Copyright is owned by the Author of the thesis.  Permission is given for 
a copy to be downloaded by an individual for the purpose of research and 
private study only.  The thesis may not be reproduced elsewhere without 
the permission of the Author. 
 
FISH OIL: REFINING, STABILITY AND 
ITS USE IN CANNED FISH FOR 
THE INDONESIAN MARKET 
A thesis presented in partial fulfilment of 
the requirements for the degree of 
Doctor of Philosophy 
in Food Process and Product Development at 
Massey University, 
New Zealand 
HARI EKO IRIANTO 
1992 
(' 
;; 
1 
J. 
Massey University Library 
Thesis Copyright Form 
Title Of thesis: FISH OIL: REFINING, STABILITY AND ITS USE IN CANNED FISH FOR THE INDONESIAN MARKET 
(1) ? I give permission for my thesis to be made available to 
readers in Massey University Library under conditions 
determined by the Librarian. 
(b) I do not wish my thesis to be made available to readers 
without my written consent for ... months. 
' 
(2) M I agree that my thesis, or a copy, may be sent to 
another institution under conditions determined by the 
(3) 
Librarian. 
(b) I do not wish my thesis, or a copy, to be sent to 
another institution without my written consent for 
months. 
I agree that my thesis may be copied for Library use. 
(b) I do not wish my thesis to be copied for Library use for 
... months. 
Signed 
Date 
The copyright of this thesis belongs to the author. Readers must 
sign their name in the space below to show that they recognise 
this. They are asked to add their permanent address. 
NAlviE AND ADDRESS ?DATE 
15 
ii 
ABSTRACT 
Fish oil has been proved to have health benefits for humans, but the utilization of fish oil for 
human consumption is very limited. 
A survey of 19 Indonesian fish oil producers showed that fish oil was produced from fish meal 
processing and fish canning. Most Indonesian fish oils, especially fish meal oil, were chemically, 
physically and organoleptically unacceptable. But, as they contain high levels of omega-3 fatty 
acids, a refining process was required to improve the quality making the oil acceptable for human 
consumption. 
The resin refining process, a no heat process, was used to refine the crude oil. Fish oil-resin 
volume ratio affected the refined fish oil quality and ratio 1:1 was recommended. The refined oil 
quality could be further improved by multiple refmings, and this method was successfully applied 
to Indonesian fish meal oil having a strong undesirable odour. The refining rate could be 
accelerated by application of vacuum pressure to the column. The height of the column showed 
a significant effect on the refined fish oil quality, but the column diameter had no effect. Resin 
refining reduced the quantity of natural antioxidants and changed the proportion of volatile flavour 
compounds. Most Indonesian fish oil producers intend to adopt the resin refining process. 
Storage tests indicated that the refined oil deteriorated faster than unrefined oil. This trend was 
shown by both Indonesian and New Zealand oils. Ter-butylhydroquinone (TBHQ) proved as the 
most effective antioxidant for fish oil, but this antioxidant is not listed as a permitted antioxidants 
for use in Indonesian foods. Butylated hydroxyanisole (BHA), as the best alternative, is 
recommended. 0.01% BHA was sufficient to recover the loss of natural antioxidant during resin 
refining. Vacuum package was very effective in reducing the deterioration rate due to autoxidation. 
Canned fish was used as a mean of delivering fish oil to Indonesian consumer. The proposed 
product type was generated through supermarket, consumer and canned fish processor surveys. The 
survey results suggested that the fish oil to be disguised in a canned fish product using sardine as 
raw material, tomato sauce as medium and 155g tall tube-can as the container. 
The most acceptable tomato sauce formula developed using mixture design is 18.6% tomato paste, 
28.0% fish oil, 46.6% water, 3.7% salt and 3.1% sugar. The canned fish should be sterilized using 
iii 
vacuum head space-can at 12l.l?C to obtain optimum protection of omega-3 fatty acids. 
The experiment using the Plackett and Burman design showed that the canned fish product should 
involve pre-cooking, vacuum head space, garlic, shallots and vinegar additions. Sterilization time 
needed to be optimized. The optimization experiment indicated that 50 minutes was recommended 
to sterilize the canned fish with disguised fish oil. Sterility and incubation tests showed that 
sterilization at 12l.l?C for 50 minutes was sufficient to produce safe product 
Consumer testing in five cities of Indonesia showed that only a minority of consumers did not like 
the developed product. Most of the consumers intend to buy the product, if the product is released 
to the market A survey of medical doctors supported the proposed product, as over 90% of them 
were willing to suggest patients consume the product for nutritional purposes. 
iv 
And if all the trees on earth were pens and the 
ocean (were ink), with seven oceans behind it 
to add to its (supply), yet would not the words 
of Allah be exhausted (in the writing):for Allah 
is Exalted in power, full of Wisdom. 
(The Holly Qur'an 31: 27) 
To my wife, Giyatmi, and my daughter, 
Husna Izzahnisa Omegita 
V 
ACKNOWLEDGEMENTS 
In the name of Allah, Most Gracious, Most Merciful. 
I would like to express my immense gratitude to Dr.Carmen C. Fernandez, my chief Supervisor, 
for her guidance, encouragement and patience during all stages of my Ph.D program. I would like 
also to thank for her efforts in upgrading my program from Master to PhD. Also my thank to 
Dr.G.J.Shaw, my eo-supervisor, for his guidance, supports and patience throughout my program. 
I would like also to express sincerely appreciation to the following: 
Prof. P.A. Munro, head of Food Technology Department, Massey University, for his full support 
in upgrading my study program from M.Tech .. to PhD. 
Dr. Suparno, director of the Research Institute of Fish Technology, Jakarta, for his supports 
throughout my study in New Zealand. 
Dr. Cecil Johnson, Crown Research Institute, for his private training in fatty acid esterification and 
introduction to gas chromatography. 
Mr. John M. Alien, Crown Research Institute, for his help in fatty acid proftles and volatile flavour 
compounds analysis. 
Mrs. M. Bewley, for her help in providing all equipments and reagents for chemical works. 
Mr. Hank van Til, for his help on computer works and canning experiment 
Mr. Garry Redford, for his help during Hunter lab colour analysis and canning experiment. 
Research staffs at the Research Institute of Fish Technology, Jakarta, especially Ir.Ijah Muljanah 
MS, Drs.Tazwir, Ir. Jamal Basmal, Ir. Mei Dwi Erlina and Ir. Murdinah MS, for their help during 
fish meal and cannery survey. 
Ir. Marlina Darmadi, Ir. Marini Gunadi, Ir. Wiwin Dyah Srie Banon, Irma Handarsari BSc and 
mbak Warni for their help during consumer product testing in Indonesia. 
vi 
All Indonesian post graduate students at Massey University and families, for their participation 
during sensory evaluation throughout my experiments. 
All people in Jakarta, Tangerang, Semarang, Sragen, and Lumajang who have participated in 
consumer survey and product testing. 
All fish meal and canned fish factories in Muncar and Bali which bave been willing to be surveyed. 
Sealord Product Ltd., Nelson, for fish oil; J W atties Foods Ltd, Hastings, for tomato paste; Dow 
Chemicals, USA, for resin; Roche, New Zealand, for DI-u-tocopherol; and Bronson and Jacobs, 
New Zealand, for Grindox 1 17. 
My parents (Siswanto and Kasmiyati) and my brothers and sister (Hero, Henny, Basuki and 
Hudba), and also Bapak and Simbok Mintopawiro for their prayers. 
vii 
TABLE OF CONTENTS 
ABSTRACI' ii 
ACKNOWLEDGEMENTS V 
LIST OF TABLES xiv 
LIST OF FIGURES xvii 
LIST OF APPENDICES xxi 
Chapter 1. INTRODUCTION 1 
Chapter 2. GENERAL LITERATURE REVIEW 3 
2.1. CHEMICAL PROPERTIES OF FISH OIL 3 
2.1.1. Triglycerides and fatty. acids 3 
2.1.2. Wax esters 5 
2.1.3. Phospholipids 5 
2.1.4. Free fatty acids (FFAs) 5 
2.1.5. Ether groups 6 
2.1.7. Sterols 6 
2.1.8. Heavy metals 7 
2.1.9. Pigments 7 
2.2. NUTRITIONAL PROPERTIES 8 
2.2.1. Essential fatty acids 8 
2.2.2. Vitamins 9 
2.3. FISH OIL AND DISEASES 9 
2.4. FISH OIL PRODUCTION 10 
2.4.1. Extraction technology 11 
2.4.2. Processing of fish oil 14 
2.5. INDUSTRIAL APPLICATION OF FISH OIL 17 
2.5.1. Fish oil application in foods 17 
2.5.2. Fish oil application in pharmaceuticals 18 
2.5 .3. Fish oil application in animal and fish feeds 20 
2.5.4. Fish oil application in non-edible uses 20 
Chapter 3. MATERIAL AND ANALYSIS METHODS 
3.1. MATERIALS 22 
viii 
3 .1.1. Fish oils 22 
3.1.2. Resin 22 
3.1.3. Fish 22 
3.1.4. Can 23 
3.3. METHODS OF ANALYSIS 24 
3.3.1. Chemical analysis 24 
3.3.2. Physical analysis 31 
3.3.3. Sensory analysis 32 
3.3.4. Canned fish analysis 35 
3.4. DATA ANALYSIS 36 
Chapter.4. FISH OIL PRODUCTION IN INDONESIA 37 
4.1. BACKGROUND 37 
4.2. OBJECTIVES 38 
4.3. METHODOLOGY 38 
4.4. RESULTS 39 
4.4.1. Position of fish meal in Indonesian fishery industry 39 
4.4.2. Raw fish used for fish mealfoil production in Indonesia 39 
4.4.3. Fish meal and fish oil processing in Indonesia 40 
4.4.4. Prices and buyers of fish oil 41 
4.4.5. Chemical, physical and sensory analysis of Indonesian fish oil 42 
4.4.6. Fatty acid profiles of fish oil 45 
4.4.7. New Zealand fish oil used as comparison with Indonesian fish oil 47 
4.5. DISCUSSIONS 48 
4.5.1. Fish oil production 48 
4.5.2. Fish oil quality 49 
4.6. CONCLUSIONS 
Chapter 5. OPTIMIZATION OF THE RESIN REFINING PROCESS OF FISH OIL 53 
5.1. BACKGROUND 53 
5.2. OBJECTIVES 55 
5.3. METHODOLOGY 55 
5.3 .1. Materials 55 
5.3.2. Experimental methods 55 
5.4. RESULTS 58 
5.4.1. Effects of fish oil-resin volume ratio on fish oil quality 58 
ix 
5.4.2. Effects of multiple refining on fish oil quality 66 
5.4.3. Effects of vacuum pressure application on fish oil quality 72 
5.4.4. Effects of column size on fish oil quality 78 
5.4.5. Effects of resin refining on natural antioxidant contents of fish oil 88 
5.4.6. Effects of resin refining on volatile flavour compounds of fish oil 88 
5.5. DISCUSSION 
5.5.1. Effects of resin refining on chemical properties of fish oil 93 
5.5.2. Effects of resin refining on physical properties of fish oil 95 
5.5.3. Effects of resin refming process on volatile flavour compounds 96 
5.5.4. Effects of resin refining on sensory properties of fish oil 98 
5.6. CONCLUSIONS 
Chapter 6. STORAGE TEST OF REFINED AND UNREFINED FISH OIL 101 
6.1. BACKGROUND 101 
6.2. OBJECTIVES 102 
6.3. METHODOLOGY 102 
6.3.1. Materials 102 
6.3.2. Methods 102 
6.3.3. Determination of the deterioration rate of fish oil during storage 103 
6.4. RESULTS 
6.4.1. Effects of storage on peroxide value (PV) of fish oil 103 
6.4.2. Effects of storage on refractive index value of fish oil 104 
6.4.3. Effects of storage on colour of fish oil 106 
6.4.4. Effects of storage on sensory properties of fish oil 107 
6.5. DETERMINATION OF RATE CONSTANTS AND ORDER REACTION 
MODEL 110 
6.6. ESTIMATION OF SHELF LIFE OF FISH OIL 112 
6.7. CORRELATION BETWEEN SENSORY RESULTS AND OTHER 
PARAMETERS 
6.8. DISCUSSION 
6.8.1. Chemical and physical changes in fish oil during storage 
6.8.2. Sensory changes in fish oil during storage 
6.8.3. Shelf life of fish oil 
6.9. CONCLUSIONS 
116 
117 
117 
119 
120 
121 
X 
Chapter 7. STABILITY IMPROVEMENT OF RESIN REFINED FISH OIL 122 
7.1. BACKGROUND 122 
7.1.1. Antioxidant and oxidation 122 
7 .1.2. Oxygen removal and oxidation 124 
7.2. OBJECTIVES 124 
7.3. METHODOLOGY 125 
7.3.1. Materials 125 
7 .3.2. Methods 125 
7.4. RESULTS 127 
7.4.1. Selection of antioxidant 127 
7.3.2. Optimisation of antioxidant level 134 
7.4.3. Use of vacuum package for fish oil stability improvement 141 
7.5. DISCUSSION 149 
7.5.1. Use of antioxidant for .fish oil stability improvement 149 
7.5.2. Use of vacuum package for fish oil stability improvement 150 
7 .6. CONCLUSIONS 
Chapter 8. APPLICATION OF RESIN REFINING TO INDONESIAN FISH OIL 153 
8.1. BACKGROUND 153 
8.2. OBJECTIVES 153 
8.3. METHODOLOGY 154 
8.3.1. Materials 154 
8.3.2. Methods 154 
8.4. RESULTS 155 
8.4.1. Effects of resin refming process on Indonesian fish oil 155 
8.4.2. Stability of refmed Indonesian fish oil 159 
8.4.3. Response of Indonesian fish oil producers to resin refining process 165 
8.5. DISCUSSION 167 
8.5.1. Effects of resin refining on chemical, physical and organoleptic 
properties of Indonesian fish oil 167 
8.5.2. Stability of refmed Indonesian fish oil 168 
8.5.3. Prospect of introduction of the resin refming process for Indonesian 
fish oil 169 
8.6. CONCLUSIONS 170 
xi 
Chapter 9. DE1ERMINATION OF CANNED FISH PRODUCT 1YPE CONI'AINING 
REFINING FISH OIL AS A MAJOR INGREDIENf 171 
9.1. BACKGROUND 171 
9.2. OBJECTIVES 172 
9.3. METIIODOLOGY 172 
9.3.1. Supermarket survey 172 
9.3.2. Cannery survey 173 
9.3.3. Consumer survey 173 
9.4. RESULTS 173 
9.4.1. Existing canned fish product in the market 173 
9.4.2. Production information for canned fish 178 
9.5.3. Consumer behaviour towards canned fish product 183 
9.6. DISCUSSION 190 
9 .6.1. Product type to be developed 190 
9.6.2. Prospects for proposed canned sardine with fish oil addition 191 
9.7. CONCLUSIONS 
Chapter 10. TOMATO SAUCE FORMULATION AND S1ERILIZATION CONDffiON 
SELECTION FOR FISH CANNING 
10.1. BACKGROUND 193 
10.1.1. Tomato sauce formulation 193 
10.1.2. Sterilization 194 
10.2. OBJECTIVES 195 
10.3. METIIODOLOGY 196 
10.3.1. Experiment 1: Tomato sauce formulation 196 
10.3.2. Experiment 2: Simulation study on the selection of sterilization 
condition for canned fish with disguised fish oil 197 
10.4. RESULTS 198 
10.4.1. Tomato sauce formulation 198 
10.4.2. Stability of fish oil during sterilization 205 
10.5. DISCUSSION 216 
10.5.1. Tomato sauce formulation 216 
10.5.2. Fish oil stability during sterilization 217 
10.6. CONCLUSIONS 219 
xii 
Chapter 11. DETERMINATION OF IMPORTANT FACTORS IN FISH CANNING AND 
CANNING PROCESS OPTIMIZATION 220 
11.1. BACKGROUND 220 
11.1. Canned fish 220 
11.1.2. Screening experimental design: Plackett and Burman 222 
11.2. OBJECTIVES 224 
11.3. METHODOLOGY 224 
11.3 .1. Materials 224 
11.3 .2. Methods 225 
11.4. RESULTS 227 
11.4 .I. Determination of important factors in canned fish processing 227 
11.4.2. Optimization of canning process 233 
11.5. DISCUSSION 238 
11.5.1. Changes in canned fish during processing 238 
11.5 .2. Canning process optimization 241 
11.6. CONCLUSIONS 242 
Chapter 12. PROSPECTS OF CANNED FISH PRODUCT WITH FISH OIL ADDITION 
IN INDONESIAN MARKET 243 
12.1. BACKGROUND 243 
12.2. OBJECTIVES 244 
12.3. METHODOLOGY 244 
12.3.1. Materials 244 
12.3.2. Methods 245 
12.4. RESULTS 245 
12.4.1. Changes during production trial 246 
12.4.2. Safety assessment of developed canned fish 250 
12.4.3. Product acceptability during consumer testing 251 
12.4.4. Opinions of Indonesian medical doctors to the product 259 
12.5. DISCUSSION 263 
12.5.1. Chemical and physical changes in canned fish during production 
trial 263 
12.5.2. Product safety and shelf life 265 
12.5.3. Prospect of developed canned fish in Indonesian market 265 
12.6. CONCLUSIONS 267 
xiii 
Chapter 13. GENERAL DISCUSSION AND CONCLUSION 268 
13.1. INTRODUCTION 268 
13.2. FISH OIL REFINING 268 
13.3. FISH OIL STABILITY 269 
13.4. DEVELOPMENT OF CANNED FISH ENRICHED WITH FISH OIL 273 
13.5. PROSPECT OF DEVELOPED CANNED FISH IN INDONESIAN 
MARKET 277 
13.6. ROLE OF SENSORY EVALUATION IN PROCESS AND PRODUCT 
DEVELOPMENT 278 
13.7. ROLE OF THE CONSUMER IN PRODUCT DEVELOPMENT 280 
13.8. RECOMMENDED FUTURE STUDIES 281 
13.8. GENERAL CONCLUSIONS 281 
REFERENCES 283 
APPENDICES 307 
xiv 
LIST OF TABLES 
Table 4.1. Raw fish used for fish meal production (as number of factories) 39 
Table 4.2. Fish oil production information obtained during survey 41 
Table 4.3. Price and buyers of fish oil (by number of factories) 42 
Table 4.4. Chemical, physical and sensory analysis of Indonesian fish oil 44 
Table 4.5. Fatty acid profJles of Indonesian fish oil (% fatty acid) 46 
Table 4.6. Chemical, physical, sensory and fatty acid profJles of New Zealand oils 47 
Table 4.7. Classification of Indonesian fish oil quality in terms of FFA value 50 
Table 5.1. Effects of fish oil - resin volume ratio on fatty acid profile of crude fish oil 
(%fatty acid) 62 
Table 5.2. Effect of fish oil - resin volume ratio on fatty acid profJle of orange roughy oil 
(%fatty acid) 63 
Table 5.3. Effect of multiple refming on fatty acid profJle of crude fish oil (%fatty acid) 70 
Table 5.4. Effect of multiple refining on fatty acid profJle of orange roughy oil(% fatty acid) 70 
Table 5.5. Effect of vacuum pressure during resin refming on fatty acid profile of crude 
fish oil(% fatty acid) 76 
Table 5.6. Effect of vacuum pressure during resin refming on fatty acid profile of orange 
roughy oil(% fatty acid) 76 
Table 5.7. Effects of height size of resin packed column on fatty acid profJle of crude 
fish oil(% fatty acid) 82 
Table 5.8. Changes of natural antioxidant content of fish oil during refining process (ppm) 88 
Table 5.9. Relative amounts of volatile flavour compounds of crude oil during refining 90 
Table 5.10. Relative amounts of volatile flavour compounds of orange rougby oil during 
refining 92 
Table 6.1. Rate constant of  zero- and first-order reactions of  each parameter during storage 
of fish oil at various storage temperatures 111 
Table 6.2. Calculated shelf life of refined and unrefined fish oil based on the odour and 
taste parameters from various storage temperatures (weeks) 
Table 6.3. Estimated shelf life of fish oil at various temperatures (weeks) 
Table 6.4. Regression analysis between odour score and other parameters (peroxide value 
colour absorbance value and refractive index value) 
113 
115 
117 
XV 
Table 8.1. Fatty acid profile changes in fish meal and canning waste oils during resin 
refining process 157 
Table 8.2. Tocopherol content of fish meal and canning waste oils dUring refming process 
(ppm) 158 
Table 8.3. Results of fish meal factory survey about the response to resin refming process 
(number of factories) 166 
Table 9.1. Percentage of canned fish product type on the Indonesian market according to 
fish species 174 
Table 9.2. Distribution of canned fish product in the market according to medium used 175 
Table 9.3. Distribution of canned fish product in the market according to can type used 176 
Table 9.4. Distribution of canned fish product based on the relation between fish species 
and can type 177 
Table 9.6. Fish species used for canned fish production 179 
Table 9.7. Medium used for canned fish production 180 
Table 9.8. Fish species used for canned fish production for local market 181 
Table 9.9. Response of canneries to the idea "canned fish with disguised fish oil" 182 
Table 9.10. Demographic characteristics of respondents 183 
Table 9.11. Consumption frequency of fish and fish product 184 
Table 9.12. Preference of respondents to consume refmed fish oil 185 
Table 9.13. Fish oil consumption suggested by respondent 186 
Table 9.14. Respondent preference for a certain fish species and medium in buying canned 
fish 187 
Table 9.15. Fish species and medium chosen by respondents in buying canned fish 187 
Table 9.16. Respondent attitude to the idea of canned fish with disguised fish oil 188 
Table 9.17. Respondent preference to medium type, can size and price for proposed 
canned fish product 189 
Table 10.1. Total organoleptic score of the tomato sauce products of the frrst formulation 200 
Table 10.2. Effects of main ingredients on sensory properties of tomato sauce 201 
Table 10.3. Total organoleptic score of tomato sauce products of the optimisation experiment 204 
Table 10.4. Fatty acid profiles changes of fish oil during sterilization 212 
Table 11.1. Design matrix for screening important factors in fish canning 223 
Table 11.2. Variables and limits for Plackett and Burman design of canned fish 226 
Table 11.3. Results of chemical analysis of fish and tomato sauce 
Table 11.4. Results of colour analysis of fish flesh 
Table 11.5. Results of sensory evaluation for fish 
228 
229 
229 
xvi 
Table 11.6. Results of sensory evaluation for tomato sauce and overall acceptability for 
canned fish product 230 
Table 11.7. The main effects and significance levels of process variables on the 
characteristic of canned fish 231 
Table 11.8. The main effects and significance levels of seasoning on the characteristic of 
canned fish 232 
Table 11.9. Chemical and physical changes in fish and tomato sauce during optimization 
experiment 234 
Table 11.10. Sensory changes in fish during optimization experiment 236 
Table 11.11. Sensory changes in tomato sauce and overall acceptability of the product 
during optimization experiment 
Table 12.1. Fish and canned fish product weight changes during production trial 
Table 12.2. Hunter-1, -a and -b values changes in both fish flesh and tomato sauce medium 
237 
247 
during production trial 248 
Table 12.3. Proximate composition changes in the caimed fish during production trial(%) 249 
Table 12.4. Results of stability study on the oil in tomato sauce medium due to 
treatment during production trial 249 
Table 12.5. Fatty acid profile changes in canned fish due to sterilization treatment 
during production trial 250 
Table 12.6. Canned fish characteristics and acceptability in consumer testing 252 
Table 12.7. Acceptability of developed canned fish product in consumer test by demographic 
characteristics 254 
Table 12.8. Buying trend of developed canned fish in consumer testing by demographic 
characteristic 255 
Table 12.9. Buying trend of developed canned fish according to consumer testing 
acceptability and consumer experience in buying canned fish product 257 
Table 12.10. Buying criterion, retain outlet, label information and price of product 
suggested by consumer testing 258 
Table 12.11. Medical doctors advising the patients to consume fish and fish oil 260 
Table 12.12. The ways advised by Indonesian medical doctors to deliver fish oil to 
consumer 261 
Table 12.13. Comments of medical doctors on the product idea and the prospect of the 
product in the market 
Table 13.1. Experimental design used for each experimental stage 
262 
274 
xvii 
LIST OF FIGURES 
Figure 5.1. Effects of fish oil-resin volume ratio on free fatty acid value of fish oil 59 
Figure 5.2. Effects of fish oil-resin volume ratio on refractive index value of fish oil 60 
Figure 5.3. Effects of fish oil-resin volume ratio of colour absorbance value of fish oil 61 
Figure 5.4. Effects of fish oil-resin volume ratio on odour score of fish oil 64 
Figure 5.5. Effects of fish oil-resin volume ratio on taste score of fish oil 65 
Figure 5.6. Effects of multiple refining on free fatty acid value of fish oil 66 
Figure 5.7. Effects of multiple refming on refractive index of fish oil 67 
Figure 5.8. Effects of multiple refming on colour absorbance value of fish oil 68 
Figure 5.9. Effects of multiple refming on odour score of fish oil 71 
Figure 5.10. Effects of multiple refining on taste score of fish oil 72 
Figure 5.11. Effects of vacuum pressure during refining on free fatty acid value of fish oil 73 
Figure 5.12. Effects of vacuum pressure during refming on refractive index value of fish oil 74 
Figure 5.13. Effects of vacuum pressure during refining on colour absorbance value of fish oil 75 
Figure 5.14. Effects of vacuum pressure during refining on odour score of fish oil 77 
Figure 5.15. Effects of vacuum pressure during refining on taste score of fish oil 78 
Figure 5.16. Effects of various height-diameter ratios of column on fatty acid value of 
?? ? 
Figure 5.17. Effects of various height-diameter ratios of column on refractive index value 
??? w 
Figure 5.18. Effects of various height-diameter ratios of column on colour absorbance 
value of fish oil 81 
Figure 5.19. Effects of various height-diameter ratios of  column on odour score of  fish oil 82 
Figure 5.20. Effects of various height-diameter ratios of column on taste score of  fish oil 83 
Figure 5.21. Effects of  various diameter sizes of  column on free fatty acid value of  fish oil 84 
Figure 5.22. Effects of various diameter sizes of column on refractive index value of fish oil 85 
Figure 5.23. Effects of various diameter sizes of column on colour absorbance value of 
fish oil 86 
Figure 5.24. Effects of various diameter sizes of column on odour score of fish oil 87 
Figure 5.25. Effects of various diameter sizes of column on taste score of fish oil 87 
Figure 5.26. Traces of volatile flavour compounds of unrefined crude oil 89 
Figure 5.27. Traces of volatile flavour compounds of refmed crude oil 89 
Figure 5.28. Traces of volatile flavour compounds of unrefined orange roughy oil 91 
xviii 
Figure 5.29. Traces of volatile flavour compounds of refmed orange roughy oil 91 
Figure 6.1. Peroxide value changes in fish oil during storage at various temperatures 104 
Figure 6.2. Refractive index changes in fish oil during storage at various temperatures 105 
Figure 6.3. Colour absorbance value changes in fish oil during storage at various 
temperatures 106 
Figure 6.4. Odour score changes in fish oil during storage at various temperatures 108 
Figure 6.5. Taste score changes in fish oil during storage at various temperatures 109 
Figure 6.6. Linear relationship between the natural logarithm of rate constant of fish 
oil and the reciprocal of absolute temperature 114 
Figure 6.7. Linear relationship between the natural logarithm of estimated shelf life and 
the reciprocal of absolute temperature 116 
Figure 7.1. Effects of various antioxidants on peroxide value changes in fish oil during 
storage at 63?2?C 127 
Figure 7.2. Effects of various antioxidants on.TBA value changes in fish oil during 
storage at 63? 2?C 129 
Figure 7.3. Effect of various antioxidants on anisidine value changes in fish oil during 
storage at 63?2?C 130 
Figure 7.4. Effects of various antioxidants on totox value in fish oil during storage 
at 63?2?C 131 
Figure 7.5. Effects of various antioxidants on colour absorbance value changes in fish oil 
during storage at 63?2?C 132 
Figure 7 .6. Effects of various antioxidants on refractive index changes in fish oil during 
storage at 63?2?C 133 
Figure 7.7. Effects of various BHA levels on peroxide value changes in fish oil during 
storage at 63?2?C 135 
Figure 7 .8. Effects of various BHA levels on TBA value changes in fish oil during 
storage at 63?2?C 
Figure 7.9. Effects of various BHA levels on anisidine value changes in fish oil during 
storage at 63? 2?C 
Figure 7.10. Effects of various BHA levels on totox value changes in fish oil during storage 
136 
137 
at 63?2?C 138 
Figure 7.11. Effects of various BHA levels on colour absorbance value changes in fish 
oil during storage at 63?2?C 
Figure 7.12. Effects of various BHA levels on refractive index changes in fish oil during 
storage at 63? 2?C 
139 
140 
xix 
Figure 7.13. Effects of vacuum package on peroxide value changes in fish oil during 
storage at 63?2?C and 3Q?.2?C 141 
Figure 7.14. Effects of vacuum package on TBA value changes in fish oil during storage 
at 63?2?C and 3Q?.2?C 142 
Figure 7.15. Effects of vacuum package on anisidine value changes in fish oil during 
storage at 63?2?C and 3Q?.2?C 143 
Figure 7.16. Effects of vacuum package on totox value changes in fish oil during storage 
at 63?2?C and 3Q?.2?C 144 
Figure 7.17. Effects of vacuum package on colour absorbance value changes in fish 
oil during storage at 63?2?C and 3Q?.2?C 146 
Figure 7.18. Effects of vacuum package on refractive index value changes in fish oil 
during storage at 63?2?C and 3Q?.2?C 147 
Figure 7.19. Effects of vacuum package on odour changes in fish oil during storage 
at 3Q?.2?C 148 
Figure 8.1. Free fatty acid value changes in fish meal and canning waste oils during 
resin refining process 155 
Figure 8.2. Refractive index value changes in fish meal and canning waste oils during 
resin refining process 156 
Figure 8.3. Colour absorbance value changes in fish meal and canning waste oils during 
resin refming process 157 
Figure 8.4. Odour score changes in fish meal and canning waste oils during resin refining 
process 159 
Figure 8.5. Peroxide value changes in both refmed and unrefmed fish meal and canning 
waste oils during storage at 63?2?C 160 
Figure 8.6. TBA value changes in both refined and unrefmed fish meal and canning 
waste oils during storage at 63.?2?C 160 
Figure 8.7. Anisidine value changes in both refmed and unrefmed fish meal and canning 
waste oils during storage at 63?2?C 161 
Figure 8.8. Totox value changes in both refined and unrefined fish meal and canning 
waste oil during storage at 63?2?C 162 
Figure 8.9. Colour absorbance value changes in fish meal and canning waste oil during 
storage at 63?2?C 163 
Figure 8.10. Refractive index value changes in fish meal and canning waste oils during 
storage at 63?2?C 
Figure 10.1. Mixture space showing areas of experiment 
Figure 10.2. Mixture space showing new areas of experiment 
164 
199 
203 
XX 
Figure 10.3. Peroxide value changes in fish oil during sterilization at various temperatures 
and times 206 
Figure 10.4. Anisidine value changes in fish oil during sterilization at various temperatures 
and times 207 
Figure 10.5. TBA value changes in fish oil during sterilization at various temperatures 
and times 208 
Figure 10.6. Totox value changes in fish oil during sterilization at various temperatures 
and times 209 
Figure 10.7. Free fatty acid value changes in fish oil during sterilization at various 
temperatures and times 210 
Figure 10.8. Colour absorbance changes in fish oil during storage at various temperatures 
and times 211 
Figure 10.9. Fishy odour score changes in fish oil during sterilization at various 
temperatures and times 213 
Figure 10.10. Rancid odour score changes in fish oil during sterilization at various 
temperatures and times 214 
Figure 10.11. Fishy taste score changes in fish oil during sterilization at various 
temperatures and times 215 
Figure 10.12. Rancid taste score changes in fish oil during sterilization at various 
temperatures and times 215 
Figure 13.1. Oxidation of glyceride leading to rancidity of oil (Sherwin, 1990) 271 
Figure 13.2. Mechanism of malonaldehyde formation (Erickson and Bowers, 1976) 272 
Figure 13.3. Experimental stages used to develop the canned fish with fish oil disguised 
in it 
Figure 13.4. Process flow of canned fish with fish oil disguised in it 
273 
276 
xxi 
LIST OF APPENDICES 
Appendix 3.1. Purging system for collection of volatile flavour compounds of fish oil 307 
Appendix 3.2. Container used for colour analysis of fish flesh and tomato sauce 308 
Appendix 4.1. Questionnaire used for fish meal factory survey 309 
Appendix 4.2. Sensory evaluation sheet for Indonesian fish oil 311 
Appendix 4.3. Factories participating in the survey 312 
Appendix 4.4. Fatty acid proftles of Indonesian fish oils 313 
Appendix 5.1. Results of chemical, physical and organoleptical analysis of fish oil as the 
effects of fish oil and resin ratio 326 
Appendix 5.2. Results of chemical, physical and organoleptical analysis of fish oil as the 
of multiple refming using resin packed column 321 
Appendix 5.3. Results of chemical, physical and organoleptical analysis of fish oil as the 
effects of vacuum pressure application during resin refining process 326 
Appendix 5.4. Results of chemical, physical and organoleptical analysis of fish oil as the 
effects of various height sizes of resin packed column 329 
Appendix 5.4. Results of chemical, physical and organoleptical analysis of fish oil as the 
effects of various diameter sizes of resin packed column 332 
Appendix 6.1. Score sheet used for sensory evaluation during fish oil storage experiment 334 
Appendix 6.3. Linear relationship between the natural logarithm of rate constant and the 
reciprocal of absolute temperature for each parameter 335 
Appendix 6.4. Linear relationship between the natural logarithm of shelf life (8) versus 
the reciprocal of absolute temperature eK) 336 
Appendix 6.5. Results of chemical, physical and organoleptic analysis of refmed and 
unrefined fish oil during storage at various temperatures 337 
Appendix 7.1. Relationship between degree of vacuum and residual oxygen content 
(CIG Ltd, 1989) 341 
Appendix 7.2. Permitted antioxidants to be used in Indonesian foods and drinks according 
to Health Ministry Regulation No.10178/NSKn4 342 
Appendix 7.3. Results of chemical and physical analysis of fish oil as the effects of various 
antioxidant addition during storage at 63?2?C 343 
Appendix 7.4. Results of chemical and physical analysis of fish oil as the effects of various 
levels of BHA addition during storage at 63?2?C 346 
Appendix 7.5. Results of chemical and physical analysis of fish oil during storage in 
vacuum package at 63?2?C and 30?,2?C 349 
xxii 
Appendix 8.1. Questionnaire used for fish meal factory survey 356 
Appendix 8.2. Results of chemical, physical and sensory analysis of Indonesian fish 
? 3? 
Appendix 8.3. Results of chemical and physical analysis of fish oil during storage at 63?2?C 359 
Appendix 9.1. Questionnaire used for supermarket survey 361 
Appendix 9 .2. Questionnaire for cannery survey 362 
Appendix 9.3. Questionnaire used for consumer survey 
Appendix 9.4. Canned fish product being available in Indonesian market 
Appendix 9.5. Dimensions of can found in the market 
Appendix 9.6. Chi-square, degree of freedom and Cramer's V of Crosstab analysis 
results for consumer survey 
Appendix 10.1. Sensory form used for evaluating tomato sauce acceptability 
Appendix 10.2. Sensory form used for evaluating sterilized fish oil 
Appendix 10.3. The example of the calculation of ingredient effects 
Appendix 10.4. Results of chemical, physical and sensory analysis of fish oil as the 
effect of sterilization treatment 
Appendix 11.1. Sensory sheet for Plackett and Burman experiment of canned fish 
Appendix 11.2. Sensory sheet for experiment on canning process optimization 
Appendix 12.1. Questionnaire form used for consumer product testing 
Appendix 12.2. Questionnaire form for medical doctor survey 
Appendix 12.3. Information on the label of the developed canned fish product distributed 
during consumer testing 
Appendix 12.4. Medical doctor's comments on developed canned fish 
Appendix 12.5. Fatty acid profile changes in fish oil and canned fish product during 
356 
359 
373 
374 
375 
376 
377 
378 
383 
385 
387 
395 
397 
398 
production trial 399 
Appendix 12.6. Chi-square, degree of freedom and Cramer' s V of crosstab analysis 
results from consumer product testing 400 
Appendix 12.7. Chi-square, degree of freedom and Cramer's V of crosstab analysis results 
from medical doctor survey 402 
1 
Chapter 1 
INTRODUCTION 
Indonesia is an archipelago consisting of approximately two million square kilometres of land and 
approximately 5.8 million square kilometre water including 2.7 million square kilometres of 
exclusive economic zone (Soekandar and Mihardjo, 1989). The potential production from marine 
resources is approximately 6.6 million tonnes per year, with a marine fishery production of 2.27 
million tonnes. Total Indonesian fishery production in 1989 was 3.04 million tonnes (Directorate 
General Of Fishery, 1991). The export volume and value of fish products in 1990 were 0.3 million 
and one billion US dollars respectively (Directorate General Of Fishery, 1991). The average 
monthly spending for an Indonesian on fish was Rp.1,188 (Central Bureau Of Statistic, 1991). 
These statistics indicate that the fisheries sector is important in the Indonesian economy. 
Fish oil consumption is still a new product for the Indonesian consumer, even though fish oil has 
been traditionally prescribed by medical practitioners because of its importance as a source of 
vitamin A. Fish oil consumption in healthy people is relatively uncommon. The benefits of fish 
oil as a health supplement has been widely reported both in the press and in scientific publications. 
Since the Indonesian fish oil industry is relatively underdeveloped, it is predicted that this market 
sector will further expand as abundant raw materials are efficiently utilized. For example, shrimp 
trawl by-catch, as reported, was approximately 0.9 million tonnes in 1984 (Directorate General of 
Fishery, 1985). If the fish oil industry can be developed properly, the industry will move from its 
current status as a by-product of the fish meal industry to the dominant processed fish product. 
The exact total volume of fish oil production is unknown. However Indonesia currently exports 
approximately 1,170 tonnes of fish oil, with a value of $US 11 million in 1990 (Directorate 
General Of Fishery, 1991). The most common problem associated with fish oil consumption is its 
undesirable odour. Indeed, it is common practice in fish canning to intentionally remove the fish 
oil in order to improve product acceptability. 
Refining is mostly used to improve fish oil quality chemically, physically and organoleptically. 
Refining does not only improve fish oil quality, but also adds value and utility. With improved 
quality, fish oil can be used to replace vegetable oils in some cases, but some technological 
considerations should be given special attention, particularly in relation to the high content of 
2 
unsaturated fatty acids. 
Since fish oil is still a by-product of fish meal manufacture, the improvements in product quality 
are not yet receiving urgent attention by fish processing factories in Indonesia. Manufactures are 
satisfied with the present conditions, since they have had an established market for the oil. With 
this in mind, establishing a suitable method for improved fish oil refmement should be carefully 
selected to ensure rapid implementation of any new technology. For example, the refining 
technology should be relatively simple, have a low labour content, and have low installation and 
operation costs. 
A new market for refmed fish oil should be created, allowing fish oil producers to benefit from the 
To date, the published scientific studies on Indonesian fish oil and fish oil industries are very 
limited. Therefore, intensive studies are required to provide information for food industries, 
particularly about fish oil quality and quality improvement methods. 
THE OBJECTIVES OF TIDS STUDY 
The four objectives of this study are 
* to develop a relatively new and improved resin based fish oil refining technology, with 
a view to improving oil value and utility; 
* to measure the chemical, physical and organoleptic improvements in fish oil quality and 
investigate the stability of the oil; 
manufactures; and 
* to develop a canned fish product with fish oil disguised in it for the Indonesian market. 
3 
Chapter 2 
GENERAL LITERATURE REVIEW 
2.1. CHEMICAL PROPERTIES OF FISH OIL 
Knowledge of the chemistry of fish oils is important for the development of high quality and high 
value fish products, and for evaluation of the nutritional importance of fish oils. Many studies have 
been reported on the chemical and nutritional properties of fish oil, highlighting the fact that fish 
oil is generally characterized as containing large groups of saturated and unsaturated fatty acids, 
mixed triglycerides (Gruger, 1967), fatty alcohol (wax ester), sterol (sterol ester), phosphoric acid 
and amines (phospholipids) (Morris and Culkin, 1989). 
2.1.1. Triglycerides and fatty acids 
According to Stansby (1979), fish oil differs from other natural oils in that the triglycerides have: 
* a much greater number of different constituent fatty acids; 
* a greater proportion of long chain fatty acids; 
* a considerable proportion of fatty acids which are in a highly polyunsaturated form (up 
to six double and 
* abundant qualities of long chain polyunsaturates of the omega-3 type. 
The carbon chain length of fish oil fatty acids commonly exceeds C-18, and a considerable portion 
of the fatty acids contain 20, 22 and, to a limited extent, 24 carbon atoms. The total proportion 
of these long chain fatty acids usually amounts to 25-30% of the total, and sometimes approaches 
50% of the total fatty acids (Stansby, 1969). 
Gruger (1967)has reported that fatty acids derived from fish oil are of three principal types: 
saturated, monounsaturated and polyunsaturated. 
Saturated fatty acids (SAFAs) normally constitute 20-30% of the total fatty acids of marine 
organisms and are thus readily available to fish from food chain organisms (Ackman, 1982). The 
4 
SAFAs have carbon chain lengths that generally range from C-12 (lauric acid) to C-24 (lignoceric 
acid). Traces of C-8 and C-10 acids may also be found in some fish oils. C-5 acids (isovaleric) 
have been identified in the jaw-bone oil of the dolphin and the porpoise (Gruger, 1967). In spite 
of the chemical variety of compounds only three saturated fatty acids are quantitatively important 
in fish oil: palmitic acid (C16:0) at 10-15%, myristic acid (C14:0) at 3-13% and stearic acid 
(C18:0) at 1-4% (Ackman, 1982). 
The monounsaturated type is comprised of monoethenoic acids and the polyunsaturated type is 
comprised of polyethenoic acids which contain from 2 to 6 ethylenic bonds per acid. The carbon 
chain lengths of the unsaturated acids range, generally, from C-14 (9-tetradecenoic acid) to C-22 
(4,7, 10,13, 16,19-docosahexaenoic acid). Small amounts of C-10 and C-12 monoenoic acids have 
been found in some fish oils. There are no naturally occurring acetylenic acids and hydroxy 
carboxylic acids presently known in fish oil (Gruger, 1967). 
Most of the polyunsaturated fatty acids (PUPAs) in fish oils occur as the omega-3 type and are 
related to linolenic acid. In contrast omega-6 type fatty acids commonly predominate in most other 
oils. These compounds are biosynthetically related to linoleic acid (Stansby, 1982). The ro-3 
PUF As in marine lipids receiving the most interest are eicosapentaenoic acid (EPA) and 
docosahexaenoic acid (DHA) (Pigott and Tucker, 1987). EPA (C20:5ro-3) and DHA (C22:6ro-3) 
are regarded as either nutritional supplements, or as therapeutic agents inhibiting a variety of 
pathological conditions in man and are popularly known as the "omega-3" fatty acids. These 
naturally make up, at most, only about 25% of fish and fatty acids (Ackman, 1988a). In the human 
body, linolenic acid can be slowly converted too both EPA and DHA, especially in the presence 
of large quantities of linoleic acid which compete for the same enzyme system (Yongmanitchai and 
Ward, 1989). Fish heads and eyes are a rich source of DHA (Anonymous, 1991). 
A number of unusual fatty acids have been found in marine fish. Odd carbon numbered fatty acids 
(C-15, C-17, C-19) have been identified in the body and gonad lipids of smelt. Baltic herring have 
been found to contain fatty acids of longer carbon chain length (C-24 to C-32) than in the usual 
marine acids (Moris and Culkin, 1989). 
The proportionate distribution of fatty acids in fish oils can be influenced by fish diet. In addition, 
the proportion of fatty acids to fish oils is also affected by environmental factors such as 
geographic location of catch, and season of the year, which may be related to water temperature 
(Gruger, 1967). Triglycerides in fish oil from different species are normally characterized by 
different fatty acid compositions (Windsor and Barlow, 1981). 
5 
2.1.2. Wax esters 
Wax esters are formed by esterification of long chain fatty acids with long-chain alcohols (Morris 
and Culkin, 1987). These compounds are found, often in abundance, in the oil fraction of a 
number of marine species, and presumably serve as an energy reserve (Malins, 1967). 
The fatty alcohol fractions of all marine wax esters are either comprised of saturated fatty acid 
16:0, for 20 - 80%, or monounsaturated, mainly 20: 1 and 22:1. The fatty acid components of wax 
eteys.??? ?????? i?l????? 
Ackman (1980) simply divided wax ester into two main classes, those rich in 16:0, and those rich 
in 22: 1 fatty acids. 
2.1.3. Phospholipids 
Phospholipids are water-insoluble compounds similar to triglycerides, but with a phosphorus 
component substituted for one fatty acid (Pigott and Tucker, 1987). Fish flesh contains about 0.5% 
phospbolipids (Sheppard et ill, 1978). The main phospholipids of marine animals are lecithins 
(phosphatidyl cholines), phosphatidyl ethanolamines, phosphatidyl serine and phosphatidyl inositol 
with minute quantities of sphingomyelin, lysophosphatylcholine, and cardiolipin (Malin, 1967; 
Moris and Culkin, 1989). Lecithins are usually present in the highest concentration. A number 
of other components common to land animal are also present in lesser amounts (Malin, 1967). 
2.1.4. Free fatty acids (FFAs) 
The splitting of the attachment between the glycerol and the fatty acids through hydrolysis results 
in the release of free fatty acids, the presence of which, in an oil, is commercially undesirable. 
Usually fish oil is sold on the basis of 3% free fatty acids with a maximum allowable content, 
normally 7%. The proportion can be up to 20% in certain cases (Windsor and Barlow, 1981). A 
partial hydrolysis will generate mono- and di-glycerides. Some glycerol produced may ultimately 
be broken up to C02 and H20 depending on the conditions (Patterson, 1989). 
6 
2.1.5. Ether groups 
Evidence accumulated suggests that the ether-containing lipids of fish occur primarily as diacyl 
glyceryl ethers in which the 1- and 2-positions of glycerol are esterified with fatty acids. The 
presence of a high percentage of glyceryl ethers in the non-saponifiable fractions of liver and body 
oil of a number of marine fish and mammals is reported, and chimyl, batyl and salicyl alcohols are 
assumed to be the principal constituents (Matins, 1967). Structurally related materials such as 
hydroxyalkyl glycerols or methoxyglyceryl ether are usually minor components in fish lipids 
(Ackman, 1980). 
2.1.6. Hydrocarbons 
Naturally occurring hydrocarbons are not usually important constituents of fish oils. It is well 
known that the highly unsaturated hydrocarbon, squalene-C3Jl50, is present in certain shark liver 
oils. The saturated isoparaffine, pristane, and the unsaturated hydrocarbon, zamene, have been 
found to accompany squalene in minute quantities (Toyama and Kaneda, 1965). The isoprenoid 
alkanes, pristane and phytane, together with odd-numbered C-15 to C-33 n-alkanes are commonly 
found in a wide range of fish species including herring, sprat, mackerel, cod, eulachon, plaice, 
gurnard, salmon, bass, whiting, and sole. The wide structural variation in this chemical class could 
be a useful chemotaxonomic tool for isolating food chains because of their relative metabolic 
stability (Morris and Culkin, 1989). 
Other unsaturated isoprenoid hydrocarbons such as C-10 and C-20 mono-, di- and tri-olefms have 
been found in some fish. Hydrocarbons usually constituent less than 1% of the total lipid 
fraction (Morris and Culkin, 1989). 
2.1.7. Sterols 
Sterols, together with vitamins A, D, and E, are the major components in the unsaponifiable portion 
of fish oils. Cholesterol itself is the most common sterol in most marine species (Kinsella, 1987), 
existing either in free form or as an ester. Typical examples of the range of cholesterol distribution 
include halibut liver oil 7.0%; Atlantic cod liver oil 0.3%; salmon egg oil 3.0%; commercial 
7 
pilchard oil 0.7%; and oil from shrimp waste 19.0% (Brody, 1965). 
2.1.8. Heavy metals 
Heavy metals which are frequently found in fish oils include mercury, lead, cadmium, arsenic and 
copper and zinc (Kinsella, 1988). The metal content of fish oil is clearly influenced by the metal 
content of fish tissue used as raw material and the type of processing and storage conditions of the 
oil. Phospholipid content of the oil increases on storage, and during production - this class of 
compounds carrying metals into the oil. Metals that are complexed by phospholipids are removed 
more easily by degumming and alkali refming (Kinsella, 1987). 
2.1.9. Pigments 
The variety of pigments which have been identified in fish oil arise from two sources: natural 
oil-soluble pigments found in fish, and colour changes occurring in fish oil either before, during 
or after processing. Carotenoids constitute the most common pigments in fish (Brody, 1965). 
These include: 
* asatacin - principal red colour in salmon oil; 
* fucoxantin - principal yellow pigment in pilchard oil; 
* xanthophylls - pilchard oil; 
* carotene - pilchard oil; and 
* toraxanthin and zeaxanthin - found in the skin of a large number of species. 
In conjugation with protein, chitin, calcium salts, or other lipids, carotenoids can yield blue, purple, 
green, orange, red, pink, and yellow colour (Pennock, 1977). 
Chlorophylls occur in pilchard oil and arise from dietary microscopic green plants (Brody, 1965; 
Cutting, 1969). 
8 
2.2. NUTRITIONAL PROPERTIES 
Nutritional factors affect health, quality of life, general well-being, and longevity in humans and 
are important criteria used by contemporary consumers to select food (Kinsella, 1988). Many 
authors have tried to show the nutritional superiority of fish oils as a human food. 
2.2.1. Essential fatty acids 
The humflJl ?1 ? ake;tll the fa y _9cids requit:?, except for e.- es?ti,U fa.t,? acids .?), 
linoleic acid (1'8:2ro-6) ancl -a-linoleni m:id 1 8?ro--3} (Fogerty, 989). 
Most fish oils are a rich source of essential polyenoic fatty acids which are reported to have 
numerous health and nutritive benefits such as eliminating various skin related disorders, promoting 
physical growth and for the integrity of cellular endoplasmic reticulum and mitochondrial 
membranes (Stansby, 1982; Kinsella, 1986). One of the most fascinating aspects of the biological 
effects of the polyenoic acids is their in vivo biosynthesis and catabolism. These reactions control 
various biological feedback mechanisms which finally regulate the natural abundances of these 
acids, e.g. the effect of acids of the omega-3, linolenic acid family, on the essential fatty acid of 
the omega-6, linoleic acid family (Jorgenson, 1967). The amount of the linoleic acid family of fatty 
acids in body oils of marine fish is around 3 %, with some species as high as 6 %. The content 
of linoleic family of fatty acids in the oil of fresh water fish is somewhat higher, perhaps in the 
range of 5-9% (Stansby, 1969). 
In EFA deficiency, the content of polyenoic acids of the oleic and palmitoleic acid families are 
significantly increased, because oleic and palmitoleic acids can be biosynthesized, while linoleic 
and linolenic can not (Jorgensen, 1967). 
According to Kinsella (1986), it is generally believed that, in humans, 1-2% of caloric intake of 
linoleic acid and up to 0.5% of linolenic acid is sufficient to provide the recommended daily 
requirement for essential fatty acids. Currently, the intake ranges from 5 to 10% of dietary 
calories. 
9 
2.2.2. Vitamins 
Fish flesh, visceral portion, and liver oils are rich in vitamin A and D (Pigott and Tucker, 1 987). 
The amounts vary widely within and between fish species. In contrast, the vitamin content of liver 
oil varies with species, age, size, sex, nutritional conditions, and spawning stage of the fish, as well 
as the geographical locale and season of catch. Variations in feeding and the reproductive cycle 
are most likely responsible for seasonal differences (Kinsella, 1987). Cod liver oil is normally used 
as vitamins A and D supplements (Gauglitz et al, 1974). 
The biologically active forms of vitamin A in fish oils are retinol (vitamin A1) and 3-dehydroretinol 
(vitamin A). Vitamin A1 is the predominant form, particularly in fish liver. Approximately 4-
20% of the total vitamin A content may be 3-dehydroretinol (Brody, 1%5 and Kinsella, 1 987). 
Most of the vitamin A in fish liver oil occurs in the form of esters which are more resistant to 
oxidation than the free-alcohol form. Some antioxidants, such as tocopherol, prevent deterioration 
of vitamin A. Pure vitamin A, extracted from fish, has a biological activity of about 4,000,000 
USP units per gram (Brody, 1965). 
Vitamin D occurs naturally in at least two major forms that show antirachitic activity (Brody, 
1965): vitamin D2? also known as "calciferol", or activated ergosterol, and vitamin D3 also known 
as "activated 7-dehydrocholesterol" and is the predominant vitamin D in fish oil. They differ in 
their physiological activity (Brody, 1965; Kinsella, 1987). Vitamin D is associated with calcium 
distribution and calcification of bone. In its active form, vitamin D promotes the synthesis protein 
involved in the distribution of calcium (Kinsella, 1988). Unlike vitamin A, vitamin D is not 
present in nonliver visceral oils in amounts exceeding liver oil content (Kinsella, 1987). 
Vitamin E (tocopherol) is also found in fish oil of which a.-tocopherol is the most abundant form. 
The content of tocopherol in fish body oil ranges from less than 1mg/100g oil to 75mg/100g oil 
(Kinsella, 1987). 
2.3. FISH OIL AND DISEASES 
Current interest in the biological effects of fish oil developed largely from studies on Greenland 
Eskimos. The main stimulus for such studies was the realisation that these people show little 
evidence of cardia vascular disease, although they eat a diet that is high in fat and animal protein. 
10  
This observation was linked to their high intake of seafoods, which are rich in omega-3 
polyunsaturated fatty acids (Carron, 1986; Burr, 1991). 
Highly unsaturated omega-3 (n-3) fatty acids in marine lipids can be important in preventing or 
reducing certain diseases (Pigott and Tucker, 1987). Only animals that are part of the food chain 
from the sea have these long chain omega-3 fatty acids, because these substances are made in the 
ftrst place, by phytoplankton - the tiny aquatic plants that serve as food for small ftsh. Actuaily 
a few plant oils have small amounts of linolenic acid, which is an omega-3 fatty acid. This 
particular omega-3 fatty acid has fewer double bonds and is shorter than the omega-3 fatty acids 
found in ftsh. Metabolically, it does not behave in the human body in the same way as fish oil 
omega-3 fatty acids do, but it can be converted to the type of omega-3s found in seafood 
(Nettleton, 1987). The best vegetable source of omega-3 fatty acids are linseed, rapeseed and 
soybean oils (Niazi, 1987). 
Omega-3 fatty acids in marine Iipids currently receiving the most interest are eicosapentanoic acid 
(EPA) and docosahexanoic acid (DHA) (Pigott and Tucker, 1987). Although linoleic acid 
competes with linolenic acid for the same enzymes in the desaturation and elongation reactions, 
the formation of arachidonic acid from linoleic acid is favoured (Kinsella, 1987). 
Unfortunately, the vital omega-3 fatty acids are susceptible to selective destruction during fish 
processing (Griggs, 1986). With this in mind, new processing methods designed for the production 
of high quality fish oils must be developed with care. 
Human diseases, which have been studied and claimed can be prevented and/or cured with fish oil, 
are cardiovascular diseases, atherosclerosis, hypertension, immune system deficiency, rheumatoid, 
arthritis, psoriasis, asthma, cancer, skin diseases, lupus, diabetes, and stroke (Robinson et al, 1985; 
Niazi, 1987; Nettleton, 1987; Kinsella, 1987; Anonymous, 1987; Pigott and Tucker, 1987; 
Anonymous, 1988, K.inderlerer, 1989; Stansby, 1990; Simopoulos, 1991 ). 
2.4. FISH OIL PRODUCTION 
Fish, used in fish meal production, shows a wide variation in the oil content within and between 
species. Intra species variation occurs with the changing seasons and with the age of ftsh (Young, 
1982). In general, the overall composition of ftsh oil in terms of its content of triglycerides, free 
1 1  
fatty acids, moisture , etc. is not species specific (Windsor and Barlow, 1981). Depending on the 
oil content, fish can be grouped into two categories: oily fish with a fat content of more than 2.5%, 
and non oily fish (lean fish or white fish) with a fat content of less than 2.5% (Barlow, 1977). The 
most important factor in the production of a high quality crude fish oil is the condition of the raw 
material available for processing. As far as possible, the fish should be undamaged and held under 
chilled conditions to effectively minimize the effect of microbial and enzymatic degradation of the 
fish tissue (Young, 1982). 
In tropical and temperate areas of the world, special consideration must be paid to developing the 
most appropriate post-harvest method. According to Barlow (1977), in cases where fish are caught 
at high temperatures and far away from the plant, chilling appears to be the most effective method 
of preserving bulk stored fish. Two methods of chilling may be considered: refrigerated or ice 
chilled water method, or mixing of ice with the fish. 
2.4.1. Extraction technology 
Processing methods normally used for the manufacture of crude oil from whole fish or waste 
products from food fish processing are now discussed. 
2.4.1.1. Wet rendering process 
This process is used almost exclusively for processing oily fish, including menhaden, herring, 
pilchard, and tuna cannery waste (Lee, 1963). The principal steps of the wet rendering process are: 
Cooking, where the oil and water in the fish can be easily separated from the solid protein. 
To obtain a satisfactory product, the cooking temperature and pressure must be 
tailored to the particular species of fish. 
Pressing, where a screw-type press squeezes both oil and water from the fish. 
Centrifugation is now used in preference to settling tanks for recovery of the oil from the 
aqueous liquid portion. 
For maximum separation and recovery, two centrifugation steps are used (Considine and Considine, 
12  
1982). The additional centrifuge steps ensure complete removal of solids and aqueous fractions 
from the oil which could ultimately lead to rapid deterioration of the oil during storage. Normally, 
hot water is added to the oil to extract remaining impurities. Final separation and isolation of the 
oil is achieved by centrifugation with close control of temperature, since specific gravity and 
viscosity of the oil are temperature dependent (Windsor and Barlow, 1981). 
2.4.1.1. Dry rendering process 
This process is used primarily for raw material which is both relatively low in oil content and 
available in relatively small quantity (Lee, 1963; Pigott, 1967). The process is usually by batch, 
rather than continuous process, and involves a combined cooking and drying step. The raw 
material is normally loaded into a large, steam-jacked cylinder drier. Inside the drier is a rotating 
scraper which brings all material into quick contact with the hot inside wall, yet prevents the 
material from sticking. The drying is done either under vacuum, or at atmospheric pressure. The 
oil is separated from the dried scrap by batch pressing in a hydraulic press. Fish oil is the only 
product recovered from this pressing operation (Pigott, 1967; Lee, 1963; Considine and Considine, 
1982). 
2.4.1.3. Solvent process 
Solvent extraction has been used in the past when flsh oil was the product of choice such as in the 
preparation of vitamin oils from flsh liver. Some technological variations of this process are 
available depending on whether raw flsh, cooked or pressed flsh, dried scrap or flsh meal 
constitutes the starting material. Raw flsh can be handled in two ways: by an azeotropic process 
using solvents that are non-miscible with water, or by direct extraction, in which case, solvents 
must remove both water and fat (Lee, 1963). 
Typical of the azeotropic extraction technique is the two-step Vio Bin process which utilizes 
ethylene dichloride in the azeotropic distillation-liquid extraction. The flrst extraction takes place 
between 71 ?C, the azeotropic temperature, and 83?C, the normal boiling point of ethylene 
dichloride. The water is removed by distillation, thus preventing the extraction of any 
water-soluble nutrients from the flsh. A second extraction with isopropanol removes any traces of 
13  
ethylene dichloride (Pigott, 1967). 
Isopropanol has been used for direct extraction of fish oil. The process is basically simple. The 
coarsely ground fish is mixed with the hot solvent. Then the oil is removed by filtration and/or by 
centrifugation. The residue is then treated with successive batches of fresh solvent until water and 
oil are reduced to the desired point (Lee, 1963). 
2.4.1.4. Enzymatic process 
Despite the variety of techniques in this process, the basic approach is to remove or homogenize 
the fish and treat with approximately 0.25-0.50% enzymes based upon the weight of the whole 
fish. Following enzymatic digestion, the reactive material is pressed and filtered. Oil that is 
liberated by this type of process will float to the surface and can be skimmed or centrifuged in a 
manner similar to that used for reclaiming oil from stick water or press cakes in the conventional 
process (Pigott, 1967). 
Shirakawa et al (1989) proposed an enzymatic technique by using a protease. Fish are incubated 
with protease at 0.001-1 .0% by weight at 45-75?C for 40-50 minutes. The slurry formed is 
separated into solid and liquid fractions in a continuous decanter to produce fish meal and fish oil. 
This method yields comparable volumes of fish oil, as a conventional process, but with retention 
of valuable trace components which otherwise would be lost due to thermal decomposition. 
2.4.1.5. Silage Process 
Fish silage is liquified fish, stabilized against bacterial decomposition by an acid, such as formic, 
propionic, sulphuric or phosphoric acids. The process involves mincing the fish followed by the 
addition of the selected acid for preservation. The enzymes in the fish hydrolyse the endogenous 
proteins into smaller soluble units (peptide components). The acid provides the optimal hydrolysis 
conditions, while preventing bacterial spoilage. Fish oil is obtained by the centrifugation of the 
silage. It is noteworthy that silage made from white fish offal contains relatively little oil. 
However when made from fatty fish, such as herring, it is necessary to remove the oil (Bimbo, 
1990). 
14 
2.4.2. Processing of fish oil 
Processing fish oil for industrial products usually involves the application of winterization, refining 
and bleaching operations. 
2.4.2.1. Winterization 
Winterization is a thermomechanical separation process where component triglycerides of fats and 
oils are crystallized from a melt. The fraction is obtained via partial solidification of certain 
triglyceride components which are subsequently separated from the oil by filtration. Fat 
crystallization occurs in two steps: the first step involves a crystal formation process called 
nucleation, the second is crystal growth. Technically, the winterization process involves two 
component fractional crystallizations in terms 0f solid and liquid fractions. The fraction consists 
mainly of higher melting triglycerides, whereas the liquid fraction is dominant in low-melting 
components (Purl, 1980). 
In the winterization process, the oil must be cooled slowly to produce readily filtered crystals, often 
a time consuming and difficult operation. The traditional crystallization process normally takes five 
days, after which the oil is filtered, when cold, either through plate and frame filter presses, or 
through a rotary vacuum filter (Bimbo, 1989). The winterization may be terminated when the 
temperature falls below the lowest temperature attained immediately preceding the rise in 
temperature (Chang, 1967). 
2.4.2.2. Refining 
There are some refining methods which can be applied to fish oils. 
2.4.2.2.1. Degumming 
Degumming is only a partial refining, since free fatty acid is not reduced and even the gums are 
not completely removed (Gunstone and Norris, 1983). Principally, degumming is the process by 
which phosphatides and certain other mucilaginous materials are removed from the fish oil by 
1 5  
treatment with 2-3% water, or with an aqueous solution of boric acid or salt such as sodium 
chloride or pyrophosphate at 30?-50"C. Phosphatides can also be insolubilized with 80% 
phosphoric acid. The sludge is then removed by centrifugation (Chang, 1967; Kinsella, 1987). 
2.4.2.2.1. Alkali refining method 
The alkali (caustic soda) refining method is carried out in batch, semi- continuous and fully 
continuous modes. Of these, the most favoured is the centrifugal, continuous, refining line. After 
the degumming treatment already described, the hot oil is treated with 4N caustic soda to neutralize 
the free fatty acid content and to solubilize phospholipids, nitrogen and sulphur containing 
compounds and some pigments. In this way a large part of these impurities can be removed in the 
aqueous phase resulting from the first centrifugation step. Oils of poor quality are then given a 
second caustic soda treatment primarily to remove more sulphur, phosphorus and colour followed 
by phase separation. The oil is then washed with water to remove soap, and centrifuged before 
drying. Average to good quality oils are given a water wash in place of the second caustic soda 
treatment. To ensure removal of the soap from the oil, phosphoric or citric acids are often added 
to the fmal wash water. The acids also act as trace metal deactivators either by chelation or by the 
formation of water soluble salts (Young, 1982). 
The minimum amount of alkali required for neutralization can be calculated from the free fatty 
acids (FFA) of the oil to be refmed, using the formula (Gunstone and Norris, 1983): 
% NaOH = % FFA x 0.142, 
where FF A is expressed in terms of oleic acid. For any desired excess of NaOH, the calculation 
is 
% NaOH = % FFA x 0.142 + % excess NaOH 
From an application view point, alkali refining or neutralization of the oil results in a product 
which, when heated, will not darken, foam or smoke, become cloudy and then form a precipitate. 
The product can also be readily bleached (Bimbo, 1989). 
16 
2.4.2.2.3. Physical refining 
The physical refining method requires a thorough degumming of the oil prior to distillative removal 
of the fatty acids, heat degradable pigments and other impurities at temperatures in the order of 
250?C at 2-SmmHg absolute pressure with open injection (Young, 1982). 
Although this method produces higher yields, it has not yet proved practical due to variability in 
quality and quantity of impurities (especially sulphur) in the oil (Kinsella, 1987). 
2.4.2.2.4. Sodium carbonate method 
Sodium carbonate was once popular, because it. neutralized free fatty acids without saponifying any 
oil. In a second step, stronger alkali could be used for colour reduction, etc. Foaming problems 
and the equipment required, have caused this method to be discontinued in most refineries 
(Gunstone and Norris, 1983). 
2.4.2.3. Bleaching 
The main objective of oil bleaching is to reduce coloured materials and natural pigments, and to 
absorb suspended mucilaginous, colloid-like matters and any traces of soap if still present (Chang, 
1967). The process leads to a fish oil product with improved colour, flavour, and oxidative 
stability, free from impurities (Bimbo, 1989). 
The process involves heating the oil to 90?-l10?C, often under vacuum condition, in the presence 
of bleaching clay (0.2-0.3% by weight depending on oil quality), for the desired period of time and 
then removal of the spent clay by filtration (Bimbo, 1989; Young 1982). Semi-continuous 
processing plants are preferred for this treatment. Efficient filtration for the removal of the clay 
is essential because impurities adhering to the clay particles remaining in the oil act as autoxidation 
catalysts. 
The activated clay, made by acid treatment of neutral clay, is more effective in bleaching. The 
activated clay may cause a rise in the free fatty acid content of bleached oil. Activated carbon is 
17 
normally used in combination with clays as an effective means of reducing the odour of fish oils 
(Chang, 1967). 
2.4.2.4. Deodorization 
The oxidation products of the highly unsaturated fatty acids of fish oils, whether free or bound in 
the triglycerides, are generally regarded as the causative agents of the characteristic fish odour of 
fish oils. One method to alleviate, but not entirely eliminate, this undesirable condition, involves 
vacuum dry oxygen free steam distillation of the oil at a relatively high temperature (Brody, 1965; 
Windsor and Barlow, 1981). 
Hydrogenation of fish oils free from non-oil .fishy material is another method for reducing fish 
odour. This is achieved by adding approximately 5% calcium hydroxide (slaked lime) and 5% 
calcium oxide (quicklime), to the oil and then agitating and filtering. Under these conditions the 
oil becomes simultaneously deodorized, decolourized, stabilized and also partially refmed (Brody, 
1965). 
2.5. INDUSTRIAL APPLICATIONS OF FISH OIL 
Numerous studies have reported on optimising the use of fish oils in food, pharmaceutical and 
animal feed industries. Further, the application of fish oils for non-edible uses have been identified. 
2.5.1. Fish oil applications in foods 
Over one million MT (metric tonnes) of marine oil are used annually in food products. Most of 
the world's marine oil production is consumed in Europe, South America and Japan, primarily in 
the forms of salad oils, frying fats, table margarines and other spreads, low caloric pastries, cakes, 
cookies, biscuits and synthetic creams. Other uses of fish oil in food include hard fats, shortening, 
pastry fats, bread fats, frying oils, cake shortenings, bread dough, biscuit fillings, canning oils, 
cooking oils, and emulsifier (Brody, 1965; Pillai, 1974; Young, 1982; Bimbo, 1987; Bimbo, 1989a; 
1 8  
Barlow et .ill_, 1990). Crude fish oil used for these purposes must be winterized, alkali refmed and 
bleached. In some cases, hydrogenation and possibly deodorization may be required (Birnbo and 
Crowther, 1990). 
2.5.2. Fish oil applications in pharmaceuticals 
Probably the most important pharmaceutical properties of fish oil is directly related to the relatively 
high polyunsaturated fatty acids (omega-3), with vitamin A and D content. 
2.5.2.1 .  Concentrate of omega-3 fatty acid 
Fish oils can be concentrated by several means, starting with the simple winterization or slow 
chilling of the oil (Ack:man, 1988a) or by using urea complexation method (Haagsma et .ill_, 1982). 
In an experiment conducted by Haagsma et al (1982), the urea complexation method was found 
to give better results with respect to yield and fatty acid composition in the concentrate. The urea 
fractionation of fatty acids is mainly based on the degree of unsaturation: the more unsaturated, the 
less they will be included in the urea crystals. 
Tocher et al (1986) described a simple and rapid method for the preparation of a marine oil fraction 
highly enriched in (ro-3)-polyunsaturated fatty acids, using cod roe. Incubation of a concentrated 
aqueous extract of the roe with porcine pancreatic phospholipase A2 (EC 3.1.1 .4) took place 
followed by extraction of the freeze-dried reaction product with diethyl ether containing formic acid 
which produced an oil yield of l .Og/lOOg wet weight. The oil contained over 95% of free fatty 
acids, with 20:5ro-3 and 22:6ro-3 accounting for up to 24% and 40%, respectively, of the total free 
fatty acids. 
Ack:man (1988a) noted that menhaden oil can be enriched without solvent from 13.9% EPA and 
9.7% DHA to 15.3% and 10.9% respectively. With a nominal concentration of one EPA or one 
DHA per triglyceride molecule, enrichment of triglycerides without enzymatic inter-esterification 
or resynthesis is not very practical beyond the 300mg/g usually listed for these two fatty acids 
alone. By adding other omega-3 fatty acid such as 16:4ro-3, 20:4ro-3, and 22:5ro-3 it is possible 
to obtain a total of 500mg/g. 
