 

1
.-

.._ i:
3‘ “
= g

a... imw
$93.
5

.1 away»;

.. 5:3...

“ﬁlm:

.wwi... ..
.v. .55... . .
. . . ”Wm?” . .3,
. .{n{1.:£ry:.. [ﬁg-1
aréﬁm... r9. .wﬁ
{a

= .
exsris...
, I;
. Lari,
...&..w - a?!

5 .,
9.
:15...

an

.
e . .
.. .. . .
z: 2.. 1%qu .wgmuﬁ .. .,...:.?.....
E‘ihk... hug: a.
£2.32. . .
.3); t

. i is.
. .rh. .n...
{3.1.

..... a
3... 2 dd...
.33.... .. ..
1... .. i...
a: .3: 2.1.3».
. “Era? ’r... ‘3‘]...
J: o ‘55; L§wl
243.5...

uﬂzhﬁuslazuvht «Fitz!
oddityrim. 5.x?

:0. 1‘.tn8ol.h§1|5!:|«
IKE Eclizpflizl: ‘
It! ‘69! x‘ .‘t.
it} .1.- a...

til‘IOlnﬁlvx
:13:

‘37:

 

la 11 v«.\¥h.r.§. a
:\qx.l.i...i3’!is Ii.
9A. .N\xv

 

if}... .9.
I: 1;}. \
.2 p. I .. .

IV... .V

gage._.§..%....$...,...mw if...»

(.371... .

 

 

TPEms

5)

LIBRARY
Michigm‘ State
Universiw

This is to certify that the
thesis entitled

Effect of Formulation and Particle Size on the
Rheological Properties of Salad Dressing

presented by

Julie Krista Branch

has been accepted towards fulﬁllment
of the requirements for

M.S. degree in We

Jam FJ%%,

Major professor

Date Jam (92. 300/

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRCIDateDuepss-p. 15

EFFECT OF FORMULATION AND PARTICLE SIZE ON THE
RHEOLOGICAL PROPERTIES OF SALAD DRESSING

By

Julie Krista Branch

A THESIS

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

MASTER OF SCIENCE
Department of Food Science and Human Nutrition
2001

pm
(161

vis

am
of 1
pic
prc
vis
6.5

par

the

ABSTRACT

EFFECT OF FORMULATION AND PARTICLE SIZE ON THE
RHEOLOGICAL PROPERTIES OF SALAD DRESSING

By

Julie Krista Branch

The effect of formulation and particle size on the rheological properties of
pourable salad dressing was studied utilizing rotational visoometry. Slightly time-
dependent and strong shear-thinning behavior was observed. The apparent
viscosity at a shear rate of 25 s'1 was strongly affected by xanthan gum.
Optimum viscosity was achieved with a formulation having 0.45% xanthan gum
and 1.5% egg yolks. Egg yolks provided emulsion stability, and increasing levels
of egg yolk had a minor influence on viscosity. Temperature was also shown to
produce minor changes in the viscosity at 25°C and 5°C. Increases in the
processing rate (differential pressure over a hydroshear) did not strongly affect
viscosity, but had a strong effect on the particle sizes ranging from 5.17 pm to
6.50pm. Xanthan gum and egg yolks were shown to only slightly influence the

particle size compared to the process rate. No correlation was found between

the particle size and the viscosity.

DEDICATION

To my ryan.

ACKNOWLEDGEMENTS

Thank you Dr. Steffe for all of your guidance and patience.
Special thanks to my committee Dr. Ustunol and Dr. Uebersax,
to Suzanne Case and Raju Borwankar from Kraft Foods,
my POST family, my ‘coach’ Jeff Feneley,
my family and friends and to Ryan for his continued support.

LIE

LIE

KE

2.5

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LiST OF TABLES ......................................................................................................... vii
LIST OF FIGURES ______________________________________________________________________________________________________ viii
KEY TO SYMBOLS OR ABBREVIATIONS ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, x
Chapter 1 —INTRODUCTION
1.1 Salad Dressings ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 1
1.2 Consumer Market--- .......................................... 2
1.3 Objectives," 4
Chapter 2 —LITERATURE REVIEW
2.1 Emulsion Systems and Stability ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 5
2-2 Ingredients ................................... 8
22-1 Fat I Oil ........................................................... 9
2.2.2 Emulsiﬁers and Stabilizers ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 11
2.2.3 Egg Yolks ,,,,,,,,, 14
2.2.4 Proteins .................... 17
22-5 Xanthan gum ........................................... 18
2.3 Manufacturing of Dressings 22
2.4 Rheology of Salad Dressings and Methodology of Fluids 28
2.4.1 Timefl'emperature Dependence 31
2.4.2 Steady State Shear Measurements __________ 37
2.4.3 Unsteady State Shear Measurements 40
2.4.4 Yield Stress 43
2.5 Particle Size Analysis _______________ 46
2-5-1 Measurement ................................................ 43
2.5.2 Control of Particle Size 51
2.5.3 Size and Composition ,,,,,,,,,, __ _ 52

 

 

Che

3.1

3.2
3.3

3.4
3.5

Ch.

 

4.2

Cl‘.

Ch

Bil

 

 

 

TABLE OF CONTENTS (cont’d)

Chapter - 3 MATERIALS AND METHODS

 

 

 

3-1 Formulas / Samples .............................................................................................. 54
3.2 Processing ............................................................... 56
3.3 Rheology ................................................................................................................. 61
3.4 Particle Size ____________________________________________________________________________________ 61
3.5 Statistical Design and Analysis ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 67

Chapter 4 - RESULTS AND DISCUSSION

 

 

 

 

 

 

 

 

4.1 Inﬂuence of Formula and Processing on Rheological Behavior ,,,,,,,,,,,,,,,, 68
4.1.1 Time dependent behavior ______________________________________________ 72
4.1.2 Xanthan Gum / Egg Yolks ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 80
4.1.3 Processing Rate _______________________________________________________________ 88
4.1.4 Temperature__ _ _______________________________________ _ 92
4.2 Particle Size lnﬂuence__ _ .. ___________ _ 94
4.2.1 Inﬂuence of pressure drop _ ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 100
4.2.2 Inﬂuence of particle size on viscosity ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 101
Chapter 5 - SUMMARY AND CONCLUSIONS ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 104
Chapter 6 — FUTURE RESEARCH ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 106
Bibliography ................................................................................................................... 108

vi

LIST OF TABLES

TABLE 1 Ingredients typically found in food emulsions.

TABLE 2 Typical fat content of dressings and sauces.

TABLE 3 Sources of emulsiﬁers.

TABLE 4 Xanthan gum functionality.

TABLE 5 Comparison of different types of emulsifying devices.

TABLE 6 Model salad dressing samples formulation by percentage of total
weight

TABLE 7 HAAKE \frscotester © 550 MVI Speciﬁcations

TABLE 8 Initial Brookﬂeld viscosity, pI-I, titratable acidity and salt analysis of the
model salad dressing samples

TABLE 9 Rheological analysis of model salad dressing formulations at 5°C

TABLE 10 Rheological analysis of model salad dressing formulations at 25°C

vii

 

 

FIGL
FIGI

FIGI
ﬁne

FIGI
FIG‘
FIG
FIG

FIG
sarr

FIG
in p

FIG
IOp

FIG

FIG
det-

FIG
f0n

You

FIC
diff
Fig
pro

F|(
Drc

LIST OF FIGURES
FIGURE 1 Emulsion instability
FIGURE 2 Breakage of cylinders of liquid into small droplets caused by shear

FIGURE 3 High pressure valve homogenizer, used to produce emulsions with
ﬁne droplet size

FIGURE 4 Timeodependent behavior of ﬂuids

FIGURE 5 Typical curves for time-independent behavior ﬂuids
FIGURE 6 Apparent viscosity of time independent ﬂuids
FIGURE 7 Static and dynamic yield stress

FIGURE 8 Overview of pilot plant salad dressing process used to manufacture
samples

FIGURE 9 Cross section of in-line emulsiﬁcation system hydroshear unit utilized
in pilot plant salad dressing system

FIGURE 10 MVI concentric cylinder testing apparatus showing to bob recessed
top and bottom to minimize end effects

FIGURE 11 Horiba principle laser measurement ﬂow diagram

FIGURE 12 Conﬁguration of the Horiba LA-500 benchtop system utilized to
determine lipid particle size

FIGURE 13 Rheological bahavior of commercial samples in comparison to
formula 1, containing 0% egg yolk and 0.15% xanthan gum; formula 5, containing
1.5% egg yolk and 0.30% xanthan gum; and formula 9, containing 3.0% egg
yolks and 0.45% xanthan gum. (All at 140 psi and 25°C)

FIGURE 14 Time-dependent ﬂuid behavior at 25°C before shear sweep at
different processing levels

Figure 15 Time dependent ﬂuid behavior at 25°C after shear sweep at different
processing levels.

FIGURE 16 Effects of aging: comparison of measurements of fresh and aged
product at two processing levels.

viii

LIST OF FIGURES (Cont’d)

FIGURE 17 Effects of increasing xanthan gum at 1.5% egg yolks and 220 psi at
25°C.

FIGURE 18 Effects of Increasing egg yolk at 0%, 1.5% and 3.0% and increasing
xanthan gum at 220 psi at 25°C.

FIGURE 19 Effect of processing rate increase 140, 180 and 220 psi with
increasing egg yolk and xanthan gum level of 0.30% at 25°C.

FIGURE 20 Effect of increasing processing rate 140, 180 and 220 psi with
increasing xanthan gum 0.15%, 0.30% and 0.45% and egg yolk level at 3.0%.

FIGURE 21 Effects of increasing process rate with 0% egg yolks and increasing
xanthan gum 0.15%, 0.30% and 0.45%at 5°C.

FIGURE 22 Inﬂuences of temperature at 0.15% and 0.45% xanthan gum and 0%
and 3.0% egg yolk levels.

FIGURE 23 Effect of increased differential pressure in the hydroshear on particle
size.

Figure 24 Effects of increasing egg yolk and xanthan gum levels on particle size
at 180 psi.

Figure 25 Effect of increased levels of egg yolk, at a constant xanthan gum level
of 0.15% on particle size.

Figure 26 Viscosity versus particle size effects with processing rate increases.

KEY TO SYMBOLS OR ABBREVIATIONS

shear stress (Pa)

O
II

K = consistency coefficient (Pa 5")

.< a
ll

shear rate (1ls)
n = ﬂow behavior index (dimensionless)
11 = apparent viscosity (Pa s)
t = time (s)
00 = yield stress (Pa)

TI. = complex viscosity (Pa s)

J = shear creep compliance (Pa'1)
Jo = instantaneous compliance (Pa‘1)
J1= retarded compliance (Pa‘1)

Ar... = retardation time (s)

p = Newtonian viscosity (Pa 5)

G' = shear storage modulus (Pa)

 

 

 

1.1

can
form

beh

in n
sale
ing:
thic

Owr

sys
of I
am
of

bet
effg
Ch;

Cor

CHAPTER 1
INTRODUCTION
1.1 Salad Dressings

A pourable emulsion salad dressing is deﬁned as a ﬂuid system, which
can be made to stream or ﬂow continuously or profusely due to gravitational
forces acting upon the system. The inﬂuence of particle size on the rheological
behavior of a pourable model system impacts consumer acceptance.

The standard of identity for pourable emulsion salad dressings, as stated
in the code of federal regulations, 21 CFR 161 (1999) provides a deﬁnition of
salad dressing: Oil in water emulsion manufactured from oil, egg yolks, acidiﬁed
ingredient and optional ingredients (salt, nutritive sweetener, stabilizer and
thickener, acid). Mayonnaise, salad dressing and French dressing all have their
own standard of identity.

Salad dressing is a thermodynamically unstable two-phase emulsion
system. It contains two immiscible liquids, one dispersed in the other in the form
of ﬁne droplets. The surrounding liquid called the continuous or external phase, is
aqueous. Oil is the dispersed or internal phase. Oil in water emulsions consists
of ﬁne droplets dispersed in an aqueous phase. The droplet size averages
between 0.1 and 100pm. Size and aggregation of these oil droplets have an
effect on the ﬂow behavior and rheological characteristics. Knowledge of these
characteristics assists in product development, production, packaging, and

consumer acceptability.

 

 

COIIIE

 

fully
agitai.
layen
applil

mode

emu
ﬁner

Woe

COW

may
is c
I0 1

leg

1.2

Th

ho

Pourable dressings include a variety of compositions, ﬂavors and oil
content. They are manufactured and formulated into two basic types. The ﬁrst is
fully emulsiﬁed which provides long-term stability, and requires little mechanical
agitation. The second type is an emulsion that separates into oil and aqueous
layers and has short-term stability, upon which mechanical agitation must be
applied to suspend the particles in solution. The current study focused on a
model system exhibiting long-term stability.

Lipid droplets are the main particle component of a salad dressing
emulsion system. These droplets variety in size and are covered with an
interfacial ﬁlm. The width of this ﬁlm ranges from 100 — 200A, and varies with the
type of dressing (Ford and others 1997). The aqueous or continuous phase
contains spices and other plant materials, which provide ﬂavor.

The continuous phase of some dressings contains starches and gums. In
mayonnaise, the continuous phase is composed of egg granules. In dressing it
is composed of hydrocolloids. These particles adhere to the interfacial ﬁlm and
to each other, forming a protein network. The formation of a protein network

results in an increase in viscosity and stability of the emulsion.

1.2 Consumer Market

Salad dressings have been part of food history for thousands of years.
The Babylonians used oil and vinegar for dressing greens over 2,000 years ago.
As early as the 1920s many of the popular brand names of today became
household names. In 1896 - Joe Marzetti opened a restaurant in Columbus,

Ohio and sewed his customers a variety of bottled dressings. In 1912 — Richard

 

 

 

Helir.
Che
the

(Ass

 

Rana
Iollov|

(Ass

the

DOUI
shel
Sau

I06

the

C0

Hellman, deli owner in New York sold “blue ribbon” mayonnaise. In 1925 - Kraft
Cheese Company purchased several regional mayonnaise manufacturers and
the Milani Company. This initiated Kraft’s pourable dressing business
(Association for Dressing and Sauces 2000).

Ranch dressing was introduced in the mid 1970s by Hidden Valley
Ranch© as a dry mix. Ranch style dressing holds reign as America’s favorite
followed by Italian, creamy Italian, Thousand Island, French and Caesar
(Association for Dressing and Sauces 2000).

As a consumer peruses down the grocery market aisle and travels through
the produce section, they can ﬁnd an overwhelming number of choices of
pourable salad dressings. These include composition, ﬂavor, size, brand, texture,
shelf stability and packaging. According to the Association for Dressings and
Sauces the amount of salad dressing sold grew from 6.3 million gallons in 1950
to 60 million gallons in 1997.

The consumer demand for product variety is a real issue in food stores.
Large food stores provide the consumer with over twenty brand names and ten
basic ﬂavors in various types, including shelf stable to refrigerated dressings in
its pourable dressings selection. This emphasizes the importance of competitive
products in the market place and drives the food industry fonlvard to develop new
and improved products.

Reduced fat products and new ﬂavors continue to be developed to satisfy
the nutrient, textural, sensory and ﬂow characteristics of a pourable dressing that

consumers accept. Average consumers judge and purchase products based on

 

 

 

bmnd
Fbvr
deper

These

brand loyalty, ﬂavor, taste, price, composition and appearance of the product.
Flow characteristics, emulsion stability, particle size, shelf life, temperature
dependency and textural attributes all contribute to consumer acceptance.
These product characteristics can be measured using rheological methodologies
to produce a consumer accepted pourable dressing. In this study, the impact of
particle size, due to formulation and production conditions, on the physical
properties of a pourable emulsion model, was evaluated using various

rheological methodologies.

1.3 Objectives

The objectives of this research were to do the following for a pourable salad
dressing:

1) Assess the effects of temperature on the rheological properties.

2) Evaluate the inﬂuence of different formulations and processing conditions on

the rheological properties.

3) Evaluate the inﬂuence of different formulations and processing conditions on

the lipid particle size.

4) Evaluate the relationship between particle size and rheological properties.

 

 

2.1

foo
imr
imr
var
99

CIE

tin
a I

an

thi

ac

Ire

Dr

CHAPTER 2
LITERATURE REVIEW

2.1 Emulsion Systems and Stability

Today's food supply is ﬁlled with natural and manufactured emulsions. A
food emulsion is deﬁned as: “A homogeneous dispersion of two dissimilar
immiscible liquid phases (Stauffer 1996)”. Dressings consist of the two
immiscible liquids of oil and water. Typical food emulsions display of a wide
variety of appearances, textures, taste and shelf life. Examples include milk and
eggs, which exist in nature. Manufactured emulsions include ice cream, coffee
creams, soups, sauces, mayonnaise, butter, margarine, and salad dressing.

