..-...Hq1 . . .13
3.321,. f. A
E

i. F 14;.
. i.

. “a?”
PMFH .

o.

. h

ubv

1 9”;
J»?

aw?!" V.
. .mL,.v!Jm“_.-.4.. r.

c

is.” H. .., . ....

H ..n‘: .. . V inf.

.

”Mgr;

"3! p“

«3.. .

:2. ill
I.

"ti-e
. ... f
at H!
u {A};
..u. :i'
7 l
1m» ‘

f.
.‘1

#‘fltlff
- :E a}: ‘ .

finisuflb . . . . ‘ ‘ . . , ‘ ., ‘ . H. , ..

L. ‘ . . ., ..H .. .. ‘ . ‘ .. ‘ ‘ , . . {Mannrl‘qfiwfir .. . _ “chm:
1‘ m.,,vJ.4.a_r.r2 . .u H. .‘ .. . z . . . .. .H u... w. n . . .. 1.. .W.

‘ F.
...
, v I
... A . .1 4L
.

.1. _.
., .7. A ‘ .
u may...“ .L .. K ‘ ha..:-;...,w
u.....$.,m£n.q_,%ififi
4 '~

.
l.‘
imam“. ¢
1-“ "
n T.

.-..JHFI.
4 :1. A0?!

 

 

El}; L ...‘urp'lrrwb A. Sulfa. 3b? Miahcflwigrt'ym [zomttalnlwiimuulbiflu3:5. .75.»:-

.’ b .51 I1.-

THESIS

(Q
8 Doc)

lllllllllllllllllllllHHllllHIHHllllillllllllllllllllllll

31293 O 1 820 0299

This is to certify that the

thesis entitled

THE RATE OF LIPID OXIDATION OF A PRODUCT MODEL SYSTEM
IUKCHCAKSEH)IFLAJWTRDNUILAITTILJPREKShDAITflDILAJMHDDATTEFTLLA

Date

STRUCTURES

presented by

Youn Suk Lee

has been accepted towards fulfillment
of the requirements for

Master . Packaging

degree in

01 L W4

Major professér

 

 

Augustll,l999

 

0—7639

 

MS U is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY

Mlchlgan State
Unlverslty

 

 

 

PLACE IN REI‘URN Box to remove this checkout from your record.
TO AVOID FINE-3 return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE I DATE DUE

DATE DUE

 

M1} 3”0“20”03|

 

£3.14! 2%?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M c/CIRCJDatoDuost-ou

-WM ._.._ » .—‘a——-—-—fi____

THE RATE OF LIPID OXIDATION OF A PRODUCT MODEL SYSTEM
PACKAGED IN AN TIOXIDANT IMPREGNATED LAMINATE FILM
STRUCTURES

By

Youn Suk Lee

A THESIS

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

MASTER OF SCIENCE

School of Packaging

1999

ABSTRACT

THE RATE OF LIPID OXIDATION OF A PRODUCT MODEL SYSTEM
PACKAGED 1N ANTIOXIDANT IMPREGNATED LAMINATE FILM
STRUCTURES
By

Youn Suk Lee

A fieeze—dried model food product system was developed as the source for the
autoxidation of linoleic acid in storage stability studies. Moisture sorption isotherms for
the model product were determined at 18, 28, 38°C. The sorption isotherms of the model
product were fitted to the Halsey model. However, for the model product with low levels
of linoleic acid, at 38°C, the Henderson model was found to give the best fit.

The efi‘ectiveness of an antioxidant impregnated film to retard autoxidation of a
packaged model product containing linoleic acid, via the evaporation-sorption
mechanism, was evaluated as a fimction of storage time and temperature. The rate of loss
of BHT from the package film structure was found to be much higher than the rate of loss
of a—tocopherol, at both storage conditions (23°C and 45°C, 50%RH).

The effectiveness of BHT and a—tocopherol impregnated laminated film pouches
_ showed no significant difl‘erent (p<0.05), due to the lack of lipid oxidation in the model
food product, when stored at 23°C, as a function of the time. The BHT hpregnated
laminate pouch showed a notable efl‘ectiveness in retarding lipid oxidation of the model
product at 45°C as a function of storage time. The control (non-antioxidants) and a-
tocopherol impregnated laminate pouch structures showed no effect on retarding lipid

oxidation of the model product during storage at 45°C.

ACKNOWLEDGMENTS

I would like to express my sincere thanks to my major professor, Dr. Jack
Giacin, for his guidance and patience in preparing this thesis.
I would also like to express sincere appreciating to my committee member, Dr.
Theron Downes and Dr. Gale Strasburg for serving on the guidance committee. I
would like to acknowledge Dr. Jack Throck Watson and Dr. Douglas Gage for use of
the GC/MS system at the Mass Spectrometer Laboratory in the Department of
Biochemistry.
Special thanks should also be attribute to many {friends at the School of Packaging for
their support and assistance. This project was supported by the Center for Food and
Pharmaceutical Packaging Research (CFPPR) at School of packaging.
Finally, I would like to express my deepest gratitude to my parents for their love,

support and encouragement.

iii

TABLE OF CONTENTS

LIST OF FIGURES ....................................................................................................... x
INTRODUCTION ......................................................................................................... 1
REVIEW OF LITERATURE ........................................................................................ 4
Lipid Oxidation in Food System ......................................................................... 4
Mechanism of Lipid Oxidation ........................................................................... 6
Secondary Product of Terminal Oxidation Reactions ...................................... 11
Factors Affecting Lipid Oxidation .................................................................... 12
Water Activity ....................................................................................... 12
Moisture Sorption Isotherm .................................................................. 13
Water Activity related with Lipid Oxidation ........................................ 15
Monolayer Moisture Content ................................................................ 17
Temperature .......................................................................................... 18
Oxygen Concentration .......................................................................... 19
Light ...................................................................................................... 20
Trace Metals .......................................................................................... 21
Antioxidants ...................................................................................................... 21
Synthetic Antioxidants .......................................................................... 23
Butylated Hydroxytoluene (BHT) ............................................. 23
Natural Antioxidants ............................................................................. 24
Tocopherol ................................................................................. 25
Antioxidants for the Polymer Plastic Film ........................................................ 27

The Mechanism of the Migration of Antioxidants from the Packaging Film
Structures to a Contained Product ..................................................................... 29
Product Model System ...................................................................................... 35
Analytical Technique for Measurement of Lipid Oxidation in Food System ..36
Gas Chromatography/Mass Spectrometry ........................................................ 37
Isolation Techniques of Volatile Compound Isolation in the Food System ..... 41
Static Headspace Analysis .................................................................... 42
Purge and Trap System (Dynamic Headspace Analysis) ..................... 42
Distillation System ................................................................................ 44
MATERIALS AND METHODS ................................................................................. 45
Product Model ................................................................................................... 45
Packaging Materials .......................................................................................... 47
Initial Moisture Content .................................................................................... 48
Sorption Isotherm .............................................................................................. 49
Moisture Equilibration System of Product Model ............................................ 50
Product Stability Studies ................................................................................... 54

iv

Analytical Procedures for Antioxidants in the Packaged Films ....................... 57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Soxhlet Extraction System ...................................................... 57
Percent Recovery of Antioxidants for Extraction System ...... 58
High Pressure Liquid Chromatography (HPLC) Analysis ..... 59
Analytical Test for Lipid Oxidation in the Model Product ......................... 61
Linoleic Acid Analysis ..................................... ................. 61
Extraction--- 61
Preparation of Methyl Esters 62
Percent Recovery of Linoleic Acid through Extraction and
Methylation ........................................ --64
Hexanal Analysis ---------------------- - -- - - - 64
A Dynamic Purge and Trap System ........................................ 64
The Analysis Procedure of Hexanal compound - - -- -68
Mass Spectrometer .................................................................. 69
Percent Recovery of Hexanal through a Dynamic Purge and
Trap System -------------- 71
RESULTS AND DISCUSSION .................................................................................. 72
Product Model ................................................................................................... 72
Initial Moisture Content .................................................................. - - 73
Equilibrium Sorption Isotherm ......................................................................... 74
The Moisture Content Equilibrium of Model Product at a Specific Water
Activity, Temperature ....................................................................................... 84
Model Product Shelf Life and Stability of Antioxidants - 84
Loss of Antioxidants from Coextruded Laminate Films 85
The rate of Loss of Antioxidants from Laminate Film Structure ..................... 90
Percent Recovery of Antioxidants from the Pouch Film .................................. 94
Level of Linoleic Acid in the Model Product for Storage at 45°C ................... 95
The Percent Recovery of Extraction for Linoleic Acid - - 97
Difl'erence in the Level of Linoleic Acid between Model Product Preparation
and Freezing-Dried Model Product 98
Product Storage Studies ---------- 99
Qualitative Identification of Hexanal Compound in Model Product using the
Selected Ion Monitoring GC/MS Procedures ................................................. 105
SUMMARY AND CONCLUSIONS ........................................................................ 107
FUTURE STUDIES ................................................................................................... 1 10
APPENDICES ........................................................................................................... 112
REFERENCES .......................................................................................................... 165

LIST OF TABLES

Table 1. Relative Amounts of Volatile Compunds from Antuoxidation of Linoleic

 

Acid at Moderated Temperature ................................................................... 10
Table 2. Components of the Model Product (A, B) on a wet weight basis ................. 72
Table 3. The Percent of Linoleic Acid in Model Product (B) calculated .................... 73
Table 4. The Initial Moisture Content of Freeze-Dried Model Products (A,B) .......... 74
Table 5. The Equilibrium Moisture Content for the Freeze-Dried Product (A)
as a Function of Water Activity at 18°C ........................................................ 76
Table 6. The Equilibrium Moisture Content for the F reeze-Dried Product (A)
as a Function of Water Activity at 28°C ........................................................ 77
Table 7. The Equilibrium Moisture Content for the Freeze-Dried Product (A)
as a Function of Water Activity at 38°C ........................................................ 78
Table 8. The Equilibrium Moisture Content for the Freeze-Dried Product (B)
as a Function of Water Activity at 18°C ....................................................... 80
Table 9. The Equilibrium Moisture Content for the F reeze-Dried Product (B)
as a Function of Water Activity at 28°C ....................................................... 81
Table 10. The Equilibrium Moisture Content for the Freeze-Dried Product (B)
as a Function of Water Activity at 38°C ........................................................ 82
Table 11. The Package Film Structures Evaluated ...................................................... 85
Table 12. The Level of Antioxidants from Coextruded Packaging Pouches as a
Function of Storage Time (23°C, 50%RH) .................................................. 86
Table 13. The Level of Antioxidants from Coextruded Packaging Pouches as a
Function of Storage Time (45°C, 50%RH) .................................................. 88
Table 14. Rate Constants for the Loss of Antioxidants from the Packaging Pouch
Structure (23°C, 50%RH) ............................................................................ 91
Table 15. Rate Constant for the Loss of BHT from the Packaging Pouch
Structure (45°C) .................... . ......... - ............. 93
Table 16. The Percent Recovery of BHT for Extraction System ................................ 94
Table 17. The Percent Recovery of a-tocopherol for Extraction System .................... 94

vi

Table 18. The Concentration of Linoleic Acid from the Model Product as a Function

of the Storage Time at 45°C, 50%RH .......................................................... 96
Table 19. The Percent Recovery of Linoleic Acid through the Extraction and
Methylation Procedures ............................................................................... 98
Table 20. The Percent Recovery of the Linoleic Acid in the Model Product Between
Before F reezing-Dry and After Freezing-Dry Condition ............................ 99
Table 21. The Hexanal Concentration of Model Product System Packged in Pouches
Fabricated from Coextruded Laminate Structures (23°C, 50%RH) ......... 100
Table 22. The Hexanal Concentration of Model Product System Packged in Pouches
Fabricated from Coextruded Laminate Structures (45°C, 50%RH) ......... 103
Table 23. The Ratio of the Selected Ion Profiles at m/z 44, 56, and 72 for
Hexanal ...................................................................................................... 106
Table 24. The Thickness of Laminate Structure for the Pouch of the Storage Studies
.................................................................................................................. 112
Table 25. The Initial Moisture Content of the Model Product (A, B) ................ , ....... 112

Table 26. The Conditions of Relative Humidity According to Each Temperature. ..113

Table 27. Experimental Data for Equilibrium Sorption Isotherm of the Model Product

(A) at 18°C (64°F) 114

Table 28. Experimental Data for Equilibrium Sorption Isotherm of the Model Product

(A) at 28°C (84°F) ...................................................................................... 114

Table 29. Experimental Data for Equilibrium Sorption Isotherm of the Model Product

(A) at 38°C (100°F) .................................................................................... 114

Table 30. Experimental Data for Equilibrium Sorption Isotherm of the Model Product
(B) at 18°C (64°F) ...................................................................................... 115

Table 31. Experimental Data for Equilibrium Sorption Isotherm of the Model Product
(B) at 28°C (84°F) ...................................................................................... 115

Table 32. Experimental Data for Equilibrium Sorption Isotherm of the Model Product
(B) at 38°C (100°F) .................................................................................... 115

Table 33. The Isotherm Mathematical Model ............................................................ 116
Table 34. The Data of Model Product (A) for Halsey Equation at 18°C (64°F) ........ 118
Table 35. The Data of Model Product (A) for Halsey Equation at 28°C (84°F) ........ 1 19

V11

Table 36. The Data of Model Product (A) for Halsey Equation at 38°C (100°F) ...... 1.
Table 37. The Data of Model Product (B) for Halsey Equation at 18°C (64°F) ........ 121 -

Table 38. The Data of Model Product (B) for Halsey Equation at 28°C (84°F) ........ 122
Table 39. The Data of Model Product (B) for Henderson Equation at 38°C (100°F) 123

Table 40. The Monolayer Values of High (A) and Low (B) Linoleic Acid Containing

Freeze-Dried Model Product According to Sorption Isotherm at 18°C ..... 125
Table 41. The Monolayer Values of High (A) and Low (B) Linoleic Acid Containing
Freeze-Dried Model Product According to Sorption Isotherm at 28°C ..... 125
Table 42. The Monolayer Values of High (A) and Low (B) Linoleic Acid Containing
F reeze-Dried Model Product According to Sorption Isotherm at 38°C ..... 125
Table 43. The Data of Model Product (A) for B.E.T. Monolayer Value at 18°C (64°F).
.................................................................................................................... 127
Table 44. The Data of Model Product (A) for B.E.T. Monolayer Value at 28°C (84°F).
.................................................................................................................... 128
Table 45. The Data of Model Product (A) for B.E.T. Monolayer Value at 38°C
(100°F) ....................................................................................................... 129
Table 46. The Data of Model Product (B) for B.E.T. Monolayer Value at 18°C (64°F).
.................................................................................................................... 1 30
Table 47. The Data of Model Product (B) for B.E.T. Monolayer Value at 28°C (84°F).
.................................................................................................................... 1 3 1
Table 48. The Data of Model Product (B) for B.E.T. Monolayer Value at 38°C
(100°F) ....................................................................................................... 132

Table 49. The Correlation Coefficient and Sums of Squares fiom the Isotherm Data
and the Respective Isotherm Models (A) at Different Temperatures (18°C,
28°C, 38°C) ................................................................................................ 133

Table 50. The Correlation Coefficient and Sums of Squares fi'om the Isotherm Data
and the Respective Isotherm Models for model product (B) at Different
Temperatures (18°C, 28°C, 38°C) .............................................................. 133

Table 51. The Mathematical Model for Equilibrium Moisture Content from the
Experimental Data of the Model Product A at 18°C (64°F) ...................... 134

Table 52. The Mathematical Model for Equilibrium Moisture Content from the
Experimental Data of the Model Product A at 28°C (84°F) ...................... 135

viii

Table 53. The Mathematical Model for Equilibrium Moisture Content from the
Experimental Data of the Model Product A at 38°C (100°F) .................... 136

Table 54. The Mathematical Model for Equilibrium Moisture Content from the
Experimental Data of the Model Product B at 18°C (64°F) ....................... 137

Table 55. The Mathematical Model for Equilibrium Moisture Content from the
Experimental Data of the Model Product B at 28°C (84°F) ....................... 138

Table 56. The Mathematical Model for Equilibrium Moisture Content fiom the
Experimental Data of the Model Product B at 38°C (100°F) ..................... 139

Table 57. Experimental Antioxidant Data Retained from the Bach Pouch as a
Function of the Storage Time at 23°C, 50 %RH ........................................ 140

Table 58. Experimental Antioxidant Data Retained from the Bach Pouch as a
Function of the Storage Time at 45°C, 50 %RH ........................................ 143

Table 59. The Ratio of the Selected Each Ion Profile at m/z 44, 56, and 72 for
Hexanal ...................................................................................................... 154

Table 60. Experimental Hexanal Data of Model Products from the Bach Pouch as a
Function of the Storage Time at 23°C, 50%RH ......................................... 154

Table 61. Experimental Hexanal Data of Model Products from the Bach Pouch as a
Function of the Storage Time at 45°C, 50%RH ......................................... 157

Table 62. Experimental Methyl Ester Data of Model Products as a Function of the
Storage Time at 45°C, 50%RH .................................................................. 158

ix

LIST OF FIGURES

Figure l. The Typical Autoxidation Pathway of Linoleic Acid .................................... 9
Figure 2. The Rate of Lipid Oxidation according to the Effect of Aw on the Food
System .......................................................................................................... 15
Figure 3. A Three Step Migration Process ................................................................... 29
Figure 4. A Schematic Diagram of the Equilibrium System ....................................... 52
Figure 5. The Moisture Equilibration System for Product Model ............................... 53

Figure 6. The Storage Arrangement of the Pouched Model Product at 23°C, 45°C.... 55

Figure 7. Flow Diagram of the Test Scheme ............................................................... 56
' Figure 8. A Schematic Diagram of the Dynamic Purge and Trap System .................. 66
Figure 9. A Dynamic Purge and Trap System for the Model Product ......................... 67
Figure 10. The Sorption Isotherm at 18°C of Model Product (A). .............................. 76
Figure 11. The Sorption Isotherm at 28°C of Model Product (A). .............................. 77
Figure 12. The Sorption Isotherm at 38°C of Model Product (A). .............................. 78
Figure 13. The Sorption Isotherms at 18, 28, 38°C for Model Product (A). ............... 79
Figure 14. The Sorption Isotherm at 18°C of Model Product (B). .............................. 80
Figure 15. The Sorption Isotherm at 28°C of Model Product (B). .............................. 81
Figure 16. The Sorption Isotherm at 38°C of Model Product (B). .............................. 82
Figure 17. The Sorption Isotherms at 18, 28, 38°C for Model Product (B) ................. 83

Figure 18. The Relative Percent Loss at BHT and a-tocopherol from Coextruded
Pouch F ihns as a Function at Storage Time (23°C) ................................... 87

Figure 19. The Relative Percent Loss at BHT and oc-tocopherol from Coextruded
Pouch Films as a Function at Storage Time (45°C) ................................... 89

Figure 20. The Relative Percent Loss of BHT and a-tocopherol from the Pouch
Films as a Function at Storage Time (23°C) .............................................. 92

Figure 21. The Relative Percent Loss of BHT and a-tocopherol from the Pouch
Films as a Function of Storage Time (45°C) ............................................. 93

Figure 22. The Concentration of Linoleic Acid in the Model Product as a Function of
Storage Time (45°C) .................................................................................. 95

Figure 23. The Concentration of the Hexanal in the Model Product System Packaged
in Pouches Fabricated from Coextruded Laminate Film Structures
(23°C, 50%RH) ........................................................................................ 101

Figure 24. The Concentration of the Hexanal in the Model Product System Packaged
in Pouches Fabricated from Coextruded Laminate Film Structures
(45°C, 50%RH) ........................................................................................ 104

Figure 25. The Plot of Linear Regression of Halsey Equation for Model Product (A)
at 18°C ....................................................................................................... 118

Figure 26. The Plot of Linear Regression of Halsey Equation for Model Product (A)
at 28°C ....................................................................................................... 119

Figure 27. The Plot of Linear Regression of Halsey Equation for Model Product (A)
at 38°C ....................................................................................................... 120

Figure 28. The Plot of Linear Regression of Halsey Equation for Model Product (B)
at 18°C ....................................................................................................... 121

Figure 29. The Plot of Linear Regression of Halsey Equation for Model Product (B)
at 28°C ....................................................................................................... 122

Figure 30. The Plot of Linear Regression of Henderson Equation for Model Product
(B) at 38°C ................................................................................................ 123

Figure 31. The Linear Regression Plot of B.E.T. Equation for the Monolayer Value of
Model Product (A) at 18°C ....................................................................... 127

Figure 32. The Linear Regression Plot of B.E.T. Equation for the Monolayer Value of
Model Product (A) at 28°C ...................................................................... 128

Figure 33. The Linear Regression Plot of B.E.T. Equation for the Monolayer Value of
Model Product (A) at 38°C ...................................................................... 129

Figure 34. The Linear Regression Plot of B.E.T. Equation for the Monolayer Value of
Model Product (B) at 18°C ...................................................................... 130

Figure 35. The Linear Regression Plot of B.E.T. Equation for the Monolayer Value of
Model Product (B) at 28°C ...................................................................... 131

Figure 36. The Linear Regression Plot of B.E.T. Equation for the Monolayer Value of
Model Product (B) at 38°C ...................................................................... 132

xi

Figure 37. Standard Calibration Curve for BHT using HPLC ................................... 145

Figure 38. Standard Calibration Curve for a—tocopherol using HPLC ...................... 145
Figure 39. HPLC-Chromatogram of an Extracted BHT from the pouched Film ...... 146
Figure 40. I-IPLC-Chromatogram of an Extracted a-tocopherol from the Pouched
Film ............................................................................................................................ 146
Figure 41. GC Chromatogram of the Methyl Ester Extracted from the Model Product
................................................................................................................... 147
Figure 42. Standard Calibration Curve of Trans- 9-12-Octadecadienoic Methyl Ester
................................................................................................................... 148
Figure 43. Standard Calibration Curve for Hexanal using GC/MS ........................... 148
Figure 44. Standard Mass Spectrum of Hexanal ....................................................... 149
Figure 45. The Model Product (B) after Freezing-Dry (0.1645 g) using SIM of Mass
Spectrum .................................................................................................. 150
Figure 46. The Linoleic acid (0.0241 g), (99%, Acros Organics) using SIM of Mass
Spectrum .................................................................................................. 150
Figure 47. The Hexanal Standard Compound using SIM of Mass Spectrum ............ 151
Figure 48. The Model Product (B) after Freezing-Dry (0.1645 g) using SIM of Mass
Spectrum .................................................................................................. 152
Figure 49. The Linoleic Acid (0.0241 g), (99%, Acros Organics) using SIM of Mass
Spectrum .................................................................................................. 152
Figure 50. The Hexanal Standard Compound using SIM using Mass Spectrum ...... 153

Figure 51. High Density Polyethylene Side of HDPE/Suryln-EVA Laminate Film
Structure with No Antioxidants ................................................................ 159

Figure 52. Heat Seal Layer (Suryln-EVA) Side of HDPE/Suryln-EVA Laminate Film
Structure with No Antioxidants ................................................................ 160

Figure 53. High Density Polyethylene Side of BHT Impregnated HDPE/Suryln-EVA
Laminate Film Structure ........................................................................... 161

Figure 54. Heat Seal Layer (Suryln-EVA) Side of BHT Impregnated HDPE/Suryln-
EVA Laminate Film Structures ................................................................ 162

xii

Figure 55. High Density Polyethylene Side of a-tocopherol Impregnated
HDPE/Suryln-EVA Laminate Film Structure .......................................... 163

Figure 56. Heat Seal Layer (Suryln-EVA) Side of a-tocopherol Impregnated
I-IDPE/Suryln-EVA Laminate Film Structure .......................................... 164

xiii

INTRODUCTION

Oxidation is a major limiting factor in the shelf life of lipid-containing food
_ products. The first products of lipid oxidation are hydroperoxides, which are colorless,
tasteless, and ordorless. Hydroperoxides, however, break down to low molecular weight
compounds, which impart flavors and odors to food products, that are associated with
rancidity. The secondary products of oxidation are free radicals, peroxides, epoxides,
aldehydes, ketones, cyclic monomers, dimers, and polycyclic aromatic hydrocarbons;
many of which are toxic (Bidlack, and Tappel, 1973).

Several factors can influence the rate of lipid oxidation. Jean (1984) reported that
factors such as light, relative humidity, temperature, and availability of oxygen affect the
production of lipid oxidation products. Quast and Karel (1971) reported that lipid
oxidation is likely the most common mechanism of oxygen uptake in fried foods, such as
potato chips. For fried foods, the extent of oxidation of the frying oil also contributes to
the product’s oxidation.

Since oxygen first attacks food at the surface, impregnating packaging materials with an
antioxidant has long been used to protect the product from oxidative rancidity and thus
prolong shelf life (Laermer et al., 1994). The proposed mechanism of antioxidant
activity involves the following 3 step process: (i) antioxidant diffusion through the
polymer bulk phase; (ii) evaporation of antioxidant from the surface of the packaging

material; and (iii) subsequent antioxidant sorption onto the surface of the packaged

product.

3,5-Di-Tert-Butyl-4-Hydroxy Toluene (BHT) has been used extensively for its
antioxidant activity. The application of BHT for its antioxidant properties in foods has
been suggested for use in a variety of products (Giese, 1996). Hoojjat et al. (1987)
demonstrated the efiectiveness of a BHT impregnated film to retard lipid oxidation of a
packaged oatmeal cereal, through the migration of antioxidant from the package to the
product via an evaporation/sorption mechanism. However, because of questions related
to safety of the migratory BHT from high density polyethylene (I-DDPE) to foods and food
simulants (Till et al., 1982), there is some concern regarding its continuous use as an
antioxidant in food packaging materials. As a result of these considerations, there has
been a growing interest in using a-tocopherol (i.e. Vitamin E) as an alternative
antioxidant for polymer processing.

Laermer and Nabholtz (1990), Laermer and Zambetti (1992), and Laermer et a1.
(1993) have recently shown that a—tocopherol, when used at low concentrations, such as
100 ppm (w/w), is an effective processing stabilizer for polyethylene and polypropylene.
(at-Tocopherol alone or in combination with synergistic additives such as tris(2,4-di-
butylphenyl) phosphite compares favorably with, and in some cases is superior to,
lrganox 1010 or 1076 as a color and melt flow stabilizer.

Further, Ho et al. (1994) have shown the efl‘ectiveness of a—tocopherol as an
antioxidant for reducing the ofi-odor and off-taste from blow molded I-IDPE bottles, and
Laermer et al. (1994) showed noticeable retention in flavor for a cereal product stored in
an a-tocopherol impregnated package structure. Compared to BHT, a-tocopherol is a
larger, less volatile molecule, and would be expected to migrate less rapidly from a

packaging material into a food or food sirnulant.

The present study will therefore focus specifically on developing an accurate
understanding of how the rate of lipid oxidation of a packaged product is affected by both
the nature and level of antioxidant incorporated within the package structure and the
temperature dependency of the oxidation process. Such knowledge is essential for
estimating the keeping quality of a packaged product, where quality is associated with

‘lipid oxidation.

The objectives of the study include:

1. Development of a freeze-dried product model using linoleic acid as the oxidizable
substrate and characterization of the product equilibrium sorption isotherms as a function
of temperature.

2. Evaluate the effectiveness of antioxidant impregnated film in preventing or
retarding lipid oxidation of the packaged model substrate via the evaporation-sorption
mechanism. The extent of fatty acid oxidation will be determined as a function of the '
nature and level of antioxidant and temperature, using hexanal as an index of oxidation.
All oxidation studies will be carried out at constant water activity. The specific water

activity level will be based on the products equilibrium sorption isotherm.

It is assumed that the results of these studies would enable better design and selection of
packaging systems for controlled transfer of antioxidant, and the concomitant reduction
in product lipid oxidation. The temperature studies should also improve our

understanding of the effectiveness of direct addition of antioxidants in products.

REVIEW OF LITERLATURE

2-1. Lipid Oxidation in Food System

Oxidation of food products is a major factor in the development of off-flavors due
to oxidative chemical changes of lipids. Lipids are one of the major constituents in many
food systems. Most food products of plant and animal origin contain measurable levels
of unsaturated fatty acids or lipids (Moran and Rajah, 1994). The main cause of lipid
degradation involves oxidative and hydrolytic reactions. The oxidative reaction, referred
to as autoxidation, can occur from progressive reactions with atmospheric oxygen to
produce free radicals. Here, autoxidation is an autocatalytic reaction, where the rate of
oxidation increases with reaction time. In such a case, the oxidation induction period or
the initiation step is a slow process and then increases as the oxidation reaction
progresses (Chan, 1987). These oxidative reactions induce unstable physical and
chemical changes in the stored food product. Frankel (1991) described the autoxidative
mechanism that leads to the formation of free radicals to produce off-flavors from
polyunsaturated edible oils. The hydrolytic reactions occur by the reaction of lipase
enzymes, which act on the triglyceride ester linkage, resulting in the hydrolysis of lipids.
Controlling temperature conditions can minimize the oxidation and hydrolytic reactions
associated with enzymatic activity.

The oxidation of lipids is catalyzed by the presence of light, trace metals, lipoxygenase
enzymes, or the effect of heat to initiate the production of free radicals. The fi'ee radicals

react with oxygen to yield the primary products of lipid hydroperoxides. Unsaturated

fatty acids are susceptible more specifically to oxygen attack leading to the free-radical
mechanism. The hydroperoxides formed from the reaction of unsaturated fatty acids with
oxygen are also catalyzed by the action of lipoxygenase (J adhav, et a1., 1996).

Also, for the case of food systems with relatively low amounts of unsaturated fats,
oxidative rancidity can still play a role 'as the main influence of deterioration of food
quality during storage. The off-flavors arising in a food system with higher amounts of
tmsaturated fatty acids can be absorbed in the lipid, since most of the ofilflavor
compormds are lipophilic, volatile compounds (Roozen, et 01., 1994).

F ujii, et 01., (1995) studied the autoxidation of methyl linoleate in emulsions containing
dextrin. As the concentration of the dextrin increased, the autoxidation of the methyl
linoleate decreased. Roozen, et al., (1994) observed that a model food system containing
linoleic acid showed a large difference in the amount and composition of volatile
oxidation products formed, depending upon the lipid concentration (low and high fatty
acids), types of emulsified oils (oleic sunflower oil, triolein and stripped corn oil), and the
presence of antioxidants (i.e., a-tocopherol).

The main concerns involving the oxidation of a stored food product occur due to the lipid
oxidation, and are typically addressed by controlling the environmental conditions of
product storage, which include: temperature, relative humidity, and light. The
incorporation of antioxidants into food products also helps to extend the product shelf
life, by retarding oxidation processes under normal storage conditions. The oxidative
process in food systems containing unsaturated fatty acids can be retarded by adding
antioxidants, either directly to the food products or by incorporating them into the

packaging materials (Coulter, 1988).

2-2. Mechanism of Lipid Oxidation

The mechanism of lipid oxidation has been well defined by a number of studies
(Chan, 1987; Frankel, 1984) and can be described as an autocatalytic chain reaction that
can be viewed as involving three major steps, namely:

(1) initiation, (2) propagation, and (3) termination. The oxidative mechanism is
influenced by several factors such as oxygen, light, heat, heavy metals, and enzymes.

Below, each step of the autoxidation mechanism is discussed briefly.