2.5.2.2. Capsule of omega-3 fatty acids 
Commercially available omega-3 fatty acidS fr 
tempeFature?with no otwio1:1s f-fe 
1 9  
288). 
Ackman et a1 (1989) analyzed seventeen brands of encapsulated fish oil or fish oil concentrate 
products consisting of two product types - triglyceride and alkyl ester oils. The alkyl ester and 
free fatty acid products showed very high levels of EPA (259-300mg) and DHA (172-254mg). In 
contrast triglyceride oils had relative low levels of EPA (80-250mg) and DHA (78-156mg) per 
gram of capsule content. 
2.5.2.3. Vitamin A And D concentrates 
Vitamin A and D from fish oils can be concentrated using commonly available methods such as 
saponification, short-path distillation and adsorption (Brody, 1965). 
2.5.2.3.1. Saponification 
Saponification effectively splits oil triglycerides into its component parts, glycerol and fatty acids, 
with the concomitant formation of soaps. Incomplete reaction leaves some unsaponified oil, 
because the presence of some unsaponified oil renders the process more efficient. After dilution 
of the soaps, vitamins can be readily extracted with a water immiscible solvent such as diethyl 
ether, or ethylene dichloride. The extract is separated from the aqueous portion, and the solvent 
is removed by distillation. In the saponification process both vitamins A and D are extracted 
simultaneously. 
20 
2.5.2.3.2. Short-path distillation 
The efficiency of the equipment used in this method has been significantly improved with the 
development of the high vacuum pump operating in the region of one micron (0.001mmHg) 
pressure. Under high vacuum thin film of oil is distributed on a heated surface. Under these 
condition the vitamin concentrate distils and condenses on a nearby cooled surface. In this process, 
vitamin A and D can be removed from the oil separately. 
2.5.2.3.3. Adsorption 
Vitamin A in fish oil is concentrated using a two step process. Vitamin A is first converted to its 
alcohol form by metbanolysis. This alcohol is -then separated from the resulting fatty-acid methyl 
ester mixture by adsorption using alumina or silicic acid. Alternatively, a vitamin A concentrate 
can also be obtained either by alcoholysis followed by adsorption on alumina or by its adsorption 
on soap formed in situ. 
2.5.3. Fish oil applications in animal and fish feeds 
For many years fish oil has been added as a supplement to animal feeds. It is an economical 
source of calories, and stimulates growth. Growth enhancement results from the high concentration 
of linolenic acid homologies - omega-3 fatty acids. The fish oil used in animal rations must be 
fresh since autoxidized oils are known to be toxic to some domestic animals (Gauglitz et&, 1974). 
Animal feeds containing fish oil include poultry, pig, cattle, fish, mink and pet feeds (Karrick, 
1967; Gauglitz et ,ill.. 1974; Barlow, 1986; Watanabe and Takeuchi, 1989). 
2.5.4. Fish oil applications in non-edible uses 
Major industrial uses take advantage either of the unique type and high degree of unsaturation of 
fish oil to produce elastic durable polymers or of the long and diverse mixture of chain length to 
add lubricity, detergency, and plasticity functions (Fineberg and Johanson, 1967). In addition, the 
21 
unique compositions of fish oils make them adaptable to a wide number of uses. The highly 
unsaturated fatty acids of fish oils allow competition with drying oils of vegetable origin. The high 
percentage of saturated fatty acids in fish oil allow competition with fats and oils such as tallow 
and vegetable oils of low iodine value (Dyer, 1967). 
Non-edible uses of fish oil include soaps, detergents, leather tanning, protective coating, lubricants, 
plasticizers, pesticides, fungicides, polyurethane foams, buffing compounds, glazing compounds, 
caulking compounds, sealing compounds, ore floatation, air entraining agent for concentrate, water 
repellents, rubber compounds, synthetic waxes, corrosion inhibitor, automotive gaskets, core oils, 
tin plate oils, rust proofmg agents, refractory compounds, linoleum, presswood fibre boards, oiled 
fabrics, ceramics deflocculans, fermentation substrates, illuminating oils, fuel oils, mushroom 
culture and fire retardants (Dyer, 1967; Fineberg and Johanson, 1967; Bimbo, 1989) 
22 
Chapter 3 
MATERIAL AND ANALYSIS METHODS 
3.1. MATERIALS 
3.1.1. Fish oils 
Fish oils were obtained from both Indonesia and New Zealand. The Indonesian oils were collected 
from several fish meal factories in Muncar, East Java and Negara, Bali. The New Zealand fish oils 
were supplied by Sealord Product Ltd., Nelson. Details of fish oils used will be given for each 
experiment in the appropriate chapters. 
3.1.2. Resin 
Macroporous strong acid cation resin used for fish oil refining was supplied by Dow Chemicals, 
USA. The resin, consisting of a styrene/divinyl benzene matrix structure had sphere form, sodium 
ionic, 1-7 meq/mllmin total exchange capacity, 150 A pore size and 42-48% water retention 
capacity. Glass columns in which resin was packed were fabricated by the Massey University 
Glass Blower. 
3.1.3. Fish 
New Zealand sardine/pilchard (Sardinops neopilchardus) was used as raw material in the canning 
experiment. The fish was supplied by Star Fish Supply Ltd., Napier, New Zealand. The fish size 
was 17.5.?0.5cm in length, by 1 .8?0.1cm thick and 44.2?4.9g in weight. The proximate 
composition of the fish was 71 .8% moisture, 22.2% protein, 2.4% fat and 3 .1% ash. The pH of 
the fish was 6.3. The fish was received frozen, and was kept at -7SC until needed. 
23 
3.1.4. Can 
Cans (4.6 cm in height, 6.6 cm in diameter) used in the fish oil experiment were obtained from 
J.Watties Foods Ltd., Hastings, New Zealand. The cans (8.2 cm height, 5.1 cm diameter) used in 
the fish canning experiment were imported from Nippon Suisan Kaisha, Ltd., Tokyo, Japan. 
3.1.5. Tomato sauce materials 
Tomato paste with 28-30% tomato soluble solids was supplied by J.Watties Foods Ltd, Hastings, 
New Zealand. Salt, sugar, shallot, garlic and white vinegar were purchased from supermarkets and 
retailers in Palmerston North. 
3.2. FISH OIL EXTRACTION 
Fish oil samples extracted from fish flesh and tomato sauce were used for fatty acid prof'Iles, 
anisidine value and TBA value analysis. Modified Bligh and Dyer method was applied to extract 
the oil (Hanson and Olley, 1963). 
A sample (15g) was placed into a Waring Blendor-jar and diluted with 30ml CHCI3 and 60 ml 
CHPH. The mixture was then homogenized for two minutes. The sample was further diluted 
with 30ml CHC13 and rehomogenised for 30 seconds. This last step was repeated using 30 ml H20. 
The fat solution was separated by centrifuging the mixture at 9000 rpm for 10 minutes. The lower 
CHC13 layer containing the fish oil was separated off and evaporated using a rotary evaporator at 
45?C. 
24 
3.3. METHODS OF ANALYSIS 
3.3.1. Chemical analysis 
3.3.1.1. Free fatty acid value 
Analysis was carried out only on fish oil using the procedure described by Fernandez (1986). 
Samples (7g) were dissolved in 50ml ethyl alcohol and titrated with O.lN NaOH using a 
phenolphthalein indicator. The blank determination was carried out using the same procedure, but 
in the absence of fish oil. The free fatty acid (FF A) values were computed by the formula: 
(S-B) X N X 28.2 
FF A value (as % oleic acid)= ----------------------------?
weight of sample 
Where, S = titration of sample (ml) 
B = titration of blank (ml) 
N = normality of NaOH 
3.3.1.2. Fatty acid profile 
Fish oil and the oils extracted from fresh fish, sterilized fish, canned fish and tomato sauce were 
esterified prior to being analyzed for fatty acid content according to the procedure described by van 
Wijngaarden (1966). One ml of 6% methanolic potassium hydroxide was added to 20-25 mg of 
fish oil in a screw cap vial and then heated for 10 minutes in a water bath at 70"C with stirring. 
Two ml of 14% BF3-methanol was then added and the reaction mixture was heated for a further 
10 minutes. Two ml hexane was added and shaken thoroughly. Six ml of distilled water was then 
added and the reaction mixture shaken again. The hexane layer was separated for methyl ester 
analysis. 
The fatty acid methyl esters were analyzed by gas chromatography using a Hewlett-Packard 
Capillary gas chromatography Model 5890 Series II, fitted with a FID detector and equipped with 
DB-Wax capillary column (30 m x 0.25 mm ID). The carrier gas was nitrogen at 10psi. This 
equipment was connected to the Hewlett-Packard 3393 integrator. The detector was kept at 260"C 
and the injection port at 220?C. The column temperature was programmed from 160?C to 250?C 
25 
at the rate of 2?C per minute and held at the upper temperature for 25 minutes. The attenuation 
used was 4. The peaks were identified using the standards of PUF A-1 and PUF A-2 obtained from 
Supelco, Inc., Bellefonte, USA. 
3.3.1.3. Tocopherol And Tocotrienols Analysis 
Quantitative analysis of tocopherol and tocotrienol in fish oil was carried out before and after 
refining. Approximately O.lg of fish oil was accurately weighed into a 5 ml volumetric flask and 
then made up with hexane containing 200 ppm BHT antioxidant. Constituents were separated by 
using Maxima 820 high performance liquid chromatography (HPLC) Model 510 (Waters 
Associates, Milford, MA, USA) equipped with a Hitachi fluorescence spectrophotometer Model 
FlOOO (Hitachi Ltd., Tokyo, Japan). The excitation was set at 295nm and emission at 330nm. The 
mobile phase was HPLC grade hexane containing 7% diisopropyl ether and 0.05% acetic acid at 
the rate of 2ml per minute. A zorbax silica column (0.5J..L, 30cm x 3 .6mm) was used to provide 
the separation of a, ?. y and 8 tocopherols, tocomonoenol and tocotrienol. The column 
temperature was maintained at ambient temperature ? 20?C. The volume injected was 50J..Ll. 
3.3.1.4. Volatile Flavour Analysis 
An all glass-purging system for collection of volatile flavour compounds of fish oil was constructed 
as shown in Appendix 3 .1 .  The size of the tube was 20cm in length and 2.3cm in diameter. The 
length of the purge tube was 15cm, and was terminated with fme nozzles at one end. A 15ml fish 
oil sample was sparged with Nitrogen (N0 at 85ml/minute through the purging tube. The purging 
unit was placed in a constant temperature water bath held at 300C. Samples were prepurged for 
two minutes to remove oxygen from the tube to avoid oxidation during sample extraction. The 
volatile components were entrained and concentrated onto a porous polymer Tenax TA trap (60/80 
mesh size, Alltech Association, Illinois, USA) attached to the exit port of the purging unit. Purging 
time was four hours. To desorb the volatile flavour components, the tenax polymer was washed 
with 7.5 ml of triply glass distilled diethyl ether, and then concentrated by carefully passing a fine 
jet of N2 over the surface of the solution to evaporate most of diethyl ether. 
The volatile flavour concentrate was analysed using a gas chromatography-mass spectrometry (GC-
26 
MS). Volatile compounds were separated on a Hewlett-Packard 5890 Series II equipped with a 
DB-wax capillary column (30 m x 0.25 mm ID). The GC condition was set the same as in the 
fatty acid profile analysis, except the column temperature was programmed from 40?C to 280?C 
at a rate of 5?C/minute. GC peak identification and quantification were carried out using a VG70-
250S high resolution mass spectrometer. 
3.3.1.5. Peroxide value 
a. For fish oil 
Peroxide value analysis for fish oil was based on the modified method described by Windsor and 
Barlow (1981). A sample (2.5 g) was weighed into a 250ml erlenmeyer flask and diluted with 
15 ml of a acetic acid-chloroform (3:2) solution. The flask was then swirled until the sample 
dissolved. 0.25 ml of saturated potassium iodide solution was added. The solution was allowed 
to stand with occasional swirling for exactly 1 minute before diluting with 15 ml of distilled water. 
The solution was titrated with 0.1N sodium thiosulphate and vigorously shaken until the yellow 
colour almost disappeared. Approximately 0.5 ml of starch indicator was added and then the 
titration continued. The flask was shaken vigorously near the end point to extract all the iodine 
from the chloroform layer until the blue colour disappeared. 
A blank determination was carried out by omitting the oil sample, but using the same procedure 
and reagents. Peroxide value (PV) was calculated using the following equation: 
(S-B)(N)(1000) 
PV (meq/kg sample) = -----------------------------?
weight of sample (g) 
where : S = titration of sample (ml) 
B = titration of blank (ml) 
N = normality of sodium thiosulphate solution 
27 
b. For fish flesh and tomato sauce 
The method developed by Pearson (1973) was used to detennine peroxide value in the tomato 
sauce. A sample (approximately 30 g) was placed in a Waring Blendor and diluted with 1 50ml 
of chlorofonn. The mixture was blended for 2 minutes before filtering through Whatman filter 
paper. The filtrate was centrifuged at lO,OOOrpm for 10 minutes. The clarified solution was used 
for PV analysis. 
10 ml of solution was pipetted into a 150 ml erlenmeyer flask and diluted with 14.8 ml of glacial 
acetic acid and 0.4 ml of freshly prepared saturated potassium iodide solution. The solution was 
allowed to stand for exactly 1 minute with occasional swirling before further diluting with 12ml 
of distilled water. The solution was titrated with 0.01N sodium thiosulphate using a starch 
indicator solution. PV was calculated using following equation: 
titration x N x 1000 
PV (meqlk:g sample) = ----------------------------?
weight of sample (g) 
where: N = nonnality of sodium thiosulphate 
The weight of the sample was detennined by pipetting 10 ml of solution into a weighed aluminium 
dish. The solvent was evaporated at 102?C for 20 minutes and then reweighed, after cooling, in 
a desiccator. 
3.3.1.6. Thiobarbituric acid ? value 
TBA values for fish oil and oil extracted from tomato sauce were determined using the method 
outlined by Fioriti et al (1974). Samples (0.12g) were dissolved in 2 ml 50%(v/v) absolute alcohol 
in 2,2,4-trimethylpentane (isooctane) to facilitate analysis. To the solution in the test tubes, 
stoppered with plastic caps, was added 5 ml isoOctane and 3 ml thiobarbituric acid solution (0.33 
g TBA in 10 m1 distilled water and 90 ml isopropyl alcohol). The capped tubes were shaken 
vigorously for approximately 30 seconds, using a Chiltern shaker, and then held in a temperature 
controlled water bath at 60?C for exactly 30 minutes. The solution was scanned against pure 
28 
isoOctane at 532 run in a Shimadzu UV Spectrophotometer Model UV-1 10-02. Recorded 
absorbances were used to compute TBA values using the formula: 
(absorbance of sample - absorbance of blank) 10 
TBA value (Jllllollg sample) = -----------------------------------------------------------------
weight of sample (g) 
The absorbance for the blank was measured using the same procedure, but in the absence of the 
sample. 
3.3.1.7. Anisidine value 
The method described by Windsor and Barlow (1981) was used to determine anisidine value in the 
fish oil and the oil extracted from the tomato sauce. 
A 0.5 g-sample was weighed into a 25 ml volumetric flask. The sample was dissolved and made 
up to volume (25 ml) with isooctane and then shaken. The absorbance for the oil solution was 
measured against pure isooctane at 350 run in a Shimadzu-UV Spectrophotometer Model UV -1 10-
02. In a 2 test tube assay system, 5 ml of the oil solution was pipetted into test tube "A", and 5 
ml isooctane was pipetted into test tube "B ". One ml of the anisidine reagent (0.25 g para-anisidine 
in 100 ml acetic acid) was added to both test tubes. The tubes were stoppered and shaken 
vigorously and left in the darlc for exactly 10 minutes. The absorbance of sample and the blank 
were measured at 350 run in the same spectrophotometer. Anisidine values (A V) were calculated 
using the formula: 
25(1 .21;, - E.) 
A V = ---------------------------
weight of sample (g) 
where: E. = the net absorbance of the oil solution 
1;, = the net absorbance of fat-anisidine-solution 
29 
3.3.1.8. Total oxidation value (Totox value) 
A combination of primary oxidation and secondary oxidation led to the formulation of an empirical 
known as the "Total Oxidation Value" (Totox Value). This gives a useful, if not a complete picture 
of the present state and past history of an oil in oxidation terms (Patterson, 1989). Since a 
peroxide group has twice the oxygen of an aldehyde group, the conversion equation states are as 
follows: 
Where : PV = peroxide value 
An. V = Anisidine Value 
3.3.1.9. Moisture content 
Totox Value = 2 PV + An.V 
Samples were prepared by chopping and grinding fresh fish and canned fish in a mortar until finely 
ground. The moisture content of 2 g samples contained in aluminium dishes was determined by 
drying in an air oven at 100 ? 2?C for approximately 16 hours. Water content was determined by 
comparison of weight difference before and after sample drying. 
3.3.1.10. Protein Content 
Protein content of the samples (fresh fish and canned fish product) was measured by the Kjeldahl 
method using a Kjeltec 1026 System Distillating Unit (Tecator, Sweden), which is a semi-automatic 
apparatus. The sample was weighed accurately and put into the digestion tube. Two Kjeltab 3 .55 
(containing 3 .5 g K2S04 and 0.0035 g Se) and 12 ml cone. H2S04 were added. Digestion was then 
carried out using a Digestion System 1007 (Tecator, Sweden) at 420?C for 30 - 45 minutes. After 
cooling, the solution was diluted with 75 ml distilled water. Distillation was then applied using 
a distillation unit programmed to use 2 ml alkali with 0.2 minute delay time and 3.6 minute 
steaming time. The liberated ammonia was collected in 25 ml 4% Boric acid solution. When 
distillation was complete, the sample was titrated with 0.1M HCl to pink end point. In calculation, 
30 
1 ml 0.1M was defined as being equivalent to 1.4 mgN and a multiplication factor of 6.25 used 
to calculate crude protein from nitrogen content. 
3.3.1.1 1. Fat content 
The fat contents of the fresh fish and canned fish were determined using the official AOAC method 
(AOAC section 18.044, 1984). The sample (8g) was weighed into 50ml beaker and mixed with 
8ml HCI. The mixture was then heated in a steam bath for 90 minutes with occasional stirring. 
After cooling, the solution was transferred to a mojonnier fat extraction flask. The beaker was 
rinsed with 7ml ethanol and then with 25ml diethyl ether. The extraction flask was subsequently 
The Mojonier flask was centrifuged at 600rpm for 5 nii;?tes and the ether - fat solution was then 
To optimize the fat extraction, the second 
extraction step was introduced using 15ml diethyl ether and 15ml petroleum ether. This time the 
centrifugation step was reduced to 3 minutes. 
The ether-fat solution was dried using a two step process. Most of the solvent was removed using 
a rotary evaporator at 45?C and then the sample was completely dried by heating in an oven at 
100?_2?C for 40 minutes. 
3.3.1.12. Ash content 
The ash content of the fresh fish and canned fish samples was determined using the official AOAC 
method (AOAC section 18.025, 1984). A 2 g sample in silica dishes were burned and then heated 
in a furnace to approximately 6000C. Ashing took three hours. 
3.3.1.13. Salt Content 
The salt content of raw fish and canned fish were determined by the official AOAC method 
(AOAC section 18.035, 1984). The sample (5-10g) was placed in a 250ml erlenmeyer and diluted 
3 1  
with 50ml O.lM AgN03 and 20ml HN03. The mixture was boiled gently for 15minutes, cooled 
and further diluted with 50ml distilled water and 5ml of ferric indicator. The mixture was titrated 
with a O.lN NH4SCN solution until the mixture became permanently light brown. The salt content 
was calculated using the equation: 
(V 2 - V I) X N X 5.85 
Sodium Chloride (g/1 OOg) = -----------------------------?
Weight of sample 
where, V 1 = volume (ml) of thiocyanate used in the sample 
V 2 = volume (ml) of thiocyanate used in the blank 
N = normality of thiocyanate 
3.3.1.14 ? .1!!! 
The pH was determined using an Orlon pH-meter Model 720 (Orlon Research, Massachusette, 
USA). Fresh fish and canned fish samples (10 g) were prepared by mixing with distilled water in 
the ratio of 1 :2. The mixtures were homogenized and the pH measured immediately by immersing. 
The pH of tomato sauce was measured directly without additional treatment. 
3.3.2. Physical analysis 
3.3.2.1 . Colour 
a. Fish oil colour 
Two methods were used to measure the colour of the fish oil during experiments. 
1. The method described by Fernandez (1986) was used to measure the absorbance of fish oil using 
a Sbimadzu Spectrophotometer Model UV-110-02 at 490 nm. The results were corrected using 
a petroleum ether as blank. Samples were equilibrated to 30?C in a water bath prior to 
measurement 
32 
2. The Photometric Method based on AOCS Official Method Cc 13c-50 (AOCS, 1973) was used 
only for the fish oil from the sterilisation experiment. Sample absorbance was measured at 460, 
550, 620 and 670 nm using the same spectrophotometer as before. The photometric colour was 
calculated using the equation: 
Photometric Colour = 1.29A460 + 69.7 A550 + 41.2A,s20 - 56.4AoJ0 
where, A = absorbance. 
b. Fish flesh and tomato sauce colour 
Hunter colour readings (L, a, b values) were determined using a Hunterlab Colourquest 
Spectrophotometer Model CQ1200k with version 2.33 software and calibrated with a white tile 
standard. For fish flesh, the sample (3.25 g) was placed in a metal container. Tomato sauce (22.5 
ml) was placed in a glass cuvette. The dimensions of container used for fish flesh and a glass 
cuvette for tomato sauce are shown in Appendix 3.2. 
3.3.2.2. Refractive Index 
Refractive Index (RI) values were determined using Bellingham Stanley Refractometer (Bellingham 
+ Stanley Limited, England) at 20?C for samples from the refming experiment (AOAC, 1984), and 
at 25?C for samples from the storage experiment (Arya, 1969). The refractometer RI range was 
1.30 - 1.74. 
3.3.3. Sensory Analysis 
Sensory analysis was conducted on fish oil, tomato sauce and canned fish. Panellists were chosen 
from among Indonesians residing in Palmerston North, because of their familiarity with the fish 
taste and fish odour. 
33  
Panellists were trained prior to participation in sensory tests. They were experienced in detecting 
fishy odour and taste levels for the refming experiment, and some rancid odour and taste levels for 
the fish oil storage experiment. During panel selection, their interests, motivation, availability, and 
attitude towards the products were considered. During evaluation, more attention was paid to 
clarify confusion, and to monitor panellists. 
The product attributes, evaluation, and sensory sheets used, will be discussed for each experiment 
in the appropriate chapters. Defmitions of each attribute listed in the sensory sheets for each 
experiment are described as follows:. 
A. For fish oil refining experiment: 
1. Odour : the overall strength of fishy and undesirable odour detected in the sample. 
2. Taste : the overall strength of fishy and undesirable taste detected in the sample. 
B. For Fish Oil Storage Experiment: 
1. Odour : the overall strength of rancid odour detected in the sample. 
2. Taste : the overall strength of rancid taste detected in the sample. 
C. For Fish Oil Sterilisation Experiment: 
1. Fishy Odour : the overall strength of fishy odour detected in the sample. 
2. Fishy Taste : the overall strength of fishy taste detected in the sample. 
3. Rancid Odour: the overall strength of rancid odour detected in the sample. 
4. Rancid Taste : the overall strength of rancid taste detected in the sample. 
D. For Tomato Sauce Formulation Experiment: 
1. Consistency : the overall impression of consistency when the product was stirred. 
2. Odour : the overall impression of odour detected in the sample. 
3. Colour : the overall impression of visual colour detected in the sample. 
34 
4. Mouth Feel : the overall impression of the sauce texture when touching all parts of the 
mouth, especially in terms of the effect of the presence of fish oil. 
5. Overall Acceptability: evaluation of the overall acceptability of the sample, and rating 
acceptability score for the sample. 
E. For Canned Fish Experiment: 
I. Fish 
1 .  Fish Surface Appearance: the smoothness, or damage to the fish surface in the canning 
process. 
2. Flesh Texture: the overall tenderness of the sample, determined by the degree of force 
required to compress the sample between the molar teeth. 
3. Softness of Bone: the overall softness of bone, determined by the degree of force 
required to compress the sample between the molar teeth. 
4. Sourness : the overall strength of sourness taste from tomato paste, and vinegar, as 
detected in the fish flesh. 
5. Saltiness : the overall strength of saltiness detected in the fish flesh. 
6. Overall Spiciness: the overall strength of garlic, and shallot taste in the fish flesh. 
7. Fishiness : the overall strength of the fish taste detected in the fish flesh. 
IT. Tomato Sauce 
1. Colour : the intensity of red colour detected in the tomato sauce. 
2. Mouthfeel : the oily impression of the sample in the mouth. 
3. Sourness : the overall strength of sourness from tomato paste, and vinegar detected 
in the tomato sauce 
4. Saltiness : the overall strength of saltiness detected in the tomato sauce. 
5. Overall Spiciness: the overall strength of garlic, and shallot taste detected in the tomato 
sauce. 
6. Fishiness : the overall strength of the fish taste of fish oil detected in the tomato 
sauce. 
35 
lll. Overall Acceptability: evaluation of the overall texture, and odour of fish and tomato sauce, 
and rating acceptability for the canned fish product. 
3.3.4. Canned fiSh analysis 
3.3.4.3. Fo determination 
Fo was used to measure total lethality of the thermal process. A wire thermocouple was mounted 
at the geometric centre of the can and injected into one of the fish pieces. The hole in the can was 
sealed with epoxy resin and hardened. The can was then filled with tomato sauce and seamed. 
During seaming, hot steam was passed over the surface of the product to facilitate the vacuum 
condition in the can. The head space in the can was about 1cm. The Fo determination was carried 
out in a pilot plant scale retort (90 cm long and 56 cm diameter). The thermocouple was connected 
to a Molytek Data Reader and computer, supported with a locally made Fo Calculation Program 
(Van Til, 1991). The Fo for the heating and cooling process was automatically calculated and 
displayed on the screen. The Fo determination was conducted at a retort temperature of 121.1?C. 
3.3.4.2. Sterility test 
Total plate count (TPC) analysis was used to determine the sterility of the canned fish. The 
analysis was carried by spreading 0.1 ml of diluted sample (10.1) on the surface of nutrient agar 
plates. The plates were incubated aerobically and anaerobically at 30 and 55?C, respectively. 
The dilution was prepared by weighing a portion of the fish flesh (25 g) and placing the portion 
in 225 ml 0.1% peptone water. The sample was obtained from two canned fish products. 
3.3.4.3. Incubation test 
The canned fish products were incubated at 30 and 55?C (5 cans at each temperature) for 14 days. 
Every seven days the cans were observed for swelling and one was opened for detection of 
undesirable odour. 
36 
3.4. DATA ANALYSIS 
Analysis of variance, t-test and crosstab were performed using Stat-Packets 1 .0 (Wallonick 
Associates, 1987). Regression analyses was carried out using both Stat-Packets 1 .0 (Wallonick 
Associates, 1987) and MUT AB/PC-Version 3.01 (Boag, 1988). The Plackett and Burman Program 
(Van Til, 1991)  was used to identify variables showing significant effects on the experimental 
canning process. 
37 
Chapter 4 
FISH OIL PRODUCTION IN INDONESIA 
4.1. BACKGROUND 
Currently, fish oil is produced as a by-product of the fish meal industry. Fish oil is produced in 
floating meal factories, floating canneries or inland fish meal factories (Tanikawa, 1971). The 
technology involved is not very complex, but some technological aspects must be seriously 
considered in order to obtain a good quality fish oil. 
According to the Food and Agriculture Organisation of the United Nations (FAO, 1986), raw 
materials used for fish meal and oil processing fall into several categories: 
* fish caught specifically for processing into meal, and oils such as menhaden, anchovy 
and sardine; 
* incidental or by-catch from another fishery, for example shrimp trawl by-catch; and 
* fish offal or waste from edible fisheries such as cutting from fll.leting operations, fish 
cannery waste, roe fishery waste and surimi processing waste. 
Several different processes have been evolved for the manufacture of meal and oil including wet 
and dry rendering, and solvent, enzymatic, and silage processes (Lee, 1%3; Pigott, 1967; Considine 
and Considine, 1982; Bimbo, 1990). These processes have been reviewed previously in Chapter 
2. 
The quality of crude fish oil is dependent on the handling and the storage of raw fish prior to 
processing, on the type and operational efficiency of the fish processing plants, and on the handling 
and storage of the crude oil prior to refining (Young, 1982). 
As information about Indonesian methods of production and quality of the product were 
unavailable, a survey and analysis of Indonesian fish oil samples were conducted. 
38 
4.2. OBJECTIVES 
The survey of Indonesian fish meal factories was aimed at determining the present status of fish 
meal and fish oil in the fishing industry. The survey was also designed to obtain technical and 
financial information about present production. 
Samples of unrefmed and refmed fish oil were collected from the producers and analyzed 
chemically, physically and organoleptically in order to determine quality. This information was 
used to decide whether Indonesian fish oil required further refinement or not, for human 
consumption. 
4.3. METHODOLOGY 
Survey of the Indonesian fish meal factories was limited to factories located around the Bali Straits, 
where most Indonesian fish meal and fish canning factories are concentrated. Since official 
information on the number of fish meal factories was unavailable, the survey was conducted of all 
fish meal processors under the guidance of local government fishery officers. Eighteen 
conventional fish meal producers and one traditional fish meal processor participated. A 
conventional fish meal producer is defmed as one using a standard, modem fish meal processing 
unit. Labour costs here are relatively low. In comparison, the traditional fish meal producer is one 
using very simple processing equipment, where high labour costs are significant. Some factories 
refused to be interviewed. The survey was carried out by direct interview of factory managers and 
staff. A copy of the questionnaire used during the survey is shown in Appendix 4.1 .  
Fish oil samples, collected during the survey, were held in polypropylene bottles with 200 ppm 
BHT antioxidant. Samples were airfreighted to Massey University for free fatty acid value, 
refractive index (20?C), colour (absorbance at 490nm) and visual colour, sensory analysis of odour, 
and for fatty acid proflling. Visual colour description was given by a trained PhD student in the 
Consumer Technology Department, Massey University. Sensory analysis on odour was undertaken 
by seven trained Indonesian students. The panellists were asked to assess the intensity of fishy and 
undesirable odour using a nine point scale (1 = non-fishy/undesirable odour, 9 = extremely strong 
fishy/undesirable odour). The score sheet used to evaluate the odour of the oil is shown in 
Appendix 4.2. 
39 
New Zealand fish oils, crude (mainly hold) and orange roughy oils, were used for comparative 
purposes. Both oils were processed using the wet rendering method. 
4.4. RESULTS 
4.4.1. Position of fish meal in Indonesian fishery industry 
Of the 19 Indonesian fish meal processors surveyed, 6 factories produced fish meal as the only 
main product, while 9 factories produced 2 main products, fish meal and canned fish. Four 
factories stated that the fish meal was a by-product. 
4.4.2. Raw fiSh used for fish meal/oil production in Indonesia 
Raw fish used to produce fish meal in the surveyed factories is shown Table 4.1 .  
Table 4.1.  Raw fish used for fish meal production (as number of factories) 
Product Status 
Raw Fish 
Fish Species Fish Meal as Fish Meal as 
Only Main One of Main 
Product Products 
Whole Sardine 6 9 
Fish 
Mackerel 1 
Mixture of fish species 1 4 
Fish Sardine 4 
Waste 
Tuna 
Note: A tacto m1 !ht use more than one raw matenal ry g type 
Fish Meal as 
By-Product 
3 
2 
40 
Generally, fish meal processors utilised whole oil sardine (Sardinella longiceps) and sardine waste 
from canneries as raw material. Other fish species used in fish meal production were mackerel and 
tuna. Where a mixture of fish species is reported these consisted primarily of sardine, scad, 
mackerel and red snapper. Only factories producing fish meal as one of two main products used 
both whole fish and fish waste as raw materials. Fish waste from canneries consisted of heads, 
tails and offal. Fish meal produced as a by-product was processed from sardine and tuna waste 
only. 
4.4.3. Fish meal and fish oil processing in Indonesia 
Three commercial processing methods were encountered during the survey, as shown in Table 4.2. 
The wet rendering process was used by 17 processors surveyed. One processor used the dry 
rendering process and one used a cooking process without pressing. In this latter process, the fish 
was boiled and then air dried, without pressing between these processing steps. 
All fish meal factories using the wet and dry rendering processes were able to separate fish oil 
from the liquid phase freed during pressing. The factory using the cooking process without 
pressing was separating fish oil from water used for cooking. In this process, the water and oil 
were allowed to separate for sometime and then the water fraction was removed. Production of 
fish oil from all types of raw materials was conducted by all factories except in 1 case where oil 
was not separated from tuna offal. 
41  
Table 4.2. Fish oil production information obtained during the survey 
Number of Factories 
a. Fish meal production method: 
1 .  Wet rendering process 17 
2. Dry rendering process 1 
3. Cooking process without pressing 1 
Total 19 
b.  Oil Separation after pressing/cooking 
1 .  Factories separating the oil 19 
2 .  Factories not separating the oil 0 
Total 19 
c. Relationship between oil separation and 
raw material: 
1 .  Separation conducted to all raw 18  
materials 
2. Separation not conducted to all raw 1 
materials Total 19 
d.  Refmement application to fish oil: 
1 .  Factories refining the oil 3 
2. Factories not refming the oil 16 
Total 19 
The alkali refining process was identified in three surveyed factories, representing each of the three 
factory types, as shown in Table 4.1.  
4.4.4. Prices and buyers of fJSb oil 
Unfortunately, some factories surveyed did not answer the questions relating to product prices, and 
buyers of their fish oil. Therefore, the results presented in Table 4.3 are limited to those factories 
which provided requested information in sufficient detail. 
42 
Table 4.3. Price and buyers of fish oil (by number of factories) 
Status of fish meal production 
The Only One Of main By-product Total 
Main products 
Product 
a. Price ?J(Rp./1): 
100 - 200 2 1 3 
201 - 300 1 9 2 12  
301 - 400 1 1 2 
401 - 500 1 1 
b Buyers: 
1 .  Fish oil 3 3 2 8 
traders 
2. Feed companies 2 5 1 8 
'1/ote: *) A tactory coUl<l gtve more man one pnce 
The price of fish oil varied from Rp.100.-/l to Rp.500.-/l. In the main, fish oil was sold at a price 
between Rp.201 .-/l-Rp.300.-/l. One factory sold fish oil according to its free fatty acid (FFA) 
value: Rp.425.-/l for fish oil having FFA value less than 5% and Rp.300.-/l for fish oil having 
FFA value more than 7%. Another factory reported that the price of canning waste oil was higher 
than the price of oil obtained from fish meal processing. The types of fish meal produced did not 
show a correlation with the fish oil price. 
Results showed that fish oil buyers were mainly local fish oil traders and animal feed companies. 
4.4.5. Chemical, physical and sensory analysis of Indonesian fish oil 
Results of chemical, physical and sensory analysis of fish oil collected from factories/processors 
during the survey is presented in Table 4.4. As a requirement of the survey conducted by an 
Indonesian Fishery Officer, the names of the factories participating in the survey need to be kept 
confidential. Thus the alphabetic codes used below do not relate in the same sequential order of 
the participating factories identified in Appendix 4.3. In general, the results showed that Indonesian 
43 
fish oil varied chemically, physically and organoleptically. All fish oil samples from either fish 
meal processing or canning operations were obtained from oil sardine and in one case, from tuna 
waste. The oil separation from fish meal production has been described previously in Section 
4.4.3. Factories having more than one main product did not separate the oil processed from the 
whole fish and fish waste. Fish oil obtained from the fish canning operation was collected during 
the pre-cooking operation. 
Factories A and E produced fish oil from fish meal processing with extremely low FF A content, 
0.08 and 0.09% respectively. However fish oil produced in other factories showed FFA value 
ranging from 1 .70 - 18.77%. The Refractive Index (RI) also varied from 1 .4767-1 .4785. 
Absorbance values of colour showed a relatively high value in the range, 1 .45-2.55. This 
corresponded to a colour change ranging from yellowish orange to brownish black. Panellists 
observed that the odour score was in the range, 4.57-7.86, indicating that a fishy or undesirable 
odour was detected. 
Factories processing fish meal as the main product, together with other main products, produced 
fish oil from canning waste during the pre-cooking stage and from the pressing stage in fish meal 
processing. Generally, the fish oil collected from canning waste had a lower FFA value than the 
oil separated from fish meal production, except the canning waste oil produced by Factory M had 
a relatively high FFA value, but the value was still lower than the FFA values analyzed in the oil 
from fish meal processing. The RI values of the oils also varied and the source of the oil did not 
give any specific indication on RI value. The colour performance and absorbance value of the oils 
were significantly influenced by the source of the oils. The canning waste fish oil had a much 
lower absorbance value than the oil from fish meal processing. The canning waste oil showed a 
yellow colour while the colour of the oils from the fish meal operation were darker, ranging from 
yellowish brown to very dark reddish brown. According to panellists, the fish oil odours were also 
correlated to the source of the oils. A low odour score was obtained for canning waste oil. The 
oil obtained from fish meal processing generally had a high score reflecting a relatively high degree 
of off-odour development. Refined oil from fish meal produced by Factory L demonstrated an 
improved quality. The quality classification given by factory H did not exactly reflect the quality 
of the oil produced in terms of FFA value. 
Table 4.4. Chemical, physical and sensory analysis of Indonesian fish oil 
Product type of fish 
meal/factories 
Main Product: 
A 
B 
C (trad.) 
D 
E 
F 
One of Main 
Products: 
G 
H 
H 
H 
H 
I 
I 
J 
K 
L 
M 
M 
M 
By-Product: 
N 
0 
0 
p 
Q 
- -? .... ..... .. 
** FM = Fish meal 
Oil 
Sources 
s?JFM ?? 
SIFM 
SIFM 
SIFM 
SIFM 
SIFM 
s1cw??? 
SIFM 
SIFM 
SIFM 
S/CW 
SIFM 
S/CW 
SIFM 
SIFM 
SIFM 
SIFM 
SIFM 
S/CW 
SIFWM+ 
S/CW 
SIFWM 
r?JFWM 
SIFWM 
-----------------
Fish oil quality FFA Refractive Index 
(% oleic (20"C) 
acid) 
- 0.08 1 .4785 
- 5.34 1.4780 
- 5.42 1.4780 
- 18.77 1.4767 
- 0.09 1.4800 
- 8.78 1 .4772 
- 0. 15 1.4790 
I 4. 18  1 .4789 
I! 21 .76 1 .4761 
Ill 10.61 1.477 1 
- 0.56 1.4786 
- 8.47 1 .4775 
- 0.06 1 .4784 
- 4.49 1.4785 
- 55.69 1 .4750 
Refined 0. 15 1.4790 
A 0.24 1.4800 
B 14.06 1.4770 
- 1 . 15 1.4780 
- 25.72 1.4760 
I 0.23 1.4785 
I! 19.58 1.4760 
- 15.2 1  1.4800 
- 6.99 1 .4780 
,., ........ ... . ....... .......... .... ' '  . ......... 
+ FWM = Fish waste meal 
------------???-?-
Colour 
(absorbance 
at 490 nm) 
1.45 
1 .81  
2.34 
2.39 
2.55 
2.43 
0.22 
1.70 
.2.55 
2.5 1 
0.34 
2.56 
0.48 
2.52 
2.52 
0.% 
2. 10 
2.45 
0.34 
1 .34 
0.27 
2.43 
1.85 
2.29 
-
---
Visual colour 
yellowish orange 
brown 
dark reddish brown 
very dark reddish brown 
brownish black 
dark brown 
yellow 
reddish brown 
dark brown 
brownish black 
yellow 
very dark reddish brown 
yellow 
greenish brown 
brownish black 
yellowish brown 
reddish brown 
brownish black 
yellow 
reddish brown 
yellow 
dark brown 
dark reddish brown 
reddish brown 
---------------
Odour 
score 
7.86 
7 14  
4.57 
7.57 
7.64 
7.00 
2.86 
6.29 
8.50 
7.29 
3.86 
7. 14 
3.71 
8.79 
8.10 
3.64 
5.64 
7.21 
2.43 
4.29 
2.86 
7.86 
5.71 
7.64 
11:> 
11:> 
45 
Producers who considered fish meal as a by-product, used fish cannery waste as the raw material. 
Only one factory produced fish oil collected from the pre-cooking stage in canning operation. High 
FFA values, 6.99-25.72%, for fish oils separated from fish waste meru, and a low FFA value, 
0.23%, for canning waste oil were noted. The RI values varied from 1 .4760 to 1 .4785. All fish 
oils obtained from fish waste meal processing tended to have a brown colour with absorbance value 
ranging from 1 .34 to 2.43; while the canning waste oil was yellow with absorbance value of 0.27. 
Panellists gave a relatively high odour score, 4.29 - 7 .86, to fish oil from fish waste meal 
production. In contrast, the canning waste oil was given a low odour score, 2.86. 
4.4.6. Fatty acid profiles of fJSh oil 
Fatty acid profiles of fish oils collected from the fish meal factories during the survey are shown 
in Table 4.5. 
Fish oil from the factories processing fish meal as the only main product produced from the whole 
fish, primarily sardine, bad 36.2-40.8% saturated fatty acids (SAFA), 28.4-33.8% monounsaturated 
fatty acids (MUFA), and 28.6-34.1%  polyunsaturated fatty acids (PUPA). The relative amounts 
of omega-3 fatty acids, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) were 24.1 -
29.5%, 9.2-20.1%  and 3.5-12.5% respectively. The single traditional processor also produced fish 
oil having a high omega-3 fatty acid content, 25.6%. 
Canning waste oils and fish oils from fish meal operation produced by factories processing fish 
meal as the main product, together with other main products, did not show any pattern in fatty acid 
profiles reflecting the effects of fish oil source. Relative amounts of SAFA, MUPA and PUPA of 
canning waste oils were 38.4-40.0%, 28.8-32.1% and 27.9-31 .6% respectively. The fish oils from 
fish meal processing had 37.5-41 .2% SAFA, 28.4-35. 1%  MUFA, and 24.9-32.8% PUPA. The 
omega-3 fatty acids, EPA and DHA of canning waste oil were 23.7-27.2%, 15 .4-17.6% and 4.9-6% 
respectively. The relative amounts of these acids in the oil from fish meal production were 1 8.5-
28.0%, 20.6-28.2%, 1 1 .5-18.6% and 5-12.2% respectively. 
Table 4.5. Fatty acid profiles of Indonesian fish oils (%) 
Product Type 
of Fish Meal! Oil Quality SAFA MUFA 
Factory Source 
Main 
Product: 
A SIFM - 30.7 29.1 
B S/FM - 40.5 28.4 
C (trad) S/FM - 40.8 29.1 
D S/FM - 38.8 31 .3 
E SIFM - 40.6 30.8 
F S/FM - 38.2 33.2 
One Of Main 
Products: 
G S/CW - 39.6 28.8 
H S/FM I 41.1 28.4 
H S/FM IT 41.2 28.9 
H S/FM m 41.1  32.7 
H S/CW - 38.4 30.2 
I S/FM - 38.9 29.1 
I S/CW - 39.6 29.1  
J SIFM - 37.8 3 1 .2 
K SIFM - 40.0 36.1 
L S/FM ref.* 39.6 29.4 
M S/FM A 37.5 29.6 
M S/FM B 38.3 30.0 
M S/CW - 40.0 32.1 
By-Product 
N SIFWM - 39.1  30.0 
0 S/CW I 39.8 28.5 
0 SIFWM IT 41.3 28.8 
p TIFWM - 38.6 27.4 
Q SIFWM - 38.5 30.6 
llote : t?at ty acids : :SAt'A - saturated at ty acids 
MUF A = monounsaturated fatty acids 
PUFA = polyunsaturated fatty acids 
ID-3 FA = omega-3 fatty acids 
EP A = eicosapentaenoic acid 
DHA = docosahexaenoic acid 
Oils *) ref. = Refined oil 
Raw materials: 
S = Sardine 
FM = Fish meal 
CW = Canning waste 
FWM = Fish waste meal 
T = Tuna 
46 
PUFA ro3FA EPA DHA 
34.1 29.5 20.1 5 .8 
3 1 .1 26.8 17.1 6.1 
30.1 25.8 17.1 5.0 
30.0 25.7 16.4 5.8 
28.6 24.1 9.2 12.5 
28.7 24.1 17.4 3.5 
3 1 .6 27.2 17.6 5 .7 
30.5 26.0 16.0 6.2 
30.0 25.5 15.2 6.4 
26.3 22.2 13.9 5.0 
3 1 .3 26.7 16.7 6.0 
32.0 27.6 18 .6 5 .3 
3 1 .3 26.8 17.2 5.8 
3 1 .0 26.3 17.4 5.2 
24.8 20.6 1 1 .5 5.3 
3 1 .0 26.4 16.6 6.1 
32.8 28.2 12.9 12.2 
3 1 .8 27.1 16 .1  7.4 
28.0 23.7 15.4 4.9 
30.9 26.5 17.2 5.8 
3 1 .8 27.4 17.6 6.0 
20.0 25.1  15.2 6.4 
34.0 29.5 5.0 22.2 
31 .0 26.1 16.0 6.0 
47 
In terms of fish species used to extract fish oil, two kinds of fish oils, sardine and tuna, were 
obtained from fish meal processors who considered fish meal as a by-product. Tuna oil contained 
38.6% SAFA, 27.4% MUFA and 34.0% PUFA; while the sardine oil contained 38.5-41.3% SAFA, 
28.5-30.6% MUFA and 30.0-31 .8% PUFA. The relative quantities of omega-3 fatty acids, EPA 
and DHA in tuna oil were 29.5%, 5.0% and 22.2%, while the quantities of those acids in sardine 
oil were 26.2-27.4%, 15 .2-17.6% and 5.8-6.4% respectively. 
The quality classification made by the above fish meal factories showed a relationship with the 
relative quantity of omega-3 fatty acids. The better the quality, the higher the omega-3 fatty acid 
content. 
4.4.7. New Zealand fish oil used as comparison with Indonesian fish oil 
Results of chemical, physical, sensory and fatty acid profiles of two New Zealand fish oils are 
shown in Table 4.6. 
Table 4.6. Chemical, physical, sensory and fatty acid proftles of New Zealand fish oils 
Parameters Crude oil Orange roughy oil 
fFA (% oleic acid) 0.67 0.40 
RI (20"C) 1 .4730 1 .4652 
Colour (absorbance 0.90 1 .96 
at 490 nm) 
Visual colour yellowish orange reddish brown 
Odour score 6.38 4.06 
Fatty acid 
proftles (%): 
SAFA 25.3 6.0 
MUFA 60.3 87.7 
PUFA 14.4 6.2 
HUFA 10.1 2.7 
Omega-3 FA 11 .6 3.5 
EPA 4.2 1.3 
DHA 5.1 1 .2 
48 
Both crude, mainly boki, and orange rougby oils bad lower FFA values compared to most 
Indonesian fish oils, but those values were higher than the FFA values of canning waste oils. 
Further, the RI values of both New Zealand oils were lower than those of Indonesian oils. The 
colour absorbance values of New Zealand oils were in the range of absorbance values analyzed in 
Indonesian fish oils. The odour score of crude oils was relatively high. On the other band, the 
orange roughy oil bad a odour score considered low. 
In terms of fatty acid profiles, Indonesian and New Zealand oils showed differences. Indonesian 
oils were significantly richer in polyunsaturated fatty acid content than New Zealand oils, 
particularly in omega-3 fatty acids content. The New Zealand oils bad a higher content of MUF A. 
However, orange rougby oil is not eatable, since the oil is predominantly wax esters (Buisson, et 
ill. 1982) 
4.5. DISCUSSION 
4.5.1. Fish oil production 
The survey of Indonesian fish meal producers and samples collection from Indonesian fish oil 
producers indicated that the fish oil could be obtained from fish meal processing and fish canning. 
Although fish meal processing factories were found to use various processing technologies, 
including wet rendering, dry rendering and cooking without pressing, most processors favoured the 
wet rendering method. It is noted that all processing methods use beat treatment as an integral part 
of fish meal production. In this case, heating or cooking is used to coagulate or denaturate fish 
protein to facilitate mechanical separation of liquids from solids. Under these conditions, fat cells 
are also ruptured, releasing the oil into the liquid phase (Bimbo, 1990; Kinsella, 1987; llyas et al, 
1985). The efficient liberation of water and oil by cooking and pressing is an important aspect in 
producing high quality fish meal (Beraquet et ill, 1984). In both wet and dry rendering methods, 
the liquid phase was released during the pressing step. In the cooking method without pressing, 
the liquids are released into cooking water from which fish oil can be subsequently separated. 
Fish oil is also found produced from pre-cooking steps during the canning operation. Pre-cooking 
49 
is normally done by steaming the fish for approximately 20 minutes. Indeed, one of the purposes 
of the pre-cooking step during fish canning is to release body Iipids if the fish are excessively oily 
or if the oil has a very strong flavour (Warne, 1988; Codex Alimentaiius Commission, 1976). It 
is known that when fish flesh is heated, a significant proportion of water is released from the 
protein. The amount varies, approximately 17.5% for tuna, 19-34% for sardines, depending on the 
endogenous fat content (Van Den Broek, 1965). 
4.5.2. Fish oil quality 
Indonesian fish oil quality varied by factory and by oil source. Even in one factory, the fish oil 
produced could vary in quality. Three factories surveyed have classified their fish oil in terms of 
quality, using different parameters: free fatty acid content, fish oil colour and fish oil source. One 
factory classified fish oil quality in terms of FFA content, in which Grade I oil bad a FFA value 
of less than 5%, Grade II oil, 5-7% and Grade Ill oil, more than 7%. This quality grade 
classification is similar to that reported by Windsor and Barlow (1981), who indicate that, usually, 
fish oil is sold on the basis of 3% FF A with a maximum allowable content, normally of 7%, which 
can, in certain cases, be up to 20%. On this basis, the Indonesian fish oil samples analysed in this 
study are classified as predominantly Grade I, 48%, but with some samples falling into Grade II, 
20%, and Grade Ill, 32%, as shown in Table 4.7. All canning waste oils were classified as Grade 
I. The two New Zealand oils were grouped as Grade 1 .  
Table 4.7. Classification of Indonesian Fish Oil Quality in Terms FFA Value 
Grade Factory 
Producer 
A 
E 
G 
H 
GRADE - I H 
(FFA value: I 
< 5 % J 
L 
M 
M 
0 
GRADE - II B 
(FFA value: C (Trad.) 
5 - 7 %) Q 
D 
F 
H 
H 
GRADE - III I 
(FFA value: K 
> 7 %  M 
N 
0 
p 
"'ote: ? - ?arc me 
FM = Fish meal 
CW = Canning waste 
FWM = Fish waste meal 
T = Tuna 
Fish Oil Factory FFA Value 
Source Grade (% oleic acid) 
S/FM - 0.08 
S/FM - 0.09 
S/CW - 0.15 
SIFM I 4.18  
S/CW - 0.56 
S/CW - 0.06 
SIFM - 4.49 
S/FM Refined 0.15 
S/FM A 0.24 
S/CW - 1 .15 
S/CW I 0.23 
S/FM - 5.34 
S/FM - 5.42 
SIFWM - 6.99 
SIFM - 18.77 
S/FM - 8.78 
S/FM II 21 .76 
S/FM Ill 10.61 
S/FM - 8.47 
SIFM - 55.69 
S/FM B 14.06 
SIFWM - 25.72 
SIFWM II 19.58 
T/FWM I 15.21 
50 
The FF A value of canning waste oils were generally lower than the value measured in the oil from 
fish meal production. It is worth re-emphasising that the source of FF A is from hydrolysis of 
triglycerides (Windsor and Barlow, 1981). In this regard, different methods of oil extraction are 
likely to play a major role in determining the fmal oil quality. A relatively short heating time of 
20 minutes at ? l00?C (Directorate General Of Fishery, 1984), and the use of the whole fish in 
the fish canning process are likely to contribute to minimal triglyceride hydrolysis. Fish oil 
produced from a fish meal processing involves more heating of the fish pulp. For example, in fish 
5 1  
meal processing, the fish pulp is cooked at 90?C for 30 minutes, and then dewatered. The liquid 
phase containing the oily emulsion is then reheated to 90?C prior to oil separation (Hoffmann, 
1989). In this process the opportunity for accelerated triglyceride hydrolysis is relatively high. The 
other major factor affecting fish oil quality is undoubtedly raw material quality. Low quality fish, 
normally processed into fish meal, may have also caused the fish oil produced to have a high FF A 
value, since, according to Young (1982), spoilage in fish is responsible for increased free fatty acid 
content by lipases/enzymatic activity. 
Canning waste oil consistently appeared lighter in colour than the oil recovered during fish meal 
processing. The difference in the colour is probably due to the method of oil extraction. It has 
been reported by Brody (1965) that excessive heat applied to fish for prolonged periods results 
in the production of darker oil. Chemically, these darker oils probably arise from heat-induced 
protein breakdown products acting as catalysts to accelerate autoxidation of the endogenous oil. 
This situation is exacerbated by the use of improperly cleaned containers contaminating the oil. 
For example, it is known that the oxides of metals such as iron, lead, and copper, when dissolved 
in oil containing water and free fatty acids, can accelerate the oxidation and darkening of fish oil 
(Brody, 1965). The raw material, oil sardine, may have affected the colour of the oil, the third 
grade oil sardine darkening the fish meal oil significantly more than the oil produced from canning 
waste, where the first grade oil sardine was used (Irianto and Fawzya, 1987). The raw material 
for fish production should be as fresh as possible in order to yield light coloured oil. Oil extracted 
from deteriorated raw material yields dark oil (Brody, 1965). In fact, the oil sardine used to 
produce fish meal is the sardine which does not meet required quality for processing into other 
products: canned fish, boiled salted-fish, and dried salted fish. 
The raw material quality used to produce these fish oils may have affected the differences in odour 
as observed by panellists. The sardine, which is normally used as the raw material in fish meal 
production, is bought from the boat without an insulated fish hold (Irianto and Fawzya, 1987), thus 
the fish receives improper handling to keep its freshness. The improper handling continues when 
the fish arrives in the factory by without refrigeration storage. For example, Total Volatile Base 
(TVB) values and Total Plate Counts (TPC) of sardine, normally processed into fish meal, were 
found to be 27.4-35.1mgN% and (60.5-88.5)103 respectively. In contrast sardine used by canneries 
usually have a TVB of 20.7-22.6mgN% and a TPC of (12.1-22.1)103 (lrianto and Fawzya, 1 987). 
Decomposition products which accumulate during fish spoilage are nitrogenous and sulphur 
compounds. Both compounds, if present in the fish at the time of processing, will pass over in part 
into the oil and occur in trace amounts sufficient to alter odour and flavour. The sulphur 
compounds possessing the more obnoxious odour cause more serious alterations in the quality of 
52 
the oil, but in well handled fish their presence is minimal (Stansby, 1990). This indicates that the 
odour of canning waste oil was better than the odour of oil obtained from fish meal processing. 
According to the panellists, the odour of canning waste oils was more uniform than the oil from 
fish meal processing. This was possibly affected by the quality of the raw fish. The results also 
showed a tendency for the oil with a darker colour to have a high odour score. From the above 
results, it can be concluded that there was a close relationship between raw material quality and 
both fish oil odour and colour. 
Fish oil produced by Indonesian factories bad relatively higher quantities of PUPA and omega-3 
fatty acids than New Zealand crude oil from, mainly, boki. The Indonesian oil bad 20.6-29.5% 
of omega-3 fatty acids and New Zealand crude oil had 1 1 .6% of omega-3 fatty acids. Analysis 
of sardine oil conducted by Setiabudy (1990) indicated that this oil contained 25.2% omega-3 fatty 
acids, which falls into the range found in this study for Indonesian oils. The level of omega-3 fatty 
acids in fish is known to be a function of fish species, age, sex and season (Moeljanto, 1982). The 
relatively high levels of omega-3 fatty acids in Indonesian fish oils suggests that the industry should 
seriously evaluate oil usage as a high quality product suitable for human consumption, rather than 
for non-human food purposes, such as animal feeds, as the survey results reveal. 
To this end, fish oil quality should be improved chemically, physically and organoleptically by 
application of a suitable and properly evaluated refining process. As the fish oil was a by-product 
from fish canning and fish meal processing, most processors did not give, proper attention to fish 
oil quality improvement Thus, the refining method to be introduced to Indonesian fish oil 
producers should be simple, low cost and labour efficient. 
4.6. CONCLUSIONS 
Fish oils in Indonesia were found to be produced from two fish processes: canning and fish meal 
processing. Chemically, physically and organoleptically, the canning waste oils were of a higher 
quality than oil recovered during fish meal processing. The levels of omega-3 fatty acids found 
in Indonesian oils is significantly higher than measured in New Zealand crude oil. However, it  was 
found that for both Indonesian and New Zealand oils to be of acceptable quality for human 
consumption there was a need for an ultimate refming step. 
Chapter 5 
OPTIMIZATION OF THE RESIN REFINING PROCESS 
OF FISH Oll..  
5.1. BACKGROUND 
53 
The commercial process of fish oil refming, such as degumming, alkali refinement, 
bleaching and deodorization, involves heat. Since fish oil is the most polyunsaturated of all 
the oils, application of these operations would be detrimental to fish oils, due, mainly, to the 
susceptibility of polyunsaturated fatty acids to heat and consequent oxidation (Banks, 1 967). 
This would probably affect the fatty acid quality of fish oil, especially ro-3 fatty acids, 
because the ro-3 fatty acids are susceptible to selective destruction during the process 
(Griggs, 1986). 
Femandez (1986) introduced the use of macroporous resin for fish oil refming, a method 
without heat involvement, based on the fact that resins are sorbent materials. The most 
known sorption operation where resins are involved is ion exchange. An ion exchanger is 
basically an electrolyte solution containing cations, anions and water, differing, however, in 
that one or other ion is bound to an insoluble microporous matrix (Patterson, 1970). The 
typical reaction occurring in the ion exchange reaction is as follows: 
AR + B<?q) ------> BR + ?aq) 
B ions, initially distributed throughout the solution, must first find their way to the surface 
of the resin grains, because A ions can not leave the resin until B ions enter, as 
electroneutrality must be preserved. This frrst transport step takes place partly by the flow 
of solution, and partly by diffusion of the ions in solution. Next, when some B ions wander 
into the pores of the resin, an equivalent number of A ions can wander out, and so the 
exchange proceeds by diffusion throughout the interior of the resin until, eventually, an 
equilibrium distribution of A and B ions is reached. Meanwhile A ions entering the solution 
are also being distributed through the whole volume by diffusion and mixing (Kitchener, 
1957). 
54 
During the ion-exchange reaction, the ion-exchange resins are concentrated into soluble 
acids, bases, or salts. Cation-exchange resins contain fixed electronegative charges which 
interact with mobile counter-ions having the opposite, or positive charge. Anion-exchange 
resins have fixed electropositive charges and exchange negatively charged anions 
(Considine, 1974). 
Early ion exchangers were zeolites consisting of aluminium silicates (Samuelson, 1953). 
However most ion exchange resins used by industry today are manufactured from uniform 
spheres of styrene-divinylbenzene (DVB) copolymers having diameters 0.3 - 1 .0 mm (20 -
50 mesh). The copolymer beads are formed by pearl polymerization and converted to ion?
exchange resins by a second processing step. Sulfonic-type cation-exchange resins are made 
by sulfonation of the polymer beads at elevated temperatures. Strong base anion-exchange 
resins are produced by means of chloromethylation and amination of the copolymer spheres 
(Considine, 1974). 
Overall exchange rate may be influenced by a change in solvent nature and content, particle 
size, temperature, and the functional group. Exchange rates are most rapid in water system, 
and become increasingly slower with less polar solvents. This is true because the solutes 
are more highly ionized in the more polar solvents as are the ion exchange resins. 
Similarly, exchange rates are more rapid in low cross-linked resins of the same inherent dry?
basis capacity because of their higher moisture content. The ion exchange process involves 
diffusion through the film of solution in close contact with the resins and diffusion within 
the resin particle. Film diffusion is rate-controlling at low concentrations and particle 
diffusion is rate-controlling at high concentrations. Whether film diffusion or particle 
diffusion is the rate-controlling mechanism, the particle size of the resin is also a 
determining factor. Elevated temperatures increase exchange rates. The kind of functional 
group and its degree of dissociation under a given set of conditions greatly affects exchange 
rates where particle diffusion is controlling, but has no effect on film diffusion. 
Furthermore, a high degree of substitution of the functional groups on the inert polymer 
matrix directly enhances overall reaction rates (Wheaton and Lefevre, 1981).  
Ion exchange between strong electrolytes can usually be carried out until most of the 
stoichiometric capacity of the exchanger has been used. Consequently, the total sorbent 
capacity is practically constant regardless of the composition of the solution being treated. 
An apparent exception arises if a weak acid or base is involved, either in the resin or in 
solution (or in both), when the apparent capacity of the resin may be much less than its 
55 
stoichiometric value (Pery et .&, 1973). 
Another operation where resins or resin-like sorbents are involved are adsorption, molecular 
sieving, and gel permeation (Perry et .&, 1973) 
Fernandez (1986) found that the macroporous resin process, which requires no application 
of temperature over 65?C during refinement, was the only technology in her experiment 
which improved fish oil sensory characteristics and maintained fish oil chemical 
characteristics. The technology has proved to be superior over molecular distillation and 
freezing fractionation. However Fernandez (1986) did not optimize the process as carried 
out in this experiment. 
5.2. OBJECTIVES 
The experiment was intended to optimize the refining process of fish oil using a 
macroporous resin packed column in order to obtain good quality fish oil for human 
consumption. The experiment also aimed to reveal changes in fish oil, before and after 
refining. 
5.3. METHODOLOGY 
5.3.1.  Materials 
Fish oils used for the investigation were New Zealand crude, mainly hold, and orange 
roughy oils. 
5.3.2. Experimental Methods 
Four aspects were investigated to determine the possible ways to optimize the resin refining 
process and fish oil quality: 
* fish oil-resin volume ratio; 
* multiple refining; 
* use of vacuum pressure during refining; and 
* height and diameter ratio of column. 
5.3.2.1. Experiment 1: Effect of fish oil ? resin volume ratio 
56 
A column 1 .15cm in diameter and 53cm long was packed with 55cc of resin. The fish oil 
volume passed through depended on the fish oil-resin volume ratio tested, at 1 ,  2, 3 and 4 
times resin volume. The column was undisturbed until all freely flowing oil passed through 
the column. The oil fraction obtained by oil free flowing from the column was called 
fraction-1 or refmed oil. Fraction-2 was defined as the amount of oil retained in the column 
and flushed out with one equivalent volume packed column (55cc) of petroleum ether. The 
column was washed with methanol (55cc) in order to clean any polar compounds that might 
be attached to the resin. Then, petroleum ether (55cc) was used to flush the packed column. 
In order to ensure that the column was completely clean, 27.5cc methanol and 27.5 cc 
petroleum ether were passed through the column in sequence. The solvents were evaporated 
using a rotary vacuum evaporator at 65?C to obtain fraction-2 oils. 
The flow rate of fish oil passing through the column was 0.42-0.65ml/min. depending on 
the fish oil type. The flow rate decreased with the increase of fish oil volume. Refining 
was conducted at ambient temperature 18-23?C. 
5.3.2.2. Experiment ll: Effect of multiple refining on fiSh oil quality 
The procedure was the same as used in Experiment I, with the addition of multiple refining 
where the oil was passed through the clean column one, two, three and four times. The 
experiment was carried out at ambient temperature 18  - 23?C. 
57 
5.3.2.3. Experiment ID: Use of vacuum pressure in resin refining process 
A column of 1 .65cm in diameter and 39cm long was used. The column was cleaned using 
the method outlined in the experiment I, but modified. 40cc of petroleum ether was passed 
into the column. Then the column was washed with methanol, 40cc. The same volumes 
of petroleum ether and methanol were used to wash the column a second time. Finally, 
40cc of petroleum ether was passed through the column in order to prepare the column for 
reused. 
The fish oil-resin volume was at ratio 1 : 1 .  The vacuum pressure applied consisted of two 
steps: vacuum pressure at 74kpa was applied to guide the oil through the column; when 
the oil reached the end of column, the vacuum pressure was reduced to 60kpa. The time 
required for the first and the second pressure was 18-22 minutes and 5-7 minutes 
respectively. The time taken in the refining without pressure application was 80-91 minutes. 
The study was performed at ambient temperature 18 - 23?C. 
5.3.2.4. Experiment IV: Effect of column size on iJSh oil quality 
Two separate experiments were performed. The first experiment varied the height of the 
column with diameter size constant, the second experiment was performed by varying the 
column diameter size with height size constant. In the first experiment, the column diameter 
was 2.6cm. The diameter-height ratios investigated were 1 : 1 ,  1 :2, 1 :3  and 1 :4. In the 
second experiment, the column height was 39 cm and various column diameter sizes 
observed were 1 .65, 2.60 and 3 .20 cm. The fish oil-resin volume ratio used was 1 : 1. The 
washing method of the column was the same as in the experiment Ill. 
5.3.2.5. Experiment V: Investigation of natural antioxidant and volatile flavour 
compounds guantity changes during refining 
58 
The oils obtained from Experiment I, with fish oil-resin volume ratio of 1:1,  were used in 
this experiment. To investigate the changes in the quantity of natural antioxidant, the 
tocopherol level of unrefmed and refined oil was measured using HPLC. The unrefmed and 
refined oils were also identified for changes in volatile flavour compounds using gas 
chromatography/mass spectrometer. 
5.4. RESULTS 
5.4.1. Effects of fish oil-resin volume ratio on fish oil quality 
5.4.1.1 Effect on free fatty acid value 
Results clearly indicate that the fish oil volume passed through the resin column affected 
the free fatty acid (FF A) value of fraction-1 and fraction-2 oils of both crude and orange 
roughy as shown in Figure 5 .1 .  The FFA values for crude and orange roughy oils decreased 
significantly at the fish oil-resin volume ratio 1 : 1 .  Increasing the fish oil-resin volume ratio 
to 4:1  did not show any further significant reduction in FF A values for crude and orange 
roughy oils. 
., 
., ... 
lL 
3.0 
2.-4 
0:0 0 1 .8 < 
1 .2 
0.8 
Untruted 1 : 1 
on 
2 : 1 3 : 1 .( : 1 
Fish Oil Volume : Res in  Volume 
0 fraction I - cruda o i l  
f2l fraction 1 1  - crude o i l  
59 
E:;3 fraction I - orange roughy oil 
? fraction 11 - oranga roughy oil 
Figure 5 . 1 .  Effects of fish oil-resin volume ratio on free fatty acid value of fish oil 
A different behaviour of FFA content was observed in the fraction-2 of crude and orange 
roughy oils, where the FFA value of the oils tended to increase with increasing fish oil 
volume. The FFA value changes in fraction-! and fraction-2 of crude and orange roughy 
oils as a result of various fish oil volumes refined showed a similar pattern. 
5.4.1.2. Effects on refractive index 
Statistical analysis indicates that the fish oil volume passed through the resin column 
significantly influenced the refractive index (RI) values of both fraction- 1  and fraction-2 
obtained from crude and orange roughy oils. The changes in RI value of fraction-1 and 
fraction-2 for both oils during refining is shown in Figure 5 .2. 
1 .475 
)( 1 .470 Cl "C 
.E 0 
Cl ? ,. 0 1 .465 ?- N -
0 -
10 10 ... --
Cl 
er 1 .460 
1 .455 
? / ? ? ? ? ? ? V.? ?? ? /:? ?? ? %? ? ?? ;/:? ?? ?? /:? ?? V? 
Untreated 1 : 1 
Oil 
? / 
? ? 
? ? 
?<: ?g ?g ? g X: 
2 : 1 
? ? ? / 
? ? V ? 
? ? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? /:? 
3 : 1 4 : 1 
Fish Oil Volume : Resin Volume 
D fr?ction I - crude oil 
fZd fr?ction 11 - crude oil 
60 
&:8 friction I - orange roughy oil 
? fraction 11 - orange roughy oil 
Figure 5.2. Effects of fish oil-resin volume ratio on refractive index value of fish oil 
Resin refinement of the oils at ratio 1: 1 decreased the RI values of both. However as the 
fish oil volume was increased to 2, 3 and 4 times the resin volume, the RI value of the 
fraction- 1 oils increased accordingly. 
Fraction-2 for both oils obtained from refining at ratio 1 : 1  had a lower RI value than the 
raw oils. When the fish oil volume was increased to 2, 3 and 4 times that of the resin 
:voitm? , me f??tien? .?1s gbtained fr0m erude oi! showed a s1gp.ificant i,llcrease in RI 
value. However the fraction-2 o??ge roughy oil showed a si nificant reduction in the RI 
value. These sjgpificancies were shown using t-test at 95% significant level witb 4 degree 
of freedom. J 
5.4.1.3. Effects on colour absorbance 
Analysis was performed only for fraction-1 oils. Absorbance values of crude and orange 
roughy oils were significantly decreased with refining at fish oil-resin volume ratio 1 : 1 .  
However, increasing the fish oil volume to 2, 3 and 4 times that of the resin volume 
resulted in a gradual increase in absorbance of both crude and orange roughy oils as shown 
in Figure 5.3.  The increase was more pronounced in orange roughy oil. 
2 
'E 
r::: 
0 
? 1 .5 
... -
::I IQ 
..S! 
0 I) 
(.) g 
IQ 
..a 1 .0 ... 0 "' 
..a 
< 
Untrea ted  1 : 1 
Oil 
2 : 1 3 : 1 " : 1 
Fish Oi l  Volume : R esin Volume 
0 fraction I - crude oi l  
? fraction I - orange roughy oi l  
Figure 5.3. Effects of fish oil-resin volume ratio of colour absorbance value of fish oil 
5.4.1.4. Effects on fatty acid profiles 
61 
Fatty acid profile behaviour of crude and orange roughy oils from fraction-I and fraction-2 
before and after resin refining can be seen in Tables 5 . 1  and 5 .2. 