Dressing emulsions are thermodynamically unstable and, given enough
time, these emulsions will separate. Emulsion stability ranges from seconds, for
a Caesar dressing to years for mayonnaise. Emulsions can be produced which
are stable for a reasonable period of time to make the product acceptable: for
minutes, hours, days, weeks, months or years. The introduction of emulsiﬁers to
the emulsion formula prior to homogenization allows for kinetic stability to be
achieved.

Emulsiﬁers are surface active molecules, which absorb to the surface of
freshly formed droplets during the homogenization process. Emulsiﬁers form a
protective membrane around the oil droplet and stabilize the emulsion against
aggregation. Surface active compounds, emulsiﬁer surfactants, position
themselves in the oil I water interface where they act to lower surface or

interfacial tension.

 

 

 

 

enc
ch;
ma
res

ﬂox

an

ﬂc
CO
de
SU

dr

Fl.

Us

Emulsions, due to their instability, will undergo phase separation given
enough time. Loss of emulsion stability results in an undesirable product due to
changes in the physical and rheological characteristics. Emulsions can be
manufactured to hold stability over a speciﬁed shelf life. This forced stability
results in a product that maintains acceptable appearance, texture, ﬂavor, and
ﬂow characteristics and ﬂavor desired and accepted by the consumer.

An emulsion can become unstable due to a number of different types of
physical and chemical processes. Physical instability results in an alteration in
the spatial distribution or stmctural organization of molecules. Chemical
instability results in an alteration in the chemical structure of molecules
(McClements 1999), and is the result of oxidation or hydrolysis (Nawar 1996).
Good stability implies that there is no change in the size distribution or the spatial
arrangement of droplets over the designated stability ﬁme of the emulsion
(Dickinson 1994).

Physical or kinetic emulsion instability includes, sedimentation,
ﬂocculation, coalescence or Ostwald ripening of lipid droplets within the
continuous phase, (Figure 1). Creaming (rise of droplets) is a result of the
density difference, i.e., and the droplets having a lower density than the
surrounding liquid. Sedimentation is the result of density difference in which the
droplets move downward.

Flocculation and coalescence both involve the aggregation of droplets.
Flocculation occurs as droplets close to each other form a ﬂoc as a result of van

der Waal’s and ionic forces. Flocculation occurs without the rupture of the

 

 

 

 

 

 

 

 

Figl

 

Kinetically
Stable
Emulsion

 

 

Phase

Inversion

 

 

 

 

Ii‘
AA

Creaming Sedimentation

 

 

 

 

F locculation

 

 

Coalescence

 

Figure 1 - Emulsion instability (McClements 1999)

 

 

 

 

prc
F lc
bu
otf

inﬁ

ﬁln
drc
ex;

ma

ten

will

2.2

saf
ion

on,

protective stabilization layer at the oil I water interface (Dickenson 1994).
F locculation results in aggregation, which typically is considered kinetic instability
but, can be beneﬁcial to an emulsion due to its effects on ﬂow behavior (Ford and
others 1997). The size and structure of the ﬂocs within an emulsion have a large
inﬂuence on the rate at which the droplets cream.

Coalescence involves the combination of two droplets into one. The thin
ﬁlm of the continuous phase between the two droplets breaks and forms a large
droplet. Ostwald ripening is a process whereby large droplets grow at the
expense of smaller ones. This physical change occurs due to the solubility of the
material in a spherical droplet increasing as the size of the droplet decreases
(Dickenson 1994).

Understanding the factors effecting stability including formulation, time,
temperature, processing conditions, mechanical agitation and storage conditions

will result in better control of emulsion stability and product integrity.

2.2 lngredlents

Ingredients play a key role in the ﬂow behavior, shelf stability, product
safety, product identity and consumer acceptance of a product. The main
formulation ingredients, which make up a salad dressing emulsion system, are
oil, emulsiﬁers and stabilizers. All of these ingredients play a functional role,

which dictates the structural and textural properties of the dressing system.

Understanding the role of each ingredient and how it effects the
structure, stability, particle size and rheology of the system is very important as
formulas are changing in today’s competitive market to met consumer and
industry pricing needs while maintaining a texturally similar and consumer
acceptable product. Food emulsions are compositionally complex materials that
contain a wide variety of different chemical constituents (Table 1).

A wide variety of natural and synthetic or modiﬁed ingredients are
available as ingredients for the formulation of a dressing system. Each
ingredient will provide unique beneﬁts and functional properties to create a
desired system. Ingredients need to be selected to create the desired
appearance, rheological characteristics, mouthfeel and product stability. This
selection is continuous through the product storage, shipping shelf life and
consumption. The appropriate selection of ingredients is critical to successful

creation the desired ﬁnished product (Mc Clements 1999).

2.2.1 FatIOil

Fats and oils or lipid compounds play an inﬂuential role on the
organoleptic and textural characteristics of an emulsion system. To form the
desired emulsion, we must understand the functionality of a lipid during
processing, manufacturing and storage. The consumer, during use and

consumption, ultimately judges this functionality relationship.

Table 1 — Ingredients typically found in food emulsions (McClements 1999)

 

 

 

 

 

 

Macrocomponents Microcomponents
Water Emulsiﬁers

Lipids Minerals
Proteins Gums
Carbohydrates Flavors
Colors

Preservatives
Vitamins

 

 

10

 

 

Sah
mes
typx

rang

 

 

122$

sun:

 

199‘
lNOd
deve
Phys
Qual
aere

She:

 

enh

USS

tell:

hue

 

Salad dressing oil is a reﬁned, bleached, deodorized and winterized oil.
Salad oil has a cloud point of a minimum of 5.5 hours, with typical good quality
dressings being 15 hours (Stauffer 1996). Salad dressing formulations are
typically at a pH of 2.8-4.0. The typical fat content of dressings and sauces
ranges from 0.1% for ketchup up to a maximum of 84% for mayonnaise (Table

2).

2.2.2 Emulsifiers and Stabilizers

Emulsiﬁers also called surfactants are molecules that promote and/or
stabilize an emulsiﬁcation or dispersion of one liquid in another liquid (Stauffer
1996). They have been used in the food industry for nearly 70 years and have
produced solutions to many problems encountered during food product
development. Emulsiﬁers can provide many key functions by controlling the
physical nature of the interfaces in the emulsion. This is crucial to produce high
quality and stable systems. Structurally they promote emulsion stability, stabilize
aerated systems, control aggregation of fat globules, modify texture, promote
shelf life, and create desirable rheological characteristics. The role of the
emulsiﬁer is to promote the long-term stability of an emulsion system or to control
destabilization under shear.

Emulsiﬁers are surface-active substances. They reduce the surface
tension of two normally immiscible liquids (phases) by absorbing at the liquid

interface. Control of this interface, through formulation with emulsiﬁers,

11

Table 2 -Typical fat contents of dressings and sauces (Ford and others

 

 

 

 

 

 

 

 

 

 

 

 

1997)
Sample Percentage
(%)
Mayonnaise 75-84
Italian 50-60
Salad Dressing (Spoonable) 30-60
Blue Cheese 3040
French 36-40
Russian 30-40
Thousand Island 30-45
Italian (low calorie) 0-3
Barbecue Sauce 1-2
Ketchup 0.1-0.2

 

 

 

12

is critical in making a high quality food product. Insufﬁcient emulsiﬁer covered
droplets result in a higher tendency of the droplets to coalesce (Stang and others
1996). Emulsiﬁers maybe components of an ingredient, such as egg yolks or
additives including monoglycerides and polysaccharides. Since emulsions are
inherently unstable, emulsiﬁers are added during manufacturing. The proper
formation and stability of the emulsion is needed for stabilization at the time of
formation and for long term product stability and increased shelf life.

Emulsion systems possess minimal stability characteristics, which can be
improved by surface active agents such as ﬁnely divided solids, Iipoproteins,
mono- and di-glycerides and polysaccharides. Each works in different ways to
maintain the oil in small droplets. Some gums, for example, actually absorb to
the oil-water interface forming a ﬁlm with good droplet coalescence (Stauffer
1996).

Emulsiﬁers are used throughout the food industry to contribute to both
long and short-term stability. Those added to the dressing system are amphilic
molecules. Surface active compounds (emulsiﬁers) operate through a
hydrophilic head group that is attached to the aqueous phase, and a Iipophilic tail
attracted to the oil phase (Hasenhuettl and Hartel 1997). These hydrophilic, or
water loving, and Iipophilic, fat loving groups are made up of hyrdocarbons that
are branched, straight -chained or cyclic. The Iipophilic part of an emulsiﬁer is
usually a long chain fatty acid. This causes formation of a threadlike network
within the intermolecular structure of the emulsion causing increases in viscosity.

These groups play a major role in the type of emulsiﬁer selected for a particular

13

application. An emulsiﬁer should be chosen based on the food system. It is
common to utilize two or three emulsiﬁers in one system to achieve multiple
functions (Hasenhuettl and Hartel 1997).

According to the FDA standards of identity, salad dressings must contain
a minimum of 30% vegetable oil and up to 0.75% by weight of emulsifying
agents. Food emulsiﬁers are esters of edible fatty acids with the most commonly
used in salad dressings including polysorbates, citric acidesters of
monoglycerides and diactyl tartaric acid esters of monoglycerides. These
common emulsiﬁers come from various sources (Table 3).

Polysorbates, polyoxyethylene sorbitan esters, are a product of the
reaction of sorbitan esters with ethylene oxide. There are three types of
polysorbates permitted by the FDA for use in food systems at limited amounts:
polysorbate 60, polysorbate 65, and polysorbate 80. Each emulsiﬁer is permitted
for use in speciﬁc foods. Polysorbate 60 can also be applied to sugar type
confection coatings, shortenings, and edible oils for baking and ﬂying

applications. In this study liquid egg yolks and polysorbate 60 were used.

2.2.3 Egg Yolks

Egg yolks have long been recognized as providing functional emulsifying
properties in dressing systems. Egg yolks are the most functional compounds of
the egg for emulsifying functionality in dressings. Emulsifying activity is a result
of the lecithin protein complex, lipoproteins (Ford and others 1997). Emulsifying

properties of egg yolk are a result of the lipoprotein fractions, which attach to the

14

 

Table 3 - Sources of emulsiﬁers (Ford and others 1997)

 

Emulsiﬁer Type

Ingredients

 

Protein

Buttermilk, sourcream, skim milk, nonfat dry milk, whole
milk, sodium caseinate, whey, whole eggs (fresh, salted,
frozen), egg whites (fresh, salted, dried), egg yolks (fresh,
salted, sugared, dried)

 

Phospholipids

Egg yolks, whole milk, sour cream

 

Particle

Mustard Flour

 

Synthetic

Polysorbate

 

 

Chemically modiﬁed

 

Proplylene glycol alginate

 

15

 

oil droplets at the salad dressing oil interface. Egg yolks provide a protective

barrier on the droplets due to steric stabilization.

Egg yolks contribute to emulsion stability by both steric and particle
mechanisms. The particle mechanism is evident from the discovery of protein
particles at the interfaces between oil droplets (Ford and others 1997). Egg yolks
have an isoelectric point of 5.3. At a neutral pH, egg yolks are negatively
charged, but since most salad dressings have a pH level of 2.8 - 4.0 they are

positively charged.

Egg yolk, a rich source of phospolipid, signiﬁcantly contributes to the
stability and viscosity of salad dressing emulsions. Carrillo and Kokini, 1988,
determined through creep analysis, steady shear and particle size analysis that
emulsions are most stable in the presence of 2% egg yolk with 0% salt, or 3%

egg yolk with 2% salt.

Carillo and Kokini (1988) determined that under steady shear
measurement both egg yolks and salt increased viscosity and showed shear-
thinning behavior. Emulsions containing egg yolk and no salt, a decrease in
apparent viscosity was observed at higher shear rates. Higher levels of egg yolk
in an emulsion lead to larger aggregates formed. These aggregates disperse
under the applied shear rate. As the continuous phase, which was entrapped
between aggregates, is released due to shear, the viscosity decreased. Salt
enhances the stability of oil / water emulsions by disrupted the egg yolk granules,
leaving more active sites exposed at the interface and lowering the interfacial

tension.

16

During the same study, Carillo and Kokini, in 1988 showed that particle
size distributions changed very little, as emulsion age increased, 0 - 60 days.
This suggests that for emulsions with egg yolks, and egg yolk and salt, the
viscosity change is caused by the build up and break down of aggregates. In
samples with egg yolk concentrations greater than 2%, they did see an increase
in apparent viscosity. These results cannot be explained through coalescence
since the droplet size distribution did not change signiﬁcantly for all egg yolk
concentrations. Therefore, rheological properties are controlled by the kinetics of

aggregate formation and breakdown.
2.2.4 Protelns

Many formulations of salad dressing contain proteins that act as an
emulsiﬁer and stabilizer. Polysaccharides are also added to a majority of
emulsion systems. This protein-polysaccharides interaction at the droplet

interface inﬂuences the stability and rheology of the emulsion aqueous phase.

To be an effective emulsiﬁer the protein being adsorbed, such as egg
yolks, must protect the oil droplets against spontaneous ﬂocculation or
coalescence. The protein forms a ﬁlm around the surface of the oil droplets
resulting in stable oil I water emulsions. The physicochemical properties of the
protein, emulsion characteristics and processing parameters all affect the
emulsiﬁcation stability. The interfacial activity of a protein molecule in an
emulsion system involves: 1) the protein molecule diffusing to the interface, 2)
the protein penetrating to the interface, 3) molecules rearranging to achieve

minimum energy in the system (Ford and others 1997).

17

 

 

5mm

 

be a

emul

E‘s——

mop?

Mnct

auep

 

can t
pnne
side
oH,u
Phas

Dhas

2.2.:

saia:
af0r
9ft)“
slab
IeXh

90m

Other factors inﬂuencing protein functionality include temperature, pH, and
salts or the presence of other ions. The viscosity of the external solution or can
be altered by adjusting the net charge on the proteins which increase the
emulsifying ability of the protein. In an oil in water emulsion, the protein coated
droplets are kept apart as a result of charged and steric stabilization. A
functional protein must be soluble; therefore proteins near their isoelectric point

are poor emulsiﬁers due to low solubility.

Surfactants, molecules that contain hydrophilic and hydrophobic regions,
can be used to decrease interfacial tension and stabilize the emulsion. When
proteins are used as surfactants, they must migrate to the interface, orientates
side chain groups to polar or nonpolar areas and form a stable ﬁlm around the
oil, to function properly. This formation of a stable ﬁlm around the oil droplet
phase, is the primary key to forming a stable emulsion that exhibits minimal

phase separation.

2.2.5 Xanthan Gum

Hydrocolloids, often referred to as gums, are widely used in production of
salad dressings. The functional properties provided by hydrated hydrocolloids in
a food system include: adhesion, binding, enhancing body, inhibition of crystal
growth, emulsiﬁcation, encapsulation, coating, ﬂocculation, ﬁlm forming, foam
stabilization, gelling, thickening, suspension, stabilization for structure and
texture, whipping and fat replacement (Phillips and Williams 1995). Xanthan

gum is a commonly utilized hydrocolloid in dressing systems. Its functional

18

 

 

ben

bkne
fenn
guni
been

shoe

(ﬂucr
sﬂucr
unhs
ade
to h)

"Ice

DSeL
rAtvg
Oihs
Thus
Sehn

(Urla

beneﬁts can be seen in (T able 4).

Xanthan gum was the ﬁrst of a new generation of polysaccharides
biotechnologically produced. First made in the early 1960s, through the
fermentation of the bacterium Xanthomonas campestn’s, this biosynthesised
gum, was able to functionally compete with natural gums. Xanthan gum has
been an approved food additive in the United States since 1969 and in Europe
since 1974 (Urlacher and Noble 1997).

The primary structure of xanthan contains a backbone of 1,4-linked p-D-
glucose with side chains. These side chains contain mannose and one
glucuronic acid and represent 60% of the total molecule. Half of these mannose
units carry a pyruvic acid residue. When hydrated in solution the xanthan gum
side chains wrap around the backbone thereby protecting the [3-1, 4 linkages due
to hydrolysis. These side chains allow the polymer to completely hydrate, even
in cold water (Urlacher and Noble 1997).

