum. The viscosities
of the pastes containing 50% levels of maltose or any of the
corn sirups were lower than the control paste and showed no
decrease in viscosity after maximum. These results appear
to be evidence for decreased hydration of the starch granules
resulting in lower hot—paste viscosities and negligible
granular rupture.

A knowledge of the composition of the corn sirups,

as shown in Table III, is helpful in understanding their

behavior (55).

TABLE III
APPROXIMATE COMPOSITION OF THE CORN SIRUPS

 

 

 

 

Trioses
Corn Sirup Glucose Maltose and Above
% % %
63.7 D.E. 37 30 33
56.3 D.E. 32 18 50

us D.E. 20 15 65

 

“ ‘1’}
\-

147

The corn sirup with the highest D.E. value showed a
Inarkedly higher viscosity than the lower converted sirups.
'The higher proportion of glucose and maltose in this sirup
Lyrobably explains this higher viscosity.

For comparison of the effects of the various single
sugars, the gelatinization curves of the pastes containing
50% sugar are reproduced in Figure N. The monosaccharides
show an appreciably different effect on gelatinization be-
havior of the cornstarch than the disaccharides, lactose
and sucrose. Maltose and the sugar alcohol, sorbitol, show
intermediate effects.

The earlier time of gelatinization, the increase in
viscosity above the control, and the decrease in viscosity
after maximum are all phenomena that give evidence that the
monosaccharides allow enough free water in the mixture to
be available for maximum swelling of the starch granules.
At the other extreme, the lactose and sucrose appear to
bind sufficient water to inhibit complete swelling of the
starch.

Maltose apparently did not tie up as much of the water
as the other disaccharides as evidenced by the higher maximum
viscosity of the pastes in which it was used. However, it
did prevent complete gelatinization of the starch granules,
resulting in a maximum viscosity lower than the pastes con-
taining monosaccharides and also lower than the control

pastes. The lack of a decreased viscosity after maximum

Wh‘it 3"...
a

 

N8

 

 

 

 

Torque
g.-cm
1N0 I Fructose
Control Glucose
120 ~
Maltose
100 . SOFPiPEigLactose
ucrose
80 t
60 r
i
110 t
20 r
O Lm-L . t....._l. .-..__L.._. L? g I l
0 10 20 30 N0

Time (minutes)

Figure N. Gelatinization curves for cornstarch pastes
containing 50% fructose, glucose, sorbitol,
lactose, maltose, and sucrose.

 

”9 _

is indicative of insufficient swelling to cause rupture of
the starch granules.

Sorbitol showed a different effect on the gelatiniza-
tion of the starch than did the monosaccharides. Possibly
the six hydroxyl groups present in the open chain molecule
of sorbitol bind more water than the monosaccharides con-
taining five hydroxyl groups in a cyclic structure; thus
sorbitol makes less water available for swelling of the
granules. The decrease in viscosity after maximum was in-
consistent. In one run there was no decrease while the
duplicate run showed a slight decrease with five minutes
additional cooking.

The influence of sucrose on the behavior of cornstarch
during gelatinization is in agreement with results reported
by other workers (23, 26, 27, 30). It is commonly accepted
that sucrose in large amounts decreases the viscosity of a
starch paste byxTntgrfering with the hydration and subsequent
swelling of the starch granules. The above comparisons show
that the other sugars affect gelatinization in a similar
manner but to a different degree.

Table IV gives the maximum hot-paste viscosity of the
duplicate pastes for all the sugars at all concentrations.

The average maximum viscosity for the duplicate pastes
as affected by the concentration of sweetening agent is
shown in Figure 5. The sugars are separated from the sirups

for ease in reading.

 

 

 

50

TABLE IV

EFFECT OF SUGARS AND SIRUPS ON MAXIMUM HOT-PASTE VISCOSITY%

 

 

Sugar Concentration (% of Water Weight)

 

 

Sugars
0% 5% 10% 20% 30% 50%
Fructose 117 131 I38 1N8 1N6 132
119 135 lbs 152 1N7 135
Glucose 111 136 1N1 1N8 1N6 123
119 135 1N2 1N8 the 131
Sorbitol 113 132 135 1N1 131 102
116 130 139 1N1 135 111
Lactose 118 I32 I3N 137 136 99
117 130 I30 132 129 91
Maltose 112 122 130 .132 130 105
115 133 1N8 INC 136 113
Sucrose 129 I35 139 1N2 I33 95
119 128 13N 139 131 88**
Sirups
Invert Sirup 120 I33 139 1N6 1N3 I21
119 130 137 139 136 116
CSU 63.7 D.E. 118 138 137 139 I35 99
IIN 130 130 1N1 122 100*%
CSU 56.3 D.E. 115 123 129 I31 125 89
117 125 132 135 128 86%*
CSU N3 D.E. 121 129 I35 130 121 89
117 123 130 123 118 75%*

 

% g.-cm. of torque ** No maximum reached

“I

. I
‘ .1
__L_._ __ —.____-ku

51

 

 