In terms of saturated fatty acids (SAFA), the relative quantities in fraction-I and fraction-2 
of crude and orange roughy oils showed a similar change trend. Fraction-I oils had a 
slightly lower SAFA value in comparison to the unrefined oils. The SAFA value found 
in fraction-2 oils tended to be higher than the values analyzed in the unrefined oils and 
fraction-! oils. 
62 
The monounsaturated fatty acid (MUF A) levels observed in the fraction-1 oils from crude 
and orange roughy oils were higher than the amounts measured in the unrefmed oils, while 
the quantity of MUFA analyzed in the fraction-2 oils tended to be lower than in the 
unrefined oils and fraction-1 oils. 
Table 5.1 .  Effects of fish oil:resin volume ratio on fatty acid profile of crude fish oil (% 
fatty acid) 
Oil : SAFA MUFA PUPA EPA DHA Omega-3 
Fraction resin f.a. 
Untreated 25.3 60.4 14.4 4.2 5.1 1 1 .6 
oil 
1 : 1 24.5 60.7 14.8 4.3 5.2 12.0 
2 :  1 25.0 60.2 14.9 4.4 5.3 12.1 
I 3 :  1 24.7 60.5 14.7 4.4 5.1  1 1 .9 
4 :  1 24.2 60.9 14.9 4.4 5.3 12.1 
1 : 1 26.0 58.9 15.1 4.4 5.4 12.2 
2 :  1 26.4 58.7 14.9 4.3 5.6 12.3 
II 3 :  1 26.1 58.2 15.3 4.7 5.5 12.5 
4 :  1 26.4 58.3 15.3 4.7 5.5 12.5 
Statistically, the polyunsaturated fatty acid (PUP A) value of crude oil was relatively constant 
during the refining process and this occurrence was revealed in both fraction-1 and 
fraction-2 oils. Fraction-1 orange roughy oil showed the same response as in fraction-1 
crude oil. Fraction-2 orange roughy oil indicated a different pattern, where the PUPA value 
was relatively constant when refmed with fish oil-resin volume ratios 1 : 1  and 2: 1 .  This 
value was significantly higher when the fish oil volume was elevated to three and four times 
resin volume. The PUP A value in fraction-2 orange roughy oil was significantly higher than 
the value in fraction-1 .  This occurrence was not noted in the crude oil. 
63 
Table 5.2. Effect of fish oil-resin volume ratio on fatty acid profile of orange roughy oil 
(% fatty acid) 
Fraction Oil : SAFA MUFA PUPA EPA DHA Omega-3 
Resin f.a. 
Untreated 6.2 87.6 6.3 1 .5 1 . 1  3 .5 
oil 
1 : 1 6.1 87.9 6.0 1 .4 1 .0 3.3 
2 :  1 5.7 88.2 6.0 1 .3 1 .0 3.3 
I 3 : 1 5.7 88.2 6.1 1 .4 1 .0 3 .3 
4 :  1 5.7 88.1 6.2 1 .5 1 .0 3.3 
1 : 1 6.7 87. 1 6.2 1 .4 1 . 1  3 .7 
2 :  1 7.0 86.8 6.1 1 .4 1 .2 3.7 
11 3 : 1  7.1 8(5.4 6.5 1 .5 1 .3 3.9 
4 :  1 7.4 85.6 7.0 1 .6 1 .5 4.1 
The resin refining process did not significantly change the omega-3 fatty acid value of crude 
oil, where unrefined, fraction-1 and fraction-2 oils from all tested fish oil-resin volume ratios 
did not show any pronounced different in omega-3 fatty acid value. The refining process 
also seemed to keep the omega-3 fatty acid value in fraction-1 orange roughy oil constant. 
The omega-3 fatty acid value in fraction-2 orange roughy oil indicated a significant increase 
when the fish oil ratio was increased to 4:1 .  In both oils the omega-3 fatty acid value in 
fraction-2 oils tended to be markedly higher than the value in fraction-1  oils. 
Eicosapentaenoic acid (EPA) values of orange rougby oil was insignificantly affected by the 
resin refining process, since the values in both fraction-1 and fraction-2 oils were relatively 
similar to the value analyzed in unrefmed oil. EPA value in fraction-1 crude oil was also 
foum:l llfti: ? 't' S'ilg ific t m of EPA: val e t'Gffiitl fffl ?01r?2 \ 
crucle oil when the fish oil volume ratio was elevated to three and four time that of resin 
olUme according to t-test value at 95% significant le.vel with 4 degtee of freed'OIIi. 
Docosabexaenoic acid (DHA) value of the crude oil did not indicate any significant change 
due to the resin refming process. Also fraction-1 orange roughy oil did not show an 
64 
obvious change, even though the fish oil volume was raised to ratio 4: 1 .  A constant DHA 
value in fraction-2 orange roughy oil was encountered until the fish oil-resin volume ratio 
was 3: 1 .  V.??ev val ignjfi, amJ ? u.?ased ??n:'M'li?'l? 
increased to 4: 1 according to t-test value at 95% significant level wjth 4 degree of freedom. 
5.4.1.5. Effects on sensory properties 
The resin refming process demonstrated a significant effect on the odour and taste scores 
of fraction- ! and fraction-2 crude and orange roughy oils as shown in Figures 5 .4 and 5.5.  
9 
B 
7 
-
G) 
.... 6 0 u 
Cl) 5 
.... :::::l 
0 4 "C 0 
3 
2 
Untreated 1 : 1 
011 
2 :  1 3 : 1 4 : 1 
Fish Oi l  Volume : Res in Volume 
CJ fnctlon I - crude oil 
E2:l fraction 11 - crude oil 
E!:tl fraction I - orange roughy oil 
- fraction 11 - orange roughy oil 
Figure 5.4. Effects of fish oil-resin volume ratio on odour score of fish oil 
Odour and taste scores of both oils were apparently improved by the resin refining process. 
In the crude oil, as shown by odour and taste scores of fraction-! oils, the best quality oil 
was obtained at fish oil-resin volume ratio 1 : 1 .  When the fish oil volume ratio increased, 
the odour and taste scores became higher, reflecting an inferior quality oil. In the 
refinement of orange roughy oil at fish oil-resin volume ratio 1 : 1, the fraction- !  oil obtained 
also improved its quality in terms of odour and taste. Increasing fish oil volume to ratio 4: 1 
did not, significantly, affect the odour and taste scores of fraction-I, refined oils. 
9 
8 
7 
? .... 6 0 0 
!!?. 5 
? -
.. .. 10" 
I-
3 
2 
Untreated 1 : 1 
Oil 
2 : 1 3 : 1 .. : 1 
Fish Oil Volume : Resiri Volume 
0 fraction I - crude oil 
IZl fraction il - crude oil 
65 
? fraction I - orange roughy oil 
? fraction 11 - orange roughy oil 
Figure 5.5. Effects of fish oil-resin volume ratio on taste score of fish oil 
Odour and taste scores of fraction-2 crude oil produced from refining with fish oil-resin 
volume ratio 1 : 1 ,  was lower than the unrefined oil. When the fish oil volume ratio 
increased to 4:1 ,  the odour and taste scores of fraction-2 oils tended to be relatively equal, 
or higher, than the scores for fraction-2 oils from previous refining. However the odour and 
taste scores of fraction-2 orange roughy oil, from all tested fish oil-resin volume ratios, were 
higher than the scores for unrefined oil. 
In general, fraction-2 oils from both crude and orange roughy oil had higher odour and taste 
scores, compared to the fraction-I oils. 
66 
5.4.2. Effects of multiple refining on fiSh oil quality 
5.4.2.1. Effects on free fatty acid value 
Both crude and orange roughy oils showed a similar response in terms of free fatty acid 
value to multiple refining treatment, as shown in Figure 5.6. 
CD :::J 
?1 
> -=o 
"C ?c::; 
?c::; < 
< 0 
>- . ., ... 0 -?J u. ? 
CD CD .... u. 
1 .3 
0.9 
0.-4 
0 
/<;: 
?? X 
?g ?g ?g X 
?g ?g 2 
Untre1 tod 
Oil 
[7 
? ;; 
? ? 
t7; 
? ? ? ? ;; ?g ?? / [X ?8 ?g ?? ?? X :X 
?g ?? ? 
;;g  ;;? ?? X V. :X ;;:x 
11 Ill 
Refining (Co lumn) 
? 
% % 
%? 
%? %? 
IV 
0 fraction I - mixed oil 
E2l fraction 11 - mixed oil 
E::3 fraction I - orange roughy oil 
? fraction 11 - orange roughy oil 
Figure 5.6. Effects of multiple refining on free fatty acid value of fish oil 
Fraction-! oils from the first, second, third and fourth refinings indicated significantly lower 
free fatty acid (FFA) values than unrefined oils. Fraction-! crude oils from the first, second 
and third refining showed an insignificant different in FF A values. The FF A values of 
fraction-! oil from the first and second refinings indicated a higher value compared to the 
value of fraction-! oil from the fourth refining. The fraction-! orange roughy oil collected 
from the first and second refinings indicated an insignificant different in FFA values. 
However the FFA values of the fraction-1 oil from the first refining was apparently higher 
than the values of fraction-1 from the third and fourth refinings. The values of fraction-1  
oils from the second and third refinings showed a significant increase over the value of the 
fraction-! oil from the fourth refining. 
67 
Each refining consistently produced fraction-1 oils with a lower FF A value in comparison 
to fraction-2 oils. The FFA value of fraction-2 for both oils from the first refining was 
markedly higher than the value of unrefined oil. That value significantly decreased in the 
second, third and fourth refinings. 
5.4.2.2. Effects on refractive index 
Multiple refining showed a significant effect on the refractive index (RI) changes in 
fraction-1 and fraction-2 for crude and orange roughy oils as shown in Figure 5.7. 
)( ? -c 
1 .475 
1 .470 
c () 1.465 
Cl ? ,. 0 
?;: N 
? -;;; 1 .460 .... --? 
a: 
1.455 
Untreated 
Oi l  
11 Ill 
Refining (Column) 
IV  
D fraction I - mhted oil 
EZJ fraction 11 - mixed oil 
g;a fraction I - orange roughy oil 
? fraction 11 - orange roughy oil 
Figure 5.7. Effects of multiple refining on refractive index of fish oil 
RI values of fraction-I oils from first, second, third and fourth refinings were significantly 
lower compared to the value observed in unrefined oils. The RI values of fraction-I oils 
from the first refming could be reduced by passing the oil through the column a second 
e o lfaction-1 oils from tl:ie 
68 
The RI values of fraction-2 for both oils showed a similar pattern. The RI values of 
fraction-2 oils from the first refming were lower than the values in unrefined oils, but the 
second, third and fourth refining resulted in fraction-2 oils having RI values which were 
relatively the same value as untreated oils. In general, the RI values of fraction-2 oils were 
obviously higher than the values of fraction-2 oils. 
5.4.2.3. Effects on colour absorbance 
Analysis of variance indicated that the fish oil colour could be improved by application of 
multiple resin refining as shown in Figure 5.8. 
E 
c 
0 
(7) 
...,. 
... -::I ... 
.!:! 
0 Cl 
0 u c 
... 
..c ... 0 
... ..c 
< 
1 .6 
1.2  
0.8 
0 . .4 
Untrelted 
Oil 
D friction I - mixed oil 
EZJ friction I - or1nge roughy oil 
11 Ill IV 
Refining (Column) 
Figure 5.8. Effects of multiple refining on colour absorbance value of fish oil 
The colour absorbance values of unrefined oils was reduced during the frrst refming. Those 
values were reduced again, when the fraction-I oils were passed through the column for the 
second time. The third refining decreased the colour absorbance value of fraction-! orange 
69 
roughy oil from the second refming. This did not occur in the crude oil. However the 
fourth refming significantly further reduced the absorbance values of both fraction-1 oils 
from the third refining. 
5.4.2.4. Effects on fatty acid profiles 
Fatty acid profile changes in fraction-1 and fraction-2 crude and orange roughy oils, as the 
result of the effect of multiple refining application, can be seen in Tables 5.3 and 5.4. 
Multiple refining did not result in any pronounced differences in the relative quantities of 
saturated fatty acids (SAFA) in fraction-1 and fraction-2 crude and orange roughy oils from 
the frrst, second, third and fourth refmings. In addition, these values were relatively the 
same as the relative value analyzed in unrefined oil. 
The relative quantities of monounsaturated fatty acids (MUF A) of fraction-1 and fraction-2 
of both oils from the frrst, second, third and fourth refinings did not indicate any significant 
different from the quantity analyzed in the unrefmed oil. The multiple refming treatments 
did not produce any significant MUFA difference between fraction-1 and fraction-2 oils. 
Statistically, the multiple refining treatments did not result in any pronounced effect on the 
relative quantities of polyunsaturated fatty acids (PUF A) for both oils. 
70 
Table 5.3. Effect of multiple refining on fatty acid proflle of crude fish oil (% fatty acid) 
Fraction Refining SAFA MUFA PUFA EPA DHA Omega-3 
f.a. 
Untreated 27.3 59.2 13.5 3 .9 5.0 1 1 .1 
oil 
I 26.1 59.3 14.6 4.2 5.7 12.2 
11 27.1 58.9 14.0 4.0 5.3 1 1 .6 
I Ill 26.7 59.0 14.3 3.9 5.6 1 1 .8 
IV 27.0 59.1 13.9 3.9 5.4 1 1 .5 
I 27.4 58.5 14.2 3.9 5.6 1 1 .6 
11 26.9 59.3 13.8 3.8 5.2 1 1 .2 
II Ill 27.5 59.2 13.3 3.8 4.9 10.8 
IV 27.3 "59.2 13.4 3.8 5.0 10.9 
Table 5.4. Effect of multiple refining on fatty acid proflle of orange roughy oil (% fatty 
acid) 
Fraction Refining SAFA MUFA PUFA EPA DHA Omega-3 
f.a. 
Untreated 8.4 86.3 5.4 1 .1  1 .2 3 .3 
oil 
I 8.3 85.3 5.5 1 .2 1 .1  3.4 
11 8.0 86.8 5.2 1 .2 1 .2 3.3 
I Ill 8.8 85.9 5.3 1 .3 1 .2 3 .4 
IV 8.7 85.9 5.3 1.3 1.1 3.3 
I 9.3 85.3 5.4 1 .1  1 .2 3.4 
11 8.3 85.8 5.4 1.3 1 .2 3.4 
11 Ill 8.2 85.6 6.2 1 .6 1 .2 4.0 
IV 8.1 86.2 5.7 1 .8 1 . 1  3.7 
71 
affected by multiple refining treatments. This result occurred in both fraction-!  and 
fraction-2 oils. The significant changes in eicosapentaenoic acid (EPA) values were 
observed only in fraction-! crude oil, where an increased EPA value was noted in fraction -
1 from the first refinings, but was followed by a decreasing pattern with further refining. 
Obvious docosahexaenoic acid (DHA) changes occurred in orange roughy oil. The increase 
in DHA values in fraction-!  oil was observed in the oil obtained from the second refining. 
The DHA value in fraction-2 orange roughy oil significantly decreased at the fourth refming. 
These significant changes were shown by t-test at 9 
freedom. 
5.4.2.5. Effects on seno;ory properties 
In general, odour and taste scores for ?Crude and orange roughy oils showed a similar 
response to the multiple refining treatment, as shown in Figures 5.9 and 5.10. 
9 
8 
7 
CD 
._ 6 0 u 
? 5 
._ :::J 
0 o4 "'0 0 3 
2 
Untreated 
Oil 
11 Ill 
Refining (Co lumn) 
IV 
D fraction I - mixed oil 
EZI f rac tion 11 - mixed oil 
g::g fraction I - orange roughy oil 
? fraction 1 1  - orange roughy oi l  
Figure 5.9. Effects of multiple refining on odour score of fish oil 
The first refining resulted in fraction- ! oils having significantly lower odour and taste 
scores, reflecting a sensory improvement. Further refining tended to give lower scores, 
indicating a better quality, but the changes were statistically insignificant. 
9 
8 
7 
-
CD .... 6 0 
u 
??. 5 
CD -
.. 4 1!1 
r-
3 
2 
Untreated 
Oi l  
11 Ill 
Refining (Column} 
IV  
0 fraction I - mixed oil 
rzj fraction 11 - mixed oil 
15::3 fraction I - or1nge roughy oi l  
? fraction 11 - orange roughy o il 
Figure 5 .10. Effects of multiple refining on taste score of fish oil 
72 
The first refining produced fraction-2 oils with odour and taste scores significantly higher 
than unrefined oils. These scores tended to be lower compared to the fraction-2 oils 
obtained from further refining. In general, multiple refining resulted in the fraction-2 oils 
acquiring a more unpleasant odour and taste quality in comparison to fraction-1 oils. 
5.4.3. Effects of vacuum pressure application on fish oil quality 
Application of vacuum pressure at one end of the column was aimed at speeding the 
refining process. 
5.4.3.1 .  Effects on free fatty acid value 
Free fatty acid (FF A) values of fraction-1 crude and orange roughy oils showed a different 
response to vacuum pressure treatment as shown in Figure 5.1 1 .  However the FFA values 
73 
of both fraction-1 oils yielded from refining with and without vacuum pressure had a 
significantly lower FFA value than unrefmed oils. 
1 .2 
t) 1 .0 
..2 
Ill 
> -
"C 0.8 -c ?u 
?u 111 
< u 
>- "ii 0.6 
... -? 0 
1.1.. H 0.-4 
t) 
t) 
... 
1.1.. 0.2 
Crude Oil Orange Roughy 
Type qf Fish Oil 
? Untreated oil 0 fraction-1 ;  without 
vacuum pressure 
? tractlon-2; without vacuum pressure 
B;ij fractlon-1 ;  with 
vacuum pressure 
m fractlon-2; with 
vacuum preuure 
Figure 5 .1 1 .  Effects of vacuum pressure during refining on free fatty acid value of fish oil 
The fraction-1 crude oil obtained from the refining with vacuum pressure application 
exhibited a lower FF A value than the fraction-1 crude oil from refining without vacuum 
pressure. The FFA values of fraction-I orange roughy oil obtained from refining with and 
without pressure did not show any significant different 
Fraction-2 crude and orange roughy oils exhibited the same response to the vacuum pressure 
treatment in terms of FFA value. The FFA values of fraction-2 oils were significantly 
higher than the value in unrefined oils. The FFA value of fraction-2 oils from refining 
without pressure was higher than the value analyzed in fraction-2 oils from refining with 
vacuum pressure. 
In general, both refining with and without vacuum pressure application resulted in fraction-2 
oils having a higher FF A values than fraction-1 oils. 
74 
5.4.3.2. Effects on refractive index 
Refractive index (RI) values of fraction-1 crude and orange roughy oils from refming with 
and without vacuum pressure was markedly lower than the RI values of unrefined and 
fraction-2 oils as shown Figure 5. 12. Crude oil refined using vacuum pressure yielded 
fraction-1 oil having a lower RI value than fraction-1 oil obtained from refining without 
pressure. However the fraction-1 orange roughy oil refmed with and without vacuum 
pressure showed no significant differences in the RI value. 
>< 
1 .-48 
1 .478 
1 .476 
C) 1 .474 
"C 
-= 0 1 .472 
C) ? 
.? ? 1 .470 -
0 -? ..!!. 1 .4118 
-
? 1..466 
1 ..464 
1 .462 
1 .46 .....__....._....__ 
Crude Oil Orange Roughy 
Type of Fish Oil 
? Unrefintd oil 0 Fraction-1 ;  without 
vacuum preuure 
? Fractlon-2; without 
vacuum preuure 
&a Fraction-1 ;  with 
vacuum preuure 
fE Fraction-2; with 
vacuum preuure 
Figure 5.12. Effects of vacuum pressure during refining on refractive index value of fish oil 
The RI values of fraction-2 oils obtained from refming with and without vacuum pressure 
were similar, and these values were relatively the same as measured in unrefined oils. 
5.4.3.3. Effects on colour absorbance 
In terms of colour absorbance value, crude and orange roughy oils showed a different 
response to vacuum pressure treatment during refining as shown in Figure 5 .13 .  
75 
2.00 
E 1 .72 c 
0 
? Untreated oil 
0 fractlon-1 ;  without 
en 
-.r 1 .44 ... .... ::I fU 0 
vacuum preuure 
? fractlon-2; without 
vacuum preuure 
0 ID 0 1 .1 6  0 c fU 
? fraction- 1 ;  with 
vacuum pressure 
.0 ... 0 0.88 .. 
m fraction-2; with 
vacuum pressure 
.0 fU -
0.60 
Crude Oil Orange Roughy 
Type o t  Fish Oil 
Figure 5.13. Effects of vacuum pressure during refining on colour absorbance value of fish 
oil 
Fraction-I oils from refining with and without pressure had a lower colour absorbance value 
compared to unrefined and fraction-2 oils, and the values of fraction-2 oils obtained from 
both refming methods were relatively similar. 
The colour absorbance values of fraction-2 oils were markedly higher than the values 
analyzed in unrefined oils. Vacuum pressure treatment resulted in fraction-2 oils having a 
colour absorbance which was relatively similar to the value observed in fraction-2 oil from 
refining without pressure application. 
5.4.3.4. Effects on fatty acid profile 
Fatty acid profile changes during refining with and without pressure are shown in Tables 
5.5 and 5.6. 
76 
Table 5.5. Effect of vacuum pressure during resin refming on fatty acid profile of crude fish 
oil (% fatty acid) 
Fraction Vacuum SAFA MUFA PUFA EPA DHA Omega-3 
treatment f. a. 
Untreated 19.4 62.6 18 .0 4.6 7.7 14.9 
oil 
Without 20.0 62.2 17.8 4.4 7.7 14.7 
I Vacuum 
With 20.0 62.0 18.0 4.4 7.7 14.8 
Vacuum 
Without 20.0 62.5 17.4 4.4 7.7 14.6 
II Vacuum 
With 19.8 62.6 17.6 4.6 7.3 14.4 
Vacuum 
Table 5.6. Effect of vacuum pressure during resin refming on fatty acid profile of orange 
roughy oil (% fatty acid) 
Fraction Vacuum SAFA MUFA PUFA EPA DHA Omega-
Treatment 3 f.a. 
Untreated 6.2 85.3 8.4 2.1 1 .7 5.5 
oil 
Without 6.3 85.6 8.0 1 .9 1 .9 5 . 1  
I Vacuum 
With 6.2 85.9 7.8 1.8 1 .9 5.0 
Vacuum 
Without 6.8 84.7 8.5 2.1 2.1 5 .4 
II Vacuum 
With 6.7 85.6 7.7 1 .8 2.1 5.0 
Vacuum 
Vacuum pressure treatment did not induce any significant effect on the fatty acid profiles 
of crude oil. Relative quantities ofMUFA, PUFA, omega-3 fatty acids and EPA in orange 
77 
roughy oil relatively unchanged during refining. However relative amounts of SAFA in 
variance indicated? that DHA yalue of both fraction-! and fraction-2 orange roughy oils f;om 
refining withi.an ?ou ?e.illiDLpfeSSut? was relatively higlie e,g 
.?1. 
5.4.3.5. Effects on sensory properties 
Fraction-I crude and orange roughy oils obtained from refming with and without vacuum 
pressure had better odour and taste properties compared 1.0 the unrefmed oils as shown in 
Figures 5.14 and 5.15 .  Vacuum pressure application did not result in any differences in 
odour and taste scores between fraction-! oil from refining with and without vacuum 
pressure. However the odour and taste scores of fraction-2 oils were higher than the scores 
of fraction-! oils, indicating that the odour and taste of fraction-2 oils were more unpleasant 
than those of fraction-I oils. 
9 
8 
7 
-
I) ... 6 0 0 5 (f) .... ... " ::J 
0 -a 3 0 
2 
0 
Crude Oil Orange Roughy 
Type of Fish Oil 
? Unrefined oil O Fraction- 1 ;  without 
ncuum preuure 
? Fraction-2; without 
vacuum preuure 
63 Fractlon-1 ;  with 
Ylcuum preuure 
ffi Fraction-2; with 
vacuum preuure 
Figure 5.14. Effects of vacuum pressure during refming on odour score of fish oil 
9 
8 
7 
? 6 
0 
? 5 
!! " 
.. 
f!J 
1- 3 
2 
Crude Oil Orange Roughy 
Type of Fish Oil 
f3 Unrofinod oil 0 Fraction-1 ;  without 
vacuum pressure 
? Fractlon-2: without 
vacuum prusurt 
&a Fraction-1 ;  with 
vacuum pressure 
B3 Fractlon-2; with 
vacuum pressure 
Figure 5 .15 .  Effects of vacuum pressure during refining on taste score of fish oil 
5.4.4. Effects of column size on fish oil quality 
5.4.4.1. Effects of various height sizes of resin packed column 
78 
The diameter of the column was kept constant at 2.60cm while the column height was 
varied to 5, 10, 15 and 20 times diameter size. 
Fraction-1 oils obtained from all diameter-height ratios had a significant lower free fatty 
acid (FFA) value compared to the unrefined oil, as shown in Figure 5 . 16. This trend 
showed that the higher the height the lower the FF A value, but varying height size in this 
79 
study did not induce any significant statistical different in the FFA value among fraction-I 
oils. However treatment of various height column sizes affected the FF A value of fraction-2 
oils: the higher the height the higher the FF A value. In general, the results indicated that 
the FFA value of fraction-2 oils was higher than the value analyzed in unrefrned and 
fraction-! oils. 
G) 
::J 
tO 3 > o:o 
-c ?u ?u < 
< 0 2 >- ?c; - -iij O  
LL ?  
., ?
G) 
.... LL 
Untrutod 1 : 5 1 : 1 0  1 : 1 5  1 : 20 
Oi l  
Diameter : Length  
0 fraction I 
E2j fraction 11 
Figure 5.16. Effects of various height-diameter ratios of column on free fatty acid value of 
fish oil 
As occurred in the FF A analysis, the RI values of fraction-1 oils were lower than the value 
analyzed in fraction-2 and unrefined oils as shown in Figure 5 .17. The various height sizes 
did not affect the RI values of fraction-I oils. The RI values of fraction-2 oils from refining 
with diameter and height ratio of 1 :5  was lower than the value analyzed in fraction-2 oil 
from refining with the height size 10, 15 and 20 times diameter size. 
>< 
(I) 
"'C 
c: -- (j  
c ? > 0 
?;::; C\1 
0 ... 
IV IV ,_ _  
-CD 
a: 
Untreated 1 : 5 1 : 1 0  1 : 1 5  1 : 20 
Oil 
Diameter : Length 
0 fraction I 
e;:a fraction 11 
80 
Figure 5.17. Effects of various height-diameter ratios of column on refractive index value 
of fish oil 
As shown in Figure 5 .18, the colour absorbance of crude oil could be reduced by passing 
the oil through the resin column with diameter-height ratio 1 :5. When the height size of the 
column was extended to 10  times diameter size, the colour absorbance value of fraction- 1  
oil was lower than the  value of fraction-1 oil from refining with column diameter-height 
ratio 5 : 1 .  The colour absorbance value would be relatively unchange, even though the 
column height size was extended to 15 and 20 times diameter size. 
E" 
c 
0 
01 
'<t 
.... .. :I la 
.? 0 G 
0 0 c la ..c ... 0 
.. ..c 
.:5 
2.6 
2.1 
1 .6 
1 .0 
0.5 
Untreated 1 : 5 1 : 1 0  1 : 1 5  1 : 20 
Oil 
Diameter : Length 
0 fraction I 
ea fraction 11 
81 
Figure 5 .18 .  Effects of various height-diameter ratios of column on colour absorbance value 
of fish oil. 
The increasing height size resulted in an increase in absorbance value of fraction-2 oils. 
The colour absorbance values of fraction-2 oil from refining with column diameter-height 
ratios 1 : 10 and 1 : 15 were insignificantly different. Those values were higher than the colour 
absorbance value of fraction-2 oils from refining with column diameter-height ratio 1 :5 ,  but 
lower than the value of fraction-2 oil from refining with column diameter-height ratio 1 :20. 
Results of fatty acid profile analysis are shown in Table 5.7. Analysis of variance showed 
that treatment of various height size did not produce any significant changes in relative 
quantities of saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, 
omega-3 fatty acids, eicosapentaenoic acid and docosahexaenoic acids in unrefined, fraction-
1 and fraction-2 oils. 
82 
Table 5.7. Effects of height size of resin packed column on fatty acid profile of crude oil 
(% fatty acid) 
Fraction 
Untreated 
oil 
I 
II 
9 
8 
7 
-G) 
0 6 0 
?! 5 
..... 
:::1 
.g .. 0 
3 
2 
diameter: SAFA MUFA PUFA 
height 
21.0 59.3 19.7 
1 :  5 22.8 58.2 18.9 
1 :  10 21 .7 58.5 19.8 
1 : 15 18.8 60.5 20.7 
1 :  20 19.5 60. 1 20.4 
1 : 5 20.3 58.8 20.9 
1 : 10 17.1 60.4 22.5 
1 : 15 17.4 60.7 21.9 
1 : 20 19.4 59.4 21 .2 
Untreated 1 : 5 1 : 1 0  1 : 1 5  1 : 20 
Oil 
Diameter : Length 
EPA DHA 
5.5 7.8 
5.1 7.5 
5.4 8.2 
6.1 8.2 
5.9 7.8 
5.8 8.1 
6.2 9.3 
5.7 8.9 
5.9 8.2 
0 fraction I 
fZl fraction 11 
Omega-3 
f. a 
17.0 
16.2 
17.1 
17.7 
17.5 
18 .1  
19.6 
18.9 
18.1  
Figure 5. 19. Effects of various height-diameter ratios of column on odour score of fish oil 
Q) .... 0 0 
!1 
Q) ... 
.. 
10 
1-
9 
8 
7 
6 
5 
? 
3 
2 
Untreated 1 : 5 1 : 1 0  1 : 1 5  1 : 20 
011 
Diameter : Length 
0 fraction I 
E'ZJ fraction 11 
83 
Figure 5.20. Effects of various height-diameter ratios of column on taste score of fish oil 
Odour and taste scores of fraction-1 and fraction-2 oils from refining with all diameter?
height ratios were lower than the scores observed in unrefined oil. However the odour and 
taste scores of fraction-1 oils were significantly lower than the scores of fraction-2 oils as 
shown Figure 5 .19 and 5.20. The diameter-height ratio treatment did not result in any 
significant difference in odour and taste scores in fraction-1 or fraction-2 oils. 
5.4.4.2. Effects of various diameter sizes of resin packed column 
In this experiment, the column height was kept constant at 39cm. The column diameter 
sizes investigated were 1.65, 2.60 and 3 .60cm. 
Figure 5 .21 shows the changes ofFFA values in fraction-1 and fraction-2 oils obtained from 
84 
refining with various column diameter sizes. FFA values of fraction-1 oils obtained from 
refining with all column diameter sizes were significantly lower than the value of unrefined 
oils, while the FFA values of fraction-2 oils were significantly higher than the values of 
unrefined oil. Refining with various diameter sizes did not result in any significant 
differences in FF A values of fraction-1 oils or fraction-2 oils. FF A values for fraction-2 oils 
obtained from refming with column diameter sizes 2.6 and 3 .2 cm were significantly higher 
than the values for unrefined oil. 
2.5 
? 2.0 
:: ?  u "'0 fll 
"(j u 1 .5 
< .G 
? 0  
? ? 1 .0 
u.. 
.. 
G) fll 
G) -
u: 0.5 
Untreated 
Oil 
1 .65 2.60 
Diameter Size (Cm) 
3.20 
0 Fraetlon-1 oil 
? Fraetlon-2 oil 
Figure 5.21.  Effects of various diameter sizes of column on free fatty acid value of fish oil 
Refractive index values of fraction-1 oils from refining with all column diameter sizes were 
significantly lower than the value measured in unrefmed and fraction-2 oils as shown in 
Figure 5.22. The values analyzed in fraction-! oil from refining with column diameters 2.6 
, and 3 .2cm were markedly lower than the value observed in fraction-! oil from refming with 
column diameter size 1 .65cm, while the RI values of fraction-2 oils from refining with all 
column diameter sizes were similar to the value of unrefined oils. 
>< 
C) 
"'C 
1 .475 
U70 
..5 0 1 .465 
C) ? 
. ? ? .... 
0 .... 
?< IV 1 .460 .._ _  
-C) 
a: 
1 .455 
Untreated 
Oil  
1 .65 2.60 
Diamster Size [Cm) 
3.20 
0 Fractlon-1 oil 
? Fractlon-2 oil 
85  
Figure 5.22. Effects of  various diameter sizes of  column on  refractive index value of  fish 
oil 
Results of colour absorbance value of untreated, fraction-1 and fraction-2 oils are shown in 
figure 5.23. All column diameter sizes resulted in fraction-1 oils with obviously lower 
absorbance values than unrefined oils. The bigger diameter sizes tended to give a lower 
colour absorbance value in fraction-I oils, but this was statistically insignificant. Colour 
absorbance value of fraction-2 oils from all refining was significantly higher than the value 
of unrefmed oil. The colour absorbance values of fraction-2 oil from refining with column 
diameters 2.6 and 3.2cm were significantly higher than the value observed in fraction-2 oils 
from refining with a column diameter of 1 .65cm. 
Qj 
..: ?< 
> 
... 
::I Cl) 
0 0 c: 0 ?< 
() .a ... 0 
"' 
.a 
? 
1 .3 
1 . 1  
0.9 
0.7 
0.5 -'---"--' 
Untreated 
Oil 
1 .65 2.60 
Diameter  Size (Cm) 
86 
0 Fractlon-1  oil 
? Fractlon-2 oil 
3.20 
Figure 5.23. Effects of various diameter sizes of column on colour absorbance value of fish 
oil 
Sensory evaluation results for odour and taste performed by eight Indonesian trained 
panellists are shown in Figures 5.24 and 5 .25. Both odour and taste scores for fraction-1 
oils, from all refmings, were significantly lower than the scores for unrefined oil. A lower 
score reflected a better acceptance by panellists. The odour and taste scores of fraction-1 
oils obtained from refining using resin column with all diameter sizes, were insignificantly 
different. In general, the fraction-1 oils showed a better performance in odour and taste 
compared to the fraction-2 oils. However the odour and taste scores for fraction-2 oils 
from all refining was insignificantly different from the scores for unrefined oil. 
? .... 0 0 
? 
.... 
:J 
0 "'0 0 
9 
8 
7 
6 
5 
" 
3 
2 
Untreated 
Oil 
1 .65 2.60 
Diameter Size (Cm) 
3.20 
Cl Fractlon- 1 oil 
? Fraction-2 oil 
Figure 5.24. Effects of various diameter sizes of column on odour score of fish oil 
9 
8 
7 
'iD' .... 6 0 0 
(/) 
5 -
11) -"' " 10:1 
1-
3 
2 
Untreated 
Oil 
1 .65 2.60 
Diameter Size (Cm) 
3.20 
0 Fractlon-1  oil 
? Fractlon-2 oil 
Figure 5.25. Effects of various diameter sizes of column on taste score of fish oil 
87 
88 
5.4.5. Effects of resin refining on natural antioxidant contents of fiSh oil 
As shown in Table 5.8, the natural antioxidants traced in crude and orange roughy oils were 
a-tocopherol and y-tocopherol accompanied by a-tocomonoenol. Both oils consisted of 
mainly a-tocopherol. The natural antioxidant content (tocopherol group) decreased 
significantly during the resin refming process. 
Table 5.8. Changes of natural antioxidant content of fish oil during resin refining process 
(ppm) (detection limit = 2-4% of value) I 
Fish oil Treatment a-tocopherol y-tocopherol a-tocomonoenol 
Unrefined 210.3 5.1  18.4 
Crude oil 
Refined 179.8 4.5 13 .1  
Unrefined 122.3 4.2 6.3 
Orange 
roughy Refined 106.8 3.6 5.0 
5.4.6. Effects of resin refining on volatile flavour compounds of fish oil 
Traces of volatile flavour compounds before and after refming treatment are shown in 
Figures 5 .26, 5 .27, 5.28, and 5.29. The relative quantity changes of volatile flavour 
compounds during refming are shown in Tables 5 .9 and 5.10. Most of those compounds 
have been identified in fish and fish oil by other researchers as cited in footnotes to each 
Table. 
!l;llll?t&--;il!lentifred :in crud oii 
19 hydrocarbons, 1 alcohol, 5 esters, 2 aldehydes and 1 acid. Before refining trea ent, 
metb I et:P I benz?.te ?s the moSl flbundant cqm_ppund ? Q.).gt;Ue fl ww, f unrefined 
?.oi, followe by 1 , 1-dimethylethyl?-!ID.ethyl propiemiCYc!eiEl"Md ethy14)(?Rzeate. ? <A:fter 
refining treatment, the relative amounts of most compounds decreased. However the 
relative amounts of methyl ethyl benzoate, 1 ,1 -dimethylethyl-2?metbyl propionic acid and 
ethyl benzoate were still high, while toluene increased significantly which indicates that this 
might be one of the most responsible compound of volatile flavour in refined crude oil. 
" 
? 
.. 
" 
a 
X 
? 
I 11 
JNII II?M Ul?l 
Orr..Jt.,..- l ... tlfitrt ? , ,ltrt 
lrst SIW'\[ a 411 orna zx tiMI mrs tal lM ea-? C 
J 
" 
" 
1 
ll 
" 
n 
" 
" 
" " 
" 
" 
L_???????????????? p 11( 
Figure 5 .26. Traces of volatile flavour compounds of unrefined crude oil 
? 
" 
? 
? 
wn 11-llll Ul?l s,s se 
o.r...a.,.. 1-.tlhn ? 1 am . hd& C 1 fi51 11L CICI(Jil9kt?l :. +t??U 
n 
,, " tl 
tlll UH P 
lt IS ? a  IU( 
Figure 5 .27. Traces of volatile flavour compounds of refined crude oil 
89 
90 
Table 5.9. Relative amounts of volatile flavour compounds of crude oil during refining 
Peak Relative amount (%) 
No. Volatile flavour Compounds 
Before After 
refining refining 
Hydrocarbons 
2 heptane b) 1 .0 3.4 
3 toluene (methyl benzene) a) b) h) 1 .7 23.2 
5 octane a) b) c) d) 3 .1  3.6 
6 ethyl benzene c) d) 0.4 1.8 
7 1,3-dimethyl benzene (m-xylene) i) 0.9 3 .0 
8 ethynyl benzene f) 1 .8 1 .0 
9 1-nonene 2.9 1 .6 
10 nonane i) 4.7 2.0 
12 1-decene h) i) 2.2 1 .1  
13 decane e) g) i) 3.8 1 .7 
14 limonene a) c) d) 3.8 1 .5 
16 1-undecene 1 .5 0.7 
17 undecane e) 3 .8 1 .7 
21 1-dodecene 0.6 0.7 
22 dodecane e) f) i) 1.4 0.6 
23 1-tridecene i) 0.4 0.2 
24 tridecane f) 0.7 0.3 
25 pentadecane e) h) 0.8 0.2 
28 hexadecane c) d) e) 1 . 1  4.9 
Alcohols 
1 1-penten-3-ol a) d) f) g) i) 2.5 1 .8 
Esters 
1 1  butyl ester-2-hydroxy propionate 0.8 0.3 
18 ethyl hexyl acetate 1 .8 0.9 
19 ethyl benzoate 10.4 7.3 
20 methyl ethyl benzoate 29.5 19.4 
26 diethylphtalate 2.4 2.2 
Carbonyls/aldehydes 
4 2-methyl-4-pentanal 2.4 0.9 
15 1-nonanal e) f) g) i) 0.8 0.3 
Acids 
27 1,1-dimethylethyl-2-methyl propionic acid 12.4 13.4 
'lote: a) identified m crayfish waste by Tanchotikul and Hsteh ( 191$9) 
b) identified in haddock flesh by Angelini and Merritt (1975) 
c) identified in crayfish tail meat by Vejaphan ? al (1988) 
d) identified in marine green algae by Sugisawa ? al (1990) 
e) identified in menhaden fish oil by Hsieh et al ( 1989) 
f) identified in whitefish by Josephson ? al (1983) 
g) identified in cod liver oil by Karahadian and Lindsay ( 1989) 
h) identified in salmon by Josephson ? al (1991)  
i)  identified in tuna oil by Crawford ? al ( 1976) 
.. 
? 
" 
I 
" 
JNU ll?l!SZ ([I?} S.,?CC! Dno.&t.,-a. 1Btl1i11"1 ? I I!IC 
lnt N. n ma trt vasna ots. ? ;at 4NM i'.lt 
n 
Figure 5.28. Traces of volatile flavour compounds of unrefined orange roughy oil 
.. 
? 
" 
.. 
" 
"" 11-asl ((I?) S.,d[l 
CJr,n.wlaf",. l*"lUi1r1 ? I lilt 
rnt?N. 11 rnz 1n ? ?? r11 ? U?-l'a scn 
l 
u 
1111 
. ., 
n 
" 
u? D 
15 ?  111( 
Figure 5.29. Traces of volatile flavour compounds of refined orange roughy oil 
91  
92 
Table 5.10. Relative amounts of volatile flavour compounds of orange roughy oil during 
refining 
Peak Relative amounts (%) 
No. Volatile flavour compounds 
Before After 
Refining Refining 
Hydrocarbons 
1 cyclohexane i) 6.4 3 .6 
2 3-methyl hexane 1 .9 0.9 
3 toluene (methyl benzene) a) b) h) 51 .5 42.3 
4 heptane b) 4.0 0.9 
6 ethyl benzene c) d) 3.3 8.3 
7 1 ,3-dimethyl benzene (m-xylene) i) 5.2 16.4 
8 1 ,4-dimethyl benzene (p-xylene) i) 2.2 6.1 
10 nonane i) 2.2 1 .1  
1 1  1-methylethyl benzene i) 0.8 3 .8 
12 1-ethyl-3-methyl benzene 0.2 1 .2 
13 1,3,5-trimethyl benzene a) c) e) 1.4 1 .0 
14 1 ,2,4-trimethyl benzene a) c) e) 1 .0 1 .8 
15 decane e) i) 1 .2 1 .0 
17 undecane e) i) 0.5 0.6 
20 dodecane e) f) i) 0.3 0.2 
Alcohols 
9 2-butoxy ethanol c) 2.0 1 .0 
Esters 
19 octyl acetate i) 1 .6 0.1 
21 diethyl phtalate 0.3 0.5 
Carbonyls/aldehydes 
16 1-nonanal e) f) g) i) 0.9 0.2 
Acids 
18 1 -methylethyl ester benzoic acid 0.7 0.3 
22 1 ,1-dimethylethyl-2-methyl propionic acid 5.7 0.9 
-
Halogens 
5 tetrachloroethene 6.8 7.5 
?ote: a) 1dentified m crayfish waste by Tanchotikul and Hsteh ( lY?Y) 
b) identified in haddock flesh by Angelini and Merritt (1975) 
c) identified in crayfish tail meat by Vejaphan et al (1988) 
d) identified in marine green algae by Sugisawa et al (1990) 
e) identified in menhaden fish oil by Hsieh et al (1989) 
f) identified in whitefish by Josephson et al (1983) 
g) identified in cod liver oil by Karahadian and Lindsay (1989) 
h) identified in salmon by Josephson et al (1991) 
i) identified in tuna oil by Crawford et al (1976) 
93 
encountered in high proportion were toluene (5 1 .5%), cyclohexane (6.4%), 1 , 1 -
dimethylethyl-2-methyl propionic acid (5.7%) and tetrachloroethene (6.8%). These results 
indicate that toluene seemed to be responsible for flavour performance. After refming, most 
volatile flavour compounds showed a decrease in relative quantities. Volatile flavour 
compounds exhibiting a significant increase in the relative quantities were m-xylene, p?
xylene and 1 -methyle?yl benzene. The relative quantity of toluene decreased, due to 
refining trea ent, b , as stil .tbe. mQst b 42.3%) in refined orange 
r?>ugb,Y. oil. 
5.5. DISCUSSION 
Refining aims to improve fish oil quality by removing impurities. These impurities can be 
broadly subdivided into three types: insoluble, colloidal and soluble. The insoluble 
impurities are moisture, rust, dirt and protein. Protein can also be present in colloidal 
suspension as phosphatides and carbohydrates. The soluble compounds are pigments, 
oxidation products, trace metals, pbosphatides, sulphur and nitrogenous chemicals, free fatty 
acids, mono and di-glycerides and unsaponiftable matter which is principally wax (Young, 
1982; Windsor and Barlow, 1981). Some of the above impurities are used as parameters 
to determine the quality of refmed fish oil in this study. 
5.5.1.  Effects of resin refining on chemical properties of fish oil 
Free fatty acid (FFA) content of crude and orange rougby oils was reduced significantly by 
refining using a resin packed column. FFA content reduction occurred at all fish oil-resin 
volume ratios, and even more reduction was achieved with multiple refining by passing the 
oil through a clean column more than once. Increasing the refining rate by application of 
vacuum pressure at one end of the column bad the same effect on FF A content reduction 
as shown in the refming without vacuum pressure application. Fraction-2 oils always bad 
a higher FF A content than the fraction-1  oils. This indication demonstrated that free fatty 
acids might be binding with the resin, rather than being trapped by resin pores. 
94 
In terms of the column size, the height of the column showed more effect on FFA content 
of the oil than the diameter size. The higher the column size, the lower the FFA content 
of the refined oil obtained. Various diameter sizes tested in this study tended to produce 
refined oil with similar FFA content. The higher column size may have given more 
opportunity for the oil to contact with the resin, since the oil bad to pass through more resin. 
The reduction of the FFA content of fish oil due to the refining process was also noted by 
Koning and Milkovitch (1984a; 1984) using ethanolamine and glycerol. The same 
occurrence was also registered in refming vegetable oils using ordinary refining procedures, 
such as in coconut oil (Gordon and Rahrnan, 199 1); soya bean oil (Sleeter, 198 1); sunflower 
oil and rapeseed oil (Balicer et .&, 1983). 
As expected, resin refining did not induce any significant change in polyunsaturated fatty 
acid (PUFA), particularly omega-3 fatty acid, in the refined oil as compared to the unrefined 
oil. The same tendency was noted by Femandez (1986) in pink salmon oil, using the same 
refining procedure. This result demonstrated the superiority of the use of cation-strong acid?
resin packed column for fish oil refming over molecular distillation, in which ffi-3 fatty acids 
were chemically modified by molecular distillation process (Fernandez, 1986). The PUFA 
and (J)-3 fatty acid of fraction-2 oil were insignificantly different from the value in unrefined 
and fmction:I oil. This indicated that the triglyceride might not bind with the resin, but 
were, rather, caught by the resin pores. 
Natural antioxidants a-tocopherol and y-tocopherol in the oils reduced in quantity as a 
result of the resin refming process. The decreased levels of both antioxidants were 
approximately 12 - 14%. a-tocomonoenol content of the oil decreased after refining, but 
the ability of a-tocomonoenol to inhibit oxidation is still unknown. The decrease of 
tocopherol in refined fish oil was also noted in cod liver oil and sprat oil refining using soda 
lye (Brzeska and Salmonowicz, 1973), and menhaden fish oil refining using different 
methods (Scott and Catshaw, 1991).  The refining process also reduced tocopherol content 
in vegetable oils such as soya oil processed using de gumming, neutralization, bleaching and 
deodorisation (Gutfmger and Letan, 1974, S leeter, 1981) ;  combined sunflower oil and 
rapeseed oil refined using acidification/neutralization, washing, drying, decolouration and 
deodorization (Ludwiki et.&, 1986); soybean oil refined with activated carbons (Bold et al, 
1991)  and coconut oil refmed using alkali refining, degumming, bleaching and deodorization 
(Gordon and Rabman, 1991).  
95 
If the storage treatment must be carried out the loss of tocopherol should be noted as a 
precaution. Brzeska and Salmonowicz (1973) reported that refmed fish oil having a lower 
natural tocopherol content than unrefined oil showed less resistance to oxidation than 
unrefined oil. 
5.5.2. Effects of resin refining on physical properties of fJSh oil 
The refractive index (RI) of fish oil decreased after refining with a cation-strong acid-resin 
packed column. The RI of an oil is characteristic within certain limits for each type of oil. 
The RI is also used as a measure of purity of the oil (Rossel, 1986). The lower the RI 
value, the higher the purity level of the oil, a purity which can be further improved by 
multiple refining. The height and diameter sizes of the column did not affect the RI value 
of refmed oil, since the values obtained from various height and diameter ratios were 
practically similar. Both refming with and without vacuum pressure applications showed 
the similar effects on the RI value of refined oils. 
trap the impurities. Operations which may be performed by resin are ion exchange, 
adsorption, molecular sieving and gel permeation (Pery et!!!., 1973). Meanwhile strong acid 
peptides arqino aci9s, cations and metals (Bio Rad 
The colour of fish oil was significantly improved by the resin refining process. The colour 
of fraction-2 oil was always worse than the colour of refmed and unrefined oils. Further 
improvement in oil colour could be obtained by using the multiple refining method. 
Vacuum pressure application resulted in refined oil with comparable colour quality to the 
refined oil obtained from resin refining without vacuum. The effect of height-diameter size 
ratio was noted until the ratio of 10: 1 .  Further height size enlargement produced oil without 
further pronounced colour improvement. Diameter size did not markedly affect the colour 
quality of the oil. Carotenoids contributing mainly to fish oil pigment have unsaturated 
linkages occurring between alternative pairs of carbon atoms in a long multiple branched 
chain (Fox, 1957). The larger number of double bonds within the molecules resulted in a 
96 
higher probability of weak bonds fonning at the intramolecular level. These intramolecular 
bonds are hydrogen bonds and/or Van der Walls force, and/or polar bonds (Bottino et al, 
1967). These have a greater tendency to fonn bonds making oil more likely to interact with 
a macroporous resin backbone (Femandez, 1986). This indication was shown in this study. 
The fraction-2 oils had a much darker colour than fraction-! oils and unrefined oils, in 
which fraction-2 oils seemed to have a higher concentration of carotenoids separated by 
resin. 
5.5.3. Effects of resin refining process on volatile flavour compounds 
5.5.3.1. Volatile flavour compounds in fish oil 
The two unrefmed New Zealand oils studied had different compounds responsible for their 
volatile flavour. As shown in Tables 5.9 and 5.10 the compounds identified in both oils 
have been identified in fish, fish oils and marine green algae by Tanchotikul and Hsieh 
(1989), Angelini and Merritt (1975), Vejaphan et al (1988), Sugisawa et al (1990), Hsieh 
et al (1989), Josephson et al (1983), Karahadian and Lindsay (1989), Josephson et al ( 1991) 
and Crawford et al (1976). The compounds contributing to the volatile flavour of crude oil 
were methyl ethyl benzoate (29.5% ), ethyl benzoate (10.4%) and 1,  1 -dimethylethyl-2-methyl 
propionic acid (12.4%). These compounds were probably responsible for the strong fishy 
odour and taste in unrefmed oil as evaluated by trained Indonesian panellists. Ethyl benzoate 
was also found as a volatile flavour in plums (Dirninger, 1989), providing aromatic odour 
(Stecher et ill, 1968). Alkane compounds of heptane, undecane and dodecane found in the 
unrefined oil were encountered in vanilla aroma (Vidal et ill, 1989). Nonane, tridecane, 
pentadecane and hexadecane detected in this fish oil were found in plums (Etievant et ill, 
1986; Dirninger et ill, 1989). Octane and dodecane traced in crude oil also contributed to 
the aroma of strawberries (Belitz and Grosch, 1987). Limonene, as analyzed in crude oil, 
was also detected in plums (Dirninger, 1989). 
orange roughy. Other compounds encountered at a? significant level were cyclohexane 
(6 4? ??dimetbylatb 
compounds, was detected as having a significantly less fishy odour and taste than crude oil. 
97 
Toluene was traced in the aroma of vanilla as well (Vidal et al, 1989) imparting a benzene 
like odour (Stecher, 1968) or a plastic like odour (Tanchotikul and Hsieh, 1989). Both m?
xylene and p-xylene detected in orange roughy were also found in plums (Dirninger et al, 
1989). Octyl acetate in volatile flavour of orange roughy also contributed to strawberry 
aroma (Belitz and Grosch, 1987). Tetrachloroethene, giving a chloroform like odour 
(Stecher et ill, 1968), was encountered in the orange roughy volatile flavour compounds 
while other halogen compounds such as dichloromethane and trichloromethane are also 
reported to contribute to the volatile flavour of fish (Van Straten and Maarse, 1983). 
Toluene, limonene, xylene and benzene derivatives in crude oil and orange roughy oil were 
probably degradation products of carotenoids (Tanchotikul and Hsieh, 1989; Belitz and 
Grosch, 1987; Josephson et ill, 1991). This degradation process may have occurred during 
the heat cooking stage of fish meal production. Diethylphtalate was traced in crude and 
orange roughy oils. This compound was also detected in the volatile flavour compound of 
plums (Dirninger et al, 1989). However diethylphtalate is odourless, as reported by Stecher 
et al (1968). Some of these volatile flavour compounds were also detected in marine green 
algae (Sugisawa et al, 1990), indicating that these compounds were probably obtained by 
fish during feeding. 
5.5.3.1. Effects of resin refining on volatile flavour compounds 
Significant changes in relative quantities of volatile flavour compounds of crude and orange 
roughy oils during resin refming were observed as shown in Tables 5.9 and 5.10. 
The relative quantities of some volatile flavour compounds in refmed crude oil showed a 
marked change in comparison to unrefined oil. The relative quantities of toluene increased 
greatly from 1 .7% to 23.2%. Ethyl benzoate and methyl ethyl benzoate indicated a 
reduction in their relative quantities, but still showed relatively high percentages, 7.3% and 
19.4% respectively. 1 ,1-dimethylethyl-2-methyl propionic acid also showed a significant 
contribution to the volatile flavour performance of refmed oil. These compounds, together 
with other minor compounds, gave a better volatile flavour perception to refined crude oil. 
This was reflected by the results of sensory evaluation, where, according to panellists as 
discussed in Section 5.5.4, refined crude oil had just a slightly fishy odour and taste. 
98 
A different occurrence was encountered in the orange roughy oil. Toluene showed a 
reduced relative quantity compared to the unrefmed oil, but it was still the compound which 
contributed the most to the refined orange roughy oil at 42.3%. Fish oil type might show 
a different effect in terms of volatile flavour compounds when passed through the resin 
packed column. 
M-xylene, p-xylene, ethyl benzene and tetrachloroethane contributed at 16.4%, 6.1 %, 8.3% 
and 7.5% respectively to volatile flavour of refined orange roughy. With these compounds 
as the main volatile flavour compounds, the improvement in odour and taste was obtained, 
as indicated by the results of sensory evaluation in Section 5.5.4. 
The above results indicated that toluene, providing a benzene like odour (Stecher, 1968) or 
a plastic like odour (Tanchotikul and Hsieh, 1989) as the important compound in the 
improvement of fish oil odour and taste, since this compound was present at the highest 
relative amount in refmed crude and orange roughy oils. 
5.5.4. Effect of resin refining on sensory properties of fish oil 
Fish oil quality was improved?n terms of sensory properties of both odour and taste. The 
changes in relative quantity of volatile flavour compounds were detected as discussed in the 
previous section, where it states that toluene tended to appear as the highest relative 
compound in the refined oil. The odour and taste qualities of fraction-2 oil was always 
inferior to the odour and taste of fraction-1.  Fernandez (1986) proved that most of 
undesirable compounds were bonding or interacting with the resins, rather than being simply 
caught in the resin pores. 
Further improvement of odour and taste could be achieved by application of multiple 
refining. This fmding provides very significant information for the Indonesian fish oil 
industry, especially as fish meal oil has a very unpleasant odour, as discussed in Chapter 
4. The odour of this oil may not have been significantly improved by refining the oil once. 
This will be discussed in detail in Chapter 8. 
The odour and taste of refined fish oil became less acceptable, when the fish oil-resin 
volume ratio increased more than 1 : 1 .  This indicated that the binding capacity of resin to 
99 
undesirable compounds decreased with increasing fish oil volumes, and consequently 
resulted in the refined fish oil having more undesirable compounds than refined oil obtained 
from refining with fish oil-resin volume ratio of 1 : 1 .  This finding suggested the use of fish 
oil-resin volume ratio of 1 : 1  to guarantee the organoleptical fish oil quality. 
Height and diameter sizes of the column did not apparently give any odour and taste 
differences in refined oils. All refmed oils from various height and diameter sizes of 
column had significantly better odour and taste properties compared to the fraction-2 oils. 
This revealed that in terms of odour and taste properties the effectiveness of macroporous 
strong acid cation resin did not change with the changes in height and diameter sizes of the 
column when the fish oil-resin volume ratio of 1 : 1  was used. 
Differences in odour and taste properties of refined fish oils obtained from the refining with 
and without vacuum pressure application were insignificant. Thus this study proves that the 
application of vacuum pressure in the resin refining process as an effective method of 
increasing the refining rate, where the refining process rate could be increased to more than 
300%. 
5.6. CONCLUSIONS 
The above results indicate that the use of the resin packed column to improve the chemical, 
physical and organoleptical qualities of fish oil, to meet required human consumption 
standard, is valid. The most important finding is that the (1}-3 fatty acids could be retained 
in the refined oil. 
In order to guarantee the quality of the refined fish oil product, the fish oil-resin volume 
ratio of 1 : 1  is recommended. Multiple refming is suggested for fish oil of inferior quality, 
particularly in terms of odour and taste quality. In general terms, the height and diameter 
sizes of column did not show any significant different in the quality of refined fish oil 
obtained. In addition, the study showed that more attention needs to be paid to the fish oil?
resin volume ratio, rather than height-diameter ratio in installing the resin refining unit. 
Vacuum pressure application is recommended to accelerate the refining rate. 
100 
Since the natural tocopherol antioxidant value of fish oil is reduced during resin refming, 
a further study on the stability of refined and unrefined oils during storage was conducted 
and is discussed in Chapter 6. 
101 
Chapter 6 
STORAGE TEST OF REFINED AND UNREFINED FISH OILS 
6.1. BACKGROUND 
Long term stability of fish oil during storage and transport is one of the most important food 
quality and safety issues to be considered, when designing suitable storage conditions including 
temperature limits and control, container type and antioxidant use. 
Autoxidation is commonly suspected as the main chemical process in reducing fish oil quality. The 
principle sites of attack by oxygen during the oxidative process are the unsaturated portions of the 
fatty acid moieties within triglycerides. Oxidation of saturated fatty acyl groups occurs too slowly 
to have any significant effect on fish oil quality. In general, fish oil is highly susceptible to 
autoxidation because of the high proportion of unsaturated fatty acids, especially those with five 
and six double bonds, such as eicosapentaenoic acid (20: 5m3) and docosahexaenoic acid (22: 6m3) 
which are very labile to oxidation (Lundberg, 1965; Kinsella, 1987; Li and Regenstein, 1990; 
Fujimoto et ill, 1990). 
The autoxidation of oils causes rancidity and the development of unpleasant flavours and odours 
in food. Rancidity can not be detected at an early stage of oxidation, because the small molecular 
weight compounds formed provides very little off-odour and off-flavour (Labuza, 1971). 
Among the established diagnostic procedures, determination of the peroxide value (PV) is 
commonly used to monitor the extent of oxidative rancidity (Quast and Karel, 1971;  Me Water, 
1971;  St. Angelo, 1977). The index provides an early warning of staleness and rancidity 
development as a result of peroxidation during storage (Read et ill, 1988; Robert et ill, 1988; 
Wallerstein et al, 1989). Hydroperoxides are generally known as primary products of lipid 
oxidation. Therefore, it seems reasonable to determine the concentration of peroxide as a measure 
of the extent of oxidation (Grey, 1978; Jackson, 1981). 
6.2. OBJECTIVES 
The objectives of this experiment were: 
* to evaluate the relative stability level of resin refined fish oil in comparison with 
unrefined oil during storage; and 
* to develop a predictive shelf life equation as the function of storage temperature. 
6.3. METHODOLOGY 
6.3.1.  Materials 
102 
Crude oil was used in this study. The refmed oil was obtained by passing the oil through a resin 
column 2.6 cm in diameter and 39 cm in height, with the fish oil-resin volume ratio 1 : 1. The flow 
rate of the oil was 2.2 ml/minute and refining was conducted at ambient temperature (18-23?C). 
6.3.2. Methods 
To test the stability of refmed and unrefmed oils during storage, 22 mi samples were held in 30 
ml polypropylene vials fitted with air tight lids. 
To investigate the fish oil stability at several temperatures, the oils were stored in temperature 
controlled rooms at 2, 20, 30 and 40?C. Samples were kept in the dark to minimize light inducing 
oxidation. The relative humidity of the storage rooms was not controlled. Samples were 
withdrawn after 0, 2, 5, 10 and 15 weeks of storage. Chemical, physical and sensory 
measurements for the samples stored at 40?C were performed until week 15. The experiment was 
conducted with two replications, as was the analysis. Seven trained Indonesian panellists 
participated in the sensory analysis. The samples were served in two ways: one at ambient 
temperature ("cold") and the other warmed to 55?C. The sensory sheet used is shown in Appendix 
6.1.  
103 
6.3.3. Determination of the deterioration rate of fish oil during storage 
The rate of reaction causing quality deterioration in fish oil, was determined by applying zero and 
first-order mathematical models. The rational behind this determination is based on reports by 
Sagui and Karel (1980) and Labuza (1982) who noted that quality deterioration of foods generally 
follow either zero- or first-order models depending on the mode of deterioration involved. 
A zero order reaction is described as : 
C = C0 ? ?t . . . . . . . . . . . . . . . . . . . .. 6. 1 .  
and, for a first-order reaction: 
In C = In C0 ? k1 t .. ... . ... . . . . . . .  6.2. 
Where: C0 = concentration of the quality factor at zero time 
C = concentration of quality factor at time t 
t = time (weeks) 
? = zero-order rate constant (concentration/time) 
k1 = first-order rate constant (time.1) 
The order of reaction was selected based on the goodness of fit of data as measured by the 
coefficient of determination (r). 
6.4. RESULTS 
6.4.1. Effects of storage on peroxide value (PV) of fiSh oil 
The PV of the refined and unrefmed fish oils during storage at various tested temperatures are 
shown in Figure 6.1 .  In general, the PV registered a progressive increase during storage at all 
temperatures. 
c. 40 -"' 
..... 35 0" 
D ?- 30 
D 25 
.=! .. 
> 20 
? 15  
";( 
? 10 
D 
c.. 
? 1 2  1 6  
S torage time (Weeks] 
IAI 
c. -"' ..... 0" D ? 
D 
" 
""ii 
> 
D "0 
";( 
? D 
c.. 
20 
104 
45 
40 
35 
30 
25 
20 
15  
1 0  
0 
0 12  16 20 
S torage Time (Week,) 
IBI 
Figure 6.1 .  Peroxide value changes in fish oil during storage at various temperature (A = refined 
fish oil; B = unrefined fish oil; ?= 2?C storage; a=  20?C storage; .A. =  30 oc storage; 
* = 40"C storage) . 
. Analysis of variance shows a significant difference in the PV due to the storage temperature. The 
PV increases with increasing temperature. This trend was most obvious for the refined oil, where 
40?C of storage temperature gave a significantly fastest rate of PV increase which declined under 
0 
30, 20 and 2?C storage temperatures. The storage temperatures of 20, 30, and 400C applied to the 
unrefined oil did not show any pronounced difference in the rate of PV increase, but the rate was 
apparently faster than in the unrefined oil stored at 2?C. A very sharp increase in PV, after two 
weeks, was noted in samples from all storage temperatures, except for the oils stored at 2?C, where 
the PV increase was gradual. 
6.4.2. Effects of storage on refractive index value of fish oil 
Figure 6.2. shows the RI value changes found for refined and unrefined fish oils during storage at 
several storage temperatures. 
105 
1.HS 1 .H5 
0 ? 
"' "' 
N N 
? 1 ...170 ? 1 .470 
)( )( 
Cl "' 
"0 "0 
-= -= 
Cl " ,. ,. U65 U65 
u u .. .. 
.. " 
a: a: 
1.?6 +----.,,-------.---,---,----, 
0 12  16  20 
Storage Time (Weeks) Storage Time (Weeks) 
fA) IBl 
Figure 6.2. Refractive index changes in fish oil during storage at various temperatures (A = refined 
fish oil; B = unrefined fish oil; ?= 2?C storage; ?= 20?C storage; A.=  30?C storage; 
* = 400C storage) 
The most important observation was that RI values for the refined oil at all temperatures did not 
exceed the values observed for the unrefined oils. During the first five weeks, the RI values 
showed a small increase, except for the oil stored at 40?C. Changes in RI due to the various 
storage temperatures could be distinguished, especially in the refined fish oil. At a storage 
were recorded over the trial. In contrast, the RI 
106 
6.4.3. Effects of storage on colour of fiSh oil 
Colour intensity changes in both refined and unrefined fish oils are shown in Figure 6.3. 
"E 2.10 "E 2.10 c c 
1.84 0 1.84 0 
"' "' 
..,. 1.57 ..,. 1.57 
? ? 
1.31  ., 1 .3 1  ., u u c 1 .05 
c 1 .05 ., ., 
-e -e 0 O.H 0 0.79 ? .. 
.0 .0 
? 0.52 ? 0.52 
? .... "' "' 0.26 0 0.26 ? 
0 0 
(.) 0 (.) 0 
0 1 2  16  20 0 ? 12  16  20 
S torage Time [Weeks} Storage Time [Weeks} 
IAl !Bl 
Figure 6.3. Colour absorbance value changes in fish oil during storage at various temperature (A= 
refined oil; B = unrefined oil; ? = 2?C storage; ?= 20?C storage; A. =  30?C storage; 
* = 400C storage) 
The results indicated that colour intensity reflected by absorbance values measured at 490nm was 
significantly affected by storage temperatures. The sharpest reduction in absorbance was observed 
in the oils stored at 40?C, followed by the oils stored at 30?C and 20?C. Again the lowest storage 
temperature (2?C) effectively minimized changes in oil colour over 20 weeks. 
The rate of absorbance decrease in the refmed oil was significantly higher than found for the 
unrefined oil. 
107 
6.4.4. Effects of storage on sensory properties of fiSh oil 
The panel data obtained from the sensory evaluation of oil odour and taste of. refmed and 
unrefined oils are shown in Figures 6.4 and 6.5. The mean score for all the sensory attributes 
tested increased with time. The increase was more noticeable in samples stored at 40"C. 
Results showed that changes in the odour and taste scores for cold and warm samples exhibited 
a similar trend. However the sensory scores of cold samples tended to be higher than the scores 
of warm samples. 
The effects of storage temperatures on the development of rancid odour and taste as perceived by 
panellists were more significant at higher temperatures. The panellists observed that during the ftrst 
five weeks of storage odour and taste scores of oils stored at all tested temperatures increased 
sharply. After this period each storage temperature exhibited a different pattern. The scores for 
oils stored at 2?C and 20?C increased gradually with five weeks of storage. The odour and taste 
scores of the oils stored at 30?C and 40?C increase sharply at the tenth week. After that period 
there was a slow increase. The increase in odour and taste scores for both the refined and the 
unrefined oils stored at 40?C was significantly more rapid in comparison oils stored at other 
temperatures. 
The development pattern of rancid odour and taste scores showed similar trends for both oils. 
However the refmed oil generally showed a slightly higher increase rate in odour and taste scores. 
Results also indicated that the increase in odour score was at a higher rate than the increase of 
taste. This trend was observed at all storage temperatures. 
H 
? 0 ..( 0 u -o 
Ul "' c 
'- ;,;:: "' "' 0 a: 
"0 
0 -o 2 0 
? 
0 +-----.-----.------r-----r----? 0 1 2  1 8  20 
Storage Time (Weeks) 
IAI 
S torage Time (Weeks) 
(C) 
108 
0 +---?r----r----?----r---? 
-;; 5 
0 
? "'0 .. 0 ., u c 
Cl) :;:  
'- ? 3 "' c 
0 :::> "0 0 ? 2 
.. 
? 1 
0 1 2  1 6  20 
Storage Time {Weeks) 
(8) 
* 
0 +-----.----.-----.-----.----. 0 1 2  2 0  
Storage Time (Weeks) 
IDI 
Figure 6.4. Odour score changes in fish oil during storage at various temperatures (A= cold refined 
oil; B=  warm refined oil; C= cold unrefined oil; D= warm refined oil; ? = 2?C storage; 
?= 20?C storage; ?= 30?C storage;-?-=  40?C storage) 
M 
? 0 .( 
0 -o u .. 
"' <: 3 ., :.= 
- " .. a: 
.. 
1- -o 2 a 
? 
o +---?----,---?r----.----. 
5 
.. 
" 0  ? 
0 -o u " 
Cl) ?= - 3 
" " 
.. <: " ::::J 
1- -o 2 
a 
? 1 
0 
0 
12  1 5  20 
Storage Time (Weeks) 
(A) 
1 2  1 5  20 
Storage Time (Weeks) 
!Cl 
? 0 ? 
0 -o 
u " Vl :_?  
? ? 
.:: ? 2 
"' 
? 
0 ?----?----.-----.-----.-----. 
.. 
0 " ? ? "'0 0 ., 
0 c 
(11 ;: 
C,l ? 3 
c 
::: ::::J 
1- E 2 
"' 
? 1 
0 
0 
12 1 5  20  
Storage Time (Weeks) 
(8) 
1 2  1 6  20 
S torage Time (Weeks) 
ID) 
109 
Figure 6.5. Taste score changes in fish oil during storage at various temperatures (A = cold refined 
oil; B = warm refined oil; C = cold unrefined oil; D = warm unrefined oil; ? = 2?C 
storage; a= 20?C storage; .A = 30?C storage; * = 40?C storage) 
1 10 
6.5. DETERMINATION OF RATE CONSTANTS AND ORDER REACTION MODEL 
Two models were evaluated, one for a zero-order reaction and one for frrst-order reac_tion. The rate 
constant for refractive index changes were not determined, since the changes were practically 
insignificant. Calculated rate constants for both zero- and frrst-order reactions are shown in Table 
6.1 .  These results show that in all cases, the rate constant is a function of temperature. 
