. _.,“, ‘_ .. ‘

370mg mam 0F #31? EWflATE
MOEST‘JRE Few

Thesis is: fine Degree of M. 8.
3mm SEMTE UMNERSSTY
MAMA CRESMMA ZULUETA B.

' 1973

ABSTRACT
STORAGE STABILITY OF AN INTERMEDIATE MOISTURE FOOD
BY

Maria Cristina Zulueta B.

Intermediate moisture foods due to their shelf stability
without refrigeration or thermal processing seem to have a
great potential in the future development of food items for
human consumption. This experiment was set up to study the
rate of some chemical reactions in an intermediate moisture
food model system.

Modifying the sugar:glycerol ratio, three water activi-
ty levels in the model system were-obtained. All samples
were stored at 37°t2°C in tightly closed glass jars. Their
water activity was determined by the electric hygrometer
and the Markower and Myers (1943) manometer; their moisture
content, by the vacuo oven and Karl Fischer titration.
Weekly assays for reduced L—ascorbic acid by the 2,6 di-
chlorophenol indophenol visual titration method, total
ascorbic acid by the adaptation of Roe (1936, 1943) method,
lipid oxidation measurement by the 2, thiobarbituric acid
(TBA) method, lysine availability by the trinitro benzene
sulfonic acid (TNBS) method and browning reaction measure-
ments by the Agtron-SOO and the Hunter-Lab Difference Color
Meter were performed.

An increase in lipid autoxidation and degradation of
ascorbic acid occurred during the browning of the model

system. These reactions adversely affected the appearance

and nutritional value of the intermediate moisture food with

storage time at 37°12°C.

STORAGE STABILITY OF AN INTERMEDIATE MOISTURE EOOD

BY
Maria Cristina Zulueta B.

A masts

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Food Science and Human Nutrition

1973

H ‘7 TABLE OF CONTENTS

u“,}
an
r H'

{j} page

H

INTRODUCTIONOCOOOOOOOOOO00......0.0.0.0000...00.......0.

LITERATIJRE REVIEWOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Intermediate Moisture Foods and Water Activity........
Non-enzymatic Browning 0f FOOdSeeeeeeeeeeeeeeeeeeeeeee
L-ascorbi:c ACidOo0.0.IOOOOOOOOOOOOOOOOOOOOOOO000......
Available LYSine.OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
RanCidity (Off-flavorS)OOOOOOOOOOOOOOO0.00.00.00.00... 13

g..-
ommww

EXPERIMENTAL METHODSeeeeeeeeeeeoeeeeeeeeeeeeeeeeeeeeeeee 17
Preparation Of the MOdel SYStemeeeeeeeeeeeeeeeeeeeeeee 17
Water Activity and Moisture Content Determination..... 17
TBA Value Determinationeoeeeeeeeeeeeeeeeeeeeeeeeeeeeee 19
Browning Measurements................................. 20
Available LYSine DeterminationSeeeeeeeeeeeeeeeeeeeeeeer20
Reduced L-ascorbic Acid Determinations................ 20
TOtal Ascorbic ACid DeterminationSeeeeeeeeoeeeeeeeeeee 20

RESIHDTSOOO...OOOOOOOOOOOCOOOO00.000.00.000.0000000000000 21

Water Activity Determinations......................... 22
MOiSture contentOOOOOOOOOCOOOOOOOOOO...OOOOOOOOOOOOOOO 22
Browning ReaCtionOOOOOOOOOOOOOOOOOOOOOOOO0.0.0.0.0.... 26
Reduced L-ascorbic ACidOOOOOOOOOOOOOOOOOOOOOOO0.00.... 30
Available Lysj-neOOOOOOOOOOIOOOOOOOO0..OOOOOOOOOOOOOOOO 37
TBA ReaCtionOOOOOOOOOOOOOOO...OOOOOIOOOOOOOOOO0.00.... 37

DISCUSSIONOOOOOIOOODOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO... 43
CONCLUSIONSOOIOOOCOOOOOO00.00....OOOQOOOOOOOCOOOOOOOOOO. 49
SUGGESTIONSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 5]-

BIBLIOGMPHY...0.0.0.0000...0.0.0.0000...OOOOOOOOOOOOOOO 52

ii

LIST OF TABLES

Table page
1. MOdel systems ReCipeSOOOOOOOOOOOOIOOOOOD000...... 18

2. Calculated and Experimental Water Activity of IMF
MOdel systemSOOOOOOOOOOOOOOOO0.0.0.....0...0...... 23

'-3. Calculated and Experimental Moisture Content on
IME‘ MOdel SystemSOOOOOOOOOOOOOO0....0.0.00.0000... 25

4. Measurement of Non-enzymatic Browning with Stor-
age Time on IMF Model Systems by the Hunter Lab
Difference COlor Meter............................ 27
5. Weekly Browning of IMF Model System.............. 28

6. Browning Reaction with Storage Time for IMF Model
Systems Expressed as CIE Parameters............... 31

7. Measurement of Non-enzymatic Browning with Stor-
age Time on IMF Model Systems by the Agtron-SOO... 33

8. Reduced L-ascorbic Acid with Storage Time on IMF
Mordel SYStemSeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 35

9. Available Lysine in Whole Casein Used for the
Model Systems Preparation......................... 38

10. Availability of Lysine with Storage Time on IMF
MOdel systemSOOO0.000000IIOOOOOOOOOOO0.0.0.0000... 39

11. 2-thiobarbituric Acid Reaction with Storage Time
on IMF MOdel S.y3temSeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 41

iii

LIST OF FIGURES

Figure page

1. Weekly Progress of the Non-enzymatic Browning
Reaction of IMF Model Systems..................... 29

2. Progress of the Browning Reaction with Storage
Time for IMF Model Systems in the CIE Chromaticity

Diagram........................................... 32

3. Development of Non-enzymatic Browning with Stor—
age Time on IMF. MOdel SYStemSeeeeeeeeeeeeeeeeeeeee 34

4. Reduced L-ascorbic Acid Loss with Storage Time in
IMF MOdel SYStemSeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 36

5. Development of the TBA Reaction with Storage Time
in IMF MOdel SYStemSeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 42

iv

INTRODUCTION

An intermediate moisture food (IMF), as defined by
various researchers is one that is moist enough to be eaten
without rehydration and at the same time is shelf-stable
without refrigeration or thermal processing. Generally,
it contains moderate levels of moisture (20-50% by weight),
but at the same time by virtue of the addition of a humec-
tant, its water activity is decreased (within 0.60-0.85),
which is below the level required for growth of most organ-
isms.

There are many food stuffs such as honey, jams and
jellies, syrups, dates,,charqui, sweetened condensed.milk
and fruitcakes, that fit the definition of IMF.

Within the last ten years there has been a great deal
of interest in IMF as a result of the commercial success
attained by moist pet foods. The acceptability of these-
products by pets and pet owners has encouraged researchers
to investigate the possibility of creating new intermediate
moisture foods for human consumption.

Although substantial research has been directed toward
IMF and their relation with low water activity, relatively
few studies have been done with respect to their storage
stability. Many chemical and physical changes in a food
occur in the range of water activity encompassed by IMF.

I

2
Some of these changes can cause deterioration, loss of nutri—
ents and poor acceptability. The object of this thesis is to
do a brief study of some of those chemical reactions that are
known to cause changes in flavor, color and nutritional value
of IMF, in relation to water activity levels, temperature and

storage time.

LITERATURE REVIEW

I. Intermediate Moisture Foods and.W3ter Activity
Intermediate moisture foods (IMF) as defined by Potter

11970) are food items that contain moderate levels of mois-
ture, 20 to 50% by weight, which is less than what is nor-
mally present in natural fruits, vegetables or meats and more
than what is left in conventional dehydrated food products.
In addition, IMF have dissolved concentrations of solutes
such that the water activity (Aw) is reduced below that
required to support microbial growth. Therefore, IMF do not
require refrigeration for preservation.

Water activity is a measure of unbound, free water in
a system which is available to support biological and chemi-
cal reactions. It is this unbound, "free" water, and not
the total moisture content, which determines the kind, extent
and rate of microbial, enzymatic and chemical changes which
may occur in foods. Foods with the same total water content
may have very different values for water activity depending
on the degree to which the water present is free or bound to
food constituents.

The water activity of a product is given by the ratio
of the partial pressure of water in food and the saturation
pressure of water at the same temperature (Taylor, 1961).

3

- p - ERH
Aw po - 100

Aw a water activity
p 2 partial pressure of water in food at a certain
temperature.

po - saturation pressure of water at the same tempera-

ture.
ERH - equilibrium relative humidity (%).

An estimate of water activity is provided by Raoult's
law of mole fraction, which involves dividing the number of
moles of water in a given solution by the total number of
moles (solute + water) in the solution (Bone, 1973).

L'-

moles H20

moles H20+ moles solute

 

For an ideal solution, water activity is independent of
temperature and, in actual practice, the water activity of a
- given solution varies only slightly within the range of shelf
storage temperatures (Scott, 1957).

Recent work (Rockland, 1969), indicates that ERH and
water activity alone may not reflect the most precise physi-
cal-chemical state or the suceptibility of a food product to
moisture dependent deterioration. Optimum stability condi—
tions are better described by the combination of both ERH and
total moisture content, as shown by a Moisture Sorption
Isotherm (MSI).

According to Labuza (1968) the MSI can be divided into
several regions depending on the state of the water present.

The first region (A) corresponds to the adsorption of a

zone
adso
tbir
the
for

inde

rang
oxic‘
oxic'

rel:

(19E
tior

5
monomolecular film of water, the second region (B), to
adsorption of additional layers over this monolayer and the
third region (C), to condensation of water in the pores of
the material. No definite relative humidity can be stated
for the cross over points from one region into the next and,
indeed, they may overlap.

