1.1::
2.: 1‘: n .2
3: an

3....

#5332...

.11) .
x33.

:1... x
25;} .1...
~ :1

1‘! i.
.333: $53.. .1.
5. 1.1 531.
ii. i

r, 25;...

.1523): I11:
.?:\X.:.v: 531:}?
5:3...

5:3. 1,)?!
3.57:: 9!.

. ‘ 7.x . . .
x24..- 3:... ‘ in;
. . (luv‘iO.
:11}:

it: .5. II
:1 7|»...xl

in}?
.. 13$...

73..
a? r z. 3v. ‘ ):I)1€£.77
$9. .1»: '11:: .33: 1:12:11
1.. 1:. . 1. 1:32.)...-

. 3.3.31. $1.”...

>31
Illfiihuz. P.

.3 .
.. ‘ 1 ...
:1: V; . :
.2. E, a...
.5135!
ii)?

 

 

 

 

 

MICHIGA STATE \ SITY LIBRARIES

\ll lllllllll \\ \llll “\lllll

3 “1293 o EEO 634

 

 

 

 

 

 

 

l

 

l

This is to certify that the

thesis entitled

CONTROL OF OFFfiFLAVOR AND ODOR DEVELOPMENT
IN PACKAGED FROZEN FISH DURING STORAGE

presented by

Lin-Bin Su

has been accepted towards fulfillment
of the requirements for

Masters Packaging

degree in

 

 

+6M<KM

Bruce R. Harte

 

Major professor

[hue June 14, 1994

 

0-7639 MSU is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE

 

 

 

 

 

 

 

 

_____—-——____‘.———-_'.—-—-——

 

,i

:L::

MSU is An Affirmative Action/Equal Opportunity Institution
Wm

 

 

 

 

CONTROL OF OFF-FLAVOR AND ODOR DEVELOPMENT
IN PACKAGED FROZEN FISH DURING STORAGE

By

Lin—Bin Su

A THESIS

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

MASTER OF SCIENCE

School of Packaging

1994

ABSTRACT

CONTROL OF OFF-FLAVOR AND ODOR DEVELOPMENT IN PACKAGED
FROZEN FISH DURING STORAGE

By

Lin-Bin Su

A 133.553.8391. abgbent to control offgflwaygrgandwodorcma
P§£§§3€9P®QDCI was investigated. Frozen fish (lake trout) was used
as the model system.

Volatile compounds were collected on Tenax GC at 2 10C from
oxidized frozen fish (lake trout). Gas chromatography/ mass
spectrometry was used to identify the volatile compounds. The
major off-flavor compounds found were butanal, cycloheptamene,
pengflflndhexanal.

The adsorbing capacity of six adsorbents, silica gel, activated
carbon, sodium bisulfite, lactose, cellulose, and pectin were
investigated by monitoring the concentration of volatile compounds
in the headspace of jars containing the above volatiles. Activated
carbon had the greatest adsorbing capacity of the adsorbents.

Performance of the adsorbent, activated carbon, to reduce the
volatiles from frozen fish packaged in glass jars was studied. The
results demonstrated that activated carbon or other adsorbents

maybe used to adsorb odors associated with storage of frozen fish.

To my mother, Sheng Huang Su

iii

 

ACKNOWLEDGMENTS

I would like to thank Dr. Bruce Harte for his guidance and
patience while serving as my major professor. His help was
invaluable and deeply appreciated.

Sincere appreciation is also extended to Dr. Jack R. Giacin and
Dr. J. Ian Gray for their assistance and for serving on my research
committee.

Special thanks to Dr. John Gill for his advice on the statistical
analysis portion of this work.

Acknowledgment is given to The Center for Food and
Pharmaceutical Packaging Research for project funding.

I would also like to thank Shaun Chen, Dr. Shu-mei Lai, and all
my friends for their never fading friendship.

My parents and sisters: For their support and understanding.

Mei-Ling Chu: For her support and patience.

Last, but not least, I would like to thank the Faculty and Staff
of the School of Packaging for making my overseas studies as much

enjoyable as they did.

iv

 

TABLE OF CONTENTS

PAGE
LIST OF TABLES viii
LIST OF FIGURES x
INTRODUCTION 1
LITERATURE REVIEW 3
Lipids in Fish 3
Lipid Oxidation in Fish 3
Deterioration of Fish Lipids during Cold Storage 4
Off-flavors Developments in Fish due to Lipid Oxidation 6
Measurement of Oxidative Off-flavors in Fish 7
Analytical Methods to Determine Level and Identity of
Volatile Compounds 9
Isolation of Volatile Compounds 9
Identification of Volatile Compounds 11
Prevention of Off-flavors Development in Fish during
Storage 12
Effect of Storage Temperature on the Quality of
Frozen Fish 13
The Activity and Application of Antioxidant in
Fish Products 13
Packaging Requirements for Frozen Fish 15
Adsorbents Having the Capability of Adsorbing Off-flavors 18
Silica Ge] 1 8
Activated Carbon 20
Sodium Bisulfite 22
Cellulose 22
Lactose 23

 

vi

Pectin
Sensory Evaluation

MATERIALS AND METHODS

Materials
Adsorbents
Volatile Standards
Packaging Material for Adsorbents
Pierce Mininert Valve

Methodology
Experiment 1. Identification of Volatile Aroma
Compounds from Oxidized Frozen Fish
Experiment 11. Distinguishing the Adsorbing Capability
of Six Perspective Adsorbents
Experiment III. Evaluation of the Selected Adsorbents

with Frozen Fish System at Low Temperature (50C)
Sensory Evaluation

RESULTS AND DISCUSSION
Major Off-flavor Volatiles of Oxidized Frozen Lake Trout

Adsorption of Volatile Standards by the Six Adsorbents
Effectiveness of Six Adsorbents on Butanal
Effectiveness of Six Adsorbents on Cycloheptatriene
Effectiveness of Six Adsorbents on Pentanal
Effectiveness of Six Adsorbents on Hexanal

Statistical Analysis
Significant Difference of Conditional Comparisons of
Volatile Standards for the Same Adsorbent
Significant Difference of Conditional Comparisons of
Adsorbents for the Same Volatile Standard

Effect of Selected Adsorbents on Oxidized Frozen
Lake Trout
Adsorption Activity of Activated Carbon for
Frozen Lake Trout
Adsorption Activity of Silica Gel for Frozen
Lake Trout

24
24
29
29
29
3O
31
31
33
33

36

4O
42

44
44
46
47
49
49
52
68
73

80

88

91

91

 

vii

Sensory Evaluation
SUMMARY AND CONCLUSIONS

APPENDIX A
GC Mass Spectrometry Analysis

APPENDIX B
Standard Calibration

APPENDIX C
Original Experimental GC Data

APPENDIX D
Statistical Analysis

APPENDIXE
Consent Form
Questionnaire for Simple Paired Comparisons Test

LIST OF REFERENCES

92

94

96

103

118

128

143
144

145

 

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

 

LIST OF TABLES

PAGE
1 Major Volatile Compounds Identified in Frozen
Lake Trout 45
2 Relative % Concentration of Volatile Standard in
Vial Headspace after 24, 48, and 72 Hours Storage
at 21°C 4 8
3 Analysis of Variance for Relative % Concentration
of Volatile Standards in the Vial Headspace 69
4 Comparisons of Volatile Standards for Each Adsorbent
on Day 1 7O
5 Comparisons of Volatile Standards for Each Adsorbent
on Day 2 71
6 Comparisons of Volatile Standards for Each Adsorbent
on Day 3 72

7 Comparisons of Adsorbents for Each Volatile Standards
on Day 1 74

8 Comparisons of Adsorbents for Each Volatile Standards
on Day 2 75

9 Comparsions of Adsorbents for Each Volatile Standards
on Day 3 76

10 Area Response of Volatile in the Headspace of Fish
Sample after 72 Hours Storage at 50C 90

11 Relative % Concentration of Volatile in the Headspace
of Fish Sample after 72 Hours Storage at 50C 90

12 Sensory Score of Frozen Lake Trout with Activated
Carbon in a Jar after Storage at 5°C for 3 Days 93

viii

Table

. Table
Table
Table
Table

Table
Table
Table
Table
Table
Table
Table
Table

Table

ix

A-l Retention Time, Area Response, of Volatile Aroma

Compounds from Oxidized Frozen Lake Trout 9 8
B-l Butanal Calibration Data 104
B-2 Cycloheptatriene Calibration Data 104
B-3 Pentanal Calibration Data 105
B-4 Hexanal Calibration Data 105

C-1 Area Response of Butanal in Cooperation with
Adsorbents after 3 Days Storage 119

C-2 Area Response of Cycloheptatriene in Cooperation
with Adsorbents after 3 Days Storage 120

03 Area Response of Pentanal in Cooperation with
Adsorbents after 3 Days Storage 121

C-4 Area Response of Hexanal in Cooperation with
Adsorbents after 3 Days Storage 122

05 Relative % Concentration of Butanal in Cooperation
with Adsorbents after 3 Days Storage 123

06 Relative % Concentration of Cycloheptatriene in
Cooperation with Adsorbents after 3 Days Storage 124

C-7 Relative % Concentration of Pentanal in Cooperation
with Adsorbents after 3 Days Storage 125

C-8 Relative % Concentration of Hexanal in Cooperation
with Adsorbents after 3 Days Storage 126

C-9 Area Response of Volatile Standards in the Headspace
after 24, 48, and 72 Hours Storage at 21°C 127

Figure

Figure

Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

LIST OF FIGURES

1 Pierce Mininert Valves

2 Glass apparatus for dynamic gas-purging of
headspace aroma volatiles onto Tenax-GC

3 Cross Sectional View of Diffusion/Adsorption Cell

4 Cross Sectional View of Diffusion/Adsorption Jar

5 Adsorption by Adsorbent of Butanal during Storage
at 210C

6 Adsorption by Adsorbent of Cycloheptatriene
during Storage at 210C

7 Adsorption by Adsorbent of Pentanal during
Storage at 210C

8 Adsorption by Adsorbent of Hexanal during
Storage at 210C

9 Relative % Concentration of Butanal Remaining

in the Vial Headspace

10 Relative % Concentration of Cycloheptatriene
Remaining in the Vial Headspace

11 Relative % Concentration of Pentanal Remaining
in the Vial Headspace

12 Relative % Concentration of Hexanal Remaining
in the Vial Headspace

PAGE
3 2

34

41

50

51

53

55

57

58

59

60

xi

Figure 13 Relative % Concentration of Volatile Standards
Remaining in the Vial Headspace with Silica
Gel as the Adsorbent 61

Figure 14 Relative % Concentration of Volatile Standards
Remianing in the Vial Headspace with Activate
Carbon as the Adsorbent 62

Figure 15 Relative % Concentration of Volatile Standards
Remaining in the Vial Headspace with Sodium
Bisulfite as the Adsorbent 63

Figure 16 Relative % Concentration of Volatile Standards
Remaining in the Vial Headspace with Cellulose
as the Adsorbent 64

Figure 17 Relative % Concentration of Volatile Standards
Remaining in the Vial Headspace wiht Lactose
as the Adsorbent 65
Figure 18 Relative % Concentration of Voaltile Standards
Remaining in the Via] Headspace with Pectin
as the Adsorbent 66

Figure 19 Relative % Concentration of Butanal Remaining
in the Vial Headspace 82

Figure 20 Relative % Concentration of Cycloheptatriene
Remianing in the Via] Headspace 83

Figure 21 Relative % Concentration of Pentanal Remianing
in the Via] Headspace 84

Figure 22 Relative % Concentration of Hexanal Remianing
in the Vial Headspace 85

Figure A-l Mass Chromatogram of Volatile Aroma Compounds
from Oxidized Frozen Lake Trout 97

Figure A-2 Mass Spectrum of Peak #1 (Retention Time:3.912) 99

Figure A-3 Mass Spectrum of Peak #2 (Retention Time:4.070) 100

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

Figure

xii

A-4 Mass Spectrum of Peak #3 (Retention Time:8.254)

101

A-5 Mass Spectrum of Peak #4 (Retention Time:27.020) 102

8-1 Butanal Calibration Curve on Dayl

B-2 Butanal Calibration Curve on Day2

B-3 Butanal Calibration Curve on Day3

B-4 Cycloheptatriene Calibration Curve on Dayl
B-5 Cycloheptatriene Calibration Curve on Day2
B-6 Cycloheptatriene Calibration Curve on Day3
B-7 Pentanal Calibration Curve on Dayl

B-8 Pentanal Calibration Curve on Day2

B-9 Pentanal Calibration Curve on Day3

B-10 Hexanal Calibration Curve on Dayl

B-ll Hexanal Calibration Curve on Day2

B—12 Hexanal Calibration Curve on Day3

106

107

108

109

110

111

112

113

114

115

116

117

INTRODUCTION

Fish is considered an inexpensive sources of animal protein
worldwide (Sawyer et al., 1988). Many studies on consumer
acceptance of fish indicate that flavor is an important criterion for
determination of acceptability (Bremner, 1977,1978; Iaslett and
Bremner, 1979; Wesson et al., 1979; Connell and Howgate, 1971;
Hamilton and Bennett, 1983,1984).

/I:ipids oxidatlon/rs one of the main factors responsible for the
magoffiflaxorinfish andQLher. hpidimnihgfonds. The
initiation of lipid oxidation and breakdown of lipid oxidation
products can be enzymatic/f nonenzymatic in nature (Hsieh and

Kinsella, 1986). Some of thgbreakdown products of lipid oxidation,

 

including aldehydes, ketones,_and alcohols, contributegtowoff-flavors,
in foods (Frankel, 1984). Fish lipids, which contain high amounts of
n-3 polyunsaturated fatty acids (PUFAs), are very susceptible to lipid
oxidation. The oxidation products of n—3 PUFAs, particularly
aldehydes with an n-3 double bond, possess low flavor thresholds
(Frankel, 1984). Therefore, once lipid oxidation is initiated, very low
concentrations of aldehydwesfwithn—B double bonds can cause
distinctive oxidative off-flavors in fish (Hsieh and Kinsella, 1986).
Application ofgfinfltjozdd‘ants to some species of fish before frozen
storage does not effectively control theformationofflozddizgdgflavors
(Sweet,1973; Deng et al., 1977, 1978; Morris and Dawson, 1979).

1

 

This may be due to the uneven distribution and penetration of
antioxidants as well as types of antioxidants employed in the system
(Josephson et al., 1985).
3/ Themducrsgihpia oxidation maynoLonlyberesponsible for
if the development ofoff—flavors in foods, -.thfeygma3¢alsoreaggwjth
; \ oWWmteins (Karel, 1 97 3). Protemzlipig
interactions can occur in some foods and feeds such as frozen and
dehydrated fish, fish meal and oilseeds, thus causing somedegreepf
gmjeinfiamage (Cheftel, 1979).Thus, usingLhQédSOIDQEéLQJeduce
eliminate or reduce offensive odors. The palatability of frozen fish
would be improved and the protein quality of the frozen fish could
be assured by preventing further reactions induced by presence of
the off-flavor volatiles.
The objectives of this study were:
1. To identify the off-flavor volatiles in the headspace of frozen fish
using GC/ MS analysis.
2. To evaluate the capacity of six adsorbents to reduce off-flavor
volatiles in sample containers.
3. To select the most promising adsorbent and to determine its

performance in a package containing frozen fish.

LITERATURE REVIEW

' i ' i h

Fish lipids are characterized as highly unsaturated lipids due to
their high concentrations of polyunsaturated fatty acids (PUFAs),
such as 20:5 n—3 (5,8,11,14,17-eicosapentaenoic acid) and 22:6 n-3
(4,7,10,13,16,19-docosahexaenoic acid). PUFAs containing 2 or more
double bonds are susceptible to oxidation (Olcott, 1962).

The composition of fish lipids is different from most other
naturally occurring oils and fats in :( 1) possessing larger quantities of
fatty acids with chain lengths exceeding 18 carbons, (2) containing a
much greater proportion of highly unsaturated fatty acids, and (3)
having polyunsaturaturates, primarily at the (0-3 rather than (0-6
position (Flick et al., 1992). In addition, the amounts and character of
the fatty acids vary with the different organs and parts of the fish
(Flick et al., 1992).

Li 1 Oxi ' in Fi h
Lipid oxidation is one of the most important factors responsible for
quality loss of fish during refrigerated and frozen storage. For
example, lipid oxidation causes changes in the flavor, color, and
texture of fish (Khayat and Schwall, 1983). lipid oxidation in fish

may be nonenzymatic, such as autoxidation and photosensitized

 

oxidation, or initiated by enzymes, such as lipoxygenases,
peroxidases and microsomal enzymes (Hsieh and Kinsella, 1986).
Lipid oxidation involves the chemical breakdown of fatty acids
in the presence of oxygen. It is a rather complex process, where
unsaturated fatty acids react with molecular oxygen via a free
radical chain mechanism to form fatty acyl hydroperoxides, generally
called peroxides. They are primary products of the oxidation process
(Gray, 1978). The hydroperoxides formed are very unstable and can
breakdown to form free radicals, which in turn can accelerate the
rate of lipid oxidation (Flick et al., 1992). The hydroperoxides may
also undergo carbon-carbon cleavage to produce volatile compounds
such as aldehydes, ketones and hydrocarbons. The free radical chain
reaction is self-generating. It continues even at freezing
temperatures, especially in fish with high fat content, and is not

especially affected by antioxidants (Flick et al., 1992).

We

low temperature storage can slow down enzyme activity and
inhibit microbial growth in fish. The loss of quality of fish prior to
freezing is due to one or all of several factors, such as physical
damage, bacterial spoilage or autolytic breakdown of the flesh
adjacent to the abdominal cavity. The main factor limiting the frozen
storage life of herring was the hydrolysis of lipids during frozen
storage (Bilinski et al., 1978). The hydrolysis of lipids in herring
during chilled or frozen storage lead to the formation of free fatty
acids which are present due to enzymatic degradation of

triglycerides and phospholipids (Bosund and Ganrot, 1969).

 

 

5
Many investigators have studied changes in the lipid fractions

in fish muscle during frozen storage and found that the increases in
free fatty acid content in fish during cold storage are primarily due
to the hydrolysis of the phospholipids which are catalyzed by
phospholipases in thall et a1. 1950; Grieg, 1968a,b). Bligh and Scott
(1966) found an increase of the free fatty acid content from 5 to 325
mg/ 100g flesh in cod muscle at -120C for nine months and reported
that this was due to the hydrolysis of phosphatidylethanolamine(PE)
and phosphatidylcholine(PC). They also found that PE hydrolysis
ceased after storage for 4 months, whereas PC hydrolysis continued
at a slower rate. Furthermore, Bosund and Ganrot (1969) studied the
formation of free fatty acids in herring and found that the increase
was primarily due to hydrolysis of lecithin, cephalin and to a lesser
extent other triglycerides. The free fatty acids formed in the dark
and white muscle as a result of phospholipid hydrolysis were 45%
and 75% of the free fatty acid formed in the dark and white muscle
respectively (Bosund and Ganrot, 1969a). In addition, the formation
of glyceryl-phosphorycholine and choline in cold stored rainbow
trout muscle is evidence for enzymatic hydrolysis of phospholipids in
fish during cold storage (Jonas and Bilinski, 1967). Nair et a1. (1979)
and Mai and Kinsella (1979) also demonstrated that the formation of
free fatty acids in frozen fish is associated with hydrolysis of
phospholipids. They found no significant changes in triglycerides and
unsaponifiable matter in fish during frozen storage at -180C.

Free fatty acids can cause denaturation of proteins in fish flesh
(Lovern et al., 1959; Olley and Lovem, 1960; Bligh, 1961; Jonas and
Tomlinson, 1962). In addition, the toughened texture, poor flavor,

 

 

 

6
and unappealing odor in poorly stored frozen seafoods have been

attributed to the binding of oxidized unsaturated lipids to proteins, a
process by which insoluble lipid-protein complexes are formed
(Khayat and Schwall, 1983).

Off-flavers Developments in Fish due [,9 lipid Oxidatien

The quality and shelflife of fish are greatly influenced by the
development of oxidative flavors (Josephson et al., 1987). Josephson
and Lindsay (1986) indicated that certain volatile compounds
responsible for the aroma of freshly harvested fish are the result of
the degradation of site-specific (i.e. enzymically formed)
hydroperoxides. These enzyme-mediated degradations are expressed
in the formation of a variety of alcohols and carbonyls which are
present in the highest concentrations in the skin and slime fractions
of fish (Josephson et al., 1987).

Volatile aldehydes have long been believed to contribute
strongly to the characteristic aroma of oxidized fish lipids (Lea, 1953;
Yu et al., 1961; Aitken and Connell, 1979; Ikeda, 1980). These
oxidized flavors have been frequently described as rancid, cod-liver-
oil-like, and painty (Yu et al., 1961). McGill et al. (1977) reported
that 4-heptenal, and 2,4-heptadienal, increase during cold storage of
cod. These compounds apparently result from the oxidation of n—3
polyunsaturated fatty acids which are released from phopholipids
(Ross and Love, 1979).

It has been reported that aldehydes with an n-3 double bond
are the critical components of oxidative off-flavors because of their
low flavor threshold values (Frankel, 1984). Fish lipids, which are

7
rich in n-3 polyunsaturated fatty acids (PUFA) are very susceptive to

lipid oxidation. Once lipid oxidation is initiated, very low
concentrations of aldehydes with n-3 double bonds can cause
distinctive oxidative off-flavors (Hsieh and Kinsella, 1986).

Badings (1973) and Ke et a1. (1975) concluded that the odor of
cold-stored fish could not be attributed to an individual compound,
but instead must be caused by a complex mixture of carbonyl
compounds. This view has been supported by the observations of
Swoboda and Peers (1977) who found that solutions containing both
C7 and C10 - 2,4-dienals possessed fishy odors, and that individual
compounds eluting through an effluent splitter after gas
chromatograph did not exhibit fishy aromas.

WWW

Headspace analysis using gas chromatography (GC) is often
used to determine flavor compounds of foods. In this procedure,
gaseous volatiles associated with food are determined by GC. In
some instances, the concentrations of volatiles in the headspace are
too low to be detected directly by GC. Therefore, collection and
concentration of these volatile substances are necessary prior to final
GC analysis (Buckholz et al., 1980). In order to concentrate trace
aroma volatiles present in headspace gases over foods, beverages,
and environmental samples, adsorption of these volatiles onto a
variety of porous solids has become an established procedure for
preparing samples for analysis by gas chromatography and mass
spectrometry ( Nunez et al., 1984; McNally and Grob, 1985a, b ).
Novotny et al. (1974a) used an adsorption precolumn packed with

 

 

8
porous polymers for high resolution GC analysis of the volatile

constituents of body fluids. Murray (1977) used a technique for
concentration of headspace, airborne, and aqueous volatiles on a
porous polymer precolumn. Recently, Tenax GC (2,6-
diphenylparaphenylene oxide polymer) has emerged as a widely
used porous polymer for these applications. According to Butler and
Burke (1976), Tenax-GC works well for high boiling components due
to its high thermal stability and low retention volume. This stability
assures no bleed on GC columns during analysis and complete
regeneration of the porous polymer by heating the column to 2600C
under a purge of helium gas. The polymer containing adsorbed
headspace volatiles is stable for up to 5 days at 00C with no
decomposition of adsorbed volatiles (Butler and Burke, 1976).
Therefore, concentration of headspace volatiles can be achieved by
repeated adsorption of volatiles on the same trap (Mussinan, 1978).
Some applications of the general properties of adsorptive techniques
in the literatures include retention volumes for different components
(Kuo et al., 1977; Krost et al., 1982; Kawata et al., 1982) and the
effects of sampling times and purging rates on retention of volatiles
on Tenax GC (Buckholz et al., 1980).

While earlier studies have employed a variety of collection
devices and procedures, most have utilized thermal desorption in
association with transfer techniques for recovery of volatiles for
subsequent GC analysis (Bertsch et al., 1975; Murray, 1977; Buckholz
et al., 1980; Nitz and Julich, 1984).Simple Tenax-GC collection

procedures using readily available materials, e.g. ethyl ether, to elute

 

9
adsorbed volatiles have been developed and evaluated (Olafsdottir et

al., 1985).

‘91 I' -. 11:10! 0 I: ‘1"L‘ ‘1. 111' I; of
V l ' C 11

Food flavor consists of many volatile compounds in very small
quantities. The aroma of a food is dependent on the vapor pressure
of the volatile compounds and interaction of these volatile
compounds with the macromolecules which constitute the food
matrix. Therefore, the isolation and identification of volatile
compounds which contribute to the flavor of a food are difficult
(Drumm, 1988).

Isolation of volatile compounds from the food matrix is a
critical step in their identification. Two approaches have been
directed in the analysis of volatile compounds isolated from foods.
First is total volatile analysis which determines the qualitative and
quantitative composition of all volatile compounds that can be
isolated from a food. Distillation and extraction are the primary
methods used in total volatile analysis (Sugisawa, 1981). Direct vapor
analysis is the second technique used to sample the actual aroma in
equilibrium with the food and is representative of the aroma smelled
(Weurman, 1969).

lsnlat'nznntL/Qlanlefimnnoimds
Isolation of volatile compounds from a fatty sample is one of

the most difficult analytic procedures. Direct extraction with an

organic solvent is practically impossible because most organic

 

10
solvents dissolve fatty materials. Thus, steam distillation is the most

common method of separating organic compounds from fatty
materials. A simultaneous steam distillation-solvent extraction
apparatus was later applied to analysis of volatiles from fat by
several researchers (Buttery et al., 1977; Ohnishi and Shibamoto,
1984). In contrast to steam distillation methods, the extracting
solvent is continuously recycled during the distillation procedure.
Therefore, only small quantities of solvent are needed for the
extraction of volatile compounds from large quantities of food. In this
method, the volatiles are extracted from the condensed steam
distillate by a low-boiling water immiscible organic solvent in a
common condensation region. The aqueous and organic phases are
continually returned to their respective flasks during the procedure
(Likens and Nickerson, 1964). The likens-Nickerson procedure and
other methods which involve distillation at atmospheric pressure,
should be avoided when characterizing the flavor constituents of
fresh, uncooked food products. The application of heat results in the
formation of artifacts which are typical of the processed, rather than
the raw product (Weurman, 1969). Extraction of flavor compounds
using Likens-Nickerson apparatus may also result in artifact
formation via lipid oxidation (McGill and Hardy, 1977; Lovegren et
al., 1979).