The rates of peroxide value, odour and taste scores changes in both oils were found to follow a 
zero-order reaction rather than the frrst-order reaction. The extent of colour loss in both oils 
followed both zero- and frrst-order reaction models, in that both models had high R square values. 
To predict the quality losses in fish oils at various temperatures, plots of the natural logarithm of 
rate constant versus reciprocal of absolute temperature COK) for each quality parameter were 
calculated. These are shown in Appendix 6.3. The trend of actual value was calculated using 
linear regression. 
1 1 1  
Table 6.1. Rate constant of zero- and ftrst-order reactions of each parameter during storage of ftsh 
oil at various storage temperatures 
Fish Oil Storage Zero Order First Order 
Parameter Sample Temp.("C) 
k r C%) k rC%) 
2 0.94 98.05 0.14 9 1 .58 
Refined 20 2.15  96.82 0.27 78.09 
Peroxide Oil 30 2.54 97.90 0.29 77.40 
Value 40 3 . 13  94.34 0.35 72.83 
2 0.84 98.95 0.12  93.41 
Unrefined 20 1 .56 96.34 0.19 80.73 
Oil 30 1 .65 93.89 0.20 77.36 
40 1 .82 86.46 0.23 70.42 
2 0.010  99.00 0.005 99.58 
Colour Refined 20 0.037 98.60 0.022 97.73 
Absorbance Oil 30 0.054 99.83 0.032 98.00 
Value 40 0.093 99.54 0.062 99.1 4  
2 0.010  68.28 0.003 62.64 
Unrefined 20 0.020 96.22 0.010 96.41 
Oil 30 0.028 98.65 0.0 14  98.37 
40 0.050 98.51 0.026 96.82 
2 0.1 3 1  85.49 0.086 78.03 
Refined 20 0.2 16  78.69 0.12 1  7 1 .70 
Odour Oil 30 0.25 1 86.95 0.1 44 79.97 
Score 40 0.296 91 .94 0.159 86.93 
from 2 0.095 92.1 2  0.056 87.38 
Cold Unrefined 20 0.1 36 9 1 .72 0.071 84.14 
Sample Oil 30 0.169 89.03 0.087 83.90 
40 0.252 93.23 0.095 9 1 .08 
2 0.126 69.33 0.068 67.1 7  
Refined 20 0.191 86.25 0.1 18 77.24 
Odour Oil 30 0.229 92.1 7  0.1 34 84.73 
Score 40 0.292 95.36 0.157 88.40 
from 2 0.079 85.94 0.044 84. 17  
Warm Unrefined 20 0.1 1 4  94.25 0.059 87.36 
Sample Oil 30 0. 150 93.78 0.070 86.97 
40 0.228 93.37 0.102 85.65 
2 0.1 12 92.75 0.088 83.75 
Refined 20 0.172 78.74 0. 1 18 7 1 .9 1  
Taste Oil 30 0.214  92.58 0.130 83.84 
Score 40 0.268 94.58 0.150 89.74 
from 2 0.098 80.70 0.062 73.3 1 
Cold Unrefined 20 0.1 19 93.28 0.070 85.68 
Sample Oil 30 0.160 93.33 0.084 89.97 
40 0.225 93.20 0.108 90.33 
2 0.106 92.01 0.082 82.86 
Refined 20 0.128 9 1.59 0.088 84.45 
Taste Oil 30 0.170 90.09 0.1 15  84.1 2  
Score 40 0.228 92.88 0.129 86. 1 1  
from 2 0.100 91 . 15  0.1 04 83.09 
Warm Unrefined 20 0.157 93.05 0.126 84. 1 8  
Sample Oil 30 0.198 93.42 0.148 83.91 
40 0.295 97.93 0.208 89.78 
Note : Units used for each rate constant are as follows: 
A. Peroxide Value: 1 .  zero order reaction: (meq/kg)/week 
2. first order reaction : week1 
B .  Colour absorbance value: 
1. zero order reaction: abs/week 
2. first order reaction: week?1 
C. Sensory scores: 1. zero order reaction: score/week 
2. first order reaction: week?1 
6.6. ESTIMATION OF SHELF LIFE OF FISH OIL 
1 12 
For information about the shelf life of a product, especially for a product developed using a new 
process, it is necessary to demonstrate the strengths and the weaknesses of the process. The 
information obtained will be used for further required action such as the need of antioxidant 
addition, and the use of a special packaging. , 
Sensory parameters can be regarded as the primary determinant of fish oil shelf life, since these 
parameters have a direct relationship to consumer acceptability of the product. Two sensory 
parameters, odour and taste, observed from cold and warm samples, were used to estimate, and to 
establish a model for predicting shelf life. 
The shelf life of both oils at each storage temperature was determined using odour and taste 
sensory parameters calculated using the equation: C = C0 ? kat. Since not all samples reached a 
quality reject point, the reject point was set when odour and taste scores reached 5. The results 
of this calculation are shown in Table 6.2. 
1 1 3  
Table 6.2. Calculated shelf life of refined and unrefmed fish oil based on the odour and taste 
parameters from various storage temperatures (weeks) 
Fish Storage Odour Taste 
Oil Temp. ec) 
Sample Cold*l Warm .. l Cold Warm 
2 3 1  3 1  36 39 
Refined 20 19 21 24 32 
30 16 17 19 24 
40 14 14 15  18  
2 39 43 39 38 
Unrefined 20 27 30 32 27 
30 22 23 24 22 
40 15  1 5  17 15 
Note: "') sample was evaluated at ambient temperature 
**) sample was warmed up to 55?C 
Calculation shows that storage temperature is the primary determinant of shelf life for both oils: 
the higher the storage temperature, the shorter the oil shelf life. In general, the shelf life of 
unrefined oils was longer than the shelf life of refined oil, except where the shelf life was 
calculated from taste scores for warm samples. Shelf life calculated from the odour scores for cold 
samples gave the lowest calculated shelf life. This suggests that the odour evaluated from cold 
samples should be used as the parameter determining the shelf life of fish oil and in establishing 
the equation for shelf life prediction. 
- 1 .0 
- 1 .3 
- 1 .6 
? 
c 
- 1 .9 
-2.2 
-2.5 
In  k = 4.62 - 1 82 1 {  1 /T) ;  / r2 = 98.73% 
? 
? Re f ined  Oil 
? Unre f i ned  Oi l  
? In k = 5. 1 5 - 2077 ( 1 /T);  
r2 = 95.6 1 %  
? 
? 
0.0030 0 .0032  0.0034  0.0036 0.0038  0.0040 
1 /T ( 'K )  
1 14 
Figure 6.6. Linear relationship between the natural logaritlun of rate constant of fish oil and the 
reciprocal of absolute temperature 
Plots for the natural logaritlun of rate constant for odour score changes observed for the cold oil 
samples versus the reciprocal of absolute temperature eK) are shown in Figure 6.6. The linear 
equation used to estimate the rate constant at various storage temperatures obtained from these plots 
for the refined oil samples is: 
In k = 4.62 - 1821 1!f ; r = 98.77% 
and, for unrefined oil samples: 
In k = 5.15 - 2077 1!f ; r = 95.61% 
1 1 5  
The estimated shelf life calculated from the above equations is shown in Table 6.3. 
Table 6.3. Estimated shelf life of fish oil at various storage temperatures (weeks) 
Fish Oil Sample Storage Temperature COC) Shelf Life (weeks) 
2 30 
Refined 20 20 
Oil 30 16 
40 13  
2 41 
Unrefined 20 26 
Oil 30 20 
40 16 
By plotting the natural logarithm of shelf life (9) versus the reciprocal of absolute temperature as 
shown in Figure 6.7, the linear equation for the prediction of shelf life for refmed oil is: 
In 9 = -3.36 + 1860 1ff ; r = 99.91% 
And the unrefmed oil: 
In 9 = -3.93 + 2102 1ff ; r = 99.88% 
4.0 
3.7 In e = -3.93 + 2 1 02 ( 1 /T) ;  rz = 99 .88%  ? 
3.4 
<D ? Re fi ned  Oi l  
c: ? = 
3 . 1  
In e -3.36 + 1 860 ( 1 /Tl :  
2.8 r2 = 99.9 1 %  
2.5 
0.0030 0.0032 0.0034 0.0036 0.0038 0.0040  
1 /T ( 'K )  
Figure 6.7. Linear relationship between the natural logarithm of estimated shelf life and the 
reciprocal of absolute temperature 
1 16 
6.7. CORRELATION BETWEEN SENSORY RESULTS AND OTHER PARAMETERS 
Investigating the correlation between sensory results and other parameters was designed to show 
whether the changes in sensory properties correlate with the changes in peroxide, colour absorbance 
and refractive index values. 
Odour scores observed from cold samples were selected to represent sensory results, because this 
evaluation was identified as the most important parameter in determining fish oil shelf life. A 
regression analysis was used to reveal the relationship between odour score and other parameters. 
The results, as shown in Table 6.4., indicate that relations between odour and peroxide value, as 
well as colour absorbance value, were found to be linear, but, in general, peroxide value showed 
a better correlation. Not all results at different storage temperatures indicated a linear relationship 
between odour score and refractive index value. In fact, a linear relationship was only found from 
data in which the oil  was stored at 40?C 
1 17 
Table 6.4. Regression analysis between odour score and other parameters (peroxide value, colour 
absorbance value and refractive index value) 
Linear Equation Fish Oil Storage a b r 
For: Sample Temp.eC) 
(%} 
2 0.79 0.1 1  86.26 
Odour Score (OS) Refined 20 0.70 0.09 78.98 
Versus Oil 30 0.63 0.11  89.41 
Peroxide Value(PV) 40 -0.03 0.15 93.77 
2 0.37 0.14 92.73 
OS = a +  b(PV) Unrefined 20 0.46 0.14 67.88 
Oil 30 -0.004 0.18 86.28 
40 -0.77 0.30 82.22 
Odour Score (OS) 2 23.06 - 1 1 .84 91 .28 
Versus Refined 20 8.91 -4.13 73.55 
Colour Absorbance Oil 30 8.17 -3.69 85.09 
Value (CAV) 40 8.03 -3.86 93.94 
2 31 .35 -14.72 65.48 
OS = a + b (CA V) Unrefined 20 19.07 -8.74 80.38 
Oil 30 15.67 -7.07 82.08 
40 13.81 -6.19 85.90 
Odour Score (OS) 2 -1918.53 1308.96 40.00 
Versus Refined 20 -1306.28 891 .78 35.61 
Refractive Index Oil 30 -2262.85 1543.66 77.79 
(RI) 40 -3088.3 1 2105.67 95.88 
2 -1 1593.51 7877.17 74.59 
OS = a +  b (RI) Unrefined 20 -3479.98 2365.30 65.36 
Oil 30 -3862.68 2625.21 65.39 
40 -6055.96 41 14.72 92.63 
6.8. DISCUSSION 
6.8.1. Chemical and physical changes in fiSh oil during storage 
The progressive formation of peroxides in fish oil during storage was evident from the increase in 
peroxide value (PV) in refined and unrefmed oils. This phenomenon in the oxidation process has 
also been observed in the storage of some edible oils (Arumugham et al, 1984), in groundnut oil 
(Narasimban et.2!,1986), in canola oil (Hawrysh et.2!, 1989) and in menhaden/soybean oil blends 
(Schnepf et al, 1991). In the production of hydroperoxides, the free radicals formed in the 
1 18 
initiation stage of oxidation react with oxygen to give a peroxy radical, which in turns react with 
the substrate to give a hydroperoxide (Hardy, 1980). 
Natural colour loss in fish oil during storage resulted in oil with .pale and clear colour. Carotenoids 
are the most common pigment composing fish oil colour. The carotenoids, embracing the red, 
orange and yellow oil soluble pigments occur naturally in a number of different plant and animal 
fats (Brody, 1965; Clydesdale and Francis, 1976). The decomposition of carotenoids would result 
in loss of colour and form a more weakly coloured product (Chou and Breane, 1972; Emods, 
1978). The unsaturated portion of carotenoids is easily affected by oxygen (Stuckey, 1972). Thus, 
the main cause of carotenoids degradation in foods is oxidation by reaction with atmospheric 
oxygen at rates dependent on light, heat and the presence of pro- and antioxidant (Clydesdale and 
Francis, 1976). Enzymatic degradation of carotenoids was also noted (Simpson et al, 1981;  
Baurnfeind, 1981). 
An increasing pattern in refractive index (RI) was observed in fish oil during storage, and a similar 
occurrence is recorded by Janick and Pokomy (1960) and Arya et al (1969) in edible oils and fats. 
The increase in RI with autoxidation is possibly attributable to conjugation known to pr?e?l 
hydroperoxide formation in the secondary stage and polymerization of partially oxidized fats in the 
tertiary stage of autoxidation (Arya et al, 1969; Grey, 1978). 
The changes in refmed oil tended to occur at higher rate than in unrefmed oil. The natural 
antioxidant lost during refming process, as mentioned in Chapter 5, was suspected as the main 
reason for this. Lower antioxidant content would lead to lesser protection capacity against 
oxidation attack. Therefore, peroxide formation, colour loss and RI increase were encountered 
faster in refined oil than in unrefmed oil. Brezka and Salmonowicz (1973) report a similar result 
when investigating the stability of soda lye refined cod liver oil and sprat oils during incubation 
at 95 - 1 oooc, in which the refined oil had a lesser resistance to oxidation compared to unrefined 
oil. The tocopherol content of these refmed oils was also found to be lower than unrefined oil. 
The storage temperature affected peroxide formation, where the higher temperature tended to give 
a higher rate of peroxide formation. Unrefined oils stored at 20, 30 and 400C showed an 
insignificant difference in peroxide value during storage period. Decomposition of peroxides into 
secondary products may have affected the total peroxide value. Previous studies have shown that 
the conversion rate of peroxides into secondary products increased with increasing temperature 
(Lea, 1962; Erickson and Bower, 1976; Karel, 1985). Possibly, under these conditions the peroxide 
formation rate in unrefined oil was nearly the same as the conversion rate into the secondary 
1 19 
products. The higher storage temperature also accelerated the colour loss and RI increase. Those 
facts are also supported by the increasing pattern of rate constant with increasing storage 
temperature. 
As described above, the changes in fish oil properties still occurred even though the oils were 
stored at a low temperature (2?C). This means that oxidation still occurred at low temperatures. 
According to Gray (1978) lowering the temperature is not particularly helpful in preventing 
oxidation, since a low energy threshold is involved. The results indicated that the oxidation 
process, especially peroxide formation, could still occur with the oxygen available even at low 
temperature. However the low temperature storage was still very effective in slowing down the 
peroxide formation, colour loss and RI increase. This experiment also proved that enzymatic 
degradation on carotenoids bad a lesser role in the decolouration in fish oil, since colour change 
still occurred at low temperature (2?C). 
6.8.2. Sensory changes in fish oil during storage 
Sensory evaluation was an important means in determining the acceptability of fish oil during 
storage, since, according to Gray (1978), the consumer uses an organoleptic evaluation to judge the 
quality of fats and oils . 
. Odour and taste scores observed from both cold and warm fish oil samples tended to increase 
during the storage testing period. This means that more rancid odour and taste was detected by 
panellists with increasing storage time. Similar results were obtained by Poble et al (1964) in fats 
and shortening, Cowan et al (1971) in blends of soybean and peanut oils, and Narasimban et al 
(1986) in groundnut oil. The objectionable flavour resulted from the accumulation of 
decomposition products of the oxidation reaction (Gray, 1978), including carbonyl compounds, 
alcobols, semi-aldehydes, acids, hydrocarbons, lactones and esters (Lilliard, 1978). Most important 
are the aldehydes, several unsaturated ketones and from derivates which exhibit a wide variety of 
aroma properties (Groscb, 1987). The odour in rancid sardine oil was found to be mainly due to 
IQ ?'ar w?igh;_tw0lattltiWC . u?ds,
_
pre??tl}: e)hano prqpi@1li ?Qi:'d, ?m ? ? :toi, ( 
ethanol, acetic acid, butanol, pentanal and similar substances, (Nakamura et&, 1980). Moreover, 
Karabadian and Lindsay (1989) found that rancid and "paint" flavour in fish oils were due to 
hexanal, t,c- and t,t-2,4-heptadienals and t,c-2,4-decadienal. 
120 
The increase of sensory scores was affected by storage temperatures, the higher the storage 
temperatures the faster the score increased. This means that a high temperature accelerated the 
formation of rancid odour and taste compounds. According to panellists, the developn:tent of rancid 
odour and taste was effectively inhibited when the oil was stored at a low temperature (2?C). This 
indicated that the high temperature would accelerate the rancid odour and taste compounds. These 
results were supported by the rate constant, in which the rate constant for these parameters 
increased by elevating storage temperatures. 
may be affected by the decrease in quantity of natural antioxidants amount. As discussed in 
wllich..pmbabl?od!ii'Y....\;'!J. 
protection agains? oxidation proees 
6.8.3. Shelf life of fish oil 
In this study, refined and unrefined oils investigated for their shelf life were contained in 
polypropylene vials. Sensory results were used to predict the shelf life of fish oil since, according 
to Jackson (1981) and Erickson and Bower (1976), whatever objective method was chosen it must 
be realized that the method must be correlated with some type of organoleptic evaluation of the 
.food. Odour score observed from cold samples was chosen to predict the shelf life of both refined 
and unrefined oils, because in general, the score resulted in shortest estimated shelf life. 
The average temperature of Indonesia is approximately 30?C (Pearce and Smith, 1984). If the oil 
stored in polypropylene containers at 30"C or below, it can be kept in acceptable quality for 16 -
30 weeks for refmed oil and 20 - 41 weeks for unrefined oil. Shorter shelf life of refined oil was 
due to a higher rate of deterioration, especially in the oxidation process. 
121 
6.9. CONCLUSIONS 
In conclusion, the results of this study show that the prediction of fish oil shelf life could be 
reasonably obtained from the changes in observed parameters which could be collected from a 
planned storage test. The shelf life of fish oil was limited by the sensory parameter, but not by 
chemical and physical parameters. The rate of fish oil deterioration was slowed by keeping the oil 
at a low temperature (2?C). At the storage temperature of 30"C in Indonesia, refined and unrefined 
fish oils bad a shelf life up to 30 and 41 weeks respectively. 
The results indicated that treatments were needed to extent the shelf life of refmed fish oil. The 
treatments recommended are antioxidant addition to replace lost natural antioxidant during refining, 
and oxygen removal. 
122 
Chapter 7 
STABILITY IMPROVEMENT OF RESIN REFINED FISH OIL 
7.1. BACKGROUND 
The resin refining process did not produce refined oil with either equal or better stability than 
unrefined oil as discussed in Chapter 6. Lower natural antioxidant in refined fish oil was suspected 
as the main reason of this occurrence. Chapter 5 illustrated how refming had decreased a?
tocopherol, which is known as a predominant antioxidant in fish oil (Karahadian and Lindsay, 
1986; Brody, 1965). 
7.1.1. Antioxidant and oxidation 
The addition of antioxidants is necessary to improve the stability of refmed oil, since antioxidants 
are able to retard deterioration, rancidity, and discolouration due to oxidation (Dziezak, 1986). 
More specifically, the antioxidants are used to protect fats and oils from developing rancid flavour 
and odour (Haumann, 1990). Many studies, including Pokomy (1987}, provide a detailed 
-description of the inhibiting . mechanism of oxidation by antioxidant. The most widely used 
rtrdioais. By? eaeti:on 
with an alkyl p?roxyl (hydroperoxide) radical, antioxidants fo? a lipid hydroperoxide, an a less 
a ".tive. intlibi?r radical: / 
ROO + InH ----> ROOH + In? 
Alkoxyl radicals can participate in an analogous way: 
RCF + InH ----> ROH + In? 
Various side reactions may proceed simultaneously, resulting in chain initiation and thus increasing 
the reaction rate. For example: 
In? + 02 ----> InOO? 
InOQ- + R-H ----> InOOH + R? 
Where In? radicals are very unstable, the following reaction is possible: 
InH + 02 ----> In + HOO? 
123 
Reaction becomes important only in high concentrations of antioxidants. In rare cases the 
antioxidant molecule can react with lipid hydroperoxides forming two free radicals: 
InH + ROOH ----> RO? + H20 + In? 
ROO-In ----> RO' + In? 
These reactions are unimportant unless antioxidant concentration is very high. In some cases the 
antioxidant may react with R radicals as well: 
InH + R? ----> In? + RH 
With phenolic antioxidants, this reaction is insignificant only when oxygen is present in traces and 
hydroperoxide concentration is also negligible (Pokorny, 1987). 
Antioxidants also minimize the oxidative destruction of certain vitamins and essential amino acids. 
Most useful are antioxidants that are soluble in fats and oils, odourless, tasteless, nontoxic at 
approved levels and effective in low concentration (Haumann, 1990). 
Antioxidants are usually aromatic compounds which are phenolic in character. Phenolic 
antioxidants permitted in edible fats and oils in many countries include tocopherol, propyl gallate 
(PG), butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT) and ter-butylhydroquinone 
(TBHQ). Permitted levels vary from country to country. The US Food and Drug Administration 
(FDA) regulations generally allow PG, BHA, BHT and TBHQ, or a combination, to be used at a 
level not exceeding 0.02% based on the weight of the fat and oil. Tocopherol and ascorbyl 
palmitate, which are used under good manufacturing practices (GMP), do not have a regulated limit 
(Haumann, 1990). BHA is the most common antioxidant used in the food industry in Indonesia 
(Sherwin, 1990). 
Very few research papers have been published regarding antioxidants used in fish oil, but intensive 
experiments have been conducted in vegetable oils. However some experiments on the use of 
124 
antioxidants in fish oil have been canied out by Jurewicz and Salmonowicz (1973), Brzeska and 
Salmonowicz (1973), Ke et al (1977) Zama et al (1979), Karahadian and Lindsay (1988) and 
Taguchi et al (1988). 
7.1.2. Oxygen removal and oxidation 
Oxidation of olefmic compounds by atmospheric oxygen is important in the development of 
rancidity, in the production of desirable and undesirable flavours, in the polymerization of highly 
unsaturated oils, and in the production of compounds of significant physiological activity (Gunstone 
and Norris, 1983). When the effect of oxygen on oxidation of lipids is considered, it is often 
simply a question of total amount available for reaction with food components. If this amount is 
limited to a level which causes no significant effect on the food, and there is no potential for 
additional oxygen coming into contact with food, then the rate of reaction is irrelevant Karel, 
1985). Thus, the most obvious precaution to take against oxidation deterioration is to remove the 
oxygen (Smouse, 1978; Belitz and Grosch, 1987; Ranken, 1989; Ranken, 1990; Hardy, 1980; 
Berger, 1989; Brookman, 1991). 
Recently, vacuum packaging has been used as a method of oxidation control. From mechanistic 
considerations based on the rancidity of the polyenoic acids present in the oil, the oxidation rate 
will be dependent on oxygen concentrations. Even if oxygen diffuses through into the pack the 
rate will be !l:educeq ?ro:etr r<Y' n.ofi-Yauu? pae M eil (Naf<ly, 1980). 
7 .2. OBJECTIVES 
1 .  The antioxidant experiments performed were intended to determine the best antioxidant to inhibit 
the oxidation process in fish oil, and to optimize its level of use. 
2. This study also aimed to investigate an alternative way of vacuum packaging for stability 
improvement instead of the use of an antioxidant. Consumers are increasingly reluctant to accept 
synthetically derived additives in foods because of concerns about health (Clark et l!!, 1990; Han 
et l!!, 1991). 
125 
7 .3. METHODOLOGY 
7 .3.1. Materials 
Crude fish oil was used as the object of the study. Antioxidants used were BHA, TBHQ, Grindox 
1 17 (propyl gallate) and Dl-a-tocopherol. BHA was supplied by Serva Feinbiochemica, 
Heidelberg, for Experiment 1 and by Bronson and Jacobs, New Zealand, for Experiment 2. TBHQ 
was obtained from Aldrich Chemical Company, Inc., USA. Grindox 117 containing 7.5% ascorbyl 
palmitate (E304), 17.5% propyl gallate (E310), 10% citric acid (E330), 30% propylene glycol and 
35% emulsifier (E471) was supplied by Bronson and Jacobs (New Zealand). Dl-a-tocophero1 was 
donated by Roche, New Zealand. 
7 .3.2. Methods 
7.3.2.1. Experiment 1 :  Antioxidant selection 
Antioxidant solutions were made by heating 20ml of refined crude oil to 80?C and then appropriate 
amounts of antioxidant were incorporated (Hawrysh et al, 1990). The amount of each antioxidant 
dissolved was 20% (w/w) of stock solution, and mixed for 30 minutes at 80?C. 
Four antioxidant agents were tested in this experiment: three of them, BHA, propyl gallate/Grindox 
1 17 and a-tocopherol, are permitted in the manufacture of Indonesian foods (Indonesian Health 
Ministry, 1974), while TBHQ is recognized as highly effective antioxidant. 0.02% of BHA and 
0.02% of TBHQ, the highest levels permitted in oils by the Indonesian Health Ministry (1974) and 
US-FDA (Haumann, 1990), were added. Grindox 1 17 used in the experiment was 0.1%  which is 
the highest level use suggested by the manufacturer. 0.1% of tocopherol was added to refined oil, 
as that level was suggested by Body (1991). Tocopherol was added in two ways: by direct addition 
of tocopherol solution, and by heating preparation as performed in other antioxidants. 
The oil samples were prepared by mixing 150ml fish oil with appropriate amounts of antioxidant 
in a 500 ml bottle, and then shaken vigorously. For the stability test, 140 m1 oil sample was placed 
in 250 m1 beakers and covered loosely with aluminium foil lids. The stability test was carried out 
using the Schaal oven method (Thompson, 1966; Stuckey, 1972; Kochhar, 1988; Buck, 1991), by 
126 
placing the samples in a forced air oven at 63?2?C for up to 16 days. The temperature used in this 
method is much lower than in most other accelerated stability testing procedures. This method can 
therefore be recommended as the one having the fewest possible problems (Ragnarsson and Labuza, 
1977). Oil samples were removed for analysis after 0, 1, 2, 4, 8, 12 and 16 days of storage. 
Immediately after each storage period, 20 ml samples were taken and placed in 35 ml 
polypropylene bottles for analysis. Each sample was made with two replications and each analysis 
was conducted in duplicate. 
7 .3.2.2. Experiment 2: Optimisation of antioxidant dosage level 
BHA antioxidant from Experiment 1 was investigated to determine its optimum level. As discussed 
in Section 7.5, BHA was the antioxidant chosen. Unrefmed oil was .used as a control in order to 
compare the effectiveness of the standard level to be achieved by the refmed oils containing 
specific quantities of the antioxidant, BHA. The BHA levels used were 0.005%, 0.01 %, 0.015% 
and 0.02%. 
The fish oil samples were prepared as in Experiment 1. The Schaal oven method was used to 
observe fish oil stability, where the incubation was conducted at 63?2?C. The samples were 
withdrawn for analysis after 0, 2, 5, 9, 13 and 17 weeks of storage. The experiment was performed 
with two replications, and each analysis was conducted in duplicate. 
7 .3.2.3. Experiment 3: Use of vacuum package for fish oil stability improvement 
Both refmed and unrefmed oils (20 ml) were placed in 6.6 cm diameter and 4.6 cm high cans and 
sealed with 75 bar pressure for vacuum package, and without pressure for non-vacuum package. 
In the first investigation, the samples were observed for stability using the Schaal oven method by 
placing them in the forced air oven at 63?2?C for up to 16 days, and withdrawn for analysis after 
0, 1 ,  2, 4, 8, 12, and 16 days of storage. In the second investigation, the samples were stored at 
30?C, the average daily temperature in Indonesia. The sampling was performed after 0, 1, 2, 4, 
8, 12 and 16 weeks of storage. Both investigations were run with two replications and each 
analysis was conducted in duplicate. Sensory analysis for the second investigation was undertaken 
by eight trained Indonesian panellists. 
127 
7.4. RESULTS 
7.4.1. Selection of antioxidant 
Six parameters of value, peroxide, thiobarbituric acid, anisidine, totox, colour absorbance and 
refractive index, were used to select the best antioxidant to improve stability against oxidation, in 
refined fish oil. 
7.4.1.1. Effects of various antioxidants on peroxide value 
Changes in peroxide value (PV) during storage is shown in Figure 7 . 1 .  The addition of antioxidant 
induced changes in the initial PV of the oil. Addition of BHA, TBHQ and Grindox 1 17 resulted 
in oils with lower initial PVs than the untreated oil used as the control. However the fish oil 
containing a-tocopherol gave a higher initial PV than the control. Direct addition of tocopherol 
gave a higher initial PV than the addition of tocopherol prepared the same way as other 
antioxidants. 
1 1 0 
1 00 IJ.. Refined Oi l  (Control) - V Refined Oii+0.02%BHA C'l 90 ? Refined Oii+0.02%TBHQ ...lll:: 
....... * Refined Oii+0.1 %Grind ox 
0" 80 o Refined Oi 1+0.1 %Tocopherol Cl.) (heated preparation) E 70 "" Refined Oi i+0.1  %Tocopherol 
Cl.) (direct addi tion) 
::ll 60 
1'0 50 > 
Cl.) 40 "'0 
>< 
0 30 
... 20 Cl.) a.. 
1 0  
0 
0 2 4 6 8 1 0  1 2  1 4  1 6  
Storage Time (days) 
Figure 7 . 1 .  Effects of various antioxidants on peroxide value changes in fish oil during storage at 
63?2?C 
128 
The PV of all oil samples increased significantly after one day's storage, except for the sample 
with a direct addition of tocopherol. On the second day of storage, the oils to which tocopherol 
was added displayed an increase in PV, but other samples, including the control, had. a practically 
constant PV. TBHQ and Grindox still showed a relatively. constant PV on the fourth day. 
Conversely, the PV in the oils containing BHA and tocopherol, increased markedly. Extension of 
the storage period until 16 days revealed that all samples indicated an increasing PV oonteot, ,but 
each exhibited a different increase rate. The PV for untreated oil increased sharply on the twelfth 
day and even exceeded the PV in the other oil samples treated with antioxidants. After four days, 
the oil containing tocopherol, using both addition methods, had approximately the same rate of PV 
increase. In general, both methods of tocopherol addition did not induce any significant different 
in PV during storage. The PV in the oil containing tocopherol was always higher in comparison 
with the PV analyzed in the oils containing other antioxidants, except at the end of storage, where 
the PV in that oil was lower than in the oil containing Grindox. The PV in the oils containing 
TBHQ and Grindox increased at a comparable rate until the twelfth day, where values were lower 
than the values in other samples. After that period, the PV in the oil containing Grindox increased 
sharply and exceeded the PV in the oils containing other antioxidants. TBHQ and BHA showed 
as the most effective antioxidants to inhibit PV increase. 
7.4.1.2. Effects of various antioxidants on thiobarbituric acid (TBA) value 
The TBA test was used to measure malonaldehyde formation as a secondary product of the 
oxidation process. TBA changes in all samples during storage are shown in Figure 7.2. The 
addition of an antioxidant did not significantly affect the initial TBA values. 
Cl) 
::I 
"' > 
< CD 
1-
1 80 
1 70 
1 60 
1 50 
1 40 
1 30 
1 20 
1 1 0 
1 00 
90 
80 
70 
60 
50 
40 
t. Refined Oil (Control) 
V Refined Oii+0.02%BHA 
? Refined  Oii+0.02%TBHQ 
* Refined Oii+0.1 %Grindox 
o Refined Oii+0.1 %TocoP.herol 
(heated preparation) 
A Refined Oii+0.1 %Tocopherol 
(direct addition) 
30 l d??::Q 20 .. 
1 0  
0 +---?--?----r---.----r---.----.---, 
0 2 4 6 8 1 0  1 2  1 4  1 6  
Storage Time (days)  
129 
Figure 7 .2. Effects of various antioxidants on TBA value changes in fish oil during storage at 
63?2?C 
TBA value of the control sample and the oil containing Grindox increased significantly after one 
day of storage, but the value remained virtually constant on the second day. The TBA value of 
the oils containing BHA, TBHQ and tocopherol (heated preparation) were statistically constant 
during the first two days. The oil containing direct addition of tocopherol had a relatively constant 
value of TBA after one day, but the value increased on the second day. On the fourth day, all oil 
samples displayed a significant increase in TBA value, and those values were relatively similar, 
except for the oil containing TBHQ which had an apparently lower TBA value. 
The control sample revealed less stability compared to treated oils on the twelfth day, where its 
TBA value significantly exceeded the value of treated oils. This occurrence was evident until the 
end of storage, but the increased rate shown by each antioxidant was different. At the end of 
storage, TBA value in the oils containing TBHQ and BHA were significantly lower than the value 
in the oils containing other antioxidants. Moreover, the oil treated with TBHQ tended to show the 
130 
lowest TBA value during investigation. The two different tocopherol additions did not exhibit any 
significant difference in TBA value. 
7.4.1.3. Effects of various antioxidants on anisidine value 
Anisidine value analysis was used to determine alpha-beta aldehyde formation as a secondary 
product of oxidation. Anisidine value changes in the fish oil treated with antioxidants are shown 
in Figure 7.3. The antioxidant addition did not cause any significant difference in the initial 
anisidine values among the samples. 
1 85 
1 75 
1 66 
1 56 
1 46 
1 36 
1 27 
1 1 7  
1 07 
97 
88 
78 
68 
58 
49 
39 
29 
1 9  
.A Refined Oi l  (Control) 
v Refined Oi l+0.02%BHA 
? Refined Oi l+0.02%TBHQ * Refined Oi 1+0.1  %Grindox 
o Refined Oii+0. 1 %TocoRherol 
(heated preparation) 
o Refined Oi 1+0.1  %Tocopherol 
(direct addi tion) 
1 0  L?llY--v o ?=-?--?--?--?--?--?--?-. 
0 2 4 6 8 1 0  1 2  1 4  1 6  
Storage  Time (days)  
Figure 7.3. Effects of various antioxidants on  anisidine value changes in  fish oil during storage at 
63?2?C 
All oils did not exhibited any significant increase in anisidine value after one day of storage. The 
increase of anisidine value began on the second day, except for the oil containing TBHQ which 
still showed insignificant change. The increase of anisidine value for all samples was at 
131  
approximately the same rate until the fourth day, except for the oil treated with TBRQ. A 
significant effect of antioxidant use in terms of anisidine value was observed on the twelfth day, 
where the anisidine value in the control sample was much higher than the value in the treated 
samples. This occurrence was also found at the end of storage, where the oils containing TBRQ 
and BRA had a lower anisidine value than oils containing other antioxidants. Significantly, the 
value of the oil containing TBRQ was the lowest. The two methods of tocopherol addition did not 
exhibit any significant difference in the anisidine value. 
7.4.1.4. Effects of various antioxidants on totox value 
Totox value changes in all samples are shown in Figure 7.4. After one day of storage, the totox 
value increase could be detected only in the control sample and in the oils containing BRA, or 
tocopherol prepared by heating. TBHQ, Grindox and tocopherol added directly kept the totox value 
of oils nearly constant on the first day. After this period, totox values in all samples increased 
steadily, but at different rates, depending on the antioxidant. 
400 
A. Refined Oi l  (ControU 360 v Refined Oii+0.02XB A 
? Refined Oi1+0.02XTBHQ 
320 * Refined Oii+0.1 % Grind ox o Refined Oii+0.1 XTocoBherol 280 (heated preparation Cl,) "' Refined Oii+0.1 XTocopherol 
::J (direct addition) 
IV 240 > 
>< 200 
0 
- 1 60 0 
1- 1 20 
80  
40  
0 
0 2 4 6 8 1 0  1 2  1 4  1 6  
Storage Time (days} 
Figure 7 .4. Effects of  various antioxidants on totox value changes in  fish oil during storage at  
63?2?C 
132 
Effect of the use of antioxidant on totox value was significant on the twelfth day. The totox value 
for the control significantly exceeded the values in treated sample. The totox value in the oil 
containing TBHQ indicated the lowest totox value at all times during the period o( observation. 
At the end of storage, oil treated with BHA showed a lower totox value than the oil treated with 
Grindox and tocopherol. The two different tocopherol additions did not exhibit any significant 
different in terms of totox value. 
7.4.1.5. Effects of various antioxidants on colour absorbance value 
The colour absorbance value of all oil samples did not display significant change after one day of 
storage. However after day one, the colour absorbance value of all oils decreased significantly as 
shown in Figure 7.5. 
2.3 
2.04 
E 
r:: 1 .79 
0 
Cl 1 .53 -.::t 
.... ..... ::I CV 1 .28 0 
0 G) 
0 0 1 .02 A. Refined Oil (Control )  r:: V Refined Oii+0.02XBHA CV 
.c 0.77 ? Refined Oii+0.02XTBHQ .... * Refined Oii+0.1 XGrindox 0 
Cll 0.5 1 o Refined Oii+0.1 %Tocopherol .c 
< (heated preparation) - 0.26 "' Refined Oii+0.1 %Tocopherol  
(direct addition) 
0 
0 2 4 6 8 1 0  1 2  1 4  1 6  
Storage T ime ( days)  
Figure 7.5 .  Effects of various antioxidants on colour absorbance value changes in fish oil during 
storage at 63?2?C 
133 
The oil containing antioxidants showed a marked lower decrease rate of  colour absorbance in  
comparison to the control. This effect was apparent after eight days in which the absorbance value 
of the control sample was significantly lower than the value scanned in the treated oils. The oil 
containing TBHQ bad the slowest decrease rate of colour absorbance value followed by the oils 
containing tocopherol. The two different tocopherol additions did not exhibit any significant 
7.4.1.6. Effects of various antioxidants on refractive index (RD value 
The antioxidant affected the initial RI value of the oils as shown in Figure 7.6. 
1 .478 
1 .477 
>< 1 .476 
Q) 
"C 
c: 1 .475 
Q) 
> 1 .474 -
0 
"' 1 .473 L.. -
Q) 
a: 1 .472 
1 .47 1  
1 .47 
0 2 4 6 
11.. Refined Oil (Control) 
v Refined Oii+0.02%8HA 
<> Refined Oi1+0.02%TBHQ * Refined Oii+0.1 %Grindox 
o Refined Oii+0.1 %Tocopherol 
(heated preparation) 
.,. Refined Oii+0.1 %Tocopherol 
(d ir?ct addit ion) 
8 1 0  1 2  1 4  1 6  
Storage Time (days) 
Figure 7.6. Effects of  various antioxidants on refractive index changes in fish oil during storage 
at 63?2?C 
The RI values of all oil samples increased with storage time. During the first day the control 
showed the highest increase rate. Up to the eighth day, all samples indicated an increasing trend 
134 
in RI value. After day eight, the RI value of the control sample tended to be higher than treated 
samples. 
?""""::a-:..??::r..""=-?dif..fu{eJ.lGe in iD Y.alue.until \ 
the twelfth day, but the RI value of the oil containing Grindox exceeded the value in the oils 
containing TBHQ and BHA at the end of storage trial. The two different tocopherol additions did 
not exhibit any effect on the changes in RI value. 
7 .3.2. Optimisation of antioxidant level 
Even though TBHQ, as described in the discussion section of this chapter, was shown to be the 
best antioxidant in the refined fish oil during storage, this antioxidant could not be used in 
Indonesian oils and foods. TBHQ is not on the list of permitted antioxidants for Indonesian oils 
and foods (Indonesian Health Ministry, 1974). BHA as the best alternative was used for the further 
experiment in the optimisation of antioxidant levels in refined fish oil. 
7.4.2.1.  Effects of BHA levels on peroxide value (PV) 
The PV changes in refmed oil containing BHA at several levels and unrefined oil during storage 
at 63?2?C are shown in Figure 7.7. 
90 
- 80 
CJ 
? 
...... 70 C" 
Cl) 
E 60 
Cl) 50 ::J 
IQ > 40 
Cl) 
30 "'C 
>< 
0 20 ..... 
Cl) 
a.. 1 0  
0 
0 
o Unrefined Oil (Control)  
D Refined Oil + 0.005% BHA 
? Refined Oil + 0 .0 1 %  BHA * Refined Oi l  + 0.0 1 5% BHA 
1J. Refined Oil + 0.02% BHA 
4 8 
Storage Time (days)  
135  
1 3  1 7  
Figure 7.7. Effects of various BHA levels on peroxide value changes in fish oil during storage at 
63?2?C 
On the second day of storage, the unrefined oil did not reveal any significant PV increase, but other 
samples treated with antioxidants showed a pronounced PV increase. A significant PV increase 
for all samples was noted on the fifth day, and the PV of the unrefmed oil was markedly lower 
than in all refined oils treated with BHA at several levels. On the ninth day, the PV increase in 
the refined oils treated with 0.015% and 0.02% BHA was insignificant, while the increase in other 
samples was significant. Unrefmed oil and refined oil treated with 0.02% BHA had no significant 
PV difference. The storage extension to 17 days resulted in a PV increase in all oil samples. The 
refined oils treateQ..with -G?l%, G.QJ,S%?and v().02% MA showed an ins-ignifieant PV 'fferen 
over untreated oil on the thirteenth day. The refined oil treated with 0.015 and 0.02% BHA bad 
no meaningful variance in PV at the end of storage, but ,this Yalu w.as lower in comP.arison to 
other oils. The PV in the refined oil treated with 0.005% B-1'1"? a lt\e ?n 
oil anhe>-end 
7.4.2.2. Effects of BHA levels on thiobarbituric acid (TBA) value 
All oil samples had an increased TBA value during storage as shown in Figure 7.8. 
1 70 
1 46 
1 21 
CD 
::I 
ea 97 > 
<( 73 aJ 
1-
49 
24 
0 
0 
o Unrefined Oil (Control)  
Cl Refined Oi l  + O.OOSX BHA 
? Refined  Oil + 0.0 1 %  BHA * Refined Oil + 0.0 1 5% BHA 
.A Refined Oi l  + 0.02% BHA 
4 8 
Storage Time (days) 
1 3  1 7  
136 
? Figure 7.8.  Effects of various BHA levels on TBA value changes in fish oil during storage at 
63?2?C 
On the fifth day of storage, the TBA value of unrefined oil was comparable to the value of the 
refined oil containing 0.02% BHA, but lower than the value in the refined oils containing 0.005%, 
0.01% and 0.015% BHA. The TBA values of the refined oil treated with 0.015% and 0.02% BHA 
were significantly lower in comparison to the values measured in other oil samples on the thirteenth 
day. At this time, the refined oil containing 0.005% BHA was the only treated sample having a 
higher TBA value than the unrefined oil. Finally, the unrefined oil and refined oil containing 
0.005% BHA exhibited a markedly higher TBA value than other treated samples at the end of 
storage, with similar TBA values in both oils. 
137 
7.4.2.3. Effects of BHA levels on anisidine value 
Figure 7.9 shows the anisidine value changes in unrefined oil and refilled oils treated with BHA 
at several levels during storage. 
Cl) 
::I 
500 
400 
? 300 
Cl) 
c 
"C 
C/1 
c 
< 
200 
1 00 
0 
o Unrefined  Oil (Con trol) 
Cl Refined Oil + 0.005% BHA 
? Refined Oil + 0.0 1 %  BHA * Refined Oil + 0.0 1 5% BHA 
1>. Refined Oil + 0.02% BHA 
4 8 
Storage Time (days )  
1 3  1 7  
Figure 7.9. Effects of various BHA levels on anisidine value changes in fish oil during storage at 
63?2?C 
All investigated oils showed a significant increase of anisidine value until the end of storage. 
Anisidine value in all oils increased at the same rate until the fifth day, and these values were 
statistically similar. However on the ninth day, the anisidine value of refmed oil treated with 
0.005% BHA was markedly higher compared to the values of other samples which bad a 
statistically insignificant different in anisidine value. On the thirteenth day, the refined oils 
containing 0.015% and 0.02% BHA exhibited a lower anisidine value than the unrefmed oil and 
refined oils containing 0.005% and 0.01% BHA. At the end of storage period, the anisidine values 
of unrefined and refmed oil containing 0.05% BHA increased sharply and both values were 
138 
obviously higher in comparison to the value analyzed in the refined oils containing 0.01 %, 0.015% 
and 0.02%. In general, the results indicated that the higher the BHA levels, the lower the anisidine 
value of the oil. 
7.4.2.4. Effects of BHA levels on totox value 
Totox values of unrefined oil and refined oils containing BHA at several levels increased during 
storage as shown in Figure 7. 10. 
650 
600 
550 
500 
Q) 450 
::I 400 ?< > 350 
>< 300 0 
- 250 0 
t- 200 
1 50 
1 00 
50  
0 
0 
o Unrefined Oil (Control) 
D Refined Oi l  + 0.005% BHA 
<> Refined Oi l  + 0.0 1 X BHA 
* Refined Oil + 0.0 1 5% BHA 
.A Refined Oi l  + 0.02% BHA 
4 8 
Storage Time (days)  
1 3  1 7  
Figure 7.10. Effects of various BHA levels on totox value changes in fish oil during storage at 
63?2?C 
Until the fifth day of storag?? the unrefined oil tended to show the lowest totox value, but at the 
ninth day that value was insignificantly different from the totox value in the refined oils containing 
0.015% and 0.02% BHA. The totox value of unrefined oil was markedly higher than the values 
139 
in the refmed oil containing 0.015% and 0.02% on the thirteenth day, and that value was lower 
than the value in the refined oils containing 0.005% and 0.0 1% BHA. At the end of storage, the 
totox values of unrefmed oil and refmed oil containing 0.005% BHA showed an insignificant 
difference, but those values were significantly higher than the values in the refmed oils containing 
0.01 %, 0.015% and 0.02% BHA. The refined oil containing a higher level of BHA showed the 
lower totox value. 
7.4.2.5. Effects of BHA levels on colour absorbance value 
The colour absorbance value in all oils tended to decrease during storage as shown in Figure 7 . 1 1 .  
-
E 
c: 
0 
en 
..q 
.... -::J (Q" 0 
0 CD 
u 0 c: 
cu 
..a 
.... 
0 
en 
..a 
<( -
0.8 
0.6 
0.4 
0.2 
0 
0 4 
o Unrefined Oi l  (Control) 
o Refined Oil + 0.005% BHA 
? Refined Oil + 0.0 1 X BHA * Refined Oil + 0.0 1 5% BHA 
A Refined Oil + 0.02% BHA 
8 1 3  
Storage Time ( days) 
1 7  
Figure 7 . 1 1 .  Effects of various BHA levels on colour absorbance value changes in fish oil during 
storage at 63?2?C 
The higher the BHA level added to the oils, the higher the initial colour absorbance value will be. 
The colour absorbance value decrease in unrefined oils occurred at a higher rate than in the refined 
oils containing BHA. The colour absorbance value reduction was found to be faster in the refined 
140 
oil containing a lower BHA level than in the oil containing higher BHA levels. The end of 
storage'shl:l'Wedi:tlni.ellie '  col() ut absorbance valu? in unrefined. oil <lecn?ased . 0,6,4, ?while)n; the 
",
'
,
'
,._ ? ?:: , , ,' ; ? ::_ ? ?'
'
,, "
,
',, : ,' ,  , , ' /' , ?>
'
, ::> ', ,' ' ' ' / ' '' ' , ' ' ; ,, , v, ? ?,, ? ;:,.<": ; ' , ' /'' , ,"':, , ,, ,' ,, ? , / - '  '' ;' ' '',' , !  
refined oils containil}g<07Q%?; 9:01 %;? O.OHo/a'?d' 0.(}2% cBHJsi::(f?reased 059,' 0359;:. 0:'63 and 
o?s?J:e:spectivel?. 
7.4.2.6. Effects of BHA levels on refractive index (RD value 
All oils exhibited an increased pattern of RI value during storage as shown in Figure 7. 12. 
1 .478 
1 .477 
>< 1 .476 
Q) 
"'C 
c: 1 .475 
Q) 
> 1 .474 
-
0 
CV 1 .473 .... 
-
Q) 
a: 1 .472 
1 .47 1 
1 .47 
0 4 
o Unrefined Oi l  (Control) 
D Refined Oil + 0.005% BHA 
o Refined Oil + 0.0 1 X BHA * Refined Oil + 0.0 1 5% BHA 
A Refined Oil + 0.02% BHA 
8 1 3  
Storage Time (days) 
1 7  
Figure 7.12. Effects of various BHA levels on refractive index changes in fish oil during storage 
at 63?2?C 
Unrefined oil did not show an increase in RI value until the fifth day of storage. A significant 
increase was noted on the ninth day. All the refined oils containing BHA had a sharp RI increase 
during the first ?ve days, an increase which continued at a slower rate until the thirteenth day. At 
141 
the end of storage, all oils exhibited a relatively sharp increase in RI value. 
7.4.3. Use of vacuum package for fiSh oil stability improvement 
7.4.3.1. Effects of vacuum package on peroxide value {PV) 
Both refined and unrefined oils stored at 63?C and 30?C showed the same response to the vacuum 
treatment in terms of PV, as shown in Figure 7.13. The patterns of PV change at storage 
temperatures of 63?C and 30?C were similar, as the PV of oils increased sharply and then gradually 
decreased. 
35 
o Refined oil, non-ucuum 
c. 30 C Refined oil, vacuum A Unrefin?d oil. non-vacuum .>< ...... * Unrefmtd od. vacuum tT ., 25 
..?. 
., 20 
:::J .. 
> 15  
c -o 
")( 1 0  
!: " 
a.. 
0 
0 2 4 6 1 0  1 2  1 4  1 6  
Storage Time (days) 
At 63?z?c 
70 
c. 60 ""' ...... tT 50 ., 
..?. 
" 40 
-= .. 30 > 
c 
-o 20 ")( 
!: c 10  a.. 
0 
0 2 4 
o Refined oil, non-vacuum 
c Refined oil, vacuum 
A. Unreflned oil, non-vacuum 
* Unrefintd oil, vacuum 
10  1 2  1 4  16  
Storage Time (weeks) 
At 3o?z?c 
Figure 7. 13. Effects of vacuum package on peroxide value changes in fish oil during storage at 
63?2?C and 3Q?2?C 
In the non-vacuum packages, the PV increased rapidly during the first two days at 63?C and after 
two weeks storage at 30?C. Then, the PV of oils stored at 63?C decreased until the end of storage. 
Until the eighth week there was a significant PV decrease in all oils stored at 30?C. From weeks 
8 to 17, all PVs were relatively constant. 
142 
In the vacuum packages, sharp increases in the PV of the oils was noted on the first day at 63?C 
and at week one for storage at 30?C. After day one, the PV of the oils stored at 63?C decreased 
during further storage, while changes in the PV of the oils stored at 306C was insignificant on the 
second week. A decreasing pattern was registered at the fourth week. Further storage for the oil 
at 30?C resulted in an insignificant change in PV. 
The PV increase of the oils stored in non-vacuum packages occurred at a markedly higher rate than 
in the oils stored in vacuum packages. The refmed oils tended to have a higher PV than the 
unrefined oils, except during the first week of storage for the oil stored at 30?C where the PV 
measured in unrefined oil was higher than in refined oil. 
7.4.3.2. Effects of vacuum package on TBA value 
The pronounced increase in TBA value of oils in non-vacuum packages at 63?C storage occurred 
on the first day as shown in Figure 7.14, but insignificant change occurred in the TBA value on 
the second day. Both refined and unrefined oils stored in non-vacuum packages at 30?C exhibited 
an increased TBA value during the first two weeks. After these periods the TBA values of oils 
decreased until the eighth day for oils stored at 63?C and the eighth week for the oils stored at 30?, 
followed by a relatively constant TBA value, to end of storage. 
68 
62 
E 55 c: .1.9 
N .1.3 "' 
? 37 
.. 
2 3 1  .. 
.> 25 
< IV  <D 
.... 1 2  
0 
0 2 " 
o Refined oil. non-vacuum 
c Refined oil, vacuum 
A Unrofined oil, non-vacuum 
* Unretined oil, vacuum 
1 0  1 2  1 4  1 6  
Storage Time [days) 
At 63?2' C 
120 
1 1 0 
100 
E c: 90 
N 80 "' 70 ? 
., 50 
:::0 50 -;;; 
> J.O 
< 30 <D 
.... 20 
10 
0 
0 2 .( 
o ReUned oil, non-vacuum 
c Refined oil. vacuum 
A Unrofinod oil, non-ucuum 
* Unrefinod oil. vacuum 
10 1 2  1 4  1 6  
Storage Time [weeks) 
At 30?2' C 
Figure 7. 14. Effects of vacuum package on TBA value changes in fish oil during storage at 63?2?C 
and 3Q?2?C 
143 
Refined and unrefined oils stored in vacuum packages at 63?C had an increase in TBA values until 
the first day of storage. Then, values decreased until the fourth day. The extension of the storage 
period resulted in relatively constant TBA values. The TBA values of the oil stored in vacuum 
packages at 30?C increased significantly until the first week. Insignificant changes in the TBA 
values are noted in the second week. The sharp decrease in TBA value is then exhibited at the 
fourth week. Further storage gave insignificant changes in the TBA values. 
The increase in TBA values in the oil stored in the vacuum package occurred at significantly lower 
rates than in the oil packed in non-vacuum conditions. In general, the refined oil had a higher 
TBA value than unrefined oils during storage. 
7.4.3.3. Effects of vacuum package on anisidine value 
The anisidine value changes in refrned and unrefined oils during storage at 60?C and 30?C as shown 
in Figure 7 . 15. Both storage temperatures gave a similar response to anisidine value changes. 
60 
54 
1.8  
" 42 " 
;;; 36  > 
.. 30 c: 
:2 24 ? ?c: 
< 1 8  0 Refined oil, non-ucuum 
1 2  0 Rtfined oil. vacuum A. Unrefined oil, non-vacuum 
* Unrefined oif. vacuum 
0 
0 2 4 10  12  11.  
Storage Time (days} 
At 63?2"C 
1.1 
1.2 
37 
.. 
2 3 1  .. 
> 26 
"' c: 21  'i5 
?;;; 
?c: 1 6  
< 
10  
0 
15  0 
o Refined oil, non-vacuum 
c Refined oil. vacyum 
A Unrelined oil, non-vacuum 
* Unrotined oil, vacuum 
10  12  14 16 
S torage Time (weeks}  
At 30?2"C 
Figure 7.15.  Effects of vacuum package on anisidine value changes in fish oil during storage at 
63?2?C and 3Q?2?C 
144 
Non-vacuum packaged oils showed a rapid anisidine value increase during the ftrst four days and 
two weeks at storage temperature 63?C and 30?C respectively. The extension of the storage period 
resulted in a relatively constant anisidine value. Vacuum packaged oils had a sharp increase in 
anisidine value during the ftrst two days for oil stored at 63?C and during the ftrst two weeks for 
the oils stored at 30?C. After these periods, the vacuum packaged oils showed an insignificant 
change in anisidine value during a further storage period. 
The increase in anisidine value in the oils stored in non-vacuum packages was at a faster rate. The 
refined oils showed a higher increase in anisidine value than that seen in unrefined oils, and this 
characteristic was shown clearly in the oil packed under non-vacuum conditions. 
7.4.3.4. Effects of vacuum package on totox value 
Totox value changes of refmed and unrefined oils stored in vacuum and non-vacuum packages at 
63?C and 30?C are shown in Figure 7.16. 
120 
1 10 
100 
90 
0 80 " -.; 
> 70 
)( 60 
.e so 
0 
I- 40 
30 
20 
1 0  
0 
0 2 
o Refined oil, non-vacuum 
c Refined oil, vacuum 
A Unrofined oil, non-ncuum 
* Unrefined oil, ncuum 
1 0  1 2  1 -4  1 6  
Storage Time (days) 
At 63?2"C 
1 80 
1 60 
140 
., 1 20 " -.; 
> 100 
)( 80 .e 
0 
I- 60 
40 
20 
0 ' 
0 2 -4 
o Refined oil, non-vacuum 
a Rehned od. vacuum 
A Unrehned oil, non-ucuum 
* Unntinod oil, ncuum 
1 0  1 2  1 -4  1 6  
Storage Time (weeks) 
At 30?2"C 
Figure 7.16. Effects of vacuum package on totox value changes in ftsh oil during storage at 63?2?C 
and 30?.2?C 
145 
Refined and unrefined oils stored in non-vacuum packages at 63?C showed a marked increase in 
totox value until the second day. The significant increase in totox values for refined and unrefined 
oils stored in vacuum packages at 63?C occurred during the first day. Further storage. of these oils 
resulted in a decrease in the totox value until the eighth day, which was followed by a relatively 
constant totox value. 
An increase in totox value in the oils stored in vacuum and non-vacuum packages at 30?C was 
shown until the second week. After that period, the totox value of oils stored in non-vacuum 
packages decreased until the eighth day of storage, and the totox value showed insignificant change 
during further storage. While the totox value in the oils stored in vacuum packages reduced 
significantly on the fourth day, changes in value were subsequently insignificant 
The increase in totox value of the oil stored in non-vacuum packages occurred at a higher rate than 
in the oils stored in vacuum packages. In addition, the refmed oil showed a higher increase in 
totox value compared to the unrefmed oil. 
7.4.3.5. Effects of vacuum package on colour absorbance value 
The changes in colour absorbance value of oils stored in vacuum and non-vacuum packages are 
shown in Figure 7.17. 
? ::1 0 
0 
() 
E 0.875 r:: 
0 0.750 
en .... 0.625 
-;;; 
"' 0.500 0 r:: "' 0.375 
-e 0 0.250 .. 
..0 
.:5 0.125 
0 
0 
o Refined oil, non-ucuum 
0 Re fined oil. vacuum 
"" Unrefined on. non-vacuum 
* Unrtfinod oil, ucuum 
' 1 0  1 2  
Storage Time (days) 
At  63?z?c 
1 4  1 6  
E 
r::: 
0 
Ol ... 
? ? ::1 0 
0 "' 
() u r::: "' 
-e 0 
M ..0 
.:5 
1.1 
0.978 
0.856 
0.733 
0.6 1 1  
0.489 
0.367 
0.2" 
0.122 
0 
0 
o Refined oil. non-ucuum 
o Re lined oil, vo?cuum 
A Unretmed oil, non-vacuum 
* Unrelined oil, vacuum 
146 
10  12  1 4  1 e  
Storage Time (weeks)  
At 3o?z?c 
Figure 7.17. Effects of vacuum package on colour absorbance value changes in fish oil during 
storage at 63?2?C and 30?.2?C 
The significant decrease of colour absorbance of oils stored in vacuum and non-vacuum packages 
at 63?C was found during the first two days, after which the values tended to be relatively constant. 
The colour absorbance changes in the oils stored at 30?C showed a different pattern. The 
decreasing pattern of the colour absorbance value in all oils stored at 30?C was noted during the 
first two weeks. Further storage for the unrefined oil in vacuum and non-vacuum packages 
indicated an increasing pattern at a slower rate than in the decreasing pattern. The refmed oil 
stored in non-vacuum packages exhibited a significant increase in colour absorbance value on the 
fourth day. Then the value decreased again on the eighth day. A further storage resulted in an 
insignificant change in colour absorbance value. The colour absorbance value of refined oil stored 
in a vacuum package was relatively constant after two weeks. 
The decrease in colour absorbance value in the oils stored in non-vacuum packages 
occurred at a higher level than in the oils stored in vacuum packages. 
147 
7.4.3.6. Effects of vacuum package on refractive index (RI) value 
In general, the pattern of refractive index (RI) changes of the oils at stOrage temperature of 63?C 
and 30?C was relatively similar as shown in Figure 7.18. 
1.475 
? 1.474 
ll'l 
? 1.473 
>< " 1.472 "0 
E 
C) 
?? 
1.471 
u 
1.470 .. 
? 
" 
o Refined oil. non-vacuum 
0 Refined oil, u.cuum 
A Unrefined oil, non-vacuum 
a: 1.46Q * Unrafinod oil, vacuum 
1.468 
0 2 .( 1 0  1 2  1 4  
Storage Time (days) 
At 63?z?c 
1..476 
u 1.475 
in UH ? 
)( 1 .473 .. "0 
E 1.472 
C) 
.::: 1.471 0 .. 
? 1 .470 
C) 
a: 1.46Q 
1.468 
1 8  0 2 .( 
g a:n?:? oil: ???;?,:.cuum 
4 Onraftna3 oil, non-vacuum 
* Unrefined oil, ucuum 
a 10  tz 14 1 e  
Storage Time (weeks) 
At 30?2?c 
Figure 7 . 18.  Effects of vacuum package on refractive index value changes in fish oil during storage 
at 63?2?C and 3Q?2?C 
RI value of the oils stored in non-vacuum package at 63?C increased until the fourth day for 
unrefined oil and the eighth day for refined oil. The RI increase in the oils stored in vacuum 
packages was significant during the first day. After these periods, the change in the RI value for 
all oils was insignificant until the end of storage. The RI changes in the oil stored in non-vacuum 
packages indicated a higher rate than in the oil stored in non-vacuum packages. 
The unrefined oil, stored at the 30?C, showed a relatively constant RI value during the first day, 
but an increase in RI was noted from the second day until the fourth day. The refined oil exhibited 
an increase in RI during the first two days. The reduction of RI values for both oils was observed 
on the eighth day. Storage extension to 12 days resulted in an increase in the RI value in both oils, 
but the increase in refined oil was registered as insignificant In addition, the RI changes shown 
at the end of storage were insignificantly. The RI changes of the oil stored in vacuum and non-
148 
vacuum packages were found to be nearly the same. 
7.4.3.7. Effects of vacuum package on odour 
Sensory observation tests for odour was performed only for the oils stored at 30?C. Eight trained 
Indonesian panellists performed the evaluation. 
6 
5 
"a;" 4 
.... 
0 
0 
e 3 
..... :::l 0 
"'0 2 0 
0 
o Refined oil, non-vacuum 
[J Refined oil, vacuum 
A Unrefined oil, non-vacuum 
* Unrefined oil, vacuum 
2 4 6 8 1 0  1 2  
Storage Time (weeks)  
1 4  1 6  
Figure 7.19. Effects of vacuum package on odour changes in fish oil during storage at 30?_2?C 
The odour scores of both refined and unrefined oils stored in vacuum and non-vacuum packages 
tended to increase during storage as shown in Figure 7.19. The increase odour score rate for the 
oils stored in vacuum packages occurred at a lower rate than in the oils stored in non-vacuum 
packages. This tendency was noted in both oils 
149 
7.5. DISCUSSION 
7.5.1. Use of antioxidants for fJSh oil stability improvement 
The results of this investigation indicate that the antioxidant induced changes in the initial peroxide 
value of fish oil. A reduction in the PV in the oils as the result of BHA, TBHQ and Grindox-1 17 
revealed that these antioxidants may have accelerated the conversion of hydroperoxide into the 
secondary products of oxidation during sample preparation. In contrast. the a-tocopherol resulted 
in an increase in peroxide value. Possibly the a-tocopherol itself was the main cause of that 
increase, since the a-tocopherol solution had a peroxide value of 1 .92 meqlkg. However 
antioxidants did not cause any significant changes for the initial values of other observed 
parameters, such as TBA, anisidine, colour absorbance and refractive index. 
All antioxidants used in this experiment showed their expected role: to reduce the incident of 
undesirable chemical and physical changes in fish oil. TBHQ and BHA demonstrated a very 
powerful,;antioxidanteffect to inhibit hydroperoxide, malonaldehyde and alP-aldehydes formation as 
shown by peroxide, TBA and anisidine value analyses respectively. TBHQ was also found superior 
in the protection of the oil from colour loss, while BHA had very poor ability to protect fish oil 
colour. The superior effectiveness of TBHQ in the protection of this food product against oxidation 
was also noted by Ke et al (1977) in mackerel skin lipid, and Tokarska et al (1986) and Hawrysh 
et al (1989) in canola oil. Sherwin (1990) states that TBHQ could interact with free amines to give 
a red colouration, and this may have given TBHQ more ability to keep the yellow-reddish colour 
of fish oil. 
Propyl gallate is also an effective antioxidant in fish oil. However this antioxidant was not 
effective over a long period of storage, exhibiting only short term capability over oxidation. After 
a period of storage there was a sharp increase in peroxide, malonaldehyde and alP-aldehyde 
formation in the oil. A similar occurrence was also encountered in oil colour loss where sharp 
decrease of colour absorbance value was noted after a period of storage. Propyl gallate has low 
oil solubility and losses its effectiveness under heat conditions (Sherwin, 1990; Dziezak, 1986; 
Coppen, 1989). These characteristics may have affected the effectiveness of this antioxidant in fish 
oil during this stability study. Propyl gallate chelates iron ions to form an aesthetically unappealing 
blue-black complex (Dziezak, 1986). However this complex did not occur in this study, as 
Grindox - 1 17 was incorporated with the chelating agent, citric acid. 
150 
DI-n-tocopherol was not an effective antioxidant in inhibiting peroxide, malonaldehyde and a/??
aldehyde formation as the primary and secondary products of oxidation in fish oil. However 
tocopherol showed as an effective protection to colour, and this indication was supported by results 
of an experiment conducted by Cort (1974) where tocopherol had the greatest effect in protection 
of carotenoids and vitamin A. Carotenoids are the main pigment forming fish oil colour. The 
protection of vitamin A was also important, because fish oil is rich in vitamin A (Brody, 1966). 
The use of tocopherol as an antioxidant would fail to provide effective protection against oxidation, 
when the oil was contaminated with trace amounts of metals, such as Fe2? or Cu2? (Valenzuela et 
ill, 1991). 
The results suggest that the order of antioxidant effectiveness was found to be TBHQ > BHA > 
Grindox-1 17 (propyl gallate) > tocopherol. Unfortunately, TBHQ is not mentioned in the list of 
antioxidants permitted in food and drink in Indonesia (Indonesian Health Ministry, 1974). Thus, 
BHA, an alternative, was used in further experiments to establish the optimum levels. In the 
optimisation experiment, the results revealed that the higher the BHA level, the more effective the 
antioxidant to protect the oil against oxidation. At high antioxidant levels, the antioxidant might 
not only react with alkyl peroxy and alkyl radicals, but also react directly with oxygen which would 
give an overall protection to the oil against oxidation. The use of unrefmed oil as control was to 
show the optimum antioxidant level necessary to recover the loss of natural antioxidant during the 
refining process. In the case of New Zealand crude oil refined using macroporous strong acid 
cation resin, 0.01% of BHA was sufficient to recover the loss of natural antioxidants. 
7.5.2. Use of vacuum package for fiSh oil stability improvement 
The results of the study of ?e va<;uum effect on fi.sh oil stability improve!llent at ?torage 
? ? ? w ? 
temperatures of 30?C and 63?C showed a similar pattern, but the peak values of PV, TBA and totox 
for oil stored at 30?C were higher than the values for oil stored at 63?C. The similar V, TBA and 
totox values analyzed in oil stored at 30?C would probably be obtained in the <Oil store? 
if the artruyses wefe ? cafried in th? ?ri&l bl?e"'en :1 ? anct;:1 days st?cik"e. 
In the oil stored in vacuum packages, where the oxygen level was limited, the peroxide formation 
was dependent on residual oxygen, and probably on oxygen in the sample itself. The recorded 
headspace vacuum in the can was approximately 17.7 - 19.2 inHg. Based on the relationship 
between the headspace vacuum and residual oxygen content, published by the CIG Gases Division 
(1989), the approximate residual oxygen content inside the cans was approximately 7.5 - 8.7%v/v. 
151  
Oxygen, or gas, can be present in  liquid samples in three forms: physically entrapped, dissolved 
or chemically bound (Fletcher, 1987). The presence of oxygen in the oil stored in vacuum 
packages was shown by the increase of peroxide value, indicating the formation of the primary 
product of oxidation, hydroperoxide. The sensory evaluation also detected the odour changes in 
the oils stored in vacuum packages, although the changes were slower than those in the oil stored 
in non-vacuum packages. 
The peroxide formation showed an increasing tendency, but a decreasing pattern was also noted. 
A similar pattern was obtained by Nair et al (1979) in skin and muscle lipids of oil sardine. This 
indicated that the decreasing pattern reflected the conversion state of peroxide into the secondary 
products of oxidation. TBA value showing malonaldehyde forming as a secondary product, 
exhibited a similar pattern as the peroxide value, where after having an increase pattern, the value 
decreased supporting the fmding of Nair et al (1979). This decrease could be due to reaction 
malonaldehyde-type compounds, with amino acids, peptides and other compounds released from 
decomposition of protein (Kwon et .ill., 1965; Finley, 1985). Malonaldehyde can cross link protein 
through a Schiff base reaction with the E-NH2 group of lysine (Belitz and Grosch, 1987; Finley, 
1985; Gillatt et .ill., 1988). Malonaldehyde can also undergo aldol-type reactions to produce a 
mixture of polymers (O'Brien, 1987). Anisidine values measuring secondary products of a/??
aldehydes increased at ftrst and then showed a relatively constant value during a further storage 
period. According to Dugan (1976) aldehydes are presumable converted to acids. If the aldehydes 
are enals or dienals, these may oxidize further to provide yet shorter chain compounds. Thus, the 
constant anisidine value was probably due to the conversion rate from peroxide into aJ?- aldehydes 
-and the conversion rate from aldehydes to other compounds, as mentioned above which were the 
same. 
Colour absorbance value of the oils stored at 30"C decreased at flrst and then tended to increase. 
This tendency was shown clearly in unreftned oil and reftned oil stored in vacuum packages. The 
increase of colour absorbance value might be due to the result of the reaction between 
hydroperoxides and their degradation products with protein inducing browning (Belitz and Grosch, 
1987). The rate of the discolouration of carotenoids was probably slower than the browning 
reaction rate which ftnally resulted in an oil of darker colour. At ftrst, the oils stored at 63?C 
showed a decreasing pattern of colour absorbance value, and then a relatively constant value was 
noted. The constant value was probably due to the browning reaction rate, and the discolouration 
rate of carotenoid which occurred at the same rate. 
This study revealed that vacuum package indicated a better protection for the oils compared to the 
152 
non-vacuum package by exhibiting slower changes chemically, physically and organoleptically. 