Torque
g o - cm a
I *zt’r") O \
1 ' ‘\
1 3 0 \.
1.11.0 T u ‘,\ \ \
i \ “in \\
1 ,io.‘ -' “*‘—+ x \‘
I 0 I "__c ____ “~ \©
1 o , ....... -0
.‘ ’;/. ---‘O' "\
t /0 ;\\ 4.0..
I \ .“A
‘ ° \ \i‘...
1" h\‘ ‘t‘u
l ’1 . '1 \
202/ , \ -,
I B‘-\ l
u \ ‘.
————— Fructose y\\ \\1s
1' H-H-y GI‘UCOSQ '\\\ ‘8 ."-
ww- Sorbitol ‘k‘.
X \
*+*--Maltose \°\ ‘
100 -——--Lactose 5;. 1
W Sucrose 0‘s 1
b
.___,.___....._L_......._.'. 3-..”, .2 .J- - . _. 1.. _- -2, -..--,-, .._ _ 2-2... _ .J
O 5 10 20 3O 50

Per cent concentration of sugars (% of water weight)
Torque

 

 

 

 

g.-cm.
1'
: ”,3.“
1 O L. ”fur" x
)4. I D”'/’ _,,¢"’WAQ\ kx‘x
f ,. +1 1’ 4k Rx \M‘-
+- > ,a ° . 3‘s \2
1 ,0}, ‘Q‘N‘ 0‘ k K" x.
' 1 ’15?! “~-\ \I a ‘\\
120 ff!" ‘9. ° " .V \o
{d \\ 0 Pk
\‘ o ‘v
I “ 0 ~
i, \\ O *
I \\\ a T y
l \\ o N
1.00 I . s\ o O
Invert Sirup ~\ on
O—p-Q—w- CSU 63 . 7 D .E . \\\ o
4)- O-o—o—a» CSU 56.3 D.E. \\ 00
------ cs0 N3 D.E. \\n
80 r
1‘... Lu...» «4.. 2...- - .. .- a .. -.- . _...._ ......_..i.---- .4
O 5 10 ~ 20 3O 50

Per cent concentration of sirups (% of water weight)

Figure 5. Effect of per cent concentration of sugars and
sirups on maximum hot-paste viscosity.

Because of the differences in the viscosity effects
between the mono- and disaccharides, the molal concentrations
of these ingredients were calculated and plotted against the
.maximum viscosity values of the pastes. This graph, Figure
6, shows the sharp differences in behavior due to the sugars
when they are compared on the basis of molal concentrations.
Not only does the effect appear to be independent of the

molality of the sugar present, but there seems to be no

.WN-uua: SW
3. _ _

 

direct relationship between the number of sugar hydroxyl
groups present and the viscosity of the pastes. At approx-
imately corresponding molal concentrations (about 1.5 molal)
the disaccharides contributed more hydroxyl groups to their
respective pastes and the hot—paste viscosities were lower.
However, the 50% concentration of a monosaccharide contained
a larger total number of sugar hydroxyl groups than this
concentration of a disaccharide. The resulting paste had a
higher maximum viscosity. The monosaccharides were not as
effective in preventing swelling of the granules as were the
disaccharides.

The stereochemistry of the sugars may play a part In
the gelatinization behavior of the starch.

The analysis of variance for the maximum viscosities,
Table V, showed that there were highly significant differ-
ences between the averages of the maximum viscosities due

to the sweetening agents and to the concentrations used.

 

120'

100 ,

(D
O

O rundfi..a-u~..v.--. .

Molal concentration (moles of sugar per 1000 g. water)

Figure 6.

.........--,c...-..4. ._.....

0.5——

e -I n--‘.« gs. v

1.0

53

1:5

 

2.0

Fructose
Glucose
Sorbitol
Maltose
Lactose
Sucrose

\ \
\ ‘
4
. 7‘0
‘4L
1
‘0
\\‘
“o
3-1-.. 2....
2.5

Effect of moles of sugar on maximum hot-paste

viscosity.

“an.

 

 

511

TABLE V

ANALYSIS OF VARIANCE FOR MAXIMUM HOT-PASTE VISCOSITY

 

Sysurce of Variance D.F. M.S. F
Total 119

IBetween sweeteners 9 N75.N2 10.00**
.Between concentrations 5 3N80.2N 73.20**
Error 105 N7.5N

 

'** Significant at the 1% level of probability

The differences between these averages were tested by
the Multiple Range Test (22). The averages pertaining to
fructose, glucose, and invert sirup were significantly dif-
ferent from the averages for all the other sweetening agents
in their influence on the maximum viscosity. The averages
for sorbitol, maltose,and sucrose were significantly differ-
ent from those for the N3 D.E. and the 56.3 D.E. corn sirups.
The averages for the corn sirup hydrolyzed to 63.7 D.E. and
for lactose were significantly different from that for corn
sirup hydrolyzed to N3 D.E. The averages for the two lower
D.E. corn sirups did not vary significantly from each other.

The concentration of sweetening agent affected the
viscosity significantly, with the average maximum viscosity
at the 50% level differing from the averages of all the other
concentrations. The 20% level gave pastes having the highest
maximum viscosities in nearly all cases and the average at
this level was significantly different from all except the

10% level. The average for the 10% level was significantly

 

55

different from the 5%, the control, and the 50% levels.

The analysis of variance for each concentration of
sweetening agent showed that no significant differences
existed in the average maximum viscosities due to the variety
of sugar at the 5% and at the 10% levels. Highly significant
differences were observed at the 20, 30, and 50% levels.

These analyses are presented in the Appendix in Tables IX

to XIII.