The development of porous polymer traps such as Tenax GC (p-
2,6-diphenyl-p-phenylene oxide) for trapping volatiles has allowed
successful collection of volatiles. Purge and trap techniques are
widely employed. Basically, the sample is heated in a glass tube

through which a stream of carrier gas flows to carry the volatiles

 

 

1 1
onto a Tenax trap. The volatiles are then thermally desorbed from

the Tenax and swept directly into the GC injection port. This method
has been successfully used by Shahidi et al. (1987) for the hexanal
analysis of cooked ground pork and by Selke and Frankel (1987) for
the analysis of soybean oil volatiles. However, one weakness of this
method is that some materials are heat labile. This may result in
some formation of a more complex mixture at the temperatures
(1200C or higher) used to drive the volatiles onto the GC column
(Waltking and Goetz, 1983).

A third method involves collection of volatiles on Tenax GC
traps by either a purge and trap method (Olafsdottir et al., 1985;
Barbut et al., 1985) or a vacuum extraction procedure (Vercellotti et
al., 1987). Following collection, the volatiles are eluted with diethyl
ether and injected onto a GC column to resolve the volatiles. Using
this method, Barbut et a1. (1985) identified hexanal as the volatile
compound primarily responsible for the aroma of oxidizing turkey

sausage.

In'iianl'lCm ns

Techniques that combine gas chromatography(GC) and mass
spectrometry(MS) have found wide application in flavor research
(Maarse, 1991). The GC—MS combination is by far the most popular
technique for the identification of volatile compounds in foods and
beverages. Many reviews have been published on this subject.
Crawford et a1. (1976) used GC-MS techniques to identify volatiles
from extracted commercial tuna oil. There were 64 compounds

identified in the oxygenated fraction and 62 in the hydrocarbon

12
fraction. Volatile compounds collected on Tenax GC at room

temperature (2 10C) from direct headspace purging of fresh uncooked
Whitefish (Coregon us clupeaformis) were separated by gas
chromatography and identified by mass spectrometry (Josephson et
al., 1983). Two distinct families of compounds associated with
cucumber and melon fruits, and mushrooms were identified as
principal contributors to fresh Whitefish aroma. Josephson et a1.
(1984) used the same techniques to identify the enzymically derived
volatile aroma compounds in saltwater and freshwater fish. Volatile
compounds collected on Tenax GC at 2 10C from oxidized Whitefish
were separated by fused silica capillary gas chromatography and
identified by mass spectrometry (Josephson, 1984). Galt and
MacLeod (1984) used a modified headspace sampling procedure
involving adsorption of the headspace vapors onto Tenax GC. Rapid
heat desorption transferred the volatiles directly into a gas
chromatography column for separation. By use of combined gas
chromatography/ mass spectrometry, 67 identifications were made,
including 8 compounds tentatively identified for the first time in

cooked beef aroma.

Eneyentien Qf fo-flayers Develepment in Fish during Sterage

Although off-flavor development has been recognized as a
major problem in frozen fish, several methods can be employed to
control or minimize this problem. For example, low temperature
storage, application of antioxidant, and packaging technique have

been researched and can be used quite successfully.

 

 

 

13
In this section, the effect of storage temperature on the quality

of frozen fish, antioxidant activity and application to fish products,
and the packaging requirements for frozen fish will be reviewed. The
role of adsorbents, in control of off-flavor development in frozen fish

during storage, will be described.

Effth Qf Sthage Temperature en the Quality Qf Frpzen Fish

The effect of storage temperature on the quality of frozen fish
has been intensively studied by many researchers (Young, 195 O;
Dyer and Morton, 1956; Peters and McLane, 1959). Their results
indicate that product shelf life is extended by lowering the storage
temperature to -l80C or below. Temperature fluctuation of —10 to
-40C has been reported by Lentz and Rooke (1960) in frozen fish
shipped by road in refrigerated trailers. A survey conducted by Lane
(1966) indicated that some frozen seafoods in retailers' freezer
cabinets reached —40C mainly due to the overloading of products in
the cabinets. Dyer (1959) pointed out that storage temperature
fluctuation above -180C caused quality deterioration and shortened
the storage life of cod fillets. Palmateer et a1. ( 1960) also found an
increase in degree of lipid oxidation caused by temperature

fluctuation in frozen rockfish fillets.

ThA ivi A 1i 'n fAn'xi inFihPr s
Freezing technology provides consumers with products that

have reasonably good acceptance. However, rancid flavor has been

considered the major factor responsible for the quality loss in frozen

mullet fillets (Saenz and Dubrow, 1959; Beaumariage et al., 1969).

 

14
Attempts have been made to stabilize seafood products with

antioxidants such as ascorbic acid (Andersson and Danielson, 1961),
tocopherols (Brown et al., 1957), and butylated hydroxyanisole-
butylated hydroxytoluene (BHA-BHT) combinations (Yu et al., 1969).
However, none of these antioxidant treatments have been found to
be effective in preventing the development of oxidative rancidity.
Stuckey (1968) reported that application of phenolic antioxidants in
frozen fish fillets is not effective in retarding oxidative rancidity.
This may be attributed to either inadequate distribution of these
water—insoluble antioxidants in the fish samples or because the
antioxidants may simply have lacked sufficient potency in the
particular application (Sweet, 1973).

However, Sweet (1973) found that tertiary butylhydroquinone
(TBHQ) has significant antioxidant effects in salmon. Due to the high
metal ion content in marine fish, he also evaluated the effectiveness
of chelating agents (citric acid and EDTA) on stability of fish lipids. He
reported that EDTA is a more effective chelator than citric acid and is
also effective even when no phenolic antioxidant is present.
However, the most potent inhibitors are combinations of EDTA or
citric acid and TBHQ or BHA. The most effective stabilizer
combination in salmon is TBHQand citric acid (Sweet, 1973).

A number of studies have also been conducted to evaluate the
effectiveness of ascorbic acid as antioxidant in fish products
(Bauernfeind et a1. 1951; Marcuse, 1954; Lilzemark, 1964; Greig,
1967; Liu and Watts, 1970; Love and Pearson, 1974; Deng et al.,
1977).The results of these studies vary considerably and are

inconclusive. Some reported that addition of ascorbic acid did not

15
improve stability of lipids and actually acted as a prooxidant. The

nature of the flesh products in which ascorbic acid is incorporated
may have direct bearing on the results because of their different
biochemical components and other inherent environmental effects
(Deng et al., 1978).

Phosphates are also widely used to assist in the preservation of
minced fish blocks although oxidative rancidity is controlled more
effectively with BHA, BHT and other phenolic antioxidants. Excessive
use of phosphates and salt in fish can lead to rubbery texture during

prolonged storage (Kirk and Sawyer, 1991).

f r Fr 2 Pi h

The materials and styles of packaging used for frozen fish
(products) are numerous and vary from flexible plastic polybag for
retail sale products to waxed paper or polyethylene-lined board for
industrial blocks of fillets. Their efficacy depends on their degree of
irnperviousness to water vapor at low temperatures and the degree
to which the product is closely and completely enclosed (Connell,
1990).

A great deal of fish is sold in the frozen food section of
supermarket. The factors affecting the quality of frozen fish are : (1)
moisture loss-dehydration, (2) oxidation, (3) rancidity, (4) change in
odor and flavor, (5) loss of volatile flavors, (6) enzymatic activity,
and (7) loss of vitamins (Sacharow and Griffin, 1980). Proper
packaging for frozen fish is essential, because efficient packaging will
help to offset the detrimental effects of oxygen and of desiccation
(Sacharow and Griffin, 1980). Therefore, the package must be

 

16
impermeable to water vapor. In addition, there should be no airspace

between the product and the package; otherwise the product will
undergo dehydration at the site of any airspace, regardless of the
irnpermeability of the packaging material to water vapor.
Furthermore, the package should also be impermeable to oxygen in
order to prevent rancidity.

Gas impermeable packaging for fresh fish should be used only
in systems where there is no chance that the product will be exposed
to temperature above 4.40C (400F). The reason for this concern lies
in the danger of an outgrowth of Mum mmljntmn, a genus of
relatively ubiquitous bacteria. These bacteria grow in the absence of
oxygen and produce a toxin that cause botulism, a potentially fatal
disease. Since quality assurance requires a commitment to keeping
the product at 00C (320F), there is no reservation in a quality
assurance program about recommending the use of gas-impermeable
packaging (Gorga, 1988).

Wax-coated cartons were the first retail packages used for
frozen fish. Problems were encountered with wax flaking off and
with dehydration through the score lines on the package. The arrival
of polyethylene-coated paperboard offered good flexibility and
better moisture protection. Although this was an improvement over
wax cartons, problems still remained in heat sealing, ink adhesion
and delarnination when in contact with fish juices.

Most cartons are coated with petroleum wax-resin blends (hot
melts). If the hot melt is used on the inside of the carton, a high gloss

is left on the product surface. This offers pleasing aesthetics. The hot

 

17
melts are easily heat—sealed. Cartons may incorporate tear strip

openings or other convenience features.

The most popular package for fresh refrigerated fish consists of
a shallow tray and transparent film overwrap. The tray may be
fabricated from molded pulp, foam polystyrene or clear polystyrene.
Foam and clear polystyrene require the use of absorbent blotters.
The overwrap is usually polyvinyl chloride (Sacharow and Griffin,
1980).

Packaging in oxygen-impermeable flexible films offers practical
extension of storage life but the oxygen in the package must be
removed before storage either by evacuating the space between film
and product or by replacing the air with an inert gas like nitrogen.
Vacuum packaging is found to be more practical, though it is only
applied to expensive, products, like shrimp and trout. The most
suitable material must maintain an excellent oxygen barrier.
Laminated structures such as cellophane-aluminum foil-polyethylene
or polyester-PVDC-polyethylene are desirable materials. Rancidity
formation decreases drastically in vacuum packs employing low
oxygen permeable laminates. Overall shelf-life is extended by almost
100%.

Since vacuum packaging does not remove all oxygen from the
fish, lipid oxidation still occurs, but at a much slower rate. Proper
freezing storage is essential for all packaged fish, whether it be
vacuum-packed or overwrapped with a plastic film(Sacharow and
Griffin,1980).

 

 

 

18
A rnHvinhC ili fA rinOff-flvr

In recent years, adsorption polymers have been used for
collection, and concentration, with subsequent gas chromatography
(GC) analyses in a wide variety of applications (Buckholz, Jr. et al.,
1980). Novotny et al. (1974a) used a porous polymer adsorption
precolumn for high-resolution GC analysis of the volatile constituents
of body fluids. Murray (1977) described a technique for
concentrating headspace, airborne, and aqueous volatiles on a porous
polymer precolumn, and Zeldes and Horton (1978) used Tenax GC
adsorption polymer to trap volatiles in cigarette smoke. The use of
adsorption polymers has become widespread in the testing of air
pollutants (Murray, 1977). Buckholz, Jr. et a1. (1980) reported the
application and characteristics of polymer adsorption method used to
analyze flavor volatiles from peanuts. Josephson et al. (1984) used
the adsorption polymer to identify volatile aroma compounds from
oxidized frozen Whitefish. Galt and MacLeod (1984) optimized the
technique for isolating genuine cooked beef aroma volatiles, using a
modified headspace sampling procedure involving adsorption of the
headspace vapors onto Tenax GC.

In this section, the adsorption capability of six adsorbents
(silica gel, activated carbon, sodium bisulfite, cellulose, pectin, and
lactose) are reviewed. These adsorbents will be discussed with

regards to their ability to adsorb volatile organics.

Silica Gel
Silica gel is a partially dehydrated form of polymeric colloidal

silicic acid. The chemical composition can be expressed as SiOz-nHzO.

 

19
The water content, which is present mainly in the form of chemically

bound hydroxyl groups, amounts to about 5% (wt./.wt.) typically. It
is a microporous structure in which the pore size is determined
mainly by the size of the original micro particles. Bond formation
between adjacent particles occurs with elimination of water between
neighboring hydroxyl groups and the final structure is therefore
physically robust (Ruthven, 1984).

The presence of hydroxyl groups imparts a degree of polarity
to the surface so that molecules such as water, alcohols, phenols,
amines (which can form hydrogen bonds) and unsaturated
hydrocarbons (which can form p-complexes) are adsorbed in
preference to nonpolar molecules such as saturated hydrocarbons
(Ruthven, 1984). Because of its selectivity for aromatics, silica gel
was used as the adsorbent in the Arosorb process for separation of
aromatics from paraffins and naphthenes.

Shantha and Ackman (1991) used silica gel to adsorb longer-
chain polyunsaturated fatty acids from food and marine lipids for the
concentration of these longer-chain polyunsaturated fatty acids. This
method did permit a rapid concentration of polyunsaturates from
most fish oil and animal lipid samples. It also could be useful for
rapid screening of various oils and/ or lipids from mixed food
products, and especially those containing fish products, for the
presence of the C20 and C22 "omega-3" polyunsaturated fatty acids
(Ackman and Ratnayake, 1989b). The information could eventually
help in predicting the oxidative stability of food fats and thus their

potential for off-flavor development (Peers and Coxon, 1986).

20
ActixatedCaan

Activated carbon is normally made by thermal decomposition
of carbonaceous material followed by activation with steam or
carbon dioxide at elevated temperature (700-1 1000C) (Ponec et al.,
1974). The activation process involves the removal of carbonization
products formed during pyrolysis, thereby opening the pores of
activated carbon (Ruthven, 1984).

The structure of activated carbon consists of elementary
microcrystallites of graphite, these microcrystallites are stacked
together in random orientation and it is the spaces between the
crystals which form the micropores (Ruthven, 1984).

The surface of carbon is essentially nonpolar although a slight
polarity may arise from surface oxidation. As a result, carbon
adsorbents tend to be hydrophobic and organophilic. They are
therefore widely used for the adsorption of organics in decolorizing
sugar, water purification, and solvent recovery systems as well as for
the adsorption of gasoline vapors in automobiles, and as a general
purpose adsorbent in range hoods and other air purification systems
(Ruthven, 1984).

In order to recover volatile components from cigarettes during
tobacco-roasting, Matsukura et al. (1984) used activated carbon to
adsorb volatiles from roasted tobacco and desorptive recovery by
ether extraction.

Sakaki et al. (1984) used activated carbon to collect the
headspce volatiles swept by helium from cut tobacco and analyzed
the volatiles by gas chromatography after desorption by extracting

the activated carbon with dichloromethane.

2 1
Peanut protein products contain phenolic compounds

(Conkerton et al., 1983; Daigle et al., 1983) which have been
implicated in off-flavor and color defects (Maga, 1978; Sosulski,
1979). Ion exchange and activated carbon treatments have been
successfully used to remove phytic acid and phenolic compounds
from soy protein extracts (Brooks and Morr, 1982; How and Morr,
1982). Seo and Morr (1985) used these procedures to remove
phytate and phenolics from laboratory and commercial peanut
proteins.

Tausig and Drake (195 9) incorporated packets of active carbon
within containers to adsorb objectionable volatiles of radiation-
sterilized beef during storage. A panel of experts judged the intensity
of irradiation flavor of cooked beef to be significantly lowered by
this carbon treatment (Tausig and Drake, 1959).

A study was made of the use of activated carbon with
irradiated cooked pork, chicken, and beef during storage by Gernon
et al. (1961). The meat products were evaluated by a consumer-type
panel for the combined characteristics of flavor, appearance, and
texture, rather than just the single parameter of irradiation flavor.

Toro-Vazquez et al. (1991) evaluated an adsorption system
using activated carbon, and the effect of adsorption time,
temperature, and concentration of H20 in the adsorption system on
the chemical characteristics of squash (Cucurbita moschata) seed 011.

An ultrafiltration-activated carbon system has been developed
to renovate contact refrigeration brines, such as used on fishing
vessels, before reuse and to salvage usable byproducts (Welsh and
Zall, 1984).

 

22
Spflium Bisulf'te

Aqueous sodium bisulfite has been evaluated as an adsorbent
for volatile chemicals in the headspace above heated pork fat
(Yasuhara and Shibamoto, 1989). The major volatile compounds
produced from the heated pork fat were hexanal, heptanal, and
pentanal (Yasuhara and Shibamoto, 1989).

Volatile carbonyls have been found to be important in fresh
fish aroma (Berra et al., 1982; Josephson et al., 1983a). Thus, the
aromas of both fresh and oxidized fish should be susceptible to
chemical modification by bisulfite which adds to aldehydes and
many ketones to form nonvolatile bisulfite-addition products
(Morrison and Boyd, 1978).

Cellulose

Cellulose is a polymer of glucose connected by [31-4 glucosidic
bonds. The polymers are laid down naturally in crystalline areas
which vary in size and are held in lateral associations by hydrogen
bonds. Between crystalline areas are non-crystalline (amorphous)
areas which can be hydrolyzable by acids and enzymes (Glicksman,
1969). Some researchers (Gilkes et al., 1992; Geluk et al., 1992 ; Jach
and Sugier, 1983; Laidsaar et al., 1981) have reported that cellulose
could be used to adsorb enzymes such as cellulase, lipase, and
glucoamylase and to immobilize the enzymes as the catalyst for
several bioconversions.

Saito et al. (1992) used cellulose to adsorb carthamine (a red
colorant) for its isolation and partial purification. Cellulose was used

as a adsorbent for quantitative headspace analysis of volatile organic

23
compounds (a-pinene, pyridine, ethyl acetate, and methyl ethyl

ketone) to investigate the nature of flavor interactions with food
ingredients (Saleeb and Pickup, 1978). Gupta et al. (1979) examined

the capacity of cellulose to adsorb diacetyl from beer.

1.3m

Lactose has served as an extender for spices and volatile
aromas (Reger, 1958). At one time it was used in the manufacture of
instant coffee to adsorb volatiles during roasting and drying. When
this lactose was incorporated into the coffee powder the flavor was
improved (Nickerson, 1979). Maier (1969, 1970, 1972) studied
binding of various vapors onto lactose, pectin, and ovalbumin. Knapp
(1973) studied a hydrocolloid-stabilized lactose-flavor system, and
Yabumoto et al. (1975) have investigated the adsorption and rate of
desorption of banana volatiles on lactose. The strengths of adsorption
of numerous small organic molecules on anhydrous oc— lactose have
been measured by McMullin et al. (1975). Adsorptive behaviors of
several low molecular weight compounds (commonly associated with
food flavor) onto stable anhydrous a—lactose were studied by Marvin
et al. (1979) and they found that aromatic hydrocarbons (toluene
and ethylbenzene) were adsorbed the least, while alcohols (1-
pentanol and l-octanol) were adsorbed in the greatest amounts;
esters and ketones showed intermediate adsorption. Nickerson
(1979) found that anhydrous lactose could adsorb a large variety of
volatile compounds. Traps containing lactose as an adsorbent were
found to be a very simple method of collecting volatile compounds

from gas chromatographic effluents for organoleptic evaluation

 

24
(Gramshaw, 1976). McMullin et al. (1975) reported that a wide

variety of organic flavor compounds including esters, aldehydes,
ketones, alcohols and hydrocarbons, were adsorbed onto stable
anhydrous oc-lactose. Nickerson and Dolby (1971) examined and
compared the adsorption of diacetyl on various forms of lactose and
other carbohydrates and concluded that for lactose the greatest

adsorption occured on anhydrous 0t lactose.

Pectin

Pectin is composed primarily of D-galacturonic acid, although it
can have other carbohydrate moieties linked to it. Partial
methylation of the carboxyl groups on the galacturonic acids imparts
important properties to pectic substances. Pectin was found to be
capable of adsorbing organic molecules, including bile acids,
cholesterol, and toxic compounds (Schneeman, 1986).

Guichard et al. (1991) studied the influence of the amount of
pectin added on strawberry jam, and the degree of esterification as
well as molecular weight of that pectin on sensory characteristics,

and on amounts of volatile compounds in the package headspace.

S n Ev
Sensory analysis is the scientific discipline that evokes,
measures, analyzes, and interprets reactions to product
characteristics through the senses of sight, smell, taste, touch, and
hearing (Stone and Sidel, 1985). The role of sensory evaluation is to
provide valid and reliable information to research & development,

production and marketing in order for management to make sound

 

25
business decisions about the perceived sensory properties of

products. The ultimate goal of any sensory program should be to find
the most cost-effective and efficient method with which to obtain the
most sensory information (Meilgaard et al., 1987).

The sensory attributes of a food item are: appearance,
odor/ aroma/ fragrance, consistency and texture, and flavor
(aromatics, chemical feeling, taste) (Meilgaard et al., 1987).

In order to measure true product difference three major
variables must be controlled. The first, test controls: the test room
environment, the use of booths or a round table, the lighting, the
room air, the preparation area, the entry and the exit areas. The
second, product controls: the equipment used, the way samples are
screened, prepared, numbered, coded and served. The third, panel
controls: the procedure to be used by a panelist evaluating the
sample in questions (Meilgaard et al, 1987).

Factors influencing sensory verdicts are physiological factors
(adaptation, enhancement or suppression), psychological factors
(expectation error, error of habitation, stimulus error, logical error,
halo effect, order of presentation of samples, mutual suggestion, lack
of motivation, and capriciousness vs. timidity), and poor physical
condition (Meilgaard et al, 1987).

The first quality judgement made by a consumer about a fish
product at the point of sale is based on its appearance. The next
judgement is about the product's texture. This is followed by an odor
judgement, and finally by taste to judge flavor (Stone and Sidel,
1985).

26
Connell and Howgate (197 1) evaluated the flavor/ texture

impact on the acceptance of cod and haddock fillets over a wide
range of freshness and concluded that flavor was a more important
criterion of quality than texture. Similarly, Hamilton and Bennett
(1983, 1984), in studies using different species of white fish,
concluded that flavor was the most significant positive determinant
of acceptability. Laslett and Bremner (1979) conducted a series of
storage trials with minced flesh from Australian fish species and
concluded that off-flavor, off-aroma and flavor were just as
important as texture variables in determining the acceptability of
frozen fish products. Kelly (1969) reported that the acceptability of
frozen cod was affected more by the development of cold storage
flavors than by textural changes.

Human perceptions of food are complex and may be influenced
significantly not only by the sensory characteristics of the food
material, but also by the personal experience of the observer such as
his or her physiological/ nutritional state, psychological state, ethnic
origin, educational level, economic level, religious convictions,
social/ cultural pressures, and so on. Apparently "normal" individuals
may differ substantially in their internal anatomy, and in various
biochemical and physiological functions (Williams, 197 1). Such
differences can have significant impact on the individual's food
behavioral patterns. How the brain integrates all of these influences,
sensory and others, determines a person's response
(acceptance/rejection; pleasantness/ unpleasantness) to the food.

Acceptability of foods, the degree of liking and disliking, is

usually estimated using a scalar method, the most common one being

 

27
the nine-point structured hedonic scale(Peryam 1952). Validity of

data generated using this method can be influenced by factors (in
addition to those already mentioned) such as unequal size category
intervals in the scale, the tendency of panelists to avoid extreme
values on the scale and to score close to the midpoint, the presence of
restrictive end points on the scale, and so on. Nevertheless, this
method, or variations of it, is probably the most widely used
technique for estimating hedonic quality of foods.

A method commonly used for consumer preference testing is
the two sample or paired comparison test. This is a very simple test
method that is easily understood by the participant. In this test, a
participant is asked only to indicate which one of the two samples he
or she prefers. Data generated using this procedure, however, yield
no information about the degree or intensity of the preference. If
information on the magnitude of acceptability, preference, and so on
is desired, some kind of scaling technique must be utilized.

Analytical testing to determine differences or similarities in
food products, to identify quality attributes, and to estimate relative
intensity of sensory attributes is usually accomplished in a
laboratory setting, under controlled conditions, with a trained panel
of subjects. There are numerous reviews in the literature that
describe in depth aspects of physical facilities, sample handling,
selection and training of panelists, data analysis, and so on(Stahl and
Einstein, 1973; Larmond, 1977; Amerine et al., 1965). Difference
tests are designed to identify samples that differ in terms of a
sensory characteristic. The attribute must be defined in the single

sample and in two samples (paired comparison) test methods, but

28
need not be defined with the duo-trio (matching one of two coded

samples with a labeled reference sample) and triangle (two samples
identical, one different) methods. These are commonly used
"difference" test methods. The information they yield is limited and
if intercomparison of several samples is needed, an extensive amount
of testing may be required, especially if tests are replicated to check
on reliability of panel performance. If sensory analytical tests are
properly conducted, with appropriate control of operational

variables, they can yield reliable and reproducible data.