7.6. CONCLUSIONS 
The above results prove that antioxidants and vacuum packages could be used to improve fish oil 
stability during storage, especially for resin refmed fish oil. Among the tested antioxidants, TBHQ 
was the best antioxidant to use in resin refined fish oil. However BHA as the second best 
alternative was used in the optimisation experiment, because TBHQ is not classified as a permitted 
antioxidant in Indonesia. 0.01% BHA was actually sufficient to recover the loss of a natural 
antioxidant due to the resin refining process. 
153 
Chapter 8 
APPLICATION OF RESIN REFINING TO INDONESIAN FISH OILS 
8.1. BACKGROUND 
As discussed in Chapter 4, most Indonesian fish oil is not suitable for human consumption, 
especially the oil obtained from fish meal processing. A further selective refinement process is 
necessary to make the oil acceptable for human consumption. 
Chapter 5 proved that resin refining could improve fish oil quality in New Zealand fish oils. In 
another experiment using salmon oil, Fernandez (1986) proved that resin refming could improve 
fish oil quality chemically, physically and organoleptically. However the use of the resin refining 
process for Indonesian fish oils has not been investigated. 
In order to introduce the resin refming method to Indonesian fish oil industries, the superiority of 
this method should be studied using Indonesian fish oils. The stability of refined Indonesian fish 
oils should also be investigated. Information about the potential value of this method to the 
Indonesian fish industry is best obtained by conducting a survey. 
8.2. OBJECTIVES 
The experiment and survey were aimed to show the possibility of the application of the resin 
refining method to Indonesian fish oil, and the prospect of this method being adopted by Indonesian 
fish oil producers. The experiment was also designed to obtain information about the stability of 
refined Indonesian oils compared to unrefmed oils. 
154 
8.3. METHODOLOGY 
8.3.1. Materials 
Two kinds of Indonesian fish oils, from fish meal processing and canning waste, were used in this 
study. Both oils were obtained from sardines. Fish meal oil was supplied by P.T.Sumber Rejeki, 
Muncar (East Java). The canning waste oil came from P.T.Karya Manunggal Tri Sukses, Muncar 
(East Java). 
8.3.2. Methods 
8.3.2.1. Fish oil refining 
The oils were refmed using a resin packed column 39 cm in length and 1 .65 cm in diameter. The 
refining was performed with a fish oil-resin volume ratio of 1 : 1 .  Canning waste oil was refined 
by passing the oil through the column once. The fish meal oil was passed through the column 
twice, since the oil had a very undesirable odour. Seven Indonesian panellists participated in the 
-sensory evaluation using the same format to evaluate fish oil samples as used in Chapter 4, and as 
shown in Appendix 4.2. 
8.3.2.1. Fish oil stability 
Fish oil stability was investigated using the Schaal oven method, also used in the experiment 
discussed in Chapter 7. The fish oil samples, each of 80 ml were contained in 100 ml beakers, 
and covered loosely with aluminium foil. The samples were stored in an oven at 63?2?C and 
withdrawn after 0, 2, 4, 7, and 1 1  days. 
155 
8.3.2.1.  Survey on the response of Indonesian fish oil producers to resin refining method 
A survey was carried out with the 19 Indonesian fish meal factories discussed in Cl:lapter 4. The 
survey obtained information about their interest in adopting the resin refining process. The 
questionnaires used for the survey is shown in Appendix 8 .1 .  
8.4. RESULTS 
8.4.1. Effects of resin refining process on Indonesian fish oil 
The fish oil quality was analyzed chemically, physically and organoleptically, using the same 
methods applied to resin refming optimisation in Chapter 5. 
Free fauy acid (FF A) values of both fish meal and canning waste oils were reduced by the resin 
refining process as shown in Figure 8 . 1 .  The FFA values of fraction-2 for both oils were higher 
than the values measured in untreated and fraction-I oils. 
Cl) ::I 
IQ "'0 > u 
"'0 111:1 
0.35 
0.30 
0.25 
0.20 
0.1 5 
0.1 0 
0.05 
0 -'-----'--
Fish Meal Oil Canning Waste Oil 
Fis h  Oil Types 
0 Unrefined oil 
? Fraction-1 oil 
H3a Fraction-2 oil 
Figure 8.1 .  Free fatty acid value changes in fish meal and canning waste oils during resin refining 
process 
156 
Refractive index (RI) values of both oils also decreased after refining process as shown in Figure 
8.2. The effect of two refinings on the fish meal oil was significant, exhibiting a higher decrease 
in RI value compared to canning waste oil which was passed through the column only once. The 
RI values of fraction-2 for both oils were higher than RI values for fraction-I oils. However the 
value measured in fraction-2 fish meal oil was equal to the RI value of untreated oil, and the value 
of fraction-2 canning waste oil was slightly lower than the value in untreated oil. 
)( 
CD 
"0 
c:: 
CD 
> 
-
0 
IQ .... -
CD 
a: 
1 .4800 
1 .4790 
1 .4780 
1 .4770 
1 .4760 
1 .4750 
1 .4740 
1 .4730 
1 .4720 
1 .47 1 0  
1 .4700 
Fish  Meal  Oil Canning Waste Oil 
Fish Oi l  Types  
D Unrefined oil 
? Fraction-1  oil 
Efi} Fraction-2 oil 
Fjgure 8.2. Refractive index value changes in fish meal and canning waste oils during resin refining 
process 
The colour of both oils was also improved by resin refining. Oils with clearer colour were 
obtained, as shown in Figure 8.3. The colour of fraction-2 for both oils was darker compared to 
the colour of untreated oils and fraction-I oils. 
157 
2.4 
e s:::: 
0 1 .8 
0'1 
"':t 0 Unrefined oil .... 
::I ... ? Frllc tion-1  oil 0 IV 
0 G) 1 .2 ? Frllction-2 oil () 0 s:::: IV 
..c ... 
0 0.6 en 
..c IV -
Fish Melll Oil Cllnning WllSte Oil  
Fish Oil Type s  
Figure 8.3. Colour absorbance value changes in fish meal and canning waste oils during resin 
refining process 
The fatty acid profile changes in both Indonesian fish oils during resin refining are shown in Table 
8.1 .  
Table 8 . 1 .  Fatty acid proflles changes in fish meal and canning waste oils during resin refming 
process 
Fish meal oil Canning waste oil 
Fatty Acids 
U.T.*) Frac-1 Frac-2 U.T.*) Frac-1 Frac-2 
SAFA 37.7 38.5 41 .9 39.2 38.7 39.4 
MUFA 29.5 29.7 29.5 29.0 29.6 29.3 
PUFA 32.6 3 1 .6 28.5 3 1 .8 3 1 .7 3 1 .0 
ro-3FA 28.1 26.9 24.1 26.7 27.2 26.5 
EPA 19.0 17.8 16.0 17.2 16.9 16.5 
DHA 5.8 5 .5 4.8 6.1 6.3 6.0 , ?  -
\lote: "') U.T. - untreated oil 
158 
Saturated fatty acid (SAFA) value of fraction-2 fish meal oil was significantly higher than the 
values in untreated and fraction-! oils. The SAF A values of untreated and fraction-! oils did not 
show any significant different However the SAFA values in untreated, fraction-! and fraction-2 
canning waste oils did not exhibit any pronounced difference . .  
Monounsaturated fatty acid (MUFA) values of untreated, fraction-! and fraction-2 for both oils 
were relatively the same. 
Polyunsaturated fatty acids (PUFA), ro-3 fatty acids and eicosapentaenoic acid (EPA) values of 
fraction-2 fish meal oil tended to be lower than the values analysed in untreated and fraction-! oils. 
The untreated, fraction-! and fraction-2 of canning waste oil showed insignificant differences in 
PUFA, ro-3FA and EPA values. Docosahexaenoic acid (DHA) values of untreated and fraction-! 
fish meal oils were markedly higher compared to the value measured in fraction-2 oil. Untreated, 
fraction-! and fraction-2 canning waste oils showed an insignificant difference in DHA value. 
Refined oils showed a lower natural tocopherol antioxidant content than in unrefined oils, especially 
in terms of a, y, o-tocopherol and ex-tocotrienol. Fish meal oil was rich in ex-tocotrienol, which 
was not traced in canning waste oil. The only natural tocopherol antioxidant detected in canning 
waste oil was ex-tocopherol, as ?shown in Table 8.2. 
Table 8.2. Tocopherol content of fish meal and canning waste oils during refming process (ppm) 
Fish meal oil Canning waste oil 
Tocopherols 
Untreated Refined Untreated Refined 
oil oil oil oil 
ex-tocopherol 25.3 14.3 7.1  4.6 
ex-tocotrienol 297.3 1 16.4 n.d n.d 
y-tocopherol 4.6 1 .8 n.d n.d 
o-tocopherol 3.2 n.d n.d n.d 
'Jote: n .d. - not detected 
Panel results indicate that the undesirable odour in Indonesian fish oils could be reduced by using 
the resin refming method, as the odour score of fraction-1 oil was lower than the odour score of 
unrefined oil as shown in Figure 8.4. The odour score for fraction-2 fish meal oil was higher than 
159 
the odour score for fraction-! oil and comparable to the odour score of untreated oil. However the 
odour score for fraction-2 canning waste oil was higher than both fraction-! and untreated oils. 
8 
6 -
Cll 
.... 
0 
(J 0 Unref ined oil Cl) - 4 ? Fraction-1 oi l  .... !m Fraction-2 o i l  :::J 0 
" 
0 
2 
0 -'------1.--
Fish Meal  Oil Canning Waste Oil 
Fish Oil Types 
Figure 8.4. Odour score changes in fish meal and canning waste oils during resin refining process 
8.4.2. Stability of refined Indonesian fish oil 
0 ? ?  J .  ? - , 
MA.-'.1 .  Chemical changes 
The changes of peroxide, TBA, anisidine and totox values of both fish meal and canning waste oils 
are shown in Figures 8.5, 8.6, 8.7 and 8.8. 
Peroxide values of both oils were relatively constant during the first two days of storage but values 
increased on the fourth day, except for the untreated fish meal oil. These PV values still showed 
an increase until the end of storage. The fastest PV increase was noted in the refined canning waste 
oil. The PV increase in the refined fish meal oil and unrefined canning waste oil occurred at 
approximately the same rate. The PV changes in unrefmed fish meal oil were very slow compared 
to other tested oils. The fastest PV increase was found between the seventh and eleventh days for 
all Oils , except for the refined canning waste oil showing the highest PV increase rate between . - -
the fourth and the seventh day. 
CD ::::J 
1 50  
120 
? 01 90 
.:.: 
CD ....._ 
"0 0" ")( CD 
a E 60 
.... 
CD 
a. 
o Unrefined fish meal oil 
? Refined fish meal oil 
c Unrefined canning WI!Ste oil 
? Refined canning Wl!ste oil 
0 2 3 4 5 6 7 8 9 1 0  1 1  
Storage Time (days) 
160 
Figure 8.5 .  Peroxide value changes in both refined and unrefined fish meal and canning waste oils 
during storage at 63?2?C 
300 
o Unrefined fish meal oil 
? Refined fish meal oi l  
240 c Unrefined canning waste oil 
? Refined canning waste oil 
CD 
:::1 1 80 i;j > 
< 
ID 1 20 
I-
0 2 3 4 5 6 7 8 9 1 0  1 1  
Storage Time (days)  
Figure 8 .6. TBA value changes in both refined and unrefined fish meal and canning waste oils 
during storage at 63?2?C 
161 
All oils exhibited a very slow change in TBA values until the fourth day, except for refined 
canning waste oil where the TBA value increased significantly. Sharp increases in TBA values was 
observed starting on the fourth day for all oils. Refined canning waste oil showed the fastest TBA 
increase, while unrefined fish meal oil had the slowest increase in  TBA value. Unrefrned canning 
waste oil and refined fish meal oil displayed a similar increase rate in TBA value. 
The linear increase in anisidine value was noted until the fourth day for refined canning waste oil, 
the seventh day for both unrefined canning waste oil and refined fish meal oils, and the end of 
storage for unrefrned fish meal oil. After these periods, unrefined and refrned canning waste oils, 
as well as refrned fish meal oil, had a sharp increase in anisidine value. At the end of storage 
those oils had insignificant anisidine value differences. Unrefrned fish meal oil showed the slowest 
anisidine value increase. 
Ill 
::J 
500 
400 
.;. 300 
Ill c 
:.0 ... 200 
?c: < 
1 00 
o Unrefined fish meal oil 
? Refined fish meal oil 
c Unrefined canning waste oil 
? Refined canning waste oil 
0 2 3 4 5 6 7 8 9 1 0  1 1  
Storage Time (days} 
Figure 8.7. Anisidine value changes in both refined and unrefined fish meal and canning waste oils 
during storage at 63?2?C 
162 
The pattern of totox value changes was similar to the pattern of anisidine value changes. However 
at the end of storage the totox value of refined canning waste oil was significantly higher in 
comparison with the values of unrefined canning waste oil, unrefined fish meal oil and refmed fish 
meal oil. The unrefined fish meal oil exhibited the lowest totox value increase at all times during 
storage. 
750 
600 
G) 
? 450 
> 
)( 0 0 300 
f-
1 50 
o Unrefined fish meal oil 
? Refined fish meal oil 
a Unrefined canning waste oil 
? Refined canning waste oi l  
0 2 3 4 5 6 7 8 9 1 0  1 1  
Storage Time (days} 
Figure 8.8. Totox value changes in both refined and unrefined fish meal and canning waste oils 
during storage at 63?2?C 
8.4.2.2. Physical changes 
The colour absorbance and refractive index (RI) values changes in both fish meal and canning 
waste oils during stability study are shown in Figures 8.9 and 8.10. 
The colour absorbance value changes in fish meal oil showed a decreasing trend during storage. 
No significant change in absorbance value of unrefined oil was noted during the first two days of 
storage. The reduction of colour absorbance in refmed oil occurred at a higher rate than in 
unrefined oil. 
163 
Canning waste oil exhibited a different trend in colour absorbance change when compared to the 
fish meal oil. Unrefined canning waste oil had a linear colour absorbance value increase until the 
fourth day. Then, colour absorbance value decreased gradually. Refined canning waste oil 
exhibited a gradual reduction pattern of colour absorbance value until the seventh day. Then, a 
relatively constant value was observed. 
E' c: 
0 
m 
"<t 
.... ... ;;:J 0 fll 
0 G) 
() u c: 
fll 
..0 .... 0 .. 
..0 
fll 
1 .5 
1 .2 
0.9 
0.6 
0.3 
0.0 
0 
o Unrefined fish meal oil 
? Refined  fish mea l  oil 
a Unrefined canning waste oil 
? Refined canning waste oil 
2 3 4 5 6 7 8 9 1 0  1 1  
Storage Time (days)  
Figure 8.9. Colour absorbance value changes in fish meal and canning waste oils during storage 
63?2?C 
1 .4800 
1 .4790 
1 .4780 
>< CD 1 .4770 
""0 
..!: 1 .4760 
CD > 1 .4750 ?;;; 0 1 .4740 IQ ..... -CD 1 .4730 a: 
1 .4720 
1 .47 1 0  
1 .4700 
0 
o Unrefined fish meal oil 
? Refined fish meal  oil 
c Unrefined canning waste  oil 
? Refined canning waste oil 
2 3 4 5 6 7 8 9 1 0  1 1  
Storage Time (days)  
164 
Figure 8. 10. Refractive index value changes in fish meal and canning waste oils during storage 
63?2?C 
The refractive index (RI) values of unrefined fish meal oil was relatively unchange, but the refined 
fish meal oil showed an RI increase trend during storage. Sharp RI increases in fish meal oil 
occurred during the first two days but after that period, the RI value decrease gradually. At the 
end of storage, the RI value of refmed oil was higher than unrefined oil. 
The RI value of unrefined canning waste oil was relatively constant until the seventh day, but 
increased by the end of storage. The refined canning waste oil showed an increase trend of RI 
value starting at the second day. The RI value of this oil exceeded the RI value of unrefined 
canning waste oil starting at the seventh day. 
165 
8.4.3. Response of Indonesian fish oil producers to resin refining process 
From the 19 factories participating in the survey, 17 factories commented that the resin refining 
process was an interesting idea The other two factories were. not interested at all. Only one of 
the interested factories did not intent to apply the refining method. This factory was the traditional 
fish meal processor. 
All factories with refming units were interested in the new refining method being developed in this 
study. Two of them intended to replace their existing refining facilities if the method proved more 
effective. Another factory intended to operate the new refining process together with existing 
method. All factories would apply the new refming method for all fish oil types produced. 
Moreover, 13 of the factories surveyed which do not have a refining unit intend to adopt this 
proposed technology for all fish oil types produced. 
166 
Table 8.3. Results of fish meal factory survey about the response to resin refining process (number 
of factories) 
1 .  Factories comments about resin refining method being 
developed: 
a. an interesting idea 
b. not an interesting idea 
2. Factories commenting that the resin refining method was an 
interesting idea intended to adopt the technology in the 
factory: 
a. intending to adopt 
b. not intending to adopt 
3. The way to adopt resin refining method among factories 
which have been facilitated with refining -unit: 
a. replace the existing method 
b. operated together with existing method 
4. Fish oil types which would be refmed using resin refining 
method among the factories which have been facilitated 
with refming unit: 
a. all fish oil types produced 
b. certain fish oil types produced 
5. Fish oil type which would be refmed using resin refming 
method among factories which have not been facilitated with 
refining unit: 
a. all fish oil types produced 
b. certain fish oil types produced 
17 
_2 
19 
16 
_1 
17 
2 
1 
3 
3 
_o 
3 
13 
_o 
13  
167 
8.5. DISCUSSION 
8.5.1. Effects of resin refming on chemical, physical and organoleptic properties of Indonesian 
fish oil 
The above results proved that Indonesian fish meal and canning waste oils could improve their 
quality chemically, physically and organoleptically by application of the resin refining process. The 
same conclusion were reached using New Zealand fish oils as discussed in Chapter 5. 
Canning waste oil, which required only one refming, was easier to refine than fish meal oil. This 
seemed that the original fish oil properties may have affected the refming process. As mentioned 
in Chapter 4, the oil produced from fish meal processing had a very undesirable odour, and this 
caused difficulties during refining, as one refming was not enough to reduce the undesirable 
strength of the odour. The application of a second refining significantly decreased the undesirable 
odour. The purity of the oil was also irtdipated by the significant decrease in RI value, a greater 
RI value reduction than occurred in canning waste oil. The colour absorbance value reduction in 
fish meal oil was also greater in value than the reduction in canning waste oil. This appeared to 
be the result of two refinings. This indicated that the multiple refining method, as investigated in 
the experiment discussed as in Chapter 5, provided important and valuable information for 
Indonesian fish oils which were mostly obtained from fish meal processing . 
. In terms of fatty acid profiles, both oils exhibited a different behaviour. The fatty acid profiles of 
canning waste oil were relatively unchanged. Changes in the fatty acid profile were noted in fish 
meal oil. Saturated fatty acid (SAFA) value in fraction-2 oil was higher than in fraction-1 oil, 
while polyunsaturated fatty acid (PUFA) and ro-3 fatty acids values in fraction-2 oil was lower than 
in fraction-1 oil. The results of fatty acid profiles in fish meal oil contradicted the results discussed 
in Chapter 5 and the results from experiments conducted by Fernandez (1986). Probably the 
natural properties of the oils affected the changing trend in fatty acid profile, but the exact reasons 
for these differences are unknown. 
168 
8.5.2. Stability of refined Indonesian fish oils 
Wbile fish oil quality could be improved using the resin refining process, fish oil stability during 
storage must be considered in determining overall quality. 
Similar results, as shown in the experiment discussed in Chapter 6, are also demonstrated in this 
experiment. Refmed oils have less stability against oxidation indicating a faster formation of 
primary and secondary products of oxidation. Generally, peroxide, malonaldehyde and a/[3-
aldehyde formation were faster in refmed oil than in unrefined oil, displaying a faster increase in 
peroxide, TBA and anisidine values. The TBA and anisidine values of fish meal oil were always 
lower than the values in canning waste oil. One reason for this is the higher free fatty acid value 
in fish meal oil, as shown in the experiment in Chapter 4, suppressed the formation of aldehydes 
(Nair, 1979). RI value changes also supported the above statement, where the increase level of RI 
value in refined oil was faster than in unrefined oil. 
G:x:i&ofcolour in refined oils occurred at faster rates than in unrefined oils and this result was also 
the same as observed in the experiment detailed in Chapter 6. Unrefined canning waste oil showed 
unusual colour absorbance changes compared to the results from the previous fish oil stability 
study. The first four days of storage resulted in the increase of colour absorbance value in 
unrefined canning waste oil, but a further storage period exhibited a decreasing trend. As 
mentioned before, the darkening process might be due to the reaction between proteins with 
hydroperoxides and their degradation products producing browning process (Belitz and Grosch, 
-1987). The increasing colour absorbance value in the canning waste oil during the first four days 
might be due to the darkening process occurred at a higher rate than the carotenoids decolouration. 
In the unrefmed oils, the natural antioxidant of tocopherol showed a higher protection of colour 
(carotenoids) against oxidation than in refined oils having a lower tocopherol content. After the 
darkening process achieved its peak, carotenoids decolouration process became more obvious 
showing a reduction in colour absorbance value. 
In general, the stability experiment revealed that canning waste oil had less ability to protect itself 
against oxidation attack in comparison to fish meal oil. The presence of a natural antioxidant, 
particularly tocopherols, was an important factor in protecting fish oil from oxidation process, as 
mentioned in Chapter 6, and proved in Chapter 7. According to the results of analysis, the 
tocopherols obtained from fish meal oil were a-tocopherol, y-tocopherol and o-tocopherol 
accompanied by a-tocotrienol, while a-tocopherol was the only natural tocopherol traced in 
canning waste oil. The presence of a-tocopherol in canning waste oil was at a lower level than 
169 
in fish meal oil. Pokomy (1987) stated that a.-tocotrienol possessed a slightly higher antioxidant 
activity than the corresponding tocopherol and this a.-tocotrienol was present in fish meal oil at a 
higher level than other tocopherols. The above indications give a reasonable explanation for why 
fish meal oil has more protection against oxidation than canning waste oil. 
8.5.3. Prospect of introduction of the resin refining process for Indonesian fish oil 
The prospects of the introduction of the resin refining process to Indonesian fish oil industry 
appears to be very promising. It was determined that 84% of fish meal factories surveyed were 
interested in installing this refming process. All factories already having refming facilities 
considered adopting this process to replace, or to operate along side of the existing refining process .  
All factories intended to use this method to refine all fish oil types. 
The factories would be more interested in this process if the method was profitable. Femandez 
(1986) has carried out economic evaluation for both pilot plant and semi-commercial scale plant, 
both proving profitable. The larger the production scale or capacity the larger the profit. The 
findings in the study of resin refining optimisation as discussed in Chapter 5 indicate that the 
application of vacuum pressure treatment to resin column could accelerate the refining process. 
This could increase the effectiveness and profitability of the process. The refming process could 
be increased to 300% without vacuum pressure application. Consequently, the flow rate was 
-increased as well. Thus production capacity was increased and the opportunity to increase profits 
is possible. 
This study does not discuss the design of a refming unit using this technology. However, from the 
survey, some considerations could be suggested when designing the resin refming unit for 
Indonesian fish oil producers, especially for the producers located in Muncar (East Java) and 
Negara (Bali). Since the fish oil was just a by-product from fish meal and canned fish processing, 
the refining unit should be designed with automatic processes, thus more efficiently operated and 
with low labour costs. The materials used in constructing the refming unit should be available 
locally for easy of maintenance. The materials must be anticorrosive, since the factories are located 
close to the sea. All of these are expected to make the refining unit more attractive to the fish oil 
producers. 
As discussed in the following chapter most canneries were interested in the production canned fish 
170 
enriched with fish oil. This could expand the market for refined fish oil. As discussed in Chapter 
9, more than 30% of respondents to the consumer survey answered that they were willing to 
consume fish oil in capsule form. This means that the pharmaceutical market will .also expand. 
Promotion to make the public aware of the health benefits of fish oil are very important to realize, 
and to open markets for fish oil. These would encourage producers to refme their oils to meet 
consumable standards making the prospect of the resin refming process adoption more feasible. 
8.6. CONCLUSION 
The above results indicate that Indonesian fish oil could be refined successfully using the resin 
packed column, where the oil qualities were improved chemically, physically and organoleptically. 
Treatments were required for refined oils to improve their stability during storage as suggested for 
New Zealand fish oils. Since the canning waste oil bad a low natural antioxidant content, the 
application of the stability improvement method should be used on the unrefmed oil as well. In 
this case, antioxidant additions, such as BHA, and vacuum package are advised. The prospect of 
adoption of the resin refining technology by the Indonesian industry is promising, and the prospects 
for new markets for refined oil appears to be an important factor in attracting producers. 
Chapter 9 
DETERMINATION OF CANNED FISH PRODUCT TYPE 
CONTAINING REFINING FISH OIL AS A MAJOR INGREDIENT 
9.1. BACKGROUND 
171 
Most companies now recognize that the key to their future survival and growth rests in a 
continuous flow of new and improved products due to the dynamic changes in market needs. The 
development of new products remains an exceedingly difficult and challenging undertaking. It is 
difficult because the process of innovation is inherently complex, requiring the close coordination 
and control of a multitude of vastly different tasks. It is challenging because important decisions, 
often involving the very survival of the enterprise, must be made on the basis of very limited 
information. Product innovation has become a vital element in corporate strategy and planning for 
a number of reasons outside the control of any single company. These include changes in 
consumer and competitor behaviour, technology and government policy (Rothberg, 1981). 
There are many ways in which a company may add products to its production for local and foreign 
markets. The most economic way, and a common practice, is to acquire a firm, or some operations 
of a firm which produces products with a potential market. A company can also add products to 
_ its own, by copying products developed successfully by others. Finally, a company can obtain new 
products by internal product development (Albaum et .ill., 1989). However commercial success is 
dependent on the following (Anderson, 1985): 
* selection of a product with a high level of consumer demand; 
* selection and definition of a product with the minimum of potential opposition from 
competitors; and 
* development of a product which ultimately embodies those characteristics which are 
required by and are acceptable to the consumer. 
A product has two key dimensions: technology and market. Technology is the fund of knowledge 
enabling economic production. Market relates to the "who" and "how" of product sale, enabling 
profitable distribution (Booz, Alien and Hamilton Inc, 1981). In product development terms, the 
food industry is adept at identifying popular preferences, and is a considerable force in persuading 
172 
consumers to try new taste sensations, thereby creating new preferences and expectations (Conning, 
1990). 
This study was conducted to obtain a new product nutritionally better than at present, by fish oil 
addition to improve ro-3 fatty acid content. As previously mentioned, health benefits of fish oil, 
particularly those provided by ro-3 fatty acids, are indisputable. In this study, product competitor, 
technological information and marlcet preferences were identified using supermarket, canned fish 
producer and consumer surveys. 
9.2. OBJECTIVES 
The supermarket, cannery and consumer surveys were aimed at obtaining information regarding 
the following: 
* existing canned fish products on the Indonesian market; 
* technological aspects of canned fish production in Indonesian canneries; 
* recent consumer behaviour relating to canned fish product; and 
* consumer acceptability about the proposed product type, using fish oil as the main 
ingredient. 
9.3. METHODOLOGY 
9.3.1. Supermarket survey 
The supermarket survey of canned fish product availability was carried out in 3 cities: Jakarta, 
Bogor (West Java) and Semarang (Central Java). Four supermarkets in Jakarta, 6 supermarkets 
in Bogor and 10 supermarkets in Semarang were surveyed. The questionnaire used is shown in 
Appendix 9 . 1 .  
173 
9.3.2. Cannery survey 
Sixteen canneries participated in the survey: 7 in Muncar (East Java), 8 factories in Ball and 1 
factory on Bitung Island. Direct interview survey was undertaken for factories in Muncar and Bali. 
A survey was mailed to the factory on Bitung Island. The questionnaire used for the survey is 
shown in Appendix 9 .2. 
9.3.3. Consumer survey 
A mail interview survey was used in this study. The questionnaire, as shown in Appendix 9.3.,  
was distributed in two large cities, Jakarta and Semarang. Sixty-five questionnaires were distributed 
in Jakarta and 160 in Semarang. Classification of high, medium and low incomes were set at 
more than Rp.SOO,OOO, between Rp.150,000-Rp.499,999, and less than Rp.149,999 respectively 
(Heruwati, 1990). 
Before the questionnaire was distributed in Indonesia, it was tested on Indonesians residing in New 
Zealand. The questionnaires were distributed to 35 people, with a 60% return. The questionnaire 
testing was expected to give provisional information on Indonesian consumer behaviour. Required 
changes in the questionnaire were made after this questionnaire testing. 
9.4. RESULTS 
9.4.1. Existing canned fish product in the market 
9.4.1.1. Origin of the canned fiSh product 
Of the 83 types of canned fish product on the Indonesian market in terms of trade mark, can size, 
fish species and medium, 43 (52%) were produced locally with the reminder (48%) imported. 
Thirty-four trade marks of canned fish are marketed consisting of 18 local and 16 imported brands. 
Countries exporting canned fish to Indonesia are Korea, People Republic of China, Mexico, 
1 74 
Australia, USA, Canada, Chill, Thailand, Portugal, Denmark and Notway. 
9.4.1.2. Fish species used in canned fish 
Three main fish species are processed internally for the Indonesian market sardine, mackerel, and 
tuna, as shown in Table 9 . 1 .  Other fish species available are imported canned salmon, herring, 
skipper and dace. 
Table 9 . 1 .  Percentage of canned fish product type on the Indonesian market according to fish 
species 
Local product Imported product Total product 
Fish species 
number % number % number % 
Sardine 26 60.5 14 35.0 40 48.2 
Mackerel 1 1  25.6 2 5.0 13 15.7 
Tuna 6 13.9 1 1  27.5 17 20.5 
Salmon - - 9 22.5 9 10.8 
Herring - - 1 2.5 1 1 .2 
Skipper - - 2 5.0 2 2.4 
Dace - - 1 2.5 1 1 .2 
Approximately 60% of the local products is processed using sardine as raw material. Imported 
canned sardine products constitute 35% of imported product type. Canned mackerel contributed 
to approximately 26% of local product type, but only 5% of imported product type. Canned tuna 
was approximately 14% of local product type and 28% of imported product type. Canned salmon 
accounts for approximately 23% of the total imported product. 
In terms of the total number of canned fish products on the market, canned sardine, mackerel, tuna 
and salmon contributed to 48%, 16%, 20% and 1 1%  of total number of product type respectively. 
175 
9.4.1.3. Medium used in canned fiSh 
As shown in Table 9.2, 86% of locally canned fish is produced using tomato sauce as a medium, 
while, 32.5% of all import varieties are canned in tomato sauce, and 71%  of import canned sardine 
uses tomato sauce as a medium. In total, 60% of all canned product types in Indonesia (imported 
and locally produced) use tomato sauce. 
Table 9.2. Distribution of canned fish product in the market according to medium used 
Local product Imported product Total product 
Type of medium 
number % number % number % 
Tomato sauce 37 86.0 13 32.5 50 60.2 
Vegetable oil 3 7.0 9 22.5 12 14.5 
Brine 2 4.7 12 30.0 14 16.9 
Veg.oil and - - 2 5.0 2 2.4 
brine mixture 
Fish oil - - 2 5.0 2 2.4 
Others 1 2.3 2 5.0 3 3.6 
Other mediums used in canned fish are vegetable oil and brine. These mediums are insignificant 
importance, contributing only 7% and 5% for the local product respectively. However these 
mediums made a 23% and 30% contribution to imported product type. The canned fish products 
of both these mediums are approximately 31% of total product type. Other mediums rarely used 
in canned fish are fish oil, a mixture of vegetable oil and brine, and sambal goreng (fried chill 
sauce). Two products having fish oil as a medium are imported from Norway, thus showing 
consumer acceptability. 
-176 
9.4.1.4. Type of can used in canned fiSh 
Four types of can were encountered in the market: tall tube, short tube; oval and rectangular. The 
size of these cans is shown in Table 9.3. 
Table 9.3. Distribution of canned fish product in the market according to can type used 
Local product Imported product Total product 
Type of can 
number % number % number % 
TaU tube can: 
155 16 36.4 2 5.1 18  21.7 
230 - - 1 2.6 1 1 .2 
425 12  27.3 2 5 .1  14 16.9 
440 - - 2 5.1 2 2.4 
Short tube can: 
105 - - 2 5.1  2 2.4 
141 - - 1 2.6 1 1 .2 
170 - - 2 5.1  2 2.4 
180 - - 2 5.1 2 2.4 
184 1 2.3 3 7.7 4 4.8 
185 4 9.1 4 10.3 8 9.6 
200 1 2.3 - - 1 1 .2 
210 - - 2 5.1 2 2.4 
220 - - 2 5.1 2 2.4 
Oval can: 
106 - - 1 2.6 1 1 .2 
125 - - 2 5.1 2 2.4 
200 - - 2 5.1  2 2.4 
213 - - 1 2.6 1 1 .2 
215 1 2.3 1 2.6 2 2.4 
227 - - 1 2.6 1 1 .2 
400 5 1 1 .4 - - 5 6.0 
425 3 9.1 3 7.7 6 7.2 
Rectangular 
can: 
1 06 - - 2 5.1  2 2.4 
125 - - 2 5.1 2 2.4 
177 
Approximately 64% of the local product is produced in tube cans. More than 55% of the cans are 
155g in size. However tall tube can only contributes approximately 18% of the imported product. 
In terms of total canned fish product, the tall tube can is used for approximately 42%, while 50% 
is produced in the 155 g can. 
The short tube can is used for approximately 14% of the local product, and approximately 46% of 
imported product. In terms of total canned fish product type, this can is utilized for approximately 
29%. Both local and imported product types showed that the 185 g can is the most commonly 
used. 
Oval can type is used equally for both local and imported products: 22.7% for local product and 
25.6% for imported product. Those product types account for approximately 24.1% of total product 
on the market. The rectangular can is used only in the imported product. 
The cross-tabulation between the can type and fish species is shown in Table 9.4. Both sardine 
and mackerel are mostly canned in tall tube and oval cans. Sardine is also canned in the 
rectangular can. Tuna is canned only in the short tube can, while salmon is canned in both tall and 
short tube cans. Herring, skipper and dace are marketed in oval cans. 
Table 9.4. Distribution of canned fish product based on the relation between fish species and can 
type 
Tall tube can Short tube can Oval can Rectangular 
Fish can 
species 
number % number % number % number % 
Sardine 25 30.1 - - 1 1  13.2 4 4.8 
Mackerel 8 9.6 - - 5 6.0 - -
Tuna - - 17 20.5 - - - -
Salmon 2 2.4 7 8.4 - - - -
Skipper - - - - 2 2.4 - -
Herring - - - - 1 1 .2 - -
Dace - - - - 1 1 .2 - -
178 
9.4.1.5. Price of canned rlSh 
The price of canned fish on Indonesian market varies with fish species, can size and product origin 
as shown in Appendix 9.4. 
Canned sardine is determined as the cheapest product. This is clearly evident in the local product. 
Can size affected the price of canned fish from the same product type: the bigger the can size, the 
more expensive the product. This occurrence can be seen clearly in the local product, where the 
price of canned sardine with tomato sauce medium in 155 g and 425 g tall tube cans is Rp.395 -
615. and Rp.960 - 1920. respectively. However this occurrence is inconsistent for imported 
products. 
In general, the price of the local product is cheaper compared to the imported product. For 
example, the differences in the price of imported canned tuna is doubled: 185g can is Rp.975 -
1915. for the local product and Rp.3100 - 5280. for the imported products. 
9.4.2. Production infonnation for canned fish 
9.4.2.1. Fish species used for canned rlSh production 
Fish species and total volume for each species used for canned fish production in all canneries 
surveyed are shown in Table 9.6. 
179 
Table 9.6. Fish species used for canned fish production 
Fish weight (ton/year) to be processed 
Fish species into: Number of 
Product for Product for 
factory 
using*) 
local export TOTAL 
market 
Sardine 24,210 600 24,810 13 
Mackerel 550 - 550 3 
Tuna 2,350 61,200 63,550 5 
Skipjack 600 46,350 46,950 3 
Scad 100 - 100 1 
'lote: one tacto ry could use more t t1an one ti.sh s pe ctes 
Most canneries use sardines as raw material for production of canned fish for the local market. 
In tenns of quantity, tuna and skipjack are consumed in greater quantity than other fish species. 
However most of the tuna and skipjack are canned for export purposes. Mackerels and scads are 
also canned, but in small quantities only. 
?.4.2.2. Medium used for canned fish production 
It was found that six types of mediums are used in the production of canned fish by Indonesian 
canneries as shown in Table 9.7. Tomato sauce is the common medium, and is used by 14 
canneries. Vegetable oil and brine are used as a medium by 7 and 5 canneries respectively. Other 
mediums reported are brine and vegetable oil mixture, vegetable broth and sambal goreng (fried 
chili sauce) . 
180 
Table 9.7. Medium used for canned fish production 
Medium Number of factory using*) 
Tomato sauce 14 
Vegetable oil 7 
Brine 5 
Vegetable oil and brine mixture 1 
Vegetable broth 1 
"Samba! goreng" (fried chili sauce) 1 
?ote: *) one factory could use more than one medium 
9.4.2.3. Canned fish marketing by canneries 
Five canneries export canned tuna and skipjack. Four factories export almost all their production 
of canned tuna and skipjack. One factory exported 25% of its canned sardine product. 
Referring to the products marketed locally, fish species and medium used the most are sardines and 
tomato sauce respectively, as shown in Table 9.8. 
181  
Table 9.8. Fish species used for canned fish production for local market 
Number of factories using*) 
A. Fish species: 
Sardine 13 
Mackerel 3 
Tuna 2 
Skipjack 1 
Scad 1 
B. Medium: 
Tomato sauce 14 
Vegetable oil 2 
"Sambal goreng" 1 
(fried chill sauce) 
to?ote: "') one tactory coU1<1 use more than one fiSh species and medium 
9.4.2.4. Canneries opinion to the idea "fish oil disguised in canned fish" 
Thirteen producers commented that the idea of disguising fish oil in canned sardine is an interesting 
one. Three canneries are not interested, because they produce mostly canned tuna for export. 
Among 13  canneries producing canned sardine, 10 are interested in the proposed product. Of the 
14 canneries using tomato sauce as medium, 1 1  are interested. In general, most of the canneries 
(10 canneries) indicated that disguising fish oil in the tomato sauce medium is an interesting idea 
as shown in Table 9.9. 
Fourteen canneries requested information about the technology for production, if the technology 
can be developed. Some of these canneries stated they would produce this product on condition 
that the product was acceptable to the market and had low production costs. 
As shown in Table 9.9, 10 canneries are interested in using the technology for 1-10% of their total 
present production. Although, one factory did not market its product locally, it planned to enter 
the market if the use of fish oil addition created a promising market. One cannery intends to 
produce this product as 21-30% of total production. Two canneries plan to produce the canned fish 
182 
with disguised ftsh oil up to 40% of total production. One producer requested technological and 
product information before any decision is taken. 
Table 9.9. Response of canneries to the idea "canned ftsh with disguised ftsh oil" 
1. Comment of canneries to the product idea "canned 
ftsh with disguised ftsh oil: 
a. interested 
b. not interested 
2. Comment of canned sardine producers to the product 
idea "canned ftsh with disguised ftsh oil":-
a. interested 
b. not interested 
3 .  Comment of canneries producing canned ftsh using 
tomato sauce medium to the idea "canned ftsh 
with disguised ftsh oil: 
a. interested 
b. not interested 
4. Medium suggested by canneries to be disguised with 
ftsh oil: 
a. tomato sauce 
b. vegetable oil 
c. vegetable oil and brine mixture 
5. Canneries requesting to be informed with the 
technology to produce this product 
a. YES 
b. NO 
6. Percentage of the product which was going to be 
produced (based on the percentage of total 
production): 
a. 1 - 10% 
b. 1 1 - 20% 
c. 21 - 30% 
d. 3 1 - 40% 
e. > 40% 
Number of factory 
13 
_3_ 
16 
10 
_3_ 
13 
11 
_3_ 
14 
10 
4 
_2_ 
16 
14 
_2_ 
16 
10 
1 
_2_ 
13 
183 
9.5.3. Consumer behaviour towards canned fish product 
9.5.3.1. Demographic characteristics of respondent 
Of the 225 questionnaires distributed 50 questionnaires (38%) were returned from Jakarta, and 80 
questionnaires (62%) from Semarang. Demographic characteristic of respondents is shown in 
Table 9.10. 
Approximately 60% of respondents participating in the survey are aged 20-29 years. Most 
respondents are in the middle income bracket. More than 79% of respondents worked in the 
private sector. 
Table 9.10. Demographic characteristics of respondents 
Number % 
Income brackets: 
high income 40 30.8 
(>Rp. 500,000/month) 
middle income 56 42.1 
(Rp. 150,000 - 500,000/month) 
low income ? 26.2 
( <Rp. 150,000/month) 130 100.0 
Ages (years): 
20 - 29 79 60.8 
30 - 39 30 23.0 
40 - 49 14 10.8 
> 50 _7_ ___M_ 
130 100.0 
Occupations: 
private sector 103 79.2 
civil servants _1:]_ 20.8 
130 100.0 
184 
9.5.3.2. Fish and fiSh product consumption 
Consumption frequency of fiSh and fish products is shown in Table 9.1 1 .  By assuming that a 
month consists of four weeks. For the respondents consuming fish and fish products more than 
twice per week, an average of three times per week was taken. Then the monthly total 
consumption frequency was calculated as shown in Table 9.1 1 .  
Table 9.1 1 .  Consumption frequency of fish and fish product 
Total 
Products % respondent consumption 
once/ twice/ >twice/ twice/ once/ 
frequency/ 
month 
week week week month month 
1. Fresh fish, 
including 33.1 17.7 25.4 3.8 3 .8 767 
frozen and 
chilled fish 
2. Processed 
products: 
- dried salted 16.1 10.8 6.1 9.2 22.3 345 
fish 
- boiled salted 16.1 3 .1  3 .8 13.1 12.3 226 
fish 
- fermented 10.8 8.5 26.1 1 .5 4.6 562 
fish/shrimp 
paste 
- pedah (moist 2.8 0.8 3 .1  4.6 16.1 101 
fermented 
fish) 
- jambal (spongy 6.1 3.1 1.5 0.8 8.5 101 
fermented 
fish) 
- fish sauce 2.3 1.5 4.6 2.3 6.9 1 15 
- canned fish 7.7 3.8 1 .5 6.9 22.3 159 
- smoked fish 3.8 1.5 2.3 7.7 10.8 106 
- softened bone 10.0 5.4 3.1 5.4 26.1 204 
fish 
- fish ball 7.7 3.8 6.1 3.8 9.2 200 
185 
The survey results showed that the respondents consumed fresh fish more often than processed fish 
products. Approximately 75% of respondents consumed fresh fish at least once a week. Among 
processed fish products, fermented fish/shrimp paste is consumed the most, but normally this 
product is consumed in small quantity as an appetizer. Other processed fish products consumed 
frequently are dried salted fish, boiled salted fish, softened bone fish and fish balls. Canned fish 
is consumed moderately,but more often when compared to moist fermented fish, spongy fermented 
fish, fish sauce and smoked fish. 
9.5.3.3. Fish oil consumption 
More than 73% of respondents were willing to consume refined fish oil as shown in Table 9.12. 
This indication was found for all income brackets, and age levels and occupation types. 
Respondents from medium and low income brackets showed more willingness to consume fish oil 
than respondents from the high income bracket. 
Table 9 .12. Preference of respondents to consume refined fish oil 
Willing to consume Not willing to 
consume 
Number % Number % 
Income brackets: 
high income 26 65.0 14 35.0 
medium income 44 78.6 12 21 .4 
low income 25 73.5 9 26.5 
Age (years): 
20 - 29 57 72.2 22 27.8 
30 - 39 22 73.3 8 26.7 
40 - 49 10 71 .4 4 28.6 
> 50 6 85.7 1 14.3 
Occupation: 
private sector 74 71 .8 29 28.2 
civil servant 21 77.8 6 22.2 
186 
The respondents who preferred to consume fish oil in capsule, salad oil, food with disguised fish 
oil and direct fish oil consumption were 33.8, 1 1 .5, 28.5 and 6.9% respectively as shown in Table 
9.13. 
Table 9.13. Fish oil consumption suggested by respondent 
Number of % respondent 
respondents*) 
Capsule 44 33.8 
Salad oil 15 1 1 .5 
Direct consumption of fish oil 9 6.9 
Food products with disguised 37 28.5 
fish oil 
Note: "') respon<lent cou1<1 choose more 1 nan one prO<luct 
9.5.3.4. Respondent preference in buying canned fish 
Most of the respondents stated that they selected a certain fish species and medium in buying 
canned fish. This indication is supported by the same tendency in all income brackets, age levels 
and occupation types, as shown in Table 9.14. 
As shown in Table 9.15 the fish species and mediums which respondents are most like to purchase 
are sardine (59.2%) and tomato sauce (51 .5%) respectively. Few respondents chose canned fish 
having other fish species (mackerel, tuna, squid and others) and other medium types (vegetable oil, 
brine, mixture of vegetable oil and brine and others). 
187 
Table 9.14. Respondent preference for a certain fish species and medium in buying canned fish 
Choose a certain Not to choose a certain 
fish medium fish medium 
species species 
No. % No. % No. % No. % 
Income brackets: 
high income 30 75.0 32 80.0 10 25.0 8 20.0 
medium income 44 78.6 52 92.9 12 21.4 4 7.1 
low income 28 82.4 32 94.1 6 17.6 2 5.9 
Age (years): 
20 - 29 62 78.5 71 89.9 17 21.5 8 10.1 
30 - 39 23 76.7 28 93.3 7 23.3 2 6.7 
40 - 49 12 85.7 12 85.7 2 14.3 2 14.3 
>50 5 71 .4 5 71 .4 2 28.6 2 28.6 
Occupation: 
private sector 85 82.5 92 89.3 18  17.5 1 1  10.7 
civil servant 17 63.0 24 88.9 10 37.0 3 1 1 . 1  
Table 9.15. Fish species and medium chosen by respondents in buying canned fish 
Number *) % 
Fish species: 
Sardine 77 59.2 
Mackerel 18  13.8 
Tuna 33 25.4 
Others (squid, shrimp, crab, 8 6.1 
milkfish and small tuna) 
Medium: 
Tomato sauce 67 51 .5 
Vegetable oil 14 10.8 
Brine 8 6.1 
Vegetable oil and brine mixture 16 12.3 
Others ("bumbu rujak", etc.) 21 16.1 
?ote: "') each respondent could choose more than one ftsh species anc medium 
188 
9.5.3.5. Attitudes of respondent towards the product idea "canned fish with disguised fish oil" 
Over 84% of the respondents were interested in the product idea of canned fish with disguised fish 
oil. Most of the respondents from each income bracket, age level and occupation type also showed 
interest in the proposed product idea, as shown in Table 9.16. Thus, the product idea was well 
accepted by all groups of respondents, 
As shown in Table 9.17 ., 56.9% of respondents suggested the use of tomato sauce to disguise the 
fish oil as a medium for the proposed canned fish product. Some other respondents suggested 
disguising fish oil in vegetable oil (16.1 %), brine (6.1 %) and a mixture of vegetable oil and brine 
(12.3%). 
Table 9.16. Respondent attitude to the idea of canned fish with disguised fish oil 
Product idea Buying trend 
Interested not Willing to Not willing 
interested buy to buy 
No. % No. % No. % No. % 
Income brackets: 
high income 32 80.0 8 20.0 28 70.0 12 30.0 
medium income 47 83.9 9 16.1 50 89.3 6 10.7 
low income 31  91 .2 3 8.8 27 79.4 7 20.6 
Age (years): 
20 - 29 69 87.3 10 12.7 63 79.7 16 20.3 
30 - 39 25 83.3 5 16.7 25 83.3 5 16.7 
40 - 49 12 85.7 2 14.3 13 92.9 1 7.1 
>50 4 57.1 3 42.9 4 57.1 3 42.9 
Occupation: 
private sector 57 83.8 1 1  16.2 82 79.6 21 20.4 
civil servant 53 85.5 9 14.5 23 85.2 4 14.8 
189 
More than 80% of respondents are willing to buy the proposed canned fish product, if the product 
becomes available. This fact is supported by respondents from all income brackets, age levels and 
occupation types as shown in table 9.16. Respondents suggesting to use can types of 155 g, 185 
g, 215 g and 415 g are 33.1 %, 31.5%, 25.4% and 7.7% respectively. Approximately 37% of 
respondents suggested a purchase price of Rp.l 000-1399. Other respondents are willing to buy the 
product if priced at Rp.400-999 (33.8%), Rp.1800-2599 (21.5%) and Rp.2600-3000 (4.6%) as 
shown in Table 9.17. 
Table 9.17. Respondent preference to medium type, can size and price for proposed canned fish 
product 
Number *) % 
Medium types: 
Tomato sauce 74 56.9 
Vegetable oil 21 16.1 
Brine 8 6.1 
Vegetable oil and brine mixture 16 12.3 
Others ("bumbu rujak", etc.) 39 30.0 
Can types: 
155 g 43 33.1 
185 g 41 3 1 .5 
215 g 33 25.4 
415 g 10 7.7 
Product price: 
Rp. 400 - 999.- 44 33.8 
Rp. 1000 - 1799.- 48 36.9 
Rp. 1800 - 2599.- 28 21.5 
Rp. 2600 - 3000.- 6 4.6 
'ljote: *) eacn respondent coUld cnoose more tnan one me< mm type, can stze and product pnce 
190 
9.6. DISCUSSION 
9.6.1. Product type to be developed 
Existing marketed canned fish shows that sardine is the most commonly used species for both 
locally produced and imported products. The market demand, exhibited by consumer survey 
results, reveals that most consumers chose canned sardine over the other species when buying 
canned fish. In addition, the cannery survey indicates that more canned sardine is produced for 
the local market than other canned fish products. The significantly lower price of canned sardine 
is probably the reason for the consumer preference. Producers could provide canned sardine at a 
lower price than other canned fish products, because the price of the raw material is significantly 
lower. As stated by one of the surveyed canneries, the prices of sardine, mackerel, skipjack and 
tuna are Rp.200-300/kg, Rp.500-800/kg, Rp.?00-900/kg and Rp.1 100-1300/kg respectively. Thus, 
the proposed canned fish product needed to use sardine as raw material, to enable the product to 
compete in the market 
Most local product types (86%) are canned in tomato sauce mediums, especially canned sardine. 
However Indonesia still imports canned sardine in tomato sauce - 4.000 tonnes in 1988 (Directorate 
General of Fishery, 1989). This was supported by the survey results where more than 30% of 
imported canned sardine are canned in a tomato sauce medium. The consumer survey also 
indicates that more than 50% of respondents purchased canned fish with tomato sauce medium. 
- In addition, more than 55% of the consumers suggested the use of fish oil disguised in tomato 
sauce as a medium for the proposed product. As a response to market demand, most of the 
canneries produce canned fish in tomato sauce medium for the local market. The canneries also 
suggest the use of tomato sauce for the proposed product The above results clearly indicate that 
the proposed canned fish product must use tomato sauce with disguised fish oil as a medium. 
Most canned sardines in the market are canned in a 155g tall tube can. If the proposed product 
is developed for the local market, the product needs to be packed in a 155g tall tube can. This fact 
is supported by consumer survey results where more respondents are willing to buy the product if 
canned in 155g can. 
The results obtained from the supermarket, cannery and consumer surveys indicated the same trend 
for the proposed product. The cannery survey is inclusive because managers were normally 
unwilling to answer questions about their activities as other factories could, possible, obtain this 
191 
confidential information. However this study proved that supermarket and consumer survey results 
are sufficient to generate the proposed product. 
9.6.2. Prospects for proposed canned sardine with fiSh oil addition 
The supermarket survey showed that the market for canned fish in tomato sauce, with disguised 
fish oil, is still open, since, to date, Indonesia imports canned sardine in tomato sauce to fulfil the 
market demand. Thus a competitive product already exists. However the proposed product is 
nutritionally better than the existing competitive product. The proposed product is richer in ro-3 
fatty acids proved to have health benefits for the consumer. This fact will probably help the 
proposed product to be more competitive and successful. 
In recent years the relationship between diet and health has received much publicity (Conning, 
1990). This would help the proposed product to acquire consumer popularity through promotion 
and publicity, as all consumers are now far more aware of how diet affects health (Dennis, 1990). 
For example, the nutrition campaign being conducted by the Indonesian government through 10  
PKK (Family Welfare Movement) programmes aims to guide Indonesians to improved family 
health and welfare. One of these concerns nutrition improvement. The nutritional benefits of the 
proposed product could be incorporated in the PKK programmes. However Conning (1990) states 
that while the food industry is satisfying consumer needs and wants, it remains true that 
-responsibility for adequate nutrition is with the individual. 
The consumer survey indicates that the respondents were willing to consume fish oil disguised in 
a canned fish product Over 80% of the respondents were willing to buy canned fish with 
disguised fish oil. These results indicate that there are potential consumers already willing to 
purchase the proposed product. Thus, there is no reason for canneries not to produce. In fact, 
most of the surveyed canneries show significant interest for the proposed product and asked to be 
informed about the processing technology when developed. Most canneries have decided about 
production levels for this proposed product, which is another positive response. 
192 
9.7. CONCLUSIONS 
The three surveys conducted, supermarket, cannery and consumer, indicate that the proposed 
canned fish product has to use sardine as raw material, tomato sauce as medium and 155 g tall tube 
can as container. 
The survey also revealed that the proposed product had a good prospect in the market, the idea 
receiving a positive response from the canned fish producers. However a consumer test for the 
final product was still necessary to assess exact consumer acceptance. 
Chapter 10 
TOMATO SAUCE FORMULATION AND 
STERILIZATION CONDITION SELECTION FOR FISH CANNING 
10.1. BACKGROUND 
10.1.1. Tomato Sauce Formulation 
193 
Tomato sauce with fish oil disguised in it was chosen as the medium for the proposed canned fish 
product. As discussed in the previous chapter, the tomato sauce was suggested as the most 
acceptable medium by respondent to the a consumer survey. Tomato paste and water are the main 
ingredients in the tomato sauce medium, while sugar, salt and spices are added to create a desirable 
tomato sauce taste (Wiahayani, 1983; Novikov, 1984). 
In the food industry, formulation plays an important role in the development of a new product, and 
must be selected carefully. In the development of a new tomato sauce with fish oil addition, with 
the objective of improving the nutritional value - particularly ro-3 fatty acids content - the 
formulation method must use a systematic determination with optimisation of ingredient levels. 
Trial and error methods, which waste time and money, must be avoided. 
Factorial designs are usually unsuitable for the development of food products involving more than 
one ingredient. In mixture design experimentation, it is impossible to vary one ingredient or 
component, while holding all others constant. As soon as the proportion of one component is 
altered so is that of at least one other components altered, since the sum of all components is 
always 1 .0. To cope with this situation a set of experimental plans, called mixture design, has been 
developed (Snee, 1971; Hare, 1974; Anderson, 1981a; Anderson and Earle, 1985). In a mixture 
experiment, the response to a blend or mixture of one or more ingredients depends only on the 
relative proportion of the ingredients present in the blend and not on the total amount of the blend 
(Comell, 1979). 
In mixture designs the sum of the variables must always be 1 .0 or 100 percent. This constraint is 
not applicable in the study of independent variables. Fewer runs may be carried out if the 
experimenter is willing to accept less information about the mixture. The results of mixture design 
194 
can be subjected to vigorous analysis to obtain mathematical models relating the ingredient levels 
to some response variables as discussed by Hare (1974). Rigorous mathematical analysis is not 
always necessary, however, as one of the advantages of mixture designs is the relative ease of 
making subjective judgements for product improvement from a visual appraisal of the responses 
throughout the mixture space (Anderson, 1981a; Anderson and Earle, 1985). 
Sensory evaluation involving trained panellists was used to determine the acceptability of the 
formula being developed. The usefulness of sensory testing in the product development is 
described by Erhardt (1978). Sensory tests provide the developer with corrective information and 
guidelines for product improvement. Tests can be used to determine which characteristics of the 
latest formulation do or do not meet the product model. Formulation can be also evaluated by 
sensory panels to determine whether optimization of the product quality has been achieved. 
Sensory evaluation can also be used to determine whether the addition of a certain ingredient 
affects the flavour of a product (Erhardt, 1978:). 
10.1.2. Sterilization 
The term "sterile" refers to a condition in which no viable microorganisms are present, a viable 
organism being one that is able to reproduce under conditions optimum for its growth (Lund, 
1975). The objective in the heat sterilization of foods is the destruction of heat resistant bacterial 
.spores. Commercial sterilization requires that the spores of all pathogens and of all organisms 
likely to grow under the anticipated conditions of storage be destroyed during processing (Brody, 
1971;  Board, 1981 ;  Adams, 1983). The process may be applied either within a sealed container, 
in the case of conventional canning, or prior to packing under aseptic conditions. Whatever the 
mode or method of hot sterilisation, the safety of a heat preserved product is not dependent upon 
the use of chemical additives or the control of temperature during storage and distribution (Hall 
and Pitcher, 1991). Profitability in sterilized products results from delivering, to the consumer, 
quality products safely and efficiently (Savage, 1984). 
The application of sufficient heat to destroy food spoilage microorganisms and enzymes also results 
in some desirable and undesirable changes in the foods (Lund, 1973). The desirable effects of 
heat processing may be summarized as follows: 
* favourable alteration of the characteristics of the product (browning reactions, textural 
changes, increased palatability, etc.); 
* destruction of enzymes (peroxidase, ascorbic acid oxidase, thiaminase, etc.); 
* improvement in availability of nutrients (gelatinization of starches and increased 
digestibility of proteins); and 
* destruction of undesirable food components. 
195 
The undesirable effects of heat processing include changes in protein and amino acids, 
carbohydrates, lipids, vitamins and minerals. According to Leonard (1986), the breakdown of heat 
labile constituents in food is approximated as a first -order chemical reaction, mathematically similar 
to the destruction of bacteria. 
The development in the processing of foods has been an attempt to optimize the thermal process 
for nutrient retention. For commercial sterilization, optimization of a thermal process is not so 
straightforward. The mode of heating within the product becomes an important factor. For those 
products which beat by convection, the high-texpperature and short-time process results in optimum 
nutrient retention. Again, a comparison of z values indicates that the rate of destruction of 
nutrients is less temperature dependent than is the rate of destruction of microbial spores. 
Therefore, the high temperature and short time (HTST) process favours nutrient retention (Lund, 
1975b). 
As the effect of sterilization on fish oil stability, particularly on the ro-3 fatty acids has not been 
investigated, a further investigation is now undertaken to obtain the results of such a study in order 
to apply them to the processing of canned fish with disguised fish oil. 
10.3. OBJECTIVES 
The aims of the experiments were to obtain: 
* selected combination levels of main ingredients in tomato sauce medium for the canned 
fish and optimal level of fish oil which could be disguised in tomato sauce to produce 
an acceptable tomato sauce; and 
* sterilization condition for processing canned fish with fish oil addition with regard to 
fish oil stability, especially ro-3 fatty acid. 
196 
10.3. METHODOLOGY 
10.3.1. Experiment 1 :  Tomato sauce formulation 
10.3.1.1. Materials 
The fish oil used in this experiment was crude fish oil. Tomato paste was obtained from J.Wattie 
Foods Ltd., Hastings. Salt, sugar, vinegar, shallots and garlic were purchased from supermarkets 
and retailers in Palmerston North. 
10.3.1.2. Formulation 
This experiment was planned to optimise the fish oil level disguised in a tomato sauce for canned 
fish, without reducing the acceptability of the product The mixture-design was used to develop 
a new tomato sauce formula. The original formula used as the base in the development of the new 
tomato sauce formula was obtained from P.T.Bangka Pioneer Industries Ltd., Bangka (Wiahayani, 
1983). The tomato sauce contains: 
In?edients Percentage {%} 
Tomato paste 18.6 
Water 74.5 
Salt 3 .7 
Sugar 3 .1  
In this experiment, salt and sugar contents were held constant, while tomato paste, fish oil and 
water quantities were subjected to change. 
Reguirements low level {% l high level {%} 
Tomato paste (T) 15 50 
Fish oil (0) 10 40 
Water (W) 40 70 
197 
Optimisation of the new formula obtained from the above levels was also performed to determine 
the acceptability of tomato sauce by panellists. The formula was evaluated organoleptically by 
nine trained Indonesian panellists in terms of consistency, odour, colour, mouth feel, appearance 
and overall acceptability. The sensory sheet used for this purpose can be seen in Appendix 10. 1 .  
The samples were served both cold and warm. The experiment was performed with two 
replications. 
10.3.1.3. Tomato sauce preparation 
To prepare the tomato sauce, water was heated to boiling and then sugar and salt added, and boiled 
until dissolved. The tomato paste was added and subsequently homogenized using a hand mixer. 
Finally, fish oil was added and mixed thoroughly using a hand mixer. 
10.3.2. Experiment 2: Simulation study on the selection of sterilization condition for canned 
fish with disguised fish oil 
Both unrefined and resin refmed fish oils were studied for their stability during the sterilization 
process. Fish oil (SOml) was contained in a can 6.6cm in diameter and 4.6cm in height. Vacuum 
.condition in the can was created by application of vacuum pressure at 17.72-19.19inHg, during 
sealing. 
Selected standard temperature and time combinations for salmon canning from the Food Processing 
Institute of the USA (Robertson, 1983) were applied to determine canning conditions suitable for 
processing canned fish with fish oil added. The temperature and time combinations were as 
follows: 
a. l l0?C and 139 minutes 
b. 1 16.7?C and 79 minutes 
c. l21.1?C and 64 minutes. 
Sterilization was carried out using a pilot plant scale retort 90 cm long and 56 cm in diameter. The 
experiment was performed with two replications. Seven trained Indonesian panellists participated 
198 
in the sensory evaluation of the oil. The sensory sheet used is shown in Appendix 10.2 .. 
10.4. RESULTS 
10.4.1. Tomato sauce formulation 
10.4.1.1 .  Tomato sauce formulation 
Figure 10.1  shows the complete space available for the mixture design. The limits on the three 
ingredients given previously, restricted the area of experimentation to the shaded, feasible region. 
From that region, the vertices suitable for the experiments are as follows: 
A = 15T + 150 + 70W 
B = 15T + 400 + 45W 
C = 20T + 400 + 40W 
D = SOT + 100 + 40W 
E = 20T + 100 + 70W 
F = 34.4T + 30.20 + 35.4W 
where, F shows the centre point of the region. 
Based on these vertices and T + 0 + W = 186.4g, the tomato sauce formula developed to 
investigate the behaviour of those ingredients in the product were as follows: 
27.96T + 27.960 + 130.48W 
27.96T + 74.560 + 83.88W 
37.28T + 74.560 + 74.56W 
93.20T + 18.640 + 74.56W 
37.28T + 18.640 + 130.48W 
64.12T + 56.290 + 65.99W 
7.4g salt + 6.2g sugar 
199 
T 
Figure 10.1 . Mixture space showing areas of experiment 
The products made using the above formulas were evaluated organoleptically and the total scores 
from all panellists for each parameter are shown in table 10. 1 .  
Table 10. 1 .  Total organoleptic score of the tomato sauce products of the flrst formulation 
Sample 
Cold 
Warm 
Note: lo = low 
hi = high 
c 
0 
d 
e 
A 
B 
c 
D 
E 
F 
A 
B 
c 
D 
E 
F 
Consistency 
53 
5 1  
56 
36 
56 
49 
50 
47 
53 
35 
54 
50 
Odour Colour Mouth 
feel 
54 47 48 
5 1  48 49 
52 56 54 
48 50 45 
54 52 53 
50 52 47 
48 47 47 
47 48 47 
48 56 53 
46 51 44 
51  54 53 
48 54 5 1  
Overall 
Appearance acceptability Coded Note 
49 50 loT;loO;hiW 
46 49 loT;hiO;loW 
' 52 54 loT;hiO;loW 
43 42 hiT;loO;lo W 
52 52 loT;loO;hiW 
50 49 mediant 
45 48 loT;loO;hiW 
42 45 loT;hiO;loW . 
51  53  loT;hiO;loW 
43 42 hiT;loO;lo W 
51  5 1  loT;loO;hiW 
52 5 1  mediant 
N 
8 
201 
10.9.1.2. Study of effects 
Table 10.2 shows the effects of each of the three ingredients used to make tomato sauce on the 
sensory properties employed to evaluate the acceptability of the products. An example of the 
calculation to evaluate these effects is shown in Appendix 10.3. 
Table 10.2. Effects of main ingredients on sensory properties of tomato sauce 
Tomato paste Fish oil Water 
Parameter 
Sample low high low high low high 
amount amount amount amount amount amount 
Consistency 54 32 47 54 46 55  
Odour 53 48 52 52 50 54 
Colour 51  50 50 52 5 1  50 
Cold Mouth feel 51  45 49 52 49 5 1  
Appearance 50 43 48 49 47 5 1  
Overall 51  42 48 52 48 5 1  
acceptability 
Consistency 51  35 46 50 45 52 
Odour 49 46 48 48 47 50 
Colour 51  5 1  5 1  52 52 5 1  
Warm Mouth feel 50 44 48 50 48 50 
Appearance 47 43 46 46 45 48 
Overall 50 42 47 50 47 50 
acceptability 
Samples served cold and warm did not show any significant effect on sensory properties, according 
to the panellists. 
A low level of tomato paste gave a better consistency, odour, colour, mouth feel, appearance and 
overall acceptability than a high level. 
202 
In the observation using both cold and warm samples, a high level of fish oil in the tomato sauce 
was preferred in terms of consistency, colour and mouth feel. In cold sample, the high amount of 
fish oil gave a better appearance, but this effect was not detected in the warm sample. However 
the fish oil level had no effect on the odour of the sauce. Overall acceptability scores indicate that 
a high level of fish oil was preferred. 
High levels of water in the tomato sauce resulted in a product with a better consistency, odour, 
mouth feel and appearance than the formula with low levels of water. A low level of water was 
preferred by panellists in terms the colour. Overall acceptability scores indicated that tomato sauce 
with a high level of water was more acceptable that the sauce with a low level of water. 
10.4.1.3. Formula optimisation of tomato sauce 
The above results suggest that the tomato paste level should be reduced and the fish oil level 
increased. The tomato paste, fish oil and water levels used for the optimisation of the tomato sauce 
formula are as follows: 
Reguirements low level {%} high level {%} 
Tomato sauce (T) 10 20 
Fish oil (0) 30 50 
Water (W) 40 60 
The new area in the mixture space is shown by the increase of fish oil level, and the reduced level 
of tomato paste. The new area of investigation to obtain the optimum level for each ingredient is 
shown in Figure 10.2. 
203 
T 
Figure 10.2. Mixture space showing new areas of experiment 
The vertices for the new area of optimisation are A, B, C and D, while and E represents the centre 
point: 
A = lOT + 300 + 60W 
B = lOT + SOO + 40W 
C = 20T + 400 + 40W 
D = 20T + 300 + SOW 
E = 1ST + 370 + 48W (centre point) 
204 
Based on the above vertices, the recipes of tomato sauce to be used for the optimisation experiment 
are as follows: 
18.64T + 55.920 + 1 1 1 .84W 
18.64T + 93.200 + 74.56W 
37.28T + 74.560 + 74.56W 
37.28T + 55.920 + 93 .20W 
27.96T + 68.970 + 89.47W 
+ 7.4g salt + 6.2g sugar 
Sensory evaluation results for the products made from the above recipes are shown in Table 10.3. 
Table 10.3. Total organoleptic score of tomato sauce products of the optimisation experiment 
c Overall-
Sample 0 Consistency Odour Colour Mouth Appearance accepta 
d feel bility 
e 
A 38 46 46 39 44 42 
B 41 47 47 44 44 43 
Cold c 61 59 62 57 61 62 
D 63 58 62 59 61 62 
E 56 54 57 54 56 56 
A 36 43 43 40 39 39 
B 36 41 46 42 43 40 
Warm c 55 55 60 55 56 57 
D 59 56 59 57 60 60 
E 53 52 55 54 55 55 
The sensory properties (consistency, odour, colour, mouth feel and appearance) of mixtures C and 
D were significantly better compared to three other mixtures. The overall acceptability showed the 
same trend, where products C and D were more acceptable to the panellists. In general, tomato 
205 
sauce processed from formula D was organoleptically better than tomato sauce produced from 
formula C using cold and warm samples. Thus, D was considered as the best formula for tomato 
sauce with fish oil disguised in it 
10.4.2. Stability of fJSh oil during sterilization 
This experiment was intended to find the sterilization condition which could provide the optimal 
protection for ro-3 fatty acid from deterioration. To reveal the changes in fish oil due to the 
sterilization, the fish oil samples were analyzed for peroxide value, anisidine value, TBA value, free 
fatty acid value, refractive index, colour value, fatty acid profiles and sensory properties. 
10.4.2.1 .  Effects of sterilization on peroxide value 
Sterilization resulted in a significant decrease in peroxide value (PV) in refined and unrefined oils 
as shown in Figure 10.3. Vacuum treatment and various sterilization condition showed a 
pronounced effect on the changes in PV. A higher PV decrease was encountered in the oil 
contained in vacuum package than in the non-vacuum container. The PV reduction in unrefined 
oil occurred at higher value than in refined oil. Refined oil sterilized at 1 10?C for 1 39 minutes had 
a significantly higher PV than the oil sterilized at 121 .6 for 64 minutes. The oil contained in 
vacuum package sterilized at 1 16.7?C for 79 minutes showed a markedly higher PV than the oil 
sterilized at 121.1 o for 64 minutes. 
5 
4.5 
Cl 
....: 4.0 ........ 
0'" 
Cll 3.5 E 
3.0 
Cll 
::I 2.5 10 
> 
2.0 
Cll 
? 1 .5 >< 0 
... 1 .0 Cll 
a.. 
0.5 
0 
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 . 1 'C, 
1 39 min. 79 min. 64 min. 