 

Temperature

wait-7" “77.75: m
i. I
1

Increasing concentrations of the various sugars and
sirups caused progressively higher paste temperatures at the
initial rise in viscosity, at maximum viscosity, and at the
terminal viscosity. With the Corn Industries Viscometer,
temperature is a function of the time required to paste the
starch mixture. Pastes which were retarded in swelling by
the sugar present tended to reach higher temperatures before
reaching maximum viscosity. The temperatures at maximum vis-

cosity are listed in Table V1 for all the sweetening agents.
Gel Strength

The effects of the different sweetening agents on the
gel strength of the cornstarch pastes, aged 18 hours in a
refrigerator, were qualitatively the same. A few of the
sweeteners increased the gel strength slightly when added

as 5% of the water weight, but in general all the sugars

 

56

TABLE VI

PASTE TEMPERATURE AT MAXIMUM VISCOSITY%

 

_.,—i

Sugar Concentration (% of Water Weight)

 

 

Sugars
0% 5% 10% 20% 30% 50%
Fructose 90.2 90.2 90.3 91.0 91.3 93.5
90.5 90.2 90.5 91.0 91.8 9N.0
Glucose 90.2 90.3 91.0 91.5 92.N 95.2
90.5 90.8 91.0 91.8 92.8 95.7
Sorbitol 90.2 90.N 90.6 92.0 93.2 96.2
90.N 90.7 91.0 91.7 93.0 96.0
Lactose 90.0 91.0 92.0 92.N 9N.5 97.7
90.3 91.0 91.5 93.0 9N.3 97.8
Maltose 90.3 90.3 91.0 92.0 93.N 96.2
90.5 91.0 91.0 92.0 93.5 96.0
Sucrose_ 90.3 90.8 91.5 92.8 9N.9 97.7
90.5 91.0 91.8 93.0 95.0 97.5**
Sirups
Invert Sirup 90.3 91.0 91.0 92.2 93.3 96.0
90.3 90.8 91.0 92.0 93.1 96.0
CSU 63.7 D.E. 90.2 90.8 91.N 92.3 9N.0 97.0
90.3 91.0 91.8 92.7 9N.3 97.0%s
csu 56.3 D.E. 90.5 90.8 91.7 93.2 9u.5 97.0
90.5 91.0 91.0 93.2 9N.3 97.0%fl
050 N3 D.E. 90.3 90.N 91.0 92.5 93.7 9u.0
90.3 91.2 92.0 92.5 93.7 96.5%%

 

% °C. ** No maximum reached

57

and\sirups progressively decreased the gel strength. At

the 50% level, the gels containing glucose and fructose had
very tender gel structure which broke easily, while the other
pastes exhibited no gel structure. Instead thick sirupy
liquids resulted. Table VII gives the gel strength of each
of the samples. Each value is an average of triplicate tests.

At the 50% concentrations, in all samples except those con- 8“.

taining monosaccharides, the torque values are actually a

measure of the consistency of the resulting sirupy liquids ‘ 9

 

rather than of gel strength.

Figure 7 shows the effect of concentration of fruc-
tose, glucose, sucrose, and invert sirup on the gel strength.
Fructose-containing gels were stronger than the others at
all levels. Glucose gave more tender gels than fructose
but observations on samples during testing showed that gel
structure was actually present in all samples. Pastes con-
taining sucrose and invert sirup gave weaker gels than the
others at corresponding sugar levels, and at the 50% level
all signs of gel structure disappeared.

Results appear to indicate a relationship between the
gelatinization history and strength of the resulting gels.
Meyer and coworkers (38) hypothesized that the formation of
starch gels involved the association of the free ends of
amylose chains partially diffused from the swollen starch
granules with one end remaining fixed in the crystallite

regions of the granule or granular segment. Thus bonds

TABLE VII

EFFECT OF SUGARS AND SIRUPS ON GEL STRENGTH*

 

 

Sugar Concentration (% of Water Weight)

 

 

Sugars
0% 5% 10% 20% 30% 50%
Fructose 1N9 166 163 1N7 96 NN
1N9 152 152 132 95 39
Glucose 1N9 1N9 1N7 101 68 2N
1N2 150 1N2 106 82 27
Sorbitol 150 162 138 117 76 26
1N3 159 1NN 101 71 25
Lactose 16N 16N 137 92 63 I7
151 150 129 9N 57 16
Maltose l5N INN 130 95 6N I8
1N1 1N2 137 92 68 21
Sucrose 157 1N9 130 98 60 18
1N5 128 123 82 N6 12
Sirups
Invert Sirup 152 159 1N6 109 73 2N
138 INN 139 101 67 26
CSU 63.7 D.E. 139 1N8 129 85 50 16
1N0 1N1 123 91 53 15
CSU 56.3 D.E. I51 135 111 77 53 IN
150 13N 109 80 N3 11
csu N3 D.E. 1N7 138 12N 91 55 11

1NN 132 11N 80 52 15

 