As shown by Salwin (1959), the monolayer region seems
to be the optimum moisture content for most foods. Lea (1958)
stated, and.Maloney et a1. (1966) confirmed, the fact that
lipid oxidation increases with decreasing water concentra-
tions below the monolayer value. .Labuza (1970) suggested
that increasing concentrations of water over the monolayer
range to as high as 50% relative humidity slow the lipid
oxidation reaction. ,Martinez (1968) demonstrated that per-
oxide production in freeze-dried salmon decreases as the
relative humidity is increased from above the monolayer to
50% relative humidity. However, Labuza, Tsuyuki and Karel
(1969) found that above 50% relative humidity, lipid oxida-
tion increases with increasing relative humidity.

Above the monolayer moisture value, non-enzymatic
browning increased with increasing relative humidity during
storage of freeze-dried meat (Sharp, 1960). Accordingly,
Labusa (1970) found that most foods show a slow rate of
browning at low humidities. This rate increases to a maximum
in the range of IMF, but decreases with further increases in
humidity due to the dilution of the reactants.

Various studies on relatively simple systems have shown

that even at low water activities, sucrose may be hydrolyzed

6
to form reducing sugars which have a potential for browning
(Karel and Labuza, 1968; Schoebel, 1969; Labuza, 1970).
Therefore, a system that was initially stable may become
susceptible to non-enzymatic browning during storage.
II. Non-enzymatic Browning of Foods

The browning that does not require enzymatic catalysis
is referred to as non-enzymatic browning (Eskin, 1971).

Three main reaction pathways have been described for such
browning (Hodge, 1953):

1. Maillard reaction or carbonyl-amino reaction, which
is the result of reaction of the carbonyl group of aldehydes,
ketones or reducing sugars and the free amino group of amino
acids, peptides and proteins (Maillard, 1912). This is con-
sidered the most important reaction leading to non—enzymatic
browning in foods (Reynolds, 1969). The Maillard reaction
involves the following steps:

a) The carbonyl-amino reaction is a condensation reac-
tion between the free amino groups and the carbonyl groups
(Ellis and Honeyman, 1955; Reynolds, 1963, 1965). A Schiff's
base is obtained as the initial condensation intermediate'
which undergoes cyclization to the corresponding N—substituted
glycosyl-amine (Partridge and Brimley, 1952).

b) The isomerization of the leubstituted glycosyl-
amine from an aldose to a ketose-sugar derivative occurs by
an.Amadori rearrangement (Hodge, 1955). To this point,
Called the induction period, the reactions are reversible and
do not contribute to visual browning since the initial conden—

sation products are colorless (Eskin, 1971).

media
zatic

Kato

inte:
meri

Hodg

diox
basi
cing

flav

free
1949
Garb

{on

th.

7

Three different pathways are believed to operate from
this point to yield the brown endrproducts (Hodge, 1967):

1) formation of 3-deoxy-hexasone derivatives as inter-
mediates followed by a series of condensations and polymeri-
zations to nitrogeneous brown pigments (Anet, 1960 and 1964:
Rate, 1962 and 1963).

ii) formation of methyl-alpha-dicarbonyl derivatives as
intermediates, ending in a series of condensations and poly—
merizations to nitrogeneous brown pigments (Hodge, 1953;
Hodge et a1. 1963; Simon and Heuback, 1965).

iii) Strecker degradation reactions which consist of the
degradation of alpha-amino acids in the presence of alpha-
dicarbonyl compounds yielding the aldehyde that corresponds
to the amino acid minus one carbon atom, lost as carbon
dioxide (Schonberg, 1948 and 1952). This reaction is not
basically a pigment producer but provides instead the redu—
cing compounds essential for pigment formation as well as
flavor and odor compounds (Eskin, 1971).

The carbonyl-amino reaction occurs in acidic or alka-
line pH, but is more rapid as pH increases (Underwood, 1959).
A linear relationship exists between the rate of reaction and
temperature from 00 to 90°C, as shown by a decrease in the
free amino-nitrogen in casein-glucose systems (Lea and Hannah,

1949). Reducing groups are essential as the source of the
C‘Urloonyl groups necessary to interact with the free amino
fir‘3\aps. Mon—reducing sugars can cleave their glycosidic

bonds yielding reducing sugars that actively participate in

the. nation (mm. and Labuza, 1968) and aldo-pentoses

 

appea

136 E
‘0“rd
.‘s

3550:

P011}
and <
:esuj
reac
(Lea

III.

def:
Vita
ith

CHER

dehy

8
appear to be more reactive than aldohexoses (Spark, 1969).

2. Caramelization reactions which occur when poly-
hydroxy-carbonyl compounds are heated to high temperatures in
the absence of amino acids (Hodge, 1953). This process was
found to proceed under acidic or alkaline conditions and is
associated with flavor changes (Eskin, 1971).

3. Oxidative reactions which convert ascorbic acid and
polyphenols into di« or poly-carbonyl compounds (Hodge, 1953)
and oxidative degradation of polyunsaturated fatty acids
resulting in the production of carbonyl compounds that can
react with the free amino groups to produce brown pigments
(Lea, 1958 and Labuza, 1970).

III. L-ascorbic Acid

Vitamin C was the first nutritional adjunct whose
deficiency was recognized as a cause of a disease. The
Vitamin C of foods and biological materials is associated
with its reduced L-ascorbic acid content (Assoc. of Vitamin
Chemists, 1966).

Reduced L-ascorbic acid and its oxidation product,
dehydro-ascorbic acid, constitute a reversible redox system
in the animal organism, though depending on the conditions
Prevailing, dehydro-ascorbic acid could be irreversibly con-
Verted to 2, 3, diketo-gulonic acid and its Vitamin C value

impaired (Penney and Zilva, 1943).
Roe (1936), reported that L-ascorbic acid exists in the
‘neSilaced form only, in the plants and animal tissues examined.
Md“JL:ls (1949) supported Roe's theory and extended this obser-

v‘tion by analyzing not only fresh but processed and

deh'y"

less

ass:

acic

brm
reai

bro'

am
am
re;

and

aci

9
dehydrated foodstuffs. He found that fresh foods contain
less than 5% of their total Vitamin C content as the inactive
diketo-gulonic acid while processed foods have more of this
form and dehydrated ones showed the greatest amount of diketo—
gulonic acid. Other studies showed that with storage time
(Mills, 1949) and with increasing temperatures (Penney, 1943)
the reduced L-ascorbic acid oxidizes readily to dehydro—
ascorbic acid which mutarotates into inactive diketo-gulonic
acid.

Joslyn and Marsh (1935) found that the non-enzymatic
browning of orange juice does not involve the carbonyl-amino
reaction, and reported ascorbic acid as the main source of
browning and carbon dioxide. Supporting this theory,
Stadtman (1948), pointed out that browning in citrus fruits
is always associated with a destruction of ascorbic acid;
Euler (1952 and 1953) and Singh (1948) reported that 2, 3,
diketo-gulonic acid formed from dehydro—ascorbic acid in an
aqueous solutionfforms colored compounds (furfurals) and
evolves carbon dioxide when heated. Guzman Barron (1936)
reported that above pH 7.0 auto- oxidation of L-ascorbic acid
and color production occurs even at 25°C.

Joslyn (1957) found that the concentration of ascorbic
mid initially present has a marked effect on the rate and

extent of browning of ascorbic acid systems. Dulkin (1956)
‘t‘ted that the rate or intensity of the browning in these
'Yatems is not increased by the addition of amino acids, thus

this browning pathway differs from the Maillard type reaction.
‘L£.:1~:lkainen (1958) studied the browning of ascorbic acid in

10
citrate—buffered solutions containing radioactive glycine and
found that a) neither Schiff bases nor volatile aldehydes
were detected, and, b) less than 3% of the carbon dioxide
evolved was derived from glycine. The data, therefore, indi-
cated that the Strecker degradation does not occur in detec-
table degree under the test conditions, in support of Dulkin's
(1956) above statement.
Joslyn (1957) suggested that when the temperature is
raised the browning mechanism changes. Lalikainen (1958)
substantiated this suggestion showing that there is a marked
change in the production of carbon dioxide to pigment forma-
tion at 37°C and 50°C, indicating that different mechanisms
may take place.
Lately, Tatum (1969) separated by GLC and identified
by spectroscopic methods, eight non—enzymatic browning
degradation products, most oflthem furan-type compounds.
IV. Availsgle Lysine
It is recognized that the protein (amino acid) content
of a protein food as determined by analysis does not neces-
sarily establish the amount of that protein "available" when
that food item is consumed by man or laboratory animals
(Calhoun, 1960). Information on: a) amounts of individual
"lino acids contained in foods, b) amounts required by the
inialal for optimum nutrition, and c) the actual amino acid
mount able to be used, are needed for a definite nutritional
""Juation of protein quality (Schweigert, 1953). Pure, dry
Proteins are relatively resistant to change, but the human

11:; uake of proteins occurs in foodstuffs where the proteins

 

are i
Llysi
some
of tt

are

ten
duri
Supp
Patt
valu
lysi
rest
shm
pIOI

toi

aci
str
lys
qre
sug
al

tai

in

and

PM

11
are in complex mixtures. Certain essential amino acids
(lysine, histidine, methionine, cysteine, threonine), and
some non-essential amino acids, react with other constituents
of the foodstuff to form enzyme resistant combinations which
are “unavailable" to the animal (Rice, 1953).