MATERIALS AND METHODS

Materials

M

lake trout (Salvelinus namaycush) is classified as high oil-low
protein fish (Stansby and Olcott, 1963). Oil content in lake trout is
about 15%. lipid oxidation in lake trout is a serious problem to
quality of fish. Four (from the same batch) whole , uncooked, gutted
and gilled frozen lake trout were purchased from Superior Seafoods
Co., Grand Rapids, Michigan. At the time of purchase, the fish were
frozen and wrapped in polyethylene film and then packed in a
corrugated box. The fish were iced in a cooler and transferred back
to E. Lansing. It was about an hour and half in the cooler. The fish
with the wrap were then removed from the cooler and stored in the
freezer at the Food Lab , Department of Food Science and Human
Nutrition. One was prepared for identification of volatile aroma
compounds, three fish were used for evaluation of the selected
adsorbents, and the other one was cut into small cubes for sensory

evaluation.

Adsorbents
1. Silica gel, ACS reagent*, blue-indicating, Aldrich Chemical

Company, Inc. (Milwaukee, Wisconsin).

29

 

3O
2. Activated Carbon, Darco® 20-40 mesh, granular, general-use

carbon from coal raw materials, Aldrich Chemical Company,
Inc. (Milwaukee, Wisconsin).

3. Sodium bisulfite, ACS reagent, granular, analytical reagent,
Mallinckrodt, Inc.(Paris, Kentucky).

4. Cellulose, powder ~20micron average particle size, Aldrich
Chemical Company, Inc.(Milwaukee, Wisconsin).

5. Pectin, analytical grade, Nutritional Biochemicals Corporation.
(Cleveland, Ohio).

6. Anhydrous a—lactose, analytical grade, Aldrich Chemical
Company, Inc.(Milwaukee, Wisconsin).

*ACS reagents meet specifications and test requirements set by

the American Chemical Society.

191mm

1. Butanal (Butyraldehyde), 99% purity, FW72.1 1, m.p.-960C,
d 0.800, b.p.750C.

2. Cycloheptatriene, 90% purity, FW92.14, cl 0.888,
b.p.l 16-1 170C.

3. Hexanal (Caproaldehyde, Hexaldehyde), 98% purity,
FW100.16, d 0.834, b.p.1310C.

4. Pentanal (Valeraldehyde), 99% purity, FW86.13,
d 0.810, b.p.1030C.

All volatile standards were purchased from Aldrich Chemical

Company (Milwaukee, Wisconsin).

3 1
Packaging Material th Adsprbents

The adsorbents were packed in Tyvek pouches (1.01n.x1.0in.)
which were sealed using the impulse heat sealer. The packaging
material, Tyvek, is made by du Pont de Nemours & Co. Tyvek is a low

oxygen barrier material.

i ° 'n r 1Mas shown in Figure 1.) (Pierce, Rockford, LL.)
The body of the valve was made of Teflon®2. The pressure seal

was made at the mouth of the cell using a system which consisted of
two rubber rings encased inside the base of the valve. Pressure was
exerted using a threaded ring. The ring was turned down on the
rubber rings which caused the base of the valve to bulge, making the
seal. It provided a pressure tight seal to at least 1.5 atmosphere. The
valve mechanism was created using a sliding cylinder which passed
through the axis of the valve at right angles. When the valve was
opened, a needle could be inserted through the valve's axis into the
diffusion/ adsorption cell. When the valve was closed, the hole in the
sliding cylinder was no longer in line with the valve axis and a
needle could not be pushed though it. To prevent gas from escaping
around the needle when it was inserted, a small silicone septum

(Pierce, Rockford,I.L.) was located above the sliding cylinder.

 

1Registered Trademark, Pierce, Rockford, IL.

2Registered Trademark, LE. du Pont de Nemours & Co.

32

Valve Opened

  
 

Needle-Seal Septa

Green Button Red Button

Valve Closed

Needle—Seal Septa

Green Button Red Button

 

Figure 1. Pierce Mininert valves: Operation is simple- just push the green
button to open, insert syringe needle and take sample, withdraw
needle, then push red button to colse. To change needle-seal septa,
simply push the old septum out with 126" diameter rod and push
a new cylindrical septum in. This is done with the valve colsed to

prevent exposure of contents.

33
Methodology

This project was designed to evaluate the capacity of six
chemicals, including silica gel, activated carbon, sodium bisulfite,
cellulose, pectin and lactose to adsorb off-flavors associated with
oxidized frozen fish. Three experiments were conducted to achieve
this. The first experiment was designed to identify the volatile
compounds of oxidized frozen fish. The next experiment was to
distinguish the ability of the six adsorbents using a model system
which incorporated the suspected off-flavor compounds identified
from experiment I with the six adsorbents. In experiment HI, the
selected adsorbent was sealed individually in Tyvek pouches and
stored with a sample of oxidized frozen fish in a glass jar. By
comparing the concentration of the volatile compounds in the
headspace, the most effective adsorbent among the six chemicals was
determined. Sensory evaluation was done to demonstrate the effect

of the selected adsorbent on the level of adsorption of off-odors.

II n'fi 'n fVl '1 Arm Cm n frm

O ' iz r z n Fi h

Extract from frozen lake trout was prepared by immersing a
whole (3 lbs), thawed, uncooked lake trout in 500 ml of a saturated
sodium sulfate solution, followed by agitating to recover most of the
slime layer in the extract. Several extracts were pooled to obtain a
large sample of the drip/slime mixture, and 100 ml subsamples were
taken for analysis using the bubble-purge configuration apparatus
(250 ml flask) as shown in Figure 2 (Olafsdottir, 1985). Headspace

volatiles from the extract were collected and concentrated by

34

Purified Nigrogen

l

 

 

 

 

 

 

 

 

/ N 5112115-
f fl
W/ \B A

 

 

 

 

 

«Q
\.‘

Figure 2. Glass apparatus for dynamic gas -purging
of headspace aroma volatiles onto Tenax-6C:
(A) Tenax-Gt: collection tube;
(B) Heat-shrinkable Teflon tubing,-
(C) Configuration of purge-tube for swept-

surface sampling;
(D) Configuration of purge-tube for bubble
sampling,- and

(13) Magnetic stirring bar.

E

 

35
purging the sample with nitrogen (100 ml / min for 3 hours) at 210C

onto Tenax GC as described by Steinke (1978).Capillary column GC in
conjunction with mass spectrometric analyses of volatiles in ethyl
ether extracts from Tenax GC traps was performed as reported by
Josephson et al., (1983a). Identification of compounds was based on
coincidence of mass spectral patterns of standards.
Mass Selective Detector (MSD), Hewlett-Packard 5970 MSD
Mass Spectroscopy, was used to identify the off-flavor compounds.
The conditions of MSD were as follows:
Detector: Picogram range
Sample compatibility: Mass range 10 - 800 amu (He/H2
excluded) Volatility stability: Max
interface T0 3200C
Sample introduction: Capillary direct from Hewlett
Packard 5890 GC (Option: open-split
interface)
Quadrapole mass filter
Solvent delay: 3.00 min.
Resulting voltage: 2000 e.v.
Ion source: 70 e.V.
Run parameter: Start time 3.00 min.
Low mass 40 amu
High mass 400 amu
Scan threshold 1000
a/d sample (211): 2
Scan/ sec 1.19

Temperature zone: Initial temperature 400C

36

Initial time 0 min.

Rate 50C/ min

Final temperature 600C

Final time 0 min.

Rate A 100C/ min

Final temperature A 2800C

Final time A 30 min.
Injection port temperature: 2500C

Transfer line temperature: 2500C

Hewlett-Packard 5890 GC: Carbowax 20M (Supelco Inc.,
Bellefonte, PA)
Fused silica capillary column
Column dimensions: 60mX0.3 1mm
Head pressure: 50 psi
Split vent flow: 1.00 ml/ min
Septum purge flow: 7 ml/ min

Exri n Di’ 'hinhA r'nC ili fSix
W

The adsorbing capacity of six perspective adsorbents: silica gel,
activated carbon, sodium bisulfite, pectin, lactose, and cellulose, were
investigated using the techniques developed by Oosting et al., (1984).

The volatile standards were butanal, cycloheptatriene,
pentanal, and hexanal which were the major off-flavor volatiles
identified from Experiment I. Essentially, 5 111 of the liquid volatile

standards were injected into a 125ml serum bottle fitted with a

37
Pierce Mininert ValveR as shown in Figure 3. The adsorbent (2.0gm),

enclosed in a Tyvek (by DuPont) pouch (1.0in.X1.01n.), was
suspended above the volatile standards. Samples of volatile
standards in the headspace were withdrawn through the Pierce
Mininert ValveR, and the uptake of the vaporized sample by the
adsorbents was monitored over several time intervals (24, 48, and

72 hours) at 210C. A 10 111 Hamilton liquid sampling syringe was
used to sample the volatile standards, and a 250ul Hamilton gas tight
sampling syringe was used to sample the headspace.

A 125ml serum bottle was used as the diffusion/ adsorption
cell. 5 ul of volatile standards were withdrawn and injected into the
diffusion/ adsorption cell. A Pierce Mininert ValveR was fitted to the
cap of the cell. To prevent gas from escaping around the needle when
it was inserted, a small silicone septum was located above the sliding
cylinder at the point where the needle was inserted.

The headspace of the diffusion/ adsorption cell was sampled
using a 250111 Hamilton gas tight syringe. The syringe was cleaned by
aspiration of acetone through the needle and heated at 700C for 30
minutes. Prior to sampling the vapor, the syringe was conditioned to
room temperature. The syringe needle was plunged through the
Pierce Mininert Valve® into the bottle. About 250111 of the headspace
was drawn into the syringe and expelled back into the bottle. Then
25111 of the vapor was withdrawn and quickly injected onto the gas
chromatographic column.

Analysis of headspace concentrations was conducted using a
gas chromatographic procedure. A Hewlett-Packard Model 5890

w/ split vent, equipped with dual flame ionization detectors and

 

36

Adsorbent

 

 

Tyvek Pouch

 

— 125 m1
Serum Bottle

 

 

 

 

 

 

L I’H‘I VOIatile
‘ fl Standard

 

Figure 3. Cross Sectional View of Diffusion/Adsorption Cell

 

 

39
interfaced to a Hewlett—Packard Model 3392A integrator was

employed. The gas chromatographic conditions were as follows:
Colurrm: DB-225 (J&W Scientific, Folsom, California)
Fused silica capillary column
Column dimensions: 30mX0.249mm
Temperature limits: 400C to 2200C
Column flow rate=0.785xD2L/tr=0.785x(0.25)2x30/ 1.87
=0.7871
Split Ratio=(4.26+0.7871)/0.7871=6.41
Temperature: Injection temperature 2000C
Detector temperature 2500C
Oven temperature 1500C
Temperature programmezlnitial temperature 400C
Initial time 5 min.
Rate 20C/min
Final temperature 480C
Final time 0 min.
Rate A 100C/ min
Final temperature A
2000C
Final Time A 40 min.
Integrator Model 3392A
Zero:0,-0.9
Attenuation: 2
Chart speed: 0.5 cm/ min
Peak width: 0.04
Threshold: 0

 

 

4O

EX!" ' ‘1 II ,_,.,nno hHSl ‘0. A OI‘n.‘ ' hFrozn
Fish System at Lew Temperature (50C)

Three whole, uncooked, frozen lake trout were thawed. Each of
the fish was ground into a large sample and 40 grams of the ground
fish sample were placed in a 450 ml glass jar with a screw cap which
had a sampling port on the top as shown in Figure 4 . Two grams
adsorbent, silica gel or activated carbon, were packed in a Tyvek
pouch (1.01n.x1.0in.) and suspended above the fish sample for three
days in the walk-in cooler (50C). A 250 111 gas tight syringe was used
to sample the headspace. 200 111 of the vapor in the headspace were
withdrawn and quickly injected onto the GC column.

Analysis of headspace concentrations was carried out using a
gas chromatographic procedure. A Hewlett-Packard Model 5890A gas
chromatograph, equipped with dual flame ionization detectors and
interfaced to a Hewlett—Packard Model 3392A integrator was
employed. The gas chromatographic conditions were as follows:

Column: SupelcowaxlO (Supelco Inc., Bellefonte, PA)

Fused silica capillary
Polar bonded stationary phase
Column dimensions 60mx0.25mmI.D.
Carrier gas: Helium at 27ml/ min
Temperature: Injection temperature 2000C
Detector temperature 2500C
Oven temperature 1500C
Temperature programmezlnitial temperature 400C

Initial time 1 min.

41

Sampling Port

   

Screw Cap —

Tyvek Pouch
w/Adsorbent

 

450ml jar —

 

. t . .

. .U} ~ —
w- e ...,
tinifi'Q‘fi
0! 090 why

‘4 gig-Fm

_ Fish Sample

 

 

 

 

 

Figure 4. Cross Sectional View of Diffusion/“Adsorption Jar

42
Rate SOC/min

Final temperature 2200C
Final time 10 min.
Integrator Model 3392A
Zero: 0, -l.1
Attenuation: 2
Chart speed: 0.5cm/ min
Peak width: 0.04
Threshold: 0
Retention time(RT) of Butanal: 2.08min
Pentanal: 3.09min
Cycloheptatriene: 4.55min
Hexanal: 5.3 2min
The standard calibration curves for butanal, pentanal,

cycloheptatriene, and hexanal are shown in Appendix B.

S Ev

A simple paired comparison test was used for sensory
evaluation. Frozen lake trout was cut into small cubes. 10 grams of
these cubes were placed in a 450ml glass jar which was previously
described in Section III. A Tyvek pouch (1.01n.x1.0in.) packed with
glass beads was suspended above the fish cubes in the glass jar as
the control.

The selected adsorbent, activated carbon, was packed in a
Tyvek pouch(1.01n.x1.0in.) and hung above the 10 grams of fish
cubes in the 450ml glass jar as the treatment.

 

43
Three pairs of samples (one control and one treatment as a

pair) were prepared for each panelist. Thus, totally 36 jars of
samples were prepared for six panelist. Also three pairs of 3-digit
random numbers were chosen to code the jars. The panelists had not
been informed the representation of the codes.

All samples were stored at 50C for 3 days and then transferred
to the Food lab in the School of Packaging where the testing sessions
were held. The Food Lab was equipped with standard indoor
fluorescent lighting and the temperature was at 2 10C.

Six panelists included faculty and graduate students from the
School of Packaging and Department of Food Science. All were semi-
trained in off-odor intensity and well experienced in sensory
evaluation.

During the test sessions (three sessions for each panelist), two
3-digit random number coded jars with the samples (one control and
on treatment) as well as the consent form and questionnaire (shown
in Appendix F) were presented to each panelist. They were asked to
follow the procedures listed in the questionnaire to sniff the samples
and identify the sample which had the stronger off-odor and to
indicate it on the questionnaire. 36 samples (three replications) had

been sniffed, a total of 18 judgments were made.

 

RESULTS AND DISCUSSION

Maier Qfl-flavpr Velatiles pf Qxigh’zed FLQzen lake Trent

Preliminary tests were conducted to identify the off—flavor
volatiles associated with oxidized frozen lake trout. The frozen lake
trout used in this study exhibited distinct oxidized aromas. Capillary
GC in conjunction with mass spectrometry was used to identify the
volatiles in ethyl ether extracts from Tenax GC at 2 10C and are
shown in Table 1. The mass spectra are shown in Appendix A.
Butanal, cycloheptatriene, pentanal, and hexanal were found to be
the major off-flavor volatiles. Volatile aldehydes have been believed
to contribute strongly to the characteristic aroma of oxidized fish
lipids (Lea, 1953; Yu et al., 1961; Aiken and Connell, 1979; Ikeda,
1980). Hexanal was found to be the most abundant volatile
compound in oxidized frozen Whitefish (Coregon us clupeafomris)
(Josephson et al.,1984). Hexanal has historically been recognized as a
principal compound formed in oxidizing lipid systems (Forss et al.,
1960; Yu et al., 1961; Badings, 1970; Matthews, 1971 ; Nobel and
Nawar, 1971; Chan et al., 1976; Warner et al., 1978; Selke et al., 1980;
Schieverle and Grosch, 1981; Frankel et al., 1982). Yu et al. (1960)
found that pentanal and butanal were major volatile compounds in
autoxidized salmon oil. Josephson et al. (1984) reported that
pentanal was a volatile aroma compound from oxidized frozen

Whitefish.
44

RESULTS AND DISCUSSION

MaienQflamerlatilesoLQxiszedfinmtlakelmt
Preliminary tests were conducted to identify the off-flavor

volatiles associated with oxidized frozen lake trout. The frozen lake
trout used in this study exhibited distinct oxidized aromas. Capillary
GC in conjunction with mass spectrometry was used to identify the
volatiles in ethyl ether extracts from Tenax GC at 2 10C and are
shown in Table 1. The mass spectra are shown in Appendix A.
Butanal, cycloheptatriene, pentanal, and hexanal were found to be
the major off-flavor volatiles. Volatile aldehydes have been believed
to contribute strongly to the characteristic aroma of oxidized fish
lipids (Lea, 1953; Yu et al., 1961; Aiken and Connell, 1979; Ikeda,
1980). Hexanal was found to be the most abundant volatile
compound in oxidized frozen Whitefish (Coregon as d upeaformis)
(Josephson et al.,1984). Hexanal has historically been recognized as a
principal compound formed in oxidizing lipid systems (Forss et al.,
1960; Yu et al., 1961; Badings, 1970; Matthews, 1971; Nobel and
Nawar, 1971; Chan et al., 1976; Warner et al., 1978; Selke et al., 1980;
Schieverle and Grosch, 1981; Frankel et al., 1982). Yu et al. (1960)
found that pentanal and butanal were major volatile compounds in
autoxidized salmon oil. Josephson et al. (1984) reported that
pentanal was a volatile aroma compound from oxidized frozen
Whitefish.

44

45

Table 1. Major Volatile Compounds Identified in Frozen Lake Trout

C 1 S E' .

Butanal G.C.1 M.S.2
Cycloheptatriene G.C. M.S.
Pentanal G.C. M.S.
Hexanal G.C. M.S.

 

1. G.C. confirmation was based on the retention time comparison to
the volatile standards.
2. MS. confirmation was based on comparison to the volatile

standards mass spectrum.

46

4.! III III I VI -. I‘ 1.10.1-0. I -1: 8'. AI Irv

Some investigators have found that silica gel was effective in
adsorbing volatile compounds (Davis et al., 1952; Ruthven, 1984;
Peers and Coxon, 1986; Ackman and Ratnayake, 1989b; Shantha and
Ackman, 1991). It has been reported that activated carbon has great
adsorption capacity for volatile organics (Tausig and Drake, 195 9;
Gernon et al., 1961; Matsukura et al., 1984; Ruthven, 1984; Sakaki et
al., 1984; Welsh and Zall, 1984; Seo and Morr, 1985; Toro-Vazquez et
al., 1991). Yasuhara and Shibamoto( 1989) used sodium bisulfite to
adsorb hexanal, heptanal, and pentanal. Some investigators have
reported that lactose has the ability to adsorb aromas ( Nickerson
and Dolby, 1971; McMullin et al., 1975; Marvin et al., 1979;
Nickerson, 1979; Ehler et al., 1979). Few studies have focused on
cellulose and pectin as a adsorbent to adsorb off-flavor compounds,
but they are food ingredients. Silica gel, activated carbon, and sodium
bisulfite do not have as wide approval to be incorporated into food.
Therefore, three food grade carbohydrates (cellulose, lactose, and
pectin), and three non-food grade chemicals (silica gel, activated
carbon, and sodium bisulfite) were chosen as adsorbents in this
preliminary report to determine the potential use of these six
compounds for adsorbing off-flavor volatile aromas in oxidized
frozen lake trout.

Analysis of the headspace in the serum bottle was done every
24 hours from time zero to 72 hours using GC. Area responses from
each GC analysis are shown in Appendix C (Table C-9).

The concentration of volatile standard in the headspace is

directly proportional to the area response. High area response means

47
high concentration, and thus adsorption of the adsorbent was less

effective. For each volatile standard, the highest recorded area
response is proportional to 100% relative concentration of the
volatile standard in the headspace. The data in Table G9 was

converted to the relative % concentrations as shown in Table 2.

E ' f ' r B

The effectiveness of six adsorbents on butanal was evaluated
every 24 hours up to 72 hours after storage at 210C and the results
were shown in Table 2.

After 24 hours storage at 210C with activated carbon as the
adsorbent, the relative % concentration of butanal in the vial
headspace was 0.39% which means that activated carbon adsorbed
most of the butanal in the vial headspace. Silica gel also adsored a
high level of butanal. As the results show, 1.74% butanal was
detected in the vial headspace. Sodium bisulfite did not demonstrate
as high adsorption capacity as did activated carbon and silica gel, but
it was considered a good adsorbent for butanal. Almost 70% of the
butanal was adsorbed by sodium bisulfite. Cellulose, lactose, and
pectin showed poor adsorption of butanal. On the second day (48
hours storage), activated carbon still had the highest level of
adsorption and the relative % concentration of butanal was reduced
from 0.3 9% to 0.19%. Silica gel also adsorbed butanal very well. The
relative % concentration of butanal was reduced from 1.74% to 0.66%,
while sodium bisulfite, cellulose, lactose, and pectin demonstrated
small reductions in the relative % concentration of butanal after 48

hours storage.

48

Table 2. Relative % Concentration of Volatile Standards in the Vial
Headspace after 24, 48, and 72 Hours Storage at 210C

Adsorbent—Him Butanal _y_c_o_entaC 1h - Pentanal Hexanal
(hrS) triene
24 1.74 0.23 ND* 58.12
Silica gel 48 0.66 0.16 ND 23.16
72 0.15 ND ND 8.66
24 0.39 ND ND 47.64
Activated 48 0.19 ND ND 12.11
carbon 72 0.10 ND ND 8.26
24 29.64 98.56 94.16 72.35
Sodium 48 21.17 81.06 78.94 36.61
Bisulfite 72 10.88 80.22 72.36 6.95
24 77.16 89.57 83.08 63.03
Cellulose 48 64.24 81.30 71.57 35.78
72 53.78 73.20 58.21 24.04
24 91.01 97.82 93.55 62.02
Lactose 48 81.20 91.91 80.70 45.19
72 71.91 79.82 60.93 30.58
24 97.29 95.12 95.95 86.95
Pectin 48 90.84 83.81 80.16 62.53
72 76.46 76.68 70.41 43.12

 

*ND: Not Detectable

 

49
Among all six adsorbents, activated carbon was the most

effective on day 3 (72 hours storage), followed by silica gel, while
sodium bisulfite, cellulose, lactose, and pectin indicated no
substantial adsorption of butanal.

Activated carbon and silica gel were the most effective
adsorption of butanal (Figure 5), followed by sodium bisulfite,

cellulose, lactose, and pectin.

Effeetiveness Qf Six Adsprbents Qn Cyelpheptatriene

Relative % concentration of cycloheptatriene in the vial
headspace are shown in Table 2. No significant adsorption occurred
with sodium bisulfite, cellulose, lactose, and pectin, while activated
carbon and silica gel demonstrated distinctive adsorption of
cycloheptatriene during 72 hours storage at 210C.

As shown in Figure 6, there were significantly different
adsorption profiles between the two groups of adsorbents. Activated
carbon and silica gel had the most significant adsorbing effect on
cycloheptatriene. Sodium bisulfite, cellulose, lactose, and pectin were

much less effective than activated carbon and silica gel.

E iv n f Six A r n P 11

As shown in Table 2, pentanal was not detectable in the
presence of silica gel or activated carbon during three days storage at
2 10C. Silica gel and activated carbon were very effective in
adsorbing pentanal. Sodium bisulfite, cellulose, lactose, and pectin
exhibited poor adsorption of pentanal after 24 hours storage at 210C.

After 72 hours, there was approximately 30% pentanal in the vial

50

 

Silica gel
Activated carbon
Sodium bisulfite

Cellulose

 

Lactose

Pectin

 

Relative % Concentration of Butanal
Remaining in the Vial Headspace

 

 

Storage Time (days)

Figure 5. Adsorption by Adsorbents of Butanal during
Storage at 21°C

51

 

   

 

 

 

“r5
.2
55;
‘5.
.95: Silica gel
'5
8 o 70- ""0“” Activated carbon
'45 8 65"

5‘ 60' ---0--- Sodium bisulfite
8 g 55 -
"g I 50- +— Cellulose
E: 13:
g > 35- WW Lactose
8 g 30- ‘
Q 25-l W Pectin
5‘ w 204
s I? 15 -
'5 N 10‘
E E 5-
04 04 " "L r

(){Ir ‘1' II]

Storage Time (days)

Figure 6. Adsorption by Adsorbents of Cycloheptatriene during
Storage at 21°C

......

52
headspace containing sodium bisulfite, and pectin had about the

same level as sodium bisulfite. Cellulose exhibited more than 40%
adsorption of pentanal after 72 hours storage at 2 10C. lactose
adsorbed approximately 40% pentanal in the vial headspace by day
3.

Silica gel and activated carbon (Figure 7) exhibited very high
adsorption of pentanal, while sodium bisulfite, cellulose, lactose, and
pectin exhibited some adsorption of pentanal. Among these four
adsorbents, cellulose was better than the other three. During the first
two days, there was no significant difference among sodium
bisulfite, lactose, and pectin. On day 3, lactose was more effective

than sodium bisulfite and pectin.

E 'v n f Six A r n H x

Generally, the relative % concentration of hexanal in the vial
headspace results (Table 2) were different from the previous volatile
standards. After one day storage (210C), almost 50% of the hexanal
was adsorbed by activated carbon, while silica gel, cellulose, and
lactose adsorbed about 40% hexanal. Sodium bisulfite adsorbed
approximately 30% hexanal after 24 hours storage at 2 10C. Pectin
showed poor adsorption of hexanal, with only 13% hexanal being
adsorbed.