+ g.—cm. of torque

Torque
g.-cm.
160f
1 , «9-
/ ,Q

120r
100"
80'
60—
no?
203

I
1

i
0Lgm_gj
5

59

/xr~”‘ ————— Fructose
. -. “Ms-+5 G 1 uco se
\

0 ‘\ ----- Invert Sirup

..
‘.‘ . \ ----+ Sucrose
Q l

\ \
s \ ‘ .x
I ‘ S
\ o
9 .
. \ ‘
9 ‘ ‘
x‘ *
\
5
o \ \
\* \
\ \,
x
5 \O‘
. a
Q .
Q \
\\‘ \‘
o \ ‘ \
\
\ t R
a \ ,,
Q \
\ I
\
I \ ‘
\ s ,
n \ ,
\
. . o
\ 7
\ I
\
o \ .
e
. h .
x
I \\ I
O t ’
\ I
\
° \ s
\
o \ o
\
o \ I
\ s
\ 5
. \
\ o
\ o
o \
\ . O
\
o \ §
\
\ 0
0 \ +
\
\
° \
\
O
a
o
0
_l _ . . 1 _. . I . _, ._ -,.__.L_.-_--__ ._._.___1

Per cent concentration of sugars (% of water weight)

Figure 7.

Effect of per cent concentration of fructose,
glucose, sucrose, and invert sirup on gel strength.

‘4‘"?
,1
I.

4-!-

‘6‘

Wmofl.‘
__ {I

 

so

would be formed between the swollen granules, giving rise
to a gel structure. A considerable degree of swelling of
the granules would be necessary before the amylose would be
sufficiently loosened from the micellar structure in the
granule for such gel formation to be possible. If the gel—
atinization history indicates such swelling with rupturing
of some of the granules, as evidenced by a decrease in vis-
cosity, then the pastes would be expected to set a gel.

Figure 1 showed that the pastes containing fructose
and glucose, the sugars giving the gels of highest strength,
also gave higher maximum viscosities than the other sugars
and sirups at all levels, except the 5% level. They also
exhibited more breakdown in viscosity than the other pastes.
These results then indicate that those pastes in which the
granules are more highly swollen, or even partially ruptured,
give gels of higher strength.

Figure 8 shows the effect of concentration of glucose,
maltose, and the three corn sirups on the gel strength. Dif-
ferences in the influence of these sweetening agents are sig-
nificant in only a few cases as noted later in the analysis
of variance. However, again the strength of the resulting
gels can be compared with the gelatinization characteristics
of the pastes, Figure l. The pastes containing glucose at-
tained a higher maximum viscosity than the others, and some-
what more breakdown in viscosity following the maximum. The

highly swollen granules with some ruptured ones acted to

ol

 

 

 

Torque
g.-cm
160
Glucose
' Maltose
3%://XL\\\ -*~F+i CSU 63.7 D. E.
‘ "Q m CSU 56.3 D. E.
Y5 \ ------ csu us D. E.
l ()9!\
u ! §3_ ‘\
‘3 a ‘0 ‘
X ‘ .
v\ \ .
v\‘ R
120 "“ °
7 \\‘~ ‘
Q“ \\ ‘ . '\\
v x n q
100 - i ‘\,- x
v \ 0 V‘.
” ‘faq
9 ‘ ‘(Q .
v qhx“ .
80 9 'o\: ' ‘x
- V X: o a
‘7 ‘t ‘ i.
v \ O
.7\ a
. k -
60 - \9 g
‘7 ‘\
Q I
“5&5 .
. x,“
Q‘ ‘9‘ .
“\ Q
to ~ < q \
Q‘\ '
{'1 \Q .
‘ \\~ .
q\\§ . b
qxo .
20 f ‘3 ;:\‘®
; w\\g
i b
O ‘ ...-...L....-..._..-1-,..--......_----.......l- ._ -_.l .,,..- __,-..__.,..,mn-.._._.___1
5 10 20 3O 50

Per cent concentration of sugars (% of water weight)

Figure 8. Effect of per cent concentration of glucose,
maltose, and three corn sirups on gel strength.

62

give some structure to the gel, even at the 50% level. The
other sweeteners in this set produced no gels at the 50%
level and weaker gels by comparison at the other levels.

A comparison of the effect of the moles of sugar on
the gel strength is made in Figure 9. This graph shows sharp
differences in the gel strength as influenced by mono— and
disaccharides. In general all of the disaccharides studied
gave the same effects. Glucose and sorbitol were very sim-
ilar, while fructose interfered the least with gel formation.
Except for the position of the glucose curve, these results
are analagous to those obtained in Figure 6 where the maximum
hot-paste viscosity is shown as affected by the molal concen—
tration of the sugars.

The degree to which the various sugars interfere with
gel formation is significantly different. The mechanism of
gel formation is not well enough understood to explain why
the sugars showed different effects on the gel characteris-
tics. Pasting history appears to offer some basis for an
explanation but the reasons for the different effects of the
various sugars on paste history is not understood. The de-
gree to which the various sugars interfere with hydrogen
bonding of the amylose chains must also be considered as a
possible factor affecting gel formation.