Initiation: RH ——-) R o + o H
RH+02 ——) Ro-HOOH

Propagation: R o + 02 —-) R00 0
ROOO+RH———> ROOH+R¢

Termination: R0 + R0 —-) R
R0 + R0. ——> ROR
R0 + ROO- ——) ROOR

The initiation step occurs when a labile hydrogen is abstracted from the unsaturated lipid
(RH) to produce the alkyl radical (R c). In the propagation step, oxygen reacts rapidly
with the alkyl radical to form the peroxyl radical (ROO 0). The peroxyl radical reacts

with a hydrogen atom abstracted fiom another unsaturated lipid molecule to form an

unstable hydroperoxide (ROOH), and another free alkyl radical (R 0). The
decomposition of hydroperoxides (ROOH) leads to the formation of aldehydes and esters,
and other secondary products. In the initial stages of the propagation step, one begins to
observe the deterioration of product flavor. In the termination step, the chain reaction can
be stopped by the formation of non-free radical reaction products. This same trend is
reflected in the length of the induction period for lipid oxidation, which was determined
by Maskan (1993) from storage stability studies using an accelerated shelf-life testing
procedure.

An alternative mechanism of lipid oxidation involves the reaction of singlet oxygen with
the double bond of a fatty acid. The reaction between a lipid and oxygen typically
proceeds with the lipid molecule in a singlet electronic state, and the oxygen molecule in
a triple state (Bradly and Min, 1992), which is the common ground state energy level for
oxygen. For initiating the autoxidation reaction involving the singlet lipid with the triple
oxygen directly, a high activation energy is required for the reaction to proceed (Jadhav,
et al., 1996). Therefore, an initiator such as a sensitizer, high temperature, or a
lipoxygenase enzyme plays an important catalytic role to provide the capability of
altering triplet oxygen to singlet oxygen for lipid oxidation. Singlet oxygen can react at
the end of the double bond of a lipid and yield a hydroperoxide (Raws and Vansanten,
1970). Singlet oxygen is generated by the transfer of energy from a sensitizer such as
chlorophyll or myoglobin with ultraviolet light. Ultraviolet light, as the main factor, is
required for this oxidation mechanism to occur. (Frankel, 1984).

Lipid oxidation is the result of a series of speCific individual reactions, with each reaction

having its own activation energy. A number of theories have been proposed, describing

the reaction mechanism for lipid oxidation (Labuza, 1971). Lipid oxidation begins with
the initial reaction process, which has a high activation energy, being about 30 to 40
kcal/mole. At this point, singlet oxygen plays a very important role in the formation of
peroxides in the initial reaction. Trace metals and ultraviolet light, with the proper
sensitizers, also function as initiators of the autoxidation process. Higher temperatures in
this reaction process affect the formation of increased singlet oxygen species. The
activation energy, under UV light, is about 4 kcal/mole for the initial reaction. Normally,
the initial process in lipid oxidation involves metal catalysis. The activation energy, in
the absence of a metal cation, is about 20 kcal/mole (Labuza, 1971). Once the initial
reaction occurs, the autoxidation mechanism involves the propagation reaction, which has
an activation energy of about 3 to 5 kcal/mole. The activation energy of the termination

reactions is about 5 to 8 kcal/mole.

 

 

+ H 0
H00 00H

> 13 \\ / —" \ / 9

50% lsomeric 50% Isomeric

hydroperoxides hydroperoxrdes

\ /
Cleavage Reaction
CHO cnzon
/\/\/ /\/\/ /\/\CHa
Hexanal Hexanal Pentane
Volatile products

Figure 1. The typical autoxidation pathway of linoleic acid (Frankel, 1984 and Labuza,
1971)

10

As shown in Figure l, the autoxidation mechanism of linoleic acid (18:2) involves the
initial abstraction of a hydrogen atom, followed by the reaction of oxygen at the 9- and
i 13- end positions of the carbon chain to produce conjugated 9- and l3-hydroperoxide
isomers. These isomers have the trans, cis-configuration which has the double allylic
methylene group on carbon-11, to generate a pentadieny radical. Linoleic acid produces
4 isomers that have 2 conjugated 9- and 13-diene hydroperoxides and 2 unconjugated 10-
and lZ-diene hydroperoxides. The acyl hydroperoxides that are formed by these
reactions, undergo subsequent reactions, leading to hydrocarbons, carbonyl and alcohols,

and various volatile compounds (Table 1).

Table 1. Relative amounts of volatile compounds from autoxidation of linoleic acid at
moderate temperature (20°C) (Chan, 1987).

 

 

Compounds Percent (weight of a compound per total weight of
all compounds identified)
Pentanal 0.7
Hexanal 66.4
Heptanal 0.7
Octanal 0.6
trans-2-Heptenal 5.9
cis-2-Octenal 13.0
trans-Z-Octenal 5.5
trans-B-Nonenal 0.4
cis-3-Nonenal 0.4
trans-Z-Nonenal 0.4
cis-2-Decenal - 0.3
trans-2,17nas-4-Nonadienal 0.4
trans-2,cis-4-Decadienal 3.2
trans-2,trnas-4-Decadienal 2.0

. l-Octen-3-one <0.1

11

2-3. Secondag Products of Terminal Oxid_ation Reactiog

 

Hydroperoxides formed in lipid oxidation decompose into secondary products. A
general scheme for the fragmentation of monohydroperoxides involves carbon-carbon
cleavage on either side of the alkoxy radical to produce two types of aldehydes, an olefin
radical, and an alkoxy radical. These radicals can react with either OH 0 or H 0. The
vinyl alcohol derived from the reaction with OH 0 is unstable and tautomerizes into
saturated aldehydes (Frankel, 1984).

The product from the reaction with H o is either an oc-olefin, or a short chain ester. These
products include carbonyls, alcohols, esters and hydroperoxides. The linoleic acid of one
substrate yields four major hydroperoxides of 9- and 13-positions in the trans-cis and
trans-trans configurations. These hydroperoxides decompose into a large number of
volatile products, including hexanal as one of the major secondary products (Martin, et
al., 1990). The qualitative data presented in most studies (Frankel, et al., 1989, Hallberg,
et al., 1991, Gardner, 1985) shows that hexanal is the major adehyde formed from
autoxidation of polyunsaturated fatty acids. Quantitatively, hexanal has also been
identified as one of the major oxidation products, resulting from autoxidation of linoleic
acid (Schieberle and Grosch, 1981; Henderson, et al., 1980). The total amounts of
hexanal and pentanal present among the volatile aldehydes formed from linoleic acid
(18:2) during autoxidation were 199.0 and 6.7 (nmole/mg fatty acid), respectively
(Esterbauer, et al., 1990). Icon and Bassette (1984) determined the levels of n-hexanal
and n-pentanal in headspace gas and found these to be good indicators for determining

the degree of rancidity and flavor quality of potato chips. The analysis for n-hexanal in

12

low fat, dehydrated foods by gas chromatography was reported by Fritsh and Gale
(1977). The quantification of hexanal in low fat food products, as a measure of oxidative
rancidity, is a simple and effective analytical method, since hexanal is a stable compound
and less reactive in the low fat model food product than other secondary products. When
the hexanal level increases above a certain concentration in dehydrated and low-fat foods
containing linoleic acid, quality deterioration by lipid oxidation is indicated.

Roozen, et al., (1994) also investigated the levels of hexanal formed from methyl
linoleate hydroperoxides prepared by autoxidation and produced by a lipoxygenase
preparation, using a low fat food model. The lipoxygenase preparation was found to

more readily yield hexanal.

2-4. Factors Affecting Lipid Oxidation

Water activity, temperature, oxygen concentration, antioxidants, trace metals, and
light are among the various factors that can influence the rate of lipid oxidation in food

products.

2-5. Water Activity

The physical and chemical deterioration of dehydrated food products is related
closely to the vapor pressure of water surrounding the product, which is commonly

expressed as water activity.

13

Water activity (Aw) is defined as the ratio of the vapor pressure (P) of water in the food

to the vapor pressure of pure water (P0) at the same temperature.

where, P = water vapor pressure equilibrated over the food
P0 = vapor pressure of pure water at same temperature

RH = relative humidity

The control of Aw in food products controls the change in food quality during storage.
The amount of moisture in a food product is related to the following physical and
chemical changes. The food system can undergo: (i) loss in nutritional quality of
proteins, and vitamins; (ii) the oxidation of lipids; (iii) the growth of microorganisms; (iv)
changes in enzymatic activity; and (v) changes in the texture of food products (Troller,

1989).

2-5-1. Moisture Sorption Isotherm

The moisture sorption isotherm can be used to describe the relationship between
the moisture content of the food product as a function of water activity. This corresponds
to a range of moisture content values and represents the change in moisture content, with

a change in the water activity. The moisture sorption isotherm of most food products

14

shows a sigmoid type curve over a broad range of water activity values. The specific
regions of the moisture sorption isotherm can be related to the physical characterization
of water within the food product and provides an important prediction of the type of
chemical reactions food products may experience.

Water contained in the food product consists of two types, referred to as ‘free’ and
‘bound‘ water. Various food products have different levels of bound water, which is
typically less than 10 %, on a wet weight basis. The B.E.T. (Brunauer, Emmett and
Teller, 1938) monolayer value represents the most stable bound water (Karmas, 1980).
The generalized moisture sorption curve or isotherm is divided into three regions, with
the level of water varying from dry to high moisture in the product. The first region has
Aw values between 0 and 0.25, which is mostly water bound by the ionic groups of the
product, such as NH;+ and COO', which are related to proteins, pectin, etc. The second
type of bound water range has Aw values between 0.25 and 0.75, which includes water
molecules H—bonded to hydroxyl and amide groups of the product. The third region of
sorbed watei has Aw values between 0.75 and 1.0, which represents the least strongly
bound and multilayer moisture levels in the food product. The zone (Aw range from 0.2
to 0.3) between the first range and second range corresponds to the monolayer value and
represented the maximum amount of water, with very strong binding to the dried product
(F ennema, 1996). The monolayer value is very useful to express the most stable
condition of a food product, based on the binding of water molecules. When the moisture
level of a food product is determined by drying, the bound water can be removed ficm
the food contents. The monolayer value from the B.E.T. equation and the moisture

sorption isotherm of food product can also be obtained.

15

26;; W§t__er Activitv related with Lipid Oxidation

 

The water activity level in dehydrated food products, containing unsaturated fatty
acids, influences the development of oxidative rancidity and is a major factor in lipid
oxidation, since a relatively narrow range of water activity can inhibit or activate lipid
oxidation of food products (Rockland and Nishi, 1980). For example, at very low levels
(below 0.3) of water activity in the food product there is a high rate of oxidation of fatty
acids, until the monolayer level of moisture content is reached. As shown in Figure 2, the
rate of oxidation of food products containing unsaturated fatty acids, at normal oxygen
concentration, increases continuously over a water activity range between Aw = 0.3 to

0.7, and is then followed by a near constant rate of oxidation (Labuza, 1971).

1.0 Lipid Oxidation

 

 

 

 

Relative ‘ 0.1
Reaction
Rate - T
0.01 l I
0.3 0.7
Water Activity

Figure 2. The rate of lipid oxidation according to the effect of Aw on the food system
(Labuza, 1971).

16

The above figure illustrates the effect of Aw on the rate of lipid oxidation for a food
model system. At Aw values below the monolayer value, the rate of lipid oxidation
decreases with increasing Aw. The rate of lipid oxidation reaches a minimum around the
B.E.T. monolayer value and increases again with a further increase in Aw, until it reaches
a constant rate of oxidation (Leung, 1987). The moisture, contained in the food product
system affects the rate of lipid oxidation by hydrating metal ions and hydrogen bonding
with peroxides formed, thus decreasing the rate of lipid oxidation by decreasing the
catalytic activity of metal cations and promoting the recombination of free radicals at low
moisture content. The hydration of hydroperoxides reduces the rate of free radical
formation.

Berends (1993) showed the efi‘ect of water activity on the rate of lipid oxidation in a food
product model system. The rate of lipid oxidation was calculated using the activation
energy at different water activity ranges. The low water activities showed a high rate of
oxidation. As the water activity increased to near the monolayer value, the rate of lipid
oxidation decreased. The lipid oxidation rate was then found to increase, up to a constant
level after the monolayer value point.

The shelf life of food products associated with flavor rancidity can be extended by
controlling the products, water activity during storage. The B.E.T. monolayer level in
food products has been shown to provide optimum conditions to minimize lipid
oxidation, which causes food rancidity.

Gopala and Prabhakar (1995) observed a change in the rate of autoxidation in raw peanut

oil by the addition of water throughout the storage period. The addition of moisture, at a

17

low level, into raw peanut oil was thought to inhibit the formation of peroxides from the
fatty acids. Furthermore, Nelson and Labuza (1992) stated the importance of the
relationship between the rate of lipid oxidation and water activity, based on the glass
transition theory in polymeric systems. The physical conditions of the food system such
as a crystalline or amorphous state can affect lipid oxidation, since the moisture content
influences the state of the food system. For a packaged product, the free volume in a
polymer matrix allows oxygen to diffuse through to the food system. Kumor (1986)
mentioned that dry cereal product containing unsaturated lipids in the presence of air
resulted in development of a rancid flavor at low water activities. Lipid oxidation in
cereal products can result in color loss and flavor deterioration associated with the

reaction of unsaturated lipids with oxygen, as well as quality loss.

2-5-3. Monolayer Moisture Content

The monolayer value can describe the amount of water required to form a
monolayer over the highly polar groups of a dehydrated food product. The monolayer
value is associated with the minimum reaction rate for lipid oxidation in a number of
dehydrated food products, when the moisture content reaches the level of the B.E.T.
monolayer value. The initial moisture content, sorption isotherm data, and the Brunauer,
Emmett, and Teller (B.E.T.) equation (Brunauer, et al., 1938) provides the monolayer
moisture content of dehydrated food products. The monolayer water content corresponds
to a specific water activity of the products’ sorption isotherm plot. The rate of lipid

oxidation at Aw below the monolayer value decreases with an increase in the products

18

water activity. The rate of lipid oxidation also increases with an increase in water
activity, after a minimum point at the monolayer value is reached (Quast and Karel,

1972)

2-6. Temperature

Temperature is the most important acceleration factor in the lipid oxidation
process. Elevated temperatures increase molecular movement and accelerate the rate of
lipid oxidation. The higher temperatures result in increased fi'ee radical activity by
providing the activation energy values required for the reactions associated with the
autooxidation process. The activation energy, determined from higher temperature level
studies, is important for the prediction of shelf life at room temperature for food products.
Berends (1993) determined the rate constants for lipid oxidation of a food system model
at three temperatures and compared the respective rate values. This comparison was
based on an assumed first order rate equation and showed that at a constant oxygen
concentration, greater amounts of hexanal are produced as the temperature increases from
23, to 40, and to 66°C. The change in temperature shows a significant efl‘ect on the rate
of lipid oxidation. Gloria (1993) investigated the effect of chemical oxidation,
photosensitized oxidation and autoxidation in a microcrystalline cellulose model system
on the formation of volatile oxidation compounds and the loss of B-carotene. The loss of
B-carotene was faster during autoxidation at 80°C than the chemical oxidation,
photosensitized oxidation and autoxidation at 20°C. Autoxidation at 20°C and chemical

oxidation yielded several volatile oxidation products from B-carotene. Autoxidation at

l9

80°C and photosensitized oxidation led to more specific oxidation products, namely;

dihydroactinidiolide and beta-ionone, respectively (Gloria, 1993).

2-7. Oxygen Concentration

Oxygen is an essential factor required to initiate formation of hydroperoxides in
the lipid oxidation reaction. Thomas (1994) studied the effect of varying initial oxygen
concentration levels, using oxygen absorbers, on the shelf life of packaged potato chips.
An initial oxygen concentration of 0.2 % (v/v) in the package headspace showed
significantly less hexanal formation than that found at an oxygen concentration of 2 %,
after 22 weeks without an oxygen absorber. Koelsch (1989) found the rate of lipid
oxidation at a headspace oxygen concentration of 15.4 % to be faster than that at 1.2 %
oxygen, using a system for measuring the rate of lipid oxidation. Berends (1993) showed
that hexanal levels, at an oxygen concentration of 8 %, followed a first order rate
equation and were greater than that for 1.5 % oxygen at accelerated temperatures, in a
model food product system containing linoleic acid. The oxygen concentration in a
package headspace, or a continually permeating low oxygen concentration can affect the
rate of lipid oxidation reactions. The oxygen concentration can be controlled by the use
of vacuum, or an inert gas and by controlling the headspace composition within the
package system. Kishida, et al., (1993) investigated the oxidation of oleic, stearic,
linoleic and linolenic acids in foods at 170°C in a closed vessel and the oxygen

consumption for autoxidation.

20

The yields of malondialdehyde (MDA) and 2-thiobarbituric acid-reactive substances
(TBA-RS) were compared with the data for oxygen consumption. TBA-RS and MDA
increased with the oxygen consumption up to 2000 ,umol/L and 700 mol/L respectively
in the autoxidation of linoleic acid. Therefore, TBA-RS reflected the reaction course
better than MDA, over a wide range of concentrations of consumed oxygen, in the

autoxidation of linoleic acid.

2-8. Light

Light is a catalyst for oxidative rancidity in food products during storage.
However, a light barrier provided by the packaging material can retard the rate of lipid
oxidation reactions. Mathluothi (1986) showed the influence of light transmittance, using
various packaging materials, on a series of physical, chemical, and sensorial
characteristics of solid whole natural yogurt during storage. High light energy afforded
considerable differences between the peroxide values of the yogurts exposed in different
packages during storage. J con and Bassette (1984) reported the changes in n-pentanal
and n-hexanal concentrations in potato chips exposed to 100 ft—candles of fluorescent
light, using headspace gas analysis. The n-hexanal concentration in the potato chips
exposed to light showed moderately higher levels than in the controls, during the first 60

hrs of storage. Rapid formation of n-hexanal occurred after 60 hrs of exposure to the

light.

21

2-9. Trape Metals

Trace metals, to include Co, Cu, Fe, and Mn, can reduce the length of the
induction period and increase the maximum rate of lipid oxidation. Metals act as
prooxidants by electron transfer, liberating radicals from hydroperoxides via oxidation
reactions. However, chelation of metal ions by food components reduces the
prooxidative effect of these ions. The addition of disodium ethylenediaminetetraacetate
(as chelating agent) in deepofiied instant noodles was found to prolong the shelf life of
the product (Rho, 1986). In biological systems, copper and iron ions have both
prooxidant and antioxidant properties which can promote free radical induced oxidative
stress under certain conditions (Johnson and Fischer, 1994). Copper and iron metal ions
catalyze the oxidation of lipids and show a decrease in the induction period, as well as an

increase in the rate of lipid oxidation (Pershern, et al., 1995).
3. Antioxidants

Antioxidants are a major class of compounds (21 Code of Federal Regulations
(CFR) 170.3) that prolong the onset of deterioration, rancidity, or discoloration of food
products, which occur due to oxidation, as defined by the United States Food and Drug
Administration (FDA) (Dziezak, 1986). Antioxidants function by retarding or inhibiting
the initiation, and propagation reactions in the free radical autoxidation process
associated with lipid oxidation.

Such a reaction scheme is shown below.

22

R0 +AH —)RH+A0
ROOO+AH-—) ROOH+A0

Here, the antioxidant (AH), which has a phenolic hydroxyl structure, can donate a
hydrogen atom to a free radical (R 0) or a peroxyl radical (ROO 0) of the lipid, thus
terminating the propagation step. Reaction of a hindered phenolic antioxidant with a free
radical forms a free phenoxy radical (A 0), which is stabilized by the electron
delocalization in the aromatic ring.

Antioxidants of the hindered phenolic type are classified as either synthetic or a natural
product. Currently, the application of a-tocopherol to a food product directly or
indirectly is increasing, as the trend toward the use of natural antioxidants increases
(Shahidi and Wanasundara, 1995). This affords a greater biological benefit and reduces
health risk, even though synthetic antioxidants applied to food products are effective and
inexpensive (Landvik et al., 1996). Many food manufactures tend to substitute synthetic
antioxidants with foods containing natural antioxidant substances (Marshell, 1974). In
general, the use of antioxidants for a food product directly is limited to 200-300 ppm-
(w/w) of butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), and tertiary
butylhydroquinone (TBHQ), or 200-500 ppm of the gallates, for the stabilization of fats
and oils (Madhavi and Salunkhe, 1996).

Porter (1977) evaluated both natural and synthetic phenolic antioxidants in a dry model
system that had linoleic acid monolayers on activated silica gel. BHA was found to be
highly effective, even though a monohydric compound. Shahidi and Wanasundara

(1995) reported that canola extracts and some of the flavonoids (morin, myricetin,

23

quercetin, and rutin) were more effective than BHA and BHT in preventing the oxidation
of canola oil. The antioxidant molecule may react with the free hydroperoxide radicals.
The free antioxidant radicals are produced after the active free radicals are deactivated.
The antioxidant free radicals are more stable and can undergo oxidation to quinones or
combine with other free radicals.

or-Tocopherol shows a good ability to inhibit lipid oxidation or to extend the induction
period of free radical oxidation, as well as to slow the propagation step of the

autoxidation process (W idicus and Kirk, 1981; Pershen, et al., 1995).

3-1. Synthetic Antioxidants

Synthetic phenolic antioxidants are the most widely used in food products and

include butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), propyl

gallate (PG) and tertiarybutyl hydroquinone (TBHQ) (King, et a1 ., 1993).

Bmlated Hydroxytoluenet BHT )

3,5-Di-T‘art-Buty1-4-Hydroxy Toluene (BHT)

 

24

BHT is one of the most extensively used antioxidants in the food industry. BHT
is used in low-fat food products, fish products, packaging materials, paraffin, and mineral
oils. BHT is also widely used in combination with other antioxidants like BHA, propyl
gallate, and citric acid for the stabilization of oils and high fat foods. BHT is a white
crystalline solid readily soluble in fats and oils and insoluble in water. It is subject to loss
during processing. BHT is Generally Recognized as Safe (GRAS) for use in food to a
maximum antioxidant concentration of 200 ppm of fat weight (21 CFR 182.3173). BHT
is permitted up to 50 ppm of maximum usage levels in dry breakfast cereals, as stated in
the Code of Federal Regualtions (21 CFR 172.115) (Coulter, 1988). BHT in high doses
has a toxic efi‘ect on the liver, lungs, and kidney, and on bloods coagulation mechanism
(Madhavi, and Salrmkhe, 1996). Cort (1974) reported that BHT and BHA were more
effective than tocopherol in their role as an antioxidant in linoleic acid. All tocopherols
were shown to be active in oleic acid, with dl-or-tocopherol equal to BHT, and dl-y-

tocopherol a more active antioxidant than BHT or BHA.

3-2. Natural Antioxidants

The list of natural antioxidants includes tocopherols, ascorbic acid, carotenoids,
flavonoids, lecithin, rosemary extract, gum guaiac, and a few others. Tocopherols have
the greatest popularity of the natural source products. They occur naturally in vegetable

oils, with the prime commercial source being soybean (Giese, 1996).

25

Tocopherol

CH3
l-D\ \

CH3
/' “(\ATKW'K
CH3 CH3 CH3 cue CH3
(II-13
dIaIpln-tocopherol (Vitamin E)

Tocopherol is designated as or-, [3-, 7-, 6- types, which are based on the number
and position of the methyl groups on the chromane ring. or-Tocopherol has methyl
groups in the 5-, 7-, and 8- positions in the aromatic ring. B—Tocopherol has methyl
groups in the 5-and 8-positions, and y-Tocopherol has methyl groups in the 7- and 8-
positions. 5—Tocopherol has a single methyl group in the 8-position of the aromatic ring
(Machlin, 1984). or-Tocopherol has the greatest biological activities in varying potencies
(Schuler, 1990). Tocopherol comes from the vegetable oil source more than fiom an
animal fat source. Cereal grains, vegetables, and fruits are also good sources of

tocopherol.

Tocopherol frmctions as a good chain-terminating antioxidant. Tocopherol has GRAS
status for use as a food preservation (21 CFR 182.3890). Tocopherol is an efiective
antioxidant in various food products for use at relatively low concentrations, in the range

of 100 to 300 ppm of fat weight (Daugherty, 1988). The most efl‘ective concentrations of

26

or - tocopherol are in the range of 100-200 ppm in the food system (Dziezak, 1986). The
loss of tocopherol in food processing is afl‘ected directly by oxygen, light, heat, and trace
metals (Eitenmiller, 1997). However, tocopherol has heat stability and is not volatile
under normal conditions of cooking, compared to the instability of some antioxidants to
heat (Daugherty, 1988).

McCord (1994) reported that vitamin E plays a role in inhibiting the progress of chain
reactions involving propagation of lipid radicals produced as oxidant productions in the
presence of iron ions. Burton, et aI. , (1993) proposed that the (Jr-tocopherol functions as
an antioxidant by rapidly transferring its phenolic hydrogen atom to a lipid peroxyl
radical, resulting in the formation of two molecules that are relatively unreactive towards
polyunsaturated lipid, a lipid hydroperoxide and the or-tocopheroxyl radical. Cort (1974)
showed that the antioxidant activity of the dl-y-tocopherol was greater than or-tocopherol,
and increased as the concentration increased in chicken, pork, and beef fats, and with
oleic acid. Pershem, et al., (1995) reported that a relationship existed between total fat,
fatty acid and a—tocopherol contents and lipoxygenase (LOX) activity and the shelf-life
of hazelnuts. The or-tocopherol content in hazelnuts played an important role in
controlling off-flavor at a level of 20mg per 100g hazelnuts. A higher a—tocopherol level

in hazelnuts improved shelf-life stability.

27

3-3. Antioxidants for the Polymer Pl§_stic Film

Currently, polymeric packaging materials are widely used due to commercial
advantages of their cost and good control of their physical and mechanical properties.
Among the polymeric food contact materials, a commercial HDPE has shown high usage
in food packaging such as for milk bottles, drinking water bottles, and as a package for
edible oil containing fatty acid components (Till, et al., 1982). Polyolefins have
incorporated antioxidants to maintain the stability of the polymer to oxidation during
processing and at environmental conditions, as well as to retard the oxidation of packaged
food products. The oxidation of polyolefins is related to the degree of chain branching in
the polymer. The oxidative degradation of polyolefins in the presence of oxygen, during
melt processing at high temperatures, may be related to the packaging article’s inability
to withstand a stress associated with aging or deteriorating efl‘ects (Birley, 1991).

The use of or—tocopherol to prevent oxidative reactions in extrusion processing for HDPE
bottles and LDPE film packaging at high temperatures has shown the effectiveness of or-
tocopherols performance as an antioxidant, through sensory evaluations. Its application
at low temperatures has also been determined (Zambetti, 1995). Laermer and Nabholz
(1990) evaluated or—toc0pherol as a highly efl‘ective melt stabilizer, alone at low
concentrations and in synergistic mixtures with other secondary antioxidants, including
phophites (such as TNPP, Hostanox PAR204, Weston 619) and thioesters (such as
A DSTDP) in polypropylene. The melt and color stabilization observed for a series of
mixtures of tocopherol with phosphite and thioester antioxidants with polypropylene and

polyethylene showed such mixtures to have excellent stability properties, as compared

28

with other antioxidants (Laermer and Zambetti, 1992). Tocopherols incorporated into
polymeric packaging afford commercial advantages to the polymer system and to
commercial applications of the polymer. Polymers stabilized with or-tocopherol were
formd to have fewer taste and odor problems, as well as better color stability, than those
with typical commercial polymer antioxidants. Further, there is a reduction of
contamination risk from the packaging components, and good thermal stability during
processing (Laermer et al., 1994). At accelerated stability tests, HDPE/EVA liner
packaging films with various concentrations of tocopherol added, showed a noticeable
reduction of flavor loss from a packaged cereal product at 37°C, over a three month
storage period (Laermer et al., 1994). Ho and Yam (1994) reported that HDPE drinking
water bottles containing tocopherol had less odor and higher acceptability than these
containing the antioxidants, Irgonox 1010 or BHT, through sensory evaluation of the OE-
flavor problem. This problem can be identified with the following possible factors
associated with odor threshold, namely: molecular weight, and polarity of the compounds
released from the polymer, such as ketones and aldehydes. Lin (1996) observed that
cereal product packaged in pouches without antioxidant showed a significantly higher
level of hexanal, as compared to the rate of oxidation of cereal product packaged in
antioxidant (or-tocopherol, BHT) impregnated structures, following the 20th week of
storage (23°C/ 5 0%RH). The hexanal levels detected in cereal product packaged in or-
tocopherol impregnated pouches and BHT impregnated pouches showed no significant

difl‘erence, over a 44 week storage period at ambient conditions (23°C/50%RH).

29

4. The Mechanism of the Mfition of Antioxidgnts from the Packaged Film 811'qu

gap Contained Product;

Antioxidant migration from a packaging film structure into a contained food
product can be described by the following basic mechanism. First, diffusion of
antioxidant molecules contained within the void volume between the polymer chains
through the polymer bulk phase to the polymer surface. Second, evaporation of
antioxidant molecules from the surface of the packaging material to the internal package
environment. Third, sorption of antioxidant molecules onto the surface of the packaged
product. High molecular weight antioxidants would be expected to migrate slowly into
the food product, as compared with low molecular weight migrants. The mechanism for

the three step migration process is illustrated in Figure 3.

 

 

 

 

 

 

Film Product

Figure 3. A three step migration process

30

Generally, migration fiom a polymeric packaging material can be described as either
global or specific migration. These terms refer to whether the total mass of migrant is
considered, or a number of restricted migrating components from the package into the
contained food, under specified conditions, are being considered (Crosby, 1981). The
migration system can be divided into the three types based on migrant diffusion, such as
described by Robertson (1993): (1) migration takes place only from the packaging
surface regardless of the presence of a food product. (2) the presence of food may affect
migration fi'om the packaging surface; However, migration is not controlled by the food
product, but by the rate of difl'hsion of the migrant through the polymer bulk phase. In
this case a diffusion coefficient can be measured under the test conditions and during a
specified contact time. (3) migration is controlled by the food product at the packaging
surface, and the food system plays an important role in the transfer of migrants from the
film structure to the contained product. Here, penetration of the film structure by a
component of the contacting food system can increase the rate of diffusion of the migrant
from the packaging film. The relationship describing the loss of antioxidants from the
packaging film structure into the food product can be expressed by the rate of migration
of the antioxidants. The migration rate of the antioxidant is afi‘ected by the following
factors (Calvert and Billingham, 1979): the additive solubility, the diffusion coemcient of
the additive, and additive volatility, which are discussed briefly below. (1) The solubility
of antioxidants in the film structures. This is described by the solubility coefficient,
which is determined by the morphology of the polymer and the chemical properties of the
antioxidants. The temperature and vapor pressure also affect the solubility of

antioxidants. (2) The diffusion coemcient of antioxidants within the film structures,

31

describes how rapidly molecules are advancing through the polymer bulk phase and the
time to reach steady state. The molecular size of the antioxidant affects the magnitude of
the diffusion coefficient.

The diffusive flux of antioxidants in a film structure can be expressed as the amount

passing through a plane of unit area normal to the direction of flow during unit time.

F = M/A -t -..---.---- (1)

Fick’s first law applies where the rate is directly proportional to the concentration

gradient.

F = — D ( aC/ax ) (2)

Where,

F = the flux (the rate of transfer of the diffusing substance per unit area)
M = the amount of the substance which passed through the film

A = area of surface at which diffusion occurs

C = the concentration of diffusing substance

X = the distance of difl‘usion

t = time

D = the coeflicient of difl‘usion

For an infinite plane or sheet, across which difl‘usion occurs in a linear direction (x) of a
thin membrane, in most cases, migration from a film structure into a food system can be

expressed by F ick’s second law (Equation 3).

32

(BC/6t): mam/62(2) ------------ (3)

The amormt of antioxidant transferred per unit time, through a film structure with
constant dimensions, can be represented by the transmission rate. Factors affecting
antioxidant diffusion within the film structure include the chemical nature of the film
structure, the size of antioxidants, the thickness of the film structure (an inverse
proportion), and environment conditions (temperatures, humidity).

(3) The loss rate of the antioxidants by volatilization from the film surface.