The adsorption by silica gel, activated carbon, and sodium
bisulfite of hexanal greatly increased, while cellulose, lactose, and
pectin demonstrated moderate changes after 48 hours storage at
210C (Table 2). With silica gel as the adsorbent, the relative %

concentration of hexanal was reduced from 58% (24 hours storage) to

53

 

   

 

 

 

7e
.5;
E o —D— Silica gel
9.. 8
o 91‘ ——0—- Activated carbon
3 2
.... o
23 :1: -—O-- Sodium bisulfite
21" Te’
g; 45- + Cellulose
6 .. :12"

'5 3 W" Lactose
a? .0 0"

t: 25‘
a; E 20— W Pectin
'5 as
53 a 15-
g" 32 10-

5
0 " f {.3
v—4 N (”'3

Storage Time (days)

Figure 7. Adsorption by Adsorbents of Pentanal during
Storage at 21°C

54
23% (48 hours storage). Activated carbon lowered the level of

hexanal in the vial headspace from 48% to 12%. The relative %
concentration of hexanal in the vial headspace was reduced from 72%
(24 hours storage) to 37% (48 hours storage) in the presence of
sodium bisulfite as the adsorbent. Cellulose showed an increase in
adsorption of hexanal from 63% (24 hours storage) to 36% (48 hours
storage). With lactose as the adsorbent, 45% hexanal was measured

in the vial headspace after 48 hours storage at 210C. Pectin exhibited
low adsorption capacity of hexanal with a 63% relative concentration
of hexanal remained in the vial headspace after 48 hours storage at
210C (Table 2).

After 72 hours storage at 210C, sodium bisulfite, activated
carbon, and silica gel had the most effective adsorption of hexanal
(7% for sodium bisulfite, 8% for activated carbon, and 9% for silica
gel) (Table 2). Cellulose exhibited good adsorption of hexanal on day
3 (72 hours storage), 24% relative concentration of hexanal (Table 2)
remaining in the vial headspace. Approximately 70% of the hexanal
was adsorbed by lactose after 72 hours storage at 210C, and about
60% hexanal was adsorbed by pectin under the same storage
conditions.

As shown in Figure 8, all adsorbents adsorbed hexanal. On day
1, activated carbon had the greatest adsorption of hexanal followed
by silica gel, cellulose (lactose), sodium bisulfite, and pectin in that
order. On day 2, activated carbon still had the highest adsorption of
hexanal. On day 3, sodium bisulfte, activated carbon, and silica gel

had the most adsorption of hexanal.

 

55

 

Silica gel

Activated carbon

 

Sodium bisulfite

 

Cellulose

Lactose

Pectin

Relative % Concentration of Hexanal
Remaining in the Vial Headspace

 

 

 

Storage Time (days)

Figure 8. Adsorption by Adsorbents of Hexanal during
Storage at 21°C

56
Overall, silica gel, activated carbon, and sodium bisulfite

exhibited outstanding adsorption of butanal (Figure 9). For
cycloheptatriene and pentanal, silica gel and activated carbon had
the greatest adsorption capacity of the adsorbents (Figure 10, 1 1). On
day 1, the effect of the adsorbents on hexanal (Figure 12) was about
the same. On day 2, activated carbon demonstrated the greatest
adsorption, while sodium bisulfite had more adsorption on day 3.

As shown in Figure 13, silica gel adsorbed butanal,
cycloheptatriene, and pentanal quite readily. Activated carbon also
had excellent adsorption of butanal, cycloheptatriene, and pentanal
(Figure 14) , while sodium bisulfite exhibited good adsorption of
butanal and hexanal (Figure 15). Cellulose, lactose, and pectin showed
almost no or irnited adsorption of any of the volatile standards
(Figure 16, 17, 18).

Adsorption involves the accumulation of substances at a
surface or interface, and occurs in large measure as a result of forces
active within surface boundaries. Two types of binding forces are
commonly distinguished; i.e., physical and chemical. Physical
adsorption results from the action of van der Waals forces, comprised
of London dispersion forces and classical electrostatic forces. The
second important category of surface interaction is that of
chemisorption. As a result of significant affinities, molecular orbital
overlap occurs between molecules in the respective phases. Transfer
and sharing of electrons take place between adsorbed solute and
adsorbent, and the chemisorptive bond can have all the
characteristics of a chemical bond. The chemisorptive bond is

localized at active centers on the adsorbent and is usually stronger

57

 

Dayl
Day2
E Day3

Relative % Concentration of Butanal

 

 

 

"‘ = 0 0 U C
0 I-o m --<
0 3 E 3 2 ‘5

H "‘ U
8 6 .2 E 3 an
E m 0 —‘
U) 'U U

2 E

<6 5

.2. '6

H O

0 U:

<1

Adsorbents

Figure 9. Relative % Concentration of Butanal Remaining
in the Vial Headspace

58

 

Dayl
Day2
W Day3

 

 

Relative % Concentration of Cycloheptatriene

"‘ t: O) 6) 0) =
O o—a m "-4
(D 8 1:. 8 g 3

H —i I—i

d.)

8 3 a 3 g m
.—( uvfl I-i
:: m 0 '4
V) 'U U

2 E

cc 5

.2 ‘6

‘-‘ O

0 (I)

<1

Adsorbents

Figure 10. Relative % Concentration of Cycloheptatriene
Remaining in the Via] Headspace

59

 

Dayl
Day2
E Day3

Relative % Concentration of Pentanal

 

   

"‘ t: 0) a.) 0 C:
O a m —
0 :3 '5 § 2 a
a a .21 .5: 8 33
IE on o 4
U} 'U U

2 E

:6 3

> ._

.... 'U

‘-' O

U U)

<

Adsorbents

Figure 11. Relative % Concentration of Pentanal
Remaining in the Vial Headspace

60

 

Day 1
Day2
E Day3

Relative % Concentration of Hexanal

 

 

"‘ = d.) d.) D t:
d) u .—
0 E E § § 8
a a a a a:
E m o .—l
U) 'U U

2 E

:6 5

.2 '6'

H O

U m

<

Adsorbents

Figure 12. Relative % Concentration of Hexanal
Remaining in the Vial Headspace

61

 

20

60
50
t:
.2
E
a 40
g Dayl
O
U 30 Day2
as
2 fl Day3
a;
B
a:

10

 

 

 

 

 

 

II
0 I l

'— I— —!
a g «s ea
c o C =
a 0- C“ (U
u t— —t K
= H C q)
m S 0 m

D. On

Q)

.=

2

0

>5

0

Volatile Standards

Figure13. Relative % Concentration of Volatile Standards Remaining
in the Vial Headspace with Silica Gel as the Adsorbent

 

62

 

 

 

 

 

 

 

 

50
c; 40
.9
i
r:
o
D Day2
&°
a; 20 . Day3
=3
2
10
I I
0 l l
'23 33' “£3
:1 g: c:
53 S :3
3 C Q)
m a: :1:

Cycloheptatriene-

Volatile Standards

Figure 14. Relative % Concentration of Volatile Standards Remaining in
the Vial Headspace with Activated Carbon as the Adsorbent

63

 

::
.2
‘5
.‘J
c:
g Dayl
o
O Day2
5°
2 E Day3
E'u'
'ES
Dd

 

 

 

_.
a:
:
cu
H
=
no

Pentanal
Hexanal

Cycloheptatriene

Volatile Standards

Figure15. Relative % Concentration of Volatile Standards Remaining in
the Vial Headspace wtih Sodium Bisulfite as the Adsorbent

64

 

Dayl
Day2
§ Day3

 

Relative % Concentraion

 

 

Butanal
Pentanal
Hexanal

Cycloheptatriene

Volatile Standards

Figure16. Relative % Concentration of Volatile Standards Remaining in
the Vial Headspacewith Cellulose as the Adsorbent

 

65

 

::
.9
E
E
8 Dayl
o
O
Day2
59
.42) E Day3
‘5
3
ad

 

 

 

Butanal
Pentanal
Hexanal

Cycloheptatriene

Volatile Standards

Figure17. Relative % Concentration of Volatile Standards Remaining in
the Vial Headspacewith Lactose as the Adsorbent

66

 

C:
.2
is?
g Dayl
C
O
0 Day2
a
E E Day3
.33
0
a:

 

 

Butanal
Pentanal
Hexanal

Cycloheptatriene

Volatile Standards

Figure18. Relative % Concentration of Volatile Standards Remaining in
the Vial Headspace with Pectin as the Adsorbent

67
than the physical van der Waals forces. Whereas physical adsorption

is usually dominant at low temperatures, chemisorption is favored
by higher temperature, since chemical reactions proceed more
rapidly at elevated temperatures than at lower temperatures (Weber
and Van Vliet, 1980).

Weber and Van Vliet (1980) concluded that activated carbon
exhibited a high degree of porosity and an extensive associated
surface area. The intraparticle surface of activated carbon is
sufficiently heterogeneous to participate in most of the various
physical interaction mechanisms (Weber and Van Vliet, 1980).
Because adsorption is essentially a surface or interfacial i
phenomenon, the surface characteristics of activated carbon are of
major import (Weber and Van Vliet, 1980). That is the reason why
activated carbon had significant adsorbing capacity of volatiles
organics.

The presence of hydroxyl groups imparts a degree of polarity
to the surface of silica gel so that molecules such as water, alcohols,
phenols, amines (which can form hydrogen bonds), and unsaturated
hydrocarbons (which can form n- complexes) are adsorbed
(Ruthven, 1984). This may explain why silica gel was able to adsorb
the volatile standards.

Cellulose, lactose, and pectin also have hydroxyl groups in their
chemical structures (McMullin et al., 1975; Schneeman, 1986). These
hydroxyl groups are the most likely locations to which a negatively
charged adsorbate center would be attracted. It is likely that a
hydrogen bond forms (McMullin et al., 1975). However, since the
availability of the hydroxyl groups to form hydrogen bonds with the

68
volatile standards is limited, and the surface areas ( on which

adsorption can occur) of cellulose, lactose, and pectin are not as many
as of activated carbon. Thus, cellulose, lactose, and pectin had less

effective adsorption of the volatile standards.

Statistical Analysis

The experiment was designed as a three factor (volatile x
adsorbent x day) experiment, which included 4 different volatile
standards, 6 adsorbents, and 3 days, in a completely randomized
model. Means, standard errors, sums of square, and mean squares I
were computed using the MSTATC microcomputer statistical program 4’
(Michigan State University, 1989).

Analysis of variance (ANOVA) of the relative concentration of
the volatile standards was performed based on a three factor, split-
plot, repeat-measure model. The ANOVA of the results (Table 3)
showed significant interactions, therefore, conditional comparisons
were required. In other words, under the condition of the same
adsorbent and the same day, comparisons between any two of the
volatile standards were necessary.Thus, for exarnpe on day 1, silica
gel as the adsorbent, comparisons between any two of the volatiles
(Table 4) had to be done in order to evaluate which volatile standard
would be readily adsorbed by silica gel. With time (day) factor
remaining the same (day 1), comparisons between any two of the
volatile standards were repeated for activated carbon, sodium
bisulfite, cellulose, lactose, and pectin independently. These
procedures were repeated for dayZ and day3 (Table 5,and 6). In the

same way, comparisons between any two of the six adsorbents on

69

Table 3. Analysis of Variance for Relative % Concentration of Volatile

Standards in the Vial Headspace

 

Source Degree of Sum of Mean F
Freedom Squares Square Value

Factor A 3 10048.009 3349.336 30.4868

(Volatile Standards)

Factor B 5 1955 15.242 39103 .048 355.9200

(Adsorbents)

AB 15 48740.349 3249.357 29.5767

Error 48 5273.371 109.862

(Bottle)

Factor C 2 17944.602 8972.301 285.9067

(Days)

AC 6 6677.277 1112.880 35.4624

BC 10 2427.733 242.773 7.7361

ABC 30 2188.714 72.957 2.3248

Error 96 3012.664 3 1.382

 

Total 215 291827.962

 

 

70

Table 4. Comparisons of Volatile Standards for Each Adsorbent on

 

Dayl
>71 - >72

Volatile Silica Activated Sodium Cellulose Lactose Pectin
Standards Gel Carbon Bisulfite
Butanal vs.
Cycloheptatriene 1.516 0.393 68.922* 12.416 6.811 2.165
Butanal vs.
Pentanal 1.745 0.393 64.527* 5.92 2.545 1.34
Butanal vs.
Hexanal 56.375* 47.244* 42.711* 14.124 28.991* 10.337
Cycloheptatriene
vs. Pentanal 0.229 0.000 4.395 6.496 4.266 0.825
Cycloheptatriene
vs. Hexanal 57.891* 47.637* 26.211* 26.540* 35.802* 8.172
Pentanal vs.
W1 58.120* 47.637* 21.816* 20.044* 31.536* 8.997

 

*MSD= 16.674, any value larger than MSD would be considered
significant difference.
**MSD calculation was shown in Appendix D.

71

Table 5. Comparisons of Volatile Standards for Each Adsorbnet on

 

Day2
7. - 72'

Volatile Silica Activated Sodium Cellulose Lactose Pectin
Standards Gel Carbon Bisulfite
Butanal vs.
Cycloheptatriene 0.552 0.190 59.889* 17.061* 10.706 7.208 ,
Butanal vs.
Pentanal 0.660 0.190 57.764* 7.326 0.507 10.674
Butanal vs. .
Hexanal 22.485* 11.954 15.435 28.463* 36.016* 28.310* 1
Cycloheptatriene
vs. Pentanal 0.108 0.000 2.215 9.735 11.213 3.646
Cycloheptatriene
vs. Hexanal 23.037* 12.144 44.454* 45.524* 46.722* 21.282*
Pentanal vs.
Hexanal 23.145* 12.144 42.329* 35.789* 35.509* 17.636*

 

*MSD= 16.674, any value larger than MSD would be considered
significant difference.
**MSD calculation was shown in Appendix D.

72

Table 6. Comparison of Volatile Standards for Each Adsorbent on

 

Day3
S’I-E

Volatile Silica Activated Sodium Cellulose Lactose Pectin
Standards Gel Carbon Bisulfite
Butanalvs.
Cycloheptatriene 0.022 0.095 69.334* 19.423* 7.908 0.218
Butanal vs.
Pentanal 0.152 0.095 61.474* 3.031 10.978 9.361
Butanal vs. i
Hexanal 8.506 8.164 3.930 29.732* 41.331* 33.341* . J
Cycloheptatriene
vs.Pentanal 0.130 0.000 7.860 16.392 18.886* 9.579
Cycloheptatriene
vs. Hexanal 8.528 8.259 73.264* 49.155* 49.239* 33.559*
Pentanal vs.
Hexanal 8.658 8.259 65.404* 32.763* 30.353* 23.980*

 

*MSD= 16.674, any value larger than MSD would be considered
significant difference.
**MSD calculation was shown in Appendix D.

 

 

73
the same day for the same volatile standards were completed and

are shown in Table 7, 8 and 9. A minimum significant difference
(MSD) was computed (as shown in Appendix D) for five percent type
I error. Any value of the comparisons larger than the MSD value was

considered a significant difference.

Significant Difference Qf CQnditiQnal Cempag'sens Qf Velatile
S f r S A r . n
The conditional comparisons of any two volatile standards
within each adsorbent at the same day (day 1, day 2, and day 3) are
shown in Table 4, S, and 6. . é

(1).Day1

On day 1 with silica gel as the adsorbent, comparisons between
any two of the volatile standards are shown in Table 4. There were
three pairs of volatile standards having significant difference, these
were butanal vs. hexanal, cycloheptatriene vs. hexanal, and pentanal
vs. hexanal. Butanal, cycloheptatriene, and pentanal were adsorbed
by silica gel more readily than hexanal, while there were no
significant adsorption difference among butanal, cycloheptatriene,
and pentanal (Figure 13).

With activated carbon as the adsorbent, comparisons of butanal
vs. hexanal, cycloheptatriene vs. hexanal, and pentanal vs. hexanal
were found to be significant different which meant that butanal,
cycloheptatriene, and pentanal were adsorbed by activated carbon

more easily than hexanal. With activated carbon as the adsorbent,

74

Table 7. Comparisons of Adsorbents for Each Volatile Standard on

Dayl
YI _ Y2

‘Q OI‘D‘II B.: C 10h..1" ‘ P'on 1 H a
Silica gel vs.
Activated Carbon 1.352 0.229 0.000 10.483
Silica gel vs.
Sodium Bisulfite 27.892* 98.330* 94.164* 14.228
Silica gel vs.
Cellulose 75.411* 89.343* 83.076* 4.912
Silica gel vs.
Lactose 89.262* 97.589* 93.552* 3.896
Silica gel vs.
Pectin 95.544* 94.895* 95.949* 28.832*
Activated Carbon
vs. Sodium Bisulfite 29.244* 98.559* 94. 164* 24.7 11*
Activated Carbon
vs. Cellulose 76.763* 89.572* 83.076* 15.395
Activatted Carbon
vs. Lactose 90.614* 97.818* 93.552* 14.379
Activated Carbon
vs. Pectin 96.896* 95.124* 95.949* 39.315*
Sodium Bisulfite
vs. Cellulose 47.519* 8.987 13.088 9.316
Sodium Bisulfite
vs. lactose 61.370* 0.741 2.612 10.332
Sodium Bisulfite
vs. Pectin 67.652* 3.435 0.215 14.604
Cellulose vs.
lactose 13.851 8.246 10.476 1.024
Cellulose vs.
Pectin 20.133* 5.552 12.873 23.920*
lactose vs.
Pectin 6.282 2.694 2.397 24.936*

 

*MSD=18.625, any value larger than MSD would be considered significant

difference.

“MSD calculation was shown in Appendix D.

O
4

Table 8. Comparisons of Adsorbents for Each Volatile Standard on

Day2

75

O

A 0 p 1 B . C 1110‘- ..I ' Pn . me
Silica gel vs.

Activated Carbon 0.470 0.108 0.000 1 1.03 1
Silica gel vs.

Sodium Bisulfite 20.512* 80.95 3* 78.936* 13.462
Silica gel vs.

Cellulose 63.581* 81.194* 71.567* 12.633
Silica gel vs.

lactose 80.545* 91.803* 80.698* 22.044*
Silica gel vs.

Pectin 90.178* 83.702* 80.164* 39.3 83*
Activated Carbon

vs. Sodium Bisulfite 20.982* 81.061* 78.936* 24.493*
Activated Carbon

vs. Cellulose 64.051* 81.302* 71.567* 23.664*
Activatted Carbon

vs. Lactose 81.015* 91.911* 80.698* 33.075*
Activated Carbon

vs. Pectin 90.648* 83.810* 80.164* 50.414*
Sodium Bisulfite

vs. Cellulose 43.069* 0.241 7.369 0.829
Sodium Bisulfite

vs. lactose 60.033* 10.85 1.762 8.582
Sodium Bisulfite

vs. Pectin 69.666* 2.749 1.228 25.921*
Cellulose vs.

lactose 16.964 10.609 9.131 9.411
Cellulose vs.

Pectin 26.597* 2.508 8.597 26.750*
Lactose vs.

Pectin 9.633 8.101 0.534 17.339

 

*MSD= 16.674, any value larger than MSD would be considered significant

difference.

**MSD calculation was shown in Appendix D.

 

76

Table 9. Comparisons of Adsorbents for Each Volatile Standard on

Day3
Yr _ Y2

A oru‘n B130. C lhow-J‘n‘ Pn v. H-xgni
Silica gel vs.
Activated Carbon 0.057 0.130 0.000 0.399
Silica gel vs.
Sodium Bisulfite 10.732 80.080* 72.358* 1.704
Silica gel vs.
Cellulose 53.623* 73.068* 56.806* 15.385
Silica gel vs.
Lactose 71.75 8* 79.688* 60.932* 21.921*
Silica gel vs.
Pectin 76.310* 76.550* 67.101* 34.463*
Activated Carbon
vs. Sodium Bisulfite 10.789 80.210* 72.358* 1.305
Activated Carbon
vs. Cellulose 53.680* 73.198* 56.806* 15.784
Activatted Carbon
vs. Lactose 71.815* 79.818* 60.932* 22.320*
Activated Carbon
vs. Pectin 76.367* 76.680* 67.101* 34.862*
Sodium Bisulfite
vs. Cellulose 42.891* 7.012 15.552 17.089
Sodium Bisulfite
vs. Lactose 61.026* 0.392 11.426 23.625*
Sodium Bisulfite
vs. Pectin 65.578* 3.530 5.257 36.167*
Cellulose vs.
lactose 18.135 6.620 4.126 6.536
Cellulose vs.
Pectin 22.687* 3.482 10.295 19.078*
Lactose vs.
Pectin 4.552 3.138 6.169 12.542

 

*MSD= 16.674, any value larger than MSD would be considered significant

difference.

**MSD calculation was shown in Appendix D.

77
similar adsorption profiles were found for butanal, cycloheptatriene,

and pentanal(Figure 14).

Only one comparison, cycloheptatriene vs. pentanal, showed no
significant difference when sodium bisulfite was the adsorbent. As
shown in Figure 15, sodium bisulfite exhibited substantial adsorption
of butanal and hexanal, and had no significant adsorption of
cycloheptatriene and pentanal.

For cellulose, the comparisons of cycloheptatriene vs. hexanal
and pentanal vs. hexanal were significantly different. Cellulose had
significant adsorption of hexanal in comparison to adsorption of
cycloheptatriene and pentanal (Figure 16).

Butanal vs. hexanal, cycloheptatriene vs. hexanal, and pentanal
vs. hexanal were observed to be significantly different with lactose
as the adsorbent (Table 4). This indicated that more higher levels of
hexanal were adsorbed than butanal, cycloheptatriene, and pentanal,
and that there were no significant differences among butanal,
cycloheptatriene, and pentanal (Figure 17).

There were no significant differences between any of the
comparisons when pectin was the adsorbent (Table 4). In other
words, the adsorption capacity of pectin was almost the same for
butanal, cycloheptatriene, pentanal, and hexanal on day 1 (Figure
18).

W

Comparisons of the volatile standards adsorbed by silica gel
showed that no significant differences were observed for butanal vs.

hexanal, cycloheptatriene vs. hexanal, and pentanal vs. hexanal

78
(Table 5). Butanal, cycloheptatriene, and pentanal were adsorbed

significantly by silica gel. Hexanal was also adsorbed by silica gel but
less so than butanal, cycloheptatriene, and pentanal (Figure 13).

There were no significant difference observed in the
comparisons of any two volatile standards with activated carbon as
the adsorbent (Table 5, Figure 14).

The results of the comparisons of butanal vs. cycloheptatriene,
butanal vs. pentanal, cycloheptatriene vs. hexanal, and pentanal vs.
hexanal exhibited significant difference (Table 5), which revealed
that sodium bisulfite had good adsorption of butanal and hexanal
with less effective adsorption of cycloheptatriene and pentanal
(Figure 15).

With cellulose as the adsorbent, the comparisons of butanal vs.
cycloheptatriene, butanal vs. hexanal, cycloheptatriene vs. hexanal,
and pentanal vs. hexanal had significant difference (Table 5).
Cellulose had more adsorption of butanal than of cycloheptatriene,
and adsorption of hexanal onto cellulose was significantly higher
than adsorption of butanal, cycloheptatriene, or pentanal onto
cellulose (Figure 16).

The amount of hexanal adsorbed by lactose was significantly
more than the amounts of butanal, cycloheptatriene, or pentanal
adsorbed (Table 5, Figure 17).

The comparisons of butanal vs. hexanal, cycloheptatriene vs.
hexanal, and pentanal vs. hexanal were significantly different when
pectin was the adsorbent (Table 5). This indicates that hexanal was
adsorbed by pectin readily, while butanal, cycloheptatriene, and

79
pentanal showed no significant difference of adsorption by pectin

(Figure 18).
W

The results of the comparisons of volatile standards for each
adsorbent on day 3 are shown in Table 6.

There were no significant differences observed in the
comparisons of volatile standards with silica gel as the adsorbent
(Table 6). As shown in Figure 13, each volatile standard was highly
adsorbed by silica gel, and thus no significant differences occurred.

No significant differences observed for the comparisons of
volatile standards for activated carbon (Table 6). Each volatile
standard was readily adsorbed by activated carbon (Figure 14).

With sodium bisulfite as the adsorbent, butanal vs.
cycloheptatriene, butanal vs. pentanal, cycloheptatriene vs. hexanal,
and pentanal vs. hexanal had significant difference (Table 6). Butanal
and hexanal were adsorbed by sodium bisulfite while
cycloheptatriene and pentanal were not (Figure 15).

The comparisons of volatile standards with cellulose as the
adsorbent on day 3 had the same significances as on day 2. Hexanal
was the most significantly adsorbed by cellulose (Figure 16). Butanal,
cycloheptatriene, and pentanal had almost the same level of
adsorption by cellulose.

In addition to the significant difference in the comparison of
cycloheptatriene vs. pentanal, there was a significant different
between hexanal and the other three volatile standards (butanal,
cycloheptatriene, and pentanal ) with lactose as the adsorbent (Table

6). Hexanal was the most highly adsorbed by lactose among the

80
volatile standards. The difference in adsorption by lactose between

pentanal and cycloheptatriene was significant. Pentanal was
adsorbed by lactose more than cycloheptatriene (Figure 17).

There was significant difference between butanal vs. hexanal,
cycloheptatriene vs. hexanal, and pentanal vs. hexanal, with pectin as
the adsorbent (Table 6). Hexanal was significantly adsorbed by
pectin by comparing to the other volatile standards (butanal,
cycloheptatriene, and pentanal ) (Figure 18).