The analysis of variance for the gel strength data,
Table Vlll, showed that the differences due to sweetening

agents and to concentrations used were highly significant.

as

 

Torque
g o "Cm o
160.. ,Q ——--—— Fructose
; °££FT"4l\ ++++4 Glucose
l J \ ‘ ------ Sorbitol
.3 '@\«;k \\‘ —————- Lactose
g. ‘Q \\ ++-ro Maltose
T o ‘ ‘ *4F*** Sucrose
lLLO .— 0 E? \w
. \$_ .
a? \‘K \
\§ \
but. \ \
. ‘, 5 .
120 " '~.'- “x \
9 “X \
v; 3 ‘
,‘ '. V.\\ \
9 -. s .
" § *Q. \
it \Q}
100 _ Q 1“ \\ x
.‘ .‘\ \b
u Q ‘ \ \
2 v\ x
O |\
. VA
-~\_ \
\ x .
. - x. \
80- s x ‘
. 5‘ 29 \
\ is
$9 ‘1 \
\ v ‘
°° \ \
60 ' U 50 ‘q \
t’ \‘ '9 \
O \ \
~ 4 \
a \Y ‘
o \ C? \9
LLO ’ O \\3
. x?
\v
0 \"v
° ED
0.
20* ‘9
G)
O L... r.-..._..._._ .. ._ --—.L....—-—~—..~..--...._.._ -1... - __. -.- _-__..__........--l-........- -._ _ __...___....J_...-~ ... L J
O 0.5 1.0 1.5 2.0 2.5

Molal concentration (moles of sugar per 1000 9. water)

Figure 9. Effect of moles of sugar on gel strength.

61:

TABLE VIII

ANALYSIS OF VARIANCE FOR GEL STRENGTH

 

 

 

Source of Variance D.F. M.S. F
Total 119

Between sweeteners 9 l2él.hO 19.SO%%
Between concentrations 5 52h30.79 810.73**
Error 105 au.s7

 

%* Significant at the 1% level of probability

When tested by the Multiple Range Test (22), the aver-
age gel strength of the fructose gels was found to be signif-
icantly different from all the other averages. The average
pertaining to the sorbitol gels was significantly different
from those of maltose, sucrose, and the corn sirups. Those
averages for glucose, invert sirup, and lactose were signif-
icantly different from those for sucrose and the corn sirups,
while the maltose average differed significantly from those
for the two lower D.E. corn sirups. The averages for sucrose
and the corn sirups did not differ significantly from each
other.

The concentration of sugars and sirups caused highly
significant differences in gel strength. The average gel
strength for each concentration above 5% was significantly
different from the average for every other concentration.

No significant differences existed between the average
strength of the control gels and that of the gels containing

5% sweetening agent.

 

65

The analysis of variance for each concentration of

sweetening agent showed highly significant differences

the averages of the gel strengths due to the variety of

in

sugar or sirup at the 10, 20, 30, and 50% concentrations.

The 5% level showed significant differences due to the
sweetening agent present. These analyses are presented

the Appendix in Tables XIV to XVIII.

in

I-' . I" .’ - .‘-c"‘. l’- .. :"
lawman";

“in“
'9

SUMMARY

The influence of ten different sugars and sirups on
the viscosity and gel strength of 5% cornstarch pastes was
studied. Fructose, glucose, sorbitol, lactose, maltose,
sucrose, invert sirup, and three corn sirups (63.7, 56.3,
and M3 D.E.) were added in such amounts that the dry sub-
stance of the sweetening agent was present as 5, IO, 20, 30,
and 50% of the total weight of the water.

With all the sweetening agents, the maximum hot—paste
viscosity increased as the sugar or sirup solids was increased
up to 10% or 20%. Above these concentrations, the viscosity
decreased. At the higher concentrations, differences in
viscosity due to the type of sweetening agent present were
obvious. The monosaccharide-containing pastes had higher
maximum viscosities than the pastes containing disaccharides
or sirups. At the 30% and 50% levels, the sugars and sirups
caused obvious increases in the temperature and time of
gelatinization.

At the 5% level, the monosaccharides and some mono-
saccharide-containing sirups caused an increase in gel
strength. Further increases of all of the sugars and sirups
caused a decrease in gel strength. These decreases were

more marked with the disaccharides and sirups than with the

66

Ii III... I I- ‘III II
II' 1"! lizlltl I 3 I I '.
a ' I 'II' I III .. I I III it‘ll

e7

monosaccharides. At the 50% concentration of sweetening

agent, gel structure was present only in the fructose and
glucose samples. Thick, sirupy liquids were formed from

the other pastes at this sugar level.

CONCLUSIONS

Small but significant differences exist in cornstarch
pastes and gels when various sugars and sirups are present
in amounts greater than 20% of the weight of the water. The
common carbohydrate sweetening agents used in food prepara-
tion produce, in general, the same qualitative effect on the
gelatinization and gel—forming properties of the starch.
Although the direction of influence is generally the same,
the results of this study indicate that the direct substi~
tution of one carbohydrate-type sweetening agent for another
in starch-containing food products should be undertaken with
caution. Aside from affecting flavor, these sweetening
agents produce pastes of different viscosities and gels of
different strengths when present in the same concentration
in a starch and water mixture. These effects might be
obscured by the presence of other ingredients in a food
mixture, but such influences remain to be investigated.

With all of the sugars and sirups studied, there was
a slight increase in maximum hot-paste viscosity up to the
10% or 20% level of concentration. Further increases in
concentration of sweetening agent produced a decrease in
viscosity. The pastes containing monosaccharides had higher

viscosities than those containing disaccharides or sirups

68

69

at comparable concentrations except the 5% level. Such
results indicate that the number of sugar units present in
the molecule apparently affects the gelatinization behavior
of the starch granules.