Among the most common reactions are those between pro-
teins and sugars (Maillard, 1912) which yield brown pigments
during heating, drying and storage of foods (Donoso, 1962).
Supporting this theory, Henry (1948), Lewis (1950) and
Patton (1950 and 1955) reported a decrease in the nutritive
value of foods due to a "lysine unavailability", that is,
lysine residues where the epsilon-amino groups are not free,
resulting from the carbonyl-amino condensation. Lea (19fl9)
showed that lysine availability was negatively affected by
processing and storage conditions and that lactose appeared
to brown less readily than glucose in the presence of casein.
The lactose reactions result in less destruction of the amino
acid than do glucose reactions under these conditions, demon-
strating that materials rich in reducing sugars will render
lysine unavailable under very mild conditions, producing
great nutritional damage, while materials low in reducing
sugars have to be exposed to severe conditions for nutrition-
a1 damage to occur. Boctor (1968) heated egg albumin con-

taining different amounts of glucose and found that the loss
1’1 available increased directly with the added glucose.

Lea (1958), working with herring meal stored under air

and. nitrogen, found that the most likely mechanism for the

rgfiuction of nutritive value of the protein of fish meal is

 

tie

Ea:

1V

12
the reaction of lipid oxidation products (peroxides, free
radicals, aldehydes) with amino groups of the protein.

Since the importance of the determination of lysine
availability was discovered, many methods have been developed
for its measurement:

1) Biological Methods:

a) Feeding studies using laboratory animals
(Schweigert, 1953; Guthneck, 1953; Kuiken, 1948 and 1952;
Gupta, 1958; Ousterhout, 1959; Carpenter, 1961; Calhoum,l960).

b) Microbiological assays (Ford, 1964; Scott, 1966).

c) Enzymatic methods (Mauron, 1955; Sheffner, 1956).

2) Chemical Methods:

Chemical methods were developed becausecof the
obvious necessity for a simple, inexpensive and faster assay.
The first chemical method was developed by Carpenter (1955)
based on Sanger's (1945) discovery that the reagent dinitro-
fluoro-benzene (DNFB) reacts with the free epsilon—amino
group of lysine, forming a stable, colored DN—lysine deriva—
tive. But DNFB does not react specifically with the free ami—
no groups; it also reacts with phenolic-hydroxyl, thiol and
imidazole groups (Sanger, 1945) giving interfering pigments.

Bruno and Carpenter (1957) used methoxy-carbonyl—
c“Lloride to make the epsilon-DNP—lysine derivative ether sol—

ukfile. The difference in color intensity of the hydrolyzate

be:Eore and after reaction with methpxy-carbonyl-chloride and

 

13
ether extraction was taken as a measurement of available
lysine. This modification also has its limitations in that
the previous colorless alpha-DNP-histidine derivatives become
colored and lysine units with both their alpha and epsilon—
amino groups free are not measured through these units are
nutritionally active (Carpenter,1960). Rao (1963) eliminated
many of theyyellow, interfering derivatives through the use
of an ion exchange column.

Handwerk (1960) found that destruction of DNP-amino
acids is associated with carbohydrates having a reducing
power or being able to yield reducing compounds on hydroly-
sis. Blom (1967) developed a very sensitive method for the
analysis of carbohydrate rich- products based on chromato-
graphic separation and polarographic detection of the eluents.

Recently, Kakade (1969) developed a simple and rapid
method for determining available lysine. It is based on the
specific reaction of 2, 4, 6, trinitro-benzene-sulfonic acid
(TNBS) with primary amino groups followed by acid hydrolysis
of the TNP-protein and ether extraction of the alpha-TN?
amino acids. The epsilon-TN? lysine derivatives remain in
the aqueous phase where they can be determined
apectrophotometrically.

V3) Rancidity (off-flavors)
Lea (1952) defines rancidity as “off—odor or flavor
thlat has developed in an oil or fat as the result of deter-
ioration or storage".

The off-flavors and odors which develop as a result of

14
secondary autoxidative changes are quite unpleasant and make
a food unpalatable.

One of the most commonly used objective measurements
to assess the extent of rancidity is the peroxide value (PV).
This determination provides a measurement of the transitory
hydrOperoxides which are intermediates in the oxidation
reactions which eventually form the saturated aldehydes, ke-
tones and alpha, beta-unsaturated carbonyls which contribute
to off-flavors and odors. Hence, a low PV is no guarantee
of lack of off-flavors.

Based onBernheim's (1948) observation that fatty
acids when aerobically incubated react with TBA to produce
a characteristic red color, Dunkley and Jennings (1951)
associated the red color produced by the reaction with the
oxidized flavors. Patton and Kurtz (1951) concluded from
spectral measurements that malonaldehyde (MA) was one of the
compounds (Sinnhuber, 19582; Kwon and Menzel, 1965) respon-
sible for the red pigment obtained when milk fat plus TBA
reagent were heated. They also stated that MA could not be
the only compound responsible, since the color production com-
pounds were only partially removed after exhaustive extraction
with hot water.

If a TBA determination is made on lipid-containing
foods, the formation of yellow interfering compounds can be
v‘53tatious (Turner, 1954; Caldwell, 1955; Briggs and Bryant,

4159553; Yu and Sinnhuber, 1957). Briggs and Bryant (1953) at-
tributed this color to carbohydrate interference

‘Afiunlaax of 4507460nm) which will produce erroneously high

15
absorption values for the red pigment, which has an Amax at
532nm due to overlapping effects. In order to apply the test
to extracts from foods containing carbohydrates, Caldwell
(1955) devised an adsorption chromatographic procedure for
separation of the yellow components which allowed the
residual red color to be estimated spectrophotometrically
(Yu and Sinnhuber, 1962).

Yu and Sinnhuber (1957) performed the TBA assay on
intact fish samples. Also Sinnhuber (1958b) reassessed the
effectiveness and sensitivity of the empiriCal TBA method
and suggested means for quantification of results. For this
purpose, the stable compound TEP (tetra-ethoxy-propane) was
used as standard and the term "TBA number" (mg MA/lOOOg of
material) was introduced.

Even though so much workahas been done on this subject
the mechanism of the reaction still is not understood and
efforts to chemically characterize the pigment have been
unsuccessful.

Tarladgis and Watts (1960) reported from oxygen uptake
measurements that MA is not accumulated as a stable end-
product, but is destroyed by oxygen.

Kurtz, Price and Patton (1951) formulated a mechanism
for the formation. of the Ma-TBA red pigment where the mono—

eholic form of MA attacks the reactive methylene group of
tiles TBA, followed by ring closure.

Kwon andeenzel (1965) found that above pH 6.5, MA

“"21.sts dissociated into its anion which could.react with

16

amino acids, proteins, glycogen and other food constituents
to form products where the MA exists in "bound form".
Supporting this theory, Lea and Parr (1958 and 1960) obtained
some evidence that MA reacts with protein in herring meal
because of high "bound lipid“ and TBA values given by oxidized
in contrast to fresh meals. Buttkus (1967) reported that MA
reacts with the epsilon-amino groups of myosin, specially
lysine, in a manner similar to the carbonyl-amino reaction.

In spite of the inconveniences and uncertainties, the
TBA method as improved through the years is capable of being
correlated to the sensory quality of milk (Patton and Kurtz,
1951; Kurtz, Price and Patton, 1951; Dunkley and Jennings,
1951; King, 1960); of fish (Lea, Parr and Carpenter, 1958
and 1960; Sinnhuber,l958b; and Yu, 1957); of pork (Turner,
1954); and of baked goods (Caldwell, 1955).

EXPERIMENTAL METHODS

1. Preparation of the Model System
Each of the three modifications shown in Table l was

prepared as follows:

a) All the dry ingredients were accurately weighed and
mixed.

b) The glycerin and oil were added to the dry, mixed
ingredients. The paste was then mixed thoroughly with a
Kitchen-aid mixer to a smooth consistency.

c) Water, as required, was slowly added to the ingre-
dients during mixing.

d) Approximately 30g of the product was placed in glass
jars and stored at 370 t 2°C in a Cenco incubator.

e) Samples of each modification were taken out as
required for the various analyses, never returning an opened
jar to the incubator.

II. Water Activity and Moisture Content Determinations

a) Initially the moisture content of each modification
was determined by drying the sample to constant weight in the
vacuo oven at 15 mm Hg of absolute pressure and 90°C over
76 hours. Subsequent moisture determinations were made by
the Karl Fischer titration. Both procedures were those
described in the manual of the Association of Official
Analytical Chemists (1970) . '

b) Water activity was measured by the manometric device

17

Table 1.
Modification A
Water
Corn oil
Casein
Sucrose
L-ascorbic acid

Sorbic acid

Modific tion B

Hater

Corn oil

casein

Glycerin
L-ascorbic acid

Sorbic acid

Modific tion C

Water

Corn oil

Casein

Sucrose
Glycerin
L-ascorbic acid

Sorbic acid

18
Model System Recipes

Non—fat basis
28.50

23.75
47.48
0.06

0.21
100.00

28.50

23.75
11.87
0.06

0.21
100.00

28.50

23.75
23.75
23.75

0.06

0.21
150.00

Regular basis

26.02
8.67
21.69
43.38
0.05

0.19
100.00

26.02
8.67
32.53
10.84
0.05

0.19
100.00

26.02
8.67
21.69
21.69
21.69
0.05
0.19

100.00

19
built by Sood (1972) and with the Electric Hygrometer-
Indicator 153001 according to the procedure provided with
this instrument. The manometric device is a modification of
the Harkower and.Myers (1943) Method.
III. TBA Value Determination

The TBA method described by Yu and Sinnhuber (1957)
was followed with the following modifications in reagents
and procedure:

a).§ggplg_p£gpsgggig§. 10g of the sample were blended
with 100ml of water for 5 minutes in a‘Waring blender.

b) TCA-TBA gesgent. 109 of TCA and lg of 2-TBA were
dissolved in 100 m1 of anhydrous ethanol with gentle heating
and stirring.

c) Procedure.