A mathematical model describing the loss of an antioxidant fi'om the polymer to the
environment in which the polymer is stored requires two parameter factors; namely: (1) a
mass transfer constant, characterizing transfer across the boundary of polymer surface-air
interface; and (2) a constant characterizing antioxidant mass transfer within the polymer
bulk phase. _

A mathematical equation has been described by Crank (1975) for the additive loss from a
film by surface evaporation with finite boundary conditions. To calculate the loss of
antioxidant concentration as a function of time, the total amormt of additive M leaving
the film at time (t) is expressed as a fraction of the corresponding amount M... leaving the

film at infinite time by the mathematical equation (4).

 

co 2 2
M, _1_Z 21; exp2( -,B:r T) __________(4)
M ,Bn (fin +L +L)

co n=l

Where, M = amount of additive leaving the film in time ( t )

33

M... = amount of additive leaving the film at infinite time
T= Dt/IZ

L = I or /D

I = half of film thickness, cm

t = time, sec

D = diflirsion coefficient of additive in film, cmZ/sec

a = mass transfer constant of additive fi'om film, cm/sec

,Bn values are the positive roots of fin tan fin = L

Calvert and Billingham (1979) reported the rate of loss of a low molecular weight
compound, such as 3,5-di-tertiary-butyl-4-hydroxytoluene (BHT), from thick polymeric
slabs was determined by bulk phase diffusion, while the loss fiom thin films was
dominated by the rate of surface evaporation.

The rate of loss of 2-tertiary-butyl-4-methoxyphenol (BHA) from a high-density
polyethylene film as a function of the storage time and temperature, based on the
theoretical model, was reported by Han, et al.,(1987).

From the following equation (5), for a first-order rate expression, the loss of BHA from

HDPE film as a fimction of storage time and temperature was analyzed.

Ln — = -Kt ---------..- (5)

where, Co : the initial concentrations of BHA in the film sample (w/w)
C : the concentrations of BHA in the film sample (w/w) at time (t)
K : the rate constant

t : the time interval

34

If the term for n = 1 is assumed in Equation 4, the following form is derived after

 

rearrangement,
1 _ M , _ 2L2 exp( — flnzT ) ____________ (6)
M,D ,Bn2(,6n2+L2+L)

01'

an

12 -------- (7)

 

2+L2+L)’
2 1:—

2L '3

Ln[(1— lax/’2“

Equation (7) is the same form as Equation (5). Since (1-Mt/Moo) = C/Co and the

equations describe the same phenomenon, the two following equations can expressed by

 

 

2
'6 D = K -----—---- (8)
[2
and
2 2 2
flw+¢+uzl “mmt)
2L2

The 5 value is used to calculate the value of D from Equation 8, and the values of L and

D are used to calculate the values ofa from Equation of L = Ior /D.

35

The above methods, as described by Calvert and Billingham (1979), give a calculation of
the diffusion coefficient, the volatilization coefficient, and the expected shelf-life of the
polymer. Han, et a1. (1987) found that the loss of BHA fi'om HDPE followed a first-
order rate expression and described a method for the calculation of the diffusion
coefficients (D) and the mass transfer coefficient (on) from sorption or desorption data for
BHA and BHT in HDPE. The controlling parameter factor for mass transfer of the

antioxidants was found to be volatilization rather than diffusion.

5. Product Model System

A number of researchers have used a model system to provide a specific, as well
as a well characterized system, to study the complex autoxidation reaction. The product
model gives a constant composition and can minimize the number of variables, which can
result in unexpected reactions. Widicus and Kirk (1981) reported the storage stability
properties of or-tocopherol using a dehydrated model food system containing methyl
linoleate for lipid oxidation. The rate of lipid oxidation was found to be dependent on the
accessibility of the surface area of the food system to oxygen. Thus, oxygen diffusion is
difficult to control in a model system. To minimize the problems associated with the
product surface area, the model system required a large surface area and a minimal
thickness. In this case, oxygen is readily accessible and the reaction rate is not dependent
on oxygen diffusion (Leung, 1987). Berends (1993) used a model product to represent an
oxidation-susceptible freeze-dried product containing linoleic acid. The uniformity of the

model product and a consistency of the oxidation rate to exposed oxygen were proved

36

from product components adapted by Koelsch (1989) and Berends (1993). The product
model used for this research has been developed to predict the rate of lipid oxidation,
based on the oxidation of linoleic acid during storage. This model system can also
provide additional information on the effectiveness of antioxidants on retarding lipid

oxidation of the model product system, according to various affected factors.

6. Analflical Technique for Measurement of Lipid Oxidation in Food System

The analytical measurements to monitor lipid oxidation have been based on a
number of techniques (Allen and Hamilton, 1994). As previously discussed, lipid
oxidation is a complex sequence of reactions, which involves the formation of
intermediate products such as hydroperoxides and peroxides, as well as a number of
different breakdown products, depending on the types of lipids in the product. The
peroxide value (PV), which is based on an iodometric titration procedure, is a common
measurement that has been used to quantify peroxide or hydroperoxide type intermediate-
products. Mate, et al. (1996) determined the peroxide value of peanut oil by reaction
with ferric thiocyanate as a colorimetric method. The arrisidine value (AV) estimates the
level of aldehydes (2-alkenals) in fat containing products, which are treated with p-
anisidine reagent in iso-octane solution, and is useful for fat containing products with a
low PV. The thiobarbituric acid (TBA) method is used to determine the level of aldehyde
products formed. Here, thiobarbituric acid reacts with malondialdehyde to produce a red
chromogen and is useful for monitoring the initial steps 'of lipid oxidation. Warner, et a1.

(1996) used a modified TBA method to determine the total extent of lipid oxidation of

37

stored ground peanuts. The 10% peanut homogenate with deionized water was added to
TCA (trichloroacetic acid) solution. The absorbance of a filtered aliquot was then
measured at 443nm using an UV-VIS spectrometer, after reaction with TCA at 60°C for
30min. The Kreis test measures a red color obtained when phloroglucinol reacts with
oxidation products in hydrochloric acid solution (Gray, 1978). The red color is related to
the reaction of epoxy aldehydes and acetals extracted from oil products. High
performance liquid chromatography (HPLC) methods have also been used to measure
aldehydic products (2,4-dinitrophenylhydrozones) of lipid peroxidation, and are most
often used with biological systems. Gas chromatography (GC) is useful for determining
volatile compormds of lipid oxidation (Frankel, et al., 1989). GC is applied to volatile
compounds in fat containing products through headspace or extraction procedures. GC
offers high resolution, in addition to providing a strong possibility of identification of

oxidation products by mass spectroscopy.

6-1. Gas chromatogrpphy / Mass metromegy

Gas chromatography/Mass spectrometry (GC/MS) is an important analytical
method to provide characteristic information ,of molecular structure and as a quantitative
tool with a high sensitivity and specificity. Gas chromatography/Mass spectrometry
(GC/MS) combines two techniques for the separation and identification of the sample
product. Gas chromatography (GC) provides a separation technique and mass
spectrometry (MS) an identification technique (Watson, 1985). A mass spectrometer

'1 provides a useful analysis function that converts sample molecules into ions and then

38

separates the ions according to their mass-to-charge ratio. Such a process provides
definitive qualitative and quantitative information on sample molecules, based on their
molecular composition. Generally, the mass spectrometer has three basic frmctions;
namely: (1) to ionize the substance; (2) to separate the constituent ions according to
mass-to-charge ratio; and (3) to detect the relative abundance of the individual mass-to—
charge fragments (i.e., ions) and record the mass spectrum of the substance.

There are a number of different ways to separate the ions of difl'erent mass-to—charge
values, when the various mass-to—charge ions from the ionized samples are formed. The
different types of mass spectrometers are: (i) magnetic-sector; (ii) double-focusing; (iii)
transmission-quadrupole; (iv) quadrupole ion traps; (v) time-of-flight; .and (vi) ion
cyclotron resonance (Watson and Sparkman, 1996). A

In the magnetic-sector mass spectrometer, ions with a positive unit charge are created
from the sample that exists in the gas phase, at a pressure of about 10'3 or 10.5 Torr. The
ions then are accelerated into a magnetic field. The magnetic field deflects the ions
according to their mass and ions with a certain mass strike the detector.

The quadrupole mass spectrometer is the most generally used instrument for the
separation of inns produced in the analysis systems such as gas chromatography/mass
spectrometry (GC/MS) and liquid chromatography/mass spectrometry (LC/MS). Radio
frequency (R1) and direct current (dc) fields applied to opposing sets of four parallel
surfaces are used to separate the ions of differing mass-to-charge ratios (m/z values) in
the quadrupole mass spectrometry.

Mass spectrometers are used in both organic and inorganic chemistry. In organic mass

spectrometry, the most widely used ionization technique for volatile compounds is

39

electron ionization (El). This technique uses a beam of electrons at 70 eV to ionize the
compormd being analyzed and at the same time impart energy to the molecular ions that
are formed. Some of the molecular ions then will fiagment. Detection of the molecular
and fragment ions produces the mass spectrum. Some molecular ions do not last long
enough to be detected, because of unstable fragment ions. This introduces some
difficulty with interpretation of the data.

In these cases, it sometimes is possible to employ chemical ionization (C1) to produce
ions representative of the impact molecule. CI is often used in conjtmction with E1 to
identity compounds. Nonvolatile ionization compounds such as sugars and peptides are
analyzed using desorption ionization techniques.

A library search is an important procedure to identify a compound by comparison of its
mass spectrum with a database of mass spectra of known compounds. Two of the most
general data bases of mass search programs that provide electron ionization (El)
generated mass spectra for use on PCs are the NTST/EPA/NIH and the Wiley database
(Watson and Sparkman, 1996). Interpretation of mass spectrum data is also used for
identification of unknown spectrum, using information of the fragmentation of molecules
related to structural elements.

Quantification based on GC-MS analyses, is usually determined using selective ion
monitoring (SIM), because it is more sensitive than full scanning. However, it produces
less information as only a limited number of ions are monitored. An SIM analysis is
obtained for specific compounds and can increase the sensitivity by 50 to 500 times, as
compared to total ion monitoring data (Watson and Sparkman, 1996). The mass

spectrometer incorporating the SIM (selected ion monitoring) mode is equipped with a

40

peak selector to record selected ions characteristic of a compound of known identity.
SIM analysis that has selected several mass ions of specific m/z values provides several
advantages which include: (1) greater sensitivity due to more time at the few selected
ions rather than total scanning; (2) shorter scan cycle times since only a few ions in the
total m/z range need to be examined; (3) more accurate quantitation since a shorter cycle
time allows the m/z range selected to be scanned more frequently; and (4) higher
resolution of overlapping peaks in a complex sample by monitoring characteristic ions of
each peak.

Dawson, et al. (1991) monitored headspace volatiles to evaluate the effect of adding
glycerophosphorylcholine (GPC) and Glycerophosphorylethanolamine (GPE) on the
oxidation of linoleic acid using GC/MS analysis that was set at 30-300 m/z scan range.
Rosiers, et a1. (1993) used the GC/MS method for the measurement of specific aldehydes,
such as malondialdehyde, hexanal, and 4—hydroxynonenal, to evaluate the extent of lipid
peroxidation. Selected ion monitoring was applied to GC/MS analysis in the positive
chemical ionzation mode for malondialdehyde and 4-hydroxynonenal, and in the electron
impact mode for other aldehydes (Rosiers, et al., 1993). Hall, et al. (1985) studied the
optimum model for describing the kinetics of the formation of hexanal in a dry milk
sample, which was stored under air and which had a low concentration of Maillard
reaction product during storage, and derived an expression to fit the experimental data
obtained by GC/MS analysis. Buttery and Ling (1995) identified hexanal, which
represented major MS ions, as the principle volatile flavor component present in tortillas

by capillary GC/MS.

41

7. Isol_ation Techniques of Volatile Compounds Present in Food System_s_

The off-flavor compounds related to various food products have a large number of
different functional groups and chemical structures, which are derived from biochemical-
metabolism or from external factors (Charalambous, 1992). Specifically, lipid oxidation
leading to the unacceptability of a food product can be estimated by the development of
rmpleasant flavors. The major secondary oxidation products from linoleic acid, among a
variety of fatty acids, are representative aldehyde type compounds, which include
hexanal, pentanal, octenal, and decadienal etc. The aldehydes and vinyl ketones formed
in the lipid phase of food products are mainly responsible for the ofi-flavors, due to their
low threshold levels for human sensory response (Kochhar, 1993). Such volatile
compormds are in low concentration and therefore must be concentrated from the food
prior to instrumental analysis. The volatile compounds resulting fiom the oxidation of
food products can be isolated by static headspace, dynamic headspace, or distillation-
solvent extraction techniques (Reineccius, 1984). The headspace sampling techniques
were found to give accurate data for the level of volatile compounds derived from a food
product. Gas chromatography/Mass spectrometry is an excellent method for analysis of
volatile compounds, due to the high resolution and sensitivity afforded by this method.

The respective procedures are reviewed briefly below.

42

7-1. Static Headspace Analysis

For static headspace analysis, the headspace gases above a food sample, in a
sealed container, are injected directly into a gas chromatograph for analysis of volatile
compounds. The direct headspace sampling procedure has been widely used, due to the
advantages of simplicity and rapidity. The direct headspace sampling procedure can
reduce the loss of volatile compounds during opening of the container. Direct headspace
sampling, however, also has the disadvantages of limited sample size, and less sensitivity
than other procedures which involve sample concentration. This system is useful for
measurement of oxidation at a particular time, as well as the extent of oxidation of foods.
Mate, et al. (1996) used the static headspace method for nut samples, using gas
chromatography, for analysis of hexanal resulting in oxidative rancidity at accelerated
tests. Moreau and Rosenberg (1996) reported the analysis of hexanal to monitor for the
oxidative stability of an anhydrous milkfat microencapsulated in whey proteins, using
headspace methods. Leino (1992) identified the volatiles. of fresh, frozen, freeze-dried,

and air-dried chive samples, using a static headspace gas analysis procedure.

7-2. Me and Trap System (Ewe Hea_dspace Analysis)

The purge and trap technique is a dynamic system designed to improve the low
sensitivity faced by the direct headspace sampling system. The volatile compounds are
purged from the sample by a stream of inert gas, such as nitrogen or helium, and are

trapped with a cartridge tube filled with a sorbant, such as Tenax, Chromosorb or

43

activated charcoal. The volatile compounds are then desorbed from the sorption tube by
rapid heating in a desorption system and are analyzed by GC or GC/MS. Sakakibara, et
al. (1988) used activated charcoal to trap the volatile flavor compounds from the
headspace of powdered dried bonito (katsuobushi) during storage. Luning, et al. (1994)
studied the differences in flavor of fresh green and red bell peppers, using Tenax sorption
to first sorb and then desorb the volatile flavor compounds by a dynamic headspace
technique. Hall, et a1. (1985) analyzed quantitatively for compounds with higher boiling
points, such as the headspace concentration of hexanal, by the dynamic purge and trap
method. Hall, et al. (1985) reported the use of Chromosorb 105 for adsorbing volatile
compounds from a dry milk sample, with fat content of 23.8-24.5 % and containing
various antioxidants, by the dynamic headspace sampling system. The purge and trap
system has a higher sensitivity for analyses involving trace levels of volatile compounds.
However, there are some disadvantages with this procedure, such as: (i) breakdown of the
very volatile compounds due to high temperature during thermal desorption; (ii) low
recovery of high boiling point compounds; and (iii) transfer of water fiom a product
sample with a high moisture content to the chromatography column. The conditions of
the purge, trap, and desorption steps must be controlled to optimize the sensitivity
according to the volatile compounds and analytical system. A Tenax trap is very widely
used for trapping of volatile compounds. Tenax has a low affinity for polar compounds

and a high affinity for nonpolar compounds (Reineccius, 1993).

44

7-3. Distillation System

The distillation system includes a high vacuum molecular distillation, stream
distillation, or simple heating of the food and taking the distilled aroma constituents into
the analytical apparatus (Reineccius, 1993). Stream distillation has two general types.
One type of distillate system involves isolation of product volatiles with a rotary
evaporator. The other is a more complex simultaneous distillation/extraction procedure
(SDE), based on the apparatus of Likens and Nickerson (1964). This system is most
commonly used for extraction of flavors today. Both approaches require distillation and
solvent extraction.

Most of the volatile compounds in food products are isolated by simultaneous
distillation/extraction. The distillation/extraction method, when compared to headspace
techniques, has the advantages of greater extraction efficiencies of high boiling point
substances and higher concentrations to allow identification by GC/MS. The Likens and
Nickerson method also has the disadvantages of the potential destruction of volatiles
during distillation, and loss of volatiles during concentration. Beal and Mottram (1994)
examined the off-odor compounds of malted barley, sampled as a function of storage
time, by GC analysis of the volatile isolates using a simultaneous distillation/extraction
procedure (a modified Likens and Nickerson apparatus). Chung and Cadwallader (1994)
reported the use of simultaneous steam distillation and vacuum simultaneous steam
distillation-solvent extraction techniques for volatiles extracts fi'om cooked blue crab
(Callinectes sapidus) claw meat. These extracts were analyzed by gas

chromatography/olfactometry for odor activity.

MATERIALS AND METHODS

3-1. Product Model

A product model was designed to simulate a low moisture content, fatty acid-
containing food product, such as a cereal product or snack food. As formulated, the
product model had a uniform thickness, density, moisture content, and surface area.
Product model ingredients included linoleic acid, to provide the substrate for lipid
oxidation, distilled and deionized water (Mallinckrodt Baker Inc., Paris, Kentucky),
Tween 20 (Fisher Scientific, Pittsburgh, PA), and sodium carboxymethycellulose (CMC)
(Aqualon Co., a division of Hercules Inc. Hopewell, VA). Linoleic acid is an unsaturated
fatty acid that has two-double bonds at the 9 and 12 carbon positions in the eighteen-
carbon chain (18:2). Generally the main source of linoleic acid is vegetable oils such as
soybean, sunflower, cottonseed, and corn oils. The linoleic acid for the product model
was obtained fi'om Acros Organics (99%, C13H3202, FW 280.45, Fisher Scientific,
Pittsburgh, PA). CMC, as cellulose gum, is an anionic water-soluble polymer derived
from cellulose. The main function of CMC is the binding of water, or to increase the
viscosity of an aqueous system. The main applications of sodium
carboxymethylcellulose are as a stabilizing agent in ice cream, and as a high water
binding agent of dietetic foods (Dreher, 1987). Tween 20 (polyoxyethylene-ZO-sorbitan
monolaurate) is an emulsifier which promotes mixing between water and oil. The

ingredients and uniformity of the mixture were based on a similar product model

45

46

described by Berends (1993). The composition of the product model formulation is
summarized in Table 2 of the Results and Discussion Section.

Linoleic acid, Tween 20, and distilled and deionized water were mixed with a
homogenizer (Model KS-A, Hobart, Kitchenaid), operated at a level 4 on the speed
control. Sodium carboxymethylcellulose was slowly added to the aqueous mixture. The
mixed ingredients were homogenized for approximately 10 minutes, or until the
formation of a uniform product that contained a constant moisture content. Samples of
the uniformly mixed product were added to a depth of approximately 15mm thick into
plastic petri dishes (100 x 15mm size). The petri dishes were weighed and tared before
adding the product samples and applying the petri dish covers. The sample products in
the petri dishes, covered by a paperboard box, were placed into a freezer (-30°C) for
fi'eezing over a 24 hour period. The frozen products were placed in a laboratory fieeze
drier (V itris Model II, Repp Industry, Gardiner, NY), after the petri dish covers were
removed. The operating conditions of the freeze-drier were 100 micro torr at -50°C, with
a plate temperature of -3 8°C. The frozen product samples were dried for 5 days. The
completely dried products were immediately covered with petri dish covers, weighted,
and placed in a paperboard carton. The freeze-dried product was then stored in a
conventional freezer (- 30°C), until samples were required for characterization and
stability studies. For storage stability studies, samples were prepared by placing the
homogenized product to a depth of 15mm thick into 60 x 15mm size aluminum weight

dishes and then freeze drying the product samples as described above.

47

3-2. Packagrp' g Materials

Three coextruded laminate films were used to fabricate the flexible pouches to be
evaluated in this study. All of the films contained a heat seal layer (Suryln-EVA)
coextruded to a high density polyethylene (HDPE) layer. The primary difference
between the three test films was in the nature and level of antioxidant incorporated within
the heat seal layer. Fihn I contained or-tocopherol in the heat seal layer, film 1] contained
3,5-di-tert-butyl-4-hydroxytoluene (BHT) in the heat seal layer, and film III had no
antioxidant in the heat heal layer. Film samples were obtained from the James River
Corporation (Flexible Packaging Group, Cincinnati, OH). The average thickness of the
test laminate films was measured using the Model 549 micrometer of Testing Machines,
Inc (Amityville, L.l., NY). The average thickness and the thickness measurements of
each film are summarized in Table 24. Each side of the film samples was identified by
Fourier Transform Infiared Spectroscopy (Perkin Elmer Model 1000). The results
obtained for each film are presented in Appendix E. The concentration levels of the
antioxidants in the film structures were determined by high pressure liquid
chromatography (HPLC) analysis. The antioxidant concentration levels determined for

the film structures are presented in Table 11.

48

3-3. Initgl Moisture Content

The initial moisture content of the model product system was determined using a
modified vacuum-oven method performed on the freeze-dried samples. This method was
based on a similar product model described by Berends (1993) and is described in
Section 28 of the omcial Method of Analysis of the Association of Official Analytical
Chemists (1975). The freeze-dried samples were added to tared aluminum weighing
dishes (50 x 15mm) and weighed. The samples in the aluminum weighing dishes were
then placed in the vacuum oven and the samples maintained at 30mmHg and 90°C for
eight hours. The samples were placed in a desiccator for 1 hour to allow equilibration to
room temperature after removing them from the vacuum oven. The samples were then
reweighed. The initial moisture content of the samples was calculated using the average
data.

The initial moisture content was obtained using the following equation:

(Ely-g1] x100 = %IMC (on dry weight basis) ----- (10)

where; Wi = initial weight of product sample (grams)
Wf = final weight of product after drying (grams)

The results for the determination of the initial moisture content of the sample are

summarized in Table 4 of the Results and Discussion Section.

49

3-4. Sorption Isotherm

A sorption isotherm was developed to determine the Equilibrium Moisture
Content (EMC) of the model product at different water activities (Aw). A series of salt
solutions were prepared using the procedure described in Hygrodynamics Technical
Bulletin No. 5 (Creating and Maintaining Humidities by Salt Solutions) and placed in
eight tightly sealed recloseable 5 gallon high density polyethylene buckets, thus creating
a series of constant relative humidity environments. The buckets were, allowed to
equilibrate for two weeks before the test began. The relative humidity inside the buckets
was monitored with a hygrometer sensor (American Instrument Company, Silver Springs,
MA) mounted in the plastic lid. F reeze-dried product was taken immediately from the
petri dish and weighed on an analytical balance into tared aluminum weighing dishes.
Three replicate samples, in aluminum weighing dishes, were then placed into each
relative humidity storage bucket. Isotherm data were obtained gravimetrically by
measuring product weight change over five day intervals, at a constant relative humidity
and temperature. Isotherm data were obtained when the freeze-dried model product
system reached an equilibrium weight, at each test condition. All humidity conditions
inside the storage buckets were shown to be constant over the course of the experiment,
as indicated by the readings from the hygrometer sensors installed in the respective
relative humidity buckets. The storage temperatures were 18, 28, and 38°C, respectively.

Isotherm results were obtained from the average of triplicate data values. The conditions

50

of relative humidity and the salts used to obtain the specific relative humidity are

presented in Table 26.

The equilibrium moisture content (EMC) was calculated according to the

following equation:

 

(”(l;.IMC))—lx100 = %EMC (on dry weight basis) -----(11)
r

where; Pf = final weight of product sample after equilibration
Pi = initial weight of product sample
IMC = gHzO/ g dry weight product

3-5. Moisture Eguilibration System of Product Model

The freeze-dried product model system was equilibrated to a constant water .
activity (AW = 0.30) under a nitrogen atmosphere, to retard lipid oxidation prior to
initiating the storage stability studies. A dynamic flow procedure was developed, which
is described below.

A multi-chamber test system was designed to equilibrate the fieeze-dried product
model to a fixed and constant relative humidity at constant temperature. The test
chambers consisted of stainless steel desiccant chambers of dimensions, 30.5cm x 28cm x
31cm (lengthx width x height). Each chamber was equipped with seven open metal-link
shelves with a 30mm interval between, to allow the exposure of multiple product samples

under constant conditions. Each test chamber was also equipped with a gas inlet and

51

outlet port to provide continuous flow of the humidified nitrogen gas stream. A
schematic diagram of the equilibrium system is given in Figure 4. The flow rate of the
humidified nitrogen gas into the inlet port of the chamber was set at 30m1/min.
Equilibration conditions in the test chambers were carried out at 23 11°C. A constant
concentration of water vapor to be flowed continually through the respective exposure
test chambers was produced by bubbling nitrogen through a gas washing bottle, which
contained distilled-deionized water, with a fiitted dispersion tube. To obtain the required
relative humidity, the humidified gas stream was mixed with another stream of dried
carrier gas (nitrogen). Flow meter settings were determined to provide the desired
relative humidity value of approximately 30% RH. A hygrometer sensor (model
No. 1243, American Instrument Company, Silver Spring, MA) was incorporated into the
system design to allow continuous monitoring of the relative humidity of the gas stream,
over the course of the equilibration process. The moisture equilibration system for the

product model is shown in Figure 5.

 

 

 

 

 

 

 

A E 0
': R4
0
ll HS

OHIIIIHIEI
3
Oflflififlfl

 

 

 

 

 

 

 

 

 

 

 

v R2
E 1 DV I DV
0
“3 HE 0D
Nitrogen Gas CB CB
w
R1: rotometer of dried gas CB: controlled chamber
R2: rotometer of humidified gas HM: hygrometer indicator
R3. R4: rotometer of controlled gas HS: hygrometer sensor
WB: water bubbler DV: needle valve

W: exit of gas flux

Figure 4. A schematic diagram of the equilibrium system

53

 

Figure 5. The moisture equilibration system for product model.

54

3-6. Product St_abilitv Studies

A known amount of the model food system (approximately 6 gm) was packaged
in 4.5” x 5” test pouches, fabricated fi'om the three test films. The pouches were sealed
by an impulse heat sealer (Sencorp Systems Inc.) at 35psi with a 0.9second heating time
and a 0.4second cooling time, and then stored in an environmentally controlled room.
The conditions of the storage room were selected to maintain the packaged product at 23
_+_ 2°C and 50 i 5% RH, which maintained the product model at constant water activity.
Accelerated storage stability studies were also performed at 45 : 1°C and 50 3t 5% RH in
a controlled chamber (Hotpack corp. Philadelphia, PA). A schematic diagram of the
storage arrangement used for the product stability studies is shown in Figure 6. Test
pouches made from the respective film structures were removed from storage at
predetermined time intervals. After sampling the pouches, the product model was
removed. A small amount of the product model was used for the hexanal analysis to
determine the extent of lipid oxidation. The pouch structure was assayed to determine
the relative concentration of antioxidant retained as a fimction of the storage time, at
given environmental conditions. Samples were stored and assayed for hexanal levels, as
a flmction of storage time and temperature and product oxidation rates compared to the

other test samples. A flow diagram of the test scheme is shown in Figure 7.

55

 

(23°C, 50 % RH) (45°C, 50 % RH)

Figure 6. The storage arrangement of the pouched model product at 23°C, 45°C.

56

 

 

 

Product Model/Package System

 

 

Sample as a Function of
Storage Time and Temperature

I
I l

 

 

 

 

 

 

 

 

 

 

 

 

I Product Model | [Package Structure
, l ,
l , l ,
Hexanal Linoleic Acid Antioxidant
Analysis Analysis Analysis

 

 

 

 

 

Figure 7. Flow diagram of the test scheme.

57

3-7. Afinalytical Procedure for Antioxidants in the Packaged F ilrns

3-7-1. Soixhlet Extraction System

A soxhlet extraction apparatus was used to extract the antioxidants from film
samples obtained either directly fiom roll stock or from the fabricated pouches, as a
function of storage time and temperature. About one gram of the film was weighed and
cut into pieces approximately 1cm x 1cm. The film samples were transferred to the
extraction thimble and were extracted with 100ml of acetonitrile solvent for 24 hours.
After cooling, the extracted solutions were transferred to 100ml volumetric flasks and
brought to volume with acetonitrile (99.9%, HPLC grade solvent, EM science). A 10ml
aliquot of the respective extraction solution was filtered using microfilter (0.45 ,u, HV,
Millipore Corp.) and transferred to 4m] glass vials (Supelco Inc, Bellefonte, PA) with
screw caps (Supelco Inc, PA) and PFTE covered septum (Waters, Millipore Corp., MA),
which were designed for the auto-sampling system of the high pressure liquid

chromatograph.

58

3-7-2. Percent Recove_ry of Antioxidants for Extraction System

The determination of antioxidant recovery was defined as the percentage loss of
antioxidants from the extraction system. Sample solutions of each antioxidant were
prepared by weighing about 0.02 gram of BHT or a-tocopherol and the weighed sample
transferred to a 200ml volumetric flask with acetonitrile solvent.

The concentration of the respective antioxidants was determined by HPLC analysis. The
area response of the antioxidant solution (i.e. control) obtained from HPLC analysis was
used as a basis for determining the percent recovery.

The antioxidant solutions were carried through the soxhlet extraction procedure, with no
film added to the extraction thimble. Here, 100ml of the antioxidant solution, of known
concentration, were transferred to the soxhlet extraction system. The conditions of the
extraction procedure were similar to those described for the extraction procedure for the
test films. Following the extraction period, the extraction solution was transferred to a
100ml volumetric flask, made up to volume and assayed for antioxidant concentration by
HPLC. The area response obtained from the extraction solution was compared with the
area response of the initially prepared antioxidant solutions. The percent recovery test of

each antioxidant was performed in triplicate.

59

3-7-3. Irgh Pressure Liguid Chromatoggaphv (HPLC) Analysis

Quantification of antioxidant levels in the test films following Soxhlet extraction
was determined by a HPLC procedure. Analyses were carried out on a Waters Model
150-C ALC/GPC equipped with a Waters 486, Tunable Absorbance Detector and
interfaced to a Waters Model 730 Data Module. The chromatographic conditions were as
follows:

The mobile phase for BHT analysis was Methanol (85%) and Water (15%) by volume,
while pure Methanol (99.9%, HPLC grade solvent, J.T.Baker Inc.) was the mobile phase
for the analysis of or-tocopherol. Absorbance was recorded at 280nm. A HPI C18
(Delta- Pakm, Waters Corporation, Milford, MA) column, 3.9mm x 150 mm length, was
used for analysis. The volume of injected sample from the 4m] single vials was 20 111.
The flow rate of the mobile phase was 1 mllmin; the total run time of one sample was 6
minutes at room temperature. The retention time for BHT was 3.58 minutes. The
retention time for or-tocopherol was 4.5 minutes. Typical peak profiles of the BHT and
(ll-tocopherol solution, obtained from the integrator are shown in Figures 39 and 40,
respectively. The sample compormds were detected with the UV detector set at 280nm
wavelength. The detector sensitivity was set at 0.03 AU; the filter of the detector was set
at 0.1 seconds. The integrator (Water, Data Module Model 730) interfaced to the
detector had a chart speed of 0.8 cm/min. The integrator conditions were set at 5 for peak
width; 15 for noise rejection; and 100 for area rejection.

a-Tocopherol (C29H5002, MW 430.72) from Eastman Kodak Company (Rochester, New

York) was used for preparation of standard solutions for calibration. BHT(C15Hz4O, MW

60

220.35) was obtained from the Eastman Chemical Company (Kingsport, TN). The
antioxidants (BHT, d-or-tocopherol) were quantified by peak area, by comparison to a
external standard calibration curve which was constructed, using a serial dilution
procedure with concentrations of 2, 4, 6, 8, 10ppm (w/v). The calibration data of each

standard antioxidant solution are presented in Figure 37 for BHT and in Figure 38 for or-

tocopherol.