There seemed to be no direct relationship between the
viscosity of the pastes and the number of sugar hydroxyl
groups present. The slight differences between the mono-
saccharides and among the disaccharides leads to the con—
clusion that the stereochemistry of the various molecules
may play a part in their influence on starch behavior.

The differences in gelatinization behavior seem to
show their effects in the resulting gels of the various
pastes. The ability of the starch granules to swell suffi—
ciently during pasting so that some amylose diffuses out
appears to be necessary for the development of a gel when
the paste is cooled. The smaller the interference of a
sugar with granular swelling, the smaller is the effect of
that sugar on the strength of the resulting gel.

While monosaccharides and some monosaccharide-con-
taining sirups appeared to cause a slight increase in gel
strength at the 5% level, subsequent increases of the
sweetening agents caused progressive decreases in gel
strength. The other sugars and sirups lowered the gel
strength at concentrations of 10% and above.

The composition of the various sirups used seemed

to explain their effects. The presence of large amounts

 

70

of fructose and glucose in the invert sirup caused its in-
fluence to be closer to that of the monosaccharides than to
that of sucrose. With the corn sirups, more comparable be-
havior to glucose and maltose was noted as the degree of
hydrolysis of the sirup and, hence, the concentration of
these sugars increased.

Further investigation is needed on the relationship
between gelatinization behavior and gel formation and also
on the reasons why the various sugars affect these properties
of starch to different degrees. The interrelation of these
effects with those of noncarbohydrate food ingredients also

requires examination.

APPENDIX

APPENDIX

TABLE IX

ANALYSIS OF VARIANCE FOR MAXIMUM HOT-PASTE VISCOSITY
AT THE 5% CONCENTRATION OF SWEETENING AGENT

.—.__..—..__‘_——.————__

 

 

 

 

 

Source of Variance D.F. M.S. F
Total 19
Between sweeteners 9 .26 l.o9
Error 10 .15

TABLE X

ANALYSIS OF VARIANCE FOR MAXIMUM HOT-PASTE VISCOSITY
AT THE 10% CONCENTRATION OF SWEETENING AGENT

 

 

 

 

 

Source of Variance D.F. M.S. F
Total 19
Between sweeteners 9 30.91 l.l9
Error 10 25.90

TABLE XI

ANALYSIS OF VARIANCE FOR MAXIMUM HOT-PASTE VISCOSITY
AT THE 20% CONCENTRATION OF SWEETENING AGENT

————-——-—

————-—-————.—-—.——- -——.‘

 

Source of Variance D.F. M.S. F
Total 19

Between sweeteners 9 98.35 8.u7**
Error 10 11.60

 

w+ Significant at the 1% level of probability

72

73

TABLE XII

ANALYSIS OF VARIANCE FOR MAXIMUM HOT-PASTE VISCOSITY
AT THE 30% CONCENTRATION OF SWEETENING AGENT

 

 

Source of Variance D.F. M.S. F
Total I9

Between sweeteners 9 lu6.68 8.h7**
Error 10 17.30

 

+4 Significant at the 1% level of probability

TABLE XIII

ANALYSIS OF VARIANCE FOR MAXIMUM HOT-PASTE VISCOSITY
AT THE 50% CONCENTRATION OF SWEETENING AGENT

 

 

Source of Variance D.F. M.S. F
Total 19

Between sweeteners 9 587.66 20.91%*
Error IO 28.l0

 

** Significant at the 1% level of probability

RI

. -
.‘Ial
I‘d-n-

._.'_N?._ .

 

VI!!!

 

m

TABLE XIV

ANALYSIS or VARIANCE FOR GEL STRENGTH AT THE
5% CONCENTRATION or SWEETENING AGENT

._.-. —-—.-.—_—.-——__._._ -—.‘---—.~._—

 

 

w c“... -5.

 

 

 

 

 

 

 

 

 

 

 

'Source of Variance D.F. M.S. F
Total 19
Between sweeteners 9 188.13 3.25%
Error 10 57.90
% Significant at the 5% level of probability
TABLE XV
ANALYSIS OF VARIANCE FOR GEL STRENGTH AT THE

10% CONCENTRATION OF SWEETENING AGENT
Source of Variance D.F. M.S. F
Total 19
Between sweeteners 9 378.22 lu.l9**
Error 10 26.65
w+ Significant at the 1% level of probability

TABLE XVI
ANALYSIS OF VARIANCE FOR GEL STRENGTH AT THE

20% CONCENTRATION OF SWEETENING AGENT
Source of Variance D.F. M.S. F
Total 19
Between sweeteners 9 592.21 11.77**
Error 10 50.30

 

xx Significant at the 1% level of probability

 

TABLE XVII

ANALYSIS OF VARIANCE FOR GEL STRENGTH AT THE
30% CONCENTRATION OF SWEETENING AGENT

*—

— q...-—_—.

 

 

 

Source of Variance D.F. M.S. F
Total 19

Between sweeteners 9 h22.08 13.52xx
Error 10 31.20

 

xx Significant at the 1% level of probability

TABLE XVIII

ANALYSIS OF VARIANCE FOR GEL STRENGTH AT THE
50% CONCENTRATION OF SWEETENING AGENT

4—— M“‘~-

 

»_—..