Aqueous solutions of xanthan exhibit a very high viscosity, very strong
pseudoplasticity, with no evidence of thixotropy, even at very low concentrations.
At very low concentrations, xanthan forms reversible entanglements, as a result
of its rod like conformation of xanthan in solution and its high molecular weight.
This behavior can be advantageous by decreasing the viscosity of a xanthan
solution with increasing shear rate, resulting in easy pouring, pumping or mixing

(Urlacher and Noble 1997). Xanthan gum is also a very resilient molecule.

19

 

 

I

Tabl'

Ti

rop

l

 

Ykﬂd

 

Pseuc

 

 

Ehabh

*

 

Table 4 - Xanthan Gum Functionality (Phillips and Williams 1995)

 

Property Function of beneﬁt

 

Yield Value Emulsion stability
Prevents runoff
Suspension of particles

 

Pseudoplasticity Improve pourability
Clean mouthfeel

 

 

Stability Uniform viscosity after high shear or temperature variations
Stable in low-pH or high-salt dressings

 

 

20

 

The secondary structure of the molecule, in which the side chains are
wrapped around the cellulose backbone, provides degradation resistance of this
hydrocolloid in the presence of high temperatures, enzymes, mixing, and
changes in pH. Under high temperatures xanthan, in the presence of salt,
remains stable and maintains viscosity. Emulsions stabilized with xanthan gum
exhibit very high viscosity at rest and are exceptionally stable under the low
shear conditions encountered during transport, storage and subsequent use.
Xanthan is therefore a very useful ingredient in salad dressings.

Xanthan, being very hydrophilic, requires special care for proper
dispersion and hydration when incorporated into a salad dressing system. Time
of hydration is a result of the effectiveness of dispersion, or separation of the
gum particles at introduction to the solution, size of the gum particles and the
other components in the solution. In 1989, Paredes and others showed that a
minimum storage or aging period is required to complete xanthan gum
solubilization and reach the desired product viscosity.

In preparation of a dressing, the main objective is to stabilize the oil /
water emulsion. Good hydration is obtained by mixing the xanthan with other
dry ingredients or by dispersion in oil. Xanthan gum dissolves readily at an
ambient temperature (20°C) and over a range of pH values and salt
concentrations. Hennock and others in 1984, proposed that xanthan gum
provides emulsion stability through two mechanisms. First, xanthan gum is

absorbed at the oil I water interface, which lowers the surface tension and

21

 

 

 

 

 

redur

by P-'

the x

 

Unde
thou]
ml
Mabr
shon
such
body
Inths
hﬁck

0fthe

2.3 l

IrTiar-
i”gre
t09er-
eXCel

enhﬂs

 

reduces the droplet size. Second, the remaining xanthan stabilizes the emulsion
by physically trapping the emulsion droplets.

In industrial production, a dispersion funnel can be utilized, which draws
the xanthan from a funnel through a venturi tube with water under a vacuum.
Under a continuous process, the xanthan powder and liquid phases can be
brought together in a cyclone chamber where dispersion and hydration can
occur.

In dressing manufacturing, the objective of using xanthan is to produce a
stabilized oil / water emulsion, for up to one year. Strong pseudoplasticity and
strong stabilizing properties of xanthan give dressings which suspend particles,
such as spice, herbs and vegetable and give the dressing its overall appearance,
body, ease of pouring, cling to salad, and textural attributes. Texture is revealed
in the ﬂow of the dressing system. Over application of xanthan can result in a
thick appearance, resulting in an unacceptable ﬂow behavior, due to the elasticity

of the xanthan gum.

2.3 Manufacturing of Dressings

Formation of a stable emulsion is the ultimate goal during the production
[manufacturing of a dressing. The manufacturing of a dressing is where the
ingredient functionality, processing equipment and production conditions come
together. Oil / water emulsions which contain a high oil content must have
excellent stabilizers due to the tendency of the system to be a water / oil

emulsion.

22

 

Not only the ingredient functionality of the emulsiﬁers, including the egg
yolks is important in maintaining a stable emulsion, but numerous processing
conditions need to be avoided to increase stability. Holding emulsion premixes
for extended periods of time, over shearing and multiple restarts of the process
can break the emulsion. Making a stable pre-emulsion is critical to creating
stable unbroken emulsions with minimal process down time.

Formation of emulsions can be accomplished with a variety of unit
operations, which range from high-energy short time systems (homogenizer), to
long-time low energy processes (mixer). The choice of the homogenizer is
dependent on the volume of batch, throughput, formulation and desired droplet
size. Various types of homogenizers can be used to achieve acceptable
products under a given set of conditions, (Table 5).

The emulsiﬁcation of a dressing system occurs under turbulent ﬂow
conditions as a result of the energy distributed by the droplets density, interfacial
tension and mass density. Input of mechanical energy, during the
homogenization process, causes an increase in the amount of interface in the
emulsion system as a result of energy input. This energy causes subdivision of
oil droplets, resulting in a greater amount of interfacial area between two phases.
Application of shear forces subdivides droplets in an emulsion, (Figure 2).

During homogenization, new droplet interfaces are formed. Emulsiﬁers
diffuse to the interfaces and lower the interfacial tension as the droplets are
formed. The emulsiﬁer acts as a protectant for the newly formed droplets. A

colloid mill, agitating vessel, homogenizer and hydroshear are the most popular

23

 

Table 5 - Comparison of different types of emulsifying devices (Mc
Clements 1999).

 

 

 

 

 

 

 

 

Throughput Relative Minimum Sample
energy droplet size viscosity
efficiency
High speed Batch Low 2pm Low to
blender
medium
Colloid mill Continuous Intermediate 1 pm Medium to
high
High pressure Continuous High 0.1“m Low to
homogenizer
medium
Ultrasonic probe Batch Low 0.1pm Low to
medium
Ultrasonic jet Continuous High 1pm Low to
homogenizer
medium
Microﬂuidization Continuous High <0.1pm Low to
medium

 

 

 

 

 

 

24

 

Shear

 

 

 

 

. i >
7010101.

00000

 

 

 

 

Figure 2- Breakage of cylinders of liquid into small droplets caused by shear

(Stauffer 1996).

25

emulsiﬁcation devices used in the dressing industry (Ford and others 1997).
Colloid mills are commonly used for mayonnaise and spoonable dressings. The
hydroshear and high-pressure homogenizers are used for pourable salad
dressings. Other homogenizer types are used for sauces and ketchup.

The high-pressure homogenizer, also called the hydroshear, is the most
commonly used method to produce ﬁne emulsions in the food industry. This
homogenizer pumps the coarse emulsion into a chamber and then forces it
through a narrow valve at the end of the chamber, (Figure 3). High-pressure
valve homogenizers can be used for a variety of different foods. Processing
parameters and conditions inﬂuence the interactions among food components
within a system. The interactions of proteins, color, ﬂavor and ingredient
functionality can all be manipulated and altered with changes applied to the
processing parameters.

Ingredient interactions have an effect on all stages of the product life,
from processing conditions, heat transfer, ﬂow properties, packaging, and
storage through the ﬁnal handling of the product by the consumer.
Understanding the interaction of the ingredients and manufacturing parameters,
such as shear rate, can result in comprehending the relationship between the
food system and processing conditions. In this study the relationship of three
processing variables were compared to the physical properties of particle size,

and ﬂow behavior of the ﬂuid system.

26

Owen-w“

 

 

Fine Emulsion
Impact Ring

\ U

 

 

 

 

 

Coarse
Emulsion O
O
l:> o
O O
O

 

 

 

 

 

 

 

 

Figure 3 - High pressure valve homogenizer, used to produce emulsions with ﬁne
droplet sizes (Mc Clements 1999).

27

 

 

2u4

of tr
App
deve
knot-
desi

ﬂovl

 

198.

 

2.4 Rheology of Salad Dressings and Methodology of Fluids

Rheology is the study of the deformation and ﬂow of matter. It is the study
of the response to an applied stress or strain, and is the material science of food.
Applications of rheological data are numerous: process engineering, product
development, quality control, shelf life and sensory evaluation (Steffe 1996). The
knowledge of the rheological properties of ﬂuid foods is essential for the proper
design and operation of product production equipment, and for understanding the
ﬂow and transport of the product during production (Rao and Anantheswaran
1982). Studying the relationship between particle size and the rheological
properties of a pourable model system, salad dressing, will provide a quantitative
contribution of the control of ﬂow behavior for different formulations and process
variables. The ﬂow behavior of salad dressings is very complex and consumer
approval of product “ﬂow” is an overall affect of all relevant textural attributes

(Ford and others 1997).

In this study, the rheological approach to characterizing salad dressing
involves a large deformation issue as a result of the consumer, gravity driven
application of pouring. The application of pouring from a bottle gives the
consumer a visual assessment of the product thickness (textural attribute) and
viscosity of ﬂow (rheological attribute). The consumer perception of the
thickness and viscosity of a salad dressing begins from the pouring out of the
bottle, the clinging of the dressing to the salad, and the ﬁnal sensory perception
of mouthfeel.

The typical shear rate range of pouring from a bottle is 101 to 102 (Steffe

28

 

 

 

lnl

and

pen

 

iudg
DOU'
cont

the

 

eﬁe:
that

and

vhcc

 

SYI’UF‘
aithr
neck
Dane

inSldr

TMs
reSulr

aggre

1996). Pourability and spreadability are important attributes of salad dressings.
In 1983, Kiosseoglou and Sherman conducted a study in which the shear stress
and shear rates of two salad dressings were associated with the sensory
perception of pourability and spreadability of the dressings, based on human
judgement. The sensory panel noted that they based their judgement of
pourability on the rate at which the samples ﬂowed down the inside of the
container, and spreadability on the area of a plate that was eventually covered by
the sample. This suggests that yield stress and shear-thinning characteristics
effect panelist judgement. Kiosseoglou and Sherman concluded from their study
that shear force and shear stress are dependent on the thickness of the sample

and the container degree of tilt by the panelist.

In 1992, Elejalde and Kokini performed a study on the psychophysics of
viscosity in which they evaluated the sensory viscosity of the unsteady ﬂow of
syrup from a bottle. It was concluded that the “degree of ﬁll' or stream thickness
at the neck of the bottle can be represented by the cross sectional area of the
neck of the bottle ﬁlled by the product stream. They also determined that the
panelists based their judgments on the rate at which the sample ﬂowed down the

inside of the bottle.

The viscosity of salad dressings is the most widely used characterization.
This study focuses on the rheological response of the emulsion system as a
result of the particle size and aggregation of the droplets. It has been shown that

aggregation of oil droplets increases emulsion viscosity. The interaction of the

29

WW

droplets determines whether the aggregates form compact or open systems.
Open aggregates pull along large amounts of continuous phase, resulting in a
higher viscosity than compact systems (Ford and others 1997). Open
aggregates occur when there is a strong attractive interaction between the
droplets and compact aggregation occurs during weak interaction. The state of
aggregation is a result of shear history. As shear rates increase, the aggregates

are broken into smaller sizes and the emulsion viscosity decreases.

Fluids and their rheological properties can be classiﬁed and characterized
under various types of rheological behavior: Newtonian, non-newtonian, power
law, pseudoplastic, shear-thinning, shear-thickening, rheopectic, thixotropic,
time-dependent, time-independent. A large number of ﬂuid foods are non-
Newtonian. The most common characterization of ﬂuid foods is achieved by
constructing a ﬂow curve in which the relationship of shear stress and shear rate
is established. Rheological behaviors of many ﬂuid food emulsions and
suspensions, including the measurement of ﬂow properties, factors inﬂuencing
the rheological behavior in relation to sensory and viscosity, were characterized

by Rao (1977).

A common instrument used for the study of the ﬂow behavior properties of
ﬂuids and semi-solid foods is the rotational viscometer. Rotational instruments
may be operated in the steady shear or oscillatory mode. Rotational type
instruments include concentric cylinder, mixer, parallel plate and cone and plate

systems. A concentric cylinder viscometer was used in the current study.

30

if. 2.1-1.4 Ian-m

2.4.1 TimeITemperature Dependence

Different temperatures and time parameters are often encountered during
processing, storage, distribution, and consumption of liquid foods (Rao 1977).
The determination of the time and temperature effects on the apparent viscosity,
for Newtonian ﬂuids (1), and power law ﬂuids (2), at a determined shear rate are

described by the following model equations:

Mimi (1)

wander/EKG)“ (2)

The power law ﬂuid model, also know as the Ostwald-de Waele model, is
the simplest and most popular model applied to non-Newtonian ﬂuid ﬂow. The
power law model has two parameters, n and K, which are the ﬂow behavior index
and consistency coefﬁcient, respectively. The ﬂow behavior index, n, provides a
convenient way to identify the shear-thinning (n < 1) or shear thickening (n > 1)

behavior.

Thixotropic behavior (Figure 4) is seen when a system experiences a
reversible decrease in shear stress or apparent viscosity at a constant shear
rate. These ﬂuids are thought to consist of asymmetrical molecules or particles,
which form a network or aggregated structure at rest. When a continuous shear
rate is applied to the system, weak bonds are broken, which results in the ﬂuid
exhibiting time—dependent thinning characteristics. A rheopectic ﬂuid, the

opposite of thixotropic, is a material, which exhibits a reversible increase in shear

31

.fers-‘a-Jﬂ

 

 

Time-Dependent Behavior
(Thixotropic

Time-Independent

i

 

 

Shear Stress, Pa

 

R
Rheopectic

 

 

Time at Constant Shear Rate, 5

 

 

 

Figure 4 - Time - dependent behavior of ﬂuids (Steffe 1996).

32

stress or apparent viscosity when sheared at a constant rate.

The viscoelasticity and/or structural changes of liquid food depend on
temperature, composition, the application of an applied shear rate or shear
stress, and the duration of shear (Rao 1977). Ideal time-dependent behavior of a
material is considered to be inelastic with a viscosity function, which depends on
time (Steffe 1996). It is important to study the rheological properties as a

function of temperature and time due to varying processing conditions, duration

 

of shear and temperatures faced during production of the material, and use by

the consumer during shaking and mixing the product.

Fluids may exhibit time-dependent ﬂow in which there are decreasing or
increasing effects on the shear stress or apparent viscosity with time during a
constant shear rate and temperature. Salad dressing, a non-Newtonian ﬂuid,
would have time-dependent behavior as a result of the weak interaction among
the components in the emulsion system resulting in easy disruption during the

shaking of the product prior to use (Tung and Paulson 1995).

In characterizing time-independent ﬂow behavior, the Herschel-Buckley

model can be used to describe the general relationship:

a = K (y)n + co (3)

This model is appropriate for many ﬂuid foods and can be utilized for
NeWtonian, power law and Bingham plastic ﬂuids, ﬁgure 5 and 6 (Steffe 1996).

33

 

 

    
 
 

Herschel

   

 

 

to
o, t

iii Shear

a) . .

b Thinning

a) Newtonian
is

cu

.:

U)

ear -
Thickening

 

Shear Rate, Us

 

 

 

Figure 5 Typical curves for time-independent behavior ﬂuids (Steffe 1996)

 

 

Apparent Viscosity, Pa 8

 

‘/ Bingham
Herschel-Bulkley
(0< n < 1.0)

___

 

Shear- ickening
/ Shear-Thinning

I?

 

'\Newtonian

 

 

 

Figure 6 - Apparent viscosity of time-independent ﬂuids (Steffe 1996)

35

 

Huhé

 

(Sm?

 

Shear

allow

 

Shear

Chara

siste

ShEa
the
apie

809‘

Fluids can be studied by subjecting them to continuous shear at a constant rate

(Steffe 1996).

In 1983, Figoni and Shoemaker studied the time-dependent ﬂow
properties of commercial mayonnaise to investigate the stress decay or structural
breakdown of the mayonnaise at shear rates of 0.530, 0.169, 0.052 and
0.0169“. They found that the initial and ﬁnal stress values increased with shear
rate, while the initial stress values were not signiﬁcantly different. Values for the
ﬁnal stress did discriminate between the different shear rates, but were not
determined to not be signiﬁcantly different. The insigniﬁcance was determined to
be a result of the thixotropic nature of the mayonnaise, with the lipid droplets
exhibiting pseudoplastic behavior due to the ﬂocculation-deﬂocculation, at high
shear rates, of the lipid droplets. Overall they found that steady shear studies
allow for reproducible calculations of rates of structural breakdown as well as

shear stress-shear rate data, to study time-dependency.