The level of antioxidants extracted fiom the films was determined by substitution into the

following equation:

(Rs x C.F. x Vtotal)
(Vinj x Wtfilm)

 

% Antioxidant (w/w) = [ ] x 100 ~---(13)

where: Rs = detector response values for the sample (A.U.)
Vm. = total volume of solution (ml)
Vin; = volume of unknown solution injection (m1)
C.F. = calibration factor from the standard calibration curve (g/A.U.)

thm = the tested film weight (grams)

61

4. Anplvticgl Test for Lipid Oxidggon in the Model Prom

4-1. Linoleic Acid Analysis
4-1-1. Extraction

Linoleic acid in the model product system was extracted with
chloroform/methanol (2:1, (v/v)) at a ratio of 20:1, solvent to sample, as described by
Nelson (1991). A 3g sample of the model product system was cut into small pieces. The
test sample was placed into a 500ml separatory funnel. Chloroform (100ml, Mallinckrodt
Chemicals Co., Paris, Kentucky) was added to the sample and the mixture shaken slowly
for 1 minute; 50ml of methanol (99.9%, HPLC solvent grade, J.T.Baker, Phillipsburg,
NJ) were then added and the mixture shaken again for 1 minute. After mixing, the
solution was stored in the separatory flmnel for 4 hours at room temperature. The
chloroform-methanol solution from the separatory flmnel was collected into a 250ml
volumetric flask. Afler collecting the chloroform-methanol solution from the test sample,
50ml of fresh chloroform were added to the separatory funnel and then 25ml of methanol
were added. The second mixture was shaken for 1 minutes and stored in the separatory
funnel for 4 hours at room temperature. The second extraction solution was collected,
added to the initial extraction solution in the volumetric flask and evaporated to dryness

using a dried nitrogen gas stream for 28 hours.

62

4-1-12. Prepmtion of Methyl Esters

The preparation of methyl esters of fatty acids for capillary gas chromatography
analysis has been reported by several researchers (Slover, 1979; Luddy, 1968; Morrison,
1964), to obtain a rapid and quantitative analysis.

To the residue remaining, following concentration to dryness, was added lml of boron
fluoride-methanol reagent (10% methanol, Supelco Inc, Bellefonte, PA) as an acidic
catalyst for the methylation of linoleic acid. The volumetric flask was then closed tightly
with the cap using a Teflon and paraffin sealing tape. The flask was heated in a boiling
water bath for 2 minutes and then cooled to room temperature. Methanol (249ml, 99.9%,
HPLC grade solvent, J.T.Baker Inc.) was added to the volumetric flask (250ml), and a
lul sample of this solution was injected with a Sul (series 800) syringe (Hamilton Co.,
Reno, Nevada) directly into the gas chromatograph, for analysis of the methyl ester of
linoleic acid. A Hewlett-Packard Model 5890A gas chromatograph equipped with dual
flame ionization detectors and a fused silica capillary SPB-S nonpolar column (30m x
0.32mm ID) was employed (Supelco Inc., Bellefonte, PA). The GC conditions were as
follows. The column temperature was programmed from the initial stage of 150°C for 4
minutes to the final stage of 220°C for 8 minutes at the rate of 5°C/min. Helium was used
as a carrier gas, at a flow rate of 2.2ml/min. Injector temperature, 220°C; and detector
temperature, 250°C.

Standard calibration solutions were prepared using a serial dilution procedure with 14,
20, 27, and 34 n mole concentrations of trans-9, 12 —octadecadienoic methyl ester

(methyl linoleate) with chloroform (99.9%, Mallinckrodt Chemical Co., Paris,

63

Kentucky). Trans-9, 12—octadecadienoic methyl ester was obtained from the Supelco
Company (Bellefonte, PA). The data obtained for standard calibration of trans-9, 12—
octadecadienoic methyl ester is shown in Figure 42. The retention time of trans-9, 12 —

octadecadienoic methyl ester was 18.7 minutes.

The level of linoleic acid in the model product was calculated by substitution into the

following equation:

(Rs x C.F.(mole/Rs) x Vt (ml)) / Vi (ml) = mole ester ----- (i)
1 mole acid with mol. wt acid equals 1 mole ester with mol. wt ester ------ (ii)
From the Equation (i) and (ii):

Concentration of linoleic acid (mole/gram product)

 

_ (Rs x C.F. x Vt) -______ (1 4)
(Wtsample x Vi)
where: Rs = detector area response values for the sample (A.U.)
CE = calibration factor fiom the standard calibration curve (mole/A.U.)
Wtwnple = the tested sample weight (grams)
Vt = the tested total volume (ml)
Vi = injection volume (ml)

mol. wt acid = Linoleic acid (C13H3202) = 280.46 (FW)
mol. wt ester = Linoleic acid methyl ester (C19H3402) = 294.5 (FW)

64

4-1-3. Perce_nt Recovery of Linoleic Acid through Extraction and Methylation

The percent recovery of linoleic acid was determined by performing the
extraction and methylation procedures with a known amount of linoleic acid. A weighed
amount of pure linoleic acid (99%, Acros Organics, Fisher Scientific) was used for
quantitative analysis. A 1141 volume from the methyl ester solution was injected directly
into the gas chromatograph for analysis. The area response obtained by GC was applied
for calculation of the amount of linoleic acid.

The amormt of linoleic acid obtained by GC analysis was used to detemrine the percent
recovery, where the determined value is compared with the amormt of the original

weighed linoleic acid. Injections of GC analysis were done in triplicate.

4-2. Hexanal Analysis

4-2-1. Apparatus for Trapping of Hexanal Commund (A _Dypamic Luge and TLap
System)

A dynamic purge and trap system was designed for headspace sampling of
volatiles from the model product system. Erlenmeyer flasks of a 250ml size were
modified with 29/42 standard taped male joints, to fit the dispersion tube assembly of a
gas washing bottle (Stopper assemblies for Coming 31770 gas washing bottles, Fisher

Scientific, Pittsburgh, PA). Modification of the dispersion tube assembly of the gas

65

washing bottles and the Erlenmeyer flasks were performed by the Glass Blowing Shop of
the Chemistry Department at Michigan State University. A schematic diagram of the
dynamic purge and trap system is shown in Figure 8. A modified Erlenmeyer flask was
interfaced to a flow meter and needle valve assembly, connections were through 1/8”
OD. copper tubing and swagelok fitting. A constant rate of flow of nitrogen gas was
controlled by the flow meters.

An amount of the model product system (approximately 6 gm), was removed
from the packaged pouch, weighed and placed into a modified Erlenmeyer flask (250ml)
equipped with an inlet and outlet port. The inlet port of the dispersion head was
connected to a flow meter and needle valve (Nu Pro type B-ZSG) to regulate the flow of
nitrogen gas. The sorption trap was connected to the exit port of the dispersion head via
swagelok adapters. The dispersion head exit port of 8 mm O.D. glass tubing was
connected by a 5/ 16” swagelok nut and a series of reducing adapters to a 1/4” male
swagelok fitting. The sorption trap was mounted to the dispersion head with a 1/4”
thumb wheel swagelok fitting (Supelco Inc., Bellefonte, PA) for easy removal and could
be affixed to both glass and metal desorption traps. Figures 8 and 9 show the trapping
cell and the complete purge and trap system. The glass thermal desorption tubes
(Carbotrap 300, 6mm ID. x 4mm ID. x 11.5cm) used in the present study were
prepacked by Supelco Inc. (Bellefonte, PA). The trapping tubes were packed with 300pg
of Carbotrap C absorbent, 200pg of Carbotrap B absorbent and 125pg of Carbosieve S-
111 absorbent. Nitrogen was flowed through the assembled purge and trap system at room
temperature to remove oxygen from the system prior to heating. After one hour, the

water bath (Blue M Constant Temperature Bath, Blue Island, IL) was turned on and

66

allowed to heat for one hour to reach 50°C. The modified Erlenmeyer flasks were then
placed in the water bath and the system purged for 24 hours. Repetitive analyses of the
model product showed no additional hexanal detected. Each sample was analyzed in
triplicate.

The thermal desorption unit (Model 890, Dynatherm Analytical Instruments, Inc supplied
by Supelco, Bellefonte, PA) conditions were as follows:

Helium was used as the carrier gas with a flow rate of 7.5ml/min. at 40psi.

The desorption temperature was set at 250°C for 6 minutes; The temperatures of the valve
and transfer line were set at 230°C: The conditioning temperature for cleaning the

sorption tubes, prior to reuse, was 280°C for 30 minutes.

 

 

 

L
V

 

 

 

 

 

 

armchair:
.“w 1.9 es"

MT

  

 

 

 

 

Oflfliflifil

 

 

 

 

 

 

 

 

B
23

L R1

 

 

\_/
Nitrogen Gas

R1. R2. R3 : rotometers MT : modified Erlenmeyer flasks
T1. T2. T3 : trapping tubes WB : water bath

Figure 8. A schematic diagram of the dynamic purge and trap system

67

 

(Dynamic and trap system) (Trapping cell apparatus)

Figure 9. A Dynamic and trap system for the model product

68

4-2-2. The Analysis Procedure of Hexanal Comppund

The levels of sorbed hexanal were quantified by a gas chromatography-mass
spectrometry (GC/MS) procedure. Volatile compormds including hexanal, sorbed on the
Carbotrap tube were desorbed from the thermal desorption unit (Dynatherm Analytical
Instruments, Inc.) and transferred to a Hewlett-Packard 5890 Gas Chromatograph, which
was interfaced to a quadrupole mass spectrometer (Model HP 5970) with a Mass
Selective Detector (MSD) and a chemstation data system. The gas chromatograph was
equipped with a SPB-5 non-polar fused silica capillary column (60m x 0.32mm ID,
Supelco Inc., Bellefonte, PA). The level of hexanal in the product model was analyzed
by GC/MS using a mass spectrometer (Model HP 5970) interfaced to a GC (Model HP
5890) and a Chemstation data system. The conditions of Gas chromatography/Mass
spectrometry analysis for hexanal were as follows:

The initial temperature was set at 40°C for 6 min; the temperature was then raised at 5
degree/min; and the final temperature and time were held at 200°C for 10 minutes.
Helium was used as a carrier gas, with a flow rate at 10 ml/min. The solvent delay was 5
minutes, and the cycles per second were 1.70. The temperature of the injection port was
set at 220°C. The temperature of the transfer line between the gas chromatograph and the
mass spectrometer was set at 25 0°C.

The total ion chromatogram peaks, or the individual ion profiles were integrated, based
on the integrator events that were set up with 0.2 as the peakwidth and 14 as the

threshold. The electron multiplier voltage was 2600eV.

69

Calibration of the GC/MS system, for quantitative analysis of hexanal, was performed by
direct injection of lul of known hexanal standard solutions onto the carbotrap tube.
Hexanal standard solutions of known concentration were prepared by a serial dilution
procedure with methanol. Hexanal for the standard calibration was obtained fi'om the
Aldrich Chemical Company, (98% grade, Milwaukee, WI). The hexanal external

standard calibration data by GC/MS are shown in Figure 43.

4-2-3. M_a_ss Spectrometer

The gas chromatograph (Model HP 5890, Hewlett Packard, Avondale, PA) was
interfaced to a quadrupole mass spectrometer (Model HP 5970) with a Mass Selective
Detector (MSD). The analysis of hexanal by the GC/MS system was carried out with the
Chemstation (Model HP 5 970), at an ionization energy of 70eV. The analysis of hexanal
was performed using a gas chromatograph/mass spectrometer at the Mass Spectrometer
Laboratory in the Department of Biochemistry at Michigan State University. The
desorbed compounds from the thermal desorption system were separated by gas
chromatography with helium carrier gas. A library search for the identification of
hexanal by the mass spectrometer was performed with the NIST Mass Spectral Search
program. Autotuning was employed when operating in the election impact mode, using
perfluorotributylamine (PFTBA) as the tuning compound. An ion group, containing five
selected ions, identified fi'om the hexanal library search, namely 29, 44, 56, 72, and 100

m/z, was used for quantification.

70

The total ion chromatogram of the five selected ions in SIM was used for the
quantification of hexanal. The ratios of the selected ions from the mass spectrum were
identified for qualitative identification of hexanal, by comparison with the standard
hexanal mass spectrum from the library search. The retention time of the
chromatographic ion peak in SIM also supports the identification of hexanal, when
compared with the retention time of the external hexanal standard. The calibration data
for the standard hexanal relations is presented in Table 58. The retention time for
hexanal from SIM analysis was 8.5 minutes.

The concentration of hexanal in the product model was determined in duplicate for each
time interval. The hexanal level of the product model was calculated by substitution into

the following equation:

----- (15)

Hexanal level (mole/ g) = [M]

Wtsample
( )

where: Rs detector response values for the sample (A.U.) obtained by GC/MS
C.F. calibration factor from the standard calibration curve (mole/AD.)

Wimp”,= the tested sample weight (grams)

71

412-4. Percent Recovery of Hexanal through a Dynamic Purge_and Trap System

The percent recovery study was carried out to determine the percentage loss of
hexanal from the dynamic purge and trap system. A sample solution of hexanal, of a
known concentration level in methanol, was added directly to the carbotrap glass tube. A
lul aliquot of the sample solution was injected into the top section of the carbotrap glass
tube. The carbotrap glass tube was then transferred to the thermal desorption rmit
interfaced with the GC/MS system for hexanal analysis. The area response obtained from
a sample of known concentration using the GC/MS analysis was used as a basis for
determining the percent recovery. An aliquot containing hexanal of the same
concentration level as used above was carried through the dynamic purge and trap
system. A In] sample of hexanal solution of known concentration was transferred to the
erlenmeyer flask of the dynamic purge and trap system. The carbotrap glass tube was
installed and the hexanal adsorbed by a constant flow of nitrogen gas for 24 hours at 50°C
in the waterbath, after flushing at room temperature for 1 hour. The carbotrap glass tube
was then transferred to the thermal desorption unit interfaced with the GC/MS system for
hexanal analysis. The area response obtained by a dynamic purge and trap system was
used to determine the percent recovery by comparison with the area response obtained
fiom the carbotrap tube by direct injection. This percent recovery test was performed in

triplicate.

RESULTS AND DISCUSSION

4-1. Product Model

Two model product systems were prepared, which were formulated with high and
low levels of lipid content. The product formulations are summarized in Table 2. The
model product systems were characterized by the relationship between water activity and
moisture content in the product. The effectiveness of antioxidant-impregnated laminate
film was determined by studying the oxidation of linoleic acid in the model product with

time. For these studies, model product system (B)(low loading level of lipid) was used.

Table 2. Components of the model product (A, B) on a wet weight basis

 

 

 

 

 

Model Product (A) Model Product (B)
Components Weiwg) % (w/w) ‘ WeM % (why
Linoleic Acid 40.59 4.06 3.6 0.36
Tween 20 0.11 0.01 0.11 0.01
CMC 130 13 130 13
Deionized 829.3 82.9 866.29 86.63
water
Total 1000 100 1000 100

 

The composition of the model product, on a wet weight basis, is shown in Table 2. The
weight of the model product, on a dry weight basis, was based on the sample weight after
freeze-drying. The weight percent of the linoleic acid in the test product, on a dry weight
basis, was determined after fi'eeze—drying. Summarized in Table 3 are the weight
percentages and actual weight values for linoleic acid for the product B formulation

samples on a wet basis, following freeze drying, sample weight following equilibrium to

72

73

30% RH and the sample weight typically used for the purge and trap analysis for hexanal

concentration.

Table 3. The percent of linoleic acid in model product (B) calculated.

 

 

Product sample Sample weight (g) % of linoleic acid (w/w)
Wet basis 33.7501 0.36
Dry basis 5.6143 2.2
Equilibrated 5.8422 2.1
To 30% RH
Test weight 0.1737 2.0797

 

The initial concentration of linoleic acid in the model product (B) following equilibrium
at 30 percent relative humidity was estimated at 0.021(g/g), based on the initial product
composition (i.e., wet basis). The initial moisture content (IMC) of the product
formulations was determined on the freeze-dried samples. Freeze dried samples were

also used to construct sorption isotherms for the respective product formulations.

4:2. @111 Moisture Content

The initial moisture content was determined and an equilibrium sorption isotherm
was developed to describe the relationship between Aw and moisture content of the
model product. The results of the initial moisture content for model products A and B are
tabulated in Table 4. The analysis was carried out in triplicate. The initial moisture
content (IMC) of the fieeze-dried products was determined by a gravimetric method and

was obtained by substitution into Equation 14 of the Material and Method section.

74

Table 4. The initial moisture content of freeze-dried model products A and B.

 

 

(gHzO/ l 00g drybasis product)
Model Product (A) Model Product (B)
IMC (%) 3.46 i 0.25 3.63 i 0.22

 

All values are the average 1 standard deviation of three replicated samples.

The initial moisture content values for each model product were 3.46 (A), and 3.63 (B)
gHzO/100g dry product, respectively. The calculations of experimental data are given in

Appendix A.
4-3. Eqailibriu_m Sorption Isotherm

The relative humidity levels for the respective test chambers were monitored with
a hygrometer (Troller and Christian, 1978) and were found to be in good agreement with
those reported by Rockland, et a1. (1980) and Labuza (1976). A constant weight of the
model products was obtained at each relative humidity condition, at three difl‘erent
temperatures.
Values for the equilibrium moisture content and the associated relative humidity values
for the freeze-dried product, determined at 18, 28, and 38°C are summarized in Tables 5 -
7 (model products A) and Tables 8 - 10 (model products B), respectively. The
equilibrium sorption isotherms for the obtained sorption data are plotted in Figures 10 -
13 (model products A), and Figures 14 —- 17 (model products B), respectively. The
calculations for the equilibrium sorption isotherms are included in Appendix A (see

Tables 26 to 31, respectively).

75

The equilibrium sorption isotherms of the model products (A, B) showed the typical
sigmoid shape at each test temperature. The equilibrium sorption isotherms, at increased
temperature, also showed a slightly higher water activity at the same moisture content for
the model product.

The equilibrium sorption isotherms of the model products (A, B), as represented by the
relationship between moisture content and water activity, provided data which allowed
for characterization of the model product, as well as for the selection of storage
conditions for the product stability studies.

With respect to the equilibrium sorption isotherms obtained for the model product
system, a series of equations developed to describe the characteristic geometric
configuration of sorption isotherm shapes were evaluated to select the best-fit model. A
detailed description of the isotherm expressions evaluated and the mathematical treatment
followed is presented in Appendix B. The linearized plots of the best-fit model are also
presented graphically in Appendix B (see Figures 25 to 30, respectively).

For all cases, the mathematical equation that best described the sorption isotherm of the
model products was the Halsey equation, except for model product B at 38°C, which was
best described by the Henderson equation. The monolayer moisture content at each
temperature was 8.57 (18°C), 9.12 (28°C), and 8.38 (38°C) for model product containing
high linoleic acid levels and 9.88 (18°C), 9.84 (28°C), and 9.23 (38°C) (g H20/100g dry

product) for model product containing low linoleic acid levels.

76

Table 5. The Equilibrium moisture content for the freeze-dried product (A) as a function
of water activity, at 18°C.

 

 

 

 

 

 

 

 

Water Activity Average EMC (%)

0.096 8.1504

0.214 10.8778

0.355 13.2473

0.46 15.5210

0.56 19.0749

0.745 30.1573

0.805 35.9295

40
35 -
g 30 «
‘E
g 25 ~
g 20 J
a is.
"35';
m 10 ~
5 .
o . . . . . . . .
0 0.1 0.2 0.3 0.4 0.5 0.8 0.7 0.8 0.9
Water Activity (Aw)

 

 

 

Figure 10. The Sorption isotherm at 18°C of model product (A).

77

Table 6. The Equilibrium moisture content for the freeze-dried product (A) as a function
of water activity at 28°C.

 

 

 

 

 

 

 

 

Water Activity Average EMC (%)

0.14 8.5393

0.25 10.5386

0.38 13.5261

0.455 15.9322

0.55 18.1299

0.63 21.9630

0.735 29.3419

0.815 34.2381

35
3o .-
§ 25 «
s
E 20 ..
g 15 i
“8; 10 ..
5 ..
0 i : : i : : : :
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
WaterActivity(Aw)

 

 

 

Figure 11. The Sorption isotherm at 28°C of model product (A).

78

Table 7. The Equilibrium moisture content for the freeze-dried product (A) as a flmction
of water activity at 38°C.

 

 

 

 

 

 

 

 

Water Activity AveraEEMC (%)
0.1 16 8.7792
0.2 10.2812
0.345 12.6412
0.445 14.9042
0.535 16.2677
0.755 27.7705
0.815 32.2779
0.905 55.0928
I)
a).
in i
i an
in.
1).
o . e E
0 01 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
“human

 

 

 

Figure 12. The Sorption isotherm at 38°C of model product (A).

79

 

 

 

’3

3

‘3

_l
01
l

 

Equllibrium Moisture Content (%)
8

_a
o

 

 

 

 

mmm

0.9

 

Figure 13. The Sorption isotherms at 18, 28, and 38°C for model product (A).

 

80

Table 8. The Equilibrium moisture content for the freeze-dried product (B) as a filnction
of water activity at 18°C.

 

 

 

 

 

 

 

 

Water Activity Average EMC (%)
0.094 8.04579
0.215 10.6671
0.35 14.361 1
0.46 17.4871
0.58 21.4555
0.756 36.3225
0.815 43.5431
50
45 ..
40 .. /
a!
V 35 ~
g 301.
g 2. ..
5 20 ..
i e l
E
10 J-
5 ..
0 A, % ; fi‘ + 2 fi‘ .
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.3 0.9
Water ActIvIty(Aw)

 

 

 

Figure 14. The Sorption isotherm at 18°C of model product (B).

81

Table 9. The Equilibrium moisture content for the freeze-dried product (B) as a function
of water activity at 28°C.

 

 

 

 

 

 

 

 

Water Activity Average EMC (%)
0.1 1 6.6965
0.235 9.3052
0.345 12.671 1
0.455 16.4800
0.545 18.8241
0.75 30.1 192
0.805 39.4153
45
4o ..
E ’5 "
g 25
a 20 *l‘
5
3 15 ..
5
3-
m 10 ..
5 ..
O a: f - f + : : 4
o 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
WaterActlvltHAw)

 

 

 

Figure 15. The Sorption isotherm at 28°C of model product (B).

82

Table 10. The Equilibrium moisture content for the freeze-dried product (B) as a flmction
of water activity at 38°C.

 

 

 

 

 

 

 

 

Water Activity Average EMC (%)
0.1 14 5.4358
0.198 9.4891
0.345 1 1.6215
0.47 15.7390
0.51 17.3351
0.755 31.3016
0.815 36.0693
40
35 ..
... 30 ..
23
g 25 ..
g 20
E
S 15 «-
i
m
10 4
5 w
0 t i : f a: ‘f e c
O 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Water Actlvlty (Aw)

 

 

 

Figure 16. The Sorption isotherm at 38°C of model product (B).

83

 

 

 

i .3

“.°°

i3

Eqilibrlum Moistln Content (%)
M
(II

J
0|
1

 

 

 

 

0 t 4 t : : c : r
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Water Activity (Aw)

 

Figure 17. The Sorption isotherms at 18, 28, and 38°C for model product (B).

 

84

4-4. The Moisture Con_tgnt Equilibrium of the M_odel Product aia Specific

Water Activity, and Temp_erature

Prior to packaging the product and initiating the storage studies, the freeze-dried
model product system was equilibrated to the desired water activity level (AW = 0.30) to
minimize the effect of water activity (Aw) on the rate of lipid oxidation. Here the dried
model product was placed in an equilibration chamber at ambient temperature (23°C),
and humidified nitrogen gas was continually flowed through the chamber to allow
equilibration of the product model to the required water activity. The results of the
hexanal analysis of the product model following equilibration showed that lipid oxidation

did not take place to a significant extent during the equilibration period.

4-5. Model Product Shelf Life and Stability of Antioxidants

A flow diagram of the test scheme followed is presented in Figure 7 in the
Materials and Methods section. The storage stability of the model product system was
determined to evaluate the ability of antioxidant impregnated films to inhibit lipid
oxidation, via the evaporation/sorption mechanism previously described. The storage
studies were performed at conditions of 23°C and 45°C, respectively. Oxidation of the
model product, stored at each condition, was monitored over a period of 28 weeks at

23°C and over a 6 day period at 45°C.

85

The pouch film was assayed over a 20 week period for the studies carried out at 23°C and
for 6 days at 45°C, to allow monitoring of the relative concentration of antioxidant

retained by the package structure with time.

4-6. Los_s of Antioxidants (_BHT, or-tocopherol) from Coextruded Laminate Films

 

The level of antioxidants present in the packaging films was determined as a
function of time and temperature. The results of the initial antioxidant concentration

levels are summarized in Table 11.

Table 11. The package film structures evaluated.

 

 

Film (HDPE/Sealant Lamination) The initial concentration of
antioxidants (wt/wt %)
l . Control No antioxidants
II . BHT 0.1137 %
(3,5-di-tert-buthyl-4- hydroxytoluene)
III . (rt-tocopherol 0.0073 %

 

The initial concentration of BHT incorporated into the laminate film structure for storage
stability studies was 0.1137 %(w/w). The initial concentration of cl-tocopherol
incorporated into the test laminate film structure was 0.0073 %(w/w).

The levels of antioxidant remaining in the packaging pouch structures, as a fimction of
storage time at 23°C, are summarized in Table 12. The relative percent loss at BHT and
rat-tocopherol from the coextruded pouch materials, as a function of storage time at 23°C,
is represented graphically in Figure 18, where the relative percent antioxidant retained is

plotted as function of storage time. Statistical analysis was performed using Sigma stat

86

1.0 (Jandel Corp., San Rafael, CA). Appropriate comparisons were made using the
Student-Newman-Keuls test for multiple comparisons by a one-way analysis of variance
(AN OVA).

Table 12. The level of antioxidants from coextruded packaging pouches as a function of
storage time (23°C, 50 %RH).

 

 

 

Storage Conc. Of antioxidants Relative percent retained
Time ppm (w/w) antioxidants(Ct/Co)x100
Weeks BHT Tocopherol % BHT % Tocopherol
0 1137:59' 73:1.0 ‘ 100 100
2 608:5.7 " 63.5:36 " 53.5 87
4 446.5:490 ° 62:67 h 39.3 85
6 279:26.1 ‘ 61512.6 b 24.5 84
8 2113589 ° 6012.4 b 18.6 82.2
10 149:4.9f 45:9.5° 13.1 61.6
12 11320.58 40.5:153 ° 9.9 55.5
14 79.5:132 " 29:11.9 ° 7.0 39.7
16 36.5123 ' 27:3.7° 3.2 37
18 32:13.2l 26:7.1 ° 2.8 35.6
20 N/A N/A N/A N/A

 

Average i standard deviation of each pouch as a function of time.
° ' ' Means with different superscripts in same column are significantly different (p<0.05).
N/A means it is not available due to a limit of the detector sensitivity.

The relative percent of antioxidant retained in the packaging pouches was 2.8% of BHT
and 35.6% of or—tocopherol, after 18 weeks of storage at 23°C. As shown, within a two
week storage period at 23°C, approximately 50 % of the included BHT was lost fiom the
packaging pouch material. The loss of or-tocopherol from the packaging pouches over

the same period of time was approximately 10%, with nearly 50% a-tocopherol

87

remaining after 12 weeks of storage at 23°C. The concentration of BHT retained in the
packaging pouches showed statistically significant differences (p<0.05) between samples,
as a flmction of storage time. After a 10 week storage period at 23°C, the level of
or—tocopherol exhibited highly significant difi‘erences (p<0.05) when compared with the
initial concentration in the packaging pouch material, with a statistically significant
difl‘erence after the first two weeks, while the BHT concentration retained in the pouch

structures, showed a highly significant difl‘erence within the first two weeks.

 

 

70-~ Tocophonfl

4o .. BHT

Penxunrxrchunedruflkumnunsflidulflkns
88

 

 

 

0 2 4 6 8 10 12 14 16 18
suxageoweeks)

 

 

 

Figure 18. The relative percent loss at BHT and ct-tocopherol from coextruded pouch
films as a flmction of storage time (23°C).

88

Comparing these results, with the data obtained by Lin (1996) for three layer laminate
films, showed similar loss rates of or—tocopherol as a funtion of storage time (23°C).

After removing the model product, the BHT and a-tocopherol content of the pouch
material was also determined as a function of storage time, for the product stability
studies carried out at 45°C and 50% RH. The results of these acclerated stability studies,
involving determination of antioxidant levels from the package structures with time, are
summarized in Table 13 and presented graphically in Figure 19.

Table 13. The level of antioxidants from coextruded packaging pouches as a function of
storage time (45°C, 50% RH).

 

 

 

Storage Time Cone. of antioxidants Relative percent of retained
ppm (w/w) antioxidants(Ct/Co)x100
Days BHT Tocopherol % BHT % Tocopherol
0 1 1681253 ‘ 7316.3 ' 100 100
2 l85:28.4 b 72512.4 ‘ 15.8 98.9
4 40:3.3 ° 7212.5 ' 3.5 98.0
6 N/A 55511.0 b N/A 74.8

 

Average 1 standard deviation of each pouch as a function of time.
" c Means with different superscripts in same column are significantly different (p<0.05).
N/A means it is not available due to a limit of the detector sensitivity.

As shown, no BHT was found in the BHT-impregnated laminate pouch structure after 6
days of storage at 45°C. For the a—tocopherol level, approximately 25% of the initial
quantity was lost, after 6 days at 45°C. The concentration of or—tocopherol retained in the
packaging pouch film structure showed no significant difference (p>0.05) until a 6 day
storage period at 45°C. The levels of BHT retained in the antioxidant impregnated
packaging pouch film indicated a significant difference (p<0.05), as a function of storage

time, until the complete loss of BHT at 45°C.

89

 

 

 

Percentofthorotainodantiorddantshthefltms
(96)
B 8 6 8 8 3 8 8

_a
O

 

 

 

O

 

100 120 1fl

o
8
5 .
8
8

 

 

 

Figure 19. The relative percent loss at BHT and a—tocopherol from coextruded pouch
films as a frmction of storage time (45°C).

The rate of loss of BHT from the package film structures was found to be much higher
than the rate of loss of or-tocopherol, at both storage conditions.

The observed difl‘erences between the rates of loss of BHT and or—tocopherol item the
laminate packaging pouch may be attributed to a higher rate of diffusion of BHT through
the laminate surface layers, or a difi‘erence in the rate of evaporation of the BHT fi'om the
HDPE and heat sealant layers, as well as to the equilibrium partition distribution of the
antioxidants between the respective layers of the lamination. Antioxidant migration is

related to the solubility of the additive within the respective layers of the lamination and

90

the diffusion of the additive through it. Differences in either the solubility or difiusivity
can affect the transmission characteristics of the respective antioxidants. The solubility
difi‘erence depends primarily on the difference in the physical and chemical nature of the
migrating molecules and the respective laminate layers and will be reflected in the
partition distribution of the antioxidant between the laminate layers. On the other hand,
the difl‘erence in antioxidant diffusivity is determined mainly by the size and shape of the
molecules, and by the degree of aggregation among the diffusing molecules within the
polymer layers. The fact that the initial concentration of the BHT was significantly

higher than the initial level of a—tocopherol impregnated into the laminate fihn structure

may also be a contributing factor to the observed lower rate loss for or—tocopherol.