1) In a graduated test tube 3 ml of the sample slurry
were mixed with 5 m1 of the TCA—TBA reagent and made
to 10 ml volume with anhydrous ethanol.

ii) The tubes are placed in a 790 t 1°C water bath for
exactly 20 minutes. The boiling point of anhydrous
ethanol is 78.5°C and when overheated projects out of
the tubes.

iii) The reaction mixture is allowed to cool to room

temperature and then 9 ml are added to the top of a

:12 cm long cellulose (Whatman standard grade) column

and the pink pigment which absorbs strongly at 535nm

was separated by the chromatographic method of Yu and

Sinnhuber (1962).

20

IV. Browning Measurements

Browning measurements and conversions were made accor-
ding to the procedures outlined in the Agtron—SOO manual and
the Hunter-Lab Color D—ZS Difference Meter manual.
V. Availablg_Lysine Determinations

Available lysine determinations were made by the pro—
cedure of Kakade and Liener (1969) and the modification
suggested by Posati (1972). All the samples, previous to
the assay, were sonicated until a complete solution in the
bicarbonate buffer was obtained.
VI. ‘Equc d L-asgorbic ApidfiDgtermigations

Reduced L-ascorbic acid determinations were obtained by
the 2, 6 dichlorophenol indophenol visual titration method,
according to the procedure of the Association of Vitamin
Chemists, Inc. (1966).
VII. Total Ascorbic Acid Determingsigg

Total ascorbic acid determination was determined by an
adaptation of Roe (1936, 1943) method according to the Asso«

ciation of Vitamin Chemists, Inc. (1966).

RESULTS

Each of the three modifications of the system was
initially assayed for moisture content and water activity
and, subsequently, weekly for reduced L-ascorbic acid, TBA
value, available lysine and browning reaction. Each of the
results reported is the average of the assay run in tripli-
cate or duplicate.

The three different modifications of the model system
were prepared in two separate batches. Batch 1 was pre-
pared first, and was intended to provide a clear idea of
the trend of the reactions. Batch 2 was meant to complete
the study and to confirm the validity of the results of
Batch 1.

The intermediate moisture food model system was pur-
posely prepared in such a way as to get high nutritional
value. This can be shown by its protein/calories ratio
when compared to some nutritionally rich food staples.

Enriched white bread.........3.llg prot/cal in 100g
Whole milk...................5.57g prot/cal in 100g
IMF model system.............7.3lg prot/cal in 1009
Raw whole eggs...............7.9lg prot/cal in 100g

Cooked roast beef............9.40g prot/cal in 100g

21

22
I. ‘Wster Activity Determination
The water activity of each modification was calculated
according to Raoult's law of mole fraction and experimentally
measured with the electric hygrometer and the Markower and
Myers (1943) manometer.

All the experimental values for the water activity were
lower than the calculated ones, as shown on Table 2. The
gray sensor of the electric hygrometer measures the highest
range of water vapor pressure, hence it was used to ini-
tially give a valid approximation of the vapor pressure of
the samples. The measurement performed with the violet
sensor of the electric hygrometer and the Markower and.uyers
(1943) manometer showed a very good correlation and were
considered indicative of the water activity of the three
modifications of the system.

Accordingly, modification A showed a water activity
value of 0.84; modification B, 0.81: and modification C,
0.18; with the total range difference in actual water ac-
tivity being 0.06 for the system.

II. MturL Content

Attempts were made to determine the moisture content
of the various modifications by conventional oven drying
to constant weight. Higher oven temperatures were unsatis-
factory due to rapid browning and degradation. Entrainment
distillation or azeotropic distillation techniques were not
satisfactory due to the extensive foaming and formation of
tenacious emulsions with the entrainer. Finally, drying in

a vacuum oven at 15 mm Hg absolute pressure and 90° to

23

.oom: ouoz muomcomoumms uoaow> paw homo one
.Hoomma Hoooe HoumoHUcHluouoEoummm ownuomam mop bus; wounmmoa muH>Huom Hopes

.AMFmH .ocomv coauomnm mace mo Boa m.uadomm an ooumaouamu

Amy
AH.

m>.o mh.o om.o mm.o o
am.o Hm.o mm.o hw.o m
om.o om.o mm.o No.0 4
HopoEocmE pmaoa> mono
mummzluw3oxumz ANVHouoEoummm UHHuoon
HmucoEHHmme Aavooumasoamu COHDMUHHHUOE N nuumm
m>.o mh.o ow.o mm.o o
Hm.o Hm.o mm.o hm.o m
~m.o . em.o om.o mm.o 4
Houoaocme poaow> mono
whomzluosoxumz ANvuouoEOHmmm UHHquHm
Hmucmeflumoxu toumasoamo coflpmosmsooz H soumm

AHV

.mamummm H0002 mzH mo MDH>HDU¢ HoumS.HmucoEfihomxm 0cm Umpmanuamo .N magma

24
essentially constant weight was chosen in an attempt to ob-
tain moisture contents consistent with the known amounts of
water added to the IMF model systems.

The quantity of moisture added to the samples was con-
stant. This accounts for the identical values for the cal-
culated per cent moisture in all batches and modifications
as shown on Table 3. The moisture content determined by
analysis correlates reasonably well with the calculated
value in modificatffin‘A. ‘However, in modification B and C,
this correlation is lost. The experimental value was, in
modification B, 6% higher than the calculated one and in
modification C, 18% higher.

The analytical problem lies in the fact that variable
amounts of glycerol, added to modifications B and C, are
lost in the extended vacuum oven treatment at 90°C.

Subsequent to the unsuccessful attempts to determine
moisture by even methods or entrainment distillation, addi-
tional batches of IMF were prepared exactly as before.
Portions of the I“? were then analyzed by the Karl Fischer
titration. In order to recover all of the moisture for ti-
tration, the samples were held immersed in absolute methanol
with constant stirring for two hours prior to titrating
with the standard reagent of iodine, sulfur dioxide, yri-
dine and methanol. The results obtained confirmed the assump-
tion that as calculated, the moisture content is constant in

the three modifications.

25

.0 om one oHSmmon ouoaomnm
mo mm as ma um co>o osom> m as nomflmz Dnmumcoo on ooflmo moumoflamflnu mo oomuo>4 " Amy

.AH magma some momflomh meoummm Hobos Eoum ooumasoamu u Adv

mm.mN
NN.©N
mN.©N

GOHDMHDHD Monumflm Humx
Amflmmn Hoasmouv whopmfioe A3\3V a

Hm.¢v
mm.Nm
Hm.mN

no>o osom> cw OCHMHQ
imammmvumanoonv onsumfloe A3\3v x

mm.¢v
om.Nm

m¢.hm

.no>o,osum> CH unflmun
imammm Hmasmouv ohsumwoe A3\3v R

.maoummm Homo: mzH mo ucoucoo whopmfloz Hmucoeflnomxm one ooumanoamu

NO.@N
No.0N
No.mN

was no mo
AHVU H H

No.0N
No.0N
No.0N

mum :0 m
AHVD H H U

No.0N
No.0N
No.0N

mum :0 m
AHVG H H U

m>.o use .0
Hm.o "as .m

em.o "so .4

>us>euu< nouns
one GOHuMUHMHUoz

z .

me.o “ 4 o
3

Hm.o " a .m

3
em.o " 4 .e

>uo>euo¢ nouns
tam coeumoomeooz

3
mp.o “ 4 .o
Hm.o use .m

3
em.o u 4 .4

mpfl>fluud Houm3
one cowumoHMHooz

m gopmm

m noumm

a noumm

.m magma

26

III.Browning Reaction

Two simple, fast and direct methods to measure the
browning reactions were used employing the Agtron—SOO and
the Hunter Lab Difference Color Meter.

The Hunter Lab Difference Color Meter decomposes the
color into three vectors (L, aL and bL) and gives values
for each one. The data obtained, (see Table 4), show that
the values for L (lightness) dropped over ten units within
the first eight days of storage at 37° 1 2°C, and then
(continued to gradually decrease with storage time. The
value for aL changed with time from negative (greenness)
to positive (redness) while the value for bL started as
positive (yellowness) and continuously increased.

From.the previous data, the weekly browning rate
(Delta 8), as well as the total browning of the model system
were calculated (see Table 5 and Fig. 1). They clearly show
that the largest amount of browning for this system occurs
within the first week of storage at 37° 3 2°C, decreases
abruptly during the second week, rises slightly during the
third week and continuously decreases in the following weeks.
0f the three modifications, modification B browned the most.

The International Commission of Illumination (CIE)
expresses its results as X (blueness), Y (greenness) and Z
(redness) parameters. Conversion of the L, aL and bL scale
to the CIE scale to the CIE scale was performed. Results
are shown on Table 6. In order to graphically show the

trend of the browning reaction with storage time, the CIE

27

Table 4. Measurement of Non—enzymatic Browning with Storage

Time on IMF Model Systems by the Hunter Lab Dif-

ference Color Meter.

standard, where
L: lightness;
aLx
be

(+) redness;

(+) yellowness;

(-) greenness;

(-) blueness.