Tiu and Boger (1974) and Figoni and Shoemaker (1983) performed a
characterization of the time-dependent ﬂow properties of the non-Newtonian ﬂuid
system mayonnaise. Using a cone and plate viscometer to measure the stress
decay, Figoni and Shoemaker measured the time dependent ﬂow properties at
shear rates of 0.530, 0.169, 0.052 and 0.0169 s". The shear stress increased as
the shear rate increased, indicating pseudoplastic ﬂow. This decrease in
apparent viscosity as the shear rate increases was also reported by Tiu and

Boger (1974), Elliott and Ganz (1977) and Figoni and Shoemaker (1983). This

36

ﬂowt

 

and J
ﬂowt
how

dehOi

 

ﬂute
Shoe
large

DIDC

244$

sub
anc
rah
‘Ws
Fio

Cyl

de

Sh

ﬂow behavior of mayonnaise can be explained by the following:

“Since the oil droplets in mayonnaise are stabilized by egg yolk (Chang
and others 1972), and droplet deformation is unimportant when considering the
ﬂow behavior of a stabilized emulsion (Van den Temple 1963), the pseudoplastic
ﬂow behavior of mayonnaise probably is due principally to the ﬂocculation-
deﬂocculation of the oil droplet. As the shear rate increases, the reaction is
shifted towards deﬂocculation, thus lowering the apparent viscosity. (Figoni and
Shoemaker 1983).” Figoni and Shoemaker also concluded the breakdown of
larger aggregates continually generates smaller aggregates during the shearing

process.

2.4.2 Steady State Shear Measurements

The measurement of a ﬂuid under steady state conditions classiﬁes a
substance according to type of ﬂuid. A speciﬁc shear rate can then be selected
and once conditions in the rheometer have stabilized, the viscosity at that shear
rate is determined. The shear rate can also be examined in a sweep test in which
viscosity data is collected over a wide shear rate range resulting in a ﬂow curve.
Rotational viscometers including cone and plate, parallel plate and concentric

cylinder, can all operate under steady shear conditions.

The apparent viscosity - shear rate relationship of shear-thinning foods is
described by Rao, 1999 as: “At sufﬁciently high polymer concentrations, most

shear-thinning biopolymer (also called a gum or hydrocolloid) dispersions, exhibit

37

and

 

IESU'.

tosh

 

 

assul
const
vehbc
lane:
E kn]

Use.

and i

SIEad

a similar three stage viscoelastic response over a wide shear rate range.” Rao
describes these three stages as (1) at low shear rates, they show Newtonian
properties with a constant zero-shear viscosity (110) over a shear range, followed
by (2) a shear-thinning range where solution viscosity decreases in accordance
with the power law relationship. At (3) high shear rates the ﬂuid shows a limiting
and constant inﬁnite shear viscosity (11...). These regions are thought to be a
result of the rearrangement in the conformation of the molecules in the ﬂuid, due

to shearing (Rao 1999).

The relationship used to describe a concentric cylinder viscometer
assumes: ﬂow is laminar and steady, end effects are negligible, temperature is
constant, there is no slip at the walls of the instrument and radial and axial
velocity components are zero (Steffe 1996). Understanding the steady shear
rate and its relationship during the typical application of pouring a salad dressing
is important to understand the ﬂow behavior exhibited during actual consumer

use.

Rheological characterization of salad dressing began in the 19705. Elliott
and Ganz in 1977 performed a preliminary rheological characterization, under
steady shear measurements to determine the ﬂow properties and estimate the
degree of structure breakdown. Paredes and others in 1988 and 1989 performed
a series of rheological characterization studies of bottled and dry salad
dressings. They evaluated the steady shear, thixotropic behavior and effect of

temperature in their ﬁrst study (1988) and the effect of storage in the second

38

smci

 

mng
bohjl
mod
me I
The*

Hahn

 

apps

eque

study (1989).

Paredes and others (1988) collected shear stress data over a shear rate
range of 30 to 1000 s". The power law equation described the data and that
bottled dressings showed a greater decrease in viscosity with an increase in
product temperature than dry dressing mix. They also found that the dressings
are pseudoplastic ﬂuids, with a ﬂow behavior index (n) between 0.43 to 0.93.
The thixotropic behavior, the rate of structural breakdown, was described by the

Hahn (4) and the Weltrnan (5) equations:

0 = A1- B1logt (4)

L09 (0 - cc) =A2 - th (5)

The rate of structural breakdown was faster at 2°C that at 10°C. The
apparent viscosity was also decribed by Paredes and others, with the Arrhenius

equation (6):

Tia = 3 exp (Ea/ Rt) (6)

Paredes and others (1989) showed that the consistency index (K) and the
ﬂow behavior of the power law model were able to describe the viscoelastic
nature of salad dressing as a function of storage time and temperatures of 1.7,
18.3, 29.4 and 378°C. They determined that the magnitude of K of a model
dressing, which decreased after 24 hour storage at 28°C, showed a great

increase during the ﬁrst seven days after production at the designated

39

 

tempe
of the
vmue
Their
least

solub

2.4.3

viscos
amplit
small
foods
939 b.

can b7

the ch
a Shea
StEadij

of stea

and Sr

 

temperatures of 18.3, 29.4 and 378°C. They also saw that after seven days all
of the viscosities increased except the one stored at 378°C. Similar results
were seen in samples pulled out of storage conditions 24 hours prior to testing.
Their study showed that a minimal storage temperature, 188°C to 294°C for at
least 24 hours maybe required, prior to a cold test at 28°C, for complete

solubilization of xanthan gum and to obtain the desired viscosity for the dressing.

2.4.3 Unsteady State Shear Measurements

Unsteady state shear measurements or dynamic evaluation of
viscoelasticity, can be deﬁned in either transient or oscillatory ﬂow. Small
amplitude oscillatory shear, where a sample is exposed to harrnonically varying
small amplitude deformations, is used to determine viscoelastic properties of
foods. Uniform shear can be achieved during measurements through a small
gap being achieved in a cone and plate geometry. Many viscoelastic properties

can be determined (Rao 1999).

Transient shear testing involves numerous measurements under which
the characteristics of a viscoelastic food under a sudden shear rate. As a result
a shear stress is exhibited which displays an overshoot before reaching a ﬁnal
steady state. Transient shear ﬂow experiments include: start-up ﬂow, cessation

of steady shear ﬂow, step strain, creep and recoil (Steffe 1996).

Bistany and Kokini (1982) performed a comparison study of steady shear

and small amplitude dynamic viscoelastic properties of ﬂuid foods, including

40

 

mayo
I
the 5*

law b
i

utilizﬁ

angle

 

stress
know'
dress
data

dress
Wpic
varie
Cree

Writi

\hi

mayonnaise. The dynamic viscosity (11*) was determined to be much larger than
the steady shear viscosities (11). Both viscosities could be described by power
law behavior. Their study focused on a shear rate range of 0.1 - 100 sec",
utilizing a cone and plate geometry with a plate radius of 1.25cm and a cone

angle of 0.04 radians for their dynamic measurements.

Another type of transient shear ﬂow experiment, where an instantaneous
stress is applied to a sample and the change in strain is observed over time is
known as creep compliance. Paredes and others in 1989 characterized salad
dressings, bottled varieties and dry mixes, at 28°C +l- 0.3°C. Creep compliance
data for the bottled dressings were collected at 55.2 dyne cm‘2 and dry salad
dressings at 22.8 dyne cm’z. These shear stress rates were chosen to develop
typical creep compliance-time response curves for each type / brand of dressing
variety: bottled creamy, reduced calorie and dry mixes. They determined that the
creep compliance model response can be described using the Burgers model,

written in terms for shear creep compliance (7) (Steffe 1996):

t
+—

ruo

ref

 

J= f0) = J0 +J,[1-exp[l;t—]
(7)
Paredes, 1989, also compared creep compliance analysis in relation to
the effects of storage. Xanthan gum was utilized in the dressing formulations.

Paredes determined that the solubilization of the xanthan gum is important in

maintaining the emulsion stability during storage. The greatest changes were

41

seen in the creep compliance time plots during the ﬁrst week of storage.

Munoz and Sherman, 1990, performed controlled stress rheometer
measurements on commercial salad dressings. Oscillatory tests were conducted
at a frequency of 1 Hz with an amplitude of applied stress of 14.93, for full fat
mayonnaise or 8.96 Pa, for reduced calorie mayonnaise and salad cream, over a
frequency sweep of 600s, at 25°C. Upon evaluation of the 6’ values, the full fat
and reduced caloric mayonnaise produced at the same manufacturer had
signiﬁcantly similar G' values. This may be due to the formula of the reduced
caloric dressing which contains modiﬁed starches and gums and the substitution
of egg whites for egg yolks. In comparing the G’ values of two dressings which
both contain xanthan gum, the dressing with which contained sugar and lemon
juice, instead of lactic acid, had a lower G’ value. It is thought that the sugar
molecules may effect the proteins which interact and form a network formation

among the oil droplets.

Overall, Munoz and Sherman, 1990, found in high fat mayonnaise that the
droplets are tightly packed together, forming a three dimensional network
between the egg protein molecules around the lipid droplets. This is responsible
for the viscoelastic properties of the high fat mayonnaise. These highly
ﬂocculated structures of lipid droplets are not formed in the salad creams

measured, due to the lower oil content, resulting in different ﬂow behavior.

Peressini and others in 1998, performed an extensive rheological

characterization of traditional and light mayonnaise. It was shown that under the

42

measurement of the oscillatory and creep recovery tests, the storage modulus
(G’), compliance and yield stress increased as the fat level increased in the
mayonnaise. These measurements were performed at 25°C, with each sample
resting for 5 minutes after being loaded to allow sample-induced stress on

sample to relax and temperature to equilibrate.

As can be seen most of the rheological testing on dressing and
mayonnaise systems involves dynamic rheological testing methods.
Measurement and evaluation of the ﬂow behavior as a consumer would use it,
needs to be further evaluated. Pouring in relationship to a variety of formulation
and processing parameters, as would be seen by a consumer, can be evaluated

through rheological steady state rotational viscometry testing.

2.4.4 Yield Stress

Yield stress (00) can be deﬁned as the minimum shear stress required to
initiate ﬂow. The yield stress of a ﬂuid is important in the engineering of systems,
quality control measurement for reproducibility of product, and comparison of
products. Yield stress is also important for dispersion food systems such as

emulsions, in order to keep the particles in suspension.

Fluids which exhibit yield stress can be described as structurally having a
network of bonds that are breaking and deforming as a result of the applied
stress (Rao and Steffe 1997). Yield stress can be characterized from this

structural breakdown as two types. An undisturbed sample is known as static

43

and in a completely broken down sample is known as dynamic yield stress, ﬁgure

7 (Steffe 1996).

Using a rotational viscometer and the Casson equation, Paredes and
others, 1989 described the yield stress ﬂow properties of bottled and dry mix and
prepared salad dressings. The yield stress, at 28°C, of the model system
increased after seven days of storage. The yield stress of salad dressing at
188°C, 294°C and 378°C remained stable over the same period of time. All of
the salad dressings stored at 21°C exhibited similar yield stresses, showing an
increase for seven days and then remaining relatively stable. Results indicated
that salad dressings need to be held for at least seven days prior to quality

control testing and sensory evaluation.

Peressini and others in 1998 examined the yield stress of mayonnaise
was measured utilizing a plate-plate geometry, diameter 40mm and gap of
1.5mm. The yield stress was shown to increase with increasing fat content in
mayonnaise. Mayonnaise with 76% fat, exhibited a yield stress of 40.6 Pa, while
the samples with 68.6% fat exhibited a yield stress of 25.4% and mayonnaise at
63.4% fat exhibited a yield stress of 19.4 Pa. The higher yield stress values in
the high fat mayonnaise is a result of the higher level of egg lipoproteins which
create a stable network at the water / oil interface, resulting in increased

equilibrium, over the lower fat mayonnaise samples.

 

 

 

I

/‘\

Shear Stress

 

21am Yield Stress

Dynamic Yield Stress

 

Shear Rate

 

Figure 7 Static and dynamic yield stress (Steffe 1996).

45

 

 

No evidence of the utilization of the vane method to determine the yield
stress of salad dressings can be found. This controlled-shear-rate and
controlled-shear-stress mode methods measure the yield stress of a ﬂuid system
through the immersion of a vane directly into a product. This method can be a
great advantage due to possible immersion of the vane directly into a packaged
product and reduction of disruption of the sample during loading for static
measurements (Steffe 1996). This method is advantageous for allowing for
quality control work to be performed directly in the packaged food, resulting in
measurements which result in yield stress measurements as would be seen by

the consumer.

2.5 Particle Size Analysis

A “particle" can be described as any object having deﬁnite physical
boundaries in all directions, without any limit with respect to size. The concept
that all matter is made up of particles is fundamental to science (Cadle, 1965). It
has also been shown that smaller particles may not be uniformly distributed
through out a material. The size of particles found in the atmosphere, or in other
gases range in diameter from about 0.001 to 100 pm. These particles can
possess different shapes chemical compositions refractive indexes electric
properties and densities. Fine particles are deﬁned as particles in which the

diameters are less than 1.0 pm.

Particle shape and size in emulsions are an important characteristic

because, they inﬂuence many other characteristics and behaviors of the

46

emulsion. The ﬂow properties as a result of the particle size and shape can play
a major role in the pouring and pumping of food systems through pipes, during

manufacturing, and during customer application of a dressing.

Most colloidal suspensions exhibit a wide range of particle sizes. A
complete knowledge, in addition to the maximum, minimum and average size
may be needed to completely understanding the material being analyzed.
Particle shapes can vary greatly unless special processes, to form uniform
shapes throughout the product are applied during processing. Shapes can
include oblate and prolate spheroid or disc shapes, which can be, either an

oblate spheroid, cylinder of a ﬂat plane.

Many foods can be classiﬁed as colloidal systems, containing particles or
macromolecules (polymers) of various kinds (Walstra, 1992). Aggregation of
these particles can be deﬁned as the staying close together for a much longer
time than would be the case in the absence of attractive forces between them.
The particle size of the droplets within an emulsion system generally exceeds

0.1 [1 (Coia and Stauffer, 1987).

According to colloidal science theory, aggregation occurring within the
emulsion I colloidal system depends upon the interaction forces between the
particles. Food systems contradict the theories of interactions of classical
emulsion / colloidal systems, which involve the interaction between two identical
homogeneous spheres. But, as we know food systems are very complex

systems and contain many types of particles, varying in size, shape and

47

homogeneity. Hence, it is essential to know the food system, the particles it

contains, size, shape and how they aggregate.

During ﬂocculation part of the continuous phase becomes immobilized
due to the formation of aggregates, therefore increasing viscosity. Upon aging of
the product, globule aggregation continues due to ﬂocculation of the remaining
globules as well as the joining of small aggregates into larger ones. A high yield
stress can be seen in the emulsion if the continuous phase becomes trapped
within the linking up of aggregates which form a loose continuous network

(Sherman, 1967).

According to studies performed by Sherman, 1964, “ﬁne” emulsions are
known to give higher viscosities than “coarse” emulsions of the same formulation.
Also, emulsions with a broad distribution of globule sizes will have a lower

viscosity than comparable emulsions with a narrow distribution of particle size.
2.5.1 Measurement

Particle size analysis measurement has begun to evolve as the modern
developments in computers, electronics, lasers and chromatography have all
continued to improve the measurement of small size emulsion droplets.
According to Orr, 1988, the phenomena traditionally used to characterize
submicroscopic entities has begun to reveal new size information. These
phenomena include, hydrodynamics, light scattering and diffraction and Browning

diffusion.

48

The methodologies used to determine emulsion systems particle size may
involve various techniques. The hydrodynamic chromatography method involves
the separation of the emulsion based on size through a porous packed column,
thus related the molecular weight and micelle size. Photon correlation
spectroscopy involves the intensity of light scattered from dispersions of particles
and macromolecules. Particles can also be measured through the analysis of
the intensity pattern created when light is diffracted from a beam, which passes

through the emulsion. This type of method is non-destructive and very rapid.

Two types of measurement that are non-destructive and rapid include
Magnetic resonance imaging and Spectroscopy. Magnetic resonance imaging
involves the measurement in the variations in signal intensity in relation to the
relaxation times of the oil and aqueous phases and the proportionality of the
projected frequencies to the spatial distribution of the sample constituents. In
spectroscopy, the percentage of oil is determined from standard curves which
relate the integrals of oil for the total spectra are correlated to the known amounts

of oil in standard preparations (Heil, 1990).