Significant Difference Qf Cenditienal Cemparisens Qf Adserbents for
S V 1 i1 S

Within the same day (day 1, day 2, and day 3), for each volatile
standard (butanal, cycloheptatriene, pentanal, and hexanal), any two
of six adsorbents (silica gel, activated carbon, sodium bisulfite,
cellulose, lactose, and pectin) were paired for comparisons. A
minimum significant difference (MSD) was computed as shown in
Appendix D. Any value of the comparisons of the adsorbents larger

than MSD was considered significantly different.

(1).Dayl

Comparisons for the adsorbents using butanal as the volatile
standard are shown in Table 7. Comparisons between silica gel and
four adsorbents (sodium bisulfite, cellulose, lactose, and pectin) were
significantly different which means that silica gel had higher
adsorption capacity of butanal than the other four adsorbents.
Activated carbon had a similar adsorption capacity for butanal as did
silica gel (Table 7). Sodium bisulfite also exhibited good adsorption of

81
butanal compared to adsorption by cellulose, lactose, and pectin

(Figure 19).

Using cycloheptatriene as the volatile standard, the comparison
results (Table 7) show significant difference for comparisons
between silica gel and four adsorbents (sodium bisulfite, cellulose,
lactose, and pectin). Activated carbon also had significant difference
with the same four adsorbents (sodium bisulfite, cellulose, lactose,
and pectin). Silica gel and activated carbon were the most effective
adsorbents of cycloheptatriene (Figure 20).

The effectiveness of the adsorbents on pentanal were the same
as previously for cycloheptatriene. Silica gel and activated carbon
were the most effective adsorbents for pentanal (Figure 21).

Results were significantly different (Table 7) for comparisons
of the adsorbents with hexanal as the volatile standard. Pectin was
the least effective adsorbent of hexanal. Silica gel, cellulose, and
lactose had similar adsorption profiles for hexanal on day 1, while
activated carbon exhibited the greatest adsorption of hexanal (Figure
22).

Mid

As shown in Table 8, the comparisons for the adsorption
capacity of the adsorbents with butanal was essentically the same as
on day 1. Silica gel and activated carbon adsorbed almost all of the
butanal in the vial headspace. Sodium bisulfite had significant
adsorption of butanal compared to cellulose, lactose, and pectin
(Figure 19).

82

 

Dayl
Day2
E Day3

Relative % Concentration of Butanal

 

 

 

'23 g 3 3 3 E.
(D .o it: o o 3
h I— —i 6.!

8 a a a 8 £3
E m o ._l

U) 'U U
0 E
E =
.2 '6‘
~ 0
0 m
<
Adsorbents

Figure 19. Relative % Concentration of Butanal Remaining
in the Vial Headspace

83

 

Dayl
Day2
E Day3

 

 

Relative % Concentration of Cycloheptatriene

"‘ C 0 O O t:
O o—a m "-1
O :3 E E .9 *5
0

8 6 .2 E 3 an
E m o ._1
U) 'U U

2 E

Q 5

.z ‘6‘

.. o

0 w

<

Adsorbents

Figure 20. Relative % Concentration of Cycloheptatriene
Remaining in the Vial Headspace

84

 

Dayl
Day2
E Day3

Relative % Concentration of Pentanal

 

 

T, G 4.) O O "E
O 8 E 8 8 3
H — —1 H
8 e .2 a a 8:
SE in o 4

U) u U
2 E
a: :1
.: '6
H O
0 (I)
<1
Adsorbents

Figure 21. Relative % Concentration of Pentanal Remaining
in the Vial Headspace

85

 

Day 1
Day2
@ Day3

Relative % Concentration of Hexanal

 

 

"" C O o O C
Q) a w "-1
0 § E E 9 3
8 5 .2 E 8 8
E m v v-'
U) u U

3 E

a =

.2. ‘6

‘-’ O

U m

<

Adsorbents

Figure 22. Relative % Concentration of Hexanal Remaining
in the Vial Headspace

86
For cycloheptatriene, significant difference was observed in the

comparisons of silica gel to sodium bisulfite, cellulose, lactose, and
pectin. Similarly comparison between activated carbon and sodium
bisulfite, cellulose, lactose, and pectin were significantly different
(Table 8). Silica gel and activated carbon were the most significant
adsorbents for cycloheptatriene (Figure 20).

With pentanal as the volatile standard ( day 2), the results
(Table 8) of comparisons for the adsorbents showed the same level of
significant difference as the results on day 1. Pentanal was not
detectable in the vial headspace with silica gel or activated carbon as
the adsorbent, while the relative % concentration of pentanal in the
vial headspace remained high when sodium bisulfite, cellulose,
lactose, or pectin was the adsorbent (Figure 2 1).

On day 2, the adsorption capacity of the adsorbents with
hexanal had changed from day 1. The adsorption capacity of silica gel
vs. lactose and pectin were significantly different. The amount of
hexanal adsorbed by silica gel was higher than the amount of
hexanal adsorbed by lactose or pectin. The comparisons between
activated carbon and sodium bisulfite, cellulose, lactose, and pectin
were significantly different, which meant activated carbon offered

the greatest adsorption of hexanal (Figure 22).

(11.272113

The significant difference in the comparisons of adsorbents
with butanal as the volatile standard (day 3) are shown in Table 9.
The comparisons between silica gel and three adsorbents (cellulose,

lactose, and pectin) were significantly different which meant that

87
silica gel had a higher adsorption capacity for butanal than the other

three adsorbents. Activated carbon had similar adsorption of butanal
as did silica gel (Table 9). Sodium biSulfite also exhibited good
adsorption of butanal compared to the adsorption of cellulose,
lactose, and pectin (Figure 19).

Using cycloheptatriene as the volatile standard, the comparison
results (Table 9) show significant difference for comparisons
between silica gel and sodium bisulfite, cellulose, lactose, and pectin.
Activated carbon also had significant difference with sodium
bisulfite, cellulose, lactose, and pectin. Silica gel and activated carbon
had the most effective adsorption of cycloheptatriene among the
adsorbents (Figure 20).

On day 3, the effectiveness of the adsorbents with pentanal
was the same as previously discussed with cycloheptatriene. Silica
gel vs. sodium bisulfite, cellulose, lactose, and pectin were
significantly different. Activated carbon vs. sodium bisulfite,
cellulose, lactose, and pectin were also significantly different (Table
9). Silica gel and activated carbon could adsorb pentanal more
effectively than the other four adsorbents (Figure 21).

Comparisons of the adsorbents with hexanal are shown in Table
9. Silica gel vs. lactose, and pectin were significantly different.
Activated carbon vs. lactose, and pectin exhibited significant
difference. Sodium bisulfite vs. lactose, and pectin were significantly
different. Significant difference was also observed in the comparison
of cellulose vs. pectin. Silica gel, activated carbon, and sodium

bisulfite adsorbed hexanal similarly and which were better than

88
lactose and pectin. The effectiveness of cellulose with hexanal was

better than pectin (Figure 22).

Overall, both silica gel and activated carbon exhibited
significant adsorption of butanal, cycloheptatriene, and pentanal in
comparison to the other four adsorbents (sodium bisulfite, cellulose,

lactose, and pectin) on day 1, 2 and 3.

Effect Qf Selected Adserbents en Oxidized Frezen le Trent

Some investigators have found that volatile aldehydes
contributed strongly to the characteristic aroma of oxidized fish
lipids (Lea, 1953; Yu et al., 1961; Aiken and Connell, 1979; Ikeda,
1980). In this study, volatile aldehydes such as butanal, pentanal,
and hexanal were found as the major off-flavor compounds in
oxidized frozen lake trout.

Low temperature storage (Young, 195 O; Dyer and Morton,
1956; Peters and Mclane, 1959), application of antioxidant (Brown et
al., 1957; Andersson and Danielson, 1961; Yu et al., 1969; Sweet,
1973; Love and Pearson, 1974; Deng et al., 1977), and packaging
technique (Griffin, 1980; Gorga, 1988; Connell, 1990) etc. can be used
to prevent off-flavor development in fish during storage. Shelf life is
extended by lowering the storage temperature (Peters and McLane,
1959). However, temperature fluctuation can cause quality
deterioration, an increase in degree of lipid oxidation, and shorten
the storage life of frozen fish (Dyer, 195 9; Lentz and Rooke, 1960;
Pahnateer et al., 1960; Lane, 1966). Application of phenolic
antioxidants in frozen fish fillets was found not effective in retarding

oxidative rancidity (Stuckey, 1968). Inadequate distribution of the

89
antioxidants may limit the usage on fish samples (Sweet, 1973). Gas

impermeable packaging needs to be associated with low temperature
(below 40C) storage, otherwise it may cause an outgrowth of
Clesmm'um 11am (Gorga, 1988). Therefore, a different
approach was used in this research to reduce concentration of off-
flavor compounds in oxidized frozen lake trout using adsorbents to
adsorb the off-flavor compounds.

The adsorbing capacity of six adsorbents was evaluated, and
activated carbon and silica gel were found to have the greatest
capacity.

The capability of activated carbon and silica gel in reducing
off-flavor development in oxidized frozen lake trout was determined.

Depositing 40 grams of fish sample (frozen lake trout) in a 450
ml glass jar with a screw cap which has a sampling port on the t0p as
shown in Figure 4 . Two grams adsorbent, silica gel or activated
carbon, was packed in a Tyvek pouch (1.0 in x 1.0 in ) and
suspended above the fish sample and stored at 50C, for three days.
Analysis of the headspace in the jar was completed by GC. The most
abundant volatile (retention time: 4.70 min.) was employed to
monitor the adsorption activity of activated carbon and silica gel. The
area response analyses of the most abundant volatile are shown in
Table 10. The area response of the control was designated 100%
relative concentration of volatile. Using this as the baseline, Table 10

was converted into Table 1 1.

90

Table 10. Area Response1 of Volatile2 in the Headspace of Fish
Samples after 72 Hours Storage at 50C

Fi S 1
mm 1 2 3
Silica gel 3075 15.0 3808.5 29545.0
Activated 3 167.0 2040.5 339.5
Carbon
Control 3 15675.0 5776.0 25 100.5

 

1. All values represent the average of two replicated experiments.
2. The most abundant volatile (retention time: 4.70 min.)

Table 1 1. Relative % Concentration1 of Volatile2 in the Headspace of
Fish Samples after 72 Hours Storage at 50C

Ei§h§amn1£
Amman; 1 2 3
Silica gel 97.41 i 5.14 65.94 i 3.49 99.99 i 1.74
Activated 1.00 i: 0.40 34.33 i 2.08 1.35 i 0.22

Carbon

 

1. All values represent the average of two replicated experiments.
2. The most abundant volatile (retention time: 4.70 min.)

91
$1 .I...n‘ ‘ .‘ .f‘ . 0. ...' v.-_.. fOOI'FI’Zn a TI‘OI

With activated carbon as the adsorbent, the relative %
concentrations of the most abundant volatile in the jar headspace
(Table 11) were 1.00%, 34.33%, and 1.35% for fish sample 1, 2, and 3
respectively after 72 hours storage at 50C. This indicates that
activated carbon has the capability of adsorbing off-flavor volatiles
and is compatible with the findings of other researchers (Tausig and
Drake, 1959; Gernon et al., 1961; Brooks and Morr, 1982; Ruthven,
1984; Sakaki et al., 1984; Tore-Vazquez et al., 1991).

The adsorption activity of activated carbon for fish sample 2
(34.33%) was less effective than for fish sample 1 (1.00%), and 3

( 1.3 5%). This variation in results could be due to experimental error.

Adsemd en Activity ef Silica Gel fer Frezen Lake Tredt

Using silica gel as the adsorbent, the relative % concentration of
the most abundant volatile in the jar headspace (Table 1 1) were
97.41%, 65.94%, and 99.99% for fish sample 1, 2, and 3 respectively
after 72 hours storage at 50C. This reveals that silica gel had poor
adsorption activity for the off-flavor volatiles from frozen lake trout.
Some investigators reported that silica gel had good adsorption
capacity for aromatics, alcohols, phenols, amines, hydrocarbons
(Davis et al., 1952; Ruthven, 1984; Chou et al., 1986; Ackman and
Ratnayake, 1989b; Shantha and Ackman, 1991). This variation could
have been caused by the moisture in the fish which may have
competed with the volatiles for surface adsorption. The presence of
hydroxyl groups in silica gel imparts a degree of polarity to the

surface, therefore water vapor can be adsorbed readily (Ruthven,

92
1984). The high capacity of silica gel to adsorb relatively large

quantities of water was known in the early work on silica gel and
this high adsorptive capacity was explained on the basis of muti-
layer adsorption (Scott, 1993). While, the adsorption of water was
greatly influenced by the hydrophobicity of the external surface and
the micropores of actived carbon. This indicated that he hydrogen-
bonded structure, by which water is generally adsorbed, was not
viable within the constricted environment of the hydrophobic
micropore of activated carbon (Bansal et al., 1988). Thus, the
adsorption capacity of activated carbon has not been affected by the

moisutre remaining in the fish.

Sensery Evaldaden

A paired comparisons test was used for sensory evaluation. Six
panelists were asked to sniff the fish samples (frozen lake trout)
with activated carbon (treatment) and with glass beads (control)
after storage at 50C for 3 days. Each panelist was asked to indicate
which sample had the stronger off-odor. This sniff-test was repeated
three times by the same group of panelists on different dates. The
corresponding sensory scores are presented in Table 12.
Five judges chose the control (with glass beads) as stronger at the
first session. All judges indicated that the control had the stronger
off-odor at the second session, and five judges gave right verdicts at
the third session (Table 12). 18 judgments were made after three
sessions, and 16 out of 18 were correct judgments. According to
larmond (1977), in a two-sample difference test one sample must be

selected 13 times, 15 times, and 16 times out of 18 judgments to be

93
significantly different at the 5%, 1%, and 0.1% level of significance

respectively. In the sensory score, 16 times out of 18 judgments
were right verdicts which was significantly different at the 0.1%
level of significance. There was detectable difference in off-odor
between the two treatments.

The results of sensory evaluation confirmed that the adsorption
activity of activated carbon was significantly effective in decreasing

off-flavor volatiles in oxidized frozen lake trout.

Table 12. Sensory Score of Frozen lake Trout With Activated Carbon
in a Jar after Storage at 50C for 3 Days

 

 

Judge lst Time 2nd Time 3rd Time
1 X R R
2 R R R
3 R R R
4 R R X
S R R R
6 R R R
Total 5R 6R 5R

=Wrong verdict
R=Right verdict

16 out of 18 correct judgments

SUMMARY AND CONCLUSIONS

Identification of the major off-flavor volatiles in frozen lake
trout was completed by GC/ MS. Butanaltcycloheptatriene, pentanal,
@gfihfiwalflwere identified as, the major off-flavormcgmpounds from
the oxidized frozen lake trout by matching the GC/ MS spectrum with
the Mass spectrum standards. The aclsprpticpn_ofsixw adsorbents on

 

fguflryfigladlewstandards was examined. Evaluationflgf the adsorbents
with frozen lake trout was carried out to deterrninewthe efficiency of
the adsorbents to control off-flavor of the frozen lake trout during
storage.

In experiment 11, Silica gel; and activated carbOn exhibited
significant adsorption effects on the volatile standards which were
butanal, cycloheptatriene, pentanal, and hexanal. Therefore, both
silica ,gel 31.19-3131“th carbon were used as adsorbents in further
studies.

Silica gel and activated carbon were included in jars containing
the frozen fish samples. Silica _gEl was ineffective in adsorbing the

7‘ "‘"J'Mm-fl'

offL-flhavorvolatiles probably due to adsorptionof the moisture, while
affilfig£arb0n suppression the off7flavor volatiles. Senspg
eyaluatipn also demonstrated the effectiveness of activated carbon.
The results showed that activated carbon has potential as an
adsorbent to Wflgr of frozen fish during storage.

"I'I'H-qh. fl ‘11....h... r'r'i
" ”‘~ nil-nun”,-

Application of activated carbon to food products needs verification.

94

95
Eliminating the moisture adsorption by silica gel may allow it to be

used. Sodium bisulfite has great adsorption of particular volatiles;
cellulose, lactose, and pectin have adsorption of some volatiles. Their
adsorption capacity may not be as good as activated carbon, they

may be suitable for particular purpose of other adsorption.

APPENDIX A

96
APPENDIX A

GC Mass Spectrometry Analysis

47 peaks were recorded for the identification of volatile aroma
compounds from oxidized frozen lake trout (Figure A—l). Four peaks
(retention time: 3.912, 4.070, 8.254, and 27.020 min.) were in the
most abundant quantity which were represented in area response
(Table A-l). The mass spectrum of these four volatile compounds
were shown in Figure A—2, A—3, A-4, and A-5. Butanal, pentanal,
cycloheptatriene, and hexanal were identified as these four volatile

compounds based on coincidence of mass spectral patterns of

standards.

 

97
Figure A-1. Mass Chromatogram of Volatile Aroma Compounds from

Oxidized Frozen Lake Trout

W359“! J

5.38 a

3883 i

«$.18 ..

 

 

 

 

3.29

 

a

 

r Ir?

 

o
O A I
‘J
Lg
L

 

 

 

H-—4

N—4

“—4

nw$jjafijmddqtqlflflll
o 82332;...233

1: _ _t? a

.3 8

~

 

fr

98

Table A-1. Retention Time, Area Response, of Volatile Aroma

Compounds from Oxidized Frozen lake Trout

Peak# Ret Time

smooxrmtnpcqm~

N—d

N—L—‘dfi—‘u—bfl
swmflmm¢~Lfl

NNI‘JNNFQ
mmkblNfl

hJ

(NU! LX101 (510401 DJ UJUJNDJ
mmflmmktflN—‘SLDCD

##JA
N—S

##Jh#¥>
\ImU'ltaLd

7's

uJUJUIthtotuiolorbibJOJUlec
:> J: J> t4 Ln Lu L» L» LN hJ ha bJ tJ hJ h

35

.912
.070
.254
.684
.557
.518
.828
.020
.428
.609
.928
.334
.537
.608
.728
.996
.043
.140
.231
.418
.498
.668
.817
.210
.366
.467
.516
.616
.753
.887
.008
.115
.366
.491
.649
.724
.008
.186
.308
.390
.653
.202
.668
.867
.099
.755
.722

Type
BU
UB
BU
PU
PU
BU
BU
PU
UU
UU
PU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
UU
BU

width
.256
.301

.088
.070
.054
.054
.072
.106
.085
.068
.103
.059
.053
.049
.069
.126
.045
.072
.097
.111

.089
.111

.080
.216
.142
.037
.049
.080
.108
.089
.069
.046
.149
.093
.118
.099
.100
.066
.084
.067
.108
.089
.065
.077
.065
.071

.073

SSSSSSGSSGGSSSSSSSSSSSGSSSSSSSGSGS8888888888888

Area

7582772058
6468088100

786325963
21684032
37359175
29620344
26627679

2384072632

25953732
13781100
12059580
35382387
14201909
20810458
21659085
28889195
8598594
22958538
24018807
32150752
33415050
55797155
103466106
201578915
158436126
28203892
47117479
72247295
113078637
83614048
233068950
36919602
175481277
75958041
189323347
88411237
110747168
370458178
92308397
437260182
81511276
85367652
285247500
171092343
198027732
110700992
47239001

Start Time

3

4
8

-o

' b1 01 U1 Lu Lu L-1 01 bl bl t;-1 t4 t-l (>1 bl (>1 0
U1¢-.t:-L\1(;~1L\1t~1ub1 NNroMerJr

.613
.036
.116
.522
.409

.323

.667
.893
.411
.524
.739

.221

.467
.571
.661
.793
.012
.064
.179
.292
.433
.545
.701
.884
.240
.446
.484
.550
.652

.801

.911

.084

.197
.412
.508
.698
.927

.062

.262

.209

.580
.098
.498
.733
.974
.634
.638

End Time
.036
.757
.623
.778
.661
.602
.893
.411
.572
.739
.001
.426
.571
.661
.793
.012
.064
.179
.292
.433
.545
.701
.884
.240
.446
.484
.550
.652
.801
.911
.084
.131
.412
.508
.698
.803
.062
.262
.384
.498
.755
.289
.775
.974
.221

.890
.836

45#«hLXIUJUJL'db-JUlrer\Jr‘-JN1\J1\J1‘J—-‘—‘-‘—‘-‘--‘“MSSSGSLDQQU‘JO’JU’ISCO-h4p

()1 ()1 L11 U! bl LN bl LI] ()1 L11 U] LN U1 (:4 U1 U1 U1 U1 U1 ()1 ()1 U1 L34 L1] 01 U1 L‘s] L4 01 "J 1“~J N R) R) Pd h.)

b 04 L111)! L's] U! U] 01
“LDCD'QCDO'J U1

 

 

99
Figure A-Z. Mass Spectrum of Peak #1 (Retention Time: 3.912)

 

 

 

 

 

 

 

 

 

 

 

u
too
751 27
72
50‘
l j ,
1 5 1 1
.! If ‘
25«
s:
wH- -

 

v I I . I I l . I I ' I I I r ' I I I . I . I I I
20 '90 60 81.1 100 120 1'10 1613 1:313 200 2‘20 H!) 260 280 300

Mhml

100
Figure A-3. Mass Spectrum of Peak #2 (Retention Time: 4.070)

 

 

 

 

 

 

 

. o H
1°01 23

7S1

5‘31

53
25
A l .-
0.4... ‘ I r .y I I. v '0. I l r I ' I. ' l ' I ' I ' I ' ‘ ' '
2') 90 50 80 IOU 1:11 190 160 13U 200 2213 290 250 230 3130

pentanal

101
Figure A-4. Mass Spectrum of Peak #3 (Retention Time: 8.254)

 

 

 

 

 

 

 

 

 

 

H

100'
0
751

56
5th 29
25‘

.1?

ga
0"- "T"""'L':.‘.".'..'.".1'.:.‘.'..
£0 '10 60 $0 100 til) 1’10 1w) luv :00 2:0 :10 :90 no son

102
Figure A-5. Mass Spectrum of Peak # 8 (Retention Time: 27.020)

 

 

 

 

31
100
O
.79
501
33

251

SS

51 J

0 I I Ii I r I I I V I I I .7 I I I I I I r . I I I I I I I
20 $0 50 80 100 120 1110 150 15'") 200 2‘20 2% 260 2‘30 300

1, 3, S-Cgc loheptatr iene

APPENDIX B

 

103
APPENDIX B

Standard Calibration

Standard calibration curves for butanal, cycloheptatriene,
pentanal, and hexanal were established. Five concentrationszloppm,
100ppm, SOOppm, 1000ppm, and 2000ppm(Vol./Vol.) were
prepared for each volatile standard in o-dichlorobenzene as the
solvent. In order to generate the data for the calibration curves,

0.5 11 l of sample was injected directly into the gas chromatography
every 24 hours up to 72 hours. The area response was recorded
under the following conditions.

Column: SupelcowaxlO: 0.25mm id x 60mm capillary column
Conditions:

Range: 4

Helium carrier gas: 27ml/ min

Column temperature: 1500C

Detector temperature: 2500C

Injection temperature: 2200C

The calibration data and the standard calibration curves for
butanal were shown in Table B-1, Figure B-1, B-2, and B-3.

For cycloheptatriene, the calibration data and the standard
calibration curves were shown in Table B-2, Figure B-4, B-5, and B-6.

The calibration data and the standard calibration curves for
pentanal were shown in Table B-3, Figure B—7, B-8, and B-9.

For hexanal, the calibration data and the standard calibration curves
were shown in Table B-4, Figure B-10, B-11, and B-12.

 

Table B-1 Butanal Calibration Data

104

Concentration W mm

mm hijectedmzfiigi 12M Daxz 12M
10 0.004 1343 1403 1292
100 0.04 20380 18855 31167
500 0.2 82765 89079 93650
1000 0.4 163680 232110 188710
2000 0.8 327270 356720 348030

 

*Retention time: 2.08min
Density: 0.80g/ml

Table B-2 Cycloheptatriene Calibration Data

(ppm, VZV)
10 0.00444
100 0.0444
500 0.222
1000 0.444
2000 0.888

Wiigl

123321 M 122323
4600 3692 4532
44158 44126 43054
212050 203250 190180
473640 408900 397060
924540 789660 965090

 

*Retention time: 4.55min.
Density: 0.888g/ ml

 

105
Table B-3 Pentanal Calibration Data

W WW

mm V Injectediiflzfiigl M 12312 23x32
10 0.00405 1809 2070 2401
100 0.0405 21975 20975 20155
500 0.2025 88543 86873 77022
1000 0.405 213780 230310 207470
2000 0.810 448480 455310 367150

 

*Retention time: 3.09min.
Density: 0.8 10g/m

Table B-4 Hexanal Calibration Data

Qcmcentration Wm WA r Ar R n *

1121201._L_1V V InjecteXmQiLgl Daxl Bax; Day}
10 0.00417 3707 3497 3534
100 0.0417 47129 37585 39639
500 0.2085 213610 202410 178010
1000 0.417 336830 316380 361080
2000 0.834 669880 649100 781880

 

*Retention time: 5.3 2min.
Density: 0.8 34g/ ml

106

 

Area Response

 

 

 

 

400(X)O
y = 407533.314x + 1159.983 1' =
1.000
300000-
200000-
100000—
0 I l l
O a If} p .—4
0' o 6

Quantity of Butanal Injected (X10-6g)

Figure B-l. Butanal Calibration Curve on Dayl

 

Area Response

107

400000

 

y = 463009.865x + 4930.126 =
0.990 D

 

 

 

 

0.75 '1

Quantity of Butanal Injected(X10-6g)

Figure B-2. Butanal Calibration Curve on Day2

 

Area Response

108

 

y = 433853.635x + 6060.725 r =
0.999

 

 

 

 

Quantity of Butanal Injected(XlO-6g)

Figure B-3. Butanal Calibration Curve on Day3

Hint-.2 ’

 

109

 

 

 

 

 

1000000
y = 1046633.798x - 3099.753 r =
1.000
750000 -
8‘
3; 500000 .1
a
<
250000 -
o It... I
0 g .—

0.25 '
0.75 ‘

Quantity of Cycloheptatriene Injected(X10-6g)

Figure B-4. Cycloheptatriene Calibration Curve on Day 1

110

 

800000

y = 891323.500x + 3496.507 r
1.000

Area Response

 

 

 

 

Quantity of Cycloheptatriene Injected(X10-6g)

Figure B-5. Cycloheptatriene Calibration Curve on Day2

 

111

 

 

 

 

1000000
y = 10677959le - 18598335 I 5'
0.995
750000 ..
a)
‘8
O
a.
6
04 500000 -
g D
<
250000 '-
0 {'1 r I l
o g V“. fl '7
0' o o'

Quantity of Cycloheptatriene Injected(X10-6g)

Figure B-6. Cycloheptatriene Calibration Curve on Day3

Nth

 

Area Response

112

 

500000

400000-

300000‘

200000-I

100000-

 

0“.-

Figure B-7. Pentanal Calibration Curve on Dayl

y = 552649.251x - 5568.973 r =
0.999

 

 

l
‘0.
o

0.25 ‘
0.75 '7

Quantity of Pentanal Injected(X10-6g)

 

Area Response

113

 

 

and»
y = 565963.748x - 5321.550 r =

0.998
400000 .