 

 

 

Source of Variance D.F. M.S. F
Total 19

Between sweeteners 9 155.93 28.09xx
Error 10 5.55

 

xx Significant at the 1% level of probability

.- ,_.—..——3— ‘7???"

10.

ll.

LITERATURE CITED

Alsberg, C. L. and Rask, C. S. On the gelatinization
by heat of wheat and maize starch. Cereal Chem.
1: 107-116 (192M)

Alsberg, Carl L. Studies upon starch. Ind. Eng. Chem.
18: 190-193 (1926) E

Alsberg, Carl L. Structure of the starch granule. II
Plant Physiol. 13: 295-330 (1938) ;'

Anker, C. A. and Geddes, W. F. Gelatinization studies .1
upon wheat and other starches with the amylograph. I
Cereal Chem. 21: 335—360 (l9hu)

 

Badenhuizen, N. P. Observations on the distribution of
the linear fraction in starch granules. Cereal
Chem. 32: 287-295 (1955)

Bates, F. L., French, D., and Pundle, R. E. Amylose
and amylopectin content of starches. J. Am. Chem.

Soc. 65: Iu2-lu8 (isus)

Bechtel, W. G. Instruction manual for the Corn Indus-
tries Viscometer. Progress Report No. 12 of Corn
Industries fellowship. (19h?) '

Bechtel, W. G. A study of some paste characteristics
of starches with the Corn Industries Viscometer.
Cereal Chem. 2h: 200-2lh (19b?)

Bechtel, W. G. and Fischer, E. K. The measurement of
starch paste viscosity. J. Colloid Sci. h: 265-
282 (19u9)

Bechtel, W. G. Measurement of properties of cornstarch
gels. J. Colloid Sci. 5: 260-270 (1950)

Binns, Gladys M. The effect of varying proportions of
starch, egg and sugar on the quality of butter-
scotch pie fillings with emphasis on large quantity
preparation. Unpublished M.S. thesis. Cornell
University. (1951)

76

12.

13.

1h.

I5.

l6.

l7.

l8.

19.

20.

21.

23.

77

Bowersox, E. M. Some measurements of physical proper-
ties of starch thickened mixtures and a comparison
of scientific and culinary terms. Unpublished
M.S. thesis. State University of Iowa. (I9h7)

Brimhall, B. and Hixon, R. M. The rigidity of starch
pastes. Ind. Eng. Chem., Anal. Ed. 11: 358-361
(1939)

Brimhall, B. and Hixon, R. M. Interpretation of viscos-
ity measurements on starch pastes. Cereal Chem.

19: h25-uu1 (19u2)

Buchanan, Ben F. and Lloyd, Robert L. Gelatinization
of starch. (To American Maize-Products Co.).
U.S. Patent 2,h06,585 (August 27, I9h6)
U.S. Patent Office, Official Gazette. 589: 605
(19kb)

Caesar, G. V. Consistency changes in starch pastes.
Ind. Eng. Chem. 2h: 1u32-1u35 (1932)

Caesar, G. V. and Moore, E. E. Consistency changes in
starch pastes. Ind. Eng. Chem. 27: Ihh7-lh5l
(I935)

Caesar, G. V. and Cushing, M. L. The starch molecule.
J. Phys. Chem. h5: 776-790 (l9hl)

Caesar, G. V. Chap. IX. The hydrogen bond in starch
as a basis for interpreting its behavior and re-
activity. Kerr, Ralph W., Editor. Chemistry and
Industry of Starch. Academic Press Inc., New

York. (179%)

 

 

Chase, Jean T. A study of the consistency of starch
gels prepared at a higher altitude. Unpublished
M.S. thesis. Montana State College. (19h2)

Chidester, Sister Mary Dominicus. The effect of added
food substances on the gels of five cereal starches.
Unpublished M.S. thesis. State University of Iowa.
<19u5)

Duncan, David B. Multiple range and multiple F tests.
Biometrics. 11: l-h2 (I955)

Ferree, Marie, J. The effect of the proportion of sucrose
on the viscosity of cornstarch pastes made with
different liquid mediums. Unpublished M.S. thesis.
Michigan State University. (1955)

2h.

26.

27.

28.

29.

31.

32.

33.

31;.

35.

French, Dexter.

78

Chap. VII. Physical properties of
starch. Kerr, Ralph W., Editor. Chemistry and
Industry of Starch. Academic

 

(19507

Glabau, Charles

ings and meringues.

(l9h8)

Hains, Ann J.

cornstarch pastes.
Michigan 8

Halliburton, Ge

flour and

 

Press Inc., New York.

A. Stabilizing custard type pie fill-

Bakers' Weekly. 139: 56-58

The effect of sucrose, citric acid and
temperature on the viscosity and gel strength of

Unpublished M.S. thesis.

tate College. (1953)

orgia Rae. The effect of sucrose on
some thickening and gel properties of a soft wheat

the flour fractions.
thesis. Pennsylvania State

Unpublished M.S.

University. (1955)

Hamer, W. J. An improved method for measurement of gel

strength and data on starch

gels. J. Research

Natl Bureau Standards. 39: 29-37 (19h?)

Harris, R. H. and Jesperson, E. A

of various

(l9u6)

Hester, E. Eliz

by sucrose

study of the effect

factors on the swelling of certain
cereal starches. J. Colloid Sci. 1: h79-h93

abeth. Some properties of selected edible
starches as thickening agents

. Unpublished Ph.D.