Optical microscopy is also a traditional type methodology, which can be
utilized to directly measure individual particle sizes. This type of methodology is
viewed as being reliable, but results in difﬁculty in the reproducibility and
accuracy for the measurement of emulsion systems. Other methodologies
include the measuring of particle size and count as they ﬂow past a sensor and a

sedimentation technique, based on gravitational separation of phases and

49

fractional creaming based on the relationship between the creaming agent and

creamed particle size.

The mechanism of light scattering measurement occurs as light strikes
particles causing (diffraction) scattering to occur. This principle is based on the
Fraunhofer diffraction and Mie scattering theories (Horiba, 1998). The laser light
beam irradiates the particles and then measures the pattern of light scattering by

the particles.

Munoz and Sherman, 1990, performed droplet size distribution analysis of
mayonnaise using a Coulter-counter. For the full fat mayonnaise the mean size
(pm) ranged between 2.53 - 3.81 pm, with a standard deviation of 1.43 - 1.78.
Reduced calorie mayonnaise had a mean diameter of 1.71 - 1.82pm and a

standard deviation of 1.35 - 1.38.

Early measurement of mayonnaise with electron microscopy began with
the work of Chang and others in 1972. The electron microscope observed that
the oil droplet in all of the mayonnaise samples to posses a continuous layer.
This layer is thought to contain coalesced low-density egg yolk Iipoproteins and
microparticles of egg yolk granules. Heil and others, 1990 utilized magnetic
resonance imaging and spectroscopy to determine the percent of oil in a French
style dressing. Both of these methodologies allow for rapid measurement and

nondestructive analysis.

In an emulsion system, the size and shape of the lipid droplets and the

50

connecting networks between them have an impact on the textural properties, as
well as appearance of the emulsion. Langton and others, 1999, studied the
relationship of the microstructure of an undiluted mayonnaise sample, in
relationship to its textural properties. The mayonnaise variables included
variable emulsiﬁcation cylinder speed, speed of viscorotor, exit temperatures and
egg yolk content. The mayonnaise microscopy was studied through the use of
confocal laser scanning microscopy and transmission electron microscopy. It
was concluded that the oil droplets form a network which has an impact on
texture. They also detected aggregates of egg yolk particles, which ranged in
diameter size of 1.5 — 7 pm. The emulsiﬁcation cylinder speed and the exit
temperature were shown to have a strong effect on the size of the egg yolk

particles.

2.5.2 Control of Particle Size

The use of macromolecules or polymers within a continuous phase may
result in the absorption of the polymers onto the particles. This would result in a
change in viscosity or the colloidal system and manipulating the formation of
aggregates. Absorption of polymers may either prevent aggregation or cause it.
This depends on the solubility of the polymer. Xanthan gum is a prime example

of a polymer used in salad dressings.

Carrillo and Kokini, 1988, conducted a study measuring the effect of egg
Yolk and egg yolk + salt on the rheological properties and particle size distribution

of model oil in water salad dressing emulsions. They found that the egg yolk

51

signiﬁcantly contributed to the stability and viscosity of the model system. Salt
also contributed to the stabilization action of the egg yolk. Creep and steady
shear analysis, along with particle size determinations showed that the emulsions
were most stable in the presence of 2% egg yolk of 3% egg yolk. These
formulas were pre-stabilized with xanthan gum and propylene glycol alginate.
The steady shear measurements showed that both the egg yolk and salt
increased viscosity radically. Also, particle size distribution changed with both

the addition of egg yolk and salt to the model system.

The effects of formulation, including pH and NaCI have been shown to
affect the mean droplet size of a mayonnaise system (Kiosseoglou and Sherman
1983). Droplet size was determined with a Joyce-Loebel disc
photosedimentometer. Kiosseoglou and Sherman determined that the oil drops
ﬂocculated and form a three dimensional network structure that exhibited
viscoelastic behavior, which is effected by pH, NaCl and temperature. The mean
particle size increased as the sample aged from 4 - 150 hours. As the pH
degreased the mean particle size increased: at 24 hours, pH 6.2, mean particle

diameter size 0.378pm, while at pH 3.5, mean particle diameter size 0.875iim.

2.5.3 Size and Composition

There are a limited number of publications dealing with the microstructure
of dressings. In 1972, Chang used electron microscopy and determined the
stability of mayonnaise was attributed to the ﬁlm membrane surrounding the

droplets. It was concluded that the ﬁlm consisted of low-density lipoprotein 9

52

(LDL) micelles from the plasma and granules of egg yolk. It was then postulated
that the stability of the droplets and plasticity of the droplets could be attributed to
the LDL and this barrier should obstruct coalescence. Hell, in 1990, used
magnetic resonance imaging and spectroscopy to determine the percentage of

edible oils in French style salad dressings.

Carrillo and Kokini, in 1988, measured and studied the particle size
distribution of a model oil-in-water emulsion dressing in relation to the effects of
egg yolk and egg yolk and salt on the rheological properties. Particle size was
measured using a Modiﬁed Coulter Counter Particle Size Analyzer with a 128 pm
oriﬁce. A shift in particle size distribution toward small values was noted upon
addition of both egg yolk and salt, and a shift to larger values was found when

egg yolk and salt were not present.

53

CHAPTER 3
MATERIALS AND METHODS

3.1. Formulas I Samples

Salad dressings are an emulsion-based system usually containing oil,
buttermilk, water, sugar, egg yolks, salt, vinegar, sorbic acid, phosphoric acid,
gum and emulsiﬁers. This research involves the evaluation of nine different
model-dressing formulas. These formulas are comprised of 59% oil which gives
the emulsion ﬂavor and texture. Formulations vary in percentage of xanthan gum
and egg yolks used for the production of each individual batch. Xanthan is
added to provide emulsion stability, prevent runoff, suspend particles, improve
pourability, mouthfeel, and create uniform viscosity under shear and temperature
variations and stability with variable pH and salt contents. Egg yolks also provide
emulsifying functionality, due to the lecithin protein complex (Iipoproteins). Use
of two emulsiﬁers, egg yolk and polysorbate 60, is common to achieve multiple

functions in the emulsion system (Hasenhuettle and Hartel 1997).

A wide range of textures have been developed with the nine formulas
tested. The amount of sugar and oil varied slightly to make up for the reduction or
addition of the xanthan gum and egg yolks. The percentage by weight of each
ingredient used in the nine formulas is given in Table 6. Four major competitive

brands of basic Ranch Dressing, (Kraft Ranch©, Hellmann’s Ranch©, Wishbone
Ranch© and Hidden Valley Ranch©), purchased from MeijerTM grocery store,

where also tested for comparison. Total weight of each batch was 13.67 kg
(30.12 lbm).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9. .2995
5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.8 .58
so: so: so: so: .522 so: so: $2.2 so: 502 58:5
s08 .22
$2.0 $2.0 s25 $2.0 $25 $25 22.0 $25 £25 25585.
s88 .5555 $8.3 $8.5 $8.55 055.55 055.8 .5555 28.8 __o 58.65
some some some some some some .53 some some 8 2252.8
$8.5 some so $85 $82 .5 $85 $82 so e__o> 8m
some some some some some some $8.0 $8.0 s03 22 298
some: $8.: $8.3 $8.3 $8.3 $8.: .55.: $8.: $82 $588
.55 $2.6 $3.0 some some son-o $2.0 s25 .53 :56 555x
seen some $8.5 $2.5 .53. $2.5 so on as 8+ 2 8.5 .835
$8.2 28.2 .53 28.2 .53 $8.2 $8.2 $8.2 38.2 :8
$8.2 $82 .552 $8.2 $8.2 $8.2 28.2 055.2 055.2 as;
as as 5 as we as 2 as a
2:83.... mas—Eon. £250". 2:55". 23:51 «35.0“. Sacco". 2:85”. 2:55". 3:05.95...

 

 

 

 

 

£995 .82 he ems—BEE E :25:th uoEEam @5323 v.23 Enos. i o 035...

55

 

3.2 Processing

Experimental dressings were prepared at a pilot plant scale in a batch
system. Three processing variables, (low, medium and high shear) were applied
to one-third of each 13.67kg batch resulting in twenty-seven different
manufacturing variables. Production of the formulas began with the initial
weighing out of all of materials needed for each batch. Next, the phosphoric acid
was mixed in with the buttermilk using a Hobart mixer while the buttermilk was
being heat to 175°F with a steam jacketed mixing bowl. Buttermilk heating keeps
the protein from precipitating due to the alteration of the pH below the isoelectric
point (4.3 pH).

A large portion of the water is then placed into a regular Hobart mixer.
The dry ingredients sugar and salt, are added to the water. Flavor and spices
would be added at this time during normal production; however, this research
does not utilize spices or ﬂavorings. Sorbic acid is then added to the mixture to
maintain microbiological stability of the ﬁnal product. Potassium sorbate and
sodium benzoate are also utilized for microbiological stability in some commercial
formulas. The solution is mixed for three minutes allowing the dry ingredients to
completely hydrated. The buttermilk and phosphoric acid mixture is then added

under continuous agitation.

The xanthan gum was mixed with a small portion of the soybean oil to
allow the xanthan gum to completely hydrate. The egg yolks, polysorbate 60 and

xanthan gum I oil mixture are then added forming a structure around the oil

56

droplets which stabilizes the emulsion. After the emulsiﬁers are added and
mixed, the soybean oil is added slowly. One Hobart® A200 Mixer was used to
form the emulsion model system, at Kraft FoodsTM pilot plant, Glenview, Illinois.
The mixer was set up with a 20 quart stainless steel bowl and wire loop whip,
and ran at 113 RPMs to form the emulsion (Hobart 2000). The agitation speed at
which the oil is incorporated into the emulsion is very important because

excessive speeds will break the emulsion.

After the ingredients are blended together, the emulsion system is placed
into a small vat and pumped through a positive displacement pump, and a
homogenization hydroshear (designed at Kraft) at 140, 180 and 220 psi (pounds
per square inch) (Figure 8). Hydroshear is an in-line pipe homogenization shear
system through which the product is pumped (Krishnamurthy and others 1998).
The hydroshear has a knife edge homogenization element located within a
closely surrounded impact ring. The product enters the hydroshear through an
oriﬁce located in the valve and seat. Product then proceeds through the in-line
pipe and oriﬁces and is impacted on the edge of the valve and seat. The
collision with the in-line stream impact ring causes a change in the direction of
ﬂow. The product is under pressure as it is forced through this homogenization
unit and is released at atmospheric pressure (Figure 9). One third of each batch
was processed at each pressure variable. Each sample was then packaged into
glass quart jars and labeled wiﬂi a batch number, production date and pressure

variable.

57

.3353 230839: 2 new: «5805 9.5.3.6 22mm :53 6.3 Co 32290 i m 239....

 

 

 

 

38.0mm

 

 

 

 

 

 

seam
552.8

 

 

 

 

are.

 

 

255.33 ca
59523:

 

 

 

 

 

 

 

aEam
on

 

 

 

 

@

 

 

5.2 0.5305

 

 

 

.055. tmnoI

 

 

 

l_ _l

 

 

 

 

 

AE

 

 

 

 

 

58

.352 228 as 258288
8223 9.586 3.3 Ema 8.3 5 355: :5 50:36.3 8055.. coa8c_m_:Eo 25...: Co c260» 320 i m 059...

 

 

5E. Convex...

 

 

 

 

 

 

 

 

 

 

/il VIIIJHE
a ll'

uEm SSE.

 

 

 

 

 

 

 

 

 

\\

 

 

 

 

 

 

59

After packaging, the viscosity, pH, titratable acidity and salt levels were
measured for the nine formulations. A standard pH meter was utilized. Also a
Brookﬁeld Viscometer was used to determine the initial viscosity. This
preliminary data was used to identify variances in initial processing formulation

variables.

The viscosity measurement, was done immediately after packaging of the
product. A Brookﬁeld Viscometer, with a #4 rotating disk or spindle, was used to
measure viscosity at 20 rpm under ambient temperatures. The torque required
to maintain constant rotation was recorded as viscosity in centipoise, cP. This
empirical information provides a quick viscosity check of the model dressing
systems prepared in relationship to a standard ranch dressing which has a

standard viscosity of 8000cP.

Titration with a standard base, also known as titratable acidity, was
performed to determine the percent acid in a sample. A 0.1 N NaOH solution of
the stock alkaline solution was used. The pH was also measured using a
standard pH meter to determine the logarithm of the reciprocal of the hydrogen

ion concentration.

3.3 Rheology
The HAAKE Viscotester© 550 (HAAKE, 1996), equipped with the MV1
concentric cylinder sensor system was utilized for this research, (Table 7). The

VT-550 is a rotational viscometer used to examine the rheological properties of

60

ﬂuid substances. The MV1 system (Figure 10) includes a rotating cylinder, which
has a recessed top and bottom to minimize end effects that may adversely
inﬂuence the torque. This system was temperature controlled with the HAAKE
CZ water bath system. Samples were stored in ambient and refrigerated
conditions prior to rheological testing. The testing protocol was applied using the
RheoWin© software program provided by HAAKE. A series of three independent
tests were performed sequentially on each dressing system: 1) steady shear at
10 s'1 for 60 s, 2) ramp up and down from a shear rate range of 1 to 50 s'1 over a
period of 6 min., 3) steady shear at 10 s’1 for 120s. Shear stress, 0' (Pa) and
apparent viscosity, 11 (Pa s) were determined from experimental data. All tests
were conducted at 5°C and 25°C, and replicated ﬁve times.

The ﬁve replications were performed after the product aged 0 — 3 months
from the date of manufacture. Two additional replications were performed once
the product aged a total of 6 months. This was performed to determine if the

emulsion viscosity was changing at the product aged.

3.4 Particle Size

The Horiba LA-500© laser scattering particle size distribution analyzer and
LA-500© software were utilized for the measurement of oil lipid particles in the
base dressings. Measurement of the particles, suspended in a liquid, is possible

with this instrument over a diameter range of 0.1 urn - 200nm. This analyzer

61

Table 7 HAAKE Viscotester© 550 MV I Speciﬁcations (HAAKE, 1996)

 

 

 

 

 

Sensor System MV1
Inner Cylinder (Rotor)
Radius Rb (mm) 20.04
Height L (mm) 60.0
Outer Cylinder (Cup)
Radius RC (mm) 21.0
Radii Ratio R31 R. 1.05
Gap Width (mm) 0.96
Sample Volume V (cm?) 34.0

 

 

62

 

 

 

 

 

 

 

 

 

_,| k_ Rc=21.0mm
——p{ '1— R b: 20.4mm

Figure 10 - MVI concentric cylinder testing apparatus showing the bob recessed
top and bottom to minimize end effects.

63

measures the particle size distribution through the application of laser diffraction.
The light intensity of the laser at the photoelectric detector and the diffracted light
pattern of particles is related to the particle radius (Figure 11). Light diffraction is
based on the various sizes of particles in a sample and the light diffraction
patterns that occur as a result of their size.

In the LA-500 system, the sample travels past the laser by a circulating
pump while being dispersed and stirred in an ultrasonic chamber, (Figure 12).
The sample is then pumped in front of a Helium - Neon laser. The diameter of
the laser beam emitted is increased by a beam expander and radiated upon the
particles suspended in the liquid. Once the light has been dispersed and
diffracted by the particles in the ﬂow cell, the laser beam passes through a
condenser lens and an image is formed on the photoelectric detector. The
intensity of the diffracted light is converted to an electrical signal which is used to
calculate the particle size distribution (Horiba Ltd. 1999). This instrument is very
easy to use on emulsion systems.
To start the particle size analysis, samples are prepared in two solutions, a 0.2%
SDS (Sodium Dodecyl Sulfate) and a 2% NaCl solutions and are placed in the
ultrasonic chamber. A 1gm sample of the base emulsion dressing system is
placed in a glass vial. To this is added 99m of either the 0.2% SDS or 2% NaCl
solution. The sample is mixed until the emulsion is dispersed.

An aliquot of each sample is added to the analyzer with a disposable

pipette until the required concentration level has been met for the analyzer to

£236 26: 29:233.: comm. 0.20:3 unto... i 2 2:9...

 

 

 

.2098
85.02922“.

 

 

 

\r.
29.. .85: 283
so: .020: El

 

\iiiliiiN

 

//%

 

 

 

225a

 

 

 

.mucmaxm
Emcm

 

 

 

 

\ .