300000-I

ZGXDO-

  

IGXDO-

 

 

 

0“.-

O

025‘
05‘
075'

Quantity of Pentanal Injected(X10-6g)

 

Figure B-8. Pentanal Calibration Curve on Day2

IIIIIIIIIIIIIIIlIllllllIll-Il--------——* glam

Area Response

114

 

y = 461093.908x + 9.275 r = 97

 

 

 

 

Quantity of Pentanal Injected(X10-6g)

Figure B-9. Pentanal Calibration Curve on Day3

 

 

Response

Area

115

 

800000

y = 794682.242x + 12479.199 r =
0.998

 

 

 

 

l
‘0.
o

0.25 "
0.75 4

Quantity of Hexanal Injected(X10-6g)

Figure B-10. Hexanal Calibration Curve on Dayl

11m...

 

Area Response

116

 

 

800000
y = 770487.399x + 8183.897 1' =
0.998
600000-
400000-
200(X)O-I
0"" I I r
o‘ O :5

Figure B-ll. Hexanal Calibration Curve on Day2

 

 

Quantity of Hexanal Injected(X10-6g)

”Iii-9.... _

Area Response

117

 

800000

y = 929608.048x - 5876.844 r
0.999

 

 

 

 

0.75 "‘

Quantity of Hexanal Injected(X10-6g)

Figure B-l2. Hexanal Calibration Curve on Day3

 

APPENDD( C

 

 

 

1 1 8
APPENDD( C

Original Experimental GC Data

The original experimental GC data were basically area responses
recorded at particular retention times for the volatile standards.

The area responses recorded for butanal, cycloheptatriene,
pentanal, and hexanal after 3 days storage, were respectively
presented in Table C-1, C-2, C-3, and C-4.

Data were transformed from area responses to relative %

concentration and shown in Table C-5, C-6, C-7, and C-8.

”I‘m...

 

1 1 9
Table C-1. Area Response of Butanal in Cooperation with Adsorbents

after 3 Days Storage

Adsorbents Dayl 22122 20123
Silica Gel 1612 384 ND
305 ND ND
9058 3767 953
Activated 346 ND ND
Carbon 1921 1193 599
206 ND ND
Sodium 73278 47932 23024
Bisulfite 30220 2221 330
82895 82959 45097
Cellulose 140450 112820 98827
160020 128330 108780
184770 162870 130590
Lactose 180700 159330 133130
190785 170235 150750
200870 181140 168370
Pectin 209470 192470 156640
194320 197940 176300
208070 180880 147940

 

*All samples were three replications.

 

-.-. .I-
.- .-
1 $8151
" {'1‘1
.1_.'
If.
. i
. "
.z .

 

120
Table C-2. Area Response of Cycloheptatriene in Cooperation with

Adsorbents after 3 Days Storage

Adsorbents 29121 12312 00123
Silica Gel 1346 802 982
686 598 ND
748 502 ND
Activated ND ND ND
Carbon ND ND ND
ND ND ND
Sodium 394350 368710 362780
Bisulfite 402720 280330 326920
396970 333000 282140
Cellulose 368530 357970 333 130
381020 328325 286060
335620 298680 267600
Lactose 382500 385680 346560
403700 342830 309920
398860 384990 310510
Pectin 373670 337800 318830
400660 350980 332460
378100 326580 277690

 

*All samples were three replications.

 

1 2 1
Table C-3. Area Response of Pentanal in Cooperation with Adsorbents

after 3 Days Storage

Adsorbents Dani 13.8122 Dax3
Silica Gel ND ND ND
ND ND ND
ND ND ND
Activated ND ND ND
Carbon ND ND ND
ND ND ND
Sodium 236720 212330 184300
Bisulfite 260540 213050 222000
240690 193230 162760
Cellulose 22 1080 199190 166110
193250 158010 126240
236730 203660 163820
lactose 23 1660 203590 171210
249200 221300 153480
257300 207530 152830
Pectin 258560 223470 185080
248070 207300 185430
245310 197470 181260

 

*All samples were three replications.

 

_ 1' I
I)‘:

 

122
Table C-4. Area Response of Hexanal in Cooperation with Adsorbents

after 3 Days Storage

Aghmuhenns 1203. 12nd: ;Dax3
Silica Gel 334560 129670 23241
258170 127320 41274
243030 76010 59992
Activated 260320 49952 13933
Carbon 220140 42357 43394
204560 81885 61438
Sodhnn. 242810 128680 10197
Bisulfite 472240 247590 36238
325310 150149 53568
Cellulose 374220 218840 138350
238640 152410 102380
293540 143240 105000
Lactose 320920 188650 102180
341320 231860 168270
229550 229300 169280
Pectin 446730 234360 173750
392230 33 1710 283070
411400 333080 163250

 

*All samples were three replications.

 

 

. '21r1's'1‘

 

123
Table C-5. Relative % Concentration of Butanal in Cooperation with

Adsorbents after 3 Days Storage.

 

Adsorbents 2.421 26122 22232
Silica Gel 0.7689 0.1832 0.0000
0.1455 0.0000 0.0000
4.3208 1.7969 0.4546
Activated 0.1650 0.0000 0.0000
Carbon 0.9163 0.5691 0.2857
0.0982 0.0000 0.0000
Sodium 34.9548 22.8643 10.9828
Bisulfite 14.4154 1.0594 0.1574
39.5422 39.5928 21.5120
Cellulose 66.9969 53.8169 47.1420
76.3321 61.2154 51.8898
88.1382 77.6916 62.2935
Lactose 86.1968 76.0029 63.505 1
91.0074 81.2048 71.9101
88.1382 77.6916 62.2935
Pectin 99.9205 91.8112 74.7197
92.6937 94.4205 84.0979
99.2527 86.2826 70.5697

 

*All samples were three replications.

 

 

124
Table C-6. Relative % Concentration of Cycloheptatriene in

Cooperation with Adsorbents after 3 Days Storage

Adsorbents 12m 1201a 129.1232
Silica Gel 0.3333 0.1986 0.2432
0.1699 0.1481 0.0000
0.1852 0.1243 0.0000
Activated 0 0 0
Carbon 0 0 0
0 0 0
Sodium 97.65 17 91.3025 89.8341
Bisulfite 99.7243 69.4172 80.9542
98.3004 82.4598 69.8655
Cellulose 91.2579 88.6430 82.4920
94.3508 81.302 1 70.8361
83.1085 73.9612 66.2650
Lactose 94.7173 95.5047 85.8176
99.9670 84.8939 76.7445
98.7685 95.3339 76.8906
Pectin 92.5307 83.6484 78.9509
99.2142 86.912 1 82.3260
93.6277 808700 68.7635

 

*All samples were three replications.

 

 

125
Table C-7. Relative % Concentration of Pentanal in Cooperation with

Adsorbents after 3 Days Storage

Adsorbents M 123122 1.3.2123
Silica Gel 0 0 0
0 0 0
0 0 0
Activated 0 0 0
Carbon 0 0 0
0 0 0
Sodium 90.6175 81.2809 70.5509
Bisulfite 99.7359 81.5565 84.2170
92.1372 73.9693 62.3052
Cellulose 84.6304 76.2508 63.5876
73.9770 60.4869 48.3252
90.6213 77.9619 62.7110
Lactose 88.6805 77.9352 65.5399
95.3949 84.7146 58.7528
96.5816 79.4434 58.5040
Pectin 98.9779 85.5453 60.9322
94.9623 79.3554 70.9834
93.9058 75.5924 69.3871

 

*All samples were three replications.

 

 

 

126
Table C-8. Relative % Concentration of Hexanal in Cooperation with

Adsorbents after 3 Days Storage

Adsorbents _aLD 1 _aLD 2 _axSD
Silica Gel 69.7974 27.0523 4.8486
53.8606 26.5621 8.6108
50.7020 15.8575 12.5158
Activated 54.3091 10.4212 2.9068
Carbon 45.9266 8.8367 9.0530
42.6762 17.0832 12.8175
Sodium 50.6561 26.8458 2.1273
Bisulfite 98.5208 51.6500 7.5601
67.8676 31.3229 11.1756
Cellulose 78.0715 45.6554 28.8632
49.7862 31.7965 21.3950
61.2396 29.8834 21.9056
Lactose 66.9918 39.3570 21.3172
71.2077 48.3717 35.1052
47.8898 47.8376 35,3160
Pectin 93.1988 48.8932 36.2485
81.8288 69.2028 59.0553
85.8281 69.4887 34.0580

 

*All samples were three replications.

127

Table C-9. Area Response of Volatile Standards in the Headspace
after 24, 48, and 72 Hours Storage at 210C

Adsottientllme Butanal chklmnta: Pentanal Hexanal
(hrs) 111920.62
24 3658,33 926.67 ND* 278586.67
Silica gel 48 1383.67 634.00 ND 111000.00
72 317.67 327.33 ND 41502.33
24 824.33 ND ND 228340.00
Activated 48 397.67 ND ND 58064.67
carbon 72 199.67 ND ND 39588.33
24 62131.00 398013.33 245983.33 346786.67
Sodium 48 44370.67 327346.67 206203.33 175473.00
Bisulfite 72 22817.00 323946.67 189020.00 33334.33
24 161746.67 361723.33 217020.00 302133.33
Cellulose 48 134673.33 328325.00 186953.33 171496.67
72 112732.33 295596.67 152056.67 115243.33
24 190785.00 395020.00 244386.67 297263.33
Lactose 48 170235.00 371166.67 210806.67 216603.33
72 150750.00 322330.00 159173.33 146576.67
24 203953.33 384143.33 250646.67 416786.67
Pectin 48 190430.00 338453.33 209413.33 299716.67
72 160293.33 309660.00 183923.33 206690.00

 

*ND: Not Detectable

Illluu

 

APPENDD( D

 

 

1 2 8
APPENDD( D

Statistical Analysis

The statistical analysis was carried out by MSTATC
microcomputer statistical program (Michigan State University, 1989).
Function: FACTOR
Experiment Model Number 5:

Completely Randomized Design for Factor A and B, Factor C is a

Split Plot on A and B
Data case no. 1 to 216
Factorial ANOVA for the factors:

 

Replication (Var 1: replicate) with values from 1 to 3
Factor A (Var 2: volatile standard) with values from 1 to 4
Factor B (Var 3: adsorbent) with values from 1 to 6

Factor C (Var 4: day) with values from 1 to 3

 

 

 

Variable 5: GC
Grand Mean= 48.066 Grand Sum= 103 82.202 Total Count= 216
TABLE OF MEANS
1 2 3 4 5 Total
* 1 * * 42.712 2306.437
* 2 * * 57.197 3088.613
* 3 * * 5 1.961 2805.904
* 4 * * 40.393 2181.248
* * 1 * 7.746 278.844
* * 2 * 5.724 206.065
* * 3 * 56.908 2048.693
* * 4 * 64.462 2320.638
* * 5 * 73.886 2659.906
* * 6 * 79.668 2868.056

 

 

129

 

 

* 1 1 * 0.852 7.670
* 1 2 * 0.226 2.034
* 1 3 * 20.565 185.081
* 1 4 * 65.057 585.5 16
* 1 S * 81.374 732.367
* 1 6 * 88.187 793.769
* 2 1 * 0.156 1.403
* 2 2 * 0.000 0.000
* 2 3 * 86.612 779.5 12
* 2 4 * 81.357 732.216
* 2 5 * 89.849 808.638
* 2 6 * 85.205 766.843
* 3 1 * 0.000 0.000
* 3 2 * 0.000 0.000
* 3 3 * 81.819 736.370
* 3 4 * 70.483 634.345
* 3 5 * 78.394 705.547
* 3 6 * 81.071 729.642
* 4 1 * 29.975 269.771
* 4 2 * 22.670 204.030
* 4 3 * 38.637 347.730
* 4 4 * 40.951 368.560
* 4 5 * 45.928 413.354
* 4 6 * 64.200 577.802
* * * 1 59.807 4306.132
* * * 2 46.801 3369.671
* * * 3 37.589 2706.399
* 1 * 1 49.538 891.684
* 1 * 2 43.05 1 774.918
* 1 * 3 35.546 639.835
* 2 * 1 63.550 1143.907
* 2 * 2 56.365 1014.574
* 2 * 3 51.674 930.131
* 3 * 1 61.123 1100.222
* 3 * 2 5 1.894 934.093
* 3 * 3 42.866 771.589
* 4 * 1 65.018 1170.319
* 4 * 2 35.894 646.085
* 4 * 3 20.629 364.844

 

 

 

 

130

 

* * 1 1 15.024 180.284
* * 1 2 5.978 71.739
* * 1 3 2.235 26.821
* * 2 1 12.008 144.091
* * 2 2 3.076 36.910
* * 2 3 2.089 25.063
* * 3 1 73.677 884.124
* * 3 2 54.444 653.327
* * 3 3 42.604 5 11.242
* * 4 1 78.209 938.5 10
* * 4 2 63.222 758.665
* * 4 3 51.955 623.463
* * 5 1 86.098 1033.181
* * 5 2 74.751 897.006
* * 5 3 60.810 729.718
* * 6 1 93.828 1125.941
* * 6 2 79.335 952.023
* * 6 3 65.841 790.092
* 1 1 1 1.745 5.235
* 1 1 2 0.660 1.980
* 1 1 3 0.152 0.455
* 1 2 1 0.393 1.180
* 1 2 2 0.190 0.569
* 1 2 3 0.095 0.286
* 1 3 1 29.637 88.912
* 1 3 2 21.172 63.5 16
* 1 3 3 10.884 32.652
* 1 4 1 77.156 231.467
* 1 4 2 64.241 192.724
* 1 4 3 53.775 161.325
* 1 5 1 91.007 273.022
* 1 5 2 81.205 243.614
* 1 5 3 71.910 215.730
* 1 6 1 97.289 291.867
* 1 6 2 90.838 272.515
* 1 6 3 76.462 229.387
* 2 1 1 0.229 0.688
* 2 1 2 0.108 0.323
* 2 1 3 0.130 0.391
* 2 2 1 0.000 0.000
* 2 2 2 0.000 0.000
* 2 2 3 0.000 0.000

 

x-x-x-4x—x-x-x—aex-x-x-x-x-x-x-4»x-x-x-x-x-x-**4»x—>+>ex—x-x-x—x-x—:ex-x~x—x—x—

.54:ka4:43kaLLAAhWWWWWWWWWWWWWWWWWWNNNNNNNNNNNN

UTAAAUJLNUJNNNHHHCNGGU'IUIMh-k-hUUWWNNNr—Ir—ll—‘O‘GGMMM-fi#4300000.)

HWNHWNHWNHWNl—‘WNHWNHWNHWNHWNHWNHWNHWNHWNHWNH

131

98.559
81.061
80.218
89.572
81.302
73.198
97.818
91.911
79.818
95.124
83.810
76.680

0.000

0.000

0.000

0.000

0.000

0.000
94.164
78.936
72.358
83.076
71.567
56.806
93.552
80.698
60.932
95.949
80.164
67.101
58.120
23.145

8.658
47.637
12.114

8.259
72.348
36.607

6.954
63.032
35.778
24.043
62.016

295.676
243.182
240.654
268.717
243.906
219.593
293.453
275.732
239.453
285.373
251.431
230.040
0.000
0.000
0.000
0.000
0.000
0.000
282.491
236.807
217.073
249.229
214.700
170.417
280.657
242.093
182.797
287.846
240.493
201.303
174.360
69.436
25.975
142.912
36.341
24.777
217.044
109.822
20.863
189.097
107.335
72.128
186.049

132

 

 

 

 

* 4 5 2 45.189 135.586
* 4 5 3 30.579 91.738
* 4 6 1 86.952 260.856
* 4 6 2 62.528 187.585
* 4 6 3 43.121 129.362
ANALYSIS OF VARLANCE TABLE
K Degrees of Sum of Mean F
Value Source Freedom Squares Square Value
2 Factor A 3 10048.009 3349.336 30.4868
4 Factor B 5 195515.242 39103.048 355.9291
6 AB 15 48740.349 3249.357 29.5767
-7 Error 48 5273.371 109.862
8 Factor C 2 17944.602 8972.301 285.9067
10 AC 6 6677.277 1112.880 35.4624
12 BC 10 2427.733 242.773 7.7361
14 ABC 30 2188.714 72.957 2.3248
-15 Error 96 3012.664 31.382
Total 2 15 291827.962

Conditional Comparisons

The judgment of significance depends on the difference between
any two means. Any value of the difference between two means
larger than minimum significant difference (MSD) will be considered
significant.

There are two MSD; one for the volatile standards and the other
for the adsorbents. The following equations are employed to calculate
MSD.

t=(y1—y2)/\/262/3

MSD=(tB,O.025,m=6,0) 2‘32 /3

t= (y, -72)/x/262 /3
MSD = (tB,o.025,m=1s,o) V 262 / 3

Equations for volatile standards

Equations for adsorbents

 

133

62 = (M5,,l + 2mm) / 3

., mug/3) (2MsE,/3)2
m1 1/{(, ,6 }

62 = [ 109.862 + 2(31.382)]/3 = 57.542

= (57,542)2 / { [(109.862/3)2 /48] + [(2x31.382/3)2 /96] }
= 101.886

C>

tB.0.025,m=6,i)= t8.0.025.m=6,101.886 = 2'692 (Gill, 1987)

-.3 007 (Gill, 1987)

tB.0.025,m=15,= v ’13 0.025.m=15101.886=

 

MSD for volatile standards = (2.692) 2675542) = 16.674

 

MSD for adsorbents = (3.007) 2675542) = 18.625

Within the same day (three days) and the same adsorbent (six
adsorbents), comparison between any two out of four volatile
standards was executed as the following. The difference between two
means (Z—Z ) corresponding to the volatile standards was compared
to MSD for significant difference.

Daxl
Within Silica Gel 171 .. y;
Butanal vs. Cycloheptatriene 1.745 - 0.229 = 1.516
Butanal vs. Pentanal 1.745 - 0.000 = 1.745
Butanal vs. Hexanal 1,745 - 58.120 = 56.375

Cycloheptatriene vs. Pentanal 0.229 - 0.000 = 0.229
Cycloheptatriene vs. Hexanal 0.229 - 58.120 = 57.891
Pentanal vs. Hexanal 0.000 - 58.120 = 58.120

 

 

134

Within Activated Carbon

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Sodium Bisulfite

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Cellulose

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Lactose

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Pectin

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Y1 "Y2
0.393 - 0.000 = 0.393
0.393 - 0.000 = 0.393
0.393 - 47.637 = 47.244
0.000 - 0.000 = 0.000

0.000 - 47.637 = 47.637
0.000 - 47.637 = 47.637

Y1 ”Y2
29.637 - 98.559 = 68.922
29.637 - 94.164 = 64.527
29.637 - 72.348 = 42.711
98.559 - 94.164 = 4.395
98.559 - 72.348 = 26.211
94.164 - 72.348 = 21.816

Y1 “Y2
77.156 - 89.572 = 12.416
77.156 - 83.076 = 5.92
77.156 - 63.032 = 14.124
89.572 - 83.076 = 6.496
89.572 - 63.032 = 26.54
83.076 - 63.032 = 20.044

Y1 "Y2
91.007 - 97.818 = 6.811
91.007 - 93.552 = 2.545
91.007 - 62.016 = 28.991
97.818 - 93.552 = 4.266
97.818 - 62.016 = 35.802
93.552 - 62.016 = 31.536

yT—E
97.289 - 95.124 = 2.165
97.289 - 95.949 = 1.340

97.289 - 86.952 = 10.337

95.124 - 95.949 = 0.825
95.124 - 86.952 = 8.172
95.949 - 86.952 = 8.997

 

DaXZ

135

Within Silica Gel

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Activated Carbon

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Sodium Bisulfite

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Cellulose

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Lactose

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Y1 “Y2
0.660 - 0.108 = 0.552
0.660 - 0.000 = 0.660
0.660 - 23.145 = 22.485
0.108 - 0.000 = 0.108

0.108 - 23.145 = 23.037
0.000 - 23.145 = 23.145

Y1 “Y2
0.190 - 0.000 = 0.190
0.190 - 0.000 = 0.190
0.190 " 12.144 = 11.954
0.000 - 0.000 = 0.000

0.000 - 12.144 = 12.144
0.000 - 12.144 = 12.144

Y1 “Y2
21.172 - 81.061 = 59.889
21.172 - 78.936 = 57.764
21.172 - 36.607 = 15.435
81.061 - 78.936 = 2.125
81.061 - 36.607 = 44.454
78.936 - 36.607 = 42.329

Y1 “Y2
64.241 - 81.302 = 17.061
64.241 - 71.567 = 7.326
64.241 - 35.778 = 28.463
81.302 - 71.567 = 9.735
81.302 - 35.778 = 45.524
71.567 - 35.778 = 35.789

Y1 “Y2
81.205 - 91.911 = 10.706
81.205 - 80.698 = 0.507
81.205 - 45.189 = 36.016
91.911 - 80.698 = 11.213
91.911 - 45.189 = 46.722
80.698 - 45.189 = 35.509

 

 

136

Within Pectin

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Silica Gel

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Activated Carbon

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Sodium Bisulfite

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Within Cellulose

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Y1 “Y2
90.838 - 83.810 = 7.028
90.838 - 80.164 = 10.674
90.838 - 62.528 = 28.310
83.810 - 80.164 = 3.646
83.810 - 62.528 = 21.282
80.164 - 62.528 = 17.636

Y1 “Y2
0.152 - 0.130 = 0.022
0.152 - 0.000 = 0.152
0.152 - 8.658 = 8. 506
0.130 - 0.000 = 0.130
0.130 - 8.658 = 8.528
0.000 - 8.658 = 8.658

Y1 “Y2
0.095 - 0.000 = 0.095
0.095 - 0.000 = 0.095
0.095 - 8.259 = 8.164
0.000 - 0.000 = 0.000
0.000 - 8.259 = 8.259
0.000 - 8.259 = 8.259

Y1 “Y2

10.884 - 80.218 = 69.334
10.884 - 72.358 = 61.474
10.884 - 6.954 = 3.930
80.218 - 72.358 = 7.860
80.218 - 6.954 = 73.264
72.358 - 6.954 = 65.404

Y1 “Y2
53.775 - 73.198 = 19.423
53.775 - 56.806 = 3.031
53.775 - 24.043 = 29.732
73.198 - 56.806 = 16.392
73.198 - 24.043 = 49.155
56.806 - 24.043 = 32.763

 

 

137

Within Lactose

Butanal vs. Cycloheptatriene
Butanal vs. Pentanal

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal
Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

Y1 “Y2
71.910 - 79.818 = 7.908
71.910 - 60.932 = 10.978
71.910 - 30.579 = 41.331
79.818 - 60.932 = 18.886
79.818 - 30.579 = 49.239
60.932 - 30.579 = 30.353

Within Pectin 271-7;
Butanal vs. Cycloheptatriene 76.462 - 76.680 = 0.218
Butanal vs. Pentanal 76.462 - 67.101 = 9.361

Butanal vs. Hexanal
Cycloheptatriene vs. Pentanal

76.462 - 43.121 = 33.341
76.680 - 67.101 = 9.579

76.680 - 43.121 = 33.559
67.101 - 43.121 = 23.980

Cycloheptatriene vs. Hexanal
Pentanal vs. Hexanal

 

Within the same day (three days) and the same volatile standard
(four volatile standards), comparison between any two out of six
adsorbents was executed as the following. The difference between
two means (17,—): ) corresponding to the adsorbents was compared to
MSD for significant difference.