D Refined Oil, Non-Vacuum Steril isation 
? Refined Oil, Vacuum Sterilisa tion 
E3 Unrefined Oil, Non-Vacuum S terilisat ion 
[I] Unrefined Oi l ,  Vacuum Steril isation 
Temperature and  Time of Steri l isation 
206 
Figure 10.3. Peroxide value changes in fish oil during sterilization at various temperatures and 
times 
10.4.2.2. Effects of sterilization on anisidine value 
As shown in Figure 1 0.4, sterilization treatment of fish oil resulted in an increase of anisidine value 
in refined and unrefmed oil from both vacuum and non-vacuum packages. The increase of 
anisidine value in the oil contained in the vacuum package was lower than in the oil contained in 
non-vacuum package. The unrefined oil had a higher anisidine value increase than the refined oil. 
Analysis of variance indicated that unrefined oil sterilized at 12l . l?C for 64 minutes had a 
significantly higher anisidine value than the unrefined oil sterilized at l l 0?C for 139 minutes, and 
1 16.7?C for 64 minutes. 
?l 
?= 
"0 
.!::! 
c: < 
1 2  
9 
6 
3 
0 
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 . 1 'C, 
1 39 min. 79 min. - 64 min. 
Temperature and  Time of Steri l isation 
207 
0 Refined Oil, Non-Vacuum Sterilisation 
? Refined Oil, Vacuum Sterilisation 
E3 Unrefined Oil , Non-Vacuum Sterilisat ion 
aJ Unrefined Oi l ,  Vacuum Sterilisation 
Figure 10.4. Anisidine value changes in fish oil during sterilization at various temperatures and 
times 
10.4.2.3. Effects of sterilization on TBA value 
A significant decrease in TBA value of refined and unrefined oils due to sterilization treatment was 
noted in the study as shown in Figure 10.5. 
1 6  
1 2  
8 
4 
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 .1 'C, 
1 39 min. 79 min. 64 min. 
Temperature and Time of S teri l isation 
0 Refined Oil, Non-Vacuum Steril isation 
? Refined Oil, Vacuum S terilisa tion 
8 Unrefined Oil, Non-Vacuum S teriJisat lon 
aJ Unrefined Oil, Vacuum Sterilisation 
208 
Figure 10.5. TBA value changes in fish oil during sterilization at various temperatures and times 
The TBA value decrease in unrefined oil was registered at a higher value than in the refined oil. 
The reduction of TBA value in the oil contained in the vacuum package occurred at a higher value 
than in the oil packed in non-vacuum container. Various sterilization conditions did not induce any 
-significant effect on the TBA values of the sterilized oils. 
10.4.2.4. Effects of sterilization on totox value 
Sterilization treatment of the fish oil resulted in the reduction of totox value for refmed and 
unrefined fish oil as shown in Figure 10.6. However the reduction was insignificant in refined oil 
contained in non-vacuum package. In general, the vacuum package yielded sterilized fish oil 
having lower totox value than the non-vacuum package. Unrefined oil tended to show a higher 
totox value decrease than refined oil. Various sterilization conditions did not show a significant 
different in totox value for sterilized oils. 
4l :::1 
117 
> 
X 0 
-0 
1-
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 . 1 'C, 
1 39 min. 79  min. 64 min. 
Temperature and Time of S teri l isation 
209 
0 Refined Oil, Non-Vacuum Sterilisation 
? Refined Oil, Vacuum Sterilisation 
El Unreflned Oil, Non-Vacuum Steril isation 
[D Unrefined Oil, Vacuum Sterilisation 
Figure 10.6. Totox value changes in fish oil during sterilization at various temperatures and times 
10.4.2.5. Effects of sterilization on free fatty acid value 
fatty acid (FF A) value of fish oil. However sterilized oils showed a tendency to have a lower FF A 
value than unsterilized oil as shown in Figure 10.7. Vacuum treatment did not?aye any significant 
effect,on?the FF A va)-ue o.(th? fish .oil . during?.sterilization. 
11) 
::I 
Ill 
> 
"0 
-:c ?u 
Ill 
?o o < ?a; 
? 0  
3.0 
2.8 
2.6 
; ? 2.4 
IJ.. 
Ill 
11) Ill 
11) 
U: 2.2 
2.0 
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 .1 'C, 
1 39  min. 79 min. - 64 min. 
Temperature and Time of Steril isat ion 
210 
D Refined Oil, Non-Vacuum Steri l isation ? Refined Oil, Vacuum Sterilisation 
B Unrefined Oil, Non-Vacuum Steril isation 
[D Unrefined Oil, Vacuum Steril isation 
Figure 10.7. Free fatty acid value changes in fish oil during sterilization at various temperatures 
and times 
10.4.2.6. Effects of sterilization on refractive index 
Refractive index (RI) value of fish oil did not change significantly during sterilization. This was 
noted in refined and unrefined oils from both vacuum and non-vacuum packages. 
10.4.2.7. Effects of sterilization on colour value 
Photometric method was used to determine the colour value of both sterilized and unsterilized fish 
oils?o The colour value of refmed and unrefined fish oils increased due to the sterilization process 
as shown in Figure 10.8. This was found in both oils contained in vacuum and non-vacuum 
packages. The colour value increase in the oil packed in non-vacuum packages was higher than 
in the oil contained in vacuum package, and this was clearly exhibited in unrefined oil. T-test 
21 1 
indicates that unrefined oil sterilized at 1 16.7?C for 79 minutes had a lower colour value than 
unrefined oil sterilized at l 10?C for 139 minutes and 121 .1?C for 64 minutes. The non-vacuum 
packed oil sterilized at 116. 7?C for 79 minutes had a significantly lower colour value than the non?
vacuum packed oil sterilized at 1 10?C for 139 minutes. 
30 
27 
Ill 
..::! 
f'CI 24 > 
.... 
;::, 0 21 0 
() 
1 8  
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 .1 'C, 
1 39 min. 79 min. 64 min. 
Temperature and Time of Steri l isation 
D Refined Oil, Non-Vacuum Sterilisat ion 
- Refined Oil Vacuum Sterilisation B Unrefined Oil, Non-Vacuum Steril isat ion 
ITJ Unrefined Oil ,  Vacuum Steril isation 
Figure 10.8. Colour value changes in fish oil during sterilization at various temperatures and times 
212 
10.4.2.8. Effects of sterilization on fatty acid profiles 
Fatty acid profile changes in fish oil due to the sterilization process is shown in Table 10.4. 
Table 10.4. Fatty acid profiles changes of fish oil during sterilization 
l lO"C, 1 16.7?C 121.1?C 
Fish Fatty Untreated 139 minutes 79 minutes 64 minutes 
Oil acids Oil 
Vac. Non- Vac. Non- Vac. Non. 
Vac. Vac. Vac. 
SAFA 21.8 21.4 21.9 22.4 22.6 21.6 23.2 
MUFA 5 1 .0 50.8 50.8 50.6 5 1 .0 50.7 50.8 
Unrefined PUFA 27.2 27.2 27.2 26.9 26.4 27.6 26.0 
oil ro-3FA 24.0 24.2 24.1 23.9 23.3 24.6 23.0 
EPA 1 1 .2 10.9 10.7 1 1 .0 10.2 1 1 .1  10.2 
DHA 5.7 5.7 5.5 5.1  5.5 5.7 5.6 
SAFA 22.0 22.0 21 .7 21.5 22.5 21.3 20.7 
MUFA 50.9 50.6 51 .2 5 1 .0 52.0 50.4 52.2 
Refined PUFA 27.1 27.3 26.7 27.4 25.5 28.3 26.9 
oil ro-3FA 24.1 24.3 23.7 24.4 22.4 25.3 23.8 
EPA 10.6 10.7 10.5 10.7 10.2 1 1 .0 10.8 
DHA 5.4 5.7 5.7 5.7 5.6 6.0 5.9 
Analysis of variance indicates that the relative values of saturated fatty acids (SAF A), 
monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA), eicosapentaenoic acid 
(EPA) and docosahexaenoic acid (DHA) of fish oil were not affected by various sterilization 
treatments. However MUFA, PUFA, ro-3 fatty acids and EPA were significantly influenced by 
vacuum treatment during sterilization. 
Vacuum package treatment yielded sterilized fish oil with a lower MUFA value, but a higher 
PUFA, ro-3 fatty acids and EPA values than non-vacuum package. In terms of condition, the 
sterilization at 121.1?C for 64 minutes yielded fish oil with higher PUFA and ro-3 fatty acid levels 
213 
than sterilization at l 10?C for 139 minutes and 1 16.7?C for 79 minutes. 
10.4.2.9. Effects of sterilization on sensory properties 
The panellists indicated that sterilization caused the reduction of fishy odour and created an 
increase in rancid odour as shown in Figures 10.9 and 10.10. Vacuum treatment did not 
significantly affect the changes in fishy or rancid odour. Sterilization did not give a significant 
different in odour among sterilized oils. 
CD 
.... 0 
(.) 
en 
3 
0 
'8 
>-
.::: 
"' i.i: 
9 
8 
7 
6 
5 
4 
3 
2 
Raw 
Fish Oil 
1 1 0'C,  1 1 6.7'C, 1 2 1 .1 'C, 
139 min. 79 min. 64 min. 
Temperature and Time of Steril isa tion 
?D 
R
Re
1
f!ned 0!1, Non-Vacuum Sterilisa tion 
e rned 011 ,  Vacuum Sterili;ation 
E3 Unrefined Oil, Non-Vacuum Steri l isation 
[D Unrefined  Oil, Vacuum Sterilisation 
Figure 10.9. Fishy odour score changes in fish oil during sterilization at various temperatures and 
times 
(I) .... 0 0 
C/) 
.... 
:::l 
0 
"8 
"C 
?o 
c:: Ill 
a: 
4 
3.75 
3.50 
3.25 
3.00 
2.75 
2.50 
2.25 
2.00 
1.75 
1 .50 
1 .25 
1 
Raw 
Fish Oil 
1 1 0'C, 1 1 6.7'C, 1 2 1 . 1 'C, 
1 39 min. 79 min. 64 min.  
Temperature and Time o f  S terilisat ion  
214 
D Refined Oil, Non-Vacuum Sterilisat ion ? Refined Oi l ,  Vacuum Sterilisation 
E3 Unrefined Oil, Non-Vacuum Sterilis ation 
[0 Unrefined Oil, Vacuum Sterilisation 
Figure 10.10. Rancid odour score changes in fish oil during sterilization at various temperatures 
and times 
Sterilization changed the taste of the oil. As shown in Figures 10. 1 1  and 10.12, the fishy taste 
decreased and the rancid taste increased. Vacuum treatment did not show any effect on taste. 
Sterilization did not induce a marked different in taste score. 
9 
8 
G) 7 .... 0 0 (/) 6 
G) - 5 "' 
IQ 
1- 4 
> 
..c:: 
"' 3 u::: 
2 
Raw 
Fish Oil 
1 1 0 'C, 1 1 6.7'C, 
1 39 min. 79 min. 
1 2 1 .1 'C, 
64  min. 
Temperature and Time of Steri l i sation 
215 
D Refined Oil, Non-Vacuum Sterili sation ? Refined Oil, Vacuum Sterilisation 
El Unrefined Oil, Non-Vacuum Steril isation 
[0 Unrefined Oil, Vacuum Sterilisat ion 
Figure 1 0.1 1 .  Fishy taste score changes in fish oil during sterilization at various temperatures and 
times 
G) .... 0 0 
(/) 
G) -
"' 
IQ 
1-
? 0 c:: 
IQ 
c: 
4 
3.75 
3.50 
3.25 
3.00 
2.75 
2.50 
2.25 
2.00 
1 .75 
1 .50 
1 .25 
1 
Raw 
Fish Oil 
1 1  o?c. 
1 3 9  min. 
1 1 6.7'C, 
79  min. 
1 2 1 .1 'C, 
64  min. 
D Ref !ned 0 !1, Non-Vacuum Steri l isation ? Refined Oil, Vacuum Steril isation 
E3 Unrefined Oil, Non-Vacuum Steri l isation 
[I] Unrefined Oil, Vacuum Sterilis at ion 
Temperature and Time of  S teril isa tion 
Figure 10.12. Rancid taste score changes in fish oil during sterilization at  various temperatures and 
times 
216 
10.5. DISCUSSION 
10.5.1. Tomato sauce formulation 
The experiment showed that the mixture design was the most successful method to use in the 
formulation purpose. The design gave information about the formula, and the behaviour of the 
investigated ingredients. Thus the changes in ingredient levels could be made immediately without 
trial and error experiments. In this study, the effects of the ingredient quantities were investigated 
using sensory evaluation. 
The experiment proved that fish oil present in tomato sauce improved the organoleptic properties 
of tomato sauce, such as consistency, colour, mouth feel and appearance. As revealed in the survey 
conducted by Fernandez (1987), the respondents pointed out that the unacceptable odour of fish 
oil was the main problem in fish oil consumption. However this was not found in this formulation 
study. Both high and low levels of fish oil, disguised in tomato sauce, showed the similar effect 
on odour properties. 
In terms of overall acceptability, the formulation experiment indicated that fish oil added to tomato 
sauce and used as the medium in canned fish was acceptable to panellists. The experiment also 
revealed that a high level of fish oil was preferred, even higher than vegetable oil levels which are 
normally added to tomato sauce in fish canning at 3.4-10% (Novikov, 1984). The vegetable oil 
- added to tomato sauce is to make the sauce shiny (Directorate General of Fishery, 1984). This was 
also evident in this study, where a higher level of fish oil gave a better appearance to the tomato 
sauce. Thus, the fish oil addition to the tomato sauce had two advantages: improved nutritional 
value, with more appealing appearance. 
The above results suggest that the fmal tomato sauce formula could be manufactured using mixture 
D, considered by panellists as the best formula: 
Tomato paste = 18.6% 
Fish oil = 28.0% 
Water = 46.6% 
Salt = 3.7% 
Sugar = 3.1% 
217 
The quantity of tomato paste contributing in the fmal formula was the same as the level in the 
original recipe from P.T.Bangka Pioneer Industries Ltd., Bangka, 18.6%. The fish oil level 
disguised in the tomato sauce just replaced the water level. 
10.5.2. Fish oil stability during sterilization 
There are three categories of products of thermal oxidation: hydroperoxides; secondary oxidation 
products like hydroxy acids, keto acids, epoxy acids and carbonyl compounds formed by 
decomposition of the hydroxides; and cyclic and polymerized fatty acids (Lang, 1970). The 
products formed when fatty materials are heated in the absence of air can be roughly divided into 
three classes: volatile substances which can readily be distilled out; compounds which still contain 
about the original number of carbon atoms in the fatty acid chain; and dimers and polymers formed 
by attaching two or more fatty acid chain, together (Artman, 1969). 
The above results reveal that the hydroperoxide content of fish oil, as shown by the peroxide value 
analysis, decreased during sterilization. Many theories have been developed to explain 
hydroperoxide behaviour in the heating treatment of fats. The formation and destruction of 
hydroperoxides are extremely rapid at high temperature (Nawar, 1985), since hydroperoxides are 
readily decomposed thermally (Hiatt and Irwin, 1968). When ethyl linolenate, for example, was 
heated in air at 250"C, hydroperoxides decomposition occurred so rapidly that a net peroxide value 
_of zero was reached in less than 30 minutes (Lomanno and Nawar, 1982). Hydroperoxides also 
undergo a variety of scission and dismutation reactions to form a wide spectrum of carbonyl 
compounds, hydroxy compounds, short chain fatty acids, dimers and polymers (Dugan, 1 968; 
Smouse, 1978). These processes may have induced the reduction of hydroperoxide value in the 
fish oil individually or collectively during sterilization. 
Malonaldehyde content measured as TBA value also decreased during sterilization. Malonaldehyde 
is an intermediary in the decomposition of lipids (Finley, 1985). Sterilization may have accelerated 
the formation and decomposition rate of malonaldehyde, where, finally the fish oil showed a 
significantly lower TBA value than unsterilized oil. Sanders (1989) stated that malonaldehyde in 
fish tends to decrease with cooking. 
Anisidine measuring a/?-aldehydes content increased during sterilization. This indicated that 
sterilization encouraged the formation of aldehyde. Nawar (1985) demonstrated that aldehyde 
218 
formation from hydroperoxides in thermal oxidative reactions. This occurrence could also be used 
to explain the reduction of hydroperoxides in the oil due to sterilization. 
Free fatty acids (FF A) content of fish oil was relatively constant during sterilization. FF A resulted 
from the splitting of the attachment between the glycerol and fatty acids (Windsor and Barlow, 
1981). Hydrolysis which is the reaction between water and triglycerides, could also produce FFA 
(Patterson, 1989). FFA can be oxidized by autoxidation and the oxidation is concerned primarily 
with the unsaturated fatty acid (Hamilton, 1989). Probably during sterilization, the rate of FFA 
formation and decomposition was insignificantly different, thus the process yielded fish oil with 
a relatively constant FF A value. 
The sterilization process caused the fish oil colour to darken. A very complex process occurred 
affecting the colour of sterilized fish oil. Carotenoids, the most abundant natural pigments in fish 
oil, are unsaturated compounds and are therefore susceptible to oxidation giving rise to rancidity 
and bleaching. In addition, two types of isomerization can occur, namely, cis-trans isomerization 
and epoxy isomerization, which can cause the colour to lighten (Hall and Pither, 1991).  Oxidized 
lipids and lipids soluble compounds may also decompose to form browning precursors (Sadler, 
1987). Maillard browning reactions might also occur in the formation of the final colour of 
sterilized oil, where amino groups react with carbonyls from oxidized fat or aldehyde groups of 
reducing sugars (Burger and Waiters, 1973). These indications suggest that browning reaction is 
a more important factor in the formation of sterilized fish oil colour than carotenoids 
decomposition. 
Relative amounts of saturated and unsaturated fatty acids were insignificantly affected by 
sterilization. Sterilization at 121.1 oc for 64 minutes tended to produce sterilized fish oil with 
higli??.r?iativ? (;}uantity 0f,ca-J:fatty a??..s tn.an .t?09tke"t 'resl'e<t ;st?ti'Hiatiob c0fidi:tfo??s. .Tll'?Jfjgher 
relative quantity of ro-3 fatty acids could be obtained by vacuum treatment in the can. The 
advantages of vacuum treatment to reduce fish oil deterioration during sterilization were also shown 
b? colGLJf ?.alue.s, ? 
ma'lmta1de1l.yde blssetvetl"' rri acumn'p'acke?oils by exhibiting a lowel' anisidine val.ue. 
Sterilization reduced the fishy odour and taste strength of fish oil followed by an increase in rancid 
odour and taste. More aldehyde compounds may have imparted the odour and taste performance 
of sterilized fish oil. The odour and taste of fish oil disguised in tomato sauce seemed not to be 
a problem as discussed in the tomato sauce formulation experiment. Fish oil odour and taste were 
suppressed by the odour and taste of the tomato sauce, sugar and salt. 
219 
10.6. CONCLUSIONS 
The tomato sauce formulation experiment proved that, organoleptically, fish oil additj.on to tomato 
sauce was acceptable by panellists and even improved the appearance and nutritional quality of the 
product. The fmal recipe obtained from the mixture design experiment is: 
Ingredients Level {%} 
Tomato paste 18.6 
Fish oil 28.0 
Water 46.6 
Salt 3.7 
Sugar 3.1 
However this tomato sauce formula could be further improved to enhance odour and taste by 
addition of spicies. 
High temperature and short time (HTST) processing seems to be the best condition for sterilization 
of canned fish with fish oil disguised in it, particularly in order to protect ro-3 fatty acids from 
intensive deterioration. Vacuum treatment in the can was also suggested to give more protection 
to ro-3 fatty acids. Therefore, for canned fish with fish oil addition sterilized at 121.1?C with 
-vacuum headspace in the can, is recommended. 
Chapter 11 
DETERMINATION OF IMPORTANT FACTORS IN 
FISH CANNING AND CANNING PROCESS OPTIMIZATION 
11.1. BACKGROUND 
11.1.1. Canned Fish 
220 
Canned fish is defmed as fish packed in containers which have been hermetically sealed and 
sufficiently heated to destroy or inactivate all microorganisms liable to multiply at any temperature, 
while the product is in storage, canning spoilage or become harmful to the consumer (Codex 
Alimentarius Commission, 1976). The hermetically sealed container used also protects against 
damage and contamination with dirt (Connell, 1990). The most common sequence for canned fish 
processing are pre-treatment, pre-cooking, filling, sealing and retorting. 
The initial treatment of fish after capture is of prime importance for all types of preservation 
including canning. The most important operation after capture is to chill the fish. This practice 
is valid for tropical fish and for fish from temperate or cold water (Perovic, 1977). Pre-treatment 
of fish covers the range of operations during which the product is prepared for canning including 
- gutting, washing, nobbing, shucking, shelling, cutting, brining and dipping (Warne, 1988). The last 
washing can be combined with salting, to remove blood and slime and to improve flesh texture. 
The salting process also stabilizes the flavour of the canned product and brings out its characteristic 
taste (Van den Brook, 1965). Brining toughens the skin of the fish. When unbrined fish are 
canned, much of the skin adheres to the can. Brining also brightens the appearance of the fish 
(Burgess et 1lJ., 1967). Then the fish is packed manually on a conveyor which has supply conveyors 
for nobbed fish, empty cans and for fish already packed in cans (Perovic, 1977). 
Pre-cooking is usually in steam, water, oil, hot air or smoke, or a combination of these. Fish are 
very often pre-cooked for the following reasons (Codex Alimentarius Commission, 1976; Warne, 
1988): 
* to free the flesh of fluid which would otherwise cook out during the heat process and 
remain in the container as an undesirable free liquid; 
* to release body oil if the fish are excessively fat or if the oil has a very strong flavour; 
* to improve texture or to condition the flesh for further processing; 
* to obtain specific textural and flavour effects such as by frying in oil; and 
* to loosen the shellfish meat from the shell and to make the meat firmer . .  
221 
The amount of water released varies depending on the fat content: 17.5% for tuna and 19-34% for 
sardine (Van den Brook, 1965), and 20-30% for blue whiting (Karl, 1984). Pre-cooking can be 
combined with a dipping process, particularly for products which require additives to impart flavour 
or colour or in order to modify texture through the surface action of brines (Warne, 1985). 
In filling containers, account must be taken of the fact that the contents will expand or contract in 
volume with changes in temperature (Codex Alimentarius Commission, 1976). Sauce or oil should 
be added to cans at a temperature in excess of 80-90?C (Bidenko et al, 1974; Novikov, 1984). 
Headspace is necessary for thermal expansion caused by heating the product. It is important that 
the temperature does not result in an excessive build up of pressure causing damage to the hermetic 
seal (Warne, 1985). 
It is generally desirable to create a partial vacuum in containers of canned fish at the time they are 
sealed. This serves two purposes: it avoids excessive pressure from entrapped gas during heat 
processing, and it reduces the likelihood that internal gas pressure, which causes metal containers 
to swell if stored in warm places or exposed to low atmospheric pressures (Codex Alimentarius 
Commission, 1976). Vacuum may be obtained by using steam exhaustion, hot fllling and steam 
vacuum closing. The best, and most consistent method of obtaining can vacuum is through the use 
-of vacuum closing equipment (Landgraf, 1963). Sealing normally uses a double seaming 
operations. It is vitally important that the side seam and the double seams are completely hermetic 
(Clucas, 1982). 
Careful control of the amount of heat treatment is extremely important. If the product is not heated 
sufficiently it may spoil, or it might become a potential health hazard. If heat treatment is too 
severe, the quality may be reduced by over-cooking. It is necessary to heat process canned fish 
at a temperature in the range 1 10-121?C. The higher temperatures are often preferred because of 
the saving in time and because it is found that texture, colour or flavour of some products may be 
preserved better by heating for a shorter time at a high temperature (Codex Alimentarius 
Commission, 1976). 
222 
11.1.2. Screening experimental design: Plackett and Burman 
There is obviously a need for a practical method of screening the large number of variables in a 
process. If efficient preliminary screening could be accomplished, a relatively small number of 
important variables could be selected for further, more detailed study. One such method is the 
Plackett and Burman design, which is based on balanced incomplete blocks (Stowe and Mayer, 
1966). 
The Plackett and Burman design makes it possible to screen N-1 variables in N experiments. It 
is impossible to measure the effects of interactions, but this is not usually important in screening 
as the main objective is to isolate the important variables (Anderson, 1981a; Anderson and Earle, 
1985). William (1963) compared various screening designs in a problem where 24 variables were 
studied. A Plackett-Burman analysis using 28 experiments was compared with a random-balance 
design using 28 experiments and with a fractional factorial using 32 experiments. The Plackett?
Burman design is generally best suited for efficient screening and accuracy in selecting the 
important variables. 
Plackett and Burman have provided the ftrst row of the design matrix for investigating various 
numbers of factors. The remainder of the design matrix is generated by shifting this ftrst row one 
space to the left N-2 times, where N equals the number of experiments. The last row of the 
matrix, a row of minus signs, is added to the bottom of the generated matrix (Stowe and Mayer, 
1974; Anderson, 1981a; Anderson and Earle, 1985). For example, this study evaluated 8 factors, 
so 12 experiments were required. The frrst row is as follows (Plackett and Burman, 1946): 
+ + - + + + - - - + -
where + signiftes the high level of a variable and - the low level. 
The full design matrix is shown in Table 1 1 . 1 .  
Table 1 1 . 1 .  Design matrix for screening important factors in fish canning 
Treatments 
ST svs Brn 
1 + + -
2 + - + 
3 - + + 
4 + + + 
5 + + -
6 + - -
7 - - -
8 - - + 
9 - + -
10  + - + 
1 1  - + + 
12 - - -
"'ote: :ST - stenltzatlon ume 
SVS = steam vacuum sealing 
Brn = brining time 
PC = pre-cooking of fish 
PC 
+ 
+ 
+ 
-
-
-
+ 
-
+ 
+ 
-
-
Variables 
GA SA VA 
+ + -
+ - -
- - -
- - + 
- + -
+ - + 
- + + 
+ + -
+ - + 
- + + 
+ + + 
- - -
liA - g aruc a<1<11t1on 
SA = shallot addition 
V A = vinegar addition 
D = dummy 
D 
-
-
+ 
-
+ 
+ 
-
+ 
+ 
+ 
-
-
223 
D D D 
- + -
+ - + 
- + + 
+ + -
+ - + 
- + + 
+ + + 
+ + -
+ - -
- - -
- - + 
- - -
Every Plackett and Burman design includes, for each variable, the same number of runs at the high 
level and the low level. To calculate the effect of any input variable, the average result at the low 
Jevel of that variable is subtracted from the average result at the high level of the same variable. 
For example the effect of the sterilization time (ST) variable would be calculated as follows: 
Effects (ST) = ------------------------------- - --------------------------------
6 6 
where X, =  value of response variable in run 1 ;  n = number of run. 
The calculated effects of the dummy variables are used to test the significance of the real effects. 
In cases where there is more than one dummy variable, their values can be pooled to estimate of 
the standard error: 
224 
(ET Standard error (effects) = ? ---?--
where Ed = dummy effects; n = number of dummy effects. 
Having obtained a measure of tbe standard error of tbe effects, tbe real effects can be subjected 
to a t-test to see if tbey are non-zero. t values are calculated for each of tbe effects by dividing 
tbe effects by tbe standard error calculated from tbe dummies: 
effect 
t = ---------------
S.E.(effect) 
These calculated t values are tben compared to tbe table value where tbe degrees of freedom are 
equal to tbe number of dummy effects making up tbe error term. It is wise to use a significance 
level which reduces tbe risk of overlooking. For tbis reason 90%, rather tban 95% or 98% 
significance is recommended (Anderson, 1981a; Anderson and Earle, 1985). 
11.2. OBJECTIVES 
This investigation was intended to identify tbe important factors in tbe processing of canned fish 
?witb fish oil disguised in it by tbe application of tbe Plackett and B urman design. This study was 
also aimed at investigating tbe optimal condition for factors selected from Plackett and Burman 
design to produce tbe most acceptable product by tbe panellists. 
11.3. 1\.fETHODOLOGY 
11.3.1. Materials 
Crude fish oil (mainly bold) was used in tbis study. The oil was refmed using macroporous strong 
acid cation resin. Pilchards (New Zealand sardine) for canning were obtained from Star Fish 
Supply Ltd., Napier. 
225 
11.3.2. Methods 
11.3.2.1. Canned ftsh processing 
The fish, received in frozen condition was thawed in running water. Head, tail and visceral parts 
were removed and then washed using clean water. In this study, some fish were treated by brining 
in 15% salt solution after washing. Four washed fish were packed in cans 8.2 cm in height and 
5.1 cm in diameter. The fish was pre-cooked using steam at 98?C for 20 minutes, and the released 
water was discarded. Tomato sauce was added to each, leaving a headspace at the top of around 
5-7.5mm. The cans were then sealed. During sealing, hot steam was passed over the surface of 
the can for approximately 2 seconds to provide vacuum headspace in the can. Finally, the canned 
fish were sterilized in the retort at 121.1 ?C. 
11.3.2.2. Experiment 1: Determination of important factors in fish canning 
Firstly, the Fo of the canned fish was determined using the method described in Chapter 3. The 
results revealed that the Fo of the product was 8.44 and the total heating and cooling time was 35 
minutes. This total time was used as the low level for sterilization in the Plackett and Burman 
experimental design. 
Table 1 1 .2 shows the eight variables screened, and the upper and lower limits for these variables. 
226 
Table 1 1 .2. Variables and limits for Plackett and Bmman design of canned fish 
Variables Upper limit Lower limit 
A. Process variables: 
Sterilization time (min.) 60 35 
Steam treatment during sealing (sec.) 2 0 
Brining time in 15% salt solution (min) 10 0 
Pre-cooking time (min.) 20 0 
B. Seasoning variables: 
Garlic addition (%) 2 0 
Shallot addition (%) 2 0 
Vinegar addition (%) 1 .2 0 
The products were analysed chemically, physically and organoleptically. The chemical analyses 
performed were pH (fish and medium), moisture, protein, fat, salt contents, and peroxide value (fish 
and sauce). Hunter-L, -a and -b values of the fish meat colour were determined. Organoleptic 
evaluation was carried out for both the fish and tomato sauce medium by the 10 trained Indonesian 
panellists. The ideal ratio scoring method, which is basically a line-scale of descriptive testing, was 
employed during sensory evaluation and the sensory form used is shown in Appendix 1 1 . 1 .  
The salt level in the tomato sauce was reduced, as suggested by panel evaluation during the tomato 
sauce formulation, Chapter 10. The tomato sauce recipe used in this experiment contained 18.78% 
tomato paste, 46.95% water, 28.17% fish oil, 3.00% salt and 3.10% sugar. The percentages of 
garlic, shallot and vinegar added to the tomato sauce were based on the total quantity of tomato 
paste, water, fish oil, salt and sugar. 
227 
11.3.2.3. Experiment 2: Optimization of canned fish processing 
As will be discussed later in this chapter, sterilization time required optimisation in a further 
investigation. The optimization experiment used sterilization times of 40, 50 and 60 minutes. The 
salt level in the tomato sauce was varied at 1 .5 and 2.5%. The experiment was performed with two 
replications. The sensory evaluation method employed in the optimization experiment was a 
descriptore-scale of descriptive method, instead of the ideal ratio scoring method, since the 
panellists commented that the ideal ratio scoring method was difficult and confusing, and required 
intensive supervision during evaluation. The sensory evaluation form used for this purpose is 
shown in Appendix 11 .2. 
11.4. RESULTS 
11.4.1. Determination of important factors in canned ftsh processing 
11.4.1.1. Results of chemical, physical and sensory analyses 
Table 1 1 .3 shows the chemical analysis results for both fish flesh and tomato sauce. The moisture 
-(MC), protein, fat and salt content of fish were in the range of 66.05-68.13, 23.33-24.19, 2.63-4.27 
and 1 .3 1-2.33% respectively. The pH of fish flesh and tomato sauce varied from 5.96 and 5 .72 
to 6.13 and 6.01 respectively. The peroxide values (PV) of fish flesh and tomato sauce were in 
the range of 23.14-433.67 and 10.63-32.65 meq/kg respectively. 
228 
Table 1 1 .3 .  Results of chemical analysis of fish and tomato sauce 
Fish flesh Tomato sauce 
Treatments M.C. Protein Fat Salt P.V. P.V. 
(%) (%) (%) (%) (meq/ pH (meq/ pH 
kg) kg) 
1 66.17 23.52 3.69 1.51 171 .43 5.96 10.75 5.72 
2. 66.05 23.54 3 .90 1 .98 272.24 6.01 15.17 5.90 
3 66. 16 23.08 4.27 2.07 563.79 6.04 16.08 5.86 
4 66.71 23.51 4.05 2.06 258.86 6.00 15.34 5.89 
5 67.68 24.19 4.03 1 .3 1  23.14 6.05 16.04 6.01 
6 67.86 23.33 3 .67 1 .40 163.73 5.97 14.51 5.90 
7 66.66 23.75 4.17 1 .56 498.75 6.04 16.82 5.81 
8 68.03 23.42 3 .73 1 .94 818.66 6.12 32.65 6.01 
9 67.96 23.73 2.89 1 .46 246.97 6.04 10.85 5.94 
1 0  66.99 23.33 4.07 2.06 240.39 5.99 10.63 5.82 
1 1  68.01 23.36 2.63 2.33 371.13 6.13 21 .52 6.00 
12 68.13 23.27 3.37 1 .34 433.67 6.1 1  27.63 5.99 
'llote: M.C. - m01sture content; P. . - pe roxtcle value 
Table 1 1 .4 shows physical colour changes in fish flesh. The results showed that Hunter-L, -a and ?
b values of fish were in the range of 44.72-51.92, 1 .19-3.56 and 12.69-14.81 respectively. 
229 
Table 1 1 .4. Results of colour analysis of fish flesh 
Treatments L - value a - value ? b - value 
1 47.72 3.53 14.81 
2 46.72 3.17 14.70 
3 49.69 1 .73 14.07 
4 46.13 3.42 14.82 
5 44.72 3.56 13.74 
6 44.92 2.88 13.34 
7 5 1 .92 1 .69 14.58 
8 49.52 2.00 14.87 
9 49.34 1 .19 12.62 
10 46.84 3.22 14.59 
1 1  49.39 1 .38 13 .29 
12 50.64 1.76 14.44 
The sensory evaluation results for fish and tomato sauce are shown in Tables 1 1 .5 and 1 1 .6 
respectively. The scores of all sensory parameters showed variation, depending on the treatment. 
Table 1 1 .5. Results of sensory evaluation for fish 
Appear Texture Bone- Salti- Spici- Sour- Fishi-
Treatment ance Soft- ness ne ss ness ness 
ness 
1 1 .35 1 .27 0.78 0.81 0.70 1 .14 1 .3 1  
2 1 .43 1 .01 0.87 1 .35 0.97 1 .19 1 .42 
3 2.62 1.26 0.49 1 .48 0.94 1 .39 1 .64 
4 1 .96 1 .28 0.65 1 .49 0.96 1 .29 1 .57 
5 3 .08 1 .00 0.68 1 .34 0.94 1 .08 1 .42 
6 2.08 0.88 0.90 1 .33 0.84 1 .27 1 .28 
7 1 .78 1 .17 0.52 1 .05 0.83 1 .03 1 .66 
8 2.87 1 .03 0.42 1 .30 1 .06 0.97 1 .63 
9 2.58 1 .27 0.42 1 . 1 1  0.98 1 .03 1 .23 
10 2.28 1 .14 0.81 1 .14 0.91 0.93 1 .22 
1 1  1 .86 0.99 0.40 1 .20 0.98 1 .1 3  1 .40 
12 1 .45 1 .21 0.31 1 .08 0.81 0.85 1 .77 
230 
Table 1 1 .6. Results of sensory evaluation for tomato sauce and overall acceptability for canned fish 
product 
Tomato Sauce Overall 
Treatment acceptability 
Colour Mouth Salti- Spici- Sour- Fish- of the 
feel ne ss ness ness ness product 
1 0.96 1 .35 1 .01 0.90 1 .14 1 .42 0.54 
2 0.90 1 .43 1 .13 1 . 1 1  1 .10 1 .42 0.58 
3 0.80 1 .43 1 .15 1 .01 1 .14 1 .34 0.45 
4 1 .01 1 .41 1 .45 0.95 0.99 1 .28 0.60 
5 0.80 1 .17 1 .22 0.86 0.94 1 .48 0.58 
6 0.89 1 .40 1 . 13  0.93 1 .21 1 .34 0.54 
7 0.71 1 .64 0.99 1 .04 1 . 11  1 . 15  0.54 
8 0.73 1 .20 1 .44 1 .16 1 .17 1 .33 0.47 
9 0.83 1 .42 0.99 1 . 1 1  1 .03 1 .19 0.54 
10 0.70 1 .05 1 .16 1 .06 1 .02 1 .29 0.70 
1 1  0.97 1 .33 1 .21 " 1 .02 1 .24 1 . 14 0.60 
12 0.83 1 . 1 1  0.91 0.93 1 .10 1 .17 0.53 
11.4.1.2. Study on the effects 
Tables 1 1 .7 and 1 1 .8 show the effects of each variable on chemical, physical and organoleptical 
?parameters of fish and tomato sauce. 
Sterilization time significantly decreased peroxide value, pH and Hunter-L value of fish flesh. This 
variable also significantly reduced the peroxide value and ideal ratio score of tomato sauce 
spiciness. Sterilization time increased Hunter-a value and the ideal ratio score of bone softness of 
fish, and increased the ideal ratio score of the fish taste of the tomato sauce. The overall 
acceptability of the product was improved by sterilization. 
The application of steam vacuum sealing reduced the peroxide value of fish and tomato sauce. The 
ideal ratio score for tomato sauce spiciness was reduced by that variable as well. Steam vacuum 
sealing improved the ideal ratio score of tomato sauce colour. 
Brining for 10 minutes in 15% salt solution reduced fish moisture content, but increased salt 
231 
content, peroxide value, the ideal ratio score for saltiness and spiciness. The brining treatment also 
increased the peroxide value, and ideal ratio scores for tomato sauce saltiness and spiciness. 
Table 1 1 .7. The main effects and significance levels of process variables on the characteristic of 
canned fish 
Sample Response variables 
Moisture content 
Protein content 
Fat content 
Salt content 
Peroxide value 
pH 
Hunter value: 
Fish L 
a 
b 
Sensory: 
appearance 
texture 
bone softness 
saltiness 
spiciness 
sourness 
fishiness 
Peroxide value 
pH 
Sensory: 
Tomato colour 
sauce mouth feel 
saltiness 
spiciness 
sourness 
fishiness 
Canned Overall acceptability 
fish 
"!ote: a - t-test s1 gn itlcance level at YO o 
b = t-test significance level at 95% 
c = t-test significance level at 99% 
SP 
-1.21 
0.66 
1 .49 
-1 .06 
-3.81 c 
-4.68 c 
-5.61 c 
14.70 a 
0.81 
-0.32 
-0.70 
7.23 
0.38 
-0.75 
0.80 
-2.21 b 
-6.97 c 
-1 .24 
1 .19 
-0.42 
0.87 
-2.03 a 
-1.12 
3 .72 b 
1 .96 a 
svs Bm PC 
-0.01 -1 .71 a -2.64 b 
0.61 -1 .27 -0. 1 1  
-0.86 0.51 0.95 
1 .25 10.41 c 0.70 
-1 .67 a 2.08 a -0.16 
-0.19 1 .12 -2.81 b 
-0.85 -0.23 1 .65 a 
0.13 0.45 -0.69 
-1 .20 1 .07 0.33 
0.52 0.23 -0.42 
1 .27 -0.18  1 .47 
-1 .39 0.10 1 .80 a 
0.28 1 .96 a -1 .26 
0.21 1 .92 a -0.69 
1 .3 1  0.80 0.19 
-0.82 0.42 -1 .18 
-4.34 c 2.39 b -7.66 c 
-0.03 0.37 -2.50 b 
1 .86 a 0.27 -1 .01 
0.37 -0.32 0.92 
0.57 2.73 b -1 .97 a 
-1 .67 a 2.38 b 1 .67 a 
-0.66 0.37 -0.32 
0.61 0.20 0.29 
-0.24 0.62 0.14 
SP - stenuzauon Ume 
SVS = steam vacuum sealing 
Bm = brining time 
PC = pre-cooking of fish 
232 
Table 11 .8. The main effects and significance levels of seasonings on the characteristic of canned 
fish 
Sample Response variables 
Moisture content 
Protein content 
Fat content 
Salt content 
Peroxide value 
pH 
Hunter value: 
Fish L 
a 
b 
Sensory: 
appearance 
texture 
bone softness 
saltiness 
spiciness 
sourness 
fishiness 
Peroxide value 
pH 
Sensory: 
Tomato colour 
sauce mouth feel 
saltiness 
spiciness 
sourness 
fishiness 
Canned Overall acceptability 
fish 
'lote: a - t-test s1 gn iti.cance level at 10% 
b = t-test significance level at 95% 
c = t-test significance level at 99% 
GA = garlic addition 
SA = shallot addition 
V A = vinegar addition 
GA SA ? vA 
1 .34 0.81 0.47 
-0.19 0.91 -0.01 
-2.19 b 0.1 1 -0.95 
0.58 1 .07 1 .90 a 
0.05 0.39 -1 .06 
-0.00 1 .12 -1 .12 
-0.56 0.64 -0. 1 1  
-1 .80 a 1 .80 a -2.89 b 
-0.99 0.72 -1 .29 
-0.33 0.36 -0.09 
-1 .23 -0.62 -0.10 
l .l2 -0.10 0.51 
-0.76 -1 .58 a -0.06 
0.37 -0.21 0.21 
0.26 -1 .18 0.10 
-2.01 a -0.54 -1 .66 a 
0.47 1 .43 -4.63 c 
0.30 -0.37 -0.43 
1.31 -1 .19 0.27 
0.42 -0.61 0.74 
0.06 0.57 0.15 
1 .67 a 0.00 0.62 
1 .70 a 0.14 0.03 
0.53 0.29 -3.15 b 
-0.62 0.91 1 .77 a 
233 
Pre-cooking significantly reduced moisture content and pH of fish. Pre-cooking increased the 
Hunter-L value and the ideal ratio score for bone softness of fish. During pre-cooking the peroxide 
value, pH and ideal ratio score of tomato sauce saltiness decreased, while the ideal ratio score of 
tomato sauce spiciness increased. 
Seasonings affected chemical, physical and organoleptical properties of fish and tomato sauce. 
Garlic reduced Hunter-a value of fish and the ideal ratio score of fishiness. Garlic increased the 
ideal ratio score of spiciness and sourness. Shallot increased Hunter-a value of the fish, but 
decreased the ideal ratio score of fish saltiness. The salt content of fish increased with vinegar 
addition. The vinegar addition reduced Hunter-a value of fish and ideal ratio score of fishiness. 
Vinegar addition resulted in the reduction of peroxide value and ideal ratio score for the tomato 
sauce fishiness. In addition, the added vinegar in canned fish improved the overall acceptability 
of the product. 
11.4.2. Optimization of canning process 
The above results indicate that all seasoning materials should be retained in the tomato sauce 
formula Sterilization time needed to be optimized in terms of overall acceptability of the product. 
Pre-cooking and vacuum headspace in the can are essential, but brining treatment is unnecessary. 
The panellists complained that the tomato sauce was too salty for them, so the salt level in the 
- sauce was varied during the optimization experiment. Thus optimization would concentrate on two 
variables: sterilization time and salt level in the tomato sauce. Optimization study for the 
sterilization time were 40, 50 and 60 minutes, while the salt levels in tomato sauce investigated 
were 1 .5 and 2.5%. 
The results of the optimization experiment are shown in Tables 1 1.9, 1 1 .10 and 1 1 . 11 .  
-Table 1 1 .9. Chemical and physical changes in fish and tomato sauce during optimization 
experiment 
Salt in Sterilization time (minutes) 
Sample Analysis tomato 
sauce 40 50 60 
(%) 
Moisture content 2.5 67.41 67.52 67.00 
(%) 
1 .5 67.55 67.76 66.74 
2.5 1 .80 1 .90 1 .86 
Salt content (%) 
1 .5 1 .46 1 .60 1 .56 
2.5 6.05 6.09 6.02 
pH 
1.5 6.07 5.98 5.95 
Fish 2.5 48.64 44.63 45.23 
L 
1 .5 46.52 44.38 44.52 
2.5 2.33 3 .38 3.65 
a 
1 .5 2.41 3 .35 3.54 
Hunter 
value 2.5 14.10 14.17 14.48 
b 
1 .5 13 .62 13 .87 14.55 
2.5 6.00 5.89 5.88 
pH 
1.5 5.98 5.90 5.90 Tomato 
sauce Peroxide value 2.5 37.12 29.67 29.36 
(meq/kg) 
1 .5 38.14 24.19 25.26 
234 
Analysis of variance shows that moisture content of fish and pH of tomato sauce were not 
significantly affected by sterilization time and salt level in tomato sauce. Tomato sauce with 2.5% 
salt resulted in the fish having pH higher than tomato sauce with 1 .5% salt. Sterilization time 
significantly influenced the Hunter-L,-a and -b values of fish. Fish sterilized for 50 and 60 minutes 
had a lower Hunter-L value than fish sterilized for 40 minutes. Fish sterilization time of 50 and 
60 minutes resulted in a higher Hunter-a value than sterilization of 40 minutes. The fish sterilized 
for 40 and 50 minutes showed a lower Hunter-b value compared to the fish sterilized for 60 
minutes. Fish canned in tomato sauce with 2.5% salt showed a higher salt content than the fish 
235 
canned in tomato sauce with 1.5%. Sterilization time also affected the peroxide value of tomato 
sauce, where tomato sauce with canned fish sterilized for 50 and 60 minutes bad a lower peroxide 
value in comparison with the tomato sauce from the canned fish sterilized for 40 minutes. 
Variables of sterilization time and salt level in tomato sauce did not affect, significantly, the 
sensory properties of the tomato sauce. Both variables did not show any pronounced effect on 
sourness, saltiness, spiciness and fishiness of the fish. However flesh texture and bone softness 
were significantly influenced by sterilization: the texture of fish sterilized for 40 minutes and 50 
minutes were more acceptable than the texture of fish sterilized for 60 minutes. Sterilization time 
of 50 and 60 minutes resulted in fish having softer bone than fish sterilized for 40 minutes. The 
bone softness of fish sterilized for both 50 and 60 minutes was more acceptable than the bone 
softness for fish sterilized for 40 minutes. Statistically, the overall acceptability of the canned fish 
product was not affected by sterilization time, but the product sterilized for 50 minutes seemed to 
be more acceptable compared to the product sterilized for 40 and 60 minutes. 
236 
Table 1 1 . 10. Sensory changes in fish during optimization experiment 
Salt in Sterilization time (minutes) 
Attributes tomato 
sauce (%) 40 50 60 
tender - not 2.5 6.6 6.9 6.4 
tender 
Fish 1.5 6.9 6.8 6.3 
texture 
2.5 7.0 7.2 6.7 
acceptability 
1 .5 7.0 7.2 6.8 
soft - not 2.5 5.4 6.1 6.5 
soft 
Bone 1 .5 4.8 6.4 6.3 
softness 2.5 5.7 6.5 6.9 
acceptability 
1 .5 5.5 6.8 6.9 
sour - not 2.5 5.4 5.1 5.7 
sour 
1 .5 5.0 5.4 5.3 
Sourness 2.5 6.5 6.7 6.7 
acceptability 
1 .5 6.4 6.8 6.2 
salty - not 2.5 5.6 5.9 5.9 
salty 
1.5 5.2 6.0 5.3 
Saltiness 2.5 6.5 6.7 6.5 
acceptability 
1 .5 6.6 6.0 6.5 
-
spicy - not 2.5 6.4 6.3 6.1 
spicy 
1 .5 6.1 6.2 6.1 
Spiciness 2.5 6.8 6.9 6.7 
acceptability 
1 .5 6.4 6.9 6.8 
fishy - not 2.5 5.2 4.7 5.3 
fishy 
1 .5 5.3 5.3 5.2 
Fishiness 2.5 6.5 7.0 6.5 
acceptability 
1 .5 6.7 6.6 6.5 
. 
237 
Table 1 1 . 1 1 .  Sensory changes in tomato sauce and overall acceptability of the product during 
optimization experiment 
Salt in Sterilization time (minutes) ? 
Attributes tomato 
sauce (%) 40 50 60 
Bright red - 2.5 6.1 5.6 6.0 
not bright 
Sauce red 1.5 5.9 6.3 5.7 
colour 
2.5 6.7 6.5 6.6 
Acceptability 
1 .5 6.7 6.8 6.5 
Oily - not 2.5 5.7 5.7 6.1 
oily 
1 .5 6.2 5.9 6.0 Mouth 
feel 2.5 6.2 6.5 5.9 
Acceptability 
1 .5 6.3 6.5 6.4 
Sour - not 2.5 4.9 5.1  5.6 
sour 
1 .5 4.9 5.5 4.8 
Sourness 2.5 6.4 6.4 6.3 
Acceptability 
1.5 6.2 6.5 6.5 
Salty - not 2.5 5.9 6.0 5.8 
salty 
1 .5 5.5 6.1 5.4 
Saltiness 2.5 6.5 6.4 6.4 
Acceptability 
1 .5 6.5 6.5 6.6 
Spicy - not 2.5 6.3 6.3 6.0 
spicy 
1 .5 6.4 6.5 6.4 
Spiciness 2.5 6.8 6.6 6.5 
Acceptability 
1.5 6.7 6.8 6.9 
Fishy - not 2.5 4.9 5.4 5.4 
fishy 
1 .5 5.3 5.1  5.2 
Fishiness 2.5 6.4 6.5 6.1 
Acceptability 
1.5 6.2 6.1 6.6 
2.5 6.5 6.8 6.5 
Overall acceptability of 
the canned fish product 1 .5 6.2 6.7 6.8 
238 
11.5. DISCUSSION 
11.5.1. Changes in canned fiSh during processing 
Brining treatment showed a decreased effect on moisture content and an increased effect of salt 
content of fish. The increment of salt content and the reduction of moisture content of fish are the 
common phenomenon in salting as reviewed by Stansby (1963), Burgess et al (1965), Voskresensky 
(1965) and Van Klaveren and Legendre (1965). The salt penetrates into the fish flesh because of 
the osmotic pressure differences in the fish and the salt surrounding the fish, and, at the same time, 
water is released from the fish. The increase of salt content was also affected by vinegar addition 
in the tomato sauce. Vinegar, which is basically acetic acid, may have accelerated penetration of 
salt from the tomato sauce into the fish flesh. According to Lee (1983), acid can induce 
denaturation of protein. Possibly, acid and sterilization treatments show a synergetic effect on 
denaturation of protein which finally allows the salt easy penetration into the fish flesh. 
Peroxide value of fish and tomato sauce decreased due to the effect of sterilization time and 
vacuum headspace treatment in the can. The optimization experiment indicated that the peroxide 
value of tomato sauce from the canned fish sterilized for 40 minutes was higher than the peroxide 
value of tomato sauce from the canned fish sterilized for 50 and 60 minutes. As discussed in 
Chapter 10, hydroperoxides decompose during heat treatment. Thus the longer the sterilization 
time the greater decomposition effects on hydroperoxides, consequently, the lower the peroxide 
- value in the product. Vacuum headspace in the can provided a much lower oxygen content 
compared to non-vacuum headspace. This condition could limit the production rate of 
hydroperoxides during sterilization. 
Peroxide value of the fish was increased by brining treatment. Salt normally contains not only 
sodium chloride, but impurities such as copper, calcium salts, magnesium salts, sulphates and 
organic matters (Rawson, 1966; Tressler and Lemon, 1960; Burgess et ill, 1965). Salt and copper 
have been known as pro-oxidants which act as catalyst agents in the oxidation process (Stuckey, 
1972; Pokorny, 1987; Buck, 1991). These pro-oxidants may have enhanced the production of 
hydroperoxides in the fish treated with brine. 
Pre-cooking of the fish decreased the peroxide value of the tomato sauce. As mentioned before, 
pre-cooking released water and oil with a high peroxide value. This study showed that the 
peroxide value of fish was significantly higher than the value in the tomato sauce. The pre-cooking 
239 
treatment could reduce the oil which might be released into the tomato sauce during sterilization. 
Thus the total peroxide value in tomato sauce was lower than the value in tomato sauce obtained 
from the canned fish without pre-cooking. The possibility of oil release during sterilization could 
also be used to explain the increase of peroxide value in the tomato sauce due to brining. The 
peroxide value of the tomato sauce also decreased due to vinegar addition. The acetic acid in 
vinegar might have the capability to act as a chelating agent removing pro-oxidants for the sauce. 
This could reduce the oxidation rate in the tomato sauce, and thus lead to the lower hydroperoxide 
content in tomato sauce with vinegar, in comparison with the tomato sauce without vinegar. 
Both sterilization and pre-cooking showed a decreasing effect on the pH of fish, while the 
decreased pH value of the tomato sauce was significantly influenced by pre-cooking. The 
optimization experiment showed that fish in tomato sauce with 2.5% salt and sterilized for 50 
minutes had a markedly higher pH than the fish with the same sauce sterilized for 60 minutes. The 
fish in tomato sauce with 1 .5% salt and sterilized for 40 minutes showed a higher pH, compared 
to the fish of the same sauce sterilized for 50 and 60 minutes. In the canning experiment using 
cat fish, ocean perch and pollack, Paredes and Baker (1987) showed that these fish exhibited 
different pH changes as the results of sterilization effect where pH of both ocean perch and pollack 
increased, while the pH of cat fish decreased. 
The Hunter-L value was decreased by sterilization, but increased by pre-cooking. The effect of 
steriJjz,ati<:>?????? ... no.ted .si?P9fltr?w,4jlec,pre,-cO()king:\?as?Jnsi?lflcant:? ? .. ?f?is;indicated'thllt;tbe 
eilfe.ct::of'sterilization time wa??st:I:onger!than.?the<eff?crof:'Pfe!;?ooking; ':The:bigher?H:uriter;:L value 
- exhibited a lighter colour. The optimization experiment demonstrated that Hunter-L values of fish 
sterilized for 50 and 60 minutes were lower than the value in the fish sterilized for 40 minutes. 
This indicated that the longer sterilization time resulted in a darker fish colour. Sterilization time 
with shallot addition revealed an increasing effect on Hunter-a value at a significant level of 90% 
for each. However garlic and vinegar additions showed a decreasing effect on the Hunter-a value 
at significant levels of 90% and 95% respectively. The optimization experiment registered that the 
fish sterilization for 50 and 60 minutes had a higher Hunter-a value than the fish sterilized for 40 
minutes. This indicated that the longer sterilization time tended to give a reddish colour to the fish. 
Hunter-b value, which was not significantly effected by any treatment in the Plackett and Burman 
experiment, was markedly influenced by sterilization time in the optimization experiment Hunter-b 
values of fish sterilized for 40 minutes and 50 minutes were lower than the values of the fish 
sterilized for 60 minutes. That result indicated that the longer sterilization time produced fish with 
yellowish colour. In terms of Hunter values, Paredes and Baker (1987) showed a similar trend in 
the canning of ocean perch and pollack. Investigations by other researchers noted that when fish 
240 
is heated from one direction, three distinct changes of opacity can be discerned. An initial increase 
in translucence is followed by two successive increases in opacity. The second increase in opacity 
is due to precipitation of thermally denaturated sarcoplasmic proteins, which appears to begin about 
45?C (Aitken and Connell, 1979). The increase in intensity of redness might be due to oxidation 
of iron pigments from ferrous to ferric derivates (English et al, 1988). A study in canned sardine 
revealed that pigments involved in the browning reaction were primarily lipophilic, and the 
presence of 02 and fat were necessary for these reactions. It was found that volatile amines, such 
as trimethylamine, dimethylamine and ammonia react directly with oxidized lipids. The reactions 
were accelerated by high temperature (Wakao and Pazos, 1984). The Hunter-a and -b values could 
be affected by oxidation of the high myoglobin to metmyoglobin, a dark brown pigment, or by 
trimethylamine ('IMAO) promoting the formation of the disulphide bond between denatured 
myoglobin and cysteine producing a green pigment (Francis and Clydesdale, 1975; English et al, 
1988). 
Panellists revealed that sterilization and pre-cooking significantly increased the bone softness of 
fish. Solubilization of organic matter of bone causes softening of fish bone due to heating process 
(S?$etiadi,? 1977; Ish?Wa' et;:illix?2.8q:) .. , ,I:Jle;il1orgafiic?watter: .. in<>ludes?calci?;fpho:%P?et'?us, 
magn?siiJ:In? .cb.l?ne,ana?fluoiine tsoe,stmaru;?19.77)Y.Pre"coo1Q.rrg.ma)i,.ihave.fesillt&l'in preliminm-y 
organic matter solubilization, further solubilization occurred duiing sterilization. Acetic acid proved 
to accelerate the solubilization rate of organic matter from the fish bone (Ishikawa et al, 1 989). 
However the panellist did not detect the effect of vinegar addition on fish bone softness in this 
study. 
Brining treatment on fish significantly increased the saltiness of both the fish and tomato sauce. 
The higher saltiness of tomato sauce from the process with brining treatment compared to the sauce 
from the canned fish without brining treatment might be due to the less salt penetration from 
tomato sauce into the fish canned after brining, compared to canned fish without brining treatment 
Brining treatment caused the fish to have a higher salt content, which finally reduced the 
penetration rate of salt from tomato sauce into the fish during sterilization. Shallot addition 
decreased the saltiness in fish. This might be because flavour compounds of shallot showed a 
stronger performance than flavour compounds of salt Pre-cooking decreased the saltiness of the 
tomato sauce. Reduction of moisture content of the fish, due to pre-cooking, may have provided 
greater possibility for salt in tomato sauce to penetrate into the fish. This resulted in the tomato 
sauce with a less salty taste. However the increase of salt content caused by pre-cooking treatment 
was noted as insignificant. 
241 
Brining improved the spicy properties of both fish and tomato sauce, but sterilization time and 
vacuum headspace decreased the spiciness properties of tomato sauce. Sterilization may have 
induced decomposition of the spicy flavour compounds into smaller compounds, which could be 
volatile by heat during sterilisation, giving a less spicy taste to the tomato sauce. Pre-cookillg 
treatment of fish could improve the spiciness of the tomato sauce. One of the purposes of pre?
cooking is to develop desirable flavour properties (Codex Alimentarius Commission, 1976; Warne, 
1988). By removal of undesirable flavour, such as fishy flavour during pre-cooking, the existence 
of a spicy flavour would be significant. Garlic significantly increased the spiciness of the tomato 
sauce. The principal chemical properties in garlic are diallyl disulphide, allicin, alliin and ajoene 
(Kritchevsky, 1991). Garlic also increased the sourness of the tomato sauce. 
The fishiness of the fish was decreased as the result of sterilization, but the fishiness of the tomato 
sauce increased during this treatment. During sterilization, the fishy flavour compounds of the fish 
might be trapped in the tomato sauce showing an increase in fishiness. Garlic and vinegar 
additions decreased the fishiness of the fish, while the fishiness of the tomato sauce significantly 
decreased due to the effect of vinegar addition. This result indicated that garlic and vinegar have 
the capability to neutralize the fishy flavour in canned fish with disguised fish oil. 
Sterilization time and vinegar addition in tomato sauce significantly increased the overall 
acceptability of the canned fish product. The above results show that sterilization had a significant 
effect on the bone softness of the fish. Thus, sterilization was necessary to the optimization 
experiment. Vinegar addition had a significant effect on the reduction of the fishiness of the fish. 
-The vinegar addition in tomato sauce had to be retained in further experiments. Brining treatment 
was removed from the fish canning process, since this treatment increased peroxide value of the 
fish, and saltiness in the tomato sauce. By removing the brining treatment, the salt in the tomato 
sauce had more opportunity to penetrate into the fish during sterilization. This was expected to 
produce a more acceptable salty taste in the tomato sauce and fish. Pre-cooking of fish and 
vacuum headspace in the can as well as garlic and shallot addition in the tomato sauce are retained 
in the canning process, because they showed desirable effects on acceptability of the canned fish 
product. 
11.5.2. Canning process optimization 
Sterilization significantly affected the acceptability of fish flesh texture and bone softness. The 
texture of fish sterilized for 40 and 50 minutes was more acceptable than the texture of fish 
242 
sterilized for 60 minutes. Bone softness of fish sterilized for 50 and 60 minutes was more 
acceptable than the bone softness of fish sterilized for 40 minutes. By considering both texture and 
bone softness acceptability, the sterilization time of 50 minutes was regarded as the optimal 
sterilization time for canned fish with fish oil disguised in it . Generally, both flesh texture and 
bone softness are the most important factors in determining the acceptability of canned fish. The 
medium is also an important factor in the acceptance of canned fish by the consumer as indicated 
by the result of consumer survey. However the optimization experiment did not reveal any 
significant effect on sensory properties of the tomato sauce due to sterilization time and salt level. 
Sterilization time and level of salt addition did not significantly influence the overall acceptability 
of the canned fish product However the trend showed that the sterilization time of 50 minutes 
generally produced canned fish with a better overall acceptability than the other two sterilization 
times. Thus, all of the above explanations suggested to the use of 50 minute sterilization time for 
processing canned fish with fish oil disguised in it. Directorate General of Fishery (1984) 
recommended sterilization of canned sardine in tomato sauce at ll7?C for 54 minutes. However 
the previous experiment. Chapter 10, proved that that temperature had less protection to ro-3 fatty 
acids in fish oil. 
The salt levels of 1.5 and 2.5% in tomato sauce did not induce pronounced differences on the 
sensory properties of fish and tomato sauce. In terms of economic consideration, the salt level of 
1 .5% was recommended for the tomato sauce in processed canned fish with disguised fish oil. 
11.6. CONCLUSIONS 
The Plackett and Burman design proved to be very useful in identifying the important factors in 
the processing of canned fish with disguised fish oil. Pre-cooking, vacuum headspace, sterilization 
time, garlic, shallots and vinegar auditions were considered as the important factors. However 
sterilization time and salt level in the tomato sauce needed optimization. 
The optimization experiment recommended sterilizing the canned fish with disguised fish oil at 
12l.l?C for 50 minutes. The optimum salt level in the tomato sauce was found to be 1.5%. 
243 
Chapter 12 
PROSPECT OF CANNED FISH PRODUCT WITH FISH Oll.. 
ADDITION IN INDONESIAN MARKET 
12.1. BACKGROUND 
The consumer survey indicated that the canned fish product being developed had good prospects 
in the Indonesian market, where approximately 80% of respondents intend buying the product. The 
supermarket survey revealed that this product was superior to the existing product, since the 
product was nutritionally better. During the development study, the sensory panel results suggested 
that the product was organoleptically acceptable. In the development process the product must now 
be exposed to a wider evaluation, using the Indonesian consumer, the target market, where more 
reliable information about the suitability of the product could be ascertained. 
Measuring consumer response to products is considered as a critical part of the development 
process, so now major emphasis is given to this activity. Acceptance testing will indicate whether 
the product can be marketed, or improvement is needed (Dethmers, 1981). 
The most practical approach to predicting consumer acceptance is through the use of sensory 
_panels. Sensory panels measure human responses to sensory stimuli in food products. Consumers 
perceive product characteristics through the senses. Sensory cues, along with behavioral influences, 
provide the consumer with a basis for a judgemental value of acceptance or rejection (Ellis, 1970). 
The consumer may not have the level of skill for a specialized sensory taste; but, the consumer can 
provide information not obtainable, in an unbiased form, from the trained panellists; for example, 
preference, purchase intent, and so forth. All participants are important and have something to 
contribute (Stone, 1988). 
A variety of tests may be employed, such as central location tests where the conditions are more 
controlled, or home placement tests where consumers can use the product under realistic conditions 
(Lyon, 1991). In this study, home placement tests were used as more questions could be asked and 
information could be obtained regarding the product and consumer attitudes toward product price, 
package label, etc; since the respondents are given enough time to answer the questionnaire 
(Gatchalian, 1981). 
244 
As the product was nutritionally better, containing a higher polyunsaturated fatty acid than existing 
products, particularly ro-3 fatty acids, a medical doctor survey was undertaken to reveal the medical 
perception of the developed product. The survey indirectly introduced the product to the medical 
doctors who were expected to contribute in the introduction of the health benefits of the developed 
product to the public, especially patients, who could benefit from its use. 
12.2. OBJECTIVES 
As the final stage in the process of product development, the objectives of this study were: 
* to investigate the physical and chemical changes of the product during production trial; 
* to show the acceptance of the product by Indonesian consumers in order to assess the 
prospect of the product in the market; and 
* to show the response and comment from the Indonesian medical doctors about the 
product and the prospect of the product from the medical point of view. 
- 12.3. METHODOLOGY 
12.3.1. Materials 
Fish, fish oil, tomato paste, salt, sugar, vinegar, shallots, garlic and cans were the same as used for 
the experiment in Chapter 1 1 .  
245 
12.3.2. Methods 
12.3.2.1. Production trial 
The canned fish was processed, using the method described in Chapter 1 1, by employing 
sterilization at 121.1 oc for 50 minutes. The product was processed at full capacity of pilot plant 
steamer for pre-cooking and retort (188 cans) in the Food Technology Department, Massey 
University. The tomato sauce formula used was 18. 16% tomato paste, 45.10% water, 27.25% fish 
oil, 1 .43% salt, 3 .00% sugar, 1 .90% shallot, 1 .90% garlic and 0.95% vinegar. Weight, physical and 
chemical changes of fish, tomato sauce and the canned fish product during the production trial were 
determined. 
12.3.2.2. Consumer test for product acceptability 
Consumer testing of the developed canned fish was undertaken using the home placement test 
method in five cities in Java, Indonesia where approximately 60% of the Indonesian population 
lives, (Central Bureau of Statistic, 1990). 
The samples for the consumer test were prepared in the Food Technology Department, Massey 
University, and flown to Indonesia for testing. Only one sample was given to each household or 
family. Three copies of the product testing questionnaire were given to each family, and one of 
- the questionnaires was designed for the housewife or the person in charge of shopping and/or 
determining the daily family menu. 
The samples were given to 158 families including 40 families in Jakarta (capital city of Indonesia), 
30 families in Tangerang (West-Java), 37 families in Semarang (capital city of the Central-Java 
province), 26 families in Sragen (Central-Java) and 25 families in Lumajang (East-Java). The 
families were given freedom to organise their own testing by following directions on the 
questionnaire. They were asked to return the questionnaire to six survey coordinators, one 
coordinator in each city except Jakarta, where there were two coordinators. The questionnaire, as 
shown in Appendix 12.1, was designed as simply as possible by using a simple 5-scale hedonic 
method in "just right" form, so it could be completed by respondents who were able to read and 
write. The questionnaire prepared for the housewife or a person responsible for shopping and/or 
preparing the daily family menu included some supplementary questions. 
246 
Both types of questionnaires, in number, were returned as listed: 
City Family Housewife Total questionnaire 
(number) (number) (number) 
Tangerang 30 30 81 
Jakarta 37 36 104 
Semarang 34 32 93 
Sragen 26 23 60 
Lumajang 24 22 64 
Return rate 95.6% 90.5% 84.4% 
12.3.2.3. Medical doctor survey 
The survey was used to reveal the attitudes of medical doctors to fish oil and the comments about 
the product being developed. The questionnaire, as shown in Appendix 12.2, was distributed to 
doctors including pathologists, paediatricians, internists, obstetricians and general practitioners at 
the "Syaiful Anwar" general hospital, Malang, East-Java, the Medical Faculty of Brawijaya 
University, Malang, East Java, and public health service centres and general hospitals in the South 
Kalimantan Province. The questionnaire was also distributed to the general practitioners attending 
an obstetrics and gynaecology course in Malang, East-Java, in 1990. The responses to the survey 
"numbered 163. 
12.4. RESULTS 
12.4.1. Changes during production trial 
12.4.1.1 .  Weight changes 
The initial raw fish weight was 35.51 kg. After preparation treatment involving heading, gutting 
and cutting, the net weight was 25.75 kg, a 27,49% weight loss. 
247 
The changes in fish weight in ten cans was monitored during the production trial. The changes are 
shown in Table 12.1 .  
Table 12.1 .  Fish and canned fish product weight changes during production trial 
Can Initial fish Pre-cooked Total fish Tomato sauce 
No. weight fish weight and sauce weight 
(g) (g) weight (g) (g) 
1 1 19.1 103.2 166.1 62.9 
2 121 .8 105.2 168.0 62.8 
3 128.8 1 14.3 165.4 51 . 1  
4 125.4 108.3 169.5 61.2 
5 127.8 1 1 1 .5 164.7 53.2 
6 .  130.6 1 14.5 166.3 51 .8 
7 124.3 106.3 173.9 67.6 
8 130.7 1 14.2 171.0 56.8 
9 131.9 1 14.9 170.3 55.4 
10 127.1 1 10.6 169.2 58.4 
X ? SD 126.8 ? 4.1 1 10.3 ? 4.3 168.4 ? 2.9 58.1 ? 5.4 
The average initial fish weight in each can was 126.76g. Pre-cooking by steaming reduced the fish 
weight to 1 10.31g. Fish weight loss due to steaming treatment was approximately 13%. The 
average weight of tomato sauce added to each can was 58.14g, approximately 34.5% of product 
net weight. 
12.4.1.2. Colour changes 
Colour changes in both fish flesh and tomato sauce medium were measured using HunterLab 
colour, and the results are shown in Table 12.2. 
248 
Table 12.2. Hunter-L, -a and -b values changes in both fish flesh and tomato sauce medium during 
production trial 
Sample L a b 
Raw fish 34.15 4.21 8.86 
Steamed fish 50.25 1 .19 12.39 
Sterilized fish 44.47 4.66 13.53 
Tomato sauce medium before 26.44 12.10 8.32 
sterilization 
Tomato sauce medium after 25.84 7.63 7.04 
sterilization 
Blend of fish and medium of canned 44.19 7.03 13.20 
fish before sterilization 
Blend of fish and medium of canned 37.47 9.23 17.28 
fish after sterilization 
In terms of the fish colour changes, pre-cooking increased Hunter-L and -b values, but reduced 
Hunter-a values. Sterilization decreased Hunter-L value of steamed fish, but significantly increased 
Hunter-a value. The Hunter-b value increased slightly. Sterilization also decreased the Hunter-L 
value of the blend of fish, but medium increased both Hunter-a and -b values. Hunter-L, -a and -
b values of tomato sauce medium decreased due to sterilization. 