 

 

 

 

39.. 520.28 W

 

 

 

 

2.0. 528280

 

 

 

 

LemmJ

 

 

 

 

 

 

65

 

.82 8.3:. 8.... «sea 2.... 2.2.28 2 58...... see... 8222 85.5 3%: m5 .6 8.55280 i 2 2:9“.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a 859.50 9.55 5.5.32.0
\ L
l—
.33 028...
PO
82350 .5322
.256 Lovemnxm :53
8...:
”85.5280 I
\
.900u8 0%0_wi0uocm v1 .1 5096050
0.5.5-8“...
1 hr

 

 

 

66

begin the analysis. Samples were circulated through the LA-500 system at two
circulation levels, high and low speed. These speeds allow for measurements at
two different circulation rates. The sample is then held at the recommended

concentration level for one minute. At this time a particle size reading is taken.

After two minutes another reading is taken. Comparison of these to time
variable readings allows for analysis of particle size changes over time. A read
out of the analysis is displayed on the main screen of the instrument. Data are
then sent to a remote computer, through the LA-500 Horiba software program,
where graphs and charts of the actual data points can be displayed, printed and

saved for future reference.

3.5 Statistical Design and Analysis

The effects of the variables, xanthan gum at 0.15%, 0.30% and 0.45%,
egg yolk at 0%, 1.5% and 3.0% and applied processing rate at 140, 180 and 220
psi and the interactions at the three shear rates on the rheological properties and
particle size analysis was analyzed with a response surface analysis factorial
design. This analysis was performed using the JMP statistical software 2.0 at
Kraft Foods, Glenview IL. The design determined the signiﬁcant effects of the

mean values of apparent viscosity and the mean particle size.

67

CHAPTER 4
RESULTS AND DISCUSSION

4.1 Effect of Formula and Processing on Rheological Behavior

Analysis of the inﬂuence of the formulation and processing rate on the
rheological properties of a model salad dressing system were determined
through viscosity testing upon production of the samples, and latter rotational
viscometry at 25°C and 5°C. The rheological data were analyzed using response
surface analysis, to identify the main ingredients or combination of ingredients

that inﬂuenced viscosity.

Nine formulations, with varying levels of xanthan (0.15%, 0.30% and
0.45%) and egg yolk (0%, 1.5% and 3.0%) were produced, under three
processing shear rates through the hydro shear unit using differential pressures
of 140, 180 and 200 psi. Initial observations showed that samples varied in color
from very white in formula 1 to pale yellow in formula 9. This results from the
xanthan gum level affecting the emulsiﬁcation of the oil droplets. As the samples
were removed and loaded into the rheometer, the texture appeared thicker as the
level of xanthan gum increased to 0.45%. No visual differences were detected

from increased levels of egg yolk or differential pressures.

Empirical rheological testing performed at the time of production showed
that the Brookﬁeld viscosity of the model systems ranged from 2240 cP - 20100

cP. This compared well to standard Ranch dressing that has an initial viscosity

68

of 8000 CF. These initial measurements indicate an increase in viscosity and
creaminess as the xanthan level increased from 0.15% to 0.45% (Table 8).
There is also a slight increase in viscosity due to the increase in egg yolks. No
pattern could be seen as the processing rate (differential pressure) increased.
The pH of the model dressing system samples ranged from 3.18 - 3.55 for the
various formulas. These samples have a pH in the typical range for salad

dressing formulations (Stauffer 1996).

The samples produced exhibit a wide range of textures. The rheological
behavior of these samples was compared to four competitive brands of basic
Ranch dressing, (Kraft Ranch ©, Hellman’s Ranch©, Wishbone Ranch ©,
Hidden Valley Ranch ©). The decreasing apparent viscosity with increasing
shear rate was shown to be very similar for all four commercial samples. When
compared to three of the model dressing systems processed at a processing rate
of 140 psi, all measured at 25°C, it can be seen that the viscosity of the
commercial dressings is similar to formula 5, which contained 0.30% xanthan
gum and 1.5% egg yolks. When compared to formula 1 containing 0.15%
xanthan gum and 0% egg yolk, the commercial dressings have a much higher
apparent viscosity at low shear rates. The opposite is true for formula 9 at the
highest xanthan gum level of 0.45% and highest egg yolk level of 3.0%, where
the apparent viscosity is lower for the commerical samples at shear rates of less

that 20 s", Figure 13.

69

Table 8 - Initial Brookfleld viscosity, pH, titratable acidity and salt analysis

of the model salad dressing samples.

 

 

 

 

 

 

 

 

 

 

 

Formula Processing Viscosity* PH Titratable Salt
Rate [(21 Acidity

1 140 2240 3.3 0.36% 2.13%
1 180 2960
1 220 3360
2 140 5120 3.32 0.36% 2.28%
2 180 5040
2 220 4880
3 140 6160 3.49 0.37% 2.46%
3 180 6080
3 220 6080
4 140 6400 3.18 0.39% 2.16%
4 180 7040
4 220 8000
5 140 7760 3.36 0.37% 2.27%
5 180 7760
5 220 14000
6 140 14400 3.5 0.38% 2.47%
6 180 12400
6 220 10700
7 140 11840 3.2 0.39% 2.13%
7 180 11600
7 220 1170
8 140 15300 3.37 0.38% 2.31%
8 180 19500
8 220 16700
9 140 15400 3.55 0.36% 2.40%
9 180 20100
9 220 17800

 

“Viscosity - measured with a Brookﬁeld at 20 rpm, base ranch = 8000cP

7O

.808 as .3 o2 .5 __<. ea 85:2 some... 25

9.6.. was ego...” 9.53:8 .m «.256. ucm .Eam 858x $8.0 ucw 50> 08 some 9.55:8 .m 53:52 .53 858x
$9... ncm 50> was $0 9.58:8 .P «.256. 2 88858 5 8.958 .m.2oEE8 .0 359.8 30658:”. 9 umber.

 

 

   

 

3:85.32»
8 on 9. 8 cm 2 o
L hi P lr _ o
Tr..."u.-...-.-...-.-.-.....-. ..... .. .N
55> 58.: I .-_....... .
, .2.
8:85.? fauna.
25:58". (ﬁr
z... a
.M.

5. .. m
amigo“. :Ev. w
-2 m
0
mainten. W
, .2 l
u
rs.

55:55.8... . . 3

. .2

. 2

. om

 

 

 

71

4.1.1 Time Dependent Behavior

Prior to the shear sweep performed on all twenty-seven samples at both
25°C and 5°C, the time—dependent behavior of the dressing was analyzed using
a controlled shear rate of 10 s’1 for 60 seconds at each temperature. This test
showed the samples became time-independent after 60 s of preshear, Figure 14.
Formulas 1, 5 and 9 were chosen to represent the increasing level of xanthan
gum and egg yolk levels of the samples while all formulas exhibited the sameﬂow

behavior.

The time-dependent behavior was small and diminished after 30 s. As the
amount of xanthan gum increased from 0.15% (formula 1) to 0.30% (formula 5)
and then to 0.45% (formula 9) the shear stress increased from 10-12 Pa (at
0.15% xanthan gum), to 35-40 Pa (0.30% xanthan gum) and 57-62 Pa (0.45%
xanthan). This increasing level of xanthan gum resulted in smaller particles and
more pronounced time-dependent behavior. Preshear eliminated the time-
dependent effects from the subsequent sweep tests, allowing for all samples to

be evaluated as time-independent fluids.

A shear rate sweep was performed, from 1 s'1 to 50 s‘1 for 180 seconds at
each temperature, 25°C and 5°C. The power law fluid model was utilized to
determine the consistency coefﬁcient (K) and the flow behavior index (n) from

resulting rheograms. Apparent viscosity was calculated at: 10 s“, 25 s‘1 and

72

 

 

.m.o>o. 9.882: “885.9 3 new?» .85 92mm 00mm 5 .0323 5:: EoucoaouéEF I 3 2:9“.

 

 

 

 

A3 o5...
8 on 9. 8 8 2 o
. _ _ . . o
- S
VN_=E.—On_ III-IllIhIiIIlilIvIII-l:-illlhIII-IIIIIIIIIIII
.ma oi. .ou
. an S
_ 09 .3 cum N
,8 u
80,“; 369:... .. .. a f§~$ «.6
m macho“. m
.3 o3 \ . ow m
.3 cm? h.‘
.3 o- . cm
a «36.5”. lullllltlll'l all WINII l I - .
a - 8
.m o: .8 o2 .3 2N

 

 

on

73

40 s", with 25 s“ being reported in Table 8 and 9.

After the shear sweep, a ﬁnal time-dependency sweep was performed at a
controlled shear rate of 10 s'1 for 60 seconds at each temperature. This test
showed that all of the samples did not exhibit residual time—dependent behavior
(Figure 15). Formulas 1, 5 and 9 can be compared to Figure 14, where it can be
seen that the samples exhibited some degree of permanent structural
breakdown. Formula 9, prior to the shear sweep, exhibited a shear stress of 57-
62 Pa 3, while after the sweep the shear stress was 48-50 Pa s. The time-
independent behavior of this ﬂuid after minimal shear is a prime product
characteristic. If these samples were highly time-dependent ﬂuids, it would
propose a larger operations challenge due to breakdown issues associated with

pumping, packaging, and use by the consumer.

The rheograms, the apparent viscosity versus shear rate, showed that the

ﬂow behavior followed the power law ﬂuid model given in equation (8):

n = r6) = K G)" I; = K (7)“ (8)

The apparent viscosity at 25 s", the consistency coefﬁcient (K) and the ﬂow
behavior index (n) at 25°C and 5°C, calculated from the shear rates and apparent
viscosity are listed in Tables 9 and 10. It can be seen that all n values are less
than 1 indicating shear-thinning behavior. The consistency coefﬁcient increased

as the level of xanthan gum increased: at 0.15% xanthan gum and 0% egg yolks,

74

A", O
L.

s.

IMP- -1
l - I
A .

411

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EN 000.0 00.0 0NN 0NN 0 0.0 0.0
SN 0000 00.0 N.0N 00v 0 0.0 0.0
EN 0000 00.0 0NN 03 0 0.0 0.0
00.N 500.0 000 N0? 0NN 0 0.0 0.?
NON 000.0 00.0 0.: 00? 0 0.0 0...
0vN 000.0 000 N5 03 0 0.0 0.?
N0.N 30.0 0Y0 0.: 0NN v 0.0 0
0N.N 000.0 00.0 0.3 00? v 0.0 0
FNN N000 0Y0 N0? 03 v 0.0 0
004 30.0 0v.0 50.0 0NN 0 09.0 0.0
P0,? 30.0 30 N50 00? 0 09.0 0.0
0.1 30.0 000 N00 03 0 0rd 0.0
00... 000.0 010 N P N 0NN N 0 .. .0 0.?
3: 30.0 :0 N05 03 N 9.0 0...
00.? 000.0 0Y0 FNN 03 N 9.0 0.?
N; 000.0 0v.0 3.0 0NN F 0 F .0 0
0: 500.0 $0 N00 00? F 0.0 0
50.0 000.0 00.0 :06 03 F 0.0 0
a me.
we 3 u 5 a e 3 pm a... :8. as as
. mm : v. a < 0.388“. :50 :mﬁcmx mv=o> 00m

 

 

00m «6 acagzgch Daznﬂgﬁ 00.: _OUOE hO 0_0>_0Cﬂ _GO_DO_O¢£¢ n G 305—.

75

VP in“... uh: 31411:“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

R0 000.0 N00 N00 0NN 0 0V.0 0.0
V0.0 000.0 0.0 0.00 00? 0 0V.0 0.0
00.0 000.0 00 V00 0V? 0 0V.0 0.0
V0.0 000.0 00.0 F.0N 0NN 0 0V.0 0. F
00.0 . 000.0 N00 NNO 00F 0 0V.0 0...
00.0 000.0 V0.0 N.0N 0V? 0 0V.0 0. F
00.0 000.0 v0.0 0.V0 0NN x. 0V.0 0
0N.0 000.0 V0.0 V.0N 00F \. 0V.0 0
0N.0 000.0 00.0 V.0N 03 \. 0V.0 0
0.. we.
7... a u . a e 3 pm we. .3. .s s
. mm : v. a < 0.35:0“. :50 :mﬁcmx mv=o> 00w

 

 

6.2.8. .. a 2.3

76

 

 

.000 0.00.8 8.00:: 3:80.80 :0 .88.:
n 020?. .000. 050008.: 0.000 n 020?. .830 :05:0x .N000.0 n 0203 2.2. 000 .000 5 005000.: 0N.0 0.0....0: I 0

”00.0v 00203 :0 .0000.0 0205 5.0803 0.0000 3:005:20 05000:: 80:200.: I o
..00.0v 0203 .5080; 0.0000 20:85:08 :50 :0£:0x I n .0000 .0 020?. 5.0803 0.0000 20:00:20.0 2.0.. 000 I 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

77

 

 

 

 

 

 

 

:.::.: 000.0 .:.« 8:: .0: 0.0. :«« 0 0.: :0
:80: 08.: 0..« :8: .0: «0. :0. 0 0: :0
9.00.: :«0: .:.« 8:: .0: 0.0. :3 0 0: :0
:80: :00: «0.. 8:: :0: :0. :«« 0 0.: 0..
3:. :80 «0.. 8:: 00.: :0. :0. 0 0.: 0..
8... «3.0 .0. «8.: 8: :0. ::. 0 0.: 0..
:80: 08: 0..« .8: «0.: 0:. :«« : 0.: :
:80: :000 «V. 8:: 8: «:. :0. : 0.: :
8...: :«00 0... 8:: 8: 0.«. o... : 0: :
8.0: :8: 8.. 8:: :0: :0: :«« 0 0.: :0
:80: :8: :«.. :8: 8.: 2.: :0. 0 0.: :0
:«:0.: «8: «... 8:: 9.: 8.0 ::. 0 0.: :0
:03: 80: :... 8:: 0:: 3.. :«« « 0.: 0..
:02: «:0: 8.. 8:: 0:: .:.0 :0. « 0.: 0..
:«8: :8: 8.. 8:: 0:: .:.0 ::. « 0.: 0..
:80: R0: :8: 8:: 0...: 8: :«« . 0.: :
:08: :.:.0 «:0: 8:: ::: «0: :0. . 0.: :
:08: «0:0 000.: 8:: «0.: 8.0 o... . 0.: :
.>00 056 Am an:

.20 8.0 e E: .0 0« u . .0 .. 3 p0 a... :3. 0: 0.

00.00.. 8.0 22:0: . N: e v. a < 0.25:“. :50 005:3. 9:2. 8:

 

 

 

 

OomN «ﬂ 02030—32552 Wcmﬂﬂeﬁ 528 _OUOE hO 0—0230 _8.WO_OO£¢ I OF 209—.

 

 

 

00.0 2020: 05.00.00 3:005:20 :0 00000
n 020>. 0.0.. 05000090 0.000 n 020>. .800 :05:0x 00000 n 020> . .0._0> 000 ”wow :_ 005000.: 00.0 000.00 I 0

.000v 0020?. :0 0000.0 020>. 300003 0.0000 3:005:20 05000.0 0.20.050 I 0
..00.0v 020>. .>._0000_> 0.00.00 3:005:20 0.50 :0:.:0x I 0 ”000.0 .0 020>. 300003 0.0000 3:005:20 0._0> 000 I 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0000.0 000.: ~00 000.0 00.0 .00 0N0 0 0:0 0.0
0000.0 :00: 00.0 000.0 0N.0 0.00 00. 0 0:.0 0.0
00.0.0 00:0 :00 000.0 0N.0 0.0m 0:. 0 0:.0 0.0
00.00 00.0 00.0 000.0 00.0 :00 0- 0 0:.0 0..
00... 000.0 000 000.0 00.0 0.00 00. 0 0:.0 0..
00... :000 000 000.0 .00 000 0:. 0 0:0 0..
0000.0 000.: :00. 000.0 00.0 0.0m 0mm 0 0:0 0
00.00 000.0 .00 000.0 .00 0.00 00. 0 0:0 0
0000.0 8:0 0:0 000.0 00.0 ~00 0:. 0 0:0 0

$00

.20 2.... .0 0:.

00.0 a 5.. .0 0« u . .0 .. 3 0.0 0:. :8. a: .0
0.0000 00.0 0.0000 . mm. : v. 0 < 0.08.00. :50 :05:0x 020> 00m

 

00.0.80. I :. 0.000

78

 

ow.

.0_0>0_ 0500000.: €0.05: .0 0002.0 .0000 .000 oo0~ .0 .0_>0...00 0.2.. .:00:0:00 0:.... I 0. 0.02“.

 

 

 

.0. 2....
0m. 00. 00 00 0: 0m 0
. . . . . . o
. 0.
. 03.0.00. . _ u __ _

.00 0 :«

. o... .0: 8. .0: :««
0 00

a! r it!"\ 1- ij‘ l ﬁ‘a‘.G 31"‘Vh— 3’3‘14"
7‘ | H I H 4

0 0.08.0". - - - .