University. (1952)

and changes induced
thesis. Cornell

Hester, E. Elizabeth, Briant, Alice M. and Personius,
J. The effect of sucrose on the proper-
me starches and flours. Cereal Chem.

Catherine
ties of so
33: 91-101

Hjermstad, E. T.

(1956)

32: 200-207 (1955)

Hughes, M. F.

Katz, J. R., De

changes in microscopical

granules 0

1258-1268

Kerr, Ralph W.
Edition.
New York.

A recording gel tester. Cereal Chem.

Private communication. (1956)

sar, M. C. and Seiberlich, J. The change
in the viscosity of starch pastes explained by the

f the paste. Trans.
(1938)

structure of the starch

Faraday Soc. 3h:

 

Chemistry and Industgy of Starch. 2nd

 

Chap. v1 and VIII.
(1950)

Academic Press Inc.,

37.

38.

39.

h0.

111.

h2.

h3.

nu.

AS.

he.

A7.

Kesler, C. C. and Bechtel, W. G. Recording viscometer
for starches. Anal. Chem. 19: 16-21 (1987)

Meyer, K. H. Recent developments in starch chemistry.
Kramer, E. O.,Editor. Advances in Colloid Science.
Vol I. pp. 1h3-l82. Interscience Publishers, New
York. (19A2)

 

Meyer, K. H., Bernfeld, P., Boissonnas, R. A.. Gurtler,
P. and Noelting, G. Starch solutions and their
molecular interpretation. J. Phys. and Colloid ,
Chem. 53: 319—33h (19h9) 3

Meyer, K. H. Natural and Synthetic High Polymers. 2nd
Edition. ppTfih56—h89. Interscience Publishing
Co., New York. (1950)

f‘fl

 

q'w' "

Morgan, W. L. Pasting and identification of starches.
Ind. Eng. Chem.,Anal. Ed. 12: 313-317 (l9h0)

Nevenzal, Gladys M. A study of the effect of varying
concentrations of sugar on starch gels. Unpub-
lished M.A. thesis. University of California.
(1931)

Nikuni, Jiro, Juwa, H. and Tatsumi, C. The effect of
sucrose on the retrogradation of alpha starch.
J. Agr. Chem. Soc. Japan. 23: 90-91 (l9h9)
C. A. as: 5129 (1950)

Osman, Elizabeth M. and Mootse, Gerda. Private commun-
ication. (1956)

Potter, A. L. and Hassid, C. Starch 1. End-group
determination of amylose and amylopectin by
periodate oxidation. J. Am. Chem. Soc. 70: 3h88-
su9o (19u8)

Potter, A. L. and Hassid, C. Starch 11. Molecular
weights of amylose and amylopectins from starches
of various plant origins. J. Am. Chem. Soc. 70:

377h-3777 (19u8)

Saare, O. and Martens, P. Ueber die Bestimmung der
Ausgiebigkeit der Stérke. Z. Spiritusind. 26:
use-837 (1903)

Sair, L. and Fetzer, W. R. Water sorption by starches.
Ind. Eng. Chem. 36: 205-208 (19AM)

88.

A9.

50.

51.

52.

53.

59.

60.

80

Schoch, T. J. Physical aspects of starch behavior.
Cereal Chem. 18: 121-128 (1981)

Schoch, T. J. Fractionation of starch by selective
precipitation with butanol. J. Am. Chem. Soc.
68: 2957-2961 <19u2)

Schoch, T. J. The fractionation of starch. Pigman,
W. W. and Wolfrom, M. L., Editors. Advances in

Carbohydrate Chemistry. Vol I. pp. 2H7-277.
Academic Press Inc., New York. (19h5)

Schoch, T. J. A decade of starch research. The Bakers
Dig. 21: 17-20, 23 (19h?) .

Schoch, T. J. Naming starch fractions. In "Words about
words". Chem. Eng. News. 30: 320A (1952)

Schoch, T. J. Chap. 6. The starch fractions. Radley,
J. 8., Editor. Starch and ts Derivatives. John
Wiley and Sons, Inc., New York. FF195h)

 

Sjostrom, O. A. Microscopy of starches and their mod-
ifications. Ind. Eng. Chem. 28: 63-7h (1936)

Snyder, E. C. Private communication. (1956)

Taylor, T. C. and Keresztesy, J. C. Granule disintegra-
tion of cornstarch. Ind. Eng. Chem. 28: 502-505
(1936)

Trempel. Larry. The use of starch in pie filling.
Bakers Dig. 20: 23-25, 28 (1986)

Whittenberger, R. T. and Nutting, G. C. Potato starch
gels. Ind. Eng. Chem. hO: 1h07 (1988)

Woodruff, Sybil and MacMasters, Majel M. Gelatinization
and retrogradation changes in corn and wheat
starches shown by photomicrographs. University
of Illinois Agr. Exp. Sta. Bull. hh5 (I938)

Woodruff, S. and Nicoli, Laura. Starch gels. Cereal
Chem. 8: 2h3-251 (1931) . ‘

\

\..

119311 USE III-IN

{72(5193‘2? ' “" 9.? U
Apr 4 '57

. J

  

U ‘

Demco—293

1W—

"‘11111111111111