12.4.1.3. Chemical changes 
As shown in Table 12.3 pre-cooking caused a significant change in the proximate composition of 
fish, especially in the significant decrease in moisture and ash content, and the increase of protein 
and carbohydrate contents. Sterilization did not markedly affect the proximate composition of pre?
cooked fish. The oil addition to the tomato sauce medium significantly increased fat and 
carbohydrate contents, and reduced the protein content of the product. 
249 
Table 12.3. Proximate composition changes in the canned fish during production trial (%) 
Samples 
Parameters 
Pre-cooked Sterilized Whole 
Raw fish fish fish canned 
fish 
Moisture content 71 .8 69.4 69.4 68.2 
Protein content 22.2 23.7 23.2 14.6 
Fat content 2.4 2.9 3.0 10.1 
Carbohydrate 0.5 1 .5 1 .9 4.4 
content (by 
difference) 
Ash content 3.1 2.5 2.5 2.7 
Stability of the oil in the tomato sauce was studied by extracting the oil and analyzing for peroxide, 
anisidine, TBA and totox values. The results are shown in Table 12.4. The increase of anisidine 
and TBA values were noted, while both peroxide and totox values registered a decreasing pattern. 
Table 12.4. Results of stability study on the oil in tomato sauce medium due to sterilization 
treatment during production trial 
Tomato sauce 
Parameter 
Before sterilization After sterilization 
Peroxide value 49.5 19.4 
(meqlkg) 
Anisidine value 18.0 41 .2 
TBA value 19.1 29.7 
Totox value 1 17.1 80.0 
Table 12.5. Fatty acid proflle changes in canned fish due to sterilization treatment during 
production trial 
Samples SAFA MUFA PUFA ro-3 FA EPA DHA 
Fish oil 28.4 48.0 23.5 20.4 6.6 8.3 
Unsterilized 29.0 45.0 26.0 22.6 7.7 8.8 
tomato sauce 
Sterilized 29.4 44.5 26.0 22.7 7.8 8.9 
tomato sauce 
Raw fish 44.9 25.4 29.7 24.1 5.7 15.2 
Sterilized fish 41.1  26.2 32.4 28.3 7.1 17.2 
Unsterilized 33.7 39.6 26.6 23.0 7.5 10.1 
canned fish 
product 
Sterilized 33.2 40.7 26.1 22.8 7.6 9.9 
canned fish 
product 
250 
Fatty acid proflles of fish, tomato sauce and the canned fish product during production trials are 
shown in Table 12.5. Fatty acid proflles of fish oil and tomato sauce with fish oil disguised in it 
showed differences in the relative amounts of monounsaturated fatty acid (MUF A), polyunsaturated 
fatty acids (PUFA) and ro-3 fatty acids. The relative amount ofMUFA of tomato sauce was lower 
. than the amount in fish oil, but the relative amount of PUFA and ro-3 fatty acid were higher than 
measured in the fish oil. The fatty acid proflles of tomato sauce were insignificantly changed 
during sterilization, but sterilization treatment increased significantly the relative amount of PUF A 
and ro-3 fatty acids, of fish, and decreased significantly the relative amount of SAF A. Fatty acid 
proflles of the whole canned fish product seemed relatively constant during sterilization. 
12.4.2. Safety assessment of developed canned fish 
As the canned fish produced from the production trial would also be used for consumer product 
testing, product safety had to be assured before releasing it to the Indonesian consumer. Sterility 
and incubation tests were used for this purpose. 
251 
The sterility test showed that no microorganisms grew on the nutrient agar plates during aerobic 
and anaerobic incubation at 30?C and 55?C. The incubation test performed by incubating five cans 
each at 30?C and 55?C for 14 days indicated that the can appearance was still in noqnal condition 
until the end of incubation period, and no undesirable changes, such as swelling were observed. 
Thus, sterilization and incubation test results revealed that the canned fish product was safe for 
consumption. 
12.4.3. Product acceptability during consumer testing 
12.4.3.1. Overall results for canned fiSh characteristics and acceptability 
The frequencies of the "just right" score for each sensory characteristic of the developed canned 
fish product are shown in Table 12.6. 
More than half of the respondents said that the texture of the fish flesh and bone softness was "just 
right", while a small number of respondents commented that the texture of fish flesh and bone was 
very strong. About 30% of respondents thought that the sauce colour was bright red, while 54% 
said that the sauce colour was "just right". Only 9% of respondents indicated that the sauce colour 
was dark red. 
- In terms of the taste of both fish flesh and sauce, over 50% of respondents said that the sourness, 
saltiness and spiciness of both fish flesh and tomato sauce medium were "just right", while 
approximately 20-35% of respondents commented that the fish flesh and tomato sauce lacked of 
sour, salt and spice taste. Only a minority of respondent thought that the product was very fishy. 
The respondents saying that the fish flesh was slightly fiShy, "just right" and slightly non-fishy 
were 39%, 32% and 14% respectively. Respondents commenting that the tomato sauce medium 
was slightly fishy, just right and slightly non-fishy were 47%, 24% and 14% respectively. Most 
of the respondents said that the mouth feel of the tomato sauce medium was slightly oily (45%) 
and "just right" (31% ). 
With regard to overall acceptability of the canned fish product, a minority of respondents (18%) 
did not like the developed product. Approximately 38% of respondents liked the product, while 
approximately 45% neither liked nor disliked. 
-252 
Table 12.6. Canned fish characteristics and acceptability in consumer testing 
Score 
Parameter 5 4 3 2 1 
N %?) N % N % N % % 
N 
FISH 
Texture?> 50 13.4 56 1 3 .6 267 66.4 25 6.3 0 0 
Bone 99 24.6 73 18.1 208 5 1 .7 22 5.6 0 0 
softness?> 
Sournessb> 6 1 .5 66 16.4 223 55.5 86 21.4 21 5.2 
Saltinessb> 12 3.0 73 18.1 224 55.8 84 20.9 9 2.2 
Spiciness b) 5 1 .2 18 4.5 235 58.4 1 31  32.7 13 3.2 
Fishiness b) 42 10.4 157 39.0 127 3 1 .6 55 13.8 21 5.2 
TOMATO 
SAUCE 
Colour> 9 2.2 1 15 28.6 218 54.2 49 12.3 1 1  2.7 
Mouthfeeld) 60 14.9 181 45.0 123 30.6 28 7.0 10 2.5 
Sourness b) 8 2.0 54 13.4 214 53.2 1 10 27.4 16 4.0 
Saltinessb> 15 3.7 66 16.4 202 50.3 109 27.1 10 2.5 
Spicinessb> 3 0.7 25 6.2 216 53.8 144 35.8 14 3.5 
Fishiness b) 30 7.5 189 47.0 98 24.4 58 14.4 27 6.7 
CANNED 
FISH 
Overall 40 10.0 109 27.1 1 80 44.8 62 15.4 1 1  2.7 
acceptabi 
lity") 
'lote: 
*) row percentage 
a) score 5 very soft, 4 slightly soft, 3 just right, 2 slightly tough, 1 very tough 
b) score 5 very strong, 4 slightly strong, 3 just right, 2 slightly lacking, 1 very lacking 
c) score 5 very bright red, 4 slightly bright red, 3 just right, 2 slightly dark red, 1 very dark 
red 
d) score 5 very oily, 4 slightly oily, 3 just right, 2 slightly non-oily, 1 very non-oily 
e) score 5 like very much, 4 like slightly, 3 neither like nor dislike, 2 dislike slightly, 1 dislike 
very much 
253 
12.4.3.2. Acceptability of developed canned fish 
The level of acceptability of the canned fish product in terms of demographic characteristics is 
shown in Table 12.7. 
Respondents from Tangerang, Semarang and Lumajang showed a similar response to the 
acceptability of the developed product. Most respondents (47-57%) said that they neither liked nor 
disliked the product. Respondents who did not like the product were less than 15%, while 62% 
of the respondents from Sragen liked the product and 10% disliked the product. Jakarta showed 
a high number of respondents (48%) who did not like the product, while only 20% liked the 
product. 
The survey results showed that the acceptability of the developed canned fish was independent of 
sex. The product was more acceptable to the consumer under 40 years of age than the consumer 
over 40 years. 
In terms of respondent career, all career groups showed a lower percentage of respondent who did 
not like the product. The respondents who liked the product were mostly private sector workers, 
government officials and housewives, but less for pupil and college/university students. Only a 
minority of consumers from each family income bracket did not like the product. Therefore, 
consumers from all family income brackets showed as potential consumers . 
. When considering the total number of consumers, the results show that the consumer accepted the 
developed canned fish product. The mean score of overall acceptability is 3.26, or close to "like 
slightly" category. 
254 
Table 12.7. Acceptability of developed canned fish product in consumer test by demographic 
characteristics 
Score 
Demographic 
characteristics 5 4 3 2 1 
N %') N % N % N % N % 
Location: 
Tangerang 1 1 .2 24 29.6 46 56.8 8 9.9 2 2.5 
Jakarta 2 1 .9 19  1 8.3 43 4 1 .3 33 3 1 .7 7 6.7 
Semarang 9 9.7 27 29.0 44 47.3 1 1  1 1 .8 2 2.2 
Sragen 15 25.0 22 36.7 l7 28.3 6 10.0 0 0 
Lumajang 13  20.3 l 7  26.6 30 46.9 4 6.2 0 0 
Sex: 
Male l l  7.1  49 3 1 .8 7 1  46.1 22 14.3 1 0.6 
Female 29 1 1 .7 60 24.2 109 43.9 40 16.2 10 4.0 
Age group 
(years): 
15 - 20 10 20.8 12  25.0 2 1  43.8 5 10.4 0 0 
2 1 - 30 13  6.9 50 26.5 86 45.5 35 1 8.5 5 2.6 
3 1 - 40 7 8.3 33 39.3 33 39.3 9 10.7 2 2.4 
4 1 - 50 5 10.9 9 19.6 24 52.2 6 1 3.0 2 4.3 
> 50 5 14.3 5 14.3 16 45.7 7 20.0 2 5.7 
Career: 
Student (pupil 9 1 3.8 lO 15.4 35 53.8 10 1 5.4 1 1 .6 
- tertiary) 
Private 1 3  6.8 59 3 1 .1 87 45.8 27 14.2 4 2.1  
sector 
Government 6 1 3.7 1 4  3 1 .8 17  38.6 7 15.9 0 0 
official 
Housewife/ 12  1 1 .7 26 25.2 4 1  39.8 1 8  17.5 6 5.8 
family helper 
Family income 
(xRp.lOOO.-): 
< 150 9 1 8.0 1 8  36.0 16  32.0 7 1 4.0 0 0 
150 - 299 1 8  1 6.4 24 21 .8  49 44.5 1 8  1 6.4 1 0.9 
300 - 500 6 5.9 28 27.5 54 52.9 12  1 1 .8 2 1 .9 
> 500 7 5.0 39 27.9 61  43.6 25 17.8 8 5.7 
ote: '") row percentage 
5 like very much, 4 like slightly, 3 neither like nor dislike, 2 dislike slightly, 1 dislike very much 
-255 
12.4.3.3. Consumer buying trend of developed canned fish 
The survey results of the buying trends in terms of demographic characteristics are shown in Table 
12.8. More than 70% of respondents from Tangerang, Sragen .and Lumajang intended to buy the 
developed product, and about 56% of respondents from Semarang indicated likewise. However 
more than 55% of respondents from Jakarta did not intend to buy the product. 
Table 12.8.  Buying trend of developed canned fish in consumer testing by demographic 
characteristic 
Buying intention 
Demographic 
Yes No Characteristics 
N %?) N % 
Location: 
Tangerang 21 70.0 9 30.0 
Jakarta 16 44.4 20 55.6 
Semarang 1 8 56.3 14 43.7 
Sragen 17 73.9 6 26.1 
Lumajang 17 77.3 5 22.7 
Age (years): 
15 - 30 35 55.6 28 44.4 
3 1  - 40 29 72.5 1 1  27.5 
41 - 50 14 63.6 8 36.4 
> 50 1 1  61.1  7 38.9 
Career: 
Private sector worker 39 62.9 23 37.1 
Government official 9 56.3 7 43.7 
Housewife/ family helper 41 63.1 34 36.9 
Family income (xRp.lOOO.-): 
<150 8 44.4 10 55.6 
150 - 299 25 62.5 1 5 37.5 
300 - 500 26 72.2 10 27.8 
>500 30 61.2 19 38.8 
"'lote: '?') row percentage 
256 
Respondents from all age groups showed a promising response to the developed product, where 
55-73% of respondents from each age group intended to buy the developed product. Respondent 
careers did not significantly affect the buying trend, since more respondents from each career group 
wanted to buy the product than respondents not intending to buy the product. 
Family income affected the buying trend. Mostly the respondents from middle and high income 
brackets planned to purchase the product. Approximately 56% of respondents from low income 
group decided not to buy the product. 
Table 12.9 shows the buying trend according to respondent opinion about product acceptability and 
respondent experiences in buying canned fish. All respondents who liked the product very much 
wanted to buy the product. On the other hand, all respondents who did not like the product very 
much did not intend to purchase. The respondents commenting "like slightly" and "neither like nor 
dislike" were approximately 85% and 59% respectively, planned to buy the product. 
Approximately 15% of respondents who "did not like the product slightly" decided to buy the 
product. 
For the respondents who have consumed present existing canned fish products, canned fish in 
tomato sauce and canned sardine products, more than 65% of them want to buy the product. More 
than 55% of respondents who have not consumed canned fish in tomato sauce and canned sardine 
intend to buy the developed canned fish product. However about 61% of respondents who have 
not bought any canned fish products decided not to buy this product. 
257 
Table 12.9. Buying trend of developed canned fish according to consumer testing acceptability and 
consumer experience in buying canned fish products 
Buying intention 
Information Yes No 
N %*) N % 
Acceptability: 
Like very much 16 100 0 0 
Like slightly 33 84.6 6 15.4 
Neither like nor dislike 37 58.7 26 41.3 
Dislike slightly 3 15.0 17 85.0 
Dislike very much 0 0 5 100 
Consume canned fish: 
Yes 80 66.7 40 33.3 
No 9 39.1 14 60.9 
Consume canned fish in 
tomato sauce: 
Yes 45 66.2 21 3 1 .8 
No 44 57.1 33 42.9 
Consume canned sardine: 
Yes 34 69.4 15 30.6 
No 55 58.5 39 41 .5 
'llote: *) row percentage 
Response to buying criterion, retail outlets, label information and selling price of the product are 
shown in Table 12. 10. A majority of consumers agreed that the major reason for purchasing the 
developed product was convenience: the product is easy to store and serve. Only 28% and 17% 
of consumers considered buying the product because of its nutritional value and health benefit 
respectively. 
258 
Table 12.10. Buying criterion, retail outlet, label information and price of product suggested by 
consumer testing 
Consumer 
Information 
%*) Number 
Buying criterion??> 
Family preference 9 10.1 
Convenience 65 73.0 
Like to eat 8 9.0 
Nutritional value 25 28.1 
Reasonable price 15 16.9 
Health benefits 1 5 16.9 
Retail outlets ??> 
Supeooarket 49 55.1 
Retail shop 23 25.8 
Food shop 21 23.6 
Local market 3 3.4 
Information on label**) 
Brand 8 9.0 
Ingredients 9 10.1 
Composition 17 19.1 
Name and address of 8 9.0 
factory 
Product superiority 14 15.7 
Net weight 1 1  12.3 
All above 57 64.0 
Price (Rp.) 
- < 500 9 10.1 
500 - 650 30 32.7 
65 1 - 800 20 22.5 
801 - 1000 16 1 8.0 
> 1000 5 5.6 
?ote: '?') tlased on the respondents mtendmg to buy the product 
**) One respondent could give more than one answer 
Most of the respondents (55%) suggested selling the product in supermarkets while others 
suggesting selling the product in retail shops and food shops 26% and 24% respectively. A 
majority of respondents said that all information mentioned on the label, as shown on Appendix 
12.3, had to be retained. Some respondents suggested adding an expiry date, "no added 
259 
monosodium glutamate" and "100% halal". The statement "100% halal" means that the product 
can be eaten by Moslems as no forbidden ingredients, according to Islamic law, are used to process 
the product. 
Half the respondents thought that the product should be sold at price between Rp. 500-800. 
12.4.4. Opinions of Indonesian medical doctors to the product 
12.4.4.1. Fish and fiSh oil 
The results of the medical doctor survey, as shown in Table 12.1 1,  indicates that many Indonesian 
doctors have not suggested their patients consume fish oil for health reasons. However more than 
80% of the respondents from all demographic groups have advised patients to consume fish for 
their health. This indicated that the benefits from eating fish have been widely recognized by 
Indonesian medical doctors. 
Table 12. 1 1 .  Medical doctors advising the patients to consume fish and fish oil 
Demographic 
characteristic 
Speciality: 
Pathologist 
Paediatrician 
Internist 
Obstetrician-
gynaecologist 
General 
practitioner 
Experience 
(years): 
<5 
5 - 10 
1 1  - 15 
16 - 20 
> 20 
Number of total 
patient/month: 
< 100 
100 - 300 
301 - 500 
501 - 700 
> 700 
Number of patient 
having heart 
problem/month: 
< 10 
10 - 30 
31 - 50 
> 50 
?ote: *) row rcenta e pe g 
Yes 
N %") 
0 0 
9 42.9 
4 21 .1  
5 23.8 
48 47.5 
40 50.6 
10 41.7 
8 25.8 
2 18 .2 
6 33.3 
10 62.5 
24 39.3 
17 34.0 
8 38.1 
7 46.7 
47 42.3 
13 39.4 
2 15.4 
4 66.7 
Fish oil Fish 
No Yes 
N % N % N 
1 100 1 100 0 
12 57.1 20 95.2 1 
15 78.9 19 100 0 
16 76.2 21 100 0 
53 52.5 87 86.1 14 
39 49.4 68 86.1  1 1  
14 58.3 23 95.8 1 
23 74.2 29 93.5 2 
9 8 1 .8 10 90.9 1 
12 66.7 1 8 100 0 
6 37.5 14 87.5 2 
37 60.7 54 88.5 7 
33 66.0 46 92.0 4 
13 61.9 19 90.5 2 
8 53.3 15 100 0 
64 57.7 99 89.2 12 
20 60.6 30 90.9 3 
1 1  84.6 13 1 00  0 
2 33.3 6 100 0 
260 
No 
% 
0 
4.8 
0 
0 
13.9 
1 3 .9 
4.2 
6.5 
9.1 
0 
12.5 
1 1 .5 
8 
9.5 
0 
10.8 
9.1 
0 
0 
Table 12.12 shows the majority of the doctors surveyed recommended food products as a mean to 
deliver fish oil to consumers in order to increase the fish oil consumption. Only paediatricians and 
medical doctors having patients between 501 - 700 people/month tended to recommend 
encapsulated oil. 
261 
Table 12.12. The ways advised by Indonesian medical doctors to deliver fish oil to consumer 
Tablespoon Disguised 
Demographic Capsule Cooking or direct ? in food 
Characteristics oil consumption products 
N %*) N % N % N % 
Speciality: 
Pathologist 0 0 0 0 0 0 1 100 
Paediatrician 14 66.6 0 0 1 4.8 6 28.6 
Internist 4 44.4 1 1 1 .1 0 0 4 44.4 
Obstetrician- 1 4.8 3 14.2 1 4.8 16 76.2 
gynaecologist 
General 25 24.8 8 7.9 0 0 68 67.3 
practitioner 
Experience 
(years): 
< 5  22 27.8 7 8.9 0 0 50 63.3 
5 - 10 5 20.8 1 4.2 0 0 1 8 75.0 
1 1 - 15 7 22.6 3 9.7 2 6.4 19 61 .3 
16 - 20 3 27.3 0 0 0 0 8 72.7 
> 20 7 38.9 1 5.6 0 0 10 55.5 
Number of total 
patient/month: 
< 100 5 31 .3 2 12.5 0 0 9 56.2 
100 - 300 1 3  21 .3 7 1 1 .5 0 0 41 67.2 
301 - 500 9 18.0 2 4.0 0 0 39 78.0 
501 - 700 1 1  52.4 1 4.8 1 4.8 8 38.0 
> 700 6 40.0 0 0 1 6.7 8 53.3 
Number of patient 
having heart 
problem/month: 
< 10 32 28.8 10 9.0 0 0 69 62.2 
10 - 30 9 27.3 1 3 .0 2 6.1 21 63.6 
3 1 - 50 1 7.7 1 7.7 0 0 1 1  84.6 
> 50 2 33.3 0 0 0 0 4 66.7 
"'ote: *) row percentage 
262 
12.4.4.2. Canned fiSh product 
Comments of medical doctors on the developed canned fish product are shown in Table 12. 1 3  and 
Appendix 12.4. About 70% of surveyed doctors said that the canned fish with fish oil disguised 
in it was a good idea to increase fish oil consumption for Indonesians. Appendix 12.4 indicates 
that paediatricians and medical doctors having 11-15 years experience did not give strong support 
to the product as only 38% and 40% of them respectively agreed with the idea Approximately 
63% of total surveyed doctors thought the product would have good prospects in the market in 
terms of medical suitability, while 34% said that maybe the product had good prospects in the 
market. In terms of demographic groups, each group of respondents had a less than 13% of 
respondents saying that the product did not have good prospects in the market. 
Table 12.13.  Comments of medical doctors on the product idea and the prospect of the product in 
the market 
Respondents 
Questions 
N % 
1 .  Is the canned fish product with fish oil 
disguised in it the good way to improve fish 
oil consumption for Indonesian? 
YES 1 14 69.9 
NO 34 20.9 
MAYBE 1 5 9.2 
2. Does this product have a good prospect in the 
market according to your medical point of view? 
YES 102 62.6 
NO 5 3.1 
MAYBE 36 34.3 
3. Will you suggest your patients to consume this 
product, if the product is available on 
the market? 
YES 147 90.2 
NO 14 8.6 
MAYBE 2 1 .2 
263 
The doctors showed strong support to the introduction of the product, as approximately 90% of 
them would suggest their patients consume this product. All demographic groups of respondents 
showed the same response by promising to advise their patients to purchase this product. 
12.5. DISCUSSION 
12.5.1. Chemical and physical changes in canned fish during production trial 
In general, the changes in the product during the production trial were similar to those occurring 
in the experiment in Chapter 1 1 .  But, some new information was revealed. 
The production trial proved that pre-cooking using steam induced weight loss in the fish, of 
approximately 13%. During cooking, water and water-soluble proteinaceous materials such as 
gelatin, nitrogen-containing extractives, and other substances are leached out of the fish (Lassen, 
1965). The removal of water from the fish tissues is attained primarily by heat coagulation and 
an ensuing shrinkage of the flesh protein. This cooking is vital to remove tissue water, which, if 
not eliminated, will later appear in the can during sterilization, giving a watery, boiled-like fish into 
which the oil does not penetrate, with large drops of water in the oil phase (Cheftel, 1965). 
Proximate analysis also indicated that the considerable changes in fish composition were found 
during pre-cooking as shown in the decrease in moisture and ash contents and the increase in 
protein, fat and carbohydrate contents. Sterilization did not induce any further changes in the 
proximate composition of the fish. This revealed that the stabilization in the fish structure has been 
formed by the application of pre-cooking. Proximate composition of the product showed that the 
fish oil addition into the medium significantly increased the fat content of the canned fish product. 
This means that the aim to deliver fish oil through the canned fish product in order to increase the 
fish oil consumption of Indonesians could be achieved. 
As found in the experiment of Chapter 1 1, the peroxide value of the tomato sauce decreased after 
the sterilization process, while anisidine and TBA values measuring the secondary products of 
oxidation increased. This result indicates that sterilization caused the quick conversion of 
hydroperoxide to secondary products such as a/?-aldehyde and malonaldehyde measured as 
264 
anisidine and TBA values respectively. Hydroperoxide may also undergo a variety of scission and 
dismutation reactions to form a wide spectrum of carbonyl compounds, hydroxy compounds, short 
chain fatty acids, dimers and polymers (Dugan, 1968; Smouse, 1978). The total oxidation (totox) 
value of tomato sauce decreased due to sterilization treatment. This indicated that the destruction 
of hydroperoxide was more obvious than the conversion into the secondary products of oxidation. 
Fat content of fish was significantly lower than the fat content of whole canned fish product due 
to the fish oil in the medium. Therefore, the changes in fatty acid profiles of fish were less 
important than the changes in tomato sauce or the whole product. The results of fatty acid profile 
analysis showed that the fatty acid profiles of tomato sauce were relatively constant during 
sterilization, while the relative quantity of fatty acid profile of fish changed due to sterilization 
process by exhibiting a higher relative amount of polyunsaturated fatty acids (PUPA) in the 
sterilized oil. The same indication was noted by Hale and Brown (1983) in the canning of spanish 
sardine, thread herring and chub mackerel. However the fatty acid profile of the whole canned fish 
product was relatively constant. This showed that the fatty acid profile of tomato sauce appeared 
to determine the fmal fatty acid profile of the product and needed more attention. Tomatoes are 
an ascorbic acid source (Hall, 1984) and ascorbic acid is known as a natural oxidation inhibitor 
(Pokomy, 1991). Ascorbic acid may have given protection to fatty acids, particularly unsaturated 
fatty acid from oxidation attacks during sterilization with remaining oxygen in the can. 
Significant changes in canned fish colour occurred starting during pre-cooking, in which according 
to Hunter-L, -a and -b values the fish flesh colour shifted to white, grey and yellow direction 
respectively. Red colour, due to the presence of carotenoid, became discoloured as a result of the 
Hunter-a value sifting to grey. Sterilization induced the changes in Hunter-L, -a and -b values to 
the direction of black, red and yellow respectively. The detail discussion about the colour changes 
in the fish flesh has been given in Chapter 1 1 .  Sterilization caused the changes in Hunter-L, -a and 
-b values of the tomato sauce medium by exhibiting the reduction of white, yellow and red 
intensity respectively. The red colour in tomato sauce is formed by carotenoids, particularly 
lycopene (Belitz and Grosch, 1987). Two types of isomerisation can occur during heat treatment 
on carotenoids, namely cis-trans isomerisation and epoxide isomerisation, which can give rise to 
lightening of the colour (Hall and Pitcher, 1991). 
265 
12.5.2. Product safety and shelf life 
Sterility and incubation testing indicated that the canned fish for consumer testing in Indonesia was 
safe. Therefore, the possibility of spoilage due to the microorganisms surviving during the 
sterilization process was not considered, and the shelf life of the product depended on post?
processing contamination. 
Contamination of can contents after processing can occur through leaks in the can. Such leaks are 
often the result of faulty seaming and excessive corrosion (Van den Broek, 1965). The measured 
overlap of the double seam of the can in the production trial was 62.6% which was higher than the 
minimum standard of 45% (Warne, 1988). This overlap level was considered safe. Non-microbial 
spoilages which can occur and affect the shelf life of canned fish products are hydrogen swells, 
corrosion of cans, carbon dioxide swells, discoloration, filling problems and tainting as reviewed 
by Murrell (1978). 
Shelf life prediction for a canned product is extremely difficult to estimate. Shelf life is determined 
by package composition, product and container compability and temperature of storage. For metal 
containers, internal corrosion and storage temperature are the most important considerations (Rees, 
1991). According to the experiment conducted by Emshanova et al (1983) on the storage trial of 
canned mackerel, hake, turbot and cavalla in tomato sauce, the products were recommended for 
storage of up to two years. Bayley (1991) stated that canned fish which passes the incubation test 
may have a shelf life up to 20 months. Based on this information, the shelf life of canned fish 
-with fish oil disguised in it is probably up to 20-24 months. 
12.5.3. Prospect of developed canned fish in Indonesian market 
The idea of incorporating fish oil into canned fish was approved by most surveyed medical doctors 
as a suitable way to deliver fish oil to the consumer. The consumer testing of the product showed 
that only a small percentage of consumers did not like the product. Most consumers commented 
"neither like nor dislike", "like slightly" and "like very much", thus indicating that the market of 
this product was promising. Crosstab analysis revealed that most consumers were willing to buy 
the product, especially the consumer who commented "neither like nor dislike",  "like slightly" and 
"like very much". The projected buying trend of the product in terms of the product acceptability 
can be described using a triangular form as follows: 
Disl ike  very mu ch 
Dislike s l ightly 
Neither l ike 
nor disli ke 
Like  s l ightly 
L ike very much 
0% 
-- - - -- 1  
Potentia l 
Consumer 
266 
Not buy 
Buying 
Trend 
Buy 
Potential consumers are expected from below the "dislike slightly" area The percentage of 
consumers who are willing to buy the product increased from the top to the bottom of the triangle. 
This study showed that the buying trend of the prospective consumer commenting "dislike very 
much" and "like very much" to the product was 0% and 100% respectively. 
Consumer experience in eating a similar product was important in influencing buying intention. 
The consumers, who have consumed similar products, gave more promising response to the product 
than the consumers who have not consumed similar products. The main reason for the consumer 
buying the product is convenience, as the product was easy to store and serve. Convenience 
products are found in more variety in large cities such as Jakarta and Semarang, providing more 
choice and product competitiveness. This may have affected the buying trend of the consumer to 
the developed product. The prospective consumers were found more in small cities (Tangerang, 
Sragen and Lumajang) than in large cities (Jakarta and Semarang). All information on the present 
label has to be retained. Since the majority of the Indonesian population is Moslem, it suggest that 
the words " 100% guaranteed halal" be added, indication that no materials or ingredients used to 
produce the product were classified "not-halal". Some respondents commented that they did not 
intend to buy the product because canned fish was normally expensive. They preferred to consume 
fresh fish which is relatively cheap and easy to purchase. Actually, these respondent liked the 
developed product. 
267 
The use of sensory evaluation during the development process of this product in the laboratory 
seemed very useful. This is reflected in the results of consumer testing. The developed canned 
fish product processed using condition and formula selected through the sensory .evaluation is 
accepted by most consumers and this revealed the usefulness of sensory evaluation to bridge 
laboratory trials and consumer desirability . .  
A majority of the surveyed gave strong support for the production of this product, as approximately 
90% of the medical doctors were willing to advise their patients to consume the product for health 
reasons. The superiority of the product, according to the medical point-of-view, over the existing 
products in the market, was significant. The role of the medical doctor to introduce this product 
to prospective consumers (patient), is obvious and strengthens the claim that the product has health 
benefits. 
12.6. CONCLUSIONS 
Consumer product testing and the medical doctor survey show that the canned fish enriched with 
disguised fish oil has good prospects in the Indonesian market. Most of the consumers 
participating in the testing liked the product and most of them intend to buy. The role of the 
medical doctor is important in the introduction of health benefits of this product. Most prospective 
consumers suggested retention of the superiority claimed by this product, as stated on the label. 
The result of consumer testing gave information about the advantages of laboratory sensory 
evaluation to develop an acceptable product to the consumer. 
268 
Chapter 13 
GENERAL DISCUSSION AND CONCLUSION 
13.1. INTRODUCTION 
Fish oil is a by-product of fish meal and canned fish production and has not received the proper 
attention during production and utilization by most fudonesian producers. At present, fish oil is 
mainly for non-food uses such as animal feed and leather tanning. No attempts have been made 
to use fish oil for human consumption. In terms of quality, most of the fish oil did not meet the 
requirements for human consumption, especially the undesirable odour which restricts use. 
Quality improvement was necessary, particularly for reduction of the undesirable odour. Refming, 
which has been widely employed for vegetable oils, could be used to remove undesirable 
compounds in fish oil. This generated the idea for this study. The resin refming method developed 
by Fernandez (1986) was used in this study, and has been proved to give a high retention of 
polyunsaturated fatty acid, especially ro-3 fatty acid, since no heat is involved during the refining 
process. 
However as also shown in the refming process of other oils using other refining methods, not only 
.undesirable compounds were reduced, but also natural antioxidant. This would affect the stability 
of the fish oil during processing and storage. 
13.2. FISH OIL REFINING 
Only a minority of fudonesian fish oil producers provided the facility of fish oil refmement. All 
used the alkali refming method involving heat processing, which could affect the stability of the 
oil. This fact indicates that the prospect of the introduction of the resin refining process in 
Indonesia was promising. Most of the fish oil producers indicated their intention to adopt the 
technology. 
Some aspects of the resin refming method have to be considered before its adoption. Quality of 
269 
the fish oil should be known before refining, especially in terms of odour. This study suggests that 
fish oil having a very strong undesirable odour must be passed through the refming process more 
than once, to reduce odour strength. The multiple refining did not oniy affect the _odour of fish 
oil, but also the free fatty acid, colour and refractive index values. Multiple refming did not 
significantly affect fatty acid profiles. This indicated the superiority of the method. 
In order to obtain the refmed oil with optimum quality, the ratio of fish oil to resin volume had to 
be considered. The fish oil-resin volume ratio of 1 : 1  was recommended. A higher fish oil ratio 
will result in a lower refined fish oil quality. This showed that there is a limited capacity for the 
resin to retain the undesirable compounds of fish oil. When the resin has achieved its maximum 
capacity to retain undesirable compounds, the remaining compounds may pass through the column 
together with refined oil resulting in a lower oil quality. High column size tended to produce better 
oil quality, allowing more opportunity for the oil to contact with the resin. However the contact 
time did not affect the refined fish oil quality. This is shown in the experiment using vacuum 
pressure to accelerate the refining time, where the vacuum pressure did not influence the quality 
obtained. 
Most of the Indonesian fish oil producers are located in an area near a beach. In constructing the 
refining unit, this has to be taken in account in the selection of materials used for the column. The 
construction materials must be anticorrosive and available locally, for easy purchase and 
maintenance. Automation would make the method more efficient. All of these requirements have 
to be considered to guarantee the success of the introduction of the resin refming method to 
.Indonesian fish oil producers. 
The resin refining process promoted toluene as the main flavour compound in the refmed fish oil, 
making the odour and taste of the oil more acceptable. The odour was the most important factor 
in fish oil consumption, as shown in the survey conducted by Fernandez (1986). The Indonesian 
panellists, in this study, commented that the odour of the fish oil was easier to evaluate than the 
taste, but the panel ratings showed that the odour score was usually similar to the taste score. 
13.3. FISH OIL STABILITY 
Study in fish oil stability is very complex, since stability is affected by many factors, as shown in 
this study. Factors which may affect fish oil stability are fish oil type, refming treatment, storage 
270 
temperature, antioxidant addition, packaging type, and processing treatment 
Different fish oils had a different stability, as shown in Chapter 8. The level of natural antioxidant 
present in the oil may affect stability. The oil obtained from the fish meal processing was more 
stable than the oil collected from the fish canning operation, and the level of natural antioxidant, 
especially tocopherols in fish meal oil; was higher than in canning waste oil. Refined fish oil also 
has less stability when compared to the unrefined oil. The tocopherols content in refined oil was 
significantly lower than in the unrefined oil. 
As storage conditions affected fish oil stability, storage condition manipulation seems to be a very 
effective way of improving stability. This study shows that the fish oil deterioration rate was the 
function of storage temperature. Chemical, physical and organoleptic deterioration could be 
significantly inhibited by storing the oil at low temperature. Oxidation is suspected as the main 
cause in fish oil deterioration, since the fish -oil is rich with unsaturated fatty acid, especially 
polyunsaturated fatty acids. Oxidation occurring on the unsaturated fatty acid is shown in Figure 
13.1 .  Oxidation could be inhibited by limiting oxygen levels in the container. In this study, the 
oxygen present in the package used to store fish oil was limited by using vacuum package. This 
method showed a significant effect in reducing oxidation rate. 
H 
? IH1 Hi-G-c(CHJ, ? '-(;? CH,),CH, 
H-c-o_g-R 
I 
Fatty Free 
Radical l+ Oxygen 
Peroxide Free 
Radicals and 
Hydroperoxides 
,J, 
Aldehydes 
Ketones 
Alcohols 
Acids 
{Double bond and a-carbon, the 
oxidation site in triglyceride 
molecules 
Initiating step where hydrogen 
is lost from the a-carbon in 
the fatty acid group. This is 
catalyzed by heat, light and 
trace metals 
Unstable form of the glyceride 
and very reactive with oxygen 
Very unstable and decompose 
ready to form short-chain 
organic compounds 
End products of glyceride 
oxidation responsible for 
RANCIDITY in oils 
Figure 13 .1 .  Oxidation of glyceride leading to rancidity of oil (Sherwin, 1990) 
271 
The sterilization process had an unusual effect on the fish oil and the changes occurring looked 
very complicated. High temperature may have accelerated all oxidation processes in the oil. 
Hydroperoxides might undergo a destruction process through scission and dismutation reaction 
resulting in carbonyl compounds, hydroxy compounds, short chain fatty acid, dimers and polymers 
including compounds known as secondary products of oxidation. The conversion of 
hydroperoxides into the secondary products of oxidation was shown by the increase of anisidine 
value measuring a/?-aldehyde formed during sterilization process. TBA value measuring 
malonaldehyde formation, which is also a secondary product of oxidation, decreased due to 
sterilization. This indicated that malonaldehyde was unstable when heated. The malonaldehyde 
formation mechanism is shown in Figure 13.2. However the sterilization process resulted in the 
increase in TBA value of the oil extracted from the tomato sauce medium. Tomato sauce may 
272 
have inhibited decomposition process of malonaldehyde. 
Figure 13 .2. Mechanism of malonaldehyde formation (Erickson and Bowers, 1976) 
decolouratiorrand""thi.s ? s 1JOSsibly one actor invo1vingin determining"the colotll' of tSh eH. 
A darkening process may also occur in fish oil. This may be due to the reaction between protein 
with hydroperoxides and their degradation products producing browning process (Belitz and 
?Grosch, 1987). The rate of both processes influences the colour of fish oil. If the former process 
is faster, the fish oil colour will be lighter. However if the later process is faster, the fish oil 
colour will be darker. Most of the oil studied for storage stability showed a lighter colour at the 
end of observation. In the study using canning waste oil, the oil colour showed a different pattern 
of colour changes, where the colour became darker at first, and then lighter during further storage. 
Sterilization resulted in a significantly different direction of colour change. The sterilized oil 
exhibited a darker colour than the unsterilized oil. This indicated that the browning reaction was 
more important than decomposition of carotenoids. 
273 
13.4. DEVELOPl\1ENT OF CANNED FISH ENRICHED WITH FISH OIL 
Canned fish was selected as a mean of delivering fish oil to the consumer, since the product 
showed the possibility of incorporating a high amount of fish oil without significantly affecting 
product acceptability. According to the survey results, this idea was well accepted by both 
potential consumers and medical doctors. Since the proposed product was market oriented, the 
medium used to disguise the fish oil had to be tested. A consumer survey endorsed the use of 
tomato sauce. 
The laboratory sequences during development of this product can be seen in Figure 13 .3, while the 
experimental designs used for each step are shown in table 1 3 .1 .  
Tomato Sauce Selection of 
Formulation ' / Sterilization , ' 
Condition 
? 
Determination 
of Important 
Processing 
Factors 
Process 
Optimization 
l 
Production Trial 
at Pilot Plant 
Scale 
Figure 13.3. Experimental stages used to develop the canned fish with fish oil disguised in it 
274 
Table 13 .1 .  Experimental design used for each experimental stage 
Experimental Stage Experimental Design 
Tomato sauce formulation Mixture design 
Selection of sterilization Factorial design 
condition 
Determination of important Plackett and Burman design 
factors in canning process 
Process optimization Factorial design 
As the fish oil was to be incorporated into canned fish through a medium, the medium received 
the first attention with experiments on a tomato sauce formulation. A mixture-design was used to 
develop the formula. This design was very helpful giving directions of change in the levels of 
ingredients in the medium, and made the formulation experiment efficient The panellists had no 
negative comments about the fish oil addition to the sauce during formulation sensory testing. 
Since the fish oil addition was to improve nutritional value, its stability during sterilization, which 
employs a high temperature, had to be investigated to determine the sterilization condition which 
ytould give optimum prevention from deterioration. This study indicated that to reduce the risk of 
quality deterioration, the fish oil had to be sterilized at high temperatures for a short period of time 
with vacuum heat space in the can. Factorial experiment design was used in this study. 
The next development step was the determination of important factors affecting product 
acceptability. The Plackett and Burman experimental design was used to select the important 
factors. This indicated that the Plackett and Burman design was a very effective tool for selecting 
the important factors over the factorial experimental design, since working with many factors, such 
as eight variables, was complex. According to the results, the variables considered important are 
pre-cooking, vacuum head space of can, garlic, shallot, and vinegar additions. These treatments 
must be retained in the processing. Sterilization time was optimized to obtain optimum 
acceptability of the product. 
The selected variables obtained from the Plackett and Burman experiment underwent optimization 
275 
using the factorial experimental design. This design was used to select the optimum sterilization 
time and salt level, because only two variables were involved. T-test was used to reveal the 
significant difference among tested levels of each variable. The optimum levels were easily 
decided: optimum sterilization time and salt level in tomato sauce medium were 50 minutes and 
1 .5% respectively. 
The final step in the development of the proposed canned fish was a production trial using a pilot 
plant scale at full capacity. The product was processed following the process flow as shown in 
Figure 13.4. The results showed that the changes in the product during processing were similar 
to those occurring in the optimization experiment. Since the product obtained from the production 
trial was to be distributed for Indonesian consumer testing, a safety assessment was undertaken. 
The results indicated that microbiologically the product was safe for consumption, as no 
microorganisms grew on nutrient agar plates during sterility testing. The incubation test revealed 
that no changes were found in the appearance of the can. 
The above results show that the process flow developed from this study could be recommended for 
processing canned sardine enriched with fish oil disguised in tomato sauce. However if the technology 
is going to be applied to Indonesian sardine (Sardinella longiceps), which is commonly processed into 
canned fish in Indonesia, some changes in the process (e.g. sterilization time) will be required to obtain 
an acceptable product. 
Can 
Washing 
Draning 
CAN 
Figure 13.4. Process flow of canned fish with fish oil disguised in it 
Fish 
Heading, gutting 
and cutting 
Washing 
Draining 
Filling 
Pre-cooking 
at 98?C, 20 min . .  
Draining 
Sauce filling 
Steam 
Sterilization 
Boil 
Stirring until 
dissolve 
Stirring properly 
(? 80?C) 
Stirring properly 
(? 80?C) 
TOMATO SAUCE 
121 ?C, 50 min 
? CANNED FISH PRODUCT 
? 
0\ 
277 
13.5. PROSPECT OF DEVELOPED CANNED FISH IN INDONESIAN MARKET 
Incorporating fish oil disguised in tomato sauce in canned fish was ainied at developing a product 
which was nutritionally better than existing products. The fish oil addition aims to increase the ro-3 
fatty acids level, as fish oil contains a high quantity of these acids. The results of the production 
trial showed that the fish oil enrichment increased the total fat content of canned sardine and 
consequently the ro-3 fatty acids level increased as well. 
Consumer testing of the product in Indonesia revealed that only a minority of respondents said that 
they did not like the product. The majority, who can be classified as potential consumers, 
responded to the product with "neither like nor dislike", "like slightly" and "like very much". 
Consumer testing indicated that the majority were willing to buy the product. Product acceptability, 
by a consumer, significantly influences buying intention, where the consumer, from the group 
showing a higher acceptable level to the product, indicated a higher percentage of consumer 
intention to buy the product. If the product was released in the market, more effort is then required 
to convert consumers commenting "neither like nor dislike" and "like slightly" to "like very much", 
by influencing them using advertisements to promote the nutritional advantages of the product. 
Sloan (1987) has warned that food companies cannot afford to assume that consumers know the 
nutritional advantages of their product. Manufactures must clearly describe these nutritional values 
to consumers. Advertisements should convey specific nutritional information, which goes beyond 
the basics, to educate consumers. By giving consumers specific additional information, 
advertisements will engage consumer interest and make the product memorable. Thus, the health 
_benefits of fish oil, particularly ro-3 fatty acids have to be mentioned clearly in promotional 
material, as well as on the label. 
Another way to convey the health benefits of the product is through the help of medical doctors. 
The prospects for this approach are shown in the medical doctor survey. More than 90% of 
medical doctors surveyed were ready to advise their patients to consume the product for health 
reasons. This means that medical doctors provide another outlet access for the product. Most of 
the doctors thought that this product had good prospects in the Indonesian market in terms of 
medicinal value. 
The consumer testing also shows that the experience of the consumer affects the buying trend. 
Before releasing the product to the market, information about the locations where high consumption 
of canned fish, especially canned sardine with tomato sauce medium occurs, has to be obtained. 
To assure success in marketing the product, the selling can be centralized to those locations first, 
278 
before other markets are developed. Most of the consumers said that the reason for buying the 
product was convenience. According to Kinsey (1992) rising incomes and the high value placed 
on time have driven consumers to buy convenience foods. This convenience reason, and the 
improving economic situation of Indonesians, would provide better opportunity for the developed 
product to be commercially successful. 
13.6. ROLE OF SENSORY EVALUATION IN PROCESS AND PRODUCT 
DEVELOPMENT 
Intensive sensory evaluation was used, as sensory characteristics can be the most critical elements 
in product success. If it does not taste "good" to consumers, the product probably will not be 
successful in the market place (Mook, 1984). -The basic concern in all sensory testing is product 
acceptance (Ellis, 1970). This indicates the importance of sensory evaluation during product 
development 
Femandez (1986) demonstrated the usefulness of sensory evaluation to select resin refining as the 
most appropriate method for refming fish oil. In this present study, sensory evaluation was used 
in optimization experiments for the refining process, storage, and sterilization. Descriptive analysis 
was undertaken using Indonesian trained panellists. The effect of the resin refining treatment can 
be shown quickly and clearly. Panellists showed that odour difference in fish oil was easier to 
.detect than taste. The sensory evaluation for fish oil recommended only odour. This indicates that 
if the odour of fish oil is unacceptable, the panellists would hesitate and even rejected taste 
evaluation. Sensory evaluation can also be used to detect the changes in fish oil during storage and 
sterilization. In the storage study, the sensory evaluation results can be used to predict shelf life. 
A method of utilizing sensory evaluation to determine product shelf life has been reviewed by 
Dethmers (1979). 
Sensory evaluation was used in all aspects of this study, as product development is a time?
consuming, costly, and risky endeavour. It has been shown how laboratory sensory testing can 
reduce the lead time for introduction of the product to the market (Moskowitz and Rabino, 1983). 
Dethmers et al (1981) stressed that product developers need information on the sensory quality and 
relative acceptability of experimental prototype samples, as an input for marketability. 
Various sensory testing methods were used during sensory evaluation. A nine-scale hedonic test 
279 
was used in the tomato sauce formulation. This test was very helpful in showing the acceptable 
level of ingredients. Descriptor-scales of descriptive testing were used in the experiment on the 
determination of the sterilization condition. The ideal ratio scoring method, which is basically a 
line-scale of descriptive testing, was used in the experiment on the determination of important 
factors of the fish canning process. However this method was not suitable for Indonesian panellists 
and they commented that the method was confusing and also required intensive supervision. 
Actually, the method has been used successfully in the Nham development for Thai market by 
Wiriyacharee (1990). In the canning process optimization experiment. two methods were used 
together: descriptor-scale of descriptive testing to evaluate product characteristics, and a nine-scale 
hedonic test to evaluate the acceptability of attributes and product. The methods seemed to be very 
convenience for the Indonesian panellists. 
Trained sensory panellists, as employed in this study, can be useful during development of the 
consumer product testing questionnaire. They can identify and describe sensory characteristics 
applicable to the test product and generate meaningful terminology which can be utilized in the test 
questionnaire (Erhardt. 1978). Furthermore, the sensory method applied in the canning process 
optimization experiment considered suitable for Indonesians was chosen to generate a questionnaire 
for consumer testing of the product. 
The "Just right" method, which is basically a descriptor-scale of descriptive testing was used in 
the consumer test. This method seemed very acceptable to the respondent. The home use testing 
method was used for this evaluation. Home use testing is the most natural testing situation, as the 
information obtained by home use testing reflects: the products assessment during an intensive 
situation, rather than a short taste-test situation; the product tasted without the standard controls; 
other influences, such as temperature and additional foods can be varied according to taste 
(Moskowitz and Chandler, 1979). 
Sensory tests can provide data to confnm that changes in the product have been made in the 
direction indicated by consumer testing (Erhardt, 1978). This study revealed that the product most 
acceptable during laboratory sensory panel testing was also accepted by consumers participating 
in product testing. 
280 
13.7 THE ROLE OF THE CONSUMER IN PRODUCT DEVELOPMENT 
In this study, the consumer was involved in two parts of the product development sequence: during 
the generation of the product type and product testing. Consumer participation is very important 
for market orientated product development. 
To reveal whether the product idea generated by the product developer was interesting or not to 
the consumer, the consumer was asked for comments. Most of the prospective consumers were 
interested in the product idea. Furthermore, the involvement of the consumer during product design 
was aimed at obtaining the right product type to meet consumer wants. In terms of product type, 
the consumer was asked to contribute ideas about fish species to be canned, medium type, and can 
size. By using this information, the product development process could be started. During 
laboratory activities, the role of the consumer was replaced by a group of panellists who 
participated in the sensory evaluation of the product. 
Further consumer participation was in the product testing involving more people than laboratory 
sensory evaluation and consumer survey. In this study, consumers participating in a consumer 
survey, laboratory sensory evaluation and consumer product testing was 130, 10 and 432 people 
respectively. According to Anderson (1981), consumer panels are an extremely important part of 
product development and marketing. They can provide invaluable information about consumer 
attitudes, which are a responses to all aspects of the marketing mix, and not simply to product 
formulation. In designing new products, consumer attitudes towards the total product image, 
- created by the physical product, packaging, promotion, price and distribution, must be taken in 
account. 
The survey on product marketability also involved the prospective consumer. Here, the consumer 
indicates information about buying trend, product price, buying motivation and market place. 
Using this information the product developer can arrange a marketing strategy. 
281 
13.8. RECOMMENDED FUTURE STUDIES 
More studies relating the importance of fish oil to consumer health and well being, are still 
required. These includes: 
* Investigation of the basic chemistry and physical process of resin action in the 
improvement of fish oil quality. So far, this information is unavailable. 
* Construction of the resin refining unit, especially towards automation. 
* The use of Indonesian sardine (Sardinella Iongiceps) for production of canned fish 
enriched with fish oil. Indonesian sardine contains up to 25% of fat which is 
significantly higher than New Zealand sardine. 
* The use of fish oil in other Indonesian food types, such as fish sauce, sausage, 
mayonnaise and salad oil. 
13.8. GENERAL CONCLUSION 
Most Indonesian fish oils, especially those obtained from fish meal processing, defmitely require 
the refining process to improve quality and acceptability. This study, using both Indonesian and 
New Zealand fish oils, proved that macroporous strong acid cation resin refmement could 
significantly improve fish oil quality and acceptability. 
The refming process affected the fish oil stability due to the loss of quantities of natural 
antioxidant, particularly tocopherols. Fish oil stability could be improved by using low storage 
temperature, antioxidant addition and vacuum package. High temperature and short sterilization 
time could reduce the deterioration of fish oil during the process and this should be applied in the 
processing of canned fish and other products containing high amounts of fish oil. 
The incorporation of fish oil into canned fish was organoleptically accepted by panellists during 
laboratory sensory evaluation and by the consumer during product testing. The prospect of the 
282 
product in the market is promising, as the majority of surveyed consumers were willing to buy the 
product. Medical doctors showed a positive response to the product, indicating their willingness 
to advise their patients to consume the canned fish for health reasons? 
283 
REFERENCES 
Ackman, R.G., 1980. Fish Lipids Part I, In Advance in fish science and technology. edited by 
Connell,JJ., p.86-103, Fishing news books Ltd., Surrey. 
Ackman, R.G., 1982. Fatty acid composition of fish oil, In Nutritional evaluation of long-chain 
fatty acids in fish oil, edited by Barlow,S.M. and Stansby,M.E., p.25-88, Academic press, 
London. 
Ackman, R.G., 1988a. The year of fish oil, Chemistry and industry 1988 (3): 139-145. 
Ackman, R.G., 1988. Concerns for utilization of marine lipids and oils, Food Tech. 42 (5): 151-
155.  
Ackman, R.G., Ratnayake,WM.N. and Macpherson,E.J., 1989. EPA and DHA contents of 
encapsulated fish oil products, JAOCS 66 (8): 1162-1 164. 
Adams, D.M. 1983. Biochemical mechanisms for thennal injury and thennal death in bacterial 
spores, In Heat sterilization of food. edited by Motohiro,T. and Hayakawa,K., p.26-38, 
Koseisha-Koseikaku CoLtd., Japan . 
. Aitken, A. and Connell, JJ., 1979. Fish, In Effects of heating on foodstuffs, edited by 
Priestley,R.J., p.219-254, Applied science publishers Ltd., London. 
Albaum, G., Strandskov, J., Duerr, E. and Dowd, L. 1989. International marketing and export 
management, Addison-Weskey publishing co., Wokingham. 
American Oil Chemists' Society, 1973. Official and tentative methods of the American oil 
chemists' society Vol.l, AOCS, Illinois. 
Association of Official Analytical Chemists, 1984. Official methods of the association of official 
analytical chemists 141b edition, AOAC, Virginia. 
284 
Anderson, AM., 198la. Experimentation, In Process improvement for small food companies in 
developing countries: a workshop manual, Compiled by Anderson, A.M., p.1 32-143, 
Massey University, Palmerston North. 
Anderson, A.M., 1981. Attitude measurement using consumer panels, Fd.Tech. in New Zealand 
16 (5): 7-16. 
Anderson, A.M., 1985. Attitude measurement and product design, In Product and process 
development in the food industry, edited by Earle,M.D. and Anderson,AM., p.99-1 12, 
Hardwood academic publishers, Chur, Switzerland. 
Anderson, A.M. and Earle, M.D., 1985. Systematic product design, In Product and process 
development in the food industry, edited by Earle, M.D. and Anderson, A.M., p.127-142, 
Hardwood academic publisher, Chur,- Switzerland. 
Angelini, P. and Merritt, C., 1975. Effect of irradiation on volatile constituents of stored haddock 
flesh, J.Fd.Scie 40: 197-199.  
Anonymous, 1987. Fish oil in rheumatoid arthritis, The Lancet 1987ii: 720-721. 
Anonymous, 1988. Fish oil, The Lancet, 1988i: 1081-1083 . 
. Anonymous, 1989. Fish oil reduce postprandial lipemia, Nutrition Review 47 (7): 21 1-214. 
Anonymous, 1991 .  Data box: scie-tech, Look Japan 1991 (2): 35. 
Artman, N.R., 1969. The chemical and biological properties of heated and oxidized fats, 
Adv.Lip.Res. 7: 245-330. 
Arya, S .S., Ramanujam, S. and Vijayaraghavan, 1969. Refractive index as an objective method for 
evaluation of rancidity in edible oils and fats, JAOCS 46: 28-30. 
Balicer, Z., Leibovitz, Z., and Ruckenstein, C., 1983. Physical (steam) refining, advantages and 
limits for soft oils, In Fat Science, p.393-403, proc. 16th ISF congress, Budapest. 
285 
Banks, A., 1967. Deteriorative changes in fish oils, In Fish oils: their chemistry, technology, 
stability, nutritional properties, and uses, edited by Stansby, M.E., p.148-163, The AVI 
publishing co.lnc., Westport, Connecticut. 
Barlow, S .M., 1977. Fish meal manufactures in the tropics, In Proceedings of the conference on 
the handling. processing and marketing of tropical fish, by Tropical Product Institute, 
p.223-229, Ministry of Overseas Development, London. 
Barlow, S .M., 1986. Fish oils in new markets, new uses, new geographical areas, Symposium 
proceedings no.:33: Fish oils and animal fats, Leatherhead Food RA, June 1986. 
Barlow, S .M., Young, F.K.V. and Dutbie, I.F., 1990. Nutritional recommendations for n-3 
polyunsaturated fatty acids and the challenge to the food industry, Proceeding of the 
nutrition society 49 (2): 13-21 .  
Bayley, C., 1991 .  Personal communication, Sealord Products Ltd., Nelson. 
Belitz, H.D. and Grosch, W., 1987. Food Chemistry, Springer-verlag, Berlin. 
Beraquet, NJ., Mann, J., and Aitken, A., 1984. Heat processing of herring: !.Release of water and 
oil, J.Food Tech. 19: 437-446. 
- Berger, K.G., 1989. Practical measures to minimize rancidity in processing and storage, In 
Rancidity in food, edited by Alien, J.C. and Hamilton, RJ., p.67-82, Elsevier applied 
science, London. 
Bidenko, M., Shendeyuk, V. and Agzhitova, L., 1974. Technology of processing of atlantic sardine, 
In Fishery products, edited by Kreuzer, R., p.208-212, Fishing News, London. 
Bimbo, A.P., 1987. The emerging marine oil industry, JAOCS 64 (5): 706-715 
Bimbo, A.P., 1989a. Food oils: past and present food uses, JAOCS 66 (12): 1717-1726. 
Bimbo, A.P., 1989. Technology of production and industrial utilization of marine oils, In Marine 
biogenic lipids, fats and oils, edited by Ackman,R.G., p.401-431 ,  CRC Press, Boca Raton, 
Florida. 
286 
Bimbo, A.P., 1990. Production of fish oil, In Fish oils in Nutrition, edited by Stansby, M.E., p.141-
180, Van Nostrand Reinhold, New York. 
Bimbo, A.P. and Crowther, 1990. The industrial uses of marine oils, In 1990 annual meeting of 
abstracts of AOCS, Inform 1 (4). 
Bio Rad Laboratories Pty Ltd., 1990. Chromatography electrophoresis immunochemistry molecular 
biology HPLC-Bio rad catalogue P, Bio Rad Laboratories Pty Ltd., Auckland. 
Boag, I.F., 1988. Mutab/PC-version 3.01, Massey University, Palmerston North. 
Board, P.W., 1981. What are canned foods?, CSIRO Fd.Res.Q. 41 : 29-3 1 .  
Body, D.R., 1991. Personal communication, Research Scientist in DSIR, Palmerston North. 
Bold, K., Wada, T. and Ohno, S., 199 1 .  Effects of flltration through activated carbons on peroxide, 
thiobarbituric acid and carbonyl values of autoxidized soybean oil, JAOCS 68 (8):561-565. 
Booz, Alien and Hamilton Inc., 198 1 .  A program for new product evolution, In Corporate strategy 
and product innovation 2nd edition, edited by Rothberg, R.R., p.177-1 84, The free press, 
New York. 
. Brody, A.L., 1971. Food canning in rigid and flexible packages, CRC Critical reviews in food tech. 
2 (7): 187-244. 
Brody, J., 1965. Fishery by-products technology, The AVI publishing co. Ltd., Westport, 
Connecticut. 
Brookman, P., 199 1 .  Antioxidants and consumer acceptance, New Zealand Fd.Tech. 26 (10): 24-28. 
Brzeska, Z. and Salmonowicz, J., 1973. The influence of tocopherol concentration in some fish oils 
upon their oxidation, Zeszyty problemowe postepow nauk rolniczych 136: 77-80, FSTA 
8): 8n473 .  
Buck, D.F., 199 1 .  Antioxidants, In Food additives user's handbook, edited by Smith,J., p.147, 
Blackie and son Ltd., Glasgow. 
287 
Buisson, D.H., Body, D.R., Dougherty, G.J., Eyres, L. and Vlieg, P., 1982. Oil from deep water 
fish species as a substitute for sperm whale and jojoba oils, JAOCS 59 (9): 390-395. 
Burger, I.H. and Waiters, C.L., 1973. The effect of processing cm the nutritive value of flesh foods, 
Proc.Nutr.Soc. 32: 1-8. 
Burgess, G.H.O., Cutting, C.L., Lovem, J.A. and Waterman, JJ., 1967. Fish handling and 
processing, Chemical publishing co Inc., New York. 
Burr, M.L., 1991 .  Is oily fish good for heart?, Trend in Fd.Scie.Tech. 1 (1): 17-20. 
Carron, K.K., 1986. Biological effects of fish oils in relation to chronic disease, Lipids 21 (12): 
731-732. 
Central Bureau of Statistics, 1991 .  Statistik Indonesia (Statistics year book of Indonesia 1990, 
Central Bureau of Statistic, Jakarta. 
Chaftel, H. 1965. The canning of the sardine, Clupea pilchardus Walbaum, In Fish as food Vol IV 
part-2, edited by Borgstrom, G., p.247-263, Academic Press, London. 
Chan, H.W.S., 1987. The mechanism of autoxidation, In Autoxidation of unsaturated lipids, edited 
by Chan, H.W.S., p.1-16, Academic press Inc., London. 
Chang, S .S., 1967. Processing of fish oils, In Fish oils, edited by Stansby,M.E., p.206-221,  The 
AVI publishing co. Ltd., Westport, Connecticut. 
CIG Gases, 1989. The gas packaging of nut meats, The commonwealth Industrial gas Ltd., New 
South Wales. 
Clark, J.P., Hunsicker, J.C, and Megreenis, CJ., 1990. Tocopherols: Nature's antioxidant, 
Fd.Australia 42 (5): 262-263. 
Clucas, IJ., 1982. Fish handling, preservation and processing in the tropics: Part-2, Tropical 
Product Institute, London. 
288 
Clydesdale, F.M. and Francis, F.J., 1976. Pigments, In Food chemistry, edited by Fennema,O.R., 
p.385-426, Marcel Dekker Inc., New York and Basel. 
Codex Alimentarius Commission, 1976. Recommended international code of practice for canned 
fish, FAO of United Nation-WHO, Rome. 
Connell, J.J., 1990. Control of fish quality, Fishery News Books, London. 
Conning, D., 1990. Towards 2000: nutrition and consumer expectation, Fd.Sci.Tech.Today 4 (1): 
26-28. 
Considine, D.M., 1974. Chemical and process technology encyclopedia, Mc.Graw Hill Book Co., 
New York. 
Considine, D.M. and Considine, G.D., 1982. Foods and food production encyclopedia, Van 
Nostrand Reinhold Publishing Co., New York. 
Coppen, P.P., 1989. The use of antioxidants, In Rancidity in foods, edited by Alien, J.C. and 
Hamilton, R.J., p.83-104, Elsevier Applied Science, London. 
Cornell, J.A., 1979. Experiments with mixtures: an update and bibliography, Technometrics 21  (1): 
95-106. 
Cort, W.M., 1974. Antioxidant activity of tocopherols, ascorbyl palmitate, and ascorbic acid and 
their mode of action, JAOCS 51  (7): 321-325. 
Cowan, J.C., Moser, H., List, G.R. and Evans, C.D., 1971 .  Organoleptic and oxidative stability of 
blends of soybean and peanut oils, JAOCS 48: 835-839. 
Craft, J.L., 1990. Statistic and data analysis for social worker, F.E.Peacock publishers, Illinois. 
Crawford, L., Kretdh, M.J. and Guadagni, D., 1976. Identification of volatile from extracted 
commercial tuna oil with a high docosahexaenoic acid content, J.Sci.Fd.Agric. 27: 531-
535. 
289 
Cutting, C.L., 1%9. Fish processing, In Food industrial manual, edited by Woollen,A., p.213-247, 
Leonard Hill, London. 
Dennis, C., 1990. Processing to provide consistent quality for the consumer, Fd.Sci.Tech.Today 4 
(1): 28-32. 
Dethmers, A.E., 1979. Utilizing sensory evaluation to determine product shelf life, Fd.Tech. 33 (9): 
40-42. 
Dethmers, A.E. and IFT Sensory Evaluation Division Committee, 198 1 .  Sensory evaluation guide 
for testing food and beverage products, Fd.Tech. 35 (11): 50-59. 
Directorate General of Fisheries, 1984. B uku petunjuk teknis pengalengan ikan seri II. Pengalengan 
ikan sardine dan mackerel di dalam saus tomat, Directorate General of Fishery, Ministry 
of Fishery, Jakarta. 
Directorate General of Fisheries, 1991.  Fisheries statistics of Indonesia 1989, Directorate General 
of Fisheries, Ministry of Agriculture, Jakarta. 
Directorate General of Fisheries, 1991.  International trade statistics of fisheries commodities 1 990, 
Directorate General of Fisheries, Ministry of Agriculture, Jakarta. 
Dirninger, N., Schaeffer, A. and Humbert, N., 1989. The flavour components of mirabelle plums: 
changes in aroma composition during ripening, Sciences des aliments 9: 725-740. 
Dugan, L.R., 1968. Processing and other stress effects on the nutritives value of lipids, World 
Rev.Nut.Diet 9: 18 1-205. 
Dyer, J.A., 1967. General industrial and potential uses of fish oils, In Fish oils, edited by 
Stansby,M., p.270-279, The AVI publishing co.Inc., Westport, Connecticut. 
Dziezak, J.D., 1986. Preservatives: antioxidants the ultimate answer to oxidation, Fd.Tech. 40 (9): 
94-102. 
Ellis, B .H., 1970. Sensory methodology for product development, Fd.prod.dev. 1970 (8/9): 86-91 .  
290 
Emodi, A. 1978. Carotenoids, properties and applications, Fd.Tech. 32 (5): 38-48. 
Emshanova, A.V., Baranova, E.I. and Kukleva, E.A., 1983. Storage tiilles for canned products in 
bulk, Rybnve Khozyaistvo 1: 64-65, FSTA 16(2): 2R117. 
English, P.M., Gerdes, D.L., Finerty, M.W. and Grodner, R.M., 1988. Effects of tripolyphosphate 
dips on the quality of thermally processed mullet <MJ!g!! cephalus), J.Food Scie. 53  (5) 
1 319-1321.  
Erhardt, J.P., 1978. The role of the sensory analyst in product development, Fd.Tech. 32 (1 1):  57-
66. 
Erickson, D.R. and Bowers, R.H., 1976. Objective determination of fat stability in prepared foods,  
In Objective methods for food evaluation: Proceedings of a symposium, 133-144, National 
Academy of Science, Washington, DC. 
Etilvant, P.X., Guichard, E.A. and Issanchow, S .N., 1986. The flavour components of mirabelle 
plums, Sciences des aliments 6: 417-432. 
FAO, 1975. The production of fish meal and oil, FAO Fisheries Technical paper, No. 142, FAO of 
United Nation, Rome . 
. Fernandez, C.C., 1986. Refinement of fish oil for human consumption: engineering investigation, 
PhD Thesis, University of Washington, Seattle. 
Fineberg, H. and Johanson, A.G., 1967. Industrial use of fish oils, In Fish oils, edited by 
Stansby,M., p.222-238, The AVI publishing co.Inc., Westport, Connecticut. 
Finley, J.W., 1985. Environmental effects on protein quality. In Chemical changes in food during 
processing, edited by Richardson, T. and Finley, J.W., The AVI publishing co.Inc., 
W estport, Connecticut. 
Fioriti, J.A., Kanuk, MJ. and Sins, R.J., 1974. Chemical and organoleptical properties of oxidized 
fats, JAOCS 51 :  219. 
291 
Fletcher, S.I., 1982. Problems caused by the presence of unwanted air in food product-A literature 
survey No. 159, Leatherhead Food RA, Surrey. 
Fogerty, A.C., 1989. Dietary fatty acids and blood lipids, Fd.Res.Ouart. 49 (3&4): 36-45. 
Fox, D.L., 1957. The pigments of fishes, In The physiology of fishes, edited by Brown, M.E., 
p.367-385, Academic Press Inc.Publisher, New York. 
Francis, F.J. and Clysdesdale, F.M., 1975. Food colorimetry: theory and applications, The AVI 
publishing co.Inc, Westport, Connecticut. 
Fujimoto, K. Mohri, S ., Hasegawa, K. and Endo, Y., 1990. Oxidative deterioration of fish meat, 
Fd.Rev.Int. 6 (4): 603-616. 
Gatchalian, M.M., 198 1 .  Sensory evaluation methods with statistical analysis, College of Home 
Economics, University of the Philippines, Quezon City. 
Gauglitz, E.J., Stout, V.F. and Wekell, J.C., 1974.Application of fish oils in the food industry, in 
Fishery products, edited by Kreuzer,R., p.l32-136, FAO of United Nation, Fishing News 
Books Ltd., Surrey. 
Gillat, P.N., Kochhar, S.P. and Rossell, J.B., 1988. A survey of the literature on the interaction of 
oxidised lipids with proteins. No.163, Leatherhead Food RA, Surrey. 
Gordon, M.H. and Rahman, LA., 1991 .  Effect of processing on the composition and oxidative 
stability of coconut oil, JAOCS 68 (8): 574-576. 
Gray, J.I., 1978. Measurement of lipid oxidation: a review, JAOCS 55: 539-546. 
Griggs, B.,  1986. The food factors, Viking-Penguin Books Ltd., Middlesex. 
Grosch, W ., 1987. Reactions of hydroperoxides-products of low molecular weight, In Autoxidation 
of unsaturated lipids, edited by Chan, H.W.-S., p.97-139, Academic Press, London. 
292 
Gruger, E.H., 1960. Fractionation and purification of triglycerides, fatty acids and methyl esters 
from fish oils, In Proceeding of the gulf and caribbean fishery institute, thirteenth annual 
session, Nov, 1960: 53-59. 
Gruger, E.H., 1967. Fatty acid composition, In Fish oils, edited by Stansby,M.E., The A VI 
publishing co.Inc., Connecticut. 
Gunstone, F.D. and Norris, F.A., 1983. Lipids in foods: chemistry, biochemistry and technology, 
Pergamon Press Ltd., Oxford. 
Gutfmger, T. and Letan, A., 1974. Quantitative changes in some unsaponiflable components of soya 
bean oil due to bleaching, J.Sci.Fd.Agric. 25: 1 143-1 147. 
Haagsma, N., van Gent, C.M., Luten, J.B ., de Jong, R.W. and van Doom, E., 1982. Preparation 
of an omega-3 fatty acid concentrated from cod liver oil, JAOCS 59 (3): 1 17-1 18. 
Hale, M.B. and Brown, T., 1983. Fatty acids and lipid classes of three underutilized species and 
changes due to canning, Marine Fish.Rev. 45 (4-6): 45-48. 