4-6-1. The Rate of Loss of MW

The rate of loss of antioxidants from coextruded pouch film was reported by

Bailey (1995) and Lin (1996) to follow a first-order or pseudo first-order rate expression:

ln(Ct/Co) = -1n ------- (17)

Where Co : the initial concentration of antioxidants in the film
Ct : the concentration (w/w, %) at time = t
k : the rate constant of the antioxidant loss
t : the time interval

A plot of log(Ct/Co) as a flmction of storage time for antioxidant losses at storage

temperatures at 23 and 45°C, indicated that a first order expression provided a good

91

description of the rates of loss of the respective antioxidants fiom the laminations
evaluated in the present study (see Figures 20 and 21).

The first-order rate equation constants (k) for the loss of antioxidants from pouch films
were determined at each temperature and are summarized in Tables 14 and 15,
respectively.

The following expressions were derived from a least square fit of the rate loss data.

23°C (t = weekiu_nl_'1)
(Ct/Co) x 100 = 87.452 * exp (-0.l925t) .......... BHT R2 :0.9887

(Ct/Co) x 100 = 109.74 * exp (-0.0625t) .......... Tocopherol R2 :0.9204

Table 14. Rate constants for the loss of antioxidants from the packaging pouch structure
(23°C).

 

Antioxidants Loss Rate Constant k (hr '1)

 

BHT 0.001 15
Tocopherol 0.00037

 

92

 

 

 

 

 

 

3 100
.5 1: ‘ A
g j: x ‘ Tocopherol
(I ,, ‘ K .-
3 3 .
8 3?
c A ..
a1 8 \
2f; 10 BHT
i! ‘T \
e E ii \
‘8 I: r
E g i \
i8 1
E 1 a 4 4. : t t t f
0 2 4 o 8 10 12 14 16 18
Storage (weeks)

 

 

 

Figure 20. Relative percent loss of BHT and a-tocopherol from the pouch fihns as a
function at storage time (23°C).

45°C (t = hours unit)

(Ct/Co) x 100 = 96.416 * exp (-0.0350t) .......... BHT R2 :0.9968

93

Table 15. Rate constant for the loss of BHT fi'om the packaging pouch film structure
(45°C).

 

Antioxidant Loss Rate Constant k (hr ’1)

 

BHT 0.0350

 

 

 

A

10 4
4»
«t

 

 

 

Relative percent of retained antioxidants In the
pouch fillms (Ct/Co) x 100

 

0 20 40 60 80 100 120 140

 

 

 

Figure 21. Relative percent loss of BHT from the pouch film as a frmction at storage time
(45°C).

As shown in Tables 14 and 15, the rate loss constant values for BHT fiom the packaging

pouch film structure, at the respective temperatures, were much greater than the rate
constant for the loss of or—tocopherol from the packaging pouch film structure. The loss

of or-tocopherol from the packaging pouch film structure at 23°C showed a similar loss

94

rate value (k= 0.00037 hr"), as compared to the loss rate constant (k= 0.0004 hr") of or-

tocopherol obtained by Lin (1996) fiom the coextruded laminate film structures.

4-6-2. Percent Recovery of Antioxidants from the Poach Film

The percent recovery of antioxidants carried through the soxhlet extraction /

HPLC analysis procedure is summarized in Tables 16 and 17, respectively.

Table 16. The percent recovery of BHT for extraction system.

 

 

 

Injection Cone. Area Response
(20 pl) 10 ppm Initial Cone. After extraction
1 547024 534472
2 551967 534732
3 545764 538658
Average 548251.7 535954
Percent recovery 97.76 %

 

Table 17. The percent recovery of a-tocopherol for extraction system.

 

 

Injection Conc. Area Response
(20 1.1.1) 10 ppm Initial Conc. Alter extraction
1 290202 282072
2 288542 287768
3 286662 289199
Average 2884687 286346.33

 

Percent recovery 99.26 %

 

95

A recovery of 98% and 99% for BHT and or-tocopherol, respectively showed good

stability of antioxidants through the extraction procedure and HPLC analysis.

4-7. Level of Linoleic Acid in the Model Product following Storage at 45°C

The level of linoleic acid remaining in the model product packaged in the
respective test pouches and stored at 45°C, as a function of time, is summarized in Table
17 and presented graphically in Figure 22, where the linoleic acid concentration is plotted

as a frmction of storage time for the two antioxidant containing package pouches and the

 

 

     

antioxidant-flee control.
30
+c6nlrol
25 +Tocqlherd
+9111'
20 1 4 4

 

Levelofllnoleicacld(mmolelg)
a".

 

 

 

 

104

5.

0 . a . j . . . L
0 20 40 60 80 100 120 140

 

 

 

Figure 22. The concentration of linoleic acid in the model product as a function of storage
time (45°C).

96

As shown, the concentration of linoleic acid in the model product packaged in the control
and in the or-tocopherol impregnated laminate pouch structures decreased rapidly during
storage over a 2 day period at 45°C. Over a 4 day storage period, there also appeared to
be no statistically significant difference (p>0.05) between the level of linoleic acid in the
model product packaged ill the control and in the a-tocopherol impregnated laminate
pouch structures. This result indicates that the or-tocopherol impregnated laminate pouch
had no effect on retarding lipid oxidation in the model product, as compared to the
control non-impregnated laminated pouch, when stored at 45°C. This was attributed to
extensive fatty acid oxidation under the conditions of storage. However, the model
product packaged in the BHT immegnated laminate pouch did not show a notable change
in linoleic acid concentration, during storage for 6 days at 45°C, suggesting minimal lipid
oxidation. These data showed a correlation between the concentration of linoleic acid
present in the model product packaged in the respective laminate film pouches and the
level of antioxidants retained by the respective packaging pouches, as a function of

storage time at 45°C, 50% RH. (see Table 18)

Table 18. The concentration of linoleic acid from the model product as a function of the
storage time at 45°C, 50% RH.

 

 

(m mole/g)
Storage fins) Control BHT Tocopherol
0 2438:028‘“ 2438:028“ 24.381028“
48 0.87:0.04M 1971:1329 0.831004“
96 0.705003M 19.42:0.28'B 051950.11“A
144 0.55:0.01“ 19.311027“B 1.271004“?

 

Average : standard deviation of each model product as a function of time.
° ' ° Means with different superscripts in same column are significantly difi‘erent (I‘D-05)-
A ' C Means with different superscripts in same row are significantly different (p<0.05).

97

The lack of observed loss of linoleic acid from the model product in the BHT
impregnated laminate pouch, as a result of oxidation, can be attributed to a higher rate of
loss of antioxidant from the pouch structure and subsequent sorption by the model
product under these condition, than the rate of lipid oxidation of the model product. By
comparison, under similar storage conditions, little or no or-tocopherol is migrated to the
model product system to inhibit lipid oxidation. Also, significantly higher initial
concentrations of the BHT than the initial level of or-tocopherol may be an important
factor contributing to the observed inhibition of linoleic acid oxidation in the model

product.

4-7-1. The Percent Recovery of Extraction for Linoleic Acid

The percent recovery for the methylation of linoleic acid was formd to be
approximately 72 %, based on the initial quantity of linoleic acid carried through the
extraction step. However, when methyl linoleate was carried through the extraction step
(see page 57 in Materials and Methods), a near quantitative recovery was observed (i.e.
97 %). Assuming a 72% recovery for the methylation step and a 97% recovery for the
extraction step, the % recovery for the methylation/extraction procedure was nearly 70%,
and while not quantitative, the percent recovery for this procedure was repeatable (see

Table 19).

98

Table 19. The percent recovery of linoleic acid through the extraction and methylation
procedures.

 

 

System Samplel Sample2 Sample3 Recovery
(%) (%) (%) Avera89(%)
1 . Methylation 71 72.9 71.4 71.77
2. Extraction 96.5 96.8 96.2 96.5
3. Extraction 68.5 70.6 68.7 69.3
and methylation

 

4-8. Differe_nce in the Level of Linoleic Acid Between Model Product P_reparation and

F reeze-Dried Model Product

The model product for the storage studies was prepared by freeze-drying to
simulate a low moisture food product, as well as to provide an initial moisture content
which would minimize the rate of lipid oxidation of the resultant freeze-dried product
system. To measure the effect of freeze drying on linoleic acid concentration in the
freeze dried product, the methyl ester value of linoleic acid in the model product after
freeze-drying was compared to the value in the original preparation (before freezing and
freeze-drying). The percent recovery of linoleic acid in the model product following the
freeze-drying procedure was 30.3%, as summarized in Table 20. The concentration of
linoleic acid in the model product after freeze-drying showed a significant loss, as
reflected by the methyl linoleate levels determined for the model product system
following freeze-drying. The observed loss of linoleic acid may be due to volatilization of

linoleic acid from the model product during the fi'eeze-drying process, which involves

99

removal of water from the product. The composition of a model product with low lipid
levels may be affected significantly by even small environment changes: such as the
fieeze-drier condition and run time.

Table 20. The percent recovery of the linoleic acid in the model product between before
fieezing-dry and after freezing-dry condition.

 

 

 

Sample conditions (wet basis product) % Recovery
Level of linoleic Before freeziridry After fieezing-dry
acid in model 0.33% (g/g) 0.1% (g/g) 30.3%
product

 

 

 

Noormarji (1990) determined the extent of lipid oxidation on both the functional and
nutritional properties of chicken breast myofibrillar proteins during different stages of
freeze drying by TBA tests. A series of experiments showed that fi'eeze drying for 48 hr

increased lipid oxidation in the control and methyl linoleate treated sample, as compared
to freeze drying for 24 hr. Freeze drying for 48 hr showed much higher lipid oxidation
levels than that for 24 hr, in the presence of oxidized methyl linoleate. The longer periods
of freeze drying increased lipid oxidation as measured by TBA numbers (Noormarji,
1990). Such results suggest that the time of fi'eeze drying of the model product system

can play an important role in product lipid oxidation.

4-9. Prodpct Storage Studies

The results of hexanal analysis carried out on model product samples packaged in
pouch I (control pouch structure), pouch II (BHT impregnated pouch structure), and
pouch IH (or-tocopherol impregnated pouch structure) and stored at 23°C/50%RH are

summarized in Table 21. The relationship between the extent of lipid oxidation and the

100

respective laminate package structures evaluated is presented graphically in Figure 23,
where hexanal concentration in the product model system is plotted as a firnction storage
time. Data were analyzed with the use of a Sigma stat 1.0 statistical analysis program
(J andel Corp., San Rafael, CA). The statistical analyses were compared using the
Student-Newman-Keuls test for all pairwise multiple comparisons by a one-way repeated

measure analysis of variance (ANOVA), with a significance difference of p<0.05.

Table 21. The hexanal concentration of model product system packaged in pouches
fabricated from coextruded laminate structures (23°C, 50 %RH).

 

 

(n mol/ggoduct system)
Storage Time Pouch I Pouch II Pouch 111
(weeks) (Control) (BHT) (a-tocopherol)

0 3.39:0.68 3.39:0.68 3.39:0.68

2 5.59:2.01 4.83:0.79 5.89:0.72

4 6.251036 5.54:2.08 6.81 :0.86

6 6.45: 1 .94 6.25:1.37 4.62:1.51

8 4.42:0.79 4.01:1 .94 4.32:1 .37

10 4.27:1.01 7.11:5.46 6.50: 1 .29

12 5.23:1 .94 7.21 :0.86 4.37:1.87

14 3.86:2.87 7.32:1.58 4.22:0.50

16 5.94:1 .51 5.28:1.44 4.93:2.80

18 5.18:1.72 5.64:1 .51 4.42:0.22

24 8.18:2.87 7.98:4.10 5.991029
26 7.87: 1 .65 8.79:0.50 7.52:0.72
28 5.84:0.50 9.50:2.37 8.18:2.37

 

Average : standard deviation of each pouch as a flmction of time.
All values in same column are not significantly difl'erent (p>0.05).

lOl

 

 

 
 
   

9 i + Control
—I— BHT
+ Tocopherol

Hexanal Gone. (11 mot/g)
0|

 

 

 

0 4 8 12 16 20 24 28
Storage (weeks)

 

 

 

Figure 23. The concentration of the hexanal in the model product system packaged in
pouches fabricated from coextruded laminate film structures (23°C, 50 %RH).

The initial concentration of hexanal in the model product at 0 day storage was 3.39 n
mole/g product. As shown up through 28 weeks of storage, the levels of hexanal detected
in the product model system packaged in these test structures were very similar. The
concentration of hexanal in the model product packaged in the respective antioxidant
impregnated laminate fihn pouches was not significantly difi‘erent (p>0.05), as a flmction
of the storage time at 23°C. Lin (1996) studied the effectiveness of or-tocopherol and
BHT impregnated laminate films to retard the oxidation of a packaged oat cereal product
and found a significant difl‘erence (p<0.05) in the level of hexanal in the cereal product

packaged in pouches without antioxidant, as compared to the cereal product packaged in

102

the antioxidant impregnated laminate film pouches after 20 weeks of storage. Storage
conditions for Lin’s study were 23°C and 50%RH.

Based on the literature, lipid oxidation of the model product packaged in non-antioxidant
impregnated laminate fihn pouches was expected to show, at the storage conditions of
test (23°C/50%RH), a slow but constant rate of oxidation, followed by a rapidly
accelerating rate of oxidation, at a certain point during the storage. This would result in
hexanal concentrations many times greater than that observed at the initial stage of the
storage stability study. However, as shown over the 28 week storage period (See Table
21), the level of hexanal in the model product packaged in non-antioxidant impregnated
laminate film pouches showed no statically significant increase, as compared to the initial
concentration of hexanal in the model product. Further, the model product packaged in
the respective antioxidant impregnated laminate film pouches showed hexanal
concentration level of less than 10 n mole/g, as compared to an initial hexanal level of
approximately 3.39 n mole/g. By comparison, the hexanal concentration in the model
product packaged in the non-antioxidant impregnated laminate film pouch and stored at
45°C and 50%RH, was over 70.7 n mole/g after 48hrs, indicating significant levels of the
lipid oxidation (see Table 22). This result suggests that the observed lack of
effectiveness of the antioxidants for the model product packaged in the respective
antioxidant impregnated laminate film pouches was due to the lack of the lipid oxidation
in the model products stored at 23°C and 50%RH, as a flmction of time.

A possible explanation for the lack of observed linoleic acid oxidation for the model
product system stored at 23°C and 50%RH is that in the absence of trace metals, which

can act to promote lipid oxidation and are typically found in a product system such as a

103

cereal product (Tjhio and Karel, 1969), the model product studied has a longer induction
period than the 28 week period over which oxidation was monitored prior to experiencing
oxidation of the fatty acid. Furthermore, the water activity of the product was adjusted to
be at or near the B.E.T. monolayer level, which would provide conditions to minimize
lipid oxidation.

The extent of oxidation of the model product packaged in the respective antioxidant
impregnated test structures and stored at 45°C was also monitored by the formation of

hexanal and the results presented in Table 22.

Table 22. The hexanal concentration of model product system packaged in pouches
fabricated from coextruded laminate structures (45°C, 50 %RH).

 

 

(n mole/gproduct system)
Storage (hrs) Control BHT Tocopherol
o 2.1+0.34' 2.1+0.34' 2.1+0.34°
48 70.76:1.10° 2.3:0.48' 125.88:15.72°
96 4.78:0.37° 27:0.79' 5.65:1.09‘
144 16.4:4.91° 1,910.27: 19.1:0.36‘

 

Average : standard deviation of each pouch as a function of time.
a ' ° Means with different superscripts in same column are significantly difi'erent (p<0.05).

As shown, the effect of the BHT impregnated laminate film structure was much greater
than the efl'ect of the a-tocopherol impregnated laminate fihn structure in inhibiting lipid
oxidation. The concentration of hexanal in the model product packaged in the BHT
impregnated laminate film pouches showed no significant difference (p>0.05) over 6

days storage at 45°C. This can be attributed to the fact that the rate of loss of BHT fi'om

the film into the model product is much faster than the rate of loss of a-tocopherol fi'om

the laminate film structure to the product model system, at 45°C.

 

 

 

 

 

 

 

 

 

200
__ t+catol
g 160 +BHT
f +Tocoplunl
3A
gin»

5

i °°‘
E

40.

o - - ' -
o 50 100 150
mum)

 

 

 

Figure 24. The concentration of the hexanal in the model product system packaged in
pouches fabricated from coextruded laminate structures (45°C, 50 %RH).

The relationship between the extent of lipid oxidation and the respective laminate
package structures that were evaluated in the present study are presented graphically in
Figure 24, where the hexanal concentration in the model product system is plotted a
function of time. The results of hexanal levels, determined in the model product showed
the BHT impregnated laminate package structure to be effective in retaining lipid

oxidation in the model product as a function of time, when stored at 45°C and 50%RH.

105

As shown in Figure 24, the highest values of hexanal were observed in the model product
packaged in the control (flee-antioxidant) and a—tocopherol impregnated pouches, after a
2 day storage period at 45°C/50%RH, which was followed by decreasing hexanal levels
over the remaining stage period. This can be attributed to the partitioning of hexanal into
the package headspace, its sorption and subsequent permeation through the package
structures, or further oxidation at the elevated test temperature.

Henderson, et a1. (1980) observed the effects of heating temperature and time on the
mechanism of linoleate decomposition. Hexanal were found in significantly greater
amounts from the propyl linoleate heated at 70°C than those heated at 180°C, 250°C. The
authors pr0posed that hexanal was the major volatile compound formed during
autoxidation at the low temperature, whereas decadienal levels increased at the high

temperatures due to thermal decomposition.

4-10. Mitative Identification of Hexanal Commund in Model Product Using the
Selected Ion Monitoring GC/MS Procedures.

 

The characteristic ions of hexanal, at selected m/z values were monitored
throughout the GC/MS analysis by operating the mass spectrometer in the SIM (selected
ion monitoring) mode. The selected ion detector monitored only the ion current selected
by the mass analyzer, for the ion current at m/z 29, 44, 56, 72 and 100, respectively. At a
retention time of 8.6 min for the selected ions, the ion currents at m/z 44, 56, 72 were
found to be from the same source. Furthermore, the ratio of the three peaks for the ion
current profiles at 44:m/z, 56:m/z and 72:m/z for the model sample was approximately 1:

0.65: 0.16. Here, m/z = 44. These ratios agreed well with that of the hexanal standard

106

compound (1: 0.63: 0.16) (see Table 23). Therefore, observation of simultaneous peaks
at the three selected ion current profiles, in the expected abundance ratio and at the
expected retention time for the model sample and hexanal standard compotmd, provided

unequivocal evidence that hexanal was present in the model sample.

Table 23. The ratio of the selected ion profiles at m/z 44, 56, and 72 for hexanal.

 

 

Hexanal standard (4 11 mole) Model product (B)
Ratio of m/z (44/44) 1 1
Ratio of m/z (56/44) 0.63 0.65
Ratio of m/z (72/44) 0.16 0.16

 

The external standard calibration curve was obtained from the total peak areas of the ion
fragments (i.e. m/z), selected as being characteristic of hexanal and determined for a
series of standard solutions of known concentration (2, 4, 6, 8 n mole). This is referred to
as the total ion chromatogram. When operating the GC/MS in the selected ion
monitoring mode, the level of hexanal in the model product was calculated using the
calibration factor obtained fiom the external standard calibration curve. The peak
obtained at retention time (8.6 min), represented the total ion chromatogram of the model
sample and was identified as being due to hexanal, based on the following. First, the
retention time for the selected ions, at m/z 29, 44, 56, 72, and 100, was identical to that of
the standard hexanal. The abundance of the peak areas from the selected ion current at
m/z 29, 44, 56, 72, and 100 in the model product sample was used to determine the

hexanal level in the sample by comparison to the response factor fi'om standard hexanal

107

solutions. A comparison of the ion ratios provided further supportive evidence for

hexanal.

SUMVIARY AND CONCLUSIONS

A lipid containing fieeze dried model product was designed to determine the
effectiveness of antioxidant impregnated laminate film pouches to retard lipid oxidation.
All of the fihns contained a heat seal layer (Suryln-EVA) coextruded to a high density
polyethylene (HDPE) layer. The type of antioxidant in the heat seal layer was based on
three test films, which were fihn I containing a-tocopherol, film II containing BHT, and
film III which had no antioxidant. The initial concentrations of a—tocopherol and BHT
incorporated into the test laminate film structure were 0.007 3 %(w/w) and 0.1137
%(w/w), respectively.

The equilibrium sorption isotherm of the model product was determined at three different
temperature conditions (18°C, 28°C, and 38°C), to allow selection of storage conditions
(i.e. Temperature and RH) which could exhibit minimal rates of oxidation. The model
product was described by the Halsey and Henderson (3 8°C, for the model product with
the low linoleic acid level) isotherm models. The monolayer moisture content at each
temperature was 8.57 (18°C), 9.12 (28°C), and 8.38 (38°C), respectively, for the model
product containing high linoleic acid levels and 9.88 (18°C), 9.84 (28°C), and 9.23 (38°C)
(g H20/100g dry product), respectively, for the model product containing low linoleic

acid levels.

108

The loss rate of antioxidant from the laminate pouch film structure was determined as a
function of storage time at 25°C/50%RH, and at 45°C/50%RH. The rate of loss of BHT
fi'om the package film structures was found to be much higher than the rate of loss of or-
tocopherol, at both storage conditions. The storage stability of the packaged model
product, at the respective storage conditions, was used to determine the effectiveness of
the antioxidant impregnated film structures in inhibiting lipid oxidation. Hexanal levels
were determined as a measure of lipid oxidation. Analyses were based on a GC/MS
procedure, operating the mass spectrometer in the selected ion monitoring mode.

The BHT impregnated laminate film was found to be more efl’ective than the oc-
tocopherol impregnated laminate film in inhibiting lipid oxidation at accelerated storage
conditions (i.e. 45°C/50%RH). This was attributed to the fact that the rate of loss of BHT
fiom the package film structure to the model product is much faster than the rate of loss
of a- tocopherol from the laminate film structure to the product model system at 45°C.
However, the level of hexanal in the model product packaged in the respective
antioxidant impregnated laminate fihn pouches showed no statically significant
difi’erences (p>0.05) following storage at 23°C/50%RH. A possible explanation for the
lack of observed linoleic acid oxidation for the model product system stored at 23°C and
50%RH is that in the absence of trace metals, which can act to promote lipid oxidation
and are typically found in a product system such as a cereal product, the model product
studied has a longer induction period than the 28 week period over which oxidation was
monitored, prior to experiencing oxidation of the fatty acid. Furthermore, the water
activity of the product was adjusted to be at or near the B.E.T. monolayer level, which

provides conditions which minimize lipid oxidation.

109

The results of this study demonstrate the ability of impregnating packaging materials with
an antioxidants to extent the shelf-life for lipid-containing food products and to give the

greatest advantages for polymeric packaging systems.

110

FUTURE STUDIES

An antioxidant used in plastic materials for food product contact is expected to be

an important factor, with respect to indirect food additives for restricted migration

regulations. Based on the results of this study, the following future studies are proposed.

1.

Various storage stability studies can be carried out to evaluate the efl‘ect of
relative humidity and temperature on the rate of transfer of antioxidant from
film structures and the oxidation of a model product stored in antioxidant
impregnated laminate film structures.

The study of the partition coefiicient and solubility of antioxidants through the
respective laminate film layers will provide thermodynamic parameters that
measure antioxidant migration fi'om the heat seal layer to HDPE layer and its
partition distribution to the food product phase. In addition, knowledge of the
diffusion coemcient will include information on the rate of loss of antioxidant
from the heat seal layer to the HDPE layer. This study will provide a better
understanding of antioxidant transfer between the respective laminate film
layers and a food product, to provide information on the levels of antioxidant
transfer to a packaged food product.

The potential effect of synergetic antioxidant combinations, or other stabilizer/
antioxidant in laminate film structures can be evaluated to provide a better

understanding relation between levels of two such components and oxidation

111

of the film structures during processing as well as product stability during
storage.

. The applications of developed HPLC/MS or Radioactive labelling techniques
for antioxidant analysis will be helpful to continue monitoring migrated
antioxidants to the fatty food model product packaged in the respective
package structures. This method will be useful for actual commercial food
products, where it is difficult to isolate antioxidants from fatty food product.

. To provide the consumer better quality for food products, a limited number of
antioxidants in the film structures falling within regulatory requirements may
be evaluated by a sensory study of the food products through the storage
stability studies. Future studies are required to develop a mass transfer
relationship between packaging films and food products for storage conditions
where other natural or commercial antioxidants such as the Ascorbic acid,

BHA, and Irganox series are considered.

APPENDICES

APPENDIX A

Table 24. The thickness of laminate structures used to fabricate the pouch structures for
storage studies.

 

 

 

unit = mil‘l
Test Control BHT a-tocopherol
l 2.3 2.25 2.3
2 2.35 2.21 2.2
3 2.38 2.01 2.25
Average 2.34 2.16 2.25

 

I"(mil = 1/1000 inch)

Table 25. The initial moisture content of the model product (A, B).

 

 

 

 

 

Model(A) Dw. Tw » Td Nw MW IMC Average
1 1.3081 4.3424 4.2359 2.9278 0.1065 3.6375 3.4606
2 1.3121 4.297 4.2021 2.89 0.0949 3.2837 10.2502

Model(B) Dw. Tw Td Nw MW IMC Average
1 1.3221 3.0732 3.0169 1.6948 0.0563 3.3219
2 1.3251 3.0678 3.0033 1.6782 0.0645 3.8434 3.6291
3 1.3261 3.0091 2.9494 1.6233 0.0597 3.6777 10.2196
4 1.3391 3.0381 2.9779 1.6388 0.0602 3.6734

 

Dw : Dish Weight (g)

Tw : Total Weight with Dish (g)

T d : Total Weight with Dish after drying (g)

Nw : Net Weight after drying (g)

Mw : Moisture Weight (g)

IMC : % Initial moisture content on a dry basis (gI-I20/g dry weight product)

112

113

Table 26. The conditions of relative humidity for the salt solutions according to each
temperature.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The model product (A)
Salt solution Relative humidig’ (%)
At 18 °C At 28 °C At 38 °C
Temperature Temerature Temperature
Lithium Chloride 9.6 14 11.6
Potassium Acetate 21.4 25 20
Magnesium 35.5 38 34.5
Chloride
Potassium 46 45.5 44.5
Carbonate
Magnesium nitrate 56 55 53.5
Sodium Nitrate 63
Sodium Chloride 74.5 73.5 75.5
Ammonium Sulfate 80.5 81.5 81.5
Potassium Nitrate 90.5
The model product (B)
Salt solution Relative humidity (%)
At 18 °C At 28 °C At 38 °C
Temperature Temperature Temperature
Lithium Chloride 9.4 11 11.4
Potassium Acetate 21.5 23.5 19.8
Magnesium 35.0 34.5 34.5
Chloride
Potassium 46.0 45.5 47.0
Carbonate
Magnesium nitrate 58.0 54.5 51.0
Sodirun Chloride 75.6 75.0 75.5

 

Ammonium Sulfate 81.5 80.5 81.5

114

Table 27. Experimental data for equilibrium sorption isotherm of the model product (A)
at18°C(64°F). Unit=g(weight)

 

Average obtained moisture content as a function of storage time

 

Aw 0 days 5 days 10 days 15 days 20 days 25 days
0.096 0.0000 0.0850 0.0876 0.0904 0.0916 0.0910
0.214 0.0000 0.1316 0.1332 0.1346 0.1348 0.1346
0.355 0.0000 0.2310 0.2309 0.2310 0.2309 0.2318

0.46 0.0000 0.2410 0.2410 0.241 1 0.2384 0.2410

0.56 0.0000 0.3100 0.3157 0.3158 0.3148 0.3125
0.745 0.0000 0.4883 0.4997 0.5006 0.5032 0.5044
0.805 0.0000 0.5896 0.6090 0.6177 0.6318 0.6449

 

‘Average of three replicate sample data

Table 28. Experimental data for equilibrium sorption isotherm of the model product (A)
at 28 °C ( 84 °F ). . Unit = g (weight)

 

Average obtained moisture content as a function of storage time

 

Aw 0 days 5 days 10 days 15 days 20 days 25 days
0.14 0.0000 0.0762 0.0805 0.0790 0.0966 0.0982
0.25 0.0000 0.0885 0.0941 0.0999 0.1432 0.1345
0.38 0.0000 0.1452 0.1495 0.1507 0.1867 0.2036
0.455 0.0000 0.1993 0.2033 0.2060 0.2453 0.2600
0.55 0.0000 0.2560 0.2598 0.2672 0.3074 0.3208
0.63 0.0000 0.3996 0.3885 0.3800 0.3778 0.3728
0.735 0.0000 0.4641 0.5007 0.5774 0.5657 0.5543
0.815 0.0000 0.5581 0.5941 0.6424 0.6250 0.6149

 

‘Average of three replicate sample data

Table 29. Experimental data for equilibrium sorption isotherm of the model product (A)
at 38 °C( 100 °F ). Unit = g (weight)

 

Average obtained moisture content as a function of storage time

 

Aw 0 days 5 days 10 (Lays 15 days 20 days 25 days
0.1 16 0.0000 0.0786 0.1203 0.1263 0.1 188 0.1 149

0.2 0.0000 0.1 127 0.1216 0.1505 0.1479 0.1425
0.345 0.0000 0.1445 0.1460 0.1758 0.1816 0.1779
0.445 0.0000 0.2046 0.2421 0.2454 0.2341 0.2403
0.535 0.0000 0.2591 0.2909 0.2994 0.2917 0.2866
0.755 0.0000 0.5498 0.5450 0.5413 0.5277 0.5225
0.815 0.0000 0.661 1 0.6642 0.6562 0.6431 0.6425
0.905 0.0000 1 .0498 1 .0769 1 .0830 l .0743 1.0691

 

‘Average of three replicate sample data

115

Table 30. Experimental data for equilibrium sorption isotherm of the model product (B)

 

 

at 18°C (64 0F). Unit= g (weight)
Average obtained moisture content as a function of storage time

Aw 0 days 5 days 10 days 15 days 20 days 25 days
0.094 0.0000 0.0779 0.0862 0.0891 0.0858 0.0888
0.215 0.0000 0.1510 0.1536 0.1563 0.1485 0.1510
0.35 0.0000 0.1945 0.2090 0.2134 0.2094 0.2108
0.46 0.0000 0.2762 0.2816 0.2850 0.2820 0.2832
0.58 0.0000 0.3408 0.3479 0.3523 0.3486 0.3486
0.756 0.0000 0.7031 0.7073 0.7144 0.7080 0.7092
0.815 0.0000 0.7932 0.8146 0.8349 0.8252 0.8304

 

‘Average of three replicate sample data

Table 31. Experimental data for equilibrium sorption isotherm of the model product (B)

 

 

at 28 °C ( 84 °F ). Unit = g (weight)
Average obtained moisture content as a fimction of storage time

Aw 0 days 5 days 10 days 15 days 20 days 25 days
0.1 1 0.0000 0.0559 0.0652 0.0632 0.0638 0.0597
0.235 0.0000 0.1014 0.1086 0.1095 0.1097 0.1 101
0.345 0.0000 0.1819 0.1853 0.1872 0.1868 0.1873
0.455 0.0000 0.2454 0.2465 0.2484 0.2480 0.2496
0.545 0.0000 0.2927 0.2921 0.2945 0.2912 0.2935
0.75 0.0000 0.5029 0.5170 0.5236 0.5195 0.5208
0.805 0.0000 0.7128 0.7095 0.7084 0.7070 0.7153

 

‘Average of three replicate sample data

Table 32. Experimental data for equilibrium sorption isotherm of the model product (B)
at 38°C( 100 °F ).