Modification and Time (days) L aL b1
Water‘Activity

A, Aw - 0.84 l 64.50 -4.78 12.52

8 52.80 0.76 18.34

15 50.12 2.34 20.00

22 48.00 6.81 20.50

29. 46.95 3.20 20.25

36 45.62 4.02 20.70

3, AW 8 0.81 1 66096 "4.88 12.42

8 51.85 —0.52 16.76

15 49.24 1.02 18.10

22 48.40 6.21 19.20

29 46.40 2.65 19.13

36 43.63 3.22 19.14

C, A" 3 0e78 1 64.10 "4e88 14el4

8 53.48 -0.15 17.20

15 49.04 1.06 17.64

22 44.82 5.81 17.58

36 45.53 2.70 18.72

Yellow tile (L: 83.0; aL: -3.5; bL: 26.5) used as

28

Table 5. Weekly Browning of IMF Model System.(l)

Modification and Browning rate
Water Activity Time (days) 2(de1ta E/time )
A, Aw30.84 l — 8 14.19

8 — 15 3.52

15 — 22 4.97

22 — 29 3.76

29 - 36 1.62

overall 22.38

B, Aw:0.81 l — 8 16.62
‘ 8 -‘ 15 ' 3e31

' 15 - 22 5037
22 — 29 3.99

29 - 36 3.02

overall 25.60

C, Aw10.78 l — 8 12.15
8 - 15 4.62

15 - 22 6.35

22 - 29 3.44

overall 20.57

(1) As ca1cu1ated from the Hunter units (L, ‘L' bL) by
the formula

on x\f(AL)2 + (A3132 + (AH-)7

 

everallx Total browning undergone by the sample in five
weeks of storage at 31° 1 2°C.

29

0»; 00a. 0»... mm

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

26 u .4 .o

 

 

 

 

 

 

 

 

 

 

3?. .o

 

 

 

 

 

 

 

 

«no... .<

 

 

 

 

 

 

 

 

 

.oaoumwm Hoooz .mzH no soauooou moan—30.3 oafimasotsos 93 mo moonooun ado—003

ON

NN

tN

DN

.ON

.w

(AB/time) Browning Difference

30
provides us with the so called Chromaticity Diagram, on
which the color can be expressed in two dimensions. Conver-
sion of the x, Y and Z parameters to the diagram.coordinates,
x and y, was carried out. Results are shown on Table 6 and
graphically in Fig. 2.

The relative spectral reflectance of the browning
reaction was followed with the Agtron-500. The results
obtained are shown in Table 7 and graphically in Fig. 3.

The Agtron-500 supports the fact, found with the Hunter Lab
Difference Color Meter, that the largest amount of browning
in the system occurred within the first week of storage at
37° 3 2°C and, from then on, increases gradually. 0n the
36th day, modification A appeared to have browned the most,
followed by modification B and modification C, the least.
IV. Reduced.L—§scorbic Acid

The reduced.L-ascorbic acid was determined in tripli-
cate by the 2, 6 dichlorophenol indophenol visual titration
method (Association of the Vitamin Chemists, Inc., 1966).
Recoveries of 99% were obtained with this visual method.

For the three modifications, the reduced L-ascorbic
acid content of the system decreased with storage time at
370 i 2°C (see Table 8 and Fig. 4), being consistently high—
est in modification A, less in B and lowest in C. By the
36th day of storage, the remaining reduced L-ascorbic acid
in modification A was 25.60 mg/100g of material; in modifi—
cation B, l9.20 mg/100g of material and in modification C,
14.40 mg/100g of material.

31

Table 6. Browning Reaction with Storage Time for IMF Model

Systems Expressed as CIE Parameters.(l)

Modification and Time
Water Activity (days) x Y Z x y

A, AW = 0e84 l 29e04 41e60 54e42 0e29 Oe3l

8 21.66 27.88 25.43 0.34 0.34

15 25.27 25.12 19.55 0.36 0.36

22 24.40 23.04 16.27 0.38 0.36

29 22.44 22.04 15.31 0.37 0.37

36 21.42 20.81 13.24 0.38 0.37

B, Aw a 0.81 1 42.11 44.83 59.64 0.29 0.30

8 26.19 26.88 26.19 0233 0.34

15 24.03 24.24 20.83 0.35 0.35

22 24.64 23.42 18.37 0.37 0.35

29 21.97 21.71 16.25 0.36 0.36

36 19.43 19.03 12.85 0.38 0.37

C Aw 0.78 1 38.51 41.08 50.95 0.29 0.31

8 27.98 28.60 27.98 0.33 0.34

15 23.85 24.05 21.16 0.34 0.35

22 21.15 20.09 16.00 0.37 0.35

29 20.74 20.52 25.56 0.36 0.36

36 21.00 -20.73 15.49 0.37 0.36

(1) As calculated from the Hunter units (L, aL, bL).

32

snag OouocHs—HSHH o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

on. o... 8.
.0613 .m 3.0» s< .<
Y \.
e820. e220
4 .N -
V. .V.
.5...» .8...» .3...»
.eeuomao musoeumaouno mHo on» as
28093 Homo: BAH now 08.3. summon—m saw... consumed moon—59.5 on» we moouooum .N .0?”

on.

0'.

33
Table 7. Measurement of Non-enzymatic Browning with Storage

Time on IMF Model Systems by the Agtron-500.(l)

Modification and

Water Activity Time (days) Relative Spectral Reflections

A, Aw = 0.84 1 58.50
8 25.00

15 18.50

22 15.50

29 14.20

36 12.50

B, AW - Oe81 1 75e00
8 29.00

15 23.00

22 18.50

29 15.50

36 13.80

C, Aw = 0.78 1 67.00
8 33.50

15 24.00

22 20.00

29 18.00

36 18.00

(1) Measurements were done on the Blue Channel using

the 00 and 41 rel. spec. ref. disks as standards.

34

Fig. 3. Development of Non-enzymatic Browning with Storage

Time on IMF Model Systems

 

80

TO

 

60

 

 

50

 

4O

 

Relative Spectral Reflectance

30

\l
\\ " '
‘5'".
20 3 "

\ \\""\«.L.........J c, Aw=0.78

‘I. I
\ \ 8, Aw = 0.8l
A, Aw: 0.84

IO J

 

 

 

 

 

 

 

 

 

 

 

time (days)

35

Table 8. Reduced L-ascorbic Acid with Storage Time on IMF

Model Systems. (1)

Modification and Time (days) mg reduced L-ascorbid
Water Activity acid/100g
A, aw = 0.84 0 59.70.
2 51.47
9 36.00
16 32.40
23 28.33
30 23.99
36 25.60
B, Aw = 0.81 0 58.50
2 53.86
9 32.80
16 31.20
23 29.16
30 17.48
36 19.20
0, AW — 0.78 0 59.20
9 54.26
16 28.00
23 22.08
30 15.45
36 14.40

(1) Measurements were done by the 2, 6 dichloroPhenol
indophenol visual titration method according to the

Association of Vitamin Chemists, Inc. (1966).

36

Fig. 4. Reduced L—ascorbic Acid Loss with Storage Time in
IMF Model Systems.

 

60

 

50

m@@

50 , / , A 7._
20 Zfl:/%:////Z/f%7///V%7 A// A, Aw=0.84
.. ///¢////¢///g 4%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:: \
.3h

30 .21.: n.'

 

 

 

 

 

 

 

 

 

 

 

 

 

 

b o o a . o u a n . . o . . . . . u . . . .
20 w -- ' ' ' : : : : : : : : : : :-:+:
. . .---". a, Aw=0.8l

 

 

mg Reduced L—ascorbic Acid/100g sample

 

 

 

 

 

 

 

 

 

 

 

 

 

C, Aw = 0.78

O 4 8 l2 IS 20 24 28 32 36

time (days)

37

Measurements of total ascorbic acid by an adaptation
of the method of Roe (1936, 1943), according to the Associa—
tien Vitamin Chemists, Inc. (1966), were performed at the
beginning and at the end of storage time. The results
showed a loss of ascorbic acid in the system with storage
time at 37° : 2°C. .Modification A lost less ascorbic acid
than modification 3 and modification 3, less than modifica-

tion Ce

V. Available Lysine
Changes in available lysine were measured colorimetri-

cally by the TNBS method of Kakade and.Liener (1969).

‘Whole casein samples were analysed to check how much
available lysine was actually present in the casein being
used for the model system preparation (see Table 9). It
was found that 3.70% of the total lysine was ”unavailable"
in the whole casein when compared to theoretical values.

The results obtained for the model system with the TNBS
method of Kakade and Liener (1969) are not very clear (see
Table 10). The three modifications suffer an initial drop
in available lysine followed by an increase to a more or
less steady state. After 31 days of storage at 37° 3 2°C
no modification showed a large change in available lysine.
VI. TBA Reaction

Following chromatographic separation of interfering
pigments (Yu and Sinnhuber, 1962) the presence of lipid oxi-
dation products was detected colorimetrically by the TBA
reacaion (Yu and Sinnhuber, 1957).

The three modifications showed an abrupt increase in

38
Table 9. Available Lysine in Whole Casein used for the

Model Systems Preparation

TheorEtical (mg Of lYSine/g Of casein)S}1..............82.0
Experimental (mg of availably leine/g of casein)......79.0
Experimental percent of available lysine in whole casein

tested (%)O...0.0...000......OOOCOOOOOOOOOOOOOOOO0.0.0.96.3

(1) Gordon, W.G., W.F. Semmett, R.S. Cable and M. Morris.
1949. A comparison of alpha and beta casein. :1;

Am. Chem. Soc. .11, 3293.

39
Table 10. Availability of Lysine with Storage Time on IMF

Model Systems.(1)

Modification and Time (days) mg avail. 1ysine/ml=mg

Water Activity avail.lys/lOmg

A, Aw = 0.84 1 157
3 157

8 147

10 146

16 129

17 168

23 153

24 160

29 144

31 ~ 153

B, Aw = 0.81 1 184
3 158

8 129

10 135

16 167

17 166

23 198

24 183

29 170

31 173

C, Aw = 0.78 1 173
3 128

8 198

10 179

16 172

17 197

23 200

24 205

29 185

31 155

(1) The TNBS method of Kakade and Liener (1969) with
modification suggested by Posati (1972) were used

for lysine availability measurements.