Dayl
Within Butanal

Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite
Silica gel vs. Cellulose
Silica gel vs. Lactose
Silica gel vs. Pectin
Activated Carbon vs. Sodium Bisulfite
Activated Carbon vs. Cellulose
Activated Carbon vs. Lactose
Activated Carbon vs. Pectin
Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. lactose
Sodium Bisulfite vs. Pectin
Cellulose vs. Lactose
Cellulose vs. Pectin
Lactose vs. Pectin

Y1 “Y2
1.745 ~ 0.393 = 1.352
1.745 - 29.637 = 27.892
1.745 - 77.156 = 75.411
1.745 - 91.007 = 89.262
1.745 - 97.289 = 95.544
0.393 - 29.637 = 29.244
0.393 - 77.156 = 76.763
0.393 - 91.007 = 90.614
0.393 - 97.289 = 96.896
29.637 - 77.156 = 47.519
29.637 - 91.007 = 61.370
29 637 - 97.289 = 67.652
77.156 - 91.007 = 13.851
77.156 - 97.289 = 20.133
91.007 - 97.289 = 6.282

 

   

Within Cycloheptatriene

Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite

138

Silica gel vs. Cellulose

Silica gel vs. Lactose
Silica gel vs. Pectin
Activated Carbon vs

Activated Carbon vs
Activated Carbon vs

Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. lactose

Sodium Bisulfite vs.

. Sodium Bisulfite

Activated Carbon vs. Cellulose
. Lactose

. Pectin

Pectin

Cellulose vs. lactose

Cellulose vs. Pectin
Lactose vs. Pectin

Within Pentanal

Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite

Silica gel vs. Cellulose

Silica gel vs. Lactose
Silica gel vs. Pectin

Activated Carbon vs. Sodium Bisulfite
Activated Carbon vs. Cellulose
Activated Carbon vs. lactose

Activated Carbon vs

Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. Lactose

Sodium Bisulfite vs.

. Pectin

Pectin

Cellulose vs. lactose

Cellulose vs. Pectin
Lactose vs. Pectin

Within Hexanal

Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite

Silica gel vs. Cellulose

Silica gel vs. lactose
Silica gel vs. Pectin

Y1 “Y2
0.229 - 0.000 = 0.229
0.229 - 98.559 = 98.330
0.229 - 89.572 = 89.343
0.229 - 97.818 = 97.589
0.229 - 95.124 = 94.895
0.000 - 98.559 = 98.559
0.000 - 89.572 = 89.572
0.000 - 97.818 = 97.818
0.000 - 95.124 = 95.124

98,559 - 89.572 = 8.987
98.559 - 97.818 = 0.741
98.559 - 95.124 = 3.435
89.572 - 97.818 = 8.246
89.572 - 95.124 = 5.552
97.818 - 95.124 = 2.694
Y1 “Y2
0.000 - 0.000 = 0.000

0.000 - 94.164 = 94.164
0.000 - 83.076 = 83.076
0.000 - 93.552 = 93.552
0.000 - 95.949 = 95.949
0.000 - 94.164 = 94.164
0.000 - 83.076 = 83.076
0.000 - 93.552 = 93.552
0.000 - 95.949 = 95.949
96.164 - 83.076 = 13.088
96.164 - 93.552 = 2.612
96.164 - 95.949 = 0.215
83.076 - 93.952 = 10.476
83.076 - 95.949 = 12.873
93.552 - 95.949 = 2.397

Y1 “Y2
58.120 - 47.637 = 10.483
58.120 - 72.348 = 14.228
58.120 - 63.032 = 4.912
58.120 - 62.016 = 3.896
58.120 - 86.952 = 28.832

Activated Carbon vs. Sodium Bisulfite 47.637 - 72.348 = 24.711

 

139

Activated Carbon vs. Cellulose
Activated Carbon vs. Lactose
Activated Carbon vs. Pectin
Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. lactose
Sodium Bisulfite vs. Pectin
Cellulose vs. lactose

Cellulose vs. Pectin

Lactose vs. Pectin

Within Butanal

Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite
Silica gel vs. Cellulose

Silica gel vs. lactose

Silica gel vs. Pectin

Activated Carbon vs. Sodium Bisulfite

Activated Carbon vs. Cellulose
Activated Carbon vs. Lactose
Activated Carbon vs. Pectin
Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. Lactose
Sodium Bisulfite vs. Pectin
Cellulose vs. lactose

Cellulose vs. Pectin

Lactose vs. Pectin

Within Cycloheptatriene

Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite
Silica gel vs. Cellulose

Silica gel vs. Lactose

Silica gel vs. Pectin

Activated Carbon vs. Sodium Bisulfite

Activated Carbon vs. Cellulose
Activated Carbon vs. lactose
Activated Carbon vs. Pectin
Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. Lactose
Sodium Bisulfite vs. Pectin
Cellulose vs. lactose

47.637 - 63.032 = 15.395
47.637 - 62.016 = 14.379
47.637 - 86.952 = 39.315
72.348 - 63.032 = 9.316
72.348 - 62.016 = 10.332
72.348 - 86.952 = 14.604
63.032 - 62.016 = 1.204
63.032 - 86.952 = 23.920
63.032 - 86.952 = 24.936

Y1 “Y2
0.660 - 0.190 = 0.470
0.660 - 21.172 = 20.512
0.660 - 64.241 = 63.581
0.660 - 81.205 = 80.545
0.660 - 90.838 = 90.178
0.190 - 21.172 = 20.982
0.190 - 64.241 = 64.051
0.190 - 81.205 = 81.015
0.190 - 90.838 = 90.648
21.172 - 64.241 = 43.069
21.172 - 81.205 = 60.033
21.172 - 90.838 = 69.666
64.241 - 81.205 = 16.964
64.241 - 90.838 = 26.597
81.205 - 90.838 = 9.633

Y1 “Y2
0.108 - 0.000 = 0.108
0.108 - 81.061 = 80.953
0.108 - 81.302 = 81.194
0.108 - 91.911 = 91.803
0.108 - 83.810 = 83.702
0.000 - 81.061 = 81.061
0.000 - 81.302 = 81.302
0.000 - 91.911 = 91.911
0.000 - 83.810 = 83.810
81.061 - 81.302 = 0.241
81.061 - 91.911 = 10.850
81.061 - 83.810 = 2.749
81.302 - 91.911 = 10.609

 

140

Cellulose vs. Pectin 81.302 - 83.810 = 2.508
lactose vs. Pectin 91.911 - 83.810 = 8.101
Within Pentanal Z—Y,
Silica gel vs. Activated Carbon 0.000 - 0.000 = 0.000
Silica gel vs. Sodium Bisulfite 0.000 - 78.936 = 78.936
Silica gel vs. Cellulose 0.000 - 71.567 = 71.567
Silica gel vs. Lactose 0.000 - 80.698 = 80.698
Silica gel vs. Pectin 0.000 - 80.164 = 80.164

Activated Carbon vs. Sodium Bisulfite 0.000 - 78.936 = 78.963
Activated Carbon vs. Cellulose 0.000 - 71.567 = 71.567

 

Activated Carbon vs. Lactose 0.000 - 80.698 = 80.698
Activated Carbon vs. Pectin 0.000 - 80.164 = 80.164
Sodium Bisulfite vs. Cellulose 78.936 - 71.567 = 7.3 69
Sodium Bisulfite vs. Lactose 78.936 - 80.698 = 1.762
Sodium Bisulfite vs. Pectin 78.936 - 80.164 = 1.228
Cellulose vs. lactose 71.567 - 80.698 = 9.131 1
Cellulose vs. Pectin 71.567 - 80.164 = 8.597 '
lactose vs. Pectin 80.698 - 80.164 = 0.534
Within Hexanal 59;;

Silica gel vs. Activated Carbon 23.145 - 12.114 = 11.031
Silica gel vs. Sodium Bisulfite 23.145 - 36.607 = 13.462

Silica gel vs. Cellulose 23.145 - 35.778 = 12.633
Silica gel vs. Lactose 23.145 - 45.189 = 22.044
Silica gel vs. Pectin 23.145 - 62.528 = 39.383

Activated Carbon vs. Sodium Bisulfite 12.114 - 36.607 = 24.493
Activated Carbon vs. Cellulose 12.114 - 35.778 = 23.664
Activated Carbon vs. lactose 12.114 - 45.189 = 33.075

Activated Carbon vs. Pectin 12.114 - 62.528 = 50.414
Sodium Bisulfite vs. Cellulose 36.607 - 35.778 = 0.289
Sodium Bisulfite vs. Lactose 36.607 - 45.189 = 8.5 82
Sodium Bisulfite vs. Pectin 36.607 - 62.528 = 25.921
Cellulose vs. Lactose 35.778 - 45.189 = 9.411
Cellulose vs. Pectin 35.778 - 62.528 = 26.750
Lactose vs. Pectin 45.189 - 62.528 = 17.339
123123 _ _
Within Butanal y1 — y,

Silica gel vs. Activated Carbon 0.152 - 0.095 = 0.057

 

141

Silica gel vs. Sodium Bisulfite

Silica gel vs. Cellulose
Silica gel vs. Lactose
Silica gel vs. Pectin

Activated Carbon vs. Sodium Bisulfite
Activated Carbon vs. Cellulose
Activated Carbon vs. Lactose
Activated Carbon vs. Pectin
Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. Lactose

Sodium Bisulfite vs. Pectin
Cellulose vs. Lactose
Cellulose vs. Pectin
lactose vs. Pectin

Within Cycloheptatriene
Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite

Silica gel vs. Cellulose
Silica gel vs. Lactose
Silica gel vs. Pectin

Activated Carbon vs. Sodium Bisulfite
Activated Carbon vs. Cellulose
Activated Carbon vs. Lactose
Activated Carbon vs. Pectin
Sodium Bisulfite vs. Cellulose
Sodium Bisulfite vs. lactose

Sodium Bisulfite vs. Pectin
Cellulose vs. Lactose
Cellulose vs. Pectin
lactose vs. Pectin

Within Pentanal
Silica gel vs. Activated Carbon
Silica gel vs. Sodium Bisulfite

Silica gel vs. Cellulose
Silica gel vs. Lactose
Silica gel vs. Pectin

Activated Carbon vs. Sodium Bisulfite
Activated Carbon vs. Cellulose
Activated Carbon vs. lactose
Activated Carbon vs. Pectin

0.152 - 10.884 = 10.732
0.152 - 53.775 = 53.623
0.152 - 71.910 = 71.758
0.152 - 76.462 = 76.310
0.095 - 10.884 = 10.789
0.095 - 53.775 = 53.680
0.095 - 71.910 = 71.815
0.095 - 76.462 = 76.367
10.884 - 53.775 = 42.891
10.884 - 71.910 = 61.026
10.884 - 76.462 = 65.578
53.775 - 71.910 = 18.135
53.775 - 76.462 = 22.687
71.910 - 76.462 = 4.552

Y1 “Y2
0.130 - 0.000 = 0.130
0.130 - 80.210 = 80.080
0.130 - 73.198 = 73.068
0.130 - 79.818 = 79.688
0.130 - 76.680 = 76.550
0.000 - 80.210 = 80.210
0.000 - 73.198 = 73.198
0.000 - 79.818 = 79.818
0.000 - 76.680 = 76.680
80.210 - 73.198 = 7.012
80.210 - 79.818 = 0.392
80.210 - 76.680 = 3.530
73.198 - 79.818 = 6.620
73.198 - 76.680 = 3.482
79.818 - 76.680 = 3.318

Y1 “Y2
0.000 - 0.000 = 0.000
0.000 - 72.358 = 72.358
0.000 - 56.806 = 56.806
0.000 - 60.932 = 60.932
0.000 - 67.101 = 67.101
0.000 - 72.358 = 72.358
0.000 - 56.806 = 56.806
0.000 - 60.932 = 60.932
0.000 - 67.101 = 67.101

 

142
Sodium Bisulfite vs. Cellulose 72.358 - 56.806 = 15.552

 

Sodium Bisulfite vs. Lactose 72.358 - 60.932 = 11.426
Sodium Bisulfite vs. Pectin 72.358 - 67.101 = 5.257
Cellulose vs. lactose 56.806 - 60.932 = 4.126
Cellulose vs. Pectin 56.806 - 67.101 = 10.295
Lactose vs. Pectin 60.932 - 67.101 = 6.169
Within Hexanal 17,—};
Silica gel vs. Activated Carbon 8.658 - 0.393 = 0.399
Silica gel vs. Sodium Bisulfite 8.658 - 6.954 = 1.704
Silica gel vs. Cellulose 8.658 - 24.043 = 15.385
Silica gel vs. Lactose 8.658 - 30.579 = 21.921
Silica gel vs. Pectin 8.658 - 43.121 = 34.463
Activated Carbon vs. Sodium Bisulfite 8.259 - 6.954 = 1.305
Activated Carbon vs. Cellulose 8.259 - 24.043 = 15.784
Activated Carbon vs. lactose 8.259 - 30.579 = 22.320
Activated Carbon vs. Pectin 8.259 - 43.121 = 34.862
Sodium Bisulfite vs. Cellulose 6.954 - 24.043 = 17.089
Sodium Bisulfite vs. Lactose 6.954 - 30.579 = 23.625
Sodium Bisulfite vs. Pectin 6.954 - 43.121 = 36.167
Cellulose vs. Lactose 24.043 - 30.579 = 6.536
Cellulose vs. Pectin 24.043 - 43.121 = 19.078

lactose vs. Pectin 30.579 - 43.121 = 12.542

 

 

APPENDD( E

143
APPENDD( E

CONSENT FORM
SCHOOL OF PACKAGING
MICHIGAN STATE UNIVERSITY

PRODUCT INGREDIENTS: FROZEN FISH (LAKE TROUT), SILICA GEL,
ACTIVATED CARBON

I, , have read the above list or
ingredients and find none that I know I am allergic to. I have also
been informed of the nature of the research, including experimental
materials and procedures, which will be used during the sensory
testing sessions. I agree to serve on this sensory panel, which is
being conducted on this ______ day of , 1993. In addition to
sniffing samples, I will be asked to complete a brief questionnaire. I
understand that I am free to withdraw my consent and to
discontinue participation in the panel at any time without penalty. I
understand that if I am injured as a result of my participation in this
research project, Michigan State University will provide emergency
medical care, if necessary, but these and any other medical expenses

must be paid from my own health insurance program.

SIGNED DATE

 

 

 

144

QUESTIONNAIRE FOR SIMPLE PAIRED COMPARISONS TEST
NAME: DATE:

 

PRODUCT: FROZEN FISH (LAKE TROUT)

Evaluate the off-odor of these two samples of frozen lake trout.

Please follow the instructions stated below:

a. When ready to begin, remove the cap carefully from the first
bottle on your left.

b. Without delay, inhale deeply from the open top of the sample
bottle using nostrils only (mouth should be closed).

c. After sniffing, replace the seal quickly.

d. Take several deep breaths before going on to the next sample.

e. Repeat steps a. through c. for the other sample.

f. Indicate which sample has stronger off—odor.

g. If no difference is apparent, enter your best guess.

Test Pairs 'h l h r ff- r

Comments:

 

 

 

 

 

 

 

LIST OF REFERENCES

 

 

LIST OF REFERENCES

Ackman, R.G. and Ratnayake, W.M.N. 1989b. Lipid analysis. In "The
Role of Fats in Human Nutrition", A.J. Vergroesen and M.Crawford
(Ed.), p.442. Academic Press. New York, NY.

Aitken, A. and Connell, J .J . 1979. Fish, in "Effects of Heating on
Foodstuffs", R.J. Preistly (Ed.), p.23 8-245. Applied Science
Publishers, Essex, UK.

Amerine, M.A., Pangbom, RM. and Roessler, EB. 1965. Principles of
Sensory Evaluation of Food, Academic Press, New York, NY.

Andersson, K. and Danielson, GE. 1961. Storage changes in frozen fish:
A comparison of objective and subjective tests. Food Technol.
15:55.

 

ASTM Committee E-18, Manual on Sensory Testing Methods, ASTM
STP 434, American Society for Testing and Materials, Philadelphia,
PA, 1968.

Badings, HT. 1970. Cold storage defects in butter and their relation to
the autoxidation of unsaturated fatty acids. Neth. Milk Dairy J.
24:147.

Badings, HT. 1973. Fishy off-flavors in autoxidized oils. J. Am. Oil
Chem. Soc. 50:334.

Bandal, R.G., Donnet, J .B., and Stoeckli, F. 1988. Characterization of
active carbons, in "Active Carbon", R.C. Bansal, J .B. Donnet, and F.
Stoeckli (Ed.), Chap. 4, p. 189-190. Marcel Dekker, Inc., New York,
NY

Barbut, S., Josephson, DB. and Maurer, A.J. 1985. Antioxidant
properrties of rosemary oleoresin in turkey sausage. J. Food Sci.
50:1356.

Bauemfeind, J .C., Smith, EC. and Siemers, G.F. 195 1. Commercial
processing of frozen fish with ascorbic acid. Food Technol. 5:24.

Beaumariage, D.S., Ingle, RM. and Joyce, E.A.Jr. 1969. Prevention of
deteriorative changes in frozen skinless mullet (Mugil cephalus)

145

 

 

146

fillets. Marine Research lab., Florida Board of Conservation, St.
Petersburg, FL.

Bertsch, W., Anderson, E. and Holzer, G. 1975. Trace analysis of
organic volatiles in water by gas chromatography mass
spectrometry with glass capillary columns. J. Chromat. 1 12: 701.

Berra, T.M., Smith, J .F., and Morrison, JD. 1982. Probable
identification of the cucumber odor of the austrlian grayling
Prototroctes maraena. Trans. Am. Fish. Soc. 11 1:78.

Bilinski, E., Jonas, R.E.E. and lau, Y.C. 1978. Chill storage and the
development of rancidity in frozen Pacific herring, CIupea
harengus pallasi. J. Fish. Res. Board Canada 35: 473.

Bligh, E.G. and Dyer, W.J. 1959. A rapid method of total lipid
extraction and purification. Can. J. Biochem. Physiol. 37: 91 1.

“mm

Bligh, E.G.1961. Lipid hydrolysis in frozen cod muscle. J. Fish. Res. Bd.
(Canada) 18: 143.

Bligh, E.G. and Scott, MA. 1966. Lipids of cod muscle and the effect of
frozen storage. J. Fish. Res. Bd. Canada 23: 1025.

Bosund, J. and Ganrot, B. 1969. Lipid hydrolysis in frozen Baltic
herring. J. Food Sci. 34: 13.

Bosund, J. and Ganrot, B. 1969a. Effect of precooking of Baltic herring
on lipid hydrolysis during subsequent cold storage. Lebensen,
Wiss. Technol. 2:59.

Bremner, H.A.1977. Production and storage of mechanically
separated fish flesh from Australian species. Proc. Fish Expo. '76
Seminar. Aust. Government Publish. Serv., Canberra. p.3 19.

Bremner, HA. 1978. Mechanically separated fish flesh from
Australian species - a summary of results of storage trials. Food
Technol. in Australia 30(10): 393.

Brooks, J .R. and Morr, C.V.1982. Phytate removal from soy protein
isolates using ion exchange processing treatments. J. Food Sci. 47:
1280.

 

 

147

Brown, W.D., Venolia, A.W., Tappel, A.L., Olcott, HS. and Stansby,
M.E.1957. Oxidative deterioration in fish and fishery products. 2.
Progress on studies concerning the mechanism of oxidation of oil
in fish tissues. Com. Fisheries Rev. 19:27.

Brys, K.L.D. 1974. The effect of microcrystalline cellulose on sensory
and physical characteristics of cakes and biscuits. MS thesis.
Michigan State University, East Lansing, MI.

Buckholz, L.L.Jr., Withycombe, DA, and Daun, Henryk 1980
Application and characteristics of polymer adsorption method

used to analyze flavor volatiles from peanuts. J. Agric. Food
Chem. 28:760-765.

Burr, M.L. 1989. Fish and the cardiovascular system. Prog. Food Nutr.
Sci. 13:291.

 

Butler, L.D. and Burke, MP. 1976. Chromatographic characterization
of porous polymers for use as adsorbents in sampling columns. J.
Chromatogr. Sci. 14:1 17.

Buttery, R.G., Ling, L.C., Teranish, R., and Mon, TR. 1977. Roasted
lamb fat: basic volatile components. J. Agric. Food Chem. 25:1227.

Chan, H.W., Prescott, RA. and Swoboda, P.A.T. 1976. Thermal
decomposition of individual positional isomers of methyl linoleate
hydroperoxide: Evidence of carbon-oxygen bond scission. J. Am.
Oil Chem. Soc. 53:572.

Cheftel, J .C. 1979. Proteins and amino acids, in "Nutritional and
Safety Aspects of Food Processing" ,S.R. Tannenbaum (Bd.), p.15 3-
215. Marcel Dekker, Inc., New York and Basel.

Chou, T.C., Lin, T.Y., Hwang, B.J., and Wang, CC. 1986. Selective
removal of H2S from biogas by a packed silica gel adsorber tower.
Biotechnology Progress, 2(4): 203-209.

Conkerton, B.J., Chapital, DC and Ory, R.L.1983. Phenolic acids in flour
from valencia and florunner peanuts: HPLC determination with
ultraviolet and electrochemical detection. Proc. 186th ACS National
Meeting, Washington, DC.

 

 

148

Connell, J .J . 1964. Fish muscle proteins and some effects on them of
processing, in "Proteins and Their Reactions", H. Schultz and A.
Anglemier (Bd.), AVI Publishing Company, Westport, CT.

Connell, J .J . and Howgate, RF. 1971. Consumer evaluation of fresh and
frozen fish. In "Fish Inspection and Quality Control", R. Kreuzer
(Bd.), p.155. Fishing News Books Ltd., London, UK.

Connell, J .J . 1990. Control of Fish Quality. 3rd Ed. p.77-82,120-121.
Fishing News Books, Oxford, England.

Crawford, L., Kretsh, M.J., and Guadagni, D. 1976. Identification of
volatiles from extracted commercial tuna oil with a high
docosahexaenoic acid content. J. Sci. Fd Agric. 27: 53 1-535.

Cronin, D.A. 1982. Techniques of analysis of flavours: Chemical
methods including sample preparation. Ch.IIa, in "Food Flavours,"
I.D. Morton and A.J. MacLeod (Bd.), p.15. Elsevier Scientific
Publishing Company, Amsterdam.

”h... __

Daigle, D.J., Conkerton, B.J., Hammons, KC. and Branch, W.D.1983. A
preliminary classification of selected white testa peanuts (Arachis
hypogaea L.) by flavonoid analysis. Peanut Sci. 10: 40.

Deng, J .C., Matthews, RF. and Watson, CM. 1977. Effect of chemical
and physical treatments on rancidity development of frozen
mullet (Mugil cephalus) fillets. J. Food Sci. 42:344.

Deng, J .C., Watson, M., Bates, RP. and Schroeder, E. 1978. Ascorbic
acid as an antioxidant in fish flesh and its degradation. J. Food Sci.
43: 45 7-460.

Drumm, TD. 1988. Characterization of the Major Components of Dry
Beans and Their Potential Relationship To Flavor. Ph.D.
Dissertation. Department of Food Science and Human Nutrition,
Michigan State University.

Dyer, W.J. and Morton, M.L. 1956a Storage of frozen plaice fillets. J.
Fish. Res. Bd. Canada 13: 129.

Dyer, W.J., Morton, M.L., Fraser, DJ. and Bligh, E.G. 1956. Storage of
frozen rosefish fillets. J. Fish. Res. Bd. Canada 13: 569.

 

 

 

149

Ehler, K.F., Bernhard, RA. and Nickerson, TA. 1979. Heats of
adsorption of small molecules on various forms of lactose, sucrose,
and glucose. J. Agric. Food Chem. 27 (5): 921-927.

Flick, G.J., Hong, Gi-Pyo, and Knobl, GM. 1992. Lipid oxidation of
seafood during storage, in "Lipid Oxidation in Food", Allen J. St.
Angelo (Bd.), American Chemical Society, Washington, DC.

Forss, D.A., Dunstone, EA. and Stark, W. 1960. Fishy flavor in dairy
products. J. Dairy Res. 27:21 1.

Frankel, E.N., Heff, We. and Selke, E. 1981. Analysis of autooxidized
fats by gas chromatography-mass spectrometry. VII. Volatile
thermal decomposition products of pure hydroperoxides from
autoxidized and photosensitized oxidized methyl oleate, linoleate,
and linolenate. Lipids 16:279.

Frankel, E.N., Heff, W.E. and Weisleder, D. 1982. Photosensitized
oxidation of methyl linoleate: Secondary and volatile thermal
decomposition products. lipids 17:1 1.

Frankel, EN .1984. Volatile lipid oxidation products. Prog. lipid Res.
61: 1098.

Fritsch, C.W., Gale, J .A. 1977. Hexanal as a measure of rancidity in
low fat foods. J. Am. Oil Chem. Soc. Vol.54 , June, 1977.

Galt, AM. and MacLeod, G. 1984. Headspace sampling of cooked beef
aroma using Tenax GC. J. Agric. Food Chem. 32:5 9-64.

Geluk, M.A., Norde, W., Kalsbeek, H.K.A.I. and Riet, K. 1992.
Adsorption of lipase from Candida rugosa on cellulose and its
influence on lipolytic activity. Enzyme and Microbial Technology.
14 (9): 748-754.

Gernon, G.D. Jr., Kraus, F.J., and Drake, MP. 1961. Effect of active
carbon on the storage stability of irradiated meats. Food Technol.
August, 1961. p.354-356.

Gilkes, N.R., Jervis, E., Henrissat, B., Tekant, B., Miller, R.C., Warren,
R.A.J. and Kilburn, D.G. 1992. The adsorption of bacterial cellulase
and its two isolated domains to crystalline cellulose. J. of Biological
Chemistry. 267 (10): 6743-6749.

 

 

 

150

Gill, J .L., 1987. "Design and analysis of experiments - in the animal
and medical sciences" Vol. 3, Appendices. The Iowa State
University Press, Ames, Iowa. p.73, 75.

Glicksman, M. 1969. Gum technology in the food industry. p. 403-
412. Academic Press, New York, NY.