Unit = 3 (weight)

 

Average obtained moisture content as a ftmction of storage time

 

Aw 0 days 5 days 10 days 15 days 20 days 25 days
0.114 0.0000 0.0421 0.0306 0.0314 0.0308 0.0353
0.198 0.0000 0.1000 0.1045 0.1075 0.1063 0.1152
0.345 0.0000 0.1500 0.1506 0.1514 0.1496 0.1505

0.47 0.0000 0.2282 0.2286 0.2345 0.2320 0.2352

0.51 0.0000 0.2665 0.2664 0.2684 0.2666 0.2669
0.755 0.0000 0.5189 0.5127 0.5177 0.4863 0.5030
0.815 0.0000 0.6321 0.6325 0.6514 0.6363 0.6228‘

 

l'Aver'age of three replicate sample data

116

APPENDIX B

Isotherm Model using the Mathematical Expression

The equilibrium sorption isotherm of the model product describes the relationship
between the product’s moisture content at a given water activity. From an initial
examination of a series of equations developed to describe equilibrium sorption

isotherms, the following models or expressions were selected.

. Henderson Equation

. Chen Equation

. Halsey Equation

. Guggenhem-Anderson-de Boer (G.A.B.) Equation
. Brunauer, Emmett and Teller (B.E.T.) Equation

These equations represented a synopsis of the characteristic geometric configurations of
sorption isotherm shapes that have been transformed to convenient linearized forms.
The linearized equations are as follows:

Table 33. Isotherm Mathematical model (Berends, 1993).

 

 

Model Equation
1. Henderson Ln [-ln(l-aw)] = nln Meq + In K
2. Chen Ln (-ln aw) = K - aMeq
3. Halsey Aw = exp ( -a/Meq )
4. G.A.B. Meq = C(K) (aw) (Wm) / ( l-Kaw ) (l-Kaw + Ckaw)
5. B.E.T. Aw/Meq (l-aw) = leC + aw ( C-l/MmC )

 

where; aw = water activity,
Meq = equilibrium moisture content,
Mm = monolayer moisture content,
n, K, C = constants

117

For the freeze-dried product model, the best fit of the experimental equilibrium
sorption data to a linearized form of the isotherm expressions was determined by linear
regression analysis. The correlation coefficient was determined and was applied to the
degree of fit among the five equations. The estimation for the parameters in the
respective expressions was calculated graphically using slope and intercept values.

By substituting these parameters back into the isotherm equations, calculated sorption
data was obtained (see Tables 34-39). A comparison of the experimental and calculated
isotherm data was given by a value of sums of squares, which provided the basis for
selection of the best fit model. 1

For all cases, the mathematical equation that best described the sorption isotherm of the
model products was the Halsey equation, except for model product B at 38°C, which was
best described by the Henderson equation. The linearized plots of the best-fit model are

presented graphically in Figures 25-30, respectively.

118

Table 34. The data of model product (A) for Halsey equation at 18°C (64°F).

 

 

 

 

 

Halsey Equation model
X axis y axis Ln(-ln(aw))=-l.601 11n(Meq)+4.1983
Aw Ln(Meq) ln(-ln(aw)) Actual Mathematical % difference
0.096 2.0981 0.8516 8.1504 8.0869 -0.7851
0.214 2.3867 0.4329 10.8778 10.5038 -3.5608
0.355 2.5838 0.0350 13.2473 13.4673 1.6335
0.46 2.7422 -0.2529 15.5210 16.1207 3.7199
0.56 2.9484 —0.5450 19.0749 19.3473 1.4077
0.745 3.4064 -l.2229 30.1573 29.5455 -2.0707
0.805 3.5816 -1.5283 35.9295 35.7532 -0.4930
1
0.5 l»
0 l
E» 8.5
l
s
-1 ..
-1.5 ..
-2 : : a 4 : r i :
2.00 2.20 2.40 2.80 2.80 3.00 3.20 3.40 3.60 3.80
MM)

 

 

 

 

 

 

Figure 25. The plot of linear regression of Halsey equation for model product (A) at

18°C.

 

Table 35. The data of model product (A) for Halsey equation at 28°C (84°F).

 

 

 

 

 

 

 

 

 

Halsey Equation model
X axis y axis Ln(-ln(aw))-—1.6719ln(Meq)-l-4.392
Aw Ln(Meq) ln(-ln(aw)) Actual Mathematical % difference
0.14 2.1447 0.6761 8.5393 9.2312 7.4950
0.25 2.3550 0.3266 10.5386 11.3769 7.3684
0.38 2.6047 -0.0330 13.5267 14.1069 4.1126
0.455 2.7683 -0.2390 15.9322 15.9566 0.1526
0.55 2.8976 -0.5144 18.1299 18.8149 3.6406
0.63 3.0894 -0.7721 21.9630 21.9501 -0.0590
0.735 3.3790 -l.1780 29.3419 27.9819 -4.8604
0.815 3.5333 -1.5869 34.2381 35.7334 4.1847
1
0.5 ..
o ..
E .05
...
E
-1 .1
o
-1.5 t .
-2 : ; t t 4r i : :
2.00 2.20 2.40 2.60 2.80 3.00 3.20 3.40 3.60 3.80
MM”)

 

 

 

Figure 26. The plot of linear regression of Halsey equation for model product (A) at
28°C.

Table 36. The data of model product (A) for Halsey equation at 38°C (100°F).

120

 

 

 

 

 

Halsey Equation model
X axis y axis Ln(-ln(aw))=-1.6886ln(Meq)+4.3592
Aw Ln(Meq) ln(-ln(aw)) Actual Mathematical % difl‘erence
0.116 2.1724 0.7674 9.2381 8.3904 -10. 1031
0.2 2.3303 0.4759 10.7463 9.9714 -7.7710
0.345 2.5370 0.0622 13.1163 12.7393 -2.9593
0.445 2.7016 -0.2111 15.3889 14.9778 -2.7443
0.535 2.7892 -0.4692 16.7582 17.4515 3.9728
0.755 3.3240 -1 .2693 28.3094 28.0283 -l.0029
0.815 3.4744 -1.5869 32.8359 33.8282 2.9335
0.905 4.0090 -2.3044 55.5797 51.7393 -7.4226
1
0.5 ..
o ..
-O.5 1 9
...
§
-1.5 h
.2 ,.
o
.3 . : : t
200 250 3.!!! 3.50 4.00 4.5)
W

 

 

 

 

 

 

Figure 27. The plot of linear regression of Halsey equation for model product (A) at

38°C.

 

121

Table 37. The data of model product (B) for Halsey equation at 18°C (64°F)

 

 

 

 

 

 

 

 

 

Halsey Equation model
X axis y axis Ln(-1n(aw))=-1.4356ln(Meq)+3.8456
Aw Ln(Meq) ln(-ln(aw)) Actual Mathematical % difference
0.094 2.0852 0.8606 8.0458 7.9989 -0.5862
0.215 2.3672 0.4299 10.6671 10.7971 1.2043
0.35 2.6645 0.0486 14.361 1 14.0817 -l.9846
0.46 2.8615 -0.2529 17.4871 17.3730 -0.6565
0.58 3.0660 -0.6075 21.4555 22.2399 3.5272
0.756 3.5924 -1.2740 36.3225 35.3807 -2.6620
0.815 3.7738 -1.5869 43.5431 43.9962 1.0298
1
0.5 ~
0 .1
g -o 5 1
l
s
.1 ..
-1.5 it
-2 1 ; , : : : : + t
2.00 2.20 2.40 2.60 2.80 3.00 3.20 3.40 3.60 3.80 4.00
ln(MOQ)

 

 

Figure 28. The plot of linear regression of Halsey equation for model product (B) at

18°C.

 

122

Table 38. The data of model product (B) for Halsey equation at 28°C (84°F).

 

 

Halsey Equation model
X axis y axis Ln(-ln(aw))=-1.3353ln(Meq)+3.3906
Aw Ln(Meq) ln(-ln(aw)) Actual Mathematical % difference
0.1 1 1.9016 0.7918 6.6965 7.0025 4.3698
0.235 2.2306 0.3703 9.3052 9.6012 3.0833
0.345 2.5393 0.0622 12.671 1 12.0927 -4.7837
0.455 2.8022 -0.2390 16.4800 15.1523 -8.7624
0.545 2.9351 —0.4993 18.8241 18.4140 -2.2274
0.75 3.4052 -1.2459 30.1 192 32.2092 6.4889
0.805 3.6742 -1.5283 39.4153 39.7938 0.9510

 

 

 

 

In(-In(aw))
6
0|

0.5 1-

 

 

 

 

220

240

260 2“)

son

320

3.40 sec

 

3.5)

 

 

Figure 29. The plot of linear regression of Halsey equation for model product (B) at

28°C.

Table 39. The data of model product (B) for Henderson equation at 38°C (100°F).

 

 

 

 

 

 

 

 

 

 

Henderson Equation model
X axis y axis Ln(-ln(1-aw))=1.4187ln(Meq)-4.4876
Aw Ln(Meq) ln(-ln( l-aw)) Actual Mathematical % difi‘erence
0.114 1.6930 -2.1 1 17 5.4358 5.3374 -1.8441
0.198 2.2502 -1.5112 9.4891 8.1497 -16.4353
0.345 2.4529 -0.8601 1 1.6215 12.8960 9.8831
0.47 2.7561 -0.4543 15.7390 17.1661 8.3138
0.51 2.8527 -0.3378 17.3351 18.6358 6.9793
0.755 3.4437 0.341 1 31.3016 30.0725 -4.0873
0.815 3.5854 0.5232 36.0693 34.1908 -5.4940
1
0.5 1
o ..
if -o.5
E
E -1 ..
-15 »
.2 ..
-2.5 t t t : r : . j
2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8
MM!!!)

 

 

 

Figure 30. The plot of linear regression of Henderson equation for model product (B) at
38°C.

124

Brunauer Emmett and Teller (B.E.T.) Monolayer V119;

 

Estimation of the B.E.T. monolayer value has been used to determine the water
activity value which could provide the maximum shelf life for the dried model product.
Also, the Aw levels generated to obtain the monolayer values fiom the B.E.T. equation
expression are derived from the actual isotherm data in the water activity range below
Aw=0.35 (C. van den berg, 1985). Below the B.E.T. monolayer value, the rate of lipid
oxidation generally increases. Above the monolayer value, the rate of lipid oxidation is
again accelerated, according to increasing water activity (Leung, 1987).

The B.E.T. equation shows a linear relationship between Aw and the equilibrium
moisture content of the model product. From the equilibrium sorption isotherm values, a

straight line relationship was obtained fiom a plot of Aw/[m (1-Aw)] against Aw.

Using the slope and y-intercept values for the B.E.T. regression equation, the

monolayer moisture content was determined by substitution into the following equation:

Monolayer value .= l / ( y-intercept + slope ) ---- (l 6)

125

The monolayer values calculated for the freeze-dried product model at 18, 28, and
38°C, respectively are summarized in Tables 40, 41, and 42. The corresponding water
activity values were also estimated and compared to the water activity values determined
from the Halsey equation and the Henderson equation (38°C, model product B), which
best described the product sorption isotherm data.

Table 40. The monolayer values of high (A) and low (B) linoleic acid containing freeze-
dried model product according to Sorption Isotherm at 18°C.

At 18°C mane

 

 

Model Product (A) Model Product (B)
Moisture Content
(gHzO/100g) of the B.E.T. 8.57 9.88
monolayer value
The Aw of the sorption isotherm
at the B.E.T. monolayer value 0.1 1 0.14

 

Table 41 . The monolayer values of high (A) and low (B) linoleic acid containing fi'eeze-
dried model product according to Sorption Isotherm at 28°C.

 

 

At 28°C temperature
Model Product (A) Model Product (B)
Moisture Content
(gH20/100g) of the B.E.T. 9.04 9.84
monolayer value
The Aw of the sorption isotherm
at the B.E.T. monolayer value 0.16 0.25

 

Table 42. The monolayer values of high (A) and low (B) linoleic acid containing freeze-
dried model product according to Sorption Isotherm at 38°C.

 

 

At 38°C temperature
Model WA) Model Product fl)
Moisture Content
(g1-120/100g) of the B.E.T. 8.38 9.23
monolayer value

The Aw of the sorption isotherm
at the B.E.T. monolayer value 0.08 0.18

 

126

The monolayer values for model product (A), at each temperature, were 8.57 (18°C), 9.04
(28°C), and 8.38 (38°C) gH20/100g dry product, respectively. The monolayer values for
model product (B), at each temperature, were calculated to be 9.88 (18°C), 9.84 (28°C),
and 9.23 (38°C) gI-120/100g dry product, respectively. The Aw levels corresponding to
the monolayer value of the model product (A) were 0.11 (18°C), 0.16 (28°C), and 0.08
(38°C), respectively. The Aw levels corresponding to the monolayer value of the model
product (B) were calculated to be 0.14 (18°C), 0.25 (28°C), and 0.18 (38°C), respectively.
The regression plots of the B.E.T. equation for monolayer values, using the data for the
low moisture range of the sorption isotherm at each temperature, are shown in Figures

31-36, and the numerical data summarized in Tables 43-48, respectively.

127

Table 43. The data of model product (A) for B.E.T. monolayer value at 18°C (64°F).

 

xaxis

 

 

 

 

 

 

 

 

 

y axis
Aw (aw) aw/(Meq(1-aw))
0.096 0.096 0.0130
0.214 0.214 0.0250
0.355 0.352 0.0416
0.46 0.46 0.0549
0.06
0.05 1
A 0.04 it
E 0.03 .
E 0.02
0.01
o : ; : : 4. : i a i
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(W)

 

Figure 31. The linear regression plot of B.E.T. equation for the monolayer value of model
product (A) at 18°C.

 

128

Table 44. The data of model product (A) for B.E.T. monolayer value at 28°C (84°F).

 

 

 

 

 

 

 

 

 

x axis y axis
Aw (aw) aw/(Meq(1-aw))
0.14 0.14 0.0191
0.25 0.25 0.0316
0.38 0.38 0.0453
0.455 0.455 0.0524
0.06
0.055 4)
0.05 4
0.045 1»
i 0.04 «»
g 0.035
‘2' 0.03 <
0.025 1»
0.02 1.
0.015 «L
0.01 c t : t t 4 :
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(W)

 

 

 

Figure 32. The linear regression plot of B.E.T. equation for the monolayer value of model
product (A) at 28°C.

129

Table 45. The data of model product (A) for B.E.T. monolayer value at 38°C (100°F).

 

 

 

 

 

 

 

 

 

x axis y axis
Aw (aw) aw/(Meq(1-aw))
0.116 0.116 0.0150
0.2 0.2 0.0243
0.345 0.345 0.0417
0.445 0.445 0.0538
0.06
0.05 «L
i 0.04 J»
E 0.03 «1
0.02 4
0.01 4. Y : t : : :
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(W)

 

 

 

Figure 33. The linear regression plot of B.E.T. equation for the monolayer value of model
product (A) at 38°C.

130

Table 46. The data of model product (B) for B.E.T. monolayer value at 18°C (64°F).

 

 

 

x axis y axis

Aw (aw) aw/(Meq(1-aw))
0.094 0.096 0.0132
0.215 0.214 0.0255
0.35 0.352 0.0383
0.46 0.46 0.0487

 

 

0.08

 

0.05 1r

W/(MU-aw»
9 .o
8 2

.o
8

0.01 v

 

 

 

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
( 8W)

 

 

 

Figure 34. The linear regression plot of B.E.T. equation for the monolayer value of model
product (B) at 18°C.

131

Table 47. The data of model product (B) for B.E.T. monolayer value at 28°C (84°F).

 

 

 

 

 

 

 

 

 

x axis y axis
Aw (aw) aw/(Meq(1-aw))
0.11 0.11 0.0185
0.235 0.235 0.0330
0.345 0.345 0.0416
0.455 0.455 0.0507
0(55
005 ~»
0045 [
0,“ 1r
E has 1
E“ O
z cos
i
0
0015 ..
0.01 4 i 4 % : ; ;
01 015 02 025 03 0:5 04 0.45 0.5
(8”)

 

 

 

Figure 35. The linear regression plot of B.E.T. equation for the monolayer value of model
product (B) at 28°C.

132

Table 48. The data of model product (B) for B.E.T. monolayer value at 38°C (100°F).

 

 

 

 

 

 

 

 

 

x axis y axis
Aw (aw) aw/(Meq(1-aw))
0.114 0.114 0.0237
0.198 0.198 0.0260
0.345 0.345 0.0453
0.47 0.47 0.0563
0.06
0.05 1
O
f 0.04 +-
‘fi 0.03 «
O
O
0.02 ‘r
0.01 : + 4 : : : t
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(W)

 

 

 

Figure 36. The linear regression plot of B.E.T. equation for the monolayer value of model
product (B) at 38°C.

133

Statistical Selection of Best Fit Model

Correlation coefficient values for the x-y components of the linearized forms of
the respective isotherm expressions were calculated fiom the isotherm data using the
Minitab Statistic Program (Minitab Inc., PA). Results for the respective isotherm models
are summarized in Tables 48-49, and formed the basis for selecting the best fit isotherm
equation.

The best-fit model was based on the best correlation between the components of the x-y
axis, and the lowest sums of squares at each temperature condition. Here, the Halsey
model showed a consistently low sum of squares and a correlation coeflicient very close
to 1 at each temperature, and was selected to describe the equilibrium sorption isotherm
for the product model. As indicated above, for model product B at 38°C, the Henderson
model was found to give the best fit.

Table 49. The Correlation Coefficient and Sums of Squares fiom the isotherm data and

the respective isotherm models for product model A at different temperatures (18°C,
28°C, 38°C).

 

 

Model Sum of Squares for Error Correlation Coemcient (p)
(A) 18°C 28°C 38°C 18°C 28°C 38°C
Henderson 26.52 16.89 1 13.7 0.973 0.981 0.950
Chen 34.95 18.81 187.6 0.974 0.983 0.949
Halsey 0.89 4.92 9.8 0.999 0.996 0.997
G.A.B. 70.20 49.80 199.50 0.919 0.934 0.870
B.E.T. 35.49 97.64 1426.2 0.973 0.991 0.671

 

Table 50. The Correlation Coemcient and Sums of Squares from the isotherm data and
the respective isotherm models for product model B at different temperatures (18°C,
28°C, 38°C).

 

 

Model Sum of Squares for Error Correlation Coemcient (p)
(B) 18°C 28°C 38°C 18°C 28°C 38°C
Henderson 36.83 21.40 8.97 0.977 0.984 0.993
Chen - 61 .1 37.73 17.90 0.973 0.978 0.989
Halsey 1.8 5.7 15.22 0.999 0.997 0.992
G.A.B. 91.26 51.42 33.56 0.932 0.955 0.971

B.E.T. 21.3 15.82 33.11 0.990 0.991 0.981

134

APPENDIX C

Table 51. The mathematical model for equilibrium moisture content fi'om the
experimental data of the model product A at 18 °C ( 64 °F ).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Henderson Equation model
x axis y axis ln(-ln(1-aw))=1.77751n(Meq)-5.6458
Aw ln(Meq) ln(-ln(1-aw)) Actual Mathematical % difference
0.096 2.098067006 -2.293368511 8.1504 6.5932 -23.6184
0.214 2.386724015 -1.423794851 10.8778 10.7537 -1.1541
0.355 2.583793758 0824384151 13.2473 15.0664 12.0742
0.46 2.742193946 0484206187 15.5210 18.2443 14.9266
0.56 2.948373335 0197255858 19.0749 21.4406 1 1.0336
0.745 3.406427017 0.312246677 30.1573 28.5577 -5.6015
0.805 3.581558685 0.491493387 35.9295 31.5877 -13.7453
Chen Equation model
X axis y axis ln(-lnaw)-—0.0807(Meq)+1.2144
Aw (Meg) ln(-lnaw) Actual Mathematical % difl‘erence
0.096 8.1504 0.851605891 8.1504 4.4956 -81.2977
0.214 10.8778 0.4329371 16 10.8778 9.6836 -12.3327
0.355 13.2473 0.035017169 13.2473 14.6144 9.3545
0.46 15.521 0252921561 15.5210 18.1824 14.6373
0.56 19.0749 0545040164 19.0749 21.8022 12.5094
0.745 30.1573 -1.222914197 30.1573 30.2022 0.1485
0.805 35.9295 -1.52825892 35.9295 33.9859 -5.7190
G.A.B Equation model
x axis y axis (aw/Meq)=0.013aw+0.0175
Aw (aw) (aw/Meq) Actual Mathematical % difl'erence
0.096 0.096 0.01 1778563 8.1504 5.1205 ~59.1705
0.214 0.214 0.019673096 10.8778 10.5512 -3.0951
0.355 0.352 0.026797914 13.2473 16.0525 17.4749
0.46 0.46 0.029637266 15.5210 19.5911 20.7754
0.56 0.559 0.029357952 19.0749 22.5989 15.5936
0.745 0.745 0.024703803 30.1573 27.4048 -10.0438
0.805 0.805 0.022404988 35.9295 28.7860 -24.8160
B.E.T. Equation model
x axis y axis Aw/(Meq(1-aw))=01394aw00054
Aw (aw) aw/(Meq(1-aw)) Actual Mathematical % difi‘erence
0.096 0.096 0013029384 8.1504 13.3036 38.7354
0.214 0.214 0025029384 10.8778 1 1.1440 2.3883
0.355 0.352 0.041547153 13.2473 12.4841 01 132
0.46 0.46 0054883825 15.5210 14.5060 -6.9969
0.56 0.559 0066722618 19.0749 17.5152 -8.9046
0.745 0.745 0096877659 30.1573 29.6748 -1.6261
0.805 0.805 0.1 14897372 35.9295 38.6475 7.0327

 

 

135

Table 52. The mathematical model for equilibrium moisture content from the
experimental data of the model product A at 28 °C ( 84 °F ).

 

 

 

 

 

 

 

 

 

Henderson Equation model
x axis y axis ln(-ln(1-aw))=1.7523ln(Meq)-5.5129
Aw ln -ln 1—aw Actual Mathematical % difi‘erence
0.14 2.144679 -1.891649046 8.5393 7.8977 -8.1240
0.25 2.3550447 -1.245899324 10.5386 11.4168 7.6921
0.38 2.6046655 0738069652 13.5267 15.2548 11.3281
0.455 2.7683422 0499276762 15.9322 17.4819 8.8647
0.55 2.8975625 0225010673 18.1299 20.4439 11.3188
0.63 3.0893592 0005764308 21 .9630 23.1687 5.2042
0.735 3.3790165 0283693217 29.3419 27.3302 -7.3608
0.815 3.5333391 0523188559 34.2381 31.3328 -9.2723
Chen Equation model
X axis y axis 1n(-1naw)——0086(Meq)+l .2497
Aw (Meq) lng-lnaw) Actual Mathematical % difference
0.14 8.5393 0676058424 8.5393 6.6703 -28.0207
0.25 10.5386 032663426 10.5386 10.7333 1.8142
0.38 13.5267 0032953009 13.5267 14.9146 9.3055
0.455 15.9322 0238945421 15.9322 17.3098 7.9587
0.55 18.1299 0514437136 18.1299 20.5132 11.6185
0.63 21.963 0772113638 21 .9630 23.5095 6.5780
0.735 29.3419 -1.178029658 29.3419 28.2294 -3.9409
0.815 34.2381 -1.586858919 34.2381 32.9832 -3.8045
G.A.B Equation model
x axis y axis (aw/Meq)=0.0117awt-00191
Aw (aw) (aw/Mg) Actual Mathematical % difi'erence
0.14 0.14 0016394786 8.5393 6.7509 -26.4914
0.25 0.25 0023722316 10.5386 11.3507 7.1549
0.38 0.38 0028092587 13.5267 16.1386 16.1843
0.455 0.455 0028558517 15.9322 18.6296 14.4791
0.55 0.55 0030336626 18.1299 21.5391 15.8278
0.63 0.63 0.028684606 21 .9630 23.7996 7.7171
0.735 0.735 0025049503 29.3419 26.5348 ~105790
0.815 0.815 0023803891 34.2381 28.4612 -20.2976

 

 

 

B.E.T. Equation model

 

 

x axis y axis Aw/(Meq(l-aw))=01526aw-0.01 18

Aw (aw) aw/(Megg l-awn Actual Mathematical % difference
0.14 0.14 0019063705 8.5393 17.0212 49.8314
0.25 0.25 0.031629755 10.5386 12.6502 16.6924
0.38 0.38 0045310625 13.5267 13.2698 -1.9364
0.455 0.455 0052400948 15.9322 14.4858 -9.9847
0.55 0.55 0067414725 18.1299 16.9447 -6.9944
0.63 0.63 0077525962 21 .9630 20.1890 -8.7868
0.735 0.735 0094526425 29.3419 27.6361 -6.1724
0.815 0.815 0128669681 34.2381 39.1352 12.5132

 

 

136

Table 53. The mathematical model for equilibrium moisture content from the
experimental data of the model product A at 38 °C ( 100 °F ).

 

 

 

 

 

 

 

 

 

 

Henderson Equation model
x axis y axis ln(-ln(1-aw))=1.54251n(Meq)—4.9409
Aw ln(Meq) ln(-ln(1-aw)) Actual Mathematical % difl‘erence
0.1 16 2.1723853 -2.093149335 9.2381 6.3356 45.8109
0.2 2.330317 -1.499939987 10.7463 9.3070 -15.4648
0.345 2.5369613 086009935 13.1 163 14.0916 6.9209
0.445 2.7016431 052969051 15.3889 17.4577 1 1.8505
0.535 2.7891815 0266941489 16.7582 20.6997 19.0416
0.755 3.3239743 0.341 102265 28.3094 30.7015 7.7915
0.815 3.4743828 0.523188559 32.8359 34.5483 4.9567
0.905 4.009019 0856064345 55.5797 42.8695 -29.6485
Chen Equation model
X axis y axis ln(-1naw)=-0.0646(Meq)+0.8703
Aw (Meq) ln(-1naw) Actual Mathematical % difierence
0.1 16 8.7792 0.767403218 9.2381 1 .5928 -479.97 82
02 10.2812 0475884995 ' 10.7463 6.1055 -76.0106
0.345 12.6412 006223355 13.1 163 12.5088 4.8568
0.445 14.9042 0211 11494 15.3889 16.7402 8.0723
0.535 16.2677 0469222283 16.7582 20.7356 19.1819
0.755 27.7705 -1.269267061 28.3094 33.1202 14.5253
0.815 32.2779 -1.586858919 32.8359 38.0365 13.6728
0.905 55.0928 -2.304383356 55.5797 49.1437 -13.0963
G.A.B Equation model
x axis y axis (aw/Meq)=00053aw+00212
Aw (aw) (aw/Meq) Actual Mathematical % difierence
0.1 16 0.1 16 0013213049 9.2381 5.3175 -73.7298
0.2 0.2 0.019452982 10.7463 ' 8.9847 -19.6065
0.345 0.345 0027291713 13.1 163 14.9814 12.4496
0.445 0.445 0029857356 15.3889 18.8891 18.5307
0.535 0.535 0032887255 16.7582 22.2587 24.7121
0.755 0.755 0.0271 87123 28.3094 29.9585 5.5047
0.815 0815 0025249474 32.8359 31.9364 -2.8165
0.905 0.905 0016426829 55.5797 34.8124 -59.6550

 

 

B.E.T. Equation model

 

 

x axis y axis Aw/(Meq(1-aw))=01895aw-00193

Aw (aw) aw/(Meq(1-avD) Actual Mathematical % difi'erence
0.1 16 0.1 16 0014946888 9.2381 48.9268 81.1 186

0.2 0.2 0024316228 10.7463 13.4409 20.0474
0.345 0.345 0041666737 13.1 163 1 1.431 1 -14.7420
0.445 0.445 0053797037 15.3889 12.3302 -24.8062
0.535 0.535 007072528 16.7582 14.0168 49.5572
0.755 0.755 0.110967849 28.3094 24.8976 -13.7036
0.815 0.815 0136483644 32.8359 32.5982 0.7290
0.905 0.905 0172913989 55.5797 62.5918 1 1.2029

 

 

Table 54. The mathematical model for equilibrium moisture content from the
experimental data of the model product B at 18 °C ( 64 °F ).

137

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Henderson Equation model
x axis y axis Ln(-ln(1-aw))=1.5833ln(Meq)-5.236
Aw ln(Meq) ln(-an -alv)) Actual Mathematical % differmce
0.094 2.085149443 -2.315508512 8.045794 6.3253 -27.199899
0.215 2.367161688 -1.418521889 10.66707 1 1.1461 4.29770234
0.35 2.664524768 0842150991 14.361 12 16.0406 104699791
0.46 2.861461895 0484206187 17.48707 20.1096 13.0410678
0.58 3.065979049 0142139113 21.45546 24.9592 14.0379253
0.756 3.592437467 0344005968 36.3225 33.9296 -7.0524779
0.815 3.773752144 0.523188559 43.54314 37.9952 -14.601799
Chen Equation model
X axis y axis ln(-lnaw)=-00644(Meq)+1.0565
Aw (Meq) ln(-lnaw) Actual Mathematical % difference
0.094 8.045793782 0860549876 8.045794 3.0427 -164.42908
0.215 1066707279 0429908747 10.66707 9.7297 06343884
0.35 14.361 12306 0048620745 14.361 12 15.6503 8.23738857
0.46 17.48707253 0252921561 17.48707 20.3326 13.9950414
0.58 21.45545766 0607470205 21.45546 25.8380 16.961766
0.756 36.32250302 -1.273987974 36.3225 36.1877 03725066
0.815 43.54313884 -1.586858919 43.54314 41.0459 00838966
G.A.B Equation model
x axis y axis (aw/Meq)=0.0072aw+0018
Aw @w) (aw/Meq) Actual Mathematical % difference
0.094 0.094 0.01 1683123 8.045794 5.0330 -59.861363
0.215 0.215 0.020155483 10.66707 10.9986 3.01398187
0.35 0.35 0024371353 14.36112 17.0565 15.8027871
0.46 0.46 0026305146 17.48707 21.5841 18.981632?
0.58 0.58 0027032749 21 .45546 26.1544 17.9661674
0.756 0.756 0020813544 36.3225 ' 32.2482 -12.634352
0.815 0.815 0018717071 43.54314 34.1461 -27.519956
B.E.T. Equation model
x axis y axis Aw/(Meq(l -aw))=0.1204aw-00019
Aw (aw) aw/(Meq(1-aw)) Actual Mathematical % difi‘erence
0.094 0.094 0.012895279 8.045794 1 1.0169 26.9686243
0.215 0.215 0025675774 10.66707 1 1.4186 6.58119986
0.35 0.35 0037494389 14.361 12 13.3813 4.3227242
0.46 0.46 0048713234 17.48707 15.9272 07935733
0.58 0.58 0064363688 21.45546 20.3285 -5.5439833
0.756 0.756 0085301408 36.3225 34.7652 4.4794007
0.815 0.815 0.101173354 43.54314 45.7819 4.88997738

 

 

Table 55. The mathematical model for equilibrium moisture content from the
experimental data of the model product B at 28 °C ( 84 °F ).