(8:) ms Janus

 

 

 

0 0.08.0".

 

.0: o... .0: :0. .0: :««

 

co

 

79

is 4.072 — 6.467, over the range of differential pressures, 0.45% xanthan gum
and 3.0% egg yolks, K is 33.427 - 33.719, over the range of differential

pressures.

To determine if the viscosity of the dressing systems was changing over
time, additional rheological testing was performed after aging. Five replications
of the shear sweep were conducted when the product was 1 to 3 months old.
Two more replications were performed to determine if the emulsion was
changing over time. These replications were performed when the product was 6
months old. In Figure 16, at 25°C it can be seen that for formula 1 and 9 at 140
psi and 220 psi there is little change between the pooled data of the ﬁve
replications and the pooled data of the two replications with aging. The
rheogram shows that the effect on apparent viscosity for the ﬁrst ﬁve replications
is very close to the two completed after the product aged. This was also seen for
formulas 2-8 at all three differential pressures. Formula 1 and 9 were chosen,
and are represented in Figure 16, because they represent the two extremes of
xanthan gum levels of 0.15% and 0.45% and egg yolk levels of 0% and 3.0%.
Paredes (1989) also showed, through measurements of creep compliance, that
salad dressing emulsions utilizing xanthan gum as a stabilizer, maintained

emulsion stability during storage.

4.1.2 Xanthan Gum / Egg Yolks

The ingredient producing greatest effect on the viscosity is xanthan gum.

Egg yolks only have a slight effect on the viscosity compared to xanthan gum.

80

.0_0>0_ 0500000.: 03. .0 8:00.: 0000 0:0 000.. .0 020505000... .0 000:0:88 ”00.00 .0 080.3 I 0.. 0.00.0.

.0:. 300 .020

 

F 030.30.
a 0.065“.

  

r
m

 

3:00.: 0000 I .0: 03
0:00.: :00... .0: o!

8:00.: 0000 I .0: omm \/

.0309: £00.. I .0: 000

Y
o
P

(82:) 07mm :umddv

{a

 

 

mm

 

81

As the xanthan gum level is increased from 0.15% to 0.30% and 0.45%, with the
egg yolk level held at 1.5%, the apparent viscosity increases an average of 0.70

Pa s at a shear rate of 25 s".

Xanthan gum, which affects the creaminess, is a viscosity builder and
provides emulsion stability (Phillips and Williams 1995). It has the greatest
impact on the viscosity due to its rod like conformation forming entanglements
(Figure 17). In 25°C temperature conditions, the apparent viscosity increases, as
the xanthan level increased from 0.15% (formula 1), to 0.30% (formula 5) and

ﬁnally 0.45% (formula 8).

The greatest apparent viscosity difference is seen at shear rates less than
20 s“. This is a result of the xanthan gum polymers forming reversible
entanglements. As stated by Urlacher and Noble in 1997, this behavior can be
advantageous, where increasing the shear rate can decrease the viscosity of a

xanthan gum solution.

This is especially important during the production of dressings which have
formulas with higher xanthan levels or if over application occurred. During
pumping and mixing a higher shear rate could be applied if too much xanthan
gum was added to the product in order to move and package the product. But
upon used by the consumer issues could arise, where a higher shear rate would
be needed to use the product, due to high levels of xanthan gum. The amount of
shear needing to be applied by the consumer would increase where the typical

shear rate during pouring is in a range of 10 -100 s“. The product would

82

 

comm .0 .0: emu 000 9.8.. 000 $04 .0 E00 c0500.. 00.0008... .0 0.0th I t 0.50.“.

.0:. 800 .020

:50 Seamx $03 .. N 032.0:

:50 55.5. $8.: I 0 035.0:

(sod) hlsoosm wonddv

:50 00500.: $00.0 I 0 03.0.0...

 

 

 

mm

83

essentially become spoonable, not pourable giving it a whole new flow behavior

proﬁle and product identity.

Xanthan gum has a strong effect on the K and n values, table 9 and 10.
The K value increases as the xanthan level increases. It can also be seen in

these tables that egg yolk also affects K, although not as signiﬁcantly as xanthan

gum.

Xanthan gum was shown to signiﬁcantly affect apparent viscosity as the
levels increased from 0.15% to 0.30% and 0.45%. This effect is independent of
egg yolk or processing rates increasing. It is the result of the xanthan gum being
dispersed in oil and then mixed in the sample to acheive hydration during
production. The increase in xanthan gum level resulted in a stable emulsion with
increased apparent viscosity. This is due to the mechanism described by
Hennock and others 1984, where it was proposed that xanthan gum is absorbed
at the oil [water inteIface, lowering surface tension, with the remaining xanthan
gum stabilizing the emulsion by physically trapping the emulsion droplets. As the
xanthan gum level increased in each formula, the level surrounding each droplet
increased and therefore increased the viscosity. The level of xanthan gum in a
dressing formula needs to be carefully controlled due to the rheological impact

and subsequent effect on production and consumption.

Egg yolks have only a slight effect on the viscosity of xanthan gum. In

addition to slightly affecting the viscosity, they provide emulsion stability due to

 

 

the emulsifying activity of the lecithin protein complex, Iipoproteins attaching to
the oil droplets at the oil interface (Ford and others 1997). Egg yolks also
provide a protective barrier around the droplets due to steric stabilization and add
color and flavor attributes to the product. Insufﬁcient egg yolk emulsiﬁer covering
droplets results in droplets having a higher tendency to coalesce (Stang and
others 1996). Therefore, an emulsiﬁer like egg yolk is necessary to properly form

a stable emulsion for long term product stability and increased shelf life.

While egg yolks were shown to statistically have a signiﬁcant effect on
apparent viscosity through response surface analysis, f value = 0.0005, in a
rheogram the increasing level of egg yolk appears to have only a slight effect on
apparent viscosity, Figure 18 and 19. In Figure 18 as the egg yolk levels
increase from 0% to 1.5% and to 3.0%, egg yolk has very little inﬂuence on
apparent viscosity at all three levels of xanthan gum, while all three formulas
exhibit shear-thinning behavior. In contrast, Carillo and Kokini (1988) found that
egg yolk and egg yolk plus salt had an effect on the rheological properties of a
model oil in water salad dressing emulsion. They found that the egg yolk
signiﬁcantly contributed to the stability and viscosity of the model system. While
in this study egg yolk provided stability but did not greatly affect viscosity.

In Figure 19, as the egg yolk level increases from 0% to 1.5% and 3.0%
and the xanthan level is held at 0.30%, there is a slight increase in apparent
viscosity. This was also seen at 0.15% and 0.45% xanthan gum. There is no

effect seen as the processing rate increases. This shows that some egg yolk is

85

 

 

 

 

.0000 .0 .0: 000 .0 E00 :05:0x 05000.05 0:0 $0.0 0:0 $0.. .$0 .0 50> 000 05000.05 .0 080.3 I 00 0.00.0.

 

om

 

. II
.I . .
Ill

' I I II II - | I I l I ' l '

0 050.5 .
N 0.08.00.
0 0_:E.o”_

$0.20 E00 :05:0x

0:90.01 505w

    

v 0_:E.00_
m 0.:E.0... .
0 03.0.0“. /

$000 :50 :05:0x
n 0:. . on.
0 0:.—E00.

0 0_:E.0....
\

$000 :50 :05:0x

row

.9

 

 

(sad) MIsoosIA Iuaieddv

86

 

.OomN .0 $00.0

.0 .05. E00 :0:.:0x 0:0 50.. 000 05000.05 5.? .0: 000 0:0 00. .03 0000.05 0.0. 0500000.: .0 .00..m I 0_. 0.00.”.

00

.0:. 30: .020
00 on on

or

 

"‘

0 0.:E.0
$0.0 0...... :00

m 0.:E.0.
$0.. 0.0.. 0:0

0 030.8.
- $0 0:0» 000

   

(sad) hlsoosm aumeddv

T
O
‘_

1 NF

 

 

 

VP

87

necessary to obtain emulsion stability in a dressing system. Carillo and Kokini
(1988) showed through creep analysis, steady shear and particle size analysis
that emulsions are most stable in the presence of 2% egg yolk with 0% salt or
3% egg yolk with 2% salt. In this study at levels greater than 1.5% the effect of
egg yolk on the apparent viscosity has a plateau effect, and egg yolks are not

controlling viscosity.

4.1.3 Processing Rate

Model dressing samples were produced under processing rates with
increasing differential pressures of 140, 180 and 220 psi. The effect of this
process variable on apparent viscosity was signiﬁcant but only slightly compared
to xanthan gum.

As can be seen in Figure 20, at all three levels of xanthan gum, the
rheogram is unchanged as the processing rate increases, while the egg yolk
level is held constant at 3.0%. The ﬂuid exhibits shear-thinning behavior, with
the apparent viscosity being strongly affected by xanthan gum. The process rate
does not have a signiﬁcant effect on the apparent viscosity in conjunction with
the other factors, or independently.

As the processing rate increase, with the egg yolk level at 0% and xanthan
gum at 0.45% in formula 7, a very slight increase can be seen in apparent

viscosity at 220 psi.

At 140 and 180 psi, no change was seen in an apparent viscosity at any

shear rate (Figure 21). The same trend, is seen in formula 1 and 4

88

 

 

 

 

 

$0.0 .0 .90. 0.0.. 000 0:0 000...:

0:0 $00.0 $0.0 E00 :0:.:0x 05000.05 5.3 .0: 000 0:0 00. .03 0.0. 0500000.: 05000.05 .0 .00..w I 00 0.00. n.

.0:; 30¢ .ucsm

 

 

 

00 00 0: 00 00 0. 0
. . I . . :
l.l. -- - - -- -III- .0
I , r .1
/
I I I / I V
I I.. //
0 0_0E.0... - $0.0 E00 :0:.:0x / ,., 0 «v
/ / r
/./ / m
/ , u
/ v 0 M
0 0_0E.00_ - $00.0 E00 :0:.:0x , ﬂ
/, . .. m.
/. m
_ . e 0
/ 0
.
0 0_0E.00_ - $00.0 E00 :0:.:0x . 0 z
. 7 0.

 

 

0..

 

89

with a little change in apparent viscosity evident as processing rate or egg yolk

levels increase.

In all the samples evaluated, a stable emulsion was formed through the
use of the hydroshear homogenization unit under a range of differential
pressures ranging from 140 - 220 psi. As the pressure variables were applied
the oil droplets formed new interfaces allowing the emulsiﬁers, egg yolks and
polysorbate 60, to diffuse to the interface and therefore lower the interfacial
tension as the droplets were formed. As seen from the rheograms, the droplets
formed and stabilized during processing through the hydroshear did not alter the

apparent viscosity.

Response surface analysis showed that egg yolk and processing rate
signiﬁcantly affect the apparent viscosity but only slightly compared to the effect
of xanthan gum. Response surface analysis also showed that egg yolk and the
processing rate independently affect viscosity. The effect of viscosity is the same

no matter what the rate or egg yolk levels are.

This study shows that the utilization of a hydroshear unit for this model
system allows for production of a dressing system under a pressure range of 140
- 220 psi to be utilized without signiﬁcantly affecting the product viscosity. This
range is large and opens up a wide range of ﬂexibility for the production of this
dressing system due to the low impact that the processing rate on the rheological

characteristics, at low and high levels of egg yolks and xanthan gum. Ultimately

90

 

. . I L _
Inﬂa’lduﬂ: a... :u}

   

I
0:0! .0: :00
00500.. $0...0 . .. 0.0.0.0. .0: .mmmv
000.00.. $0.0 . 0 0.0.0.0.
.0: 00m
.0: 00.
.08.:
E00 00500.. $00.0 . .. 0.0.0.0.

 

10H

.0.

 

 

.000

.0 $000 0:0 $00.0 $0.0 E00 :0....:0x 05000.05 000 05.0.. 000 $0 0...... 0.0. 00000.: 05000.05 .0 0.00..w I .0 0.00. n.
.0:. 0.0: .020

(sed) Ausoosm Iaweddv

91

operations could run as high as 220 psi, resulting in an increase in throughput

without negatively affecting the product integrity.
4.1.4 Temperature

Temperature can greatly affect the ﬂow behavior of ﬂuid systems. In this
study, the effects of 5°C and 25°C on the rheological behavior were determined.
Temperatures were chosen to mimic the typical temperatures that consumers

encounter during consumption of salad dressing.

In the initial time-dependent test, temperature dependency was also
evaluated, under a controlled shear rate of 10 s‘1 for 60 seconds at 5°C and
25°C. The apparent viscosity was determined to be only slightly higher for the
same sample at the colder temperature of 5°C than at 25°C, Figure 22. This

shows that the model dressing system is slightly temperature dependent.

As can be seen from the rheogram, Figure 22, the apparent viscosity of
the dressing is only slightly dependent upon temperature especially at 0.45%
xanthan gum and 3.0% egg yolk levels at low shear rates, where the ﬂuid
exhibits shear thinning behavior. At the 0.15% xanthan gum level and 0% egg
yolks the apparent viscosity becomes constant at higher shear rates. This shows
that the apparent viscosity is only slightly affected by the temperature. A

consumer would see little difference during pouring.

These temperature effects are important to note in developing new

dressing systems and for improving upon existing systems. The effect of the

92

 

 

 

 

.0_0>0_ 50> 000 $0.0 0:0 $0 000 E00 000.00.. $00.0 0:0 $0.0 .0 0.0.0.0:E0. .0 0080000. I «N 0.00. n.

.02.. 30m. .0000

 

. . 1 , 0
.0: o... , I
.0: 00. .0: 00. /
.0::0: .0: 8. _/ / , / «w
I 00: .0: 0:: . / / a
I .0. u
00 / , w
. .0: o: . A
50> 000 $0 E00 :00.:0x $0. 0 .0: 00. / m.
I . 0.0000. .0800 _ m
I 000 .0: o... _ .0. m.
.0: :0. . w...
.0: :00 .0
I 00
50> 000 $0.0 E00 00...:0x $000 .8

I 0 0.0.0.0.

 

 

 

mm

93

varying temperature conditions is also important in maintaining similar product

characteristics during the van'ety of temperatures seen by consumers.

4.2 Particle Size Inﬂuence

The size of the lipid particles in salad dressing is a very important
characteristic due to its impact on appearance, texture, rheology, stability, and
ﬂavor. Reduction of droplet size causes a signiﬁcant increase in interfacial area.
Newly formed droplets will coalesce if not stabilized by emulsiﬁers. Analysis of
the inﬂuence of the formulation including, xanthan gum, egg yolk level, and the
processing rate (differential pressure) on the particle size was determined
through the measurement of the lipid particle size of each formula using a light

scattering particle size distribution analyzer.

To measure the particle size of the lipid droplets in each formula required
dilution in two solutions. The samples were ﬁrst diluted in SDS to obtain a
uniform dispersion of droplets, including the breaking up of any aggregates.
Particles distributed in SDS were analyzed and a mean particle size and the
surface area was determine through the average of two readings per sample.
The ﬁrst reading was at one minute, and the second at three minutes (Table 9).
Mean particle size of formulas with 0% egg yolks and 0.15% xanthan gum
ranged from 5.432 pm at a processing rate of 140 psi to 4.877 pm at 220 psi.

The use of SDS as a diluent for particle size analysis was successful.

The greatest range of particle size seen in a formula as a result of

94

ﬁll.“

 

 

increased processing rates from 140 psi to 220 psi was found in formula 5 (1 .5%
egg yolk, 0.3% xanthan gum). At 140 psi the particle size was 6.442 pm and at
220 psi it was 4.550 pm, giving a 1.892 pm difference. These results show that
overall as the processing rate increased the particle size decreased (Figure 23).
Overall response surface analysis showed that in an $08 solution, egg yolk,
xanthan gum and processing rate all signiﬁcantly affected the particle size. The
processing rate (differential pressure) had the strongest effect on particle size

(Table 9).

Particle size of each formula at each processing rate was also measured
in a NaCl solution. Measurements were taken under slow and fast circulation
past the laser and through the circulation pump. The mean particle size was
determined through the average of one and three minute readings. Three

replications were taken as a result of the variability seen during the readings.

At slow and fast circulation, the lipid particles began to aggregate
spontaneously in the NaCl solution, especially under slow circulation. This
caused an inaccurate measurement of the lipid particle size. The formation of
these aggregates in the NaCl solution is a result of the amphilic lipid molecules
being associated physically and aggregating due the NaCl ions. The mean
particle sizes varied no matter how much shear was applied (by the fast or slow
circulation) to break aggregates. Overall the results show that the use of NaCl as
a diluent not effective due to the ionic charge of the particles in the solution

resulting in spontaneous aggregation of the lipid droplets.