Gorga, C. 198 8. Quality Assurance of Seafood. Avi Book Pub. New
York.

Gramshaw, J .W. 1976. Use of lactose in flavour assessment of gas
chromatographic effluents. Chemistry and Industry. 24: 1072.

Gray, J .I.1978. Measurement of lipid oxidation. J. Am. Oil Chem. Soc.
55:539-546.

Greig, R.A.1967. Extending the shelf life of frozen chub (Leuchich thys
hoyi) fillets through the use of ascorbic acid dips. Fish. Ind. Res.
4(1): 23.

Greig, R.A. 1968a Extending the shelf life of frozen chub
(Leucihchthys hoyi) fillets through the use of ascorbic acid dips.
Fish. Ind. Res. 4(1): 23.

Greig, R.A. 1968b Extending the shelf life of frozen white bass
(Roccus chrysops) through the use of ascorbic acid dips. Fish. Ind.
Res. 4: 45.

Guichard, E., Issanchou, S., Descourvieres, A., and EtievantP. 1991.
Pectin concentration, molecular weight and degree of
esterification: Influence on volatile composition and sensory
characteristics of strawberry jam. J. Food Sci. 56(6): 1621-1627.

Gupta, K.G., Jain, AK. and Surinder, D. 1979. Removal of diacetyl from
beer by adsorbents and diacetyl reductase. Biotechnology and
Bioengineering. 21 (4): 649-65 7.

Hamilton, M. and Bennett, R. 1983. An investigation into consumer
preferences for nine fresh white fish species and the sensory
attributes which determine acceptability. J. Food Technol. 18:75.

 

151

Hamilton, M. and Bennett, R. 1984. Consumer preferences for fresh
white fish species. J. Consumer Studies & Home Economics 8: 243.

How, J .S.L. and Morr, C.V. 1982. Removal of phenolic compounds from
soy protein extracts using activated carbon. J. Food Sci. 47: 933.

Hsieh, R.J. and Kinsella, J .E., 1986. lipoxygenase-catalyzed oxidation
of N-6 and N-3 polyunasturated fatty acids: relevance to and
activity in fish tissue. J. Food Sci. 51(4): 940-945, 996.

Ingalls, R.L., Klocke, J.F., Rafferty, J .P., Greensmith, R.E., Chang,
M.L.,Tack, RI. and Ohlson, MA. 1950. Nutritive value of fish from
Michigan waters. Mich. State Univ. Tech. Bull. 219.

Ikeda, S. 1980. Other organic components and inorganic components,
in "Advances in Fish Science Technology," J .J . Connell (Bd.), p.11 1-
123. Fishing News Books Ltd., Farnham, England.

Jach, M. and Sugier, H. 1983. Adsorption of glucoamylase on DEAE-
cellulose. Starch/Staerke 35 (12): 427-430.

Jonas, R.E.E. and Tomlinson, N. 1962. The phospholipid content of ling
cod muscle during frozen storage. J. Fish. Res. Bd. Canada 19:733.

Jonas, R.E.E. and Bilinski, E. 1967. Glycerylphosphorylcholine and
related compounds in rainbow trout muscle. J. Fish. Res. Bd.
Canada 24: 273.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1983. Bisulfite
suppression of fish aromas. J. Food Sci. 48: 1064-1067.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1983a. Identification
of compounds characterizing the aroma of fresh Whitefish
(Coregon us clupeaformis). J. Agric. Food Chem. 3 1:326.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1984. Identification of
volatile aroma compounds from oxidized frozen Whitefish
(Coregonus clupeaforml's). Can. Inst. Food Sci. Technol. J. 17(3):
178-182.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1984a. Variations in
the occurrences of enzymically-derived volatile aroma compounds
in salt and freshwater fish. J. Agric. Food Chem. 32:1344.

 

 

152

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1984b. Biogenesis of
lipid-derived volatile aroma compounds in the emerald Shiner
(Notropl's atherinoides). J. Agric. Food Chem. 32:1347.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1985. Volatile
compounds characterizing the aroma of fresh Atlantic and Pacific
oysters. J. Food Sci. 50:5.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1985. Effect of
handling and packaging on the quality of frozen Whitefish. J. Food
Sci. 50:1.

Josephson, D.B., Lindsay, RC. 1986. Enzymic generation of fresh fish
volatile aroma compounds. In "Biogeneration of Aromas". T.H.
Parliment and R.Croteau (Bd.), p.201. ACS Symposium #3 17.

Josephson, D.B., Lindsay, R.C., and Stuiber, D.A. 1987. Enzymic
hydroperoxide initiated effects in fresh fish. J. Food Sci. 52(3):
596-600.

Karel, M.1973. Symposium: Protein interactions in biosystems.
Protein-lipid interactions. J. Food Sci. 38: 756.

Karel, M. 1975. Protective packaging of foods, in "Principles of Food
Science, Part 11: Physical principles of food preservation" O.R. M.
Fennema (Bd.), Dekker Inc., p.422-466.

Kawata, K., Uemura, I.K., Tominaga, Y. and Oikawa, K. 1982.
Breakthrough volumes of organic vapors on Tenax GC adsorbent.
Bunseki Kagaku (Japanese Analyst) 8: 453.

Ke, P.E., Ackman, R.G., and linke, BA. 1975. Autoxidation of
polyunsaturated fatty compounds in mackerel oil: formation of
2,4,7-decatrienals. J. Am. Oil Chem. Soc. 52:349.

Kelly, TR. 1969. Quality in frozen cod and limiting factors on its shelf
life. J Food Technol. 4: 95.

Khayat A.; Schwall, D. 1983. Lipid oxidation in seafood. Food Technol.
37(7):130-140.

 

 

153

Kirk, RS. and Sawyer, R.1991. Fish, in "Pearson's Composition and
Analysis of Foods" 9th Ed. Longman Scientific & Technical. 1991.
p.505-529.

Knapp, W.A. 1973. Stabilized flavoring compositions, US. Patent,
3,736,149.

Krost, K.J., Pellizari, E.D., Walburn, S.G. and Hubbard, SA. 1982.
Collection and analysis of haxardous organic emissions. Anal.
Chem. 54:810.

Kuo, P.P.K., Chian, E.S.K., DeWalle, RB. and Kim, J.H. 1977. Gas
stripping, sorption and thermal desorption procedures for
preconcentrating volatile polar water-soluable organics from
water samples for analysis by gas chromatography. Anal. Chem.
7: 1023.

 

Laidsaar, R.B., Lasle, E., and Khoshke, A. 1981. Immobilization of
glucoamylase on organic carriers. Tallinna Poluetehnilise
Instituudi Toimetised; No. 5 10, p.1 15-123.

Lane, J .P. 1966. Time-temperature tolerance of frozen seafoods. II.
Temperature conditions during commercial distribution of frozen
fishery products. Food Technol. 20: 549.

Larmond, E. 1977. Laboratory Methods for Sensory Evaluation of
Food, Publication 1637, Rood Research Institute, Canada
Department of Agriculture, Ottawa, Canada.

Larmond, E.1977. Sensory evaluation can be objective, in "Methods
in Food Quality Assessment", J.G. Kapsalis (Bd.), p.3-14. CRC Press,
Boca Raton, FL.

Laslett, GM. and Bremner, HA. 1979. Evaluating acceptability of fish
minces and fish fingers from sensory variables. J .Food Technol.
14:389.

Lea, OH. 1953. Recent developments in the study of oxidative
deterioration of lipids.Chem. and Ind. (London) 49:1303.

Lee, I., Nickerson, TA. and Bernhard, RA. 1975. Adsorption of low-
molecular-weight organic compounds by stable anhydrous a-
lactose. J. Dairy Sci. 58: 319.

 

154

Lentz, C.P. and Rooke, EA. 1960. Temperratures in frozen fish
shipped by road in refrigerated trailers. Can. Food Ind. 3 1(2): 26.

Lilzemark, A. 1964. Cold storage of retailed-packed fillets of
mackerel and herring. Food Technol. 18:122.

Liu, H. and Watts, BM. 1970. Catalysis of lipid peroxidation in meats.
3. Catalysts of oxidative rancidity in meats. J. Food Sci. 35:596.

Love, J .D. and Pearson, AM. 1974. Metmyoglobin and nonheme iron
as prooxidants in cooked meat. J. Agric. Food Chem. 22:1032.

Lovegren, N .V., Fisher, G.S., Legendre, MG, and Schuller, W.H. 1979.
Volatile constituents of dried legumes. J. Agric. Food Chem. 27:85 1.

Lovem, J .A., Olley, J. and Watson, H. 1959. Changes in the lipids of
cod during storage in ice. J. Sci. Food Agric. 10: 327.

Maarse, H. 1991. Introduction, in "Volatile Compounds in Foods and
Beverages", Henk Maarse (Bd.), Chap.1 , p.1-39. Marcel Dekker,
Inc., New York, NY.

MacBain, J .W.1932. The sorption of gases and vapours by solids.
Routledge, London. Chap. 5.

Maga, J .A.1978. Simple phenol and phenolic compounds in food
flavor. CRC Crit. Rev. Food Sci. Nutr. 11: 323.

Mai, J. and Kinsella, J .E. 1979. Changes in lipid composition of cooked
minced carp (Cyn’n us carpino) during frozen storage. J. Food Sci.
44:1619.

Maier, H.G. 1969. The binding of volatile aroma constituents in foods.
I. Sensory method. Z. Lebensum-Untersuch-Forsch. 141: 65.

Maier, H.G. 1969. The binding of volatile aroma constituents in foods.
II. Desiccator method. Z. Lebensum-Untersuch—Forsch. 141: 332.

Maier, H.G. 1970. The binding of volatile aroma constituents in foods.
III. Gas chromatrography study. Z. Lebensm-Untersuch-Forsch.
143:24.

 

 

155

Maier, H.G. 1972. The binding of volatile aroma constituents in foods.
Lebesm.-Wiss. und Tech. 5: 1.

Marcuse, R. 1954. Experiments on fat stabilization by ascorbic acid
with two Herring biotypes (Fladen and Iceland). Svenska Inst. for
Konserverings for shrimp. Publication vol. 100. Sect.1 1.

Marvin, J .W., Bernhard, RA. and Nickerson, T.A. 1979. Interactions of
low molecular weight adsorbated on lactose. J. of Dairy Sci. 62
(10): 1546-1557.

Matsukura, M., Takahashi, K., Ishiguro, S., and Matsushita, H. 1984.
Adsorption of volatiles from roasted tobacco on activated carbon

and desorptive recovery by ether extraction. Agric. Biol. Chem.,
48(4): 971-975.

Matthews, RF. 1971. The autoxidation of 2,4-decadienal. MS. Thesis,
Oregon State University, Corvallis, OR.

McGill, AS. and Hardy, R. 1977. Artifact production in the Likens-
Nickerson apparatus when used to extract the volatile flavorous
components of cod. J. Sci. Food Agric. 28:89.

McGill, A.S., Hardy, R., and Gunstone, RD. 1977. Futher analysis of the
volatile components of frozen cold stored cod and the influence of
these on flavor. J. Sci. Food Agric. 28:200.

McMullin, S.L., Bernhard, RA, and Nickerson, T.A. 1975. Heats of
adsorption of small molecules on lactose. J. Agric. Food Chem. 23
(3): 452-458.

McNally, ME. and Grob, R.L. 1985a. A review: Current applications of
static and dynamic headspace analysis: Part one: Environmental
applications. Amer. Lab. 17(1): 106.

McNally, ME. and Grob, R.L. 1985b. Current applications of static and
dynamic headspace analysis: A review: Part two:
Nonenvironmental applications. Amer. Lab. 17(2): 106.

Meilgaard, M., Civille, G.V. and Carr, ET 1987. Sensory Evaluation
Techniques. Vol. I & II. CRC Press, Inc. Boca Raton, FL.

156

Morris, D.M. and Dawson, L.E.1979. Storage stability of mechanically
deboned sucker (Catostomidage) flesh. J. Food Sci. 44: 1093.

Morrison, RT. and Boyd, RN. 1978. Aldehydes and ketones, in
"Organic Chemistry," 3rd ed., p.639. Allyn and Bacon Inc., Boston,
MA.

Murray, K.E.1977. Concentration of headspace, airborne and aqueous
volatiles in Chromosorb 105 for examination by gas
chromatography and gas chromatography-mass spectrometry. J.
Chromat. 135: 49.

Mussinan, C. 1978. International Flavor & Fragrances, Union Beach,
NJ.

Nair, P.G.V., Anthony, RD. and Gopakumar, K. 1979. Oxidative
rancidity in the skin and muscle lipids of soil sardine (Sardinella
longiceps). J. Food Sci. Tech. 16: 151.

Nickerson, T.A. 1979. lactose chemistry. J. Agric. Food Chem. 27 (4):
672-677.

Nickerson, T.A. and Dolby, RM. 1971. Adsorption of diacetyl by
lactose and other sugars. J. of Dairy Sci. 54 (8): 1212-1214.

Niediek, EA. 1988. Effect of processing on the physical State and
aroma sorption properties of carbohydrates. Food Technology. 42
(11): 80, 82-83, 85.

Nitz, S. and Julich, E. 1984. Concentration and GC-MS analysis of trace
volatiles by sorption-desorption techniques, in "Analysis of
Volatiles", P. Schreier (Bd.), p.15 1. Walter de Gruyter & Co., Berlin-
New York.

Novotny, M., Lee, M.L., and Bartle, KB. 1974. Some analytical aspects
of the chromatographic headspace concentration method using a
porous polymer. Chromatographia 7:333.

Nunez, A.J., Gonzales, LP. and Janak, J .1984. Pre-concentration of
headspace volatiles for trace organic analysis by gas
chromatography. J. Chromat. 300: 127.

 

 

157

Ohnishi, S. and Shibamoto, T.1984. Volatile compounds from heated
beef fat and beef fat with glycine. J. Agric. Food Chem. 32:987.

Olafsdottir, G., Steinke, J .A., and lindsay, R.C., 1985. Quantitative
performance of a simple Tenax-GC adsorption method for use in
the analysis of aroma volatiles. J. Food Sci. 50:143 1.

Olcott, HS. 1962. In "Fish in Nutrition", E. Heen and R.Kreuzer (Bd.),
p.1 12-116. Fishing News Books Ltd., London, England.

Olley, J. and Lovern, J .A.1960. Phospholipid hydrolysis in cod flesh
stored at various temperatures. J. Sci. Food Agric. 1 1: 644.

Olley, J ., Farmer, J. and Stephan, E. 1969. The rate of phospholipid
hydrolysis in frozen fish. J. Food Technol. 4: 27.

Oosting, E.M., Gray, J .I., and Grulke, E.A., 1984. Correlating diffusion
coefficients in concentrated carbonhydrate solutions. AIchE
Journal 31(5): 773.

Palmateer, R.E., Yu, TC. and Sinnhuber, R.O.1960. An accelerated
oxidation method for the estimation of the storage life of frozen
seafoods. Food Technol. 14(10): 1.

Peers, KB. and Coxon, D.T. 1986. Simple enrichment procedure for the
estimation of minor polyunsaturated fatty acids in food fats. J.
Food Technol. 2 1:463.

Peters, J .A. and McLane, D.T.1959. Storage life of pink shrimp held in
commercial cold storage room. Com. Fish. Rev. 21(9): 1.

Peryam, DR. and Girardot, NP. 1952. Advanced taste test method.
Food Eng., 24(7):58-61, 194.

Peryam, DR and Pilgrim, F.J. 1957. Hedonic scale method of
measuring food preferences. Food Technol., 1 1(9, supplement):
9-14.

Ponec, V., Knor, Z., and Cemy, S. 1974. Adsorption on Solids.
Butterworth and Co., London, Chap. 14.

Reger, J .V. 1958. New aspects of old sugar-lactose. Cereal Sci. Today.
3 (10): 270.

 

158

Reineccius, GA. and Anandaraman, S. 1984. Analysis of volatile
flavors. Ch.5, in "Food Constituents and Food Residues", J .F.
lawrence (Bd.), p.195. Marcel Dekker, Inc., New York, NY.

Ross, D.A. and Love, RM. 1979. Decrease in the cold store flavour
developed by frozen fillets of Storved cod (Gadus morh ua L.)J.
Food Technol. 14: 115.

Ruthven, D.M. 1984. Physical adsorption and the characterization of
porous adsorbents, in "Principles of Adsorption and Adsorption
Processes", p.29-30. A Wiley-Interscience Publication, John Wiley
& Sons, New York, NY.

Sacharow, S., and Griffin, R.C. Jr. 1980. Fish and shellfish, in
"Principles of Food Packaging", 2nd Ed. Avi Pub. Co., Westport, CT.

Saenz, W. and Dubrow, D.L. 1959. Control of rancidity in vacuum-
packaged frozen mullet fillets. p. 1 1.The Marine Laboratory, Univ.
of Miami, Annual Report 1958-1959.

Saito, K., Yamamoto, T. and Miyamota, K. 1992. Isolation and partial
purification of carthamine: an instrumentation manual. Zeitschrift
fuer Lebensmittel Untersuchung und Forschung. 195 (6): 550-555.

Sakaki, T., Niino, K., Sakuma, H., and Sugawara, s. 1984. Analysis of i
tobacco headspace volatiles using Tenax GC or active carbon.
Agric. Biol. Chem. 48(2). 3121-3128.

Sakodynskii, K., Panina, L., and Klinskaya, N. 1974. A study of some
properties of Tenax, a porous polymer sorbent. Chromatographia
7:339.

Saleeb, F.Z. and Pickup, J .G. 1978. Flavor interactions with food
ingredients from headspace analysis measurements, in "Flavor of
foods and beverages" p.1 13-130.

Sawyer, F. Miles, 1987. Sensory methodology for estimating quality
attributes of seafoods, in "Seafood Quality Determination", D.E.
Kramer and J. Liston(Ed.), Elsevier, NY.

Sawyer, F.M., Cardello, A.V., and Prell, RA. 1988 Consumer evaluation
of the sensory properties of fish. J. Food Sci. 53(1):12-24.

 

159

Schneeman, ED. 1986. Physical and chemical properties, methods of
analysis, and physiological effects. Food Technol. p. 104-109. Feb.
1986.

Scott, R.P.W. 1993. The silica gel surface, in "Silica Gel and Bonded
Phases- Their Production, Properties and Use in LC", R.P.W. Scott
(Ed.), Chap. 4, p.74-75. John Wiley & Sons, Inc., New York, NY.

Selke, E. and Frankel, EN. 1987. Dynamic headspace capillary gas
chromatographic analysis of soybean oil volatiles. J. Am. Oil Chem.
Soc. 64: 749.

Selke, E., Rohwedder, W.K. and Dutton, H]. 1980. Volatile components
from trilinolein heated in air. J. Am. Oil Chem. Soc. 57:25.

Seo, A. and Morr, C.V. 1985. Activated carbon and ion exchange
treatments for removing phenolics and phytate from peanut
protein products. J. Foods Sci. 50: 262-263.

Shantha, N .C. and Ackman, R.G. 1991. Silica gel thin-layer
chromatographic method for concentration of longer-chain
polyunsaturated fatty acids from food and marine lipids. Can.
Inst. Sci. Technol. J. 24(3/4): 156-160.

Shimada, Y., Roos, Y., and Karel, M. 1991. Oxidation of methyl
linoleate encapsulated in amorphous lactose-based food model.
Americal Chemical Soceity. 39(4): 63 7-641.

Sosulski, F.1979. Organoleptic and nutritional effects off phenolic
compounds on oilseed protein products: A review. J. Am. Oil Chem.
Soc. 56: 7 1 1.

Stahl, W.H. and Einstein, MA. 1973. Sensory testing methods, in
"Encyclopedia of Industrial Chemical Analysis", Vol.17, F.D. Snell
and LS. Ettre, (Bd.), p.608-644. John Wiley and Sons, Inc., New
York, NY.

Stansby, ME. and Olcott, HS. 1963. Composition. of fish, in
"Idndustrial Fishery Technology", M.E. Stansby, (Bd.), p.339.
Reinhold, N.Y.

160

Steinke, J ., 1978. Isolation and identification of volatile constitutes
from off-flavored Lake Michgan Salmon, PhD Dissertation,
University of Wisconsin, Masison, W.I.

Stone, H., and Sidel, J ., 1985. Sensory Evaluation Practices, Academic
Press Inc., Orlando, FL., p.25 8-260.

Stuckey, B.N .1968. Antioxidants as food stabilizers, in "Handbook of
Food Additives", p.209. The Chemical Rubber Co., Cleveland, Ohio.

Sugisawa, H. 1981. Sample preparation: Isolation and concentration
Ch.2, in "Flavor Research - Rescent Advances", R. Teranishi, R.A.
Flath, and H. Sugisawa (Bd.), p.11. Marcel Dekker, Inc., New York,
NY.

Sweet, C.W. 1973. A research note: Activity of antioxidants in fresh
fish. J. of Foos Sci. Vol.38:1260—1261.

Swoboda, P.A.T., and Peers, KB. 197 7. Metallic odour caused by vinyl
ketones formed in the oxidation of butter fat. The identification of
Octa—l, cis-S-dien-3one. J. Sci. Food Agric. 28: 1019.

Tausig, F., and Drake, MP. 1959. Activated carbons as odor
scavengers for radiation-sterilized beef. Food Research 24: 224.
Toro-Vazquez, J .F., Garcia-L, OE, and Guerrero-E, LL. 1991.
Adsorption Isotherms of Squash (Cucurbita moschata) Seed Oil on
Activated Carbon. JAOCS 68(8):5 96-5 99.

Umano, K. and Shibamoto, T. 1987. Analysis of headspace volatiles
from overheated beef fat. J. Agric. Food Chem. 35:14-18.

Vercellotti, J .R., Kuan, J .C.W., liu, R.H., Legendre, M.G., St. Angelo, A.J.,
and Dupuy, HP. 1987. Analysis of volatile heteroatomic meat
flavor principles by purge-and-trap/ gas chromatography-mass
spectrometry. J. Agric. Food Chem. 35: 1030.

Vercellotti, J .R., St. Angelo, A.J., Legendre, M.G., Sumrell, G., Dupuy,
HP, and Flick, G]. 1988. Analysis of trace volatiles in food and
beverage products involving removal at a mild temperature under
vacuum. J. Food Comp. Anal.

161

Waltking, A.E. and Goetz, AG. 1983. Instrumental determination of
flavor stability of fatty foods and its correlation with sensory
flavor responses. CRC Crit. Rev. in Food Sci. and N utr. 19: 99.

Warner, K., Evans, C.D., list, G.R., Dupuy, H.P., Wadsworth, J .I., and
Goheen, GE. 1978. Flavor score correlation with pentanal and
hexanal contents of vegetable oil. J. Am. Oil Chem. Soc. 55:252.

Watts, B.M. and Wong, R. 1951. Some factors affecting the antioxidant
behavior of ascorbic acid with unsaturated fats. Arch. Biochem.
30: 1 10.

Weber, Jr. W.J., and Van Vliet, BM, 1980. Fundamental concepts for
application of activated carbon in water and wastewater
treatment. in "Activated Carbon Adsorption of Organics from the
Aqueous Phase", Vol. 1. Irwin H. Suffet and Michael J. McGuire
(Bd.), Chap. 1. p. 16-19. Ann Arbor Science Publishers Inc./ The
Butterworth Group, Ann Arbor, MI.

Welsh, F.W. and Zall, RR. 1984. An ultrafiltration activated carbon
treatment system for renovating fishery refrigeration brines.
Can. Inst. Food Sci. Technol. J. 17(2): 92-96.

Wesson, J .B., Lindsay, KC. and Stuiber, D.A. 1979. Discrimination of
fish and seafood quality by consumer populations. J. Food Sci.
44:878.

Westendorf, R.G. 1985. Automated analysis of volatile flavor
compounds. Ch. 10, in "Characterization and Measurement of
Flavor Compounds", D.D. Bills and OJ. Mussinan (Bd.), ACS Symp.
Ser. 289, p.138. American Chemical Society, Washington, D.G.

Weurman, C. 1969. Isolation and concentration of volatiles in food
odor research. J. Agric. Food Chem. 17:370.

Williams, R.J. 1971. You Are Extraordinary, Pyramid Books, New
York, NY.

Yabumoto, K., Jennings, W.G. and Pangbom, RM. 1975. Evaluation of
lactose as a transfer carrier for volatile flavor constituents. J. Food
Sci. 40: 105.

162

Yasuhara, A. and Shibamoto, T. 1989. Analysis of aldehydes and
ketones in the headspace of heated pork fat. J. Foood Sci. 54(6):
1471-1472,1489.

Young, O.C. 1950. Quality of fresh and frozen fish and facilities for
freezing, storing, and transporting fishery products. Food Technol.
4( 1 1) : 447.

Yu, T.C., Day, E.A., and Sinnhuber, R.O. 1961. Autoxidation of fish oils.
1. Identification of volatile monocarbonyl compounds from
autoxidized salmon oil. J. Food Sci. 26:192.

Yu, T.C., Landers, MK and Sinnhuber, R.O. 1969. Storage life
extension of refrozen silver salmon steaks. Food Technol. 23:1602.

Yu, T.C., Sinnhuber, R.O., and Crawford, D.L. 1973. Effect of packaging
on shelf life of frozen silver salmon steaks. J. Food Sci. 38
pp.1197-1199.

Zeldes, S.G. and Horton, AD. 1978. Trapping and determination of
labile compounds in the gas phase of cigarettte smoke. Anal.
Chem. 50(6):779.

.31).

I. 1...: .9

:xitlf .11

 

 

31:7}
{45.1.1.1}:
, . . .. .

 

I: L... : 7.
: z...— .1. .