40

O.D. at 535nm by the 8th day (see Table 11 and.Pig. 5),
followed by an abrupt decrease to almost initial levels and
a subsequent increase in the reaction with storage time.
On the 36th day of storage at 37° 1 2°C, modification.A
showed the highest O.D. at 535nm, while modification B was
less and modification c the least.

In the three modifications, rancid off—odors started
being perceptible en the 8th day of storage at 37° 1 2°C

and were obvious on the 15th day.

41

Table 11. 2-thiobarbituric Acid Reaction with Storage Time

on IMF Model Systems.(l)

Modification and Time (days) 0.D. at 535nm
Water Activity
A; Aw = 0.84 1 0.060
8 0.090
15 0.060
22 0.082
29 0.110
36 0.167
8 0.091
15 0.063
22 0.054
29 0.093
36 0.127
C, Aw = 0.78 1 0.058
8 0.091
15 0.061
22 0.064
29 0.067
36 0.090

(1) Measurements were done according to Yu and Sinnhuber
(1957) and the separation of interfering pigments by
chromatographic adsorption on cellulose was performed

according to Yu and Sinnhuber (1962).

Fig. 5.
J
on
no
es

5 .02
m
M
In
4.)
o
"E
o

42

Development of the TBA Reaction with Storage Time

in IMF Model Systems.

 

/ A,Aw:0.04
/

 

 

 

N

 

‘L

 

V

 

 

Roncldity
obvious

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.0.

 

 

 

\Wx
*\

 

 

 

 

 

 

 

 

 

 

 

 

‘\
.00
Roactdlty
obvious
.04
.02
c, Aw = O. 78

time (days)

8, Aw=0.8l

DISCUSSION

o t 2°C over a period of

The effects of storage at 37
time were studied in an intermediate moisture food model
system. This model system was studied at three different
water activity levels. Different levels of the water acti—
vity were obtained by decreasing the sugarzglycerol ratio
while maintaining a constant moisture content.

Glycerol is called a "humectant" because it associates
with water molecules when added to a product containing
moisture. This property also effectively lowers the water
activity, A", of the system. In this way, the replacement
of about 11% sucrose by glycerol in the system should
theoretically decrease its water activity value by 0.05
units, but it is actually lowered by 0.11 units (see
Table 2). This difference between calculated and observed
values when glycerol is added to the system is due to a
property of glycerol called "activity coefficient", which
is a function of concentration (Bone, 1973).

From Table 3 one can observe that, in modification A,
containing no glycerol, the calculated and experimental
moisture values coincide. However, when glycerol was added
to the system the moisture content showed an apparent increase.

The three modifications were simultaneously submitted to the
43 '

44
same drying conditions, suggesting that the extra loss in
weight comes from the almost total volatilization of gly-
cerol. Therefore, the moisture determinations for modifi-
cations B and C as accomplished by vacuum oven drying at
90°C cannot be considered valid. The data are displayed to
demonstrate that such losses do occur.

The subsequent Karl Fischer titration data correlated
very well with the theoretically calculated values in all
three modifications. They confirmed the assumption that
the moisture content is actually held constant in the three
modifications. Therefore, the calculated moisture contents
were assumed to reflect the final moisture content of the IMF.

Labusa (1970) defines Intermediate Moisture Foods as
those food items whose moisture content varies from 20-40%
and whose water activity is therefore above 0.50. Our
samples fulfill thses requirements, and can therefore be
considered IMF by this definition.

It is well known that the carbonyl—amino reaction
(Maillard,l912) produces brown pigments as end products.

In a carbohydrate—protein-lipid system, this reaction can
occur between the amino groups of the amino acids, nitrogen—
containing phospholipids, peptides or proteins, and the
reducing groups of carbohydrates, derivatives of carbohy-
drates or lipid oxidation products. Carbohydrates, especially
pentoses and hexoses and ascorbic acid are capable of
ultimately producing brown pigments by decomposing to
furfural—like cupounds (Tami: 1969) and subsequent

45
polymerization.

It can be clearly seen from the data in Fig. 1 and
Fig. 3 that the largest amount of browning occurs within
the first week of storage at 37° 3 2°C. This correlates with
a decrease in lysine availability (Table 10), with an abrupt
increase in 0.D. at 535nm of the TBA reaction products
(Fig. 5) and a decrease of over 40% of the reduced ascorbic
acid content of the system (Fig. 4). This suggests that all
these three reactions are occurring during the browning of
the system.

After the first week, the browning rate sharply de-
creased and then diminished gradually with time. This
coincided with the gradual decrease in loss of reduced
ascorbic acid and with the increase in 0.D. at 535nm in the
TBA reaction. However, browning somewhat surprisingly did
not correlate with the available lysine as measured
chemically (Table 10).

A continuously decreasing available lysine level during
storage would seem consistent with continuous browning of the
system. Instead, a variable curve was obtained. The initial
decrease in lysine availability has a logical explanation,
due to the reaction of epsilon-amino groups with reducing
groups, but reasons for the subsequent increase in availa-
bility are not clear. The chance of arginine (4.2% in whole
casein) and perhaps histidine (3.0% of whole casein) reaction
with the TNBS reagent and of carbohydrate interference
regardless of the internal standardization (Posati, 1972)

carried out, is one possibility for these data.

J

46

The abrupt increase in the TBA values during the first
eight days of storage of the system at 370 1 2°C suggests
that the lipid oxidation reaction is very active under these
conditions (see Fig. 5).-

Lipid oxidation yields reducing compounds which are
very reactive. The simultaneous increase of the TBA value
and the browning reaction, supports the hypothesis of the
involvement of lipid oxidation products reacting with amino
groups with the formation of brown pigments.

It has been shown that malonaldehyde is not a very
stable end—product. It can be destroyed by oxygen
(Tarladgis and.Watts, 1960), or can strongly react with other
food constituents (Lea and Parr, 1958; 1960); Buttkus, 1967).
The decrease in TBA values in the second week of storage, can
be explained in that, malonaldehyde or other TBA reactive
material may be destroyed by the oxygen available in the
headspace of the jar. Hence, the TBA values obtained during
this storage period did not entirely reflect the actual
lipid oxidation that occurred. Possibly, when the available
oxygen in the headspace was exhausted, such reaction ceased
and the TBA value again reflected the lipid oxidation occur—
ring in the system.

However, no headspace oxygen determinations were made
and there were no control samples stored in inert gas, so
this explanation is purely speculative.

On the browning results obtained by the Hunter Lab
Difference Color Motor, modification B browned to the great-

est extent followed by A and finally C. ‘When browning was

47

measured as relative spectral reflectance with the Agtron-SOO,
modification A was the brownest, followed by B and C,
respectively. This difference in results can be explained by
the fact that the Hunter Lab Difference Color Meter breaks
down the reflected light into three parameters, as previously
explained, and measures each. On the other hand, the Agtron-
500 records only the amount of light the product is able to
reflect. Therefore, if we consider the values obtained with
the Agtron-SOO as expressions of the visible extent of
browning, a direct correlation between the browning under-
gone in the system and its water activity levels can be
made. The higher the water activity value, within the water
activity range of 0.78:to 0.84, the more browning the sample
undergoes. These results agreezwith those of Labuza (1970).

The CIE (Commission of International Illumination)
chromaticity diagram is an axis of coordinates in which one
of the three chromaticity coordinates, which is a ratio of
each of the three primary colors to the total sum, is plotted
against another chromaticity coordinate. This enables one
to follow the trend of a color reaction (see Fig. 2).
During storage, the color of the three samples studied shifts
toward the orange-yellow part of the chromaticity diagram,
becoming progressively farther from the illuminant point as
they darken perceptibly in color.

The good correlation between the browning reaction and
the loss of reduced ascorbic acid has been previously stated.
The loss of reduced L-ascorbic acid for this particular

system is inversely related to its water activity content

48
(see Fig. 4), suggesting a protective effect of the free
water molecules on reduced.L-ascorbic acid oxidation. On
the 36th day of storage at 37° 1 2°C, 24.32% of the reduced
ascorbic acid remained in the modification having the lowest
water activity (o.78) level, while 32.82% and 42.88% remained
in the modifications having 0.81 and 0.84 water activity
contents, respectively.

Analyses of the total ascorbic acid of the system at
the beginning and end of storage time, demonstrated the
disappearance of most of the ascorbic acid from the system.
The degree of its disappearance was directly related to the
loss of the reduced ascorbic acid and as with reduced
ascorbic acid, was inversely related to the water activity
content of the system.

The disappearance of ascorbic acid from the system
can proceed by degradation to furfural—like compounds
(Tatum, 1969), or by reaction with nitrogeneous food consti-
tuents (Dulkin, 1956). Both processes can contribute to the

overall browning of the product.

CONCLUSIONS

1. Intermediate moisture foods of three different
levels of water activity were prepared by maintaining con-
stant levels of moisture in all three systems while adjusting
the sucrosezglycerol ratio.

2. For this particular model system the determination
of moisture by drying to constant weight at absolute pres-
sure of 15 mm Hg and 90°C is not appropriate because of the
high volatility of glycerol.

° i 2°c browned

3. The model systems stored at 37
within the first week of storage. The browning was directly
related to the water activity content of the systems and
seems to be the overall result of carbonyl-amino reaction
and ascorbic acid decomposition.

4. There was a definite loss of Vitamin C activity
and a parallel disappearance of total ascorbic acid in the

O 1 2°C. These losses, inversely rela-

systems stored at 37
ted to the water activity levels of the system, suggested a
protective effect of the free water and also that degradation
products of ascorbic acid easily decompose into furan—type
~compounds which are known to react rapidly in the browning of

food systems.

5. Lipid oxidation reactions were very active in the
49

50
product during storage at 370 i 2°C. TBA values showed
good correlation to the appearance of browning and off-odors
in the IMF.