95

Wh‘f in n_

 

0 0.0E.00.

 

 

 

 

.0: 8:0
.0: 00. I

.0: o!-

 

 

.0~.0 0.0...0: :0 00000.00. 0... 5 05000.: 0000.050 00000.00. .0 .00..m I 00 0.00. h.

0 0.0E.00.

 

 

.. 0.0E.0...

 

0 0.0E.0...

'9. -£'1.iT-“r- m."

 

' ' '- .-‘l‘; mtz'ﬂIf.!7“T E‘.‘Aav.w-

 

 

0 0.0E.0....

 

 

v 0.0E.0.,.

 

 

0 0.0E.00.

 

 

N 0.0E.00.

 

. 0.0E.00.

 

 

 

96

Xanthan gum and egg yolks were shown to have a small effect on particle
size. The particle size is the smallest at high levels of xanthan gum (0.45%) and
egg yolk (3.0%). Particle size was not affected at 0% and 3% egg yolk as
xanthan increased. At 1.5% egg yolk, as the xanthan gum level increases the
particle size increases. The particle size for all of the samples was shown to be
the largest (6.399 um) at 1.5% egg yolks and 0.45% xanthan gum at 180 psi
(Figure 24).

As egg yolk level increased from 0, to 1.5% and 3.0% the particle size was
shown to decrease at any differential pressure (Figure 25). Figure 25 is a
representative example, for all xanthan gum levels, showing a decrease in
particle size as egg yolk levels increase at a constant xanthan gum level. These
results are consistant with the study performed by Carillo and Kokini (1988).
They determined that a shift in the particle size distribution towards small values
was seen with addition of egg yolk and salt, and a shift towards larger values

when egg yolk and salt were not present.

In this study the effect of the xanthan gum on the particle size was shown
to be signiﬁcant, in SDS. It was also shown that xanthan gum and egg yolks
together signiﬁcantly inﬂuence particle size at speciﬁc levels. At high levels of

xanthan gum and high levels of egg yolk the particle size is smaller.

Xanthan gum is an effective stabilizer that affects the creaminess,
viscosity and emulsion stability while the egg yolk is an effective emulsiﬁer. The

various xanthan gum and egg yolk levels resulting in different interaction effects

97

 

.0: 00. .0 00.0 0.500: :0 0.05. E00 000.00.. 000 50> 000 05000.05 .0 0.00..m I 00 0.00. H.

.0. .26.. 50> 000
0 0..

 

 

 

 

(um) ezls opined

 

 

 

 

$00.0 $00.0

 

$00.0

 

 

 

E00 000.00.. .0090: n

98

.050 0.500: :0 $0.0 .0 _0>0. E00 000.00.. .00.0:00 0 .0 .50> 000 .0 0.05. 00000.00. .0 .00..m I 00 0.00.0.

00:. 20000.: .0..00.0...n.

 

 

 

 

 

 

000 00. 0...
I 0
I .
w
I N mu
0
01
I 0 My.
7.
O
- m,
.. I0.
- 0
000.00.. $0. 0 50> 000 $0 0... m
. 0

 

000.00.. $0. 0 50> 000 $0 .-

000.:0x $0.0 50> 00m. $0.”. ,

99

on the particle size is important to understand to maintain the desired emulsion
stability. As the lipid droplet was sheared through the hydroshear unit the lipid
particle surface area increased due to the formation of new droplets. This
allowed for the egg yolk to be absorbed on the droplet surface. The absorption
of the emulsiﬁer to the interface minimized coalescence of the newly formed

droplets.

In addition to the egg yolk acting as an emulsiﬁer, the dressing
formulations also included the emulsiﬁer polysorbate 60. This was held at a
constant level in all of samples. Polysorbate 60 is a very fast emulsiﬁer due to
its small molecular size compared to the larger molecules of the egg yolk
Iipoprotein complex. The level of the xanthan gum in conjunction with the level of
the egg yolk signiﬁcantly inﬂuencing the overall emulsion stability, ﬂow behavior,
with minimal effects on the particle size. Overall xanthan gum and egg yolk do
not affect the particle size as signiﬁcantly as the processing rate (differential

pressure).

4.2.1 Inﬂuence of pressure drOp

Particle size decreases as the differential pressure in the hydroshear
increases. Response surface analysis showed that processing rate
independently affects particle size. This independence can be seen in the
interaction of egg yolk and process rate: as the egg yolk increases, regardless of

rate, the particle size decreases, Figure 25.

100

 

 

Control and reduction of particle size as a result of changing the
differential pressure is important in forming a stable emulsion system. Optimizing
the shear proﬁle of the hydroshear or any other emulsiﬁcation device is important
in reducing the oil droplet size. This may make it possible to reduce the total oil

used while maintaining a target rheological proﬁle.

4.2.2 Inﬂuence of particle size on viscosity

As the processing rate (differential pressure over the hydroshear) was
increased causing a decrease in particle size, no effect was seen on the
viscosity. The particle size decreases as the processing rate (differential
pressure) increases. Even as the particle size decreased the viscosity does not
change, but remains ﬂat. No correlation was seen between the particle size and
the viscosity as is depicted in Figure 26. This correlation was seen with all nine

formulations.

101

 

 

00000005 0.00 050000000 0.05 0.00:0 000 22:00 0:000> 300005 I mm 0590.

 

 

 

 

 

 

m 0.0—Eon. m 0.3.50... 0 03.50". m 030:0". m 0_:E..0u_ N 038.0...
0 I - o
, ,
0... . F
F . . 0
d
A .
q 0. . m w
3
w a.
S
a m
.. I
a 0 . .. v m
0.0 , ., 0
000- 080510.: ,
8. - £082>Ill
03 - .0080;qu
. 0

000 - 000 0.2000” 0 .
our - 0N5 0_0_.._0a|
o: - 000 20.000I

 

 

0.0.. I.-- I-I.I .II III.. I.

102

Chapter 5

Summary and Conclusions

The effect of formulation and particle size on the rheological properties of
salad dressing suggests that the viscosity at 25 s'1 for this slightly time-
dependent, shear-thinning behavior salad dressing is greatly inﬂuenced by

xanthan gum. Particle size is affected by processing rate.

Temperature was shown to have a minor effect on the viscosity. Salad
dressing exhibited temperature dependency as the temperature was increased
from 5°C to 25°C. A 1.0 Pa 3 increase in viscosity was seen between the two
temperatures. At a shear rate of 25 s", for a dressing containing 3.0% egg yolks
and 0.45% xanthan gum, at 220 psi, the viscosity was 2.72 Pa s at 25°C and

3.77 Pa s at 5°C.

The optimal rheological behavior was found with a formulation of 0.45%
xanthan gum and 1.5% egg yolks. In this formula the particle size observed was
6.534 pm at 140 psi, 6.388 pm at 180 psi, and 5.172 pm at 220 psi. Varying the
level of xanthan gum in the formulation affects the viscosity of the dressing
system; hence, this ingredient must be very carefully controlled during
production. While the egg yolks provide emulsion stability, as the egg yolk level
is increased, concurrently with increasing xanthan gum levels, only a slight effect

is seen on the apparent viscosity. Egg yolks have a very small effect on the

103

 

viscosity compared to xanthan gum.

Increasing the processing rate (differential pressure over the hydroshear)
did not affect viscosity, but had a strong effect on the particle size. During
processing, a differential pressure range of 140 — 220 psi could be utilized with
minimal effects on viscosity. This fact means the manufacturer can operate over
a wide range of pressure while still providing the desired product viscosity. The
increase in process rate, due to an increase in differential pressure across the
hydroshear, had the greatest effect on the mean lipid particle size, for all

formulas.

Xanthan gum, egg yolks and processing rate all signiﬁcantly affect particle
size, however the processing rate has the greatest effect. High levels of xanthan
gum and high levels of egg yolk result in a smaller particle sizes. As the egg yolk
level was decreased, while the xanthan gum and processing rate were held
constant the particle size decreased. Also as the processing rate (differential
pressure over the hydroshear) was increased the particle size decreased and
there was no change in viscosity. Even as the particle size decreased the
viscosity does not change. Overall no correlation was seen between the particle

size and the viscosity.

104

 

 

 

 

Chapter 6

Future Research

Future research in rheological characterization of salad dressings due to
formulation and processing conditions could be expanded to include: further
particle size testing on the existing system, measurement of the stability provided
by egg yolks, measurement and comparison of other types of model dressing

systems, ingredients, and emulsiﬁcation equipment.

The comparison of the measurement of the particle size prior to the shear
sweep and after the shear sweep, to determine the degree of permanent
structural breakdown due to shear would provide further information on the
system. This could also determine any aggregation, or change in particle size,

that is occurring as a result of the shear rate sweep.

The stability provided by the egg yolks at each level could also be
evaluated. This would yield a better understanding of the exact level of egg
yolk. lf proven to be signiﬁcant or provide a cost savings the use of a variety of
emulsiﬁers in dressing formulation could also be tested for product development

or for product improvement.

The application and study of a wider variety of emulsiﬁcation equipment,
including the standard homogenizer, could be utilized to produce a product under

a variety of processing conditions. This would determine the effects of different

105

 

processes on a control formulation. It would also be interesting to understand
the relationship of formulation to multiple processing conditions and the resulting

viscosity.

106

 

Bibliography

Association for Dressings and Sauces. Facts and Forklore — The History of Salad
Dressings. [Online] Available

ht_tg:/I_wy_v\_r.dressings-saucesorg/folklore.html, October 11, 2000.
Bistany KL, Kokini JL. 1983. Comparison of steady shear rhoelogical properties

and small amplitude dynamic viscoelastic properties of ﬂuid food materials. J
Texture Studies 14(2): 113-124.

Cadle RD. 1965. Particle Size. New York: Reinhold Publishing. p 1-30.

Carrillo AR, Kokini JL. 1988. Effect of egg yolk and egg yolk + salt on rheological
properties and particle size distribution of model oil-in-water salad dressing
emulsions. J Food Sci 53(5): 1352-1366.

Chang CM, Powrie WD, Fennema O. 1972. Electron microscopy of
mayonnaise. J. Inst Can Sci Tech 5(3) 134-137.

Code of Federal Regulations. 21 .CFR.161.

Coia KA, Stauffer KR. 1987. Shelf life study of oil/water emulsions using various
commercial hydrocolloids. J Food Sci 52(1): 166-172.

Dickenson E. 1994. Emulsion stability. In: Nishinari K, Doi E, editors. Food
hydrocolloids: Structures, properties and functions. New York: Plenum Press. p
387-396.

Elliott JH, Ganz AJ. 1977. Salad dressings- preliminary rheological
characterization. J Texture Studies 8(3): 359-371.

Elejalde CC, Kokini JL. 1992. The psychophysics of pouring, spreading and in-
mouth viscosity. J Texture Studies 23(3): 315-336.

Figoni Pl, Shoemaker CF. 1983. Characterization of time dependent ﬂow
properties of mayonnaise under steady shear. J Texture Studies 14(4): 431-442.

Ford LD, Bomankar R, Martin Jr. RW, Holcomb DN. 1997. Dressings and
sauces. In: Fn'berg SE, Larsson K, editors. Food emulsions. 3'6 ed. New York:
Marcel Dekker, Inc. p 361-412.

HAAKE. 1996. Viscotester VT 550 manual. Paramus, NJ.

107

 

Hasenhuettl GL, Hartel RW. 1997. Food emulsiﬁers and their applications. New
York: Chapman 8 Hill. p 1-9.

Heil JR, Perkins WE, McCarthy MJ. 1990. Use of magnetic resonance
procedures for measunnent of oil in french-style dressings. J Food Sci 55(35):
763-884.

Hennock M, Rahalkar RR, Richmond P. 1984. Effect of xanthan gum upon the
rheology and stability of oil-water emulsions. J Food Sci 49(5): 1271-1274.

Hobart Corporation, Food Equipement. 2000. A200 Mixer. Ohio. p 1-4.

Horiba LTD. 1999. Laser scattering particle size distribution analyzer, LA-920.
BulletinzHRD-36398.

Kiosseoglou VD, Sherman P. 1983. The rheological conditions associated with
judgement of pourability and spreadability of salad dressings. J of Texture
Studies 14(3): 277-282.

Kiosseoglou VD, Sherman P. 1983. Inﬂuence of egg yolk Iipoproteins on the
rheology and stability of OM emulsions and mayonnaise 1. Viscoelasticity of
goundnut oil-in-water emulsions and mayonnaise. J of Texture Studies 14(3):
397-417.

Krishnamurthy RG, Witte VC. 1998. Methods of making a reduced fat emulsiﬁed
dressing. United States Patent 5,795,614. p 1-5.

Langton M, Astrom A, Herrnansson A. 1999. Microstructure in relation to the
textural properties of mayonnaise. In: Dickenson E, Patino editor. Food
emulsions and foams: Interfaces, interactions and stability. New York: Plenum
Press. p 367-375.

McClements DJ. 1999. Food emulsions: Principles, practice and techniques.
New York: CRC Press.

Munoz J, Sherman P. 1990. Dynamic viscoelastic properties of some
commercial salad dressings. J Texture Sci 21(4): 411-426.

Narwar, W. 1986. Lipids. In: Fennema O editor. Food chemistry. 2nd ed. New
York: Marcel Dekker, Inc. p 139-244.

Orr C. 1988. Determination of particle size. In: Becher P, editor. Enclyclopedia
of emulsion technology, volume 3 basic theory, measurement, applications. New
York: Marcel Decker. p 137-169.

108

 

Paredes MDC, Rao MA, Boume MC. 1988. Rheological characterization of
salad dressings 1: Steady shear, thixotrophy and effect of temperature. J Texture
Sci 19(3): 247-258.

Paredes MDC, Rao MA, Bourne MC. 1989. Rheological characterization of
salad dressings 2: Effect of storage. J Texture Sci 20(2): 235-250.

Peressini D, Sensidoni A, Cindio BD. 1998. Rheological characterization of
traditional and light mayonnaises. J Food Eng 35(4): 409-417.

Phillips GO, Williams PA. 1990. Gums and Stabilisers for the food industry 5.
Oxford: Oxford Univeristy Press.

Rao, MA. 1999. Rheology of ﬂuids and semisolid foods, principles and
applications. Aspen Publishers. Maryland. p219—250.

Rao, MA, Steffe JF. 1997. Measuring yield stress of ﬂuid foods. Food
Technology. 51(2):50-52.

Rao, MA, Anantheswaran RC. 1982. Rheology of ﬂuids in food processing.
Food Technology. 36(2) 116 -126.

Rao, MA. 1977. Rheology of liquid foods - a review. J Texture Studies
8(2):135—168.

Sherman P. 1963. Rheology of Emulsions. Oxford: Pergamon Press.

Sherman P. 1967. Changes in the rheological properties of emulsions on aging
II. J. Colloid & Int. Sci. 24:67.

Stang M, Karbstein H, Schubert H. 1996. Inﬂuence of emulsiﬁer adsorption
kinetics and emulsiﬁcation machine construction on dispersity of oil-in-water
emulsions. In: Dickenson E, editor. Food Colloids: Proteins, Lipids and
Polysaccharides. New York: Plenum Press. p 382-392.

Stauffer CE. 1996. Fats and oils. Minnesota: Eagan Press. p 101-107.

Steffe JF. 1996. Rheological methods in food process engineering, 2nd ed. East
Lansing, Michigan: Freeman Press.

Tiu C, Boger DV. 1974. Complete rheological characterization of time-
dependent food products. J Texture Studies 5(3):329-338.

109

 

Tung MA, Paulson AT. 1995. Rheological concepts for probing ingredient
interactions in food systems. In: Gaonkar AG, editior. Ingredient Interactions
Effects on Food Quality. New York: Marcel Dekker, Inc. p 45-83.

Urlacher B, Nobel O. 1997. Xanthan gum. In: lmeson editor. Thickening and
gelling agents for food. 2"‘1 ed. London: Blackie Acedemic 8 Professional. p
285-311.

Van Den Temple M. 1963. Effect of droplet ﬂocculation on emulsion viscosity. In:
Sherman P, editor. Rheology of emulsions. New York: Mac Millan Company.

Walstra P. 1996. Emulsion stability. In: Becher P, editor. Encyclopedia of
emulsion technology. Volume 4. New York: Marcel Dekker, Inc. p 1-59.

110

 

I. Irl..ll.l|!|..[