138

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Henderson Equation model
x axis y axis ln(-ln(1-aw))=1 .4626ln(Meq)-4.6786
Aw ln(Meq) ln(-ln(1—aw)) Actual Mathematical % difference
0.1 1 1.9015788 -2.14957378 6.696459 5.6358 -1 8.821047
0.235 2.2305698 -1.317218231 9.305167 9.9565 6.5415964
0.345 2.5393259 086009935 12.67113 13.6094 6.89443584
0.455 2.8021453 0499276762 16.47996 17.4173 5.3814109
0.545 2.9351401 0238945421 18.82414 20.8104 9.5446271 1
0.75 3.4051634 0.32663426 30.11922 30.6352 1.68416215
0.805 3.674155 0491493387 39.41534 34.2904 -14.94575
Chen Equation model
X axis y axis Ln(-lnaw)=—00696(Meq)+1.0006
Aw (Meg) ln(-lnaw) Actual Mathematical % difi'aence
0.1 1 6.6964587 0.791758684 6.696459 3.0006 -123.17113
0.235 9.3051669 0370300528 9.305167 9.0560 -2.7510958
0.345 12.671 126 0.06223355 12.671 13 13.4823 6.01642128
0.455 16.479963 0238945421 16.47996 17.8096 7.46563734
0.545 18.824141 0499276762 18.82414 21.5500 12.6488107
0.75 30.1 19215 -1.245899324 30.1 1922 32.2773 6.68604402
0.805 39.415336 -1.52825892 39.41534 36.3342 04800488
G.A.B Equation model
x axis y axis (aw/Meq)=00036awt-00227
Aw (aw) (aw/Meq) Actual Mathematical % difi‘erence
0.1 1 0.1 1 0016426593 6.696459 4.7627 40601281
0.235 0.235 0025254786 9.305167 9.9805 6.76618762
0.345 0.345 0027227257 12.671 13 14.4098 12.0660582
0.455 0.455 0027609285 16.47996 18.6950 11.8484945
0.545 0.545 0.028952185 18.82414 22.0988 14.818173
0.75 0.75 0024901047 30.1 1922 29.5276 -2.0037427
0.805 0.805 0020423522 39.41534 31.4478 -25.335872
B.E.T. Equation model
x axis y axis Aw/(Meq(1-aw))=01262aw00003
Aw (aw) aw/Qdequ my) Actual Mathematical % difference
0.1 1 0.11 0.018456846 6.696459 9.0999 204121286
0.235 0.235 0033012792 9.305167 10.4639 11.0738661
0.345 0.345 0041568331 12.67113 12.1815 4.0190916
0.455 0.455 0050659238 16.47996 14.6157 -12.755348
0.545 0.545 0.063631 175 18.82414 17.4915 -7.618631 1
0.75 0.75 0099604188 30.11922 31.7965 5.27506765
0.805 0.805 0104736012 39.41534 40.7559 3.28923392

 

 

139

Table 56. The mathematical model for equilibrium moisture content from the
experimental data of the model product B at 38 °C ( 100 °F ).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chen Equation model
X axis y axis ln(-lnaw)=-00751(Meq)+l.0469
Aw (Meg) ln(-lnaw) Actual Mathematical % difference
0.1 14 5.4358275 0775444344 5.435828 3.6146 -50385759
0.198 9.4891382 0.4821 10203 9.489138 7.5205 -26.176904
0.345 1 1.621493 0.06223355 1 1.62149 13.1 114 11.3634758
0.47 15.738987 0281007617 15.73899 17.6819 109879427
0.51 17.335143 0395498114 17.33514 19.2064 9.74272588
0.755 31 .30162 -1.269267061 31 .30162 30.8411 -1.4931811
0.815 36.069276 -1.586858919 36.06928 35.0700 -2.8493017
Halsey Equation model
X axis y axis Ln(-ln(aw))=-l .28711n(Meq)+3.1837
Aw Ln(Meq) ln(-ln(aw)) Actual Mathematical % difi‘erence
0.1 14 1.69301 18 0.775444344 5.435828 6.4952 16.3107258
0.198 2.2501478 0.4821 10203 9.489138 8.1578 -16.319848
0.345 2.4528562 0.06223355 1 1.62149 1 1.3044 -2.8048968
0.47 2.7561409 0281007617 15.73899 14.7593 0637975
0.51 2.8527358 03954981 14 17.33514 16.1323 -7.4560959
0.755 3.4436698 -1.269267061 31.30162 31.8071 1.5892567
0.815 3.5854414 -1.586858919 36.06928 40.7086 1 1.3964767
G.A.B Equation model
x axis y axis (aw/Meq)=00026aw+00242
Aw Jaw) (aw/Meq) Actual Mathematical % difl‘erence
0.114 0.114 . 0020971968 5.435828 4.6537 -16.805444
0.198 0.198 0020865962 9.489138 8.01 14 -18.445532
0.345 0.345 0029686375 1 1 .62149 13.7467 15.4595342
0.47 0.47 0.029862151 15.73 899 1 8.4879 14.8688265
0.51 0.51 0029420006 17.33514 19.9796 13.235911 1
0.755 0.755 0.024120158 31 .30162 28.8575 «8.4694401
0815 0.815 0022595408 36.06928 30.9662 -16.479421
B.E.T. Equation model
x axis y axis Aw/(Meq(1-aw))=01355aw-00004
Aw (aw) aw/(Meq(1-aw)) Actual Mathematical - % difference
0.1 14 0.1 14 0023670392 5.435828 8.5511 36.431 1343
0.198 0.198 0026017409 9.489138 9.3414 -1.5819819
0.345 0.345 004532271 1 1.62149 1 1.3645 -2.2610951
0.47 0.47 0056343682 15.73899 14.0127 -12.3l9603 '
0.51 0.51 ' 0060040828 17.33514 15.1491 -l4.430467
0.755 0.755 0098449623 31.30162 30.2410 05072524
0.815 0.815 0122137338 36.06928 40.0373 9.91084162

 

 

140

APPENDIX D

Table 57. Experimental data for retained antioxidant levels in pouch structures as a
function of storage time at 23 °C, 50 % RH.

 

 

 

 

 

 

Storage Pouch Type Sample Area Injected Antioxidants
(Weeks) number weight (9) Response volume(ml) Abgraeg'e
0 1 Vit. E 1 25445 0.02
Vit. E 2 1.0641 25025 0.02
V11. E 3 24773 0.02 73 ppm
2 W. E 1 24231 0.02 (7.019)‘
Vit. E 2 1.0457 24835 0.02
Vit. E 3 24489 0.02
0 3 BHT 1 641805 0.02
BHT 2 1.0356 648445 0.02
BHT 3 647065 0.02 1137 ppm
4 BHT 1 656513 0.02 (3.646)*
BHT 2 1.0486 650126 0.02
BHT 3 654936 0.02
2 5 Vit. E 1 13433 0.02
Vit. E 2 0.7635 12511 0.02
Vit. E 3 14710 0.02 63.5 ppm
6 W. E 1 12736 0.02 (6.836)*
Wt. E 2 0.6809 12536 0.02
V11. E 3 12824 0.02
2 7 BHT 1 279284 0.02
BHT 2 0.8519 278473 0.02
BHT 3 279546 0.02 608 ppm
8 BHT 1 289245 0.02 (3.836)*
BHT 2 0.8678 288077 0.02
BHT 3 289925 0.02
4 9 V11. E 1 10179 0.02
Vit. E 2 0.6898 12048 0.02
V11. E 3 14222 0.02 62 ppm
10 Vit. E 1 13207 0.02 (6.83)‘
W. E 2 0.7019 13446 0.02

141

 

 

 

 

 

 

Vit. E 3 13007 0.02
4 11 BHT 1 96393 0.01
BHT 2 0.7871 94345 0.01
BHT 3 98122 0.01 446.5 ppm
12 BHT 1 128865 0.01 (3.29)’
BHT 2 0.9354 128634 0.01
BHT 3 126018 0.01
6 13 Vit. E 1 14035 0.02
Vit. E 2 0.8191 15017 0.02
Vtt. E 3 15871 0.02 61.5 ppm
14 W. E 1 12875 0.02 (6.83)*
Vit. E 2 0.706 12461 0.02
Vrt. E 3 12470 0.02
6 15 BHT 1 188052 0.02
BHT 2 0.9542 186209 0.02
BHT 3 186475 0.02 279 ppm
16 BHT 1 148028 0.02 (3.09)*
BHT 2 0.8803 143680 0.02
BHT 3 144687 0.02
8 17 Vit. E 1 8525 0.02
Vit. E 2 0.5001 8379 0.02
Vlt. E 3 8355 0.02 60 ppm
18 W. E 1 11517 0.02 (6.88)*
Vtt. E 2 0.6214 11114 0.02
Vit. E 3 11370 0.02
8 19 BHT 1 110723 0.02
BHT 2 0.7768 110146 0.02
BHT 3 109976 0.02 211 ppm
20 BHT 1 91421 0.02 (3.06)‘
BHT 2 0.7011 91412 0.02
BHT 3 94181 0.02
10 21 Vlt. E 1 11054 0.02
V11. E 2 0.7018 10484 0.02
Vit. E 3 12120 0.02 45 ppm
22 Vit. E 1 7539 0.02 (6.63)*
Vit. E 2 0.6592 7756 0.02
Vit. E 3 6578 0.02

142

 

 

 

 

 

 

 

10 23 BHT 1 78223 0.02
BHT 2 0.8008 78299 0.02
BHT 3 77509 0.02 149 ppm
24 BHT 1 73716 0.02 (2.98)*
BHT 2 0.7151 72383 0.02
BHT 3 74857 0.02
12 25 V11. E 1 4464 0.02
W. E 2 0.5718 4807 0.02
V11. E 3 4041 0.02 40.5 ppm
26 W. E 1 12224 0.02 (6.85)‘
Vit. E 2 0.7251 10337 0.02
Vtt. E 3 11749 0.02
12 27 BHT 1 40743 0.02
BHT 2 0.7255 39951 0.02
BHT 3 39642 0.02 112 ppm
28 BHT 1 66554 0.02 (3.37)*
BHT 2 0.8454 64777 0.02
BHT 3 65252 0.02
14 29 W. E 1 24190 0.01
W. E 2 0.7462 24037 0.01
W. E 3 23599 0.01 29 ppm
30 Vit. E 1 9660 0.01 (6.24)*
Vit. E 2 0.6513 9290 0.01
Vit. E 3 8684 0.01
14 31 BHT 1 36321 0.02
BHT 2 0.8373 34572 0.02
BHT 3 32443 0.02 79.5 ppm
32 BHT 1 48733 0.02 (3.31)*
BHT 2 0.8513 45573 0.02
BHT 3 46045 0.02
16 33 V11. E 1 11719 0.01
Vit. E 2 0.7261 11078 0.01
W. E 3 10082 0.01 27 ppm
34 W. E 1 16397 0.01 (6.24)’
Vit. E 2 0.6734 16727 0.01
Vit. E 3 15682 0.01
16 35 BHT 1 18942 0.02

143

BHT 2 0.8401 18051 0.02

 

 

BHT 3 16568 0.02 36.5 ppm
36 BHT 1 18068 0.02 (3.30)*
BHT 2 0.7898 17988 0.02
BHT 3 18459 0.02
18 37 V11. E 1 27220 0.02
Vit. E 2 0.7086 33612 0.02
Vit. E 3 33215 0.02 26 ppm
38 Vit. E 1 20536 0.02 (6.55)‘
Vit. E 2 0.7154 23469 0.02
Vtt. E 3 21031 0.02
18 39 BHT 1 24496 0.02
BHT 2 0.8957 23463 0.02
BHT 3 23328 0.02 32 ppm
40 BHT 1 7880 0.02 (3.26)‘
BHT 2 0.8058 11526 0.02
BHT 3 9918 0.02

 

( )* = x 10‘” calibration factor

Table 58. Experimental data for retained antioxidant levels in pouch structures as a
function of storage time at 45 °C, 50 % RH.

 

 

Storage Type Sample Area Injected Antioxidants Antioxidants
(Days) weight (9) Response volume(ml) (ppm) Avleevrzlge
0 W. E 1 1.3591 16553 0.01 82
13910 0.01 69
13813 0.01 69 73 ppm
W. E 2 1.3658 1422 0.01 70
16230 0.01 80
13808 0.01 68
BHT 1 1 .4165 537600 0.01 1 142
550836 0.01 1 171
545862 0.01 1160 1168 ppm
BHT 2 1.6847 679814 0.01 1215
645585 0.01 1 153

652225 0.01 1 165

145

 

 

 

 

 

 

 

4 5 6 7 8 9 10
(cum

 

Figure 37. Standard Calibration Curve for BHT using HPLC.

 

 

 

 

 

 

 

4 e e 10 12 14 1e 18 20
(g x10'°)

 

Figure 38. Standard Calibration Curve for ct-tocopherol using HPLC.

 

 

146

)— 3.57

 

 

 

 

) L

Standard BHT compound BHT impregnated laminated film
(8 PP“) (2.0209 g)

Figure 39. HPLC-chromatogram of an extracted BHT fiom the laminated film.

0
If)
v
0
In
M

Standard (14000le compound aotocopherol impregnated laminated film
(4 ppm) (2.0502 g)

Figure 40 HPLC-chromatogram of an extracted cr-tocopherol from the laminated film.

147

 

 

r" 18.71?

L

 

Standard linoleic acid (6 ppm)

7
18 576

 

 

 

EL. 3

Lino1eieeeidinmoderptoduot(4daye.arrn

Figure 41. GC chromatogram of the methyl ester extracted from the model product.

148

 

 

 

 

 

 

 

 

 

Figure 42. Standard Calibration Curve of trans- 9-12-octadecadienoic methyl ester.

 

80000

 

  
 
  
  
 

70000 «
60000«
50000 1 2
R = 0.9893
3 40000 -
§ 30000.

20000 4

10000 J

 

 

0

 

14 16 18 20 22 24 26
(1110‘ m mole)

 

 

Figure 43. Standard Calibration Curve for Hexanal using GC/MS.

 

l4l

 

 

 

 

 

 

Vit. E 3 13007 0.02
4 11 BHT 1 96393 0.01
BHT 2 0.7871 94345 0.01
BHT 3 98122 0.01 446.5 ppm
12 BHT 1 128865 0.01 (3.29)*
BHT 2 0.9354 128634 0.01
BHT 3 126018 0.01
6 13 Vit. E 1 14035 0.02
Vit. E 2 0.8191 15017 0.02
Vit. E 3 15871 0.02 61.5 ppm
14 Vit. E 1 12875 0.02 (6.83)*
Vit. E 2 0.706 12461 0.02
Vit. E 3 12470 0.02
6 15 BHT 1 188052 0.02
BHT 2 0.9542 186209 0.02
BHT 3 186475 0.02 279 ppm
16 BHT 1 148028 0.02 (3.09)’
BHT 2 0.8803 143680 0.02
BHT 3 144687 0.02
8 17 Vit. E 1 8525 0.02
Vit. E 2 0.5001 8379 0.02
Vit. E 3 8355 0.02 60 ppm
18 Vit. E 1 11517 0.02 (6.88)’
Vit. E 2 0.6214 11114 0.02
Vit. E 3 11370 0.02
8 19 BHT 1 110723 0.02
BHT 2 0.7768 110146 0.02
BHT 3 109976 0.02 211 ppm
20 BHT 1 91421 0.02 (3.06)*
BHT 2 0.7011 91412 0.02
BHT 3 94181 0.02
10 21 Vit. E 1 11054 0.02
Vit. E 2 0.7018 10484 0.02
Vit. E 3 12120 0.02 45 ppm
22 V11. E 1 7539 0.02 (6.63)*
Vii. E 2 0.6592 7756 0.02
Vit. E 3 6578 0.02

142

 

 

 

 

 

 

 

10 23 BHT 1 78223 0.02
BHT 2 0.8008 78299 0.02
BHT 3 77509 0.02 149 ppm
24 BHT 1 73716 0.02 (2.96)*
BHT 2 0.7151 72383 0.02
BHT 3 74857 0.02
12 25 Vit. E 1 4464 0.02
Vit. E 2 0.5718 4807 0.02
Vit. E 3 4041 0.02 40.5 ppm
26 W. E 1 12224 0.02 (6.85)*
Vit. E 2 0.7251 1037 0.02
Vit. E 3 11749 0.02
12 27 BHT 1 40743 0.02
BHT 2 0.7255 39951 0.02
BHT 3 39642 0.02 112 ppm
28 BHT 1 66554 0.02 (3.37)*
BHT 2 0.8454 64777 0.02
BHT 3 65252 0.02
14 29 Vit. E 1 24190 0.01
Vit. E 2 0.7462 24037 0.01
Vit. E 3 23599 0.01 29 ppm
30 Vit. E 1 9660 0.01 (6.24)*
Vit. E 2 0.6513 9290 0.01
Vit. E 3 8684 0.01
14 31 BHT 1 36321 0.02
BHT 2 0.8373 34572 0.02
BHT 3 32443 0.02 79.5 ppm
32 BHT 1 48733 0.02 (3.31)*
BHT 2 0.8513 45573 0.02
BHT 3 46045 0.02
16 33 Vit. E 1 11719 0.01
Vit. E 2 0.7261 11078 0.01
Vit. E 3 10082 0.01 27 ppm
34 Vit. E 1 16397 0.01 (6.24)*
Wt. E 2 0.6734 16727 0.01
Vit. E 3 15682 0.01
16 35 BHT 1 18942 0.02

143

BHT 2 0.8401 18051 0.02

 

 

BHT 3 16568 0.02 36.5 ppm
36 BHT 1 18068 0.02 (3.30)*
BHT 2 0.7898 17988 0.02
BHT 3 18459 0.02
18 37 Vit. E 1 27220 0.02
Vit. E 2 0.7086 33612 0.02
Vit. E 3 33215 0.02 26 ppm
38 Vit. E 1 20536 0.02 (6.55)*
Vit. E 2 0.7154 23469 0.02
Vit. E 3 21031 0.02
18 39 BHT 1 24496 0.02
BHT 2 0.8957 23463 0.02
BHT 3 23328 0.02 32 ppm
40 BHT 1 7880 0.02 (3.26)*
BHT 2 0.8058 11526 0.02
BHT 3 9918 0.02

 

( )* = x 10'13 calibration factor

Table 58. Experimental data for retained antioxidant levels in pouch structures as a
fimction of storage time at 45 °C, 50 % RH.

 

 

Storage Type Sample Area Injected Antioxidants Antioxidants
(Days) weight (9) Response volume(ml) (ppm) Avlee:|ge
0 Vit. E 1 1.3591 16553 0.01 82
13910 0.01 69
13813 0.01 69 73 ppm
W. E 2 1.3658 14222 0.01 70
16230 0.01 80
13808 0.01 68
BHT 1 1.4165 537600 0.01 1 142
550836 0.01 1 171
545862 0.01 1160 1168 ppm
BHT 2 1.6847 679814 0.01 1215
645585 0.01 1 153

652225 0.01 1 165

144

 

 

 

2 W. E 1 0.8055 34588 0.01 73
34465 0.01 72
35577 0.01 75 72.5 ppm
W. E 2 0.8458 34016 0.01 68
37106 0.01 74
36484 0.01 73
BHT 1 0.8405 186051 0.01 167
174310 0.01 156
171047 0.01 153 185 ppm
BHT 2 0.8993 248335 0.01 208
253494 0.01 212
249700 0.01 209
4 W. E 1 1.1996 51035 0.01 72
50591 0.01 71
51975 0.01 73 72 ppm
Vit. E 2 1.0331 40811 0.01 67
44750 0.01 73
45414 0.01 74
BHT 1 0.8892 51275 0.01 43
50579 0.01 43
51671 0.01 44 40 ppm
BHT 2 0.9422 46120 0.01 37
47807 0.01 38
45992 0.01 37
6 Vlt. E 1 0.7184 23460 0.01 55
23069 0.01 54
23439 0.01 55 55.5 ppm
Vit. E 2 0.9165 29913 0.01 55
30710 0.01 57
29612 0.01 55
BHT 1 0.8887 NIA 0.01 NIA
NIA 0.01 NIA
NIA 0.01 NIA NIA
BHT 2 0.951 NIA 0.01 NIA
NIA 0.01 NIA
NIA 0.01 NIA

 

Vit. E = calibration factor (6.77 x 10"3 (g/Area Response»
BHT = calibration factor (3.01 x 10"3 (g/Area Response»

145

 

 

 

 

 

 

 

 

 

 

50000 ..
O t c : c i
4 5 e 7 e 9 10
(9 x10‘)
Figure 37. Standard Calibration Curve for BHT using HPLC.
350000

 

 

 

 

 

(oxw‘)

 

Figure 38. Standard Calibration Curve for a-tocopherol using HPLC.

 

 

145

 

 

 

Area Response

 

 

 

 

(gx10‘°)

10

 

Figure 37. Standard Calibration Curve for BHT using HPLC.

 

 

 

 

    

 

 

3 1500001. R2 = 0.9993
1m .9
50000
0 7 * Y . ¢
4 6 8 10 12 14 16 18 20

(gx10‘)

 

Figure 38. Standard Calibration Curve for (at-tocopherol using HPLC.

 

 

146

—=-—— 3,57

 

 

 

 

 

l l.

Standard BHT compound BHT . edl . 1 film
(8 PP“) (2.0209 g)

Figure 39. HPLC-chromatogram of an extracted BHT from the laminated film.

0
U)
V
O
In
M

Standard a-tocophero] compound a~tocopherol 233%? laminated film

(4 9999)

Figure 40. HPLC-chromatogram of an extracted a-tocopherol from the laminated fihn.

147

 

 

r 18.717

L

Standard linoleic acid (6 ppm)

18.576

 

 

 

F‘L' n

Iinoleicaoidinmodelpmductfidmmfl‘)

Figure 41. GC chromatogram of the methyl ester extracted from the model product.

148

 

 

 

 

 

 

2 3 4 5 e 7 e
(g x 10")

 

 

 

Figure 42. Standard Calibration Curve of trans- 9-12-octadecadienoic methyl ester.

 

 

80000
70000 «

60000 .

  

50000 4 2
R = 0.9893

   

40000 -
30000 «

Arearuponse

20000 «

100001

 

 

 

14 16 18 20 22 24 26
(x10‘ m mole)

 

 

 

Figure 43. Standard Calibration Curve for Hexanal using GC/MS.

149

Hand

Emma-1120

r mmm
uawuauu
wumumuan
anaamnn

“Abundance:
2
26
32

cm DHENNEBERG. m-Puucx NSTITUTE. MN. WEST GERMANY

68
67
441 68
04 60
8 70
1 71

5 1

MW: 100GASt86-25-1 N81”: mmmmr 08: M
0010! 086: Fm. TSCA, RTECS. HOOOC. NIH. ENECS. R08

53

44

1W4

O

.51

82

1W

7' 'YY'V'V'V

 

7911.5

"VV'V'IV

 

72
r

 

07
m

 

A

 

 

. Jill ,.

 

 

 

41

 

 

Figure 44. Standard Mass Spectrum of Hexanal.

150

The total ion cm'rents selected at m/z 29 44 56 72 and 100 in SIM mode for hexanal

 

 

 

 

 

 

TIC
1
9.91:9? §
‘ 06
o 1 l
3 4.0561
fl
‘ 1
§ 2.9:91
if I -A‘*“ Aafihnn¢£:15;;lé
9" —I ' 'r ‘—I 7
6 6 18 12 14 16 18
Time (010.)

 

 

Figure 45. The model product (B) after freezing-dry (0.1645 g) using SIM of Mass
spectrum.

 

 

 

 

 

TIC
3.8E5‘ or
on
. «a
Q
3 2.0:51\‘
c <
‘ 1
t 1
51.8125-
8 .
a I Y r V 'j vi v 1 I 1
6 8 10 12 14 16 18
Ttne thin.)

 

 

Figure 46. The linoleic acid (0.0241 g), (99%, Acros Organics) using SIM of Mass
spectrum.

15]

 

 

 

 

 

 

 

 

 

 

 

Scan 318 (8.398 ntnl
. 56
'\
”1.06139 4‘ /
U
s 1 72
'2 5.0541 \\\ 100
a 1
8 Bl -. . .- I l...
30 40 SO 60 79 69 90 100
Mass/Charge
TIC
1
3.91251 E
u 1 06
g 2.9:51
d
'2 .
= 1.0[5‘
i :\“_’1
a I 1 'V'fi 'Ti 7 ffi"
6 7 8 9 10 11
Time lain.)

 

Figure 47. The hexanal standard compound using SIM of Mass spectrum.

152

The rofile of the selected ion cm-rent ran e in SIM mode for hexanal

 

 

 

 

 

Ion 44.” um.
2.866 M 2.866
c
L {l .
no
8" A ‘ ‘* ‘ H
Ion $6.” mu.
2.866 2.866
v
a O
o \O
s 1 A .5
t
ion 72.” am.
2.866 2.866
m
3
on
a +1 ' " 1:5: Tfiw f """" I V
6 8 18 12 14 16 18
Time lain.)

 

 

 

Figure 48. The model product (B) after freezing-dry (0.1645 g) using SIM of Mass

spectrum.
W m.
2.265H N
9 A ‘

Ion 58.” am.

 

 

'2.265

 

 

 

 

2.265 2.265
V
a co
1:
g 9’ AA A __A- ‘0
F
Ion 72.” mu.
2.265 2.265
o
co
"1
N
a it * :A‘Ai r * ‘AI‘AA‘ A: A “‘1'
6 8 18 12 14 16 18

 

Tine lain.)

 

 

Figure 49. The linoleic acid (0.0241 g), (99%, Acros Organics) using SIM of Mass
spectrum.

153

 

 

 

Ion 44.” am.
o
32
1.865 .6 1.91:5
5.864 5.864
0 ‘ ‘9
Ion 58.88 am. a
3 1.955 :3 1.91:5
S 5.01:4 A 3.91:4
“B
5 8- fig
:2
Ion 72.” am.
1.865 33 1.865
"I
5.864 ‘° 5.864
8
6 7 8 9 18 11
Time (Mm)

 

 

 

Figure 50. The hexanal standard compound using SIM of Mass Spectrum.

154

Table 59. The ratio of the selected ion profiles at m/z 44, 56, and 72 for hexanal.

(l) Hexanal standard

 

 

 

 

 

 

 

Hexanal Total ions Ion 44 Ion 56 Ion 72 At mlz At mlz At mlz
Standard (44/44) L55/44) (72/44)
(ppm) Area Response Ratio of each ion
200 13103931 5539125 3624348 889171 1 0.65 0.16
400 28108899 12317988 7729692 1986165 1 0.63 0.16
600 52183827 20944542 13377271 3275147 1 0.64 0.16
800 75446973 28780401 18051896 4818452 1 0.63 0.17
(2) Sample product
Total ions Ion 44 Ion 56 Ion 72 At mlz At mlz At mlz
44I44 56144 72144
Area Response Ratio of each ion
Linoleic 39577467 18109765 11098908 2682988 1 0.61 0.15
acid
Model 1976506482 778529724 502088472 122492128 1 0.65 0.16
product
(B)

 

Table 60. Experimental hexanal data for model products from each package structure as a
function of storage time at 23 °C, 50 % RH.

 

 

 

 

Storage Pouch Types Tested Area Hexanal Average
(Weeks) Number weight(g) Response (ppm)
0 g 1 Initial 0.2268 11992503 0.26
2 Initial 0.1556 748211 1 0.39 0.33 ppm
3 Initial 0.1729 8777682 0.35
2 4 Control1 0.1865 25825613 0.69 0.55 ppm
5 Control2 0.1883 15431 150 0.41
6 Vit. E 1 0.1622 17274000 0.53 0.58 ppm
7 W. E 2 0.1737 21900202 0.63
8 BHT 1 0.1805 15124836 0.42 0.48 ppm
9 BHT 2 0.1682 17879940 0.53
4 10 Control1 0.1621 20828770 0.64 0.60 ppm
1 1 Control2 0.1878 22033331 0.59
12 Vit. E 1 0.1686 24701755 0.73 0.67 ppm

155

 

 

 

 

 

 

13 Vit. E 2 0.1519 19495463 0.61
14 BHT 1 0.1625 22386569 0.69 0.55 ppm
15 BHT 2 0.1522 12082418 0.40

6 16 Control1 0.2253 22537950 0.50 0.64 ppm
17 Contro12 0.2478 39220047 0.77
19 Vit. E 1 0.1599 11096964 0.35 0.46 ppm
19 Vit. E 2 0.1713 19163524 0.56
20 BHT 1 0.1659 17265690 0.52 0.62 ppm
21 BHT 2 0.2059 29171667 0.71

6 22 Control1 0.1674 12648215 0.39 0.44 ppm
23 Contro12 0.1762 17155443 0.49
24 Vit. E 1 0.1729 11571040 0.33 0.43 ppm
25 Vit. E 2 0.1801 19694963 0.52
26 BHT 1 0.1763 16795256 0.53 0.40 ppm
27 BHT 2 0.1796 9445695 0.26

10 29 Control1 0.1921 12662707 0.35 0.42 ppm
29 Control2 0.1746 17042790 0.49
30 V11. E 1 0.1617 26437902 0.73 0.64 ppm
31 Vit. E 2 0.1772 19536651 0.55
32 BHT 1 0.1793 11456329 0.32 0.7 ppm
33 BHT 2 0.1769 36019935 1.09

12 34 Control1 0.1941 21136465 0.36 0.52 pme"
35 Contro12 0.1952 36391962 0.65
36 Vit. E 1 0.1794 16072924 0.30 0.44 ppm")
37 V11. E 2 0.1913 32419547 0.56
39 BHT 1 0.2174 42550000 0.65 0.71 ppm")
39 BHT 2 0.2295 53094529 0.77

14 40 Control1 0.9075 29777419 0.19 0.38 ppm
41 Contro|2 0.9124 93690026 0.59
42 Vit. E 1 0.9954 92114231 0.45 0.42 ppm
43 Vit. E 2 0.9942 76629027 0.36
44 BHT 1 1.0292 169919996 0.93 0.72 ppm
45 BHT 2 1.0109 122427791 0.61

16 46 Control1 1.0413 252051564 0.49 0.59 ppm”—
47 Contro|2 1.002 346646746 0.69
46 Vit. E 1 1.0149 146532190 0.29 0.49 ppm“)
49 Vit. E 2 1.0039 341457224 0.66

156

 

 

 

 

 

 

50 BHT 1 1.0019 309377090 0.62 0.52 ppm“)
51 BHT 2 1.0776 226936539 0.42

16 52 Control1 0.9941 77107701 0.39 0.51 ppm
53 Contro12 0.9166 115926397 0.63
54 V11. E 1 1.3075 117362665 0.45 0.29 ppm
55 V11. E 2 1.0902 91209532 0.42
56 BHT 1 1.0162 134601322 0.66 0.56 ppm
57 BHT 2 1.0251 92546647 0.45

20 56 Control1 1.0736 435213710 2.03 1.66me
59 Contro12 1.0446 269770596 1.29
60 v11. E1 1.0606 222323293 1.05 1.34ppm‘°’
61 V11. E 2 1.0931 354066479 1.62
62 BHT 1 1.0791 443459227 2.06 1.46ppm‘°’
63 BHT 2 1.0934 197199975 0.96

22 64 Control1 1.1099 299634916 1.31 112me
65 Control2 1.0903 265470691 1.22
66 V11. E 1 1.1671 314271199 1.35 0.99ppm‘°’
67 V11. E 2 1.0737 153559103 0.72
69 BHT 1 1.1071 299590596 1.35 1.03ppm‘°’
69 BHT 2 1.0021 140132819 0.70

24 70 Control1 1.0794 139299955 0.65 0.91 ppm
71 ControlZ 1.0466 201706739 0.96
72 Vit. E 1 1.0309 125905960 0.61 0.59 ppm
73 V11. E 2 1.0474 120299053 0.57
74 BHT 1 1.131 242592464 1.07 0.79 ppm
75 BHT 2 1.0626 105930947 0.50

26 76 Control1 1.2749 453192673 0.89 0.79 ppm
77 Contro|2 1.5641 413136267 0.66
76 V11. E 1 1.4419 395655627 0.69 0.74 ppm
79 V11. E 2 1.5607 499550769 0.79
60 BHT 1 1.7554 592417499 0.93 0.87 ppm
61 BHT 2 1.5921 573393593 0.90

29 92 Control1 1.069 133139751 0.61 0.56 ppm
93 Controlz 1.0043 109451346 0.54
94 V11. E 1 1.0355 131729567 0.64 0.67 ppm
95 Vit. E 2 1.0125 196992761 0.97
86 BHT 1 1.0156 222915997 1.10 0.94 ppm

87

BHT 2

Calibration factor = S x 10‘” (g/A.U)

(a) = 3.33 x 10'” (g/A.U) or.
(b) = 2 x 10''5 (g/A.U) CF.