6. The TNBS method for lysine availability did not
appear to be applicable to this particular system. The
results obtained were inconsistent with the continuous brown-
ing reaction undergone by the system.

7. When these intermediate moisture food model systems
were stored at 370 t 299, lipid oxidation, browning and loss
of ascorbic acid occurred, adversely affecting their appear-
ance and nutritional value.

8. In this range of water actiVity (0.84-0.78) all
degradative reactions (browning, loss of ascorbic acid,
rancidity, decrease of lysine availability) took place
rapidly and to approximatelyfithe same extent. Therefore,
this model system is not stable at these water activity
levels when stored at 370 t 2°C. Hence, to benefit from
the application of intermediate water activity principles to
food stuff, further studies on different water activity
levels as well as of the various, interlinked, chemical

reattions and the methods of minimizing or inhibiting these

reactions need to be explored.

SUGGESTIONS

1. The determination of total reducing and non-
reducing sugars might yield useful information on the in-
volvement of the sucrose in the IMF.

2. Experiments involving packaging the IMF in inert
gas and the addition of various levels of antioxidents to
the system would be worthy of future investigation.

3. Some observations were made regarding the improve-
ment of the Markower and.Myers (1943).manometerx

a) The neck of the sample tubes is too narrow.

7 Hence, the introduction of the sample into the vial
is a difficult task, with much opportunity of the
sample to dry out.

b) A precision thermometer installed inside the
sample vial is necessary in order to be able to mea-
sure with the highest degree of accuracy, the right

temperature at which the sample reaches equilibrium.

51

l.

2.

3.

6.

7.

8.

9.

10.

11.

12.

BIBLIOGRAPHY

Anet, E.F.L.T. 1960. Degradation of carbohydrates.
I. Isolation of 3-deoxyhexosones. Aust. J. Chem.
,aa, 396. '

Anet, B.F.L.T. 1964. 3-deoxyglucosuloses (3-
deoxyglycosones) and the degradation of carbohy—

drates. Adv. in Caabohydrate Chem.,;2, 181.
Association of Official Analytical Chemists (AOAC).
L 11970. Official Methods of Analysis. ‘Washington D.C:
William orwitz, ed. 1 t edi on. pp. 11, 538.
Association of Vitamin Chemists, Inc. 1966. Methods

f of Vitamin Assay. New York: Interscience Publishers.
3rd edition. pp. 294, 320.

Bernheim, F., M.L.C. Bernheim and.K.M.‘Wilbur. 1948.
Reaction between TBA and the oxidation products of
cert‘in lipids. Jo BiOle Chemo 174’ 257s

Blom, L., P. Hendricks and J. Caris. 1967. Determi—
nation of available lysine in foods. Analytical
Biochem.,;;, 382. '

Boctor, A.M., AJ. Harper. 1968. Measurements of
available lysine in heated and unheated foodstuffs
by chemical and biological methods. J. Mutr.,2g,289.

Bone, D. 1973. Importance of water activity. Food
T.Che £1, 71o

Briggs, D.A. and.L.R. Bryant. 1953. The TBA test for
butterfat oxidation. Canadiaa J, Tech. ,aa, 138.

Bruno, D. and K. J. Carpenter. 1957. {Modified proce-
dure for the estimation of available lysine in food
proteins. Biochg. J.‘ _§_7_, 139.

Buttkus, H. 1967. Reaction of myosin with malonalde-
hyde. J. Food Science. .33: 432.

Caldwell, B.F. and B. Grogg, 1955. Application of the
TBA test to cereal and baked products. Food Tech.,g,
185.

S2

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

53

C‘lhou.’ W.K., Fame H.pburn “d ".3. Br‘u‘y' 1960e
Availability 01 lysine in IhCIt, £10112, bread Ind

gluten. .1, "3.1.3.:- _7_9_, 337.

Carpenter, K.J. 1960. Estimation of the available
lysine in animal protei foods. Biocag. .J. 77,
604.

Carpenter, K.J. and 0.1!. Bllinger. 1955. Estimation
of available lysine in protein concentrates. Bao-
chem. J. A, XI. '

Carpenter, K.J. and 8.3. March. 1961. Availability
of lysine in groundnut biscuits used in treatment

of kwashiorkor. Brat. J. Mug: 3.3, 403.

D0n°.°, Go. 1.1!. I‘d-'1" Dog. “111.: ma P.R. P‘YIICo
1962. Effect of heat treatment on the nutritive
value of proteins: chemical and balance studies.

J: Sci. Fooa Er. 3.3: 192.

Dulkin, 8.1. and JJ. Friedemm. 1956. Role of dehy-
droascorbic and dehydroreductic acids in the browning

reaction. Fooa Res: 3,1,, 519.

Dunkley, W.L. and W.G. Jennings. 1951. A procedure
for application of the TBA test to milk. J: ”31:52
Sea. 35, 1064.

lllis, (LP. and J. Honeyman. 1955. Glycosylamines.
EVe 13 carbOhxgstCI ewe fl, 95o

lskin, N.A.M., H.M. Henderson and R.J. Townsend. 1971.
Bioch%sa§y of foog. New York: academic Press,
99- -

Ruler, H.V. and H. .Hasselquist. 1952. Mechanism of
d.cubaxy1‘ti°n s Me A. 4 79 e

Euler, H.V. and H. Hasselquist. 1953. Ascorbic acid,
dehydroascorbic acid and dioxogulonic acid. 555;
Remi. a, 49.

Ford, J .8. 1964. A microbiological method for assessing
the nutritional value of proteins. 3. Further
studies on the measurement of available amino acids.

héte J: Huge a, 44’s

Cup“, J.D., Adi. DmOuny' 1.3. KW“ ma Cele RlVChjwe
£958. Biological availability of lysine. Ja ESE.
4, 259.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

54

Guzman Barron, B.S., R.H. DeMeio, F. Klemperer. 1936.
Studies on biological oxidations. V. COpper and
hemochromogens as catalysts for the oxidation of
ascorbic acid. The mechanism of the oxidation.

J. Biol. Cheaa ,aaa, 625.

Guthneck, B.T., B.A. Bennet and B.S. Schweigert. 1953.
Utilization of amino acids from foods by the rat.
II. Lysine. J. nutr.,gg, 289.

Handwerk, V., E. Bujard and J. Mauron. 1960. Reduction
of aromatic nitrogeneous groups during acid hydroly-
sis in the presence of carbohydrates and its bearing
on the colorimetric determination. Biochem. J.

_7_§_, S4P.

Henry, K.M., S.K. Kon, C.H. Lea and J.C.D.White.
1948. Deterioration on storage of dried skim milk.

J. Dairy Res. ,;§, 293.

Hodge, J.E. 1953. Chemistry of browning reactions in
model systems. J. Ag . Food Chem. .1: 928.

Hodge, J.E. 1955. Amadori re-arrangement. Adv. In
Carbohydrates Chem. .;9, 169.

Hodge, J.E., B.S. Fisher and E.C. Nelson, 1963.
Dicarbonyls, reductones and heterocyclics produced
by the reactions of reducing sugars with secondary
amino salts. Amer. Soc. Brew.Chem. Proc. pp 84.

Hodge, J.E. 1967. Non-enzymatic browning. Chemistry
and Physiology of Flavours. H.W. Schultz, ed.
Westport, Conn.: Avi Publ. Co. pp. 465.

Joslyn, M.A. and G.L. Marsh. 1935. Browning of orange
juice. Ind. Eng. Chem..gl, 186.

Joslyn, M.A. 1957. Role of amino acids in the browning
of orange juice. Food Res. .ag, 1.

Kakade, M.L. and 1.8. Liener. 1969. Determination of
available lysine in proteins. Analytical Biochem.
31, 273.

Karel, M. and T.P. Labuza. 1968. Non-enzymatic brown-
ing in model systems containing sucrose. J. Agr.
Food Chem. ,;9, 717.

Kato, H. 1962. Chemical studies on amino-carbonyl
reactions. I. -Iso1ation of 3-deoxy-hexosones
formed by browning degradation of N—glycosides.

Agr. Biol. Chem. ,;9, 187.

A 39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

55

Kato, H. 1963. Chemical studies on amino-carbonyl
reactions. II. Identification of D—glucosone formed
by oxidative browning degradation of Hen-glucoside.

ME: Biol. game A1, 461e

King, R.L. 1962. Oxidation of milk fat globule membrane
material. I. TBA reaction as a measure of oxidized
flavor in milk and model systems. J, DEE Sci. _4_§_,
1165.

Kuiken, M.A. and C.A. Lyman. 1948. Availability of
some amino soids in some foods. J. Hug. 92, 359.

Kuiken. M.A. 1952. Availability of the essential
amino acids in cottonseed meal. J, nu§£.,gg, 13.

Hurts, G.W., B.F. Price and S. Patton. 1951. Studies
on z-TBA as a reagent for detecting milk fat oxi-
dation. Paper M-38. Presented at the 46th Annual
Meeting of the American Dairy Science Association.
Knoxville, Tenn. Abstracted in J, ”Bi-.12 Sci. A1, 484.

wae '1'.W., 13.3. ”1.1. and HeSe OICOtte 1965e R.‘Cti-
vity of malonaldehyde with food constituents. _-_I_,_

3°95 SC e 3.0-, 808.

Labuza, T.P. ,1968. Sorption phenomena in foods. Food
TCChe 2;, 263e

Labuza, T.P., H. Tsuyuki and M. Karel. 1969. Kinetics
of linoleate oxidation in model systems. J, Agr.

011 Chaa. Soc. ,gg, 406.

11ml" Tops, 8.3. Tmm‘u “4 ll. m‘1e 197°e
Intermediate moisture foods. Food Tach. 33,, 543.