(0) = multiplier : 3000 in mass spectrum

157

1.0358

1 58806077

0.77

Table 61. Experimental hexanal data for model products from each package structure as a
function of the storage time at 45 °C, 50 % RH.

 

 

 

 

 

Storage Pouch Types Tested Area Hexanal Average
(days) Number weight(g) Response (ppm)
0 1 Control 1 .0861 40297132 0.19
2 Control 1.1917 46175336 0.19 0.21 ppm
3 Control 1.0325 51060730 0.25
2 4 Control1 1.0071 1440785559 7.2 7.1 ppm
5 ControlZ 1.0375 1451991086 7.0
6 V11. E 1 1.0298 2363682208 12 13 ppm
7 Vit. E 2 1.0277 2815748133 14
8 BHT 1 1.0638 56125009 0.26 0.23 ppm
9 BHT 2 1.0118 39574553 0.20
4 10 Control1 1.0469 513358678 2.5 2.5 ppm
1 1 Control2 1.0841 542953058 2.5
12 Vit. E 1 1.0891 575261315 2.6 2.6 ppm
13 Vit. E 2 1.0146 504732764 2.5
14 BHT 1 1.0145 42771370 0.21 0.27 ppm
15 BHT 2 1.0084 64890482 0.32
6 16 Control1 1.0475 270144442 1.29 1.64 ppm
17 Control2 1.0015 397307843 1.98
18 W. E 1 1.0574 408603505 1.93 1.91 ppm
19 Vrt. E 2 1.0132 381085947 1.88
20 BHT 1 1.0131 41022760 0.20 0.19 ppm
21 BHT 2 1.0027 32993965 0.17

 

Calibration factor = 5 x 10'" (g/A.U)

158

Table 62. Experimental methyl ester data of model products as a function of the storage
time at 45 °C, 50 % RH.

 

 

 

 

 

Storage Tested Injection Area linoleic acid Average
(days) Types weight(g) Response % (wt/wt) °/o (gig)
0 Control 1 770029 0.7260
Control 2.5253 2 758612 0.7152 0.72
Control 3 752789 0.7097
2 Control 1 42378 0.0258
Control 3.1262 2 39735 0.0242 0.03
Control 3 43615 0.0266
BHT 1 691275 0.5383
BHT 3.0071 2 796271 0.6123 0.57
BHT 3 718910 0.5599
Vit. E 1 42378 0.0248
W. E 3.2527 2 39735 0.0233 0.02
Vlt. E 3 43615 0.0255
4 Control 1 28746 0.0200
Control 2.7337 2 30656 0.0214 0.02
Control 3 29013 0.0202
BHT 1 658786 0.5771
BHT 2.71 75 2 655984 0.5747 0.57
BHT 3 641332 0.5619
Vit. E 1 13135 0.0109
Vit. E 2.2947 2 15490 0.0129 0.01
Vit. E 3 19143 0.0159
6 Control 1 20954 0.0140
Control 2.8417 2 27389 0.0184 0.02
Control 3 23633 0.0158
BHT 1 688227 0.5735
BHT 2.8568 2 670899 0.5591 0.56
BHT 3 685829 0.5715
Vrt. E 1 45552 0.0308
V11. E 2.8155 2 42683 0.0289 0.03
Vit. E 3 44196 0.0299

 

Calibration factor = l x10'”(g/A.U)

159

APPENDIX E

1%

SI
h OZ
' CZ

0E

SC

09

59
F 05

$5

09

S9
0L

'5 -
I-
o- J."
a 5....
... a. ‘1'

00000

“ ““‘W r61
1
“3.4

009C

00::
\
u‘mz
KV

--- ........
---------~------ --
_-- -n-go‘

pm
0081 000: 009:
I919“
‘Ioumo um

00}!
SL (LN
/
. 'l o
l

00;!

000!

 

 

Figure 51. High density polyethylene side of HDPE/Suryln-EVA laminate film structure
with no antioxidants.

160

 

 

 

 

 

at- 11.1. / 83> 82.99.
88.8 w
a.
u m _
. W m 2,, ..
8. M m \/\
n . .2 ..
, .u u u u .3 .
a . u _ n _ "Er _ .
. u . _ m .
u _ _ . . j .. :
. _ _ "_m n .u u _ .58.:
S . ._ _ .m u a _u u .
. ” ...: _ < m M Wm __
_ Wm. _ u .m .2. _n
_ . u. u .. u“
a. . .“__ _ a 52...: u m." "n
. u _ . n . r
_ .. . . . m" n.—
_:_ __ . u .. w Cm.
s _ . __ u . J
1; __ . T. 39...: \.
_ ..__ _u _ u N . up: __
I _ u
_. _ u u .
we. 2 r u.“ 1 :3.
_. m "u
.n - ..
Nu . "n are .m
w .
u. m.“
8.. u m
.
3...: M / .398
~‘.0 1:. Ill. .141 I1 llll J / L a ”2.: q - a q q u <
533... ”39¢ 38 was N25 N80 38 _So :8 _NS 38 .8

a...—

Figure 52. Heat seal layer (Suryln-EVA) side of HDPE/Suryln-EVA laminate fihn

structure with no antioxidants.

16]

1.96

 

E: z: '5 ... s a: a a s f4 is 343
51'
l
9
l
l
l
u
5" s
U
/'8 ._-..-..___________-----....-.
.5 F‘FZ:TT:;;;:_:: L‘£—_.'_'— -—_-----»_«_~__
§fi§x \)

[one
0091 009! 000: 09!!
w
u'zm
/
E
erg»:
1 1 mm
ma

W"
"'19"
\

 

00' 0001 00:1
91 0a
/

0 009
SE 6".
\

 

Figure 53. High density polyethylene side of BHT impregnated HDPE/Suryln-EVA

laminate film structure.

162

 

3
II. ‘3
g: 3 ’6’ '03 3 3 .3 3‘. 8 ‘12 S a mu;
'owf L
l l

..1
§1 ..

5-

l:
‘64
8

HID
0091 0031 000: 00}: 003:

Q"

 

0011!

0001

 

 

0009

Figure 54. Heat seal layer (Suryln-EVA) side of BHT impregnated HDPE/Suryln-EVA
laminate fihn structure.

163

1%

0 0009

l 691

oz

9:

* or
51:
or
50
05
cs
09

00%

00“
-l . . ..
1.99161

__-—-—------------------—---_.—— ......
" ‘0-

mpg

009l
It’ll."

\
W— —. _— —_-—-----------
I

0091
tram

Nil

000!

 

 

‘ ...
; 8- -—~‘£I—:_—-;‘._-’———-—_—————"'
r_._ ......t,___——__——__-—-_—-__ __ __.__.-__----
1‘ “ N
‘—
. ‘..
' o
| u
.1

Figure 55. High density polyethylene side of a-tocopherol impregnated HDPE/Suryln-
EVA laminate film structure.

164

”CI 19‘

_—.———-——--

§ 02915:
km

“’9‘

random.

PD
0001 1

- -..-
-— —- —. c—I — — _— — -.. ————
- —s$$.’_-_"_: :_=_-____- ----_--------...._..

”I 00?!
scam
I
J

..--.-
coco------
--———--———————-----
-—
—---—--—--—

—
°
—
— 'n t“ N
§‘ 3
§4
f: _________-___,,)
o ..--‘m—J i‘gf-— — --
u —“
‘3

 

§A .41;.-::;:‘.'2... .
'0

Figure 56. Heat seal layer (Suryln-EVA) side of a-tocopherol impregnated
HDPE/Suryln-EVA laminate film structure.

REFERENCES

Allen, J. C. and Hamilton, R. J. 1994. Rancidity in Foods. Third edition. Blackie
Academic & Professional.

AOAC. 1975. Oflicial Methods of Analysis, 12th ed. Association of Official Analytical
Chemists, Washington, DC.

Bailey, L. A. 1995. Mass Transfer of BHT from a Multi-Layer Lamination. M.S. Thesis,
Michigan State University, East Lansing.

Beal, Andrew D. and Mottram, Donald S. 1994. Compounds of Contributing to the
Characteristic Aroma of Malted Barley. J oumal of Agricultural Food Chemistry. 42:
2880-2884.

Berends, C. L. 1993. Measure of the Effect of Water Activity on the Rate of Lipid
Oxidation at Constant Oxygen Concentration. M.S. Thesis, Michigan State University,
East Lansing.

Bidlack, W. R. and Tappel, A. L., 1973. Fluorescent Products of phospholipids during
Lipid Peroxidation. Lipid, 8:203-207.

Birley, A. W., Haworth, B., Batchelor, J. 1991. Physics of Plastics: Processing,
Properties, and Materials Engineering. Oxford University Press, New York.

Bradly, D. and Min, D. B. 1992, Singlet Oxygen Oxidation of Foods. Crit. Rev. in Food
Science and Nutrition, 31: 211.

Bnmauer, S., Emmett, H. P. and Teller, E. 1938. Adsorption of Gases in Multimolecular
Layers. J. Amer. Chem. Soc. 60: 309 - 319.

Buttery, R. G. and Ling, L. C. 1995. Volatile Flavor Components of Com Tortillas and
Related Products. Journal of Agricultural and Food Chemistry. 43: 1878-1882.

Burton, G. W., Hughes, L., Foster, D. 0., Pietlzak, B., Goss-Sampson, M. A., and Muller,
David P. R. 1993. Free Radicals: From Basic Science to Medicine. G. Poli, E. Albano &
M.U. Dianzani (eds.), Birkhauser Verlag Basel, Switzerland, 388-399.

Calvert, P. D. and Billingham, N. C. 1979. Loss of Additives from Polymers: A
Theroetical Model, Journal of Applied Polymer Science, vol. 24, 357-370.

Chan, H. W.-S. 1987. The Mechanism of Autoxidation, In Autoxidation of Unsaturated
Lipids (Chan, H.W.-S., edited) Academic Press, London: 1-16.

165

166

Charalambous, George 1992. Ofi-flavors in Foods and Beverages. (Developments in
Food Sciences; 28). Elsevier Science Publishers, Amsterdam, The Netherlands.

Chung, H. Y. and Cadwallader, K. R. 1994. Aroma Extract Dilution Analysis of Blue
Crab Claw Meat Volatiles. Journal of Agricultural Food Chemistry. 42: 2867-2870.

Cort, W. M. 1974. Antioxidant Activity of Tocopherols, Ascorbyl Palmitate, and
Ascorbic Acid and their mode of action. Journal of American Oil Chemist’s Society.
51 :321- 325.

Coulter, Roberta Buford 1988. Extending Shelf Life by Using Traditional Phenolic
Antioxidants. Cereal Foods World. February, vol. 33, No.2 :207-210.

Crank, J. 1975. The mathematics of difi‘usion, 2‘“1 edition, Clarendon Press, Oxford,
England.

Crosby, N. T. 1981. Food Packaging Materials, Applied Science Publishers, London,
England.

C. van den berg. 1985. Development of B.E.T. - Like Models for Sorption of Water on
Foods, Theory and Relevance. In Properties of Water in F oods' (edited by D. Simatos and
J .L.Multon, Martinus Nijhofl‘ Publishers, Dordrecht/Boston/Lancaster, NATO ASI
Series, No. 90), 119-131.

Dawson, P. L., Sheldon, B. W., Larick, D. K., and Turner, B. E. 1991. Role of
Glycerophophoryloholine and Glycerophosphorylethanolamine in Linoleic Acid
Oxidation. Journal of Agricultural Food Chemistry, 39, 12-16.

Dougherty, Marion E. Jr. 1988. Tocopherols as Food Antioxidants. Cereal Foods World.
February, vol. 33, No.2:222-223.

Dziezak, J. D. 1986. Antioxidants: The Ultimate Answer to Oxidation. Food Technology.
September: 94-102.

Dreher, M. L. 1987. Handbook of Dietary Fiber. Dietary Fiber Ingredients and Food
Uses. Marcel Dekker Inc., New York, Chapter 1 1: 381-441.

Eitenmiller, R. R. 1997. Vitamin E Content of Fats and Oils-Nutritional Implications.
Food Technology. May, vol.51, No.5 : 78-81.

Esterbauer, H., Zollner, H., and Schaur, R. J. 1990. Aldehydes Formed by Lipid
Peroxidation: Mechanism of Formation, Occurrence, and Determination, In Membrane
Lipid Oxidation (edited by Carmen Vigo-Pelfi'ey, CRC press, Inc., Boca Raton, Florida)
Chapter 11: 239-252.

167

Fennema, O. R. 1996. Water and Ice. In Food Chemistry (edited by Owen R. Fennema),
Third edition, Marcel Dekker, Inc., New York, Chapter 2: 17-94.

Frankel, E. N. 1984. Lipid oxidation: Mechanisms, Products and Biological Significance.
Journal of American Oil Chemist’s Society, 61 (12): 1908 - 1915.

Frankel, E. N. 1991. Recent Advances in Lipid Oxidation. Review. Journal of the Science
of Food and Agriculture, 54: 495-511.

Frankel, E. N., Hu, M. -L., and Tappel, A. L. 1989. Rapid Headspace Gas
Chromatopgraphy of Hexanal as a Measure of Lipid Peroxidation in Biological Samples.
Lipids, 24: 976-981.

Fritsch, C. W. and Gale, J. A. 1977. Hexanal as a Measure of Rancidity in Low-Fat
Foods. Journal of American Oil Chemist’s Society, 54: 225 - 228.

F ujii, Hamano, and Yuasa, 1995. Inhibition of the Autoxidation of Methyl Linoleate by
Protein and Highly Concentrated Dextrin in an Emulsion System. Bioscience-
Biotechnology-and-Biochemistry, 59 (9) 1761-1763.

Gardner, H. W. 1985. Oxidation of Lipids in Biological Tissue and Its Significance. In
Chemical Changes in Food during Processing, edited by TR Richardson and J.W.
Finley, Van Nostrand Reinhold/AVI, New York.

Giese, James 1996. Antioxidants: Tools for Preventing Lipid Oxidation, Food
Technology. November: 73-81.

Gloria, M. B. A., Grulke, E. A., Gray, J. I. 1993. Efl‘ect of Type of Oxidation on Beta-
Carotene Loss and Volatile Products Formation in Model Systems. Food Chemistry, 46
(4) 401-406.

Gopala, G. and Prabhakar, V. 1995. Retarding Autoxidation in Raw Peanut Oil by
Addition of Water. Journal of American Oil Chemists' Society. 72 (10) 1219-1221 .

Gray, J. I. 1978 Measurement of Lipid Oxidation: A Review. Journal of the American Oil
Chemists’ Society, June, vol. 55, 539-546.

Han, J. K., Miltz, J., Harte, B. R., Giacin, J. R. 1987. Loss of 2-Tertiary-Burtyl-4-
Methoxy Phenol (BHA) fiom High-Density Polyethylene Film. Polymer Engineering and
Science, July, vol. 27, No. 13, 934-938.

Hall, Gunnar; Andersson, Jonas; Lingnert, Hans; Olofsson, Berti]. 1985. Flavor Changes
in Whole Milk Power during Storage (II. The Kinetics of the Formation of Volatile Fat
Oxidation Products and Other Volatile Compounds). Journal of Food Quality. 7: 153-
190.

168

Hallberg, Marianne Lilja and Lingnert, Hans. 1991. Lipid Oxidation in Potato Slices
under Conditions Simulating the Production of Potato Granules. Journal of American Oil
Chemist’s Society, vol. 68, no. 3: 167-170.

Henderson, S. K., Witchwoot, A. and Nawar, W.W. 1980. The Autoxidation of
Linoleates at Elevated Temperatures. Journal of American Oil Chemist’s Society, 5 7:
409-413.

Ho, Y. C. and Yam, K. L. 1994. Comparison of Vitamin E , lrganox 1010 and BHT as
Antioxidants on Release of Off-Flavor fiom HDPE bottles. Journal of Plastic Film &
Sheeting, 10: 194 - 212.

Hoojjat, P., Harte, B. R., Hernandez, R. J. and Giacin, J. R. 1987. Mass Transfer of BHT
fiom HDPE Film and Its Influence on Product Stability. Journal of Packaging
Technology.1(3) : 78.

Jadhav, S. J ., Nimbalkar,S. S., Kulkarni, A. D. ,and Madhavi,D. L. 1996. Lipid
oxidation in biological and food systems, In Food antioxidants (Madhavi, D. L. et a1 .,
ed.) Chapter 2: 5-63.

Jeon, I. J. and Bassette, R. 1984. Analysis of n-Pentanal and n-Hexanal as Indices of
Potato Chip Shelf-Life. Journal of Food Quality, 7: 97-105.

Johnson, Mary, and Fischer, J. G. 1994. Role of Minerals in Protection Against Free
Radicals. Food Technology. May: 112-120

Karmas, Endel 1980. Techniques for Measurement of Moisture Content of Foods. Food
Technology. April: 52-58.

King, D. L., Hahm, Tae S., Min, Davis B. 1993. Chemistry of Antioxidants in Relation to
Shelf Life of Foods. In Shelf-Life Studies of Foods and Beverages. Charalambous, G.
(Bd.) Elsevier Science Publishers: 629-705.

Kishida, E., Tokumaru, S., Ishitani, Y., Yamamoto, M., Oribe, M., Iguchi, H., Kojo,
S. 1993. Comparison of the Formation of Malondialdehyde and Thiobarbituric Acid-
Reactive Substances from Autoxidized Fatty Acids Based on Oxygen Consumption.
Journal of Agricultural and Food Chemistry. 41(10): 1598-1600.

Kochhar, S. P. 1993. Oxidative Pathways to the Formation of Off-Flavors. In Food Taints
and Ojf-Flavors. (edited by Saxby, M.J.), Blackie Academic & Professional, Glasgow,
UK.

Koelsch, C. M. 1989. A System for the Measurement of the Rate of Lipid Oxidation at
Concentration at Constant Oxygen Concentrations and Relative Humidity. M.S. Thesis,
Michigan State University. East Lansing.

169

Kumor, E. A. 1986. The Characterization and Stability of the Lipids in Oat Cereal. M.S.
Thesis, Michgan State University, East Lansing.

Labuza, T. P. 1971. Kinetics of Lipid Oxidation in Foods. CRC Critical Reviews in Food
Technology. October: 355-405.

Laermer, S. F. & Nabholtz, F. 1990. Use of Biological Antioxidants as Polypropylene
Stabilizers. Plastics and Rubber Processing and Applications. 14: 235 - 239.

Laermer, S. R, Young, S. S., and Zambetti, P. F. 1994. Could Your Packaging Use A
Dose of Vitamin E ?. Food Processing. June. Vol. 55: 35 - 36.

Laermer, S. F ., Young, S. S., and Zambetti, P. F. 1993. Vitamin E: New Developments
in Polyolefin Antioxidant Technology. Proc. in SPO ‘93, pp. 541-585.

Laermer, S. F. and Zambetti, P. F. 1992. Alpha-Tocopherol (Vitamin E)-The Natural
Antioxidant for Polyolefins. Journal of Plastic & Sheeting. 8: 228 - 248.

Landvik, S. V., Diplock, A. T., and Packer L. 1996. Efficacy of Vitamin E in Human
Health and Disease. In Handbook of Antioxidants (edited by Cadenas and Packer ),
Marcel Dekker, Inc., New York, Chapter 4, 463-87

Leino, M. E. 1992. Effect of Freezing, F reeze-Drying, and Air-Drying on Odor of Chive
Characterized by Headspace Gas Chromatography and Sensory analyses. Journal of
Agricultural Food Chemistry, 40: 1379-1384.

Leung, Henry K. 1987. Influence of Water Activity on Chemical Reactivity. In Water
Activity: Theory and Applications to Food. (edited by Rockland, Louis B. and Beuchat,
Larry R.), IF T Basic Symposium Series, Chicago, Illinois, 27-54.

Likens, S. T. and Nickerson, G. B. 1964. Detection of Certain Hop Oil Constituents in
Brewing Products. Am. Soc. Brewing Chemists Proc., 5-15.

Lin, J. F. 1996. Mass Transfer of a-Tocoherol (Vitamin E) from a Multi-Layer Laminate
Structure and Its Influence on Product Stability. M.S. Thesis, Michigan State University,
East Lansing.

Luddy, F. B., Barford, R. A., Herb, S. F ., and Magidman, P. 1968. A Rapid and
Quantitative Procedure for the Preparation of Methyl Esters of Butter Oil and Other Fats.
Journal of the American Oil Chemists’ Society. 45. 549-552.

170

Liming, P. A., Rijk, T. D., Wichers, H. J ., and Roozen, J. P. 1994. Gas Chromatography,
Mass Chromatdgraphy, and Sniffing Port Analyses of Volatile Compounds of Fresh Bell
Peppers (Capsicum annuum) at Different Ripening Stages. Journal of Agricultural Food
Chemistry. vol. 42, no. 4, 977-983.

Machlin, L. J. 1984. Vitamin E. In Handbook of Vitamins. Marcel Decker, Inc., New
York, Chapter 3, 99- 145.

Madhavi, D. L. and Salimkhe, D. K. 1996. Toxicological Aspects of Food Antioxidants.
In Food antioxidants. Marcel Dekker, Inc., New York, 267-359.

Marion E. Dougherty, Jr. 1988. Tocopherols as Food Antioxidants. Cereal Foods World,
33: 222 - 223.

Marshell, W. E. 1974. Health Foods, Organic Foods, Natural Foods (what they are and
what makes them attractive to consumers), Food Technology. February, 28: 50-56

Martin, R. A.; Richard, C.; and Rousseau-Richard, C. 1990. Oxidation of Linoleic Acid
and Related or Similar Compounds. In Membrane Lipid Oxidation (edited by Carmen
Vigo-Pelfrey), Chaper 4, 63-99.

Maskan, M., Oner, M.D., and Kaya, A. 1993. Storage Stability and Accelerated Shelf-
Life Testing of Margarine Samples. Journal of Food Quality. 16: 175 -186.

Mate, J. I., Saltveit, M. E., and Krochta, J. M. 1996. Peanut and Walnut Rancidity:
Effects of Oxygen Concentration and Relative Humidity. Journal of Food Science
61 (2): 465-467.

Mathluothi, M. 1986. The influence of Light Transmittance and Gas Permeability of
Various Packing Materials on the Quality of Whole Natural Yoghurt during Storage. In
Food Packaging and Preservation (edited by J .O.Bosset, etc.), Elsevier Applied Science
Publishing Co. Inc., New York.

McCord, Joe M. 1994. Free Radicals and Prooxidants in Health and Nutrition. Food
Technology. May, 106-111.

Moran, D. P. J., Rajah, K. K. 1994. Fats in Food Products. First edition. Blackie
Academic & Professional, Glasgow, UK.

Moreau, D. L. and Rosenberg, M. 1996. Oxidation Stability of Anhydrous Milkfat
Microencapsulated in Whey Proteins. Journal of Food Science, Vol. 61, No. 1, 39-43.

Morrison, William R. and Smith, Lloyd M. 1964. Preparation of Fatty Acid Methyl
Esters and Dimethylacetals from Lipid with Boron Fluride-Methanol. Journal of Lipid
Research, Vol. 5, 600-608.

171

Nelson, Gary J. 1991 . Isolation and Purification of Lipids fi'om Biological Matrices. In
Analyses of Fats, Oils and Lipoproteins (edited by Edward G. Perkins), the American Oil
Chemist’s Society, Champaign, Illinois.

Nelson, Katherine A. and Labuza, Theodore P. 1992. Relationship Between Water and
Lipid Oxidation Rates (Water Activity and Glass Transition Theory). In Lipid Oxidation
in Food (edited by Allen J. St. Angelo), the American Oil Chemist’s Society, Champaign,
Illinois.

Noormarji, H. (Sheiler) 1990. Efl‘ect of Lipid Oxidation on Functional and Nutritional
Properties of Chicken Myofibrillar Proteins Stored at Different Water Activities. Ph.D.
Thesis, Michigan State University, East Lansing.

Packer, L. 1992. New Horizons in Vitamin E Research — the Vitamin E Cycle,
Biochemistry, and Clinical Applications. Lipid-Soluble Antioxidants: Biochemistry and
Clinical Applications (edited by A.S.H. Ong and L. Packer), l —16.

Pershem, A. S., Breene, W. M., and Lulai, E. C. 1995. Analysis of Factors Influencing
Lipid Oxidation in Hazelnuts ( CORYL US SPP. ). Journal of Food Processing and
Preservation. 19: 2-26.

Porter, W. L., Levasseur, L. A., Henick, A. S. 1977. Evaluation of Some Natural and
Synthetic Phenolic Antioxidants in Linoleic Acid Monolayers on Silica. Journal of Food
Science. vol. 42 (6): 1533-1535.

Pokomy, Jan. 1971. Stability of Fats by Phenolic Antioxidant. Can.Inst. Food Technol. J.
4(2): 68 - 74.

Quast, D. and Karel, M., 1972. Effects of Environmental Factors on the Oxidation of
Potato Chips. Journal of Food Science. 37: 584-588.

Quast, D. and Karel, M., 1971. Effects of Oxygen Diffusion on Oxidation of Some Dry
Foods. Journal of Food Technology, 6: 95-106.

Ragnarsson, J. O. and Labuza, T. P. 1977. Accelerated Shelf-life Testing for Oxidative
Rancidity in Foods- A Review. Food Chemistry. 2: 291 - 308.

Ragnhild Reinton and Astri Rogstad. 1981. Antioxidant Activity of Tocopherols and
Ascorbic Acid. Journal of Food Science. 46: 970 - 973.

Raws, H. R. and Vansanten, P. J ., 1970, A Possible Role of Singlet Oxygen Oxidation in
the Initiation of Fatty Acid Autoxidation. Journal of American Oil Chemistry’s Society,
47: 121-125.

172

Reineccius, G. A. 1984. Determination of Flavor Compounds. In Modern Methods of
Food Analysis (edited by Stewart, Kent K and Whitaker, John R.), AVI Publishing
company, Westport, Connecticut.

Reineccius, G. A. 1993. Biases in Analytical Flavor Profiles Introduced by Isolation
Method. In Flavor Measurement (edited by Ho, Chi-ng and Manley, C. H.), Marcel
Dekker, Inc., New York.

Rho, K. L., Selb, P. A., Chung, O. K. and Chung, D. S. 1986. Retardation of Rancidity in
Deep-Fried Instant Noodles (Ramyon). Journal of American Oil Chemists' Society. 63
(2): 251 - 256.

Roberta B. C. 1988. Extending Shelf Life by Using Traditional Phenolic Antioxidants.
Cereal Foods World. 33: 207 - 210.

Robertson, Gordon L. 1993. Food Packaging: Principles and Practice, Marcel Dekker,
Inc. New York.

Rockland, L. B. and Nishi, S. K. 1980. Influence of Water activity on Food Product
Quality and Stability. Food Technology. April, 42-51.

Roozen, J. P., Frankel, E. N. and Kinsella, J. E. 1994. Enzymic and Autoxidation of
Lipids in Low Fat Foods: Model of Linoleic Acid in Emulsified Hexadecane. Food
Chemistry. 50: 33—38.

Roozen, J. P., Frankel, E. N. and Kinsella, J. E. 1994. Enzymic and autoxidation of
Lipids in Low Fat Foods: Model of Linoleic Acid in Emulsified Triolein and Vegetable
Oils. Food Chemistry. 50: 39-43.

Roseiers, C. D., Rivest, M.-J., Boily, M.-J., Jette, M., Carrobe-Cohen, A., Kumar, A.
1993. Gas Chromatographic-Mass Spectrometric Assay of Tissue Malondialdehyde, 4-
Hydroxynonenal, and Other Aldehydes after Their Reduction to Stable Alcohols.
Analytical Biochemistry. 208, 161-170.

Rossell, J. B. 1994. Measurement of Rancidity. In Rancidity in Foods (edited by Allen,
J .C. and Hamilton, 1U.) Third edition. Blackie Academic & Professional, Glasgow, UK,
Chapter 2, 22-53.

Sakakibara, H., Yanai, T., Yajima, I., and Hayashi, K. 1988. Changes in Volatile Flavor
Compounds of Powered Dried Bonito (Katsuobushi) during Storage. Agric. Biol. Chem.,
52 (11): 2731-2739.

Schieberle, P. and Grosch, W. 1981. Model Experiments about the Formation of Volatile
Carbonyl Compounds. Journal of American Oil Chemist’s Society. 58: 602-607.

173

Schuler, Peter 1990. Natural Antioxidants Exploited Commercially. In Food Antioxidants
(edited by Hudson, B.J.F.), Chapter 4. Elsevier Applied Science, New York.

Scott, W. J. 1957. Water Relations of Food Spoilage Microorganisms. Adv. Fd Res., 7:
83-127.

Shahidi, F. and Wanasmdara, U. 1995. Effect of natural antioxidants on the stability of
canola oil. In Food Flavors: Generation, analysis and Process Influence (edited by
Charalambous, G.), Elsevier Science, Amsterdam, The Netherlands.

Slover, H. T. and Lanza, E. 1979. Quantitative analysis of food fatty acids by capillary
gas chromatography. Journal of the American Oil Chemists’ Society. 56. 933-943.

Thomas, D. B. 1994. Rate of Lipid Oxidation at Various Initial Oxygen Concentrations
using Absorbing Sachets in the Packages. M.S. Thesis, Michigan State University, East
Lansing.

Till, D. E., Ehntholt, D. J ., Reid, R. C., Schwartz, P. S., Sideman, K. R., Schwope, A. D.
and Whelan, R. H. 1982. Migration of BHT Antioxidant fi'om High Density Polyethylene
to Foods and Food Simulants. Ind. Eng. Chem. Prod. Res. Dev., 21: 106-113.

Tjhio, K. H. and Karel, M. 1969. Autoxidation of Methyl Linoleate in Freeze-Dried
Model Systems. 4. Effects of Metals and of Histidine in the Absence of Water. Journal of
Food Science, 34: 540 - 543.

Troller, John A. 1989. Water Activity and Food Quality. In Water and Food Quality
(edited by Thehna M. Hardman), Chapter]: 1-31.

Troller, J. A. and Christian, J. H. B. 1978. Water Activity Methods. In Water Activity and
Food, Academic Press, New York, Chapter 2: 15-28.

Young, L.G. ; Lun, A.; J. Pos ; Forshaw, R. P. ;Edmeades, D. 1975. Vitamin E Stability
in Corn and Mixed Feed. Journal of Animal Science, 40: 497 - 499.

Warner, K. J. H., Dimick, P. S., Ziegler, G. R., Mumma, R. O., and Hollender, R. 1996.
‘Flavor-Fade’ and Off-Flavors in Ground Roasted Peanuts as Related to Selected
Pyrazines and Aldehydes. Journal of Food Science. 61(2): 469-472.

Watson, J. T. 1985. Introduction to Mass Spectrometry. Raven Press, New York.

Watson, J. T. and Sparkman, O. D. 1996. Mass Spectrometry: Principles and Practice.
American Chemical Society Short Course, Michigan State University, East Lansing.

Widicus, W. A. and Kirk, J. R. 1981. Storage Stability of Alpha-Tocopherol in a
Dehydrated Model Food System containing Methylene Linoleate. Journal of Food
Science. 46: 813 - 816.

174

Wolf, J. C., Thompson, D. R., Warthesen, J. J., and Reineccius, G. A. 1981. Relative
Importance of Food Composition in Free Lysine and Methionine Losses during Elevated
Temperature Processing. Journal of Food Science. 46: 1074.

Zambetti, P. F. 1995. Vitamin E: Effective Antioxidant for Extrusion Coating Polymers.
Packaging Technology & Engineering. September: 56-59.