L‘lmme To. M.A. Joslyn .114 0.0. ChinCh..t.re
1958. Mechanism of browning of ascorbic acid, citric

acid and glycine system. J, m. Fooa Chg. g, 135.

Lea, C.H. and.R.S. Hannah. 1949. Studies of the reac-
tion between proteins and reducing sugars in the
“dry“ state. I. The effect of the activity of
water, pH, temperature on the primary reaction be-

tween casein and glucose. Biocag, Bioahys, Acta
_3_, 313.

Lea, C.H. 1952. Food Science. A symposium on quality
and preservation of foods. Univ. Press. Cambridge,
6.8.

Lea, c.a., L.J. Parr and K.J. Carpenter. 1958. Chemical
and nutritional changes in stored herring meal. 1.

Brite ‘Je Hug. a, 297m

56

52. 10.8, CeHe’ LeJe Put ma KeJe Carpentdr. 1960. Cth‘l
and nutritional changes in stored herring meal. 2.
Brit. J. Nutr. ,;3, 91.

53. Lewis, V.M. and C.H. Lea. 1950. A note on the relative
rates of reaction of several reducing sugars and
sugar derivatives with casein. .Biochaa. Bioghys.
ACtae i, 5320

54. Maillard, L.C. 1912. Action des acides amines sur les
sucres; formation des melanoidines par voie metho-
dique. C.R. Acad. Sci. .agg, 66.

55.

Maloney, J.F., T.P. Labuza, D.H. Wallace and.M. Karel.
1966. Autoxidation of methyl linoleate in freeze-
dried model systems.

I. Effect of water on the
sutocatalyzed oxidation. J. Food Sci. ,aa, 878.

56. Markower, B. and S. Myers.

1943. A new method for
the determination of moisture in dehydrated vege-
tables.

Proc. Inst. Food Technologists. 4th Con-
ference. pp. 156.

57. Martinez, F. and T.P. Labuza. 1968. Effect of moisture
concentration on rate of deterioration of freeze-
dried salmon. J. Foog Sci. ,1;, 241.

58. Mauron, J.

E. Mottee, B. Bujard and R.S. Egli.
Availabil

1955.
ity of lysine, methionine and tryptophane
in condensed milk and milk powder.

In vitro diges-
tion studies. Arch. Biochem. Bioahys. ,aa, 433.
Mills, M.B., D.M. Damron and J.H. Roe, 1949.

Ascorbic
acid, dehydroascorbic acid and diketogulonic acid.
Anal. Chfim‘.2$: 107.

59.

60. GuaterhOUt, L.E., C.Re Grau and Bone Lundh01me

1959.
Biological availability of amino acids in fish meals
and other protein sources.

 

 

J. Nutr..§g, 65.
61. Partridge, S.M. and R.C. Brimley. 1952. Displacement
chromatography of synthetic ion exchange resins.
8. A systematic method for the separation of amino
acids. Biochem. J..§;, 628.
62. Pattontws, andaG; W. Kurtz; . 1951. 2-TBA as a reagent
for detecting milk fat oxidation. J.Dairy Sci. 23,
669.
63.

Patton, A.R. 1950.

Present status of heat processing
damage to protein foods. _§utr. Rev..§, 193.

 

64. Patton, S. 1955. Browning and associated changes in
milk and its products: a review.
457.

Jo 1331;! SCie lg,

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

57

Penney, J.R. and 8.8. Zilva, 1943. Chemical behavior

of dehydro
Biochem. J.

Posati, L.P.,
P‘llmSChe

L-ascorbic acid in vitro and in vivo.
31, 403.

V.H. Holsinger, R.D. DeVilbiss and.M.J.
1972. Factors affecting the determina-

tion of available lysine in whey with 2, 4, 6, tri-
nitrobenzene sulfonic acid. J. Daiay Sci.,§§, 12.

Potter, N. Nov. 1970. Intermediate moisture foods:

principles
_4_, 38.

and technology. Food Product Develogaent

Rao, S.R., F.L. Carter and V.L. Frampton. 1963.
Determination of available lysine in oilseed meal

proteins.

Reynolds, T.
browning.

Anal. Chem. 3;, 1927.

1963. The chemistry of non-enzymatic
I. The reaction between aldoses and

amines. Adv. infiFood Res..;g, 1.

Reynolds, T.
browning.

Reynolds, T.

1965. The chemistry of non-enzymatic
II. Adv. in,Food.Res. ,;3, 167.

1969. Non-enzymatic browning. Sugar—

amin interactions. In Syagosium on Fooasx Caabo-
hydrates and their roles. H. . Sc ultz, .F. Cain
and R.W.‘Wrolstad, eds.‘Westport, Conn.: The Avi

Publishing’

Co., Inc. pp. 219.

Rice, 3.3. “d J.F. B.uke 1953. Th. C££CCta o: he‘t
upon the nutritive value of protein. Adv. in Food
Res. _4_, 233.

ROCkl‘nd’ Lane

1969. Water activity and storage sta-

bility. FoodfiTech. .ga, 1241.

Roe, J.H. 1936. Determination of ascorbic acid as
furfuralcand a comparison of the results obtained
by this method and by indophenol titration. ,7,

Biol. Chem.

Roe, J.H. and

116, 609.

C.A. Kuether. 1943. Determination of

ascorbic acid in whole blood and urine through the
2, 4 dinitrophynyl-hydrazine derivative of dehydro-
ascorbic acid. J. Biol. Chem. 147, 399.

Salwin, H. 1959. Defining minimum moisture contents
for dehydrated foods. Food Tech. .li: 594.

Sanger, F. 1945. The free amino groups of insulin.

lgiochem. J.

33, 507.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

58

Schoebel, T., S.R. Tannenbaum and T.P. Labuza. 1969.
Reaction at limited water concentration. 1. Sucrose
hydrolysis. J. of Fogd Sci. .;1, 324.

Schonberg, A., R. Monbaster and.A. Mostafa. 1948.
Degradation of alpha amino acids to aldehydes and
ketones by interaction with carbonyl compounds.
J. Chem. Soc., London. pp. 176.

Schonberg,.A. and R. Monbaster. 1952. The Strecker
degradation of amino acids. Chem..Rev.,§Q, 261.

Schweigert, B.S. and B.T. Guthneck. 1953. Utilization
of amino acids from foods by the rat. I. Methods
of testing for lysine. J. Nutr. .22, 277.

Scott,‘W.J. 1957. ‘Water relations of food spoilage
organisms. 9.9.2. Food Res. 1, 83.

Scott, J.A. and H. Smith. 1966. .Microbial assay of
protein quality with Tetrahyaena ayriformis w. 4.
Measurement of available lysine, methionine, argi—
nine, histidine. Brit. J. Nutr. ,;Q, 663. ‘

Sharp, J.G. 1960. Non-enzymatic browning deterioration
in dehydrated meats. Recent Adv. in Food Sci..;, 65.

Sheffner, A.L., 0.3. Fckfeldt .nde. Specter. 1956.
The pepsin-digest—residue (PDR) amino acid index of
net protein utilization. J. Nutr. .QQ, 105.

Simon, H. and G. Heuback. 1965. Formation of alicyclic
and open chain nitrogeneous reductones by reaction of
secondary amin salts on monosaccharides. Chem.g§a§.
£31, 3703.

Singh B., C.R. Dean and S.M. Cantor. 1948. Role of.
S—(hydroxy-methyl)~furfural in the discoloration of

sugar solutions. J. gaer. Chem. Soc. 13, 517.

Sinnhuber, R.C. and T.C. Yu. 1958a. Characterization
of red pigment formed in the Z-TBA reaction. Food
Research. .aa, 626.

Sinnhuber, 3.0. and T.C. Yu. 1958b. 2—thiobarbituric
acid in fish. II. Quantitative determination of
malonaldehyde. FoodgTech. lag, 9. '

Sood, V.C. 1972. Rapid water activity evaluation of
dehydrated dairy products. M.S. thesis, Michigan
State UniverSity.

Spark, A.A. 1969. Role of amino acids in non-enzymatic
browning. J, Sci. Food.Agric. .gg, 308.

92.

93.

94.

95.

96.

97.

98.

99.

59

Stadtman, S.R. 1948. Non-enzymatic browning in fruit
products. Adv. in Food Res. _1,, 325.

Tul‘inB, BeGe ma 3.“. W‘tt‘e 1960e n‘lon‘ldChYd.
production during the controlled oxidation of pure

unsaturated fatty acids. J. Aaer. Oil Chem. Soc.
37, 403.

T‘tum. JeHe. Pogo Sh.w ma R.S. BMe 1969e Dfiqr‘d“
tion products from ascorbic acid. J. Agric. Food
Chemo a. 38.

Taylor, AeA. 1961. Determination of moisture equili—
bria in dehydrated foods. Food Tech. lg, 536.

Turner, R.W., W.D. Paynter, R.J. Montie, M.W. Bessert,

‘..G.M. Struck and.F.C. Olson. 1954. Use of 2-TBA
reagent to measure rancidity in frozen pork. Foog
TQChe 9, 326a

undQflOOd, JeCm, HeGe Lmto ma C..O. Willits. 1959s
Browning of sugar solutions. 3. Effect of pH on
the color produced in dilute glucose solutions con-
taining amino acids with the amino group in differ-
ent positions in the molecule. Food Res._gg, 181.

Yu, T.C. and R.C. Sinnhuber. 1957. 2-TBA method for
the measurement of rancidity in fishery products.
FOOd T.Ch e 11,104e

Yu, T.C. and R.C. Sinnhuber. 1962. Removal of inter-
fering pigments in determining malonaldehyde by the
2-TBA reaction. Food Tech. ,;§ (6), 115.

MICHIGAN STATE UNIVERSITY LIBRARIE

mill! |||||||l |||