Hall, M., 1984. Feeding your children, Piatkus, London. 
Hall, M.N. and Pitcher, RJ., 1991 .  The effect of heat preservation on product quality, In Processing 
and packaging of heat preserved foods, edited by Rees,J.A.G. and Bettison, J., p.221-237, 
Blackie and son Ltd., Glasgow-London. 
Hamilton, R.J., 1989. The chemistry of rancidity in foods, In Rancidity in foods 2nd edition, edited 
by Alien, J.C. and Hamilton, RJ., p.l-21, Elsevier Applied Science, London. 
Han, D., Yi, O.S. and Shin, H.K., 1991 .  Solubilization of vitamin C in fish oil and synergetic effect 
with vitamin E in retarding oxidation, JAOCS 68 (10): 740-743. 
Hanson, S .W.F. and Olley, J., 1963. Application of the Bligh and Dyer method of lipid extraction 
to tissue homogenates, Biochem.J. 89: 101p-102p. 
Hardy, R., 1980. Lipids part-2, In Advance in science and technology, edited by Connell,JJ., p.1 03-
1 1 1, Fishing News Books Ltd., Surrey. 
293 
Hare, L.B ., 1974. Mixture designs applied to food formulation, Fd.Tech. 1974 (3): 50-56. 
Haumann, B .F., 1990. Antioxidants: ftrms seeking products they can label as 'natural' ,  Inform 1 
(12): 1002-1013. 
Hawrysh, Z.J., Shand, P.J., Tokarska, B. and Lin, C., 1988. Effects of tertiary butylhydroquinone 
on the stability of canola oil. I. Accelerated storage, Can.Inst.Fd.Sci.Technol.J. 21 (5): 549-
554. 
Hawrysh, Z.J., Shand, P.J., Tokarska, B. and Lin, C., 1989. Effects of tertiary butylhydroquinone 
on storage stability of canola oil. II. Practical storage, Can.Ins.Fd.Sci.Technol.J. 22: 40-
45. 
Hawrysh, Z.J., Mc.Mullen, L.M., Lin, C., Tokarska, B. and Hardin, RT., 1990. Effects of tertiary 
butylhydroquinone on canola oil thermal stability, Can.Inst.Fd.Sci.Technol.J. 23 (2/3):  94-
100. 
Heruwati, E.S., 1990. Personal communication, Senior research scientist, Research Institute for 
Fishery Technology, Jakarta. 
Hiatt, R and Irwin, K.C., 1968. Homolytic decomposition of hydroperoxide. V.Thermal 
decomposition, J.Org.Chem. 33: 1436-1441. 
Hoffmann, G., 1989. The chemistry and technology of edible oils and fats and their high fat 
products, Academic Press, London. 
Hsieh, T.C.Y., William, S .S. ,  Vejaphan, W. and Meyers, S .P., 1989. Characterization of volatile 
components of menhaden ftsh (Brevortia tyrannus) oil, JAOCS 66 (1): 1 14-1 17. 
Ilyas,S ., Saleh, M. and Irianto, H.E., 1985. Teknologi pengolahan tepung ikan, In Prosiding rapat 
teknis tepung ikan, p.109-116, Puslitbangkan, Jakarta. 
Indonesian Health Ministry, 1974. Peraturan Menteri Kesehatan RI. tentang zat antioxidant yang 
diijinkan bagi makanan dan minuman No.10178/A/SK/74, In Himpunan perundan 
undangan nasional dibidang hygiene-makanan 2ndedition, Ditjen.Farmasi, DepKes. 
RI., Jakarta. 
294 
Irianto, H.E. and Fawzya, Y.N., 1987. Pengaruh penggunaan palka berinsulasi pada perahu motor 
purse seine terhadap pemanfaatan basil tangkapan ikan di Muncar, Jawa-Timur, Media 
Teknol. Pangan 3 (1-2): 1-5. 
Ishikawa, M., Mori, S., Watanabe, H. and Sakai, Y., 1987. Softening of fish bone. !.Relation 
between softening rate and solubilization rate of organic matter from fish bone, 
J.Fd.Proc.Pres. 1 1 :  277-287. 
Ishikawa, M., Mori, S., Watanabe, H. and Sakai, Y. 1989. Softening of fish bone. IT. Effect of 
acetic acid on softening rate and solubilization rate of organic matter from fish bone, 
J .Fd.Proc.Pres. 13:  123-132. 
Jackson, H.W., 1981 .  Techniques for flavour and odour evaluation of soy oil, JAOCS 58: 227-231 .  
Janicek, G .  and  Pokorny, J., 1960. Variation of refractive index as a rapid method for measuring 
fat oxidation, Papers of the Inst. of Chem.Tech. Praque 4-1, 157-183, In JAOCS 40 (7): 
28. 
Jorgensen, E.A., 1967. Fish oil as a source of essential fatty acids, In Fish oils, edited by 
Stansby,M.E., p.300-321, The AVI publishing co.Inc, Westport, Connecticut. 
Josephson, D.B., Lindsay, R.C. and Stuiber, D.A., 1983. Identification of compounds characterizing 
the aroma of fresh whitefish (Coregonus clupeaformis), J.Agric.Fd.Chem. 31 :  326-330. 
Josephson, D.B., Lindsay, R.C. and Stuiber, D.A., 1991 .  Volatile carotenoid related oxidation 
compounds contributing to cooked salmon flavour, Lebensm.Wiss.U.Technol. 34: 424-432. 
Jurewicz, I. and Salmonowicz, J., 1973. Pro- and antioxidant effects of some amino acids upon fish 
oil, Zeszyty problemove postepow nauk rolniczych 136: 119-122, FSTA 5(8): 8N472. 
Karahadian, C. and Lindsay, R.C., 1989. Role of oxidative process in the formation and stability 
of fish flavours, In Flavour chemistry: trends and developments, edited by Teranishi, R.,  
Buttery, R.G. and Shahidi, F.,  p.60-75, American Chemical Society, Washington. 
Karahadian, C. and Lindsay, R.C., 1988. Evaluation of the mechanism of dilauryl thiodipropionate 
antioxidant activity, JAOCS 65 (7): 1159-1165. 
295 
Karel, M., 1985. Environmental effects on chemical changes in foods, In Chemical changes in food 
during processing, edited by Richardson,T. and Finley, J.W., p.483-501, The A VI 
publishing company Inc., Westport, Connecticut. 
Karl, H., 1984. Canned products from blue whiting, Arbeiten aus dem Intitute fiir Biochemie 
Technologie, No.1 ,  FSTA 19 (1987): 1 1R47. 
Karrick, N.C., 1967. Nutritional value as animal feed, In Fish oils, edited by Stansby,M., p.362-
382, The A VI publishing co Inc., W estport, Connecticut. 
Ke, P.J., Nash, D.M. and Ackman, R.G., 1977. Mackerel skin lipids as an unsaturated fat model 
system for the determination of antioxidative potency of TBHQ and other antioxidant 
compounds, JAOCS 54 (10): 417-420. 
Khayat, A. and Schwall, D., 1983. Lipid oxidation in seafood, Fd.Tech. 37 (7): 1 30-140. 
Kinderlerer, J., 1989. Fish oil, Brith.Food J. 91 (1): 32. 
Kinsella, J.E., 1986. Food components with potential therapeutic benefits: the n-3 polyunsaturated 
fatty acids of fish oils, Fd.Tech. 40 (2): 89-97. 
Kinsella, J.E., 1987. Seafoods and fish oils in human health and disease, Marcel Dekker Inc, New 
York. 
Kinsella, J.E., 1988. Fish and seafoods: nutritional implications and quality issues, Fd.Tech. 42 (5): 
146-150. 
Kinsey, J. 1992. Quest for convenience: a matter of time, Cereal Fd.World 37 (4): 305-310. 
Kitchener, J.A., 1957. Ion-exchange resins, John Wiley and Sons Inc, London. 
Klaui, H. and Bauernfeind, J.C., 1981 .  Carotenoids as food colors, In Carotenoids as colorants and 
vitamin A precursors, edited by Bauernfeind, J.C., p.47-317., Academic Press Inc., New 
York. 
296 
Kochhar, S.P., 1988. Influence of temperature on accelerated stability testing of oils and fats: a 
literature survey no.161, Leatherhead Food RA, Surrey. 
Koning, A.J.de and Milkovitch, S. 1984. Preliminary trials intended to reduce free fatty acid (FFA) 
content of an anchovy oil by reaction with glycerol, Annual report No.38, p.97-98, Fishing 
Industry Research Institute, Cape Town. 
Koning, A.J.de and Milkovitch, S., 1984. Refming of fish oil with ethanolamine, In Annual report 
No.38, p.93-96, Fishing Industry Research Institute, Cape Town. 
Kritchevsky, D. 1991.  The effect of dietary garlic on the development of cardiovascular disease, 
Trends in Fd.Scie.Tech. 1 (6): 141-143. 
Kwon, T.W., Menzel, D.B. ? and Olcott, H.S-., 1965. Reactivity of malonaldehyde with food 
constituents, J.Fd.Scie. 30: 808-813. 
Labuza, T.P., 1971. Kinetics of lipid oxidation in foods, CRC Cri.Rev.Fd.Tech. 2 (3): 355-405. 
Labuza, T.P., 1984. Application of chemical kinetics to deterioration of foods, J.Chem.Ed. 61 (4): 
348-358. 
Landgraf, R.G., 1963. Canned fishery products, In Industrial fishery technology, edited by 
Stansby,M.E. and Dassow, J.A., p.309-322, Reinhold Publishing Co., London. 
Lang, K., 1970. Influence of cooking on foodstuffs, World Rev.Nut.Diet 12: 260-317. 
Lassen, S., 1965. Tuna canning and the preservation of the raw material through brine refrigeration, 
In Fish as food Vol IV part-2, edited by Borgstrom, G., p.207-245, Academic Press, New 
York and London. 
Lee, C.F., 1963. Processing of fishmeal and oil, In Industrial fishery technology, edited by 
Stansby,M.E., Reinhold Publishing Co., New York. 
Leonard, S .J., Merson, R.L., Marsh, G.L. and Heil, J.R., 1986. Estimating thermal degradation in 
processing of foods, J.Agric.Fd.Chem. 34 (3): 392-396. 
297 
Li, Y.T. and Regenstein, 1990. Use ofmenbaden fish oil in mayonnaise, In Advances in fisheries 
technology and biotechnology for increased profitability, edited by Voigt, M.N. and 
Botta,J.R., p.l51-161, Technomic Publishing Co.Inc., Lancaster. 
Lillard, D.A., 1978. Chemical changes involved in the oxidation of lipids in foods, In Lipids as a 
source of flavour, edited by Supran, M.K., p.68-80, American Chemical Society, 
Washington, D.C. 
Lomanno, S .S. and Nawar, W.W., 1982. Effects of heating temperature and time on the volatile 
oxidative decomposition of linolenate, J.Fd.Scie. 47: 744-766+ 752. 
Lund, D.B ., 1973. Effects of heat processing, Fd.Tech. 27 (1): 16-18. 
Lund, D.B., 1975a. Effects of blanching, ?  pasteurization, and sterilization on nutrients, In 
Nutritional evaluation of food processing, edited by Harris, R.S. and Karmas, E., p.205-
240, The AVI publishing Co.Inc., Westport, Connecticut. 
Lund, D.B., 1975b. Heat processing, In Principles of food science part II: Physical principles of 
food preservation, edited by Fennema, O.R., Marcel Dekker, Inc, New York. 
Lundberg, W.O., 1967. General deterioration reactive, In Fish oils: their chemistry, technology, 
stability, nutritional properties, and uses, edited by Stansby, p. 141-147, The AVI 
publishing Co.Inc., Westport, Connecticut. 
Ludwicki, J., Tayeb, LB . and Dillon, J.C., 1986. Losses of a tocopherol and a- and ?-carotenes 
at various stages of the refining process of a mixed vegetable oil, Sciences des Aliments 
6 (2): 287-299. 
Lyon, D.H., 1991. The role of sensory analysis in product development, The Food Technologist 
21 (3): 8-13. 
Malins, D.C., 1967. The classes of lipids in fish, In Fish oils, edited by Stansby, M.E., p.31-42, 
The AVI publishing Co.Inc., Westport, Connecticut. 
298 
Moeljanto, R., 1982. Pemanfaatan lemak dalam hubungannya dengan pemanfaatan lemuru secara 
optimal, In Prosiding seminar perikanan lemuru, Banyuwangi 18-21 Januari. 1982, 
Puslitbangkan, Jakarta. 
Mook, J.H., 1984. Correlation of consumer and professional sensory description, Cereal Fd.World 
29 (7): 403-405. 
Moris, R.J. and Culkin, F., 1989. Fish, In Marine biogenic lipids, fats, and oils Vol.II, edited by 
Ackman, R.G., CRC Press Inc, Florida. 
Moskowitz, H.R. and Chandler, J., 1979. Notes on consumer oriented sensory evaluation, J.Food 
Qual. (2): 269-276. 
Moskowitz, H. and Rabino, S ., 1983. Alternative strategies for product optimization, Advance in 
strategic management 2: 99-123. 
Murrell, W.G., 1978. The spoilage of canned foods, Fd.Tech.Aust. 30 (10): 381-384. 
Nakamura, K., Iida. H. and Tokunaga, T., 1980. Separation and identification of odour in oxidized 
sardine oil, Bull.Japanese Soc.Scie.Fish. 46 (3): 355-360. 
Narasimhan, S., Raghuveer, K.G., Arumugham, C., Bhat, K.K. and Sen, D.P., 1986. Oxidative 
rancidity in groundnut oil-evaluation by sensory and chemical indices and their correlation, 
J.Fd.Sci and Tech. 23: 273-277. 
Nair, P.G.V., Antony, P .D. and Gopakumar, 1979. Oxidative rancidity in the skin and muscle lipids 
of oil sardine (Sardinella longiceps), J.Fd.Sci and Tech. 16: 151-154. 
Nawar, W.W., 1985. Thermal and radiolytic decomposition of lipids, In Chemical changes in food 
during processing. edited by Richardson, T. and Finley, J.W., p.79-105, The AVI 
publishing Co.Inc., W estport, Connecticut. 
Nettleton, J.A., 1987. Seafood and health for New Zealanders, David Bateman, Auckland. 
Niazi, S.K., 1987. The omega connection, Esquire Books Inc, Illinois. 
299 
Novikov, V.M., 1984. Handbook of fishery technology, A.A.Balkema, Rotterdam. 
O'Brien, P.J., 1987. Oxidation of lipids in biological membranes and rntracellular consequences, 
In Autoxidation of unsaturated lipids, edited by Chan,RW.S., p.233-280, Academic Press, 
London. 
Odumosu, O.T., Sinha, J. and Hudson, B.J.F., 1979. Comparison of chemical and sensory methods 
of evaluating thermally oxidized groundnut oil, J.Sci.Fd.Agric. 30: 515-520. 
Paredes, M.D.C. and Baker, R.C., 1987. Physical, chemical and sensory changes during thermal 
processing of three species of canned fish, J .Fd.Pros.Pres. 12: 71-81 .  
Paterson, R., 1970. An introduction to ion exchange, Heyden and Son Ltd., London. 
Patterson, H.B.W., 1989. Handling and storage of oilseeds, oils, fats and meal, Elsevier Applied 
Science, London. 
Pearce, E.A. and Smith, G.C., 1984. The world weather guide, Hutchinson, Sydney. 
Pennock, J.F., 1977. Terpenoids in marine invertebrates, In Biochemistry of lipids II, edited by 
Goodwin, T.W., p.153-213, University Park Press, Maryland. 
Perovic, V., 1977. The canning of fish in the tropic, In Proceedings of the conference on the 
handling, processing and marketing of tropical fish, Tropical Product Institute, Ministry 
of Overseas Development, London. 
Perry, R.H., Green, D.W. and Maloney, J.D., 1973. Perry's chemical engineer's handbook sixth 
edition, Me Graw-Hill Book Co., New York. 
Pigott, GM., 1967. Production of fish oil, In Fish oils, edited by Stansby,M.E., p.183-1 92, 
W estport, Connecticut. 
Pigott, G.M. and Tucker, B.W., 1987. Science opens new horizons for marine lipids in human 
nutrition, Fd.Rev.lnter. 3 (1 & 2): 105-138. 
300 
Pillai, V.K., 1974. Utilization of sardinella resources in India. In Fishery products, edited by 
Kreuzer, R., Fishing News Books, Surrey. 
Plackett, R.L. and Burman, J.P., 1946. The design of optimum multifactorial experiments, 
Biometrika 33: 305-325. 
Pohle, W.D., Gregory, R.L., Weiss, T.J., van Giessen, B.,  Taylor, J.R. and Ahern,J.J., 1964. A 
study of methods for evaluation of the stability of fats and shortening, JAOCS 41:  795-
798. 
Pokorny, J., 1987. Major factors affecting the autoxidation oflipids, In Autoxidation of unsaturated 
lipids, edited by Chan, H.W.S., p.141-206, Academic Press, London. 
Pokorny, J., 1991. Natural antioxidants for food use, Trends in Fd.Scie.Tech. 2 (9): 223-226. 
Puri, P.S., 1980. Winterization of oils and fats, JAOCS 54 (1 1): 91 1-917. 
Ragnarsson, J.O. and Labuza, T.P., 1977. Accelerated shelf-life testing for oxidation rancidity in 
foods- a review, Fd.Chem 2: 291-308. 
Ranken, M.D., 1989. Rancidity in meats, In Rancidity in foods, Alien, J.C. and Hamilton, R.J., 
p.225-236, Elsevier Applied Science, London. 
Ranken, M.D., 1990. Rancidity in meat and fish product, In Proceeding of a symposium on 
"Rancidity in foods" No.45, Leatherhead Food RA, Surrey. 
Rees, J.A.G., 1991.  Introduction, In Processing and packaging of heat preserved foods, edited by 
Rees, J.A.G. and Bettison, J., p.l-14, Blackie, Glasgow, London. 
Robertson, G.R., 1983. An introduction to the fundamental of thermal process evaluation, 
Department of food technology, Massey University, Palmerston North. 
Robinson, D.R., Prickett, J.D., Polisson, R., Steinberg, A.D. and Levine,L., 1985. The protective 
effect of dietary fish oil on marine lupus, Prostaglandins 30 (1): 51-71 .  
301 
Rossell, J.B ., 1986. Classical analysis of oils and fats, In Analysis of oils and fats, edited by 
Hamilton, R.J. and Rossell, J.B ., p.1-90, Elsevier Applied Science Publisher, London. 
Rossell, J.B ., 1989. Measurement of rancidity, In Rancidity in foods, edited by Alien, J.C. and 
Hamilton, R.J., p.23-52, Elsevier Applied Science Publisher , London. 
Rothberg, R.R., 1981. Product innovation in perspective , In Corporate strategy an product 
innovation, Edited by Rothberg, R.R., p.3-13, The Free Press, New York. 
Sadler, G.D., 1987. Aseptic chemistry, In Principles of aseptic processing and packaging, edited 
by Nelson, P.E., Chambers, J.V. and Rodriguez,J.H., p.45-61, The Food Processors 
Institute, Washington, D.C. 
Sagui, I. and Karel, M., 1980. Modelling of quality deterioration during food processing and 
storage, Fd.Tech. 34 (2): 78-85. 
Samuelson, 1953. Ion exchangers in analytical chemistry, John Wiley and Sons, Inc, New York. 
Sanders, T.A.B., 1989. Nutritional aspects of rancidity, In Rancidity in foods, edited by Alien, J.C. 
and Hamilton, R.J., p.125-139, Elsevier Applied Science, New York. 
Savage, D., 1984. Processing sterilized products, In Profitability of food processing. Symposium 
series No.84, Institute of Chemical Engineer, p.307-316, Pergamon Press, Oxford. 
Scott, K.C. and Latshaw, J.D., 1991 .  Effects of commercial processing on the fat-soluble vitamin 
content of menhaden fish oil, JAOCS 68 (4): 234-236. 
Setiabudi, E., 1990. Pengaruh waktu penyimpanan dan jenis filter pada jumlah omega-3 dalam 
minyak limbah basil pengalengan dan penepungan ikan lemuru, Master thesis, Bogor 
Agricultural University, Bogor. 
Sheppard, A.J., Iverson, J.L. and Weihrauch, J.L., 1978. Composition of selected dietary fats, oils, 
margarines, and butter, In Handbook of lipid research 1. Fatty acids and glycerides, edited 
by Kuksis,A., p.341-374, Plenum Press, New York. 
302 
Sherwin, E.R., 1990. Antioxidants, In Food additives, edited by Branen, AL., Davidson, P.M. and 
Salminen, S., p.139-193, Marcel Dekker Inc., New York. 
Shirakawa, Y., Minowa, Y., Azumi, T. and Hisano, J., Process for producing protein-rich fish meal 
and/or fish oil, European patent application EPO 301 795 AI, FSTA 21 (10): p.218. 
Simopoulos, A.P., 1991.  Omega-3 fatty acids in health and disease and in growth and development, 
Am.J.Clin.Nutr. 54: 438-463. 
Simpson, K.L., Katayama, T. and Chichester, C.D., 1981. Carotenoids in fish feeds, In Carotenoids 
as colorants and vitamin A precursors, edited by Bauemfeind,J .C., p.463-538, Academic 
Press Inc., New York. 
Sleeter, R.T., 1981 . Effects of processing on quality of soybean oil, JAOCS 58 (3): 239-247. 
Sloan, A.E., 1987. The next generation of nutrition claims, Cereal Fd.World 32 (12): 870-872. 
Smouse, T.H., 1978. Oil stability, In Encyclopedia of food science, edited by Peterson, M.S.P. and 
Johnson, A.H., p.577-580, The AVI Publishing Co.Inc., Westport, Connecticut. 
Snee, R.D., 1971 .  Design and analysis of mixture experiments, J .Qual.Tech. 3 (4): 159-169 . 
. Soesetiadi, D., 1977. Anatomi veteriner-Alat gerak-penuntun praktikum osteology, Anatomi 
Department, Veterinary Science Faculty, Bogor Agricultural University, Bogor. 
Stansby, M.E., 1963. Cured fishery products, In Industrial fishery technology, edited by 
Stansby,M.E. and Dascow, J.A., p.323-335, Reinhold Publishing Co., London. 
Stansby, M.E., 1969. Nutritional properties of fish oil, World Rev .Nut. and Diet 1 1 : 46-105.  
Stansby, M.E., 1979. Marine-derivated fatty acids or fish oils as raw material for fatty acid 
manufacture, JAOCS 56: 793A-796A. 
303 
Stansby, M.E., 1982, Properties of fish oils and their application to handling of fish and to 
nutritional and industrial use, In Chemistry and biochemistry of marine food products, 
edited by Martin, R.E., Flick, GJ., Hebard, CE. and Ward, D.R., p.75-92, The A VI 
Publishing Co.Inc., Westport, Connecticut. 
Stansby, M.E., 1990. Deterioration, In Fish oils in nutrition, edited by Stansby,M.E., p.120-140, 
Van Nostrand Reinhold, New York. 
Stansby, M.E., 1990. Nutritional properties of fish oil for human consumption-early development, 
In Fish oils in nutrition, edited by Stansby,M.E., p.268-288, Van Nostrand Reinhold, New 
York. 
Stecher, P.G., Windholz, M., Leahy, D.S., Bolton, D.M. and Eaton, L.G., 1968. The merck Index, 
Merck & Co.Inc., New York. 
Stone, H., 1988. Using sensory resources to identify successful products, In Food acceptability, 
edited by Thomson, D.M.H., p.283-296, Elsevier Applied Science, London. 
Stowe, R.A. and Mayer, R.P., 1966. Efficient screening of process variables, Ind.Eng.Chem. 58 (2): 
36-40. 
Stuckey, B.N., 1972. Antioxidants as food stabilizers, In CRC Handbook of food additives, edited 
by Furia, T E., CRC Press, Florida. 
Sugisawa, H., Nakamura, K. and Tamura, H., 1990. The aroma proflle of the volatile in marine 
green algae, Fd.Rev.Inter. 6 (4): 573-589. 
Sukandar, A. and Mihardjo, I.M., 1989. Menuju sasaran ekspor dengan pengolahan untuk 
pengalengan ikan tuna, Paper presented in lokakarya perikanan tuna, Jakarta 5-6 June 
1989. 
Suparno, Poernomo, A., Sarnianto, P., Setiabudi, E. and Subroto, W., 1989. Analysis suplai bahan 
baku dan lokasi pabrik kapsul omega-3 :  Laporan survey, Research Institute for Fishery 
Technology, Jakarta. 
304 
Taguchi, K., Iwami, K., Kawabata, M. and Ibuki, F., 1988. Antioxidant effect of wheat gliadin and 
hen's egg white in powder model system: protection against oxidative deterioration of 
safflower oil and sardine oil, Agr.and Bio.Chem 52 (2): 539-545. 
Tanchotikul, U. and Hsieh, T.C.Y., 1989. Volatile flavor components in crayfish waste, J.Fd.Scie. 
54: 1515-1420. 
Tanikawa, E., 1971 .  Marine products in Japan, Kosheisa-Kosheikaku Co, Tokyo. 
Thomson, S.W., 1966. Survey of accelerated tests for determining the stability of oils and fats, In 
Laboratory handbook for oil and fat analysis, edited by Cocks, L.V. and van Rede, C, 
p.340-345, Academic Press, London. 
Tocher, D.R., Webster, A. and Sargent, J.R., 1986. Utilization of porcine pancreatic phospholipase 
A2 for the preparation of a marine flsh oil enriched in (n-3) polyunsaturated fatty acids, 
Biotech, and app.biochem. 1986 (8): 83-95. 
Tokarska, B.,  Hawrysh, Z.J. and Claudinin, M.T., 1986. Study of the effect of antioxidants on 
storage stability of canola oil using gas liquid chromatography, Can.Ins.Fd.Scie.Tech. 19  
(3): 130-133. 
Toyama, Y. and Kaneda, T. 1965. Nutritive aspects of flsh oils, In Fish as foods Vol.II, edited by 
Borgstrom, G., p.149-173, Academic Press, New York. 
Valenzuela, A., Nieto, S. ,  Cassels, B .K. and Speisky, H., 1991 .  Inhibitory effect of boldine on fish 
oil oxidation, JAOCS 68 (12): 935-937. 
Van den Brook, C.J.H., 1965. Fish canning, In Fish as foods vol IV part 2, edited by Borgstrom,G., 
p.127-205, Academic Press, New York. 
Van Klaveren, F.W. and Legendre, R., 1965. Salted cod, In Fish as food vol. HI, edited by 
Borgstrom, G., p.133-163 ,  Academic Press, New York. 
Van Straten, S .  and Maarse, H., 1983. Volatile compounds in food-qualitative data, Division for 
Nutrition and Food Research, TNO, Utrechtseweg, Netherlands. 
305 
Van Til, H., 1991. Fo calculation program, Department of Food Technology, Massey University, 
Palmerston North. 
Van Til, H., 1991 .  Plackett and Burman Program, Department of Food Technology, Massey 
University, Palmerston North. 
Van Wijngaarden, D., 1967. Modified rapid preparation of fatty acid esters from lipids for gas 
chromatographic analysis, Anal.Chem. 39 (7): 848-849. 
Vejaphan, W., Hsieh, T.C.Y. and Williams, S .S ., 1988. Volatile flavor components from boiled 
crayfish (Procambarus clarkii) tail meat, J.Fd.Scie. 53 (6): 166-1670. 
Vidal, J.P., Fort, J.J., Gaulsier, P. and Richard, H., 1989. Vanilla aroma extraction by dense carbon 
dioxide, Sciences des Aliments 9: 89-100. 
Voskresensky, N.A., 1965. Salting of herring, In Fish as foods vol Ill, edited by Borgstrom, G., 
p.107-131,  Academic Press, New York. 
Wakao, A. and Pazos, M., 1984. Browning reaction in canned sardines: the role of lipids and 
arnines, B uletin de Investigation, Institiuto Tecnologico Pesquero 2 (1 ): 63-71 ,  FST A 
1987 19 (1): 1R23. 
-Walonick, D.S., 1987. Stat-Packets, Walonick Associates Inc., Minneapolis, M.N. 
Warne, D., 1988. Manual on fish canning. FAO Fish Tech.Pap. (285), FAO of United Nation, 
Rome. 
Wheaton, R.M. and Lefevre, LJ., 1981 .  Ion exchange, In Encyclopedia of chemical technology vol 
.!l, edited by Mark, H.F., Otbmer, D.F., Overberger,C.G. and Seaburg, G.T., p.678-705, 
John Willey and Sons, New York. 
Wiabayani, S.,  1983. Pengalengan ikan di PT Bangka Pioneer Industries Ltd., Akademi Usaba 
Perikanan, Jakarta. 
Williams, K.R., 1963. Comparing screening designs, Ind.Eng.Chem, 55 (6): 29-32. 
306 
Windsor, M. and Barlow, S ., 1981 .  Introduction to fishery by-products, Fishing News Books Ltd., 
Surrey. 
Wiriacharee, P. 1990. The systematic development of a controlled fermentation process using mixed 
bacterial starter culture for nham, a Thai product, PhD Thesis, Massey University, 
Palmerston North. 
Yongwanitchai, W. and Ward, O.P., 1989. Omega-3 fatty acids: alternative sources of production, 
Proc.Biochem. 1989 (8): 1 17-125. 
Young, F.V.K., 1982. The production and use of fish oils, In Nutritional evaluation of long-chain 
fatty acids in fish oil, edited by Barlow,S.M. and Stansby, M.E., p.1-24, Academic Press, 
London. 
Zama, K., Takama, K. and Mizushima, Y., 1979. Effect of metal salts and antioxidants on the 
oxidation of fish lipids during storage under the conditions of low and intermediate 
moisture, J.Fd.Proc,Prev. 3: 249-257. 
APPENDICES 
Appendix 3 .1 . Purging system for collection of volatile flavour compounds of fish oil 
T : ?? 
- l 
I I I I I I I I I I I I 
L.-- r- -,-..J  
I I I I I 
I I 
I I 
20 Cm I I I I lY" I I I I 
I I 
I I I 
2 . 3  Cm-
L 1 
17 
?N2 
?t:?t.(;?\)?Y:?:?t-?'j;!:? 
Cm 
0 
Tena:x-'fA 
head space 
,IL, water V Bath 
307 
l 
308 
Appendix 3.2. Container used for colour analysis of fish flesh and tomato sauce 
A .  Container for fish flesh 
5 Cm  
// .0 
' 
I 
I 
I 
I 
5 . 5  Cm I I 
I 
I 
I 
I I 
I 
I 
,.J- - - - - - - - - - - - - -
?, ? 1 Cm 
B .  Glass cuvette for tomato sauce 
Appendix 4.1 .  Questionnaire used for fish meal factory survey 
FISH MEAL FACTORY SURVEY 
1 .  What is the position of fish meal production in your company? 
( ) main product 
( ) main product together with other main products 
( ) by-product 
309 
2. Raw materials used for fish meal production in your factory (your answer can be more than 
one): 
( ) one species: ( ) sardine 
( ) mackerel 
( )  scads 
( ) others, .. . . . . . . . . . . . . . ..... . 
( )  mixture of fish species, mainly ... ... . . . . . . . . . . . . . . . . . . . . 
( ) fish wastes: ( ) canning waste, mainly ..... . . . . . . . . . . . . . 
( ) industry .... . . . . . . . . . , mainly ... . . . . . . . . . . 
( ) others, . . . . .. .. .. . . . . . . . . . . . . . . 
3. In terms of the above raw materials, please specify the percentage of each raw material used in 
your factory . 
. . .. % one species: .. .. sardine 
.. . . mackerel 
.... scads 
.... others, .. . . . . . . . . . . . . . . . . . . . . . . . . . 
. ... % mixture of fish species 
. . .. % fish waste: .... canning waste 
.... industry . .. . . . . . . . . . . . . . . . . 
.... % others, ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
? 4. What is the method used for fish meal production in your factory? 
( ) wet rendering method 
( ) dry rendering method 
( ) other methods, please specify ... . . . . . . . . . . . . . . . . . . 
5. Do you separate the oil from press liquor? 
( ) Yes 
( ) No 
If "YES", do you separate oil from all types of raw material? 
( ) Yes 
( ) No 
If "NO", what are raw materials which you do not separate their oils? 
Please specify, . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 
6. Do you refine the oil to make it consumable for human? 
( ) Yes 
( ) No 
If "YES", what is the method being used to refine fish oil in your factory? 
( ) alkali refme method 
( ) molecular distillation 
( ) thawing fractionation 
( ) others, . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 
================================= 
Additional Information: 
Factory 
Location 
Fish oil packaging 
Fish oil selling price 
Fish oil buyers : ( ) food companies 
( ) pharmaceutical company 
( ) flsh oil traders 
( ) exporter 
( ) others, .. . . . . . . . . . . . . . . . . . . . . 
Raw material price: - sardine Rp .. . . .. . . . . . . . . . . . 
- mackerel Rp . .. .. . . . . . . . . . . . . 
- scad Rp ..... . . . . . . . . . . . . 
- mixture of fish Rp ..... . . . . . . . . . . . . 
- fish waste Rp . . . . . . .  ?? ?? ? ? ? ? ? ?  
310 
Appendix 4.2. Sensory evaluation sheet for Indonesian fish oil 
SENSORY EVALUATION OF INDONESIAN FISH OIL 
Name: ... . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Date: . . . .. . . . . . . . . .... .......... . . .......... . 
Instruction: 
There are some samples in front of you and you are asked to evaluate their odour. 
Score the samples using the following scale: 
Score Descriptions 
1 none fishy/undesirable odour 
2 
3 slightly fishy/undesirable odour 
4 
5 moderately fishy/undesirable odour 
6 
7 strong fishy/undesirable odour 
8 
9 extremely strong fishy/undesirable odour 
Sample Code Odour Score 
3 1 1  
Appendix 4.3. Factories participating in the survey 
Fish Meal Factories 
1 .  C. V. Jaya Baru 
2. P.T. Sinar Tratas 
3 .  P.T. Avila Prima 
4. Bpk. Salim 
5.  C.V. Bali Indah 
6. P.T. Sumber Rejeki 
7. P.T. Bali Maya Permai 
8. P.T. Maya Muncar 
9. C.V. Samudra Raya 
10. P.T. Indo Bali 
1 1 .  C. V. Harapan Lancar 
12. P.T. Pengambengan Raya 
13.  P.T. Sinar Laut Jaya 
14. P.T. Karya Manunggal Prima Sukses 
15. N.V. Muncar 
16. P.T. Sumber Yala Samudra 
17. P.T. Sumino 
-18 .  P.T. Bali Raya 
19. P.T. Blambangan Raya 
3 12  
Appendix 4.4. Fatty acid profiles of Indonesian Fish oil 
Fatty Factory/Fish Oil Type/Quality 
Acid 
Profiles A/FM/- BIFM!- CIFM!- DIFM/- FJFM/- FIFM!-
14:0 12.6 13.0 13.5 12.4 9.4 13.0 
15:0 0.6 0.8 0.8 0.8 1 .3 0.8 
16:0 19.6 22.1 21.9 21.3 22.4 20.2 
16: 1 14.8 14.3 15.5 15.9 10.4 18.0 
17:0 0.6 0.8 0.9 0.8 1 .7 0.8 
17: 1 4.0 4.0 4.1 4.4 3 .4 4.8 
18:0 3.2 3.8 3.7 3.5 5.75 3.4 
18: 1  7.3 8.0 7.9 7.9 14.2 8.0 
18:2 1 . 1  1.2 1 .3 1 .1  1 .3 1 .2 
18:3 1 .1 1 .2 1 .3 1 . 1 1 .2 1 .1 
18:4 1 .9 2.0 2.1 1.7 1 .2 1 .9 
20: 1 0.9 0.6 0.6 1 .0 1 .3 0.8 
20:3 0.1 0.1 0.1 0.1 0.2 0.1 
20:4 2.7 2.3 2.2 2.3 2.1 2.5 
20:5 20.0 17.1 17.1 16.4 9.2 17.3 
22: 1 2.1 1 .4 0.9 2.0 1 .5 1 .6 
22:4 0.1 0.1 0.1 0.1 0.2 0.1 
22:5 1 .2 1 . 1  0.9 1.3 0.8 0.9 
22:6 5.8 6.1 5.0 5.7 12.5 3.5 
?ote: J:<M - Ftsh meal 
CW = Canning waste 
FWM = Fish waste meal 
G!CW!- H!FM/1 
13.2 13.1 
0.8 0.8 
21.2 22.3 
14.9 14. 1 
0.8 1 .0 
4. 1 4.0 
3.5 3.8 
8.2 8.2 
1.3 1 .3 
1 .3 1 .3 
2.2 2.0 
0.7 0.7 
0.1 0.1 
2.3 2.3 
17.6 16.0 
0.9 1 .5 
0.1 0.1 
1 .0 1 . 1  
5.8 6.3 
HJFMIII 
13.8 
0.8 
21.5 
13.8 
0.9 
3.8 
4.1 
8.1 
1.3 
1 .3 
1 .8 
1 . 1  
0.1 
2.2 
15.2 
2.1 
0.1 
1 .4 
6.4 
HIFMII I 
13.8 
0.9 
21.8 
16.6 
1 .0 
4.5 
3.6 
7.8 
1 .2 
1 . 1  
1 .7 
1 .4 
0.1 
2.1 
13.9 
2.3 
0.1 
1.1 
5.0 
w 
-
w 
Continuation appendix 4.4 
--------------------- ---------
Fatty 
Acid 
Profiles WCW/-
14:0 1 1 .5 
15:0 0.9 
16:0 21.3 
16: 1 15.3 
17:0 1 .0 
17:1  4.4 
18:0 3 .9 
18: 1  8.7 
18:2 1 .4 
18:3 1 .4 
18:4 2.3 
20: 1 0.8 
20:3 0.1 
20:4 2.3 
20:5 16.7 
22: 1 1 .0 
22:4 0.1 
22:5 1 .0 
22:6 5.9 
"''ote: .t<M - l?tsh meal 
CW = Canning waste 
1/FM/-
13.3 
0.7 
20.7 
15.6 
0.7 
4.3 
3.4 
7.2 
1 .2 
1 .2 
2.0 
0.7 
0.1 
2.4 
18.6 
1 .4 
0.1 
1 .0 
5.3 
FWM = Fish waste meal 
Factory/Fish Oil Type/Quality 
1/CW/- JIFMI- KJFM/-
12.8 1 1 .6 12.5 
0.8 0.8 0.9 
21.3 20.7 21.1 
14.8 16.4 16.3 
0.9 0.8 0.9 
4.1 4.5 3.2 
3.7 3.8 4.6 
8.2 8.1 1 1 .5 
1 .3 1 .3 1.2 
1 .3 1 .2 1 . 1  
2.1 2.1 1 .7 
0.8 0.8 1 .4 
0.1 0.1 0.1 
2.3 2.5 2.2 
17.2 17.4 1 1.5 
1 .2 1 .5 2.6 
0.1 0.1 0.2 
1 . 1  1 . 1  1 .7 
5.8 5.2 5.3 
- -----??????????-??-?-??---
UFM/ref M/FM/A M!FMIB 
12.4 8.8 10.9 
0.9 0.9 1 .0 
21.4 21.6 21.0 
14.6 1 1 .7 14.3 
1 .0 1 .3 1 . 1  
4.0 3 .7 4.5 
4.0 5.0 4.3 
8.4 12.0 9.6 
1.4 1 .3 1 .4 
1.4 1 .2 1 .3 
2.0 1 .6 2.0 
0.8 1 .0 0.7 
0.1 0.2 0.1 
2.3 2.3 2.3 
16.6 12.8 16. 1  
1.5 1 .2 0.8 
0.1 0.2 0.1 
1 . 1  1 .0 1 .0 
6.1 12.2 7.4 
M/CW/-
1 1.8 
0.8 
22.9 
15.8 
0.9 
4.4 
3.5 
8.1 
1 .3 
1 .2 
1 .9 
1 .5 
0.1 
2.1 
15.4 
2.3 
0.1 
0.9 
4.9 
NIFWM/-
12.8 
0.8 
21.3 
14.8 
0.9 
4.1 
3.7 
8.2 
1 .3 
1 .3 
2.1 
0.8 
0.1 
2.3 
17.2 
1 .2 
0.1 
1 . 1  
5.8 
I 
w 
...... 
.j:.. 
Continuation of appendix 4.4 
Fatty Factory/Fish Oil Type/Quality 
Acid 
Profiles 0/CW/1 0/FWM/II PIFWM/-
14:0 13.0 14.5 3.9 
15:0 0.7 0.7 1 .4 
16:0 21.6 21.5 23.8 
16: 1 14.5 13.6 5 .4 
17:0 0.8 0.9 2.3 
17: 1 4.1 3.6 2.5 
18:0 3.6 3.7 7.2 
18 :1  8.3 8 .1  18.4 
18:2 1 .3 1 .2 1 .5 
18:3 1 .2 2.2 1 .2 
18:4 2.1 1 .8 0.9 
20: 1 0.7 1 .4 0.8 
20:3 0.1 0.1 0.2 
20:4 2.3 2.3 1 .9 
20:5 17.5 15.2 5.0 
22: 1 0.9 2.2 0.3 
22:4 0.1  0.6 0.2 
22:5 1 .3 1 .2 0.8 
22:6 6.0 6.4 22.2 
Note: FM - Ftsh meal 
CW = Canning waste 
FWM = Fish waste meal 
QIFWM/-
11 . 1  
1 .0 
21. 1 
14.6 
1 .1  
4.6 
4.1 
9.1 
1.5 
1 .4 
2.1 
0.9 
0.1 
2.4 
16.0 
1 .3 
0.1 
1 .3 
6.0 
w 
....... 
V! 
3 16 
Appendix 5 .1 .  Results of chemical, physical and organoleptic analysis of fish oil as the effects of 
fish oil and resin ratio 
Analysis : Free fatty acid value - Crude oil (% oleic acid) 
Fish oil Resin Fraction -1 Fraction-2 
Untreated fish oil 0.66 0.66 
1 : 1 0.43 1 .12 
2 :  1 0.42 1 .72 
3 :  1 0.43 2.30 
4 :  1 0.43 2.66 
Analysis : Free fatty acid value - Orange rohghy oil (%  oleic acid) 
Fish oil : Resin Fraction -1 Fraction-2 
Untreated fish oil 0.40 0.40 
1 : 1 0.25 0.94 
2 : 1  0.26 1 .22 
3 :  1 0.27 1 .79 
4 :  1 0.25 2.34 
Analysis : Refractive Index value - Crude oil (at 20?C) 
Fish oil : Resin Fraction -1 Fraction-2 
Untreated fish oil 1 .4673 1 .4673 
1 : 1 1 .4667 1 .4710 
2 :  1 1 .4689 1 .4715 
3 :  1 1 .4685 1 .4720 
4 :  1 1 .4700 1 .4728 
317 
Analysis : Refractive Index value ? Orange roughy oil (at 20?C) 
Fish oil : Resin Fraction -1 Fraction-2 
Untreated fish oil 1 .4652 1 .4652 
1 : 1 1 .4586 1 .4641 
2 :  1 . 1 .4619  1 .4632 
3 :  1 1 .4620 1 .4631 
4 :  1 1 .4630 1 .4620 
Analysis : Sensory analysis for odour ? Crude oil 
Fish oil Resin Fraction -1 Fraction-2 
Untreated fish oil 6.38 6.38 
1 : 1 2.78 5.10 
2 :  1 4.06 6.17 
3 : 1  4.14 7.07 
4 :  1 4.17 7.22 
Analysis : Sensory analysis for taste - Crude oil 
Fish oil : Resin Fraction -1 Fraction-2 
Untreated fish oil 5.88 5.88 
1 : 1 2.88 4.19 
2 :  1 3 .22 6.28 
3 : 1 3 .64 5.98 
4 :  1 3 .79 6.50 
3 18  
Analysis : Sensory analysis for odour - Orange roughy oil 
Fish oil Resin Fraction -1 Fraction-2 
Untreated fish oil 4.06 4.06 
1 : 1 3.64 5.79 
2 :  1 3 .44 6.38 
3 : 1 3.71 6.57 
4 :  1 3 .86 6.86 
Analysis : Sensory analysis for taste - Orange roughy oil 
Fish oil Resin Fraction -1 Fraction-2 
Untreated fish oil 4.15 4.15  
1 : 1 3.19 5.71 
2 :  1 3 .44 5.50 
3 :  1 3.43 6.63 
4 :  1 3 .57 6.93 
Analysis : Colour absorbance at 490 run 
-
Fish oil : resin Fraction -1 Fraction-2 
Crude oil Orange roughy oil 
Untreated fish oil 0.90 1 .96 
1 : 1 0.81 1 .7 1  
2 :  1 0.82 1 .74 
3 :  1 0.82 1 .75 
4 :  1 0.83 1 .76 
319  
Analysis : Fatty acid profiles - Crude oil 
Fish oil : Resin 
Untreated 
F.A. Fish 1 1 2 1 3 1 4 1 
Oil F-1 F-2 . F-1 F-2 F-1 F-2 F-1 F-2 
14:0 4.4 4.3 5.3 4.6 5.5 4.4 5.4 4.6 5 .4 
15:0 0.7 0.6 0.7 0.6 0.7 0.6 0.7 0.6 0.7 
16:0 17.2 16.9 17.3 16.7 17.5 16.8 17.3 16.0 17.5 
16:1 8.1 7.7 8.3 7.4 8.3 7.5 8.3 7.5 8.3 
17:0 0.3 0.4 0.3 0.4 0.3 0.4 0.3 0.4 0.3 
17:1 2.3 2.3 2.3 2.3 2.3 2.3 2.3 2.3 2.3 
18:0 2.6 2.6 2.4 2.6 2.4 2.5 2.4 2.6 2.4 
18 :1  34.2 35.1 33.1 34.0 33.1 34.4 32.9 34.4 32.9 
18:2 1 .6 1 .6 1 .6 1 .6 1 .6 1 .6 1 .6 1 .6 1 .6 
18:3 0.9 1 .0 1 . 1  0.9 0.8 0.9 0.8 0.9 0.8 
18:4 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 
20:1 10.8 1 1 .7 10.2 1 1 .4 10.2 1 1 .3 10.0 1 1 .5 10.0 
20:3 0.3 0.3 0.3 0.3 0.3 0.4 0.3 0.3 0.3 
20:4 0.6 0.6 0.5 0.6 0.5 0.6 0.7 0.6 0.7 
20:5 9.2 4.4 4.4 4.4 4.3 4.4 4.7 4.4 4.7 
22: 1 4.9 5.0 4.9 5.1 4.8 5.0 5.0 5.2 4.8 
22:4 - - - - - - - - -
22:5 0.8 0.9 0.9 0.9 0.9 0.8 0.8 0.9 0.9 
22:6 5.1 5.3 5.4 5.3 5.6 5 .1  5.5 5.3 5.5 
?ote: .F - Fraction 
320 
Analysis : Fatty acid profiles ? Orange roughy oil 
Fish oil : Resin 
Untreated 
F. A. Fish 1 1 2 1 3 1 4 1 
Oil F-1 F-2 F-1 F-2 F-1 F-2 F-1 F-2 
14:0 1 .8 1 .8 2.2 1 .7 2.2 1 .7 2.3 1 .7 2.4 
15:0 0.1 0.1 0.1 0.1 0.2 0.1 0.2 0.1 0.2 
16:0 3 .3 3 .5 3.7 3.2 4.0 3.2 4.1 3.2 4.4 
16:1 1 1 .4 12.1 13.7 1 1 .9 13.8 1 1 .5 13.9 1 1 .9 13 .6 
17:0 0.3 0.3 0.2 0.3 0.2 0.2 0.2 0.2 0.2 
17:1 2.5 2.5 2.6 2.5 2.6 2.5 2.6 2.6 2.6 
18:0 0.6 0.4 0.4 0.4 0.4 0.5 0.4 0.3 0.4 
18:1  52.6 52.9 6.0 53.3 52.0 53.3 5 1 .7 52.1 5 1 .3 
18:2 1.4 1.4 1 .4 1 .4 1 .4 1 .4 1 .4 1 .4 1 .5 
18:3 1 . 1  1 . 1  1 .0 1 . 1  1 .0 1 . 1  0.9 1 . 1  1 .0 
18:4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 
20:1 14.8 14.7 1 3.2 14.8 13 .6 15.0 12.8 14.6 12.8 
20:3 0.3 0.3 0.2 0.2 0.2 0.3 0.2 0.3 0.2 
20:4 0.4 0.3 0.4 0.3 0.3 0.4 0.5 0.4 0.5 
20:5 1 .4 1 .4 1.4 1 .3 1 .4 1.4 1 .5 1 .5 1 .7 
22:1 5.7 5.6 5.6 5.8 5.4 5.8 5.4 5.9 5.3 
22:4 - - - - - - - - -
22:5 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 
22:6 1 . 1  1 .0 1 . 1  1 .0 1 .2 1 .0 1 .3 1 .0 1 .5 
321 
Appendix 5.2. Results of chemical, physical and organoleptic analysis of fish oil as the effects of 
multiple refining using resin packed column 
Analysis : Free fatty acid value (% oleic acid) - Crude oil 
Refining/column Fraction -1 Fraction-2 
Untreated fish oil 0.63 0.63 
I 0.48 1 .04 
II 0.41 0.71 
m 0.35 0.62 
IV 0.25 0.44 
Analysis : Free fatty acid value (% oleic acid) - Orange roughy oil 
Refining/column Fraction -1 Fraction-2 
Untreated fish oil 0.67 0.67 
I 0.47 1 .14 
II 0.40 0.71 
m 0.32 0.63 
IV 0.26 0.50 
Analysis : Refractive Index value - Crude oil (at 20?C) 
Refining/column Fraction - 1  Fraction-2 
Untreated fish oil 1 .4740 1 .4740 
I 1 .4665 1 .471 8  
II 1 .4625 1 .4739 
III 1 .4601 1 .4739 
IV 1 .4543 1 .4737 
322 
Analysis : Refractive Index value - Orange roughy oil (at 20?C) 
Refining/column Fraction -1  Fraction-2 
Untreated fish oil 1 .4659 1 .4659 
I 1 .4593 1 .4630 
II 1 .4555 1 .4649 
Ill 1 .4550 1 .4651 
IV 1 .4506 1 .4660 
Analysis : Sensory evaluation for Odour - Crude oil 
Refining/column Fraction -1  Fraction-2 
Untreated fish oil 5.70 5.70 
I 3.19 6.75 
II 2.90 6.20 
Ill 2.70 6.00 
IV 2.38 4.13 
Analysis : Sensory evaluation for Taste - Crude oil 
Refining/column Fraction -1  Fraction-2 
Untreated fish oil 6.00 6.00 
I 3.00 6.94 
II 2.90 5.78 
III 2.90 5.84 
IV 2.50 4.00 
323 
Analysis : Sensory evaluation for Odour - Orange roughy oil 
Refining/column Fraction -1 Fraction-2 
Untreated fish oil 6.74 6.74 
I 4.76 7.03 
II 4.70 5.85 
m 4.53 5.79 
IV 4.14 5.64 
Analysis : Sensory evaluation for Taste - Orange roughy oil 
Refining/column Fraction -1 Fraction-2 
Untreated fish oil 5.72 5.72 
I 4.60 6.92 
II 4.60 5.73 
m 4.1 1  5.51 
IV 4.06 5.03 
Analysis : Colour absorbance at 490 run 
Refining/column Fraction -1 Fraction-I 
Crude Oil Orange roughy oil 
Untreated fish oil 0.49 1 .52 
I 0.43 1 .37 
II 0.40 1 .29 
m 0.40 1 .23 
IV 0.36 1 .1 3  
324 
Analysis: Fatty acid profiles - Crude oil 
Refining - Column 
Untreated 
F. A. Fish I II Ill IV 
Oil F-1 F-2 F-1 F-2 F-1 F-2 F-1 F-2 
14:0 6.2 5.7 5.9 5.8 6.1 5.7 6.1 5.8 6.0 
15:0 0.8 0.7 0.8 0.8 0.8 0.7 0.8 0.8 0.6 
16:0 18.5 17.7 18.2 18.0 18.3 17.8 18 .6 18 .0 1 8.3 
16:1 9.0 8.5 8.8 8.6 8.9 8.5 9.0 8.6 8.9 
17:0 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 
17:1  2.7 2.5 2.6 2.6 2.7 2.5 2.7 2.5 2.6 
18:0 1 .5 1 .6 2.1 2.2 1 .5 2.2 1 .7 2.1 2.1  
18:1  32.9 32.6 32.3 32.7 33.0 32.8 33.0 32.8 33 .1  
18:2 1 .4 1 .4 1 .4 1 .4 1 .5 1 .4 1 .4 1 .4 1 .4 
18:3 0.6 0.6 0.7 0.6 0.7 0.6 0.6 0.6 0.7 
18 :4 0.7 0.7 0.7 0.8 0.7 0.7 0.7 0.7 0.7 
20: 1  10.5 10.8 10.5 10.6 10.5 10.8 10.4 10.7 1 0.4 
20:3 0.3 0.3 0.3 0?3 0.3 0.3 0.3 0.3 03 
20:4 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.5 
20:5 3 .9 4.2 3.9 3.9 3 .8 3.9 3.8 3 .9 3.8 
22:1 4.2 4.7 4.2 4.3 4.2 4.4 4.0 4.4 4.2 
22:4 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 
22:5 0.9 1 .0 0.9 1 . 1  0.9 0.9 0.9 0.9 0.9 
22:6 5.0 5.7 . 5.6 5.3 5.2 5.6 4.9 5.4 5.0 
?ote: F - fraction 
325 
Analysis : Fatty acid profiles - Orange roughy oil 
Refining - Column 
Untreated 
F. A. Fish I II m IV 
Oil F-1 F-2 F-1 F-2 F-1 F-2 F-1 F-2 
14:0 2.3 2.4 2.6 2.4 2.4 2.4 2.4 2.4 2.5 
15:0 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 
16:0 4.8 4.6 5.0 4.7 4.8 4.8 4.7 4.7 4.8 
16: 1  13.0 12.5 13.2 13.0 13.0 12.8 12.7 12.9 13 . 1  
17:0 - - - - - - - - -
17: 1 2.3 2.6 2.6 2.7 2.6 2.6 2.6 2.6 2.6 
18:0 1 .0 1 .5 1 .5 0.6 0.9 1 .3 1 .0 1 .4 0.6 
18:1  52.7 52.3 52.0 52.6 52.0 52.6 5 1 .8 52.2 5 1 .8 
18:2 1 .5 1 .5 1 .4 1 .4 1.4 1 .4 1 .5 1 .4 1 .4 
18:3 0.5 0.5 0.5 0.4 0.5 0.4 0.5 0.4 0.5 
18:4 0.4 0.4 0.4 0.4 0.4 0.4 0.6 0.4 0.2 
20:1 13.4 13.7 13.0 13.7 13.8 12.9 13 .7 13 .0 1 3.3 
20:3 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 
20:4 0.3 0.3 0.3 0.2 0.3 0.3 0.3 0.3 0.3 
20:5 1 . 1  1 .2 1 .1 1 .2 1 .2 1 .3 1 .6 1 .3 1 .8 
22:1 4.9 5 .1  4.5 4.8 4.8 5.0 4.9 5.2 5 .4 
22:4 - - - - - - - - -
22:5 0.2 0.3 0.3 0.2 0.3 0.2 0.2 0.2 0.2 
22:6 1.2 1 . 1  1 .2 1 .2 1 .2 1 .2 1 .2 1 . 1  1 . 1  
'llote: F - tracuon 
326 
Appendix 5.3. Results of chemical, physical and organoleptic analysis of fish oil as the effects of 
vacuum pressure application during resin refining process 
Analysis : Free fatty acid value (% oleic acid) 
Crude Oil Orange roughy oil 
Treatment 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
Untreated fish oil 0.63 0.63 0.79 0.79 
Without vacuum pressure 0.55 0.94 0.66 1 .20 
With vacuum pressure 0.48 0.76 0.68 0.94 
Analysis : Refractive Index (20?C) 
Crude Oil Orange roughy oil 
Treatment 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
Untreated fish oil 1 .4731 1 .4731 1 .4651 1 .4651 
Without vacuum pressure 1 .4710  1 .4732 1 .4631 1 .4655 
With vacuum pressure 1 .4691 1 .4735 1 .4630 1 .4660 
Analysis : Colour absorbance value 490 nm 
Crude Oil Orange roughy oil 
Treatment 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
Untreated fish oil 0.76 0.76 1 .59 1 .59 
Without vacuum pressure 0.73 0.80 1 .48 1 .76 
With vacuum pressure 0.72 0.80 1 .51  1 .72 
327 
Analysis : Sensory evaluation for Odour 
Crude Oil Ofange roughy oil 
Treatment 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
Untreated fish oil 6.06 6.06 6.31 6.31 
Without vacuum pressure 4.25 5.56 3 .63 6.00 
With vacuum pressure 4.63 5.44 4.00 4.88 
Analysis: Sensory evaluation for Taste 
Crude Oil Orange roughy oil 
Treatment 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
Untreated fish oil 5.87 5.87 5.50 5.50 
Without vacuum pressure 4.31 5.56 3 .75 6.00 
With vacuum pressure 3 .94 5.00 3.75 5.19 
Analysis: Fatty acid profiles 
Untreated Oil Crude Oil 
Fatty 
Crude Orange Without Vacuum With Vacuum Acids 
Oil roughy Pressure Pressure 
oil 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
14:0 3.5 1 .5 3.9 3.9 3 .9 3.8 
15:0 0.5 - 0.5 0.5 0.5 0.5 
16:0 14.9 3 .7 15.2 15.2 15.2 15.0 
16: 1 7.0 9.9 7.2 7.1 7.1 7.0 
17:0 0.4 0.2 0.4 0.4 0.4 0.4 
17: 1 2.2 2.1 2.0 2.0 2.1 2.1 
18:0 - 0.8 - - - -
18: 1 36.0 51 .4 35.9 36. 1 35.7 36.5 
18:2 1 .8 1 .5 1 .7 1 .6 1 .7 1 .7 
18:3 1 . 1  1 .3 1 . 1  0.9 1 .1  1 .0 
18:4 1 .0 0.6 0.9 0.8 0.9 0.9 
20: 1 12.0 15.6 1 1 .7 1 1 .8 12.0 1 1 .9 
20:3 0.4 0.4 0.4 0.3 0.4 0.4 
20:4 0.6 0.3 0.6 0.5 0.6 0.6 
20:5 4.6 2.1 4.4 4.4 4.4 4.6 
22: 1 5.4 6.5 5.3 5.5 5.1 5.2 
22:4 - - - - - -
22:5 0.9 0.5 1 . 1  1 .0 1 . 1  1 .0 
22:6 7.7 1 .2 7.7 7.7 7.7 7.3 
--------------- -- ----???-
Orange Roughy Oil 
Without Vacuum 
Pressure 
Frac.-1 Frac.-2 
1 .5 1 .6 
- -
3.7 4.0 
10.1  10.0 
0.4 0.3 
2.1 2.3 
0.8 0.9 
5 l .3 50.4 
1.4 1.4 
1 .2 1 . 1  
0.4 0.5 
15.8 15.6 
0.4 0.4 
0.4 0.4 
1 .9 2.1 
6.3 6.3 
- -
0.4 0.4 
1 .9 2.1 
With Vacuum 
Pressure 
Frac.- 1 Frac.-2 
1 .5 1 .6 
- -
3 .6 4.0 
9.9 10. 1 
0.3 0.3 
2.1 2.1 
0.8 0.8 
51 .6 51.4 
1 .4 1 .4 
1 . 1  1 . 1  
0.4 0.4 
15.9 15.5 
0.4 0.3 
0.4 0.3 
1 .8 1 .8 
6.5 6.4 
- -
0.4 0.4 
1 .9 2.1 
V) 
? 
329 
Appendix 5.4. Results of chemical, physical and organoleptic analysis of fish oil as the effects of 
various height sizes of resin packed column 
Analysis : Free fatty acid value (% oleic acid) 
Diameter = 2.6 cm 
Diameter : height Fraction-I Fraction-2 
Untreated oil 2.62 2.62 
1 : 5 2.20 2.71 
1 : 10 2.21 3.21 
1 : 15 2.16 3 .33 
1 :  20 2.07 3.73 
Analysis : Refractive Index value (20?C) 
Diameter = 2.6 cm 
Diameter : height Fraction-I Fraction-2 
Untreated oil 1 .4706 1 .4706 
1 :  5 1 .4650 1 .4703 
1 : 10 1 .4645 1 .4709 
1 : 15 1 .4650 1 .4710  
1 :  20 1 .4650 1 .4710  
Analysis : Colour absorbance value 490 run 
Diameter = 2.6 cm 
Diameter : height Fraction-I Fraction-2 
Untreated oil 1 .52 1 .52 
1 : 5 1 .36 1 .92 
1 :  10 1 . 18  2.36 
1 :  15 1 . 18  2.34 
1 :  20 1 . 15  2.53 
330 
Analysis : Sensory evaluation for Odour 
Diameter = 2.6 cm 
Diameter : height Fraction-! Fraction-2 
Untreated oil 6.72 6.76 
1 : 5 4.1 1  5.04 
1 : 10 3 .64 5.32 
1 : 15 3 .82 5.25 
1 :  20 3.36 5.39 
Analysis : Sensory evaluation for Taste 
Diameter = 2.6 cm 
Diameter : height Fraction-! Fraction-2 
Untreated oil 6.06 6.06 
1 : 5 3 .43 4.43 
1 : 10 3.36 4.43 
1 : 15 3.36 4.40 
1 :  20 3.21 4.50 
Analysis : Fatty acid profiles 
Diameter : height 
Fatty Untreated 
Acids Oil 1 : 5 1 10 
Frac.-1 Frac.-2 Frac.-1 Frac.-2 
14:0 5.2 6.3 5.2 5.7 3.5 
15:0 - - - - -
16:0 13.4 14.2 12.8 13.6 1 1.2 
16: 1 8.4 8.9 8.2 8.3 7.0 
17:0 - - - - -
17: 1 2.5 2.6 2.3 2.5 2.2 
18:0 2.4 2.3 2.3 2.3 2.4 
18: 1 34.0 33.3 34.0 32.6 34.9 
18:2 1 .8 1 .8 2.0 1 .8 1 .9 
18:3 0.8 0.8 0.9 0.7 0.8 
18:4 1 .6 1 .5 1 .7 1 .5 1 .7 
20: 1 9.7 9.2 9.5 10.3 10.7 
20:3 - - - - -
20:4 0.8 0.8 0.8 0.8 1 . 1  
20:5 5.5 5.1 6.3 5.4 6.2 
22: 1 4.7 4.3 4.7 4.8 5.6 
22:4 - - - - -
22:5 1 .3 1 .3 1 .5 1 .4 1 .6 
22:6 7.8 7.6 8.2 8.2 9.3 
1 15 
Frac.-1 Frac.-2 
4.1 3.5 
- -
12.2 1 1 .3 
7.6 7.0 
- -
2.4 2.2 
2.6 2.6 
34.7 34.9 
2.1 2.1 
0.7 1 . 1  
12.3 1.8 
10.5 1 1.0 
- -
0.9 0.9 
6. 1 5.7 
5.2 5.5 
- -
1.4 1 .4 
8.2 8.9 
1 
Frac.-1 
4.5 
-
12.4 
7.9 
-
2.4 
2.6 
34.3 
1 .9 
0.9 
1 .6 
10.4 
-
0.9 
5.9 
5.2 
-
1 .4 
7.8 
20 
Frac.-2 
4.5 
-
12.4 
7.8 
-
2.3 
2.5 
34.1 
2.0 
1 .0 
1 .7 
10.3 
-
1 . 1  
5.9 
4.8 
-
1 .4 
8.2 
w 
w 
....... 
. 
332 
Appendix 5.4. Results of chemical, physical and organoleptic analysis of fish oil as the effects of 
various diameter sizes of resin packed column 
Analysis : Free fatty acid value (%  oleic acid) 
Height = 39 cm 
Diameter (Cm) Fraction-1 Fraction-2 
Untreated oil 1 .82 1 .82 
1 .65 1 .53 2.01 
2.60 1 .44 2.13 
3.20 1 .53 2.23 
Analysis : Refractive Index Value (20?C) 
Height = 39 cm 
Diameter (Cm) Fraction-1 Fraction-2 
Untreated oil 1 .4700 1 .4700 
1.65 1 .4670 1 .4700 
2.60 1 .4655 1 .4700 
3.20 1 .4660 1 .4700 
Analysis : Colour absorbance value 490 nm 
Height = 39 cm 
Diameter (Cm) Fraction-1 Fraction-2 
Untreated oil 0.99 0.99 
1 .65 0.91 1 .07 
2.60 0.90 1 . 13  
3.20 0.88 1 . 18  
333 
Analysis : Sensory analysis for Odour 
Height = 39 cm 
Diameter (Cm) Fraction-1 Fraction-2 
Untreated oil 5.43 5.43 
1 .65 3.71 5.64 
2.60 3.57 4.78 
3.20 3 .57 4.71 
Analysis : Sensory analysis for Taste 
Height = 39 cm 
Diameter (Cm) Fraction-1 Fraction-2 
Untreated oil 5.36 5.36 
1 .65 3.71 6.28 
2.60 3.50 5.50 
3.20 3 .78 6.00 
Appendix 6.1 .  Score sheet used for sensory evaluation during flsh oil storage 
experiment 
SENSORY EVALUATION FOR FISH Oll.. STORAGE 
Name : ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Date : ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Instruction: Please indicate the score that best reflects your attitude about the product. 
Description Scale 
B?d 0 
Suspicion of off-odour/taste 1 
Noticeable but very slight off-odour/taste 2 
More noticeable off-odour/taste 3 
Distinct off-odour/taste 4 
Disagreeable off-odour/taste, rancid 5 
Markedly disagreeable off-odour/taste, very rancid 6 
Sample Code Odour Taste 
334 
335 
Appendix 6.3.  Linear relationship between the natural logarithm of rate constant and the reciprocal 
of absolute temperature for each parameter 
Fisb Oil Storage 1/f('K) Ink linear equatioo 
Parameter Sample Temp?'C) (x l?) 
2 3.64 .{).06 
Refined 20 3.41 0.76 In 1: = 9.80 - 2690.7 1/f ; 
Peroxide oil 30 3.30 0.93 r' = 96.38% 
Value 40 3.19 1.14 
2 3.64 .{).17 
Unrefined 20 3.41 0.44 In k =  6.22 . 1737 1/f ; 
oil 30 3.30 0.50 r' = 90.88% 
40 3.19 0.60 
2 3.64 -4.60 
Coloor Refined 20 3.41 -3.30 In 1: = 13.35 . 4918 1/f ; 
Absorbance oil 30 3.30 -2.92 r' = 99.22% 
Value 40 3.19 -2.37 
(zero 2 3.64 -4.60 
order Unrefmed 20 3.41 -3.91 In 1: = 7.94 ? 3462 1/f ; 
reaction) oil 30 3.30 -3.58 r' = 98.14% 
40 3.19 -3.00 
2 3.64 -5.30 
Coloor Refined 20 3.41 -3.82 In k =  14.83 ? 5514 1/f ; 
Absorbance oil 30 3.30 -3.44 r' = 99.01% 
Value 40 3.19 -2.78 
(first 2 3.64 -5.81 
order Unrefmed 20 3.41 -4.60 In k =  1 1.34 ? 4705 1/f ; 
reaction) oil 30 3.30 -4.7:1 r' = 99.29% 
40 3.19 -3.65 
2 3.64 -2.03 
Odour score Refined 20 3.41 -1 .53 In k = 4.62 ? 1821 1/f ; 
from: oil 30 3.30 -1 .38 r' = 98.73% 
40 3.19 -1.22 
cold 2 3.64 -2.35 
sample Unrefmed 20 3.41 -1.99 In k =  5.1 5 .  2077 1/f ; 
oil 30 3.30 -1.78 r' = 95.61'To 
40 3.19 -1.38 
2 3.64 -2.07 
Odour Refined 20 3.41 -1 .65 In k =  4.62 . 1841 1/f ; 
score oil 30 3.30 -1.47 r?"'" .. 
from: 40 3.19 -1.23 
warm 2 3.64 -2.54 
sample Unrefmed 20 3.41 -2.17 In 1: = 5.67 ? 2373 1/f ; 
oil 30 3.30 -1.90 r' = 95.61% 
40 3.19 -1 .48 
2 3.64 -2.19 
Taste Refined 20 3.41 -1 .76 In k = 4.839 ? 1 .932 1/f ; 
score oil 30 3.30 -1.54 r' = 99.95'To 
from: 40 3.19 -1.32 
cold 2 3.64 -2.32 
sample Unrefmed 20 3.41 ?2.13 In k =  4.1 2 .  1793 1/f ; 
oil 30 3.30 -1 .83 r' = 90.38% 
40 3.19 -1.49 
2 3.64 -2.24 
Taste Refined 20 3.41 ?2.06 In k =  3.75 . 1666 1/f ; 
score oil 30 3.30 -1.77 r' = 91 .50'To 
from: 40 3.19 -1 .48 
warm 2 3.64 -2.30 
sample Unrefmed 20 3.41 -1 .85 In k =  6.1 2 .  2325 1/f ; 
oil 30 3.30 -1 .62 r' = 97.77% 
40 3.19 ?1.22 
Appendix 6.4. Linear relationship between the natural logarithm of shelf life (9) versus the 
reciprocal of absolute temperature eK) 
Fish Oil Storage l!f ("I() In 9 Linear Equation 
Sample Temp ("C) (x 1Q?3) 
2 3.64 3.401 
Refined 20 . 3.41 2.996 In 9 = -3 .36 + 
oil 30 3.30 2.773 1869.01 1!f ; 
40 3.19 2.565 r = 99.91% 
2 3.64 3.714 
Unrefined 20 3.41 3 .258 In 9 = - 3.93 + 
oil 30 3.30 2.996 2102.12 1!f ; 
40 3.19 2.773 r = 99.88% 
336 
337 
Appendix 6.5. Results of chemical, physical and organoleptic analysis of refmed and unrefmed fish 
oils during storage at various temperature 
Analysis: Peroxide value (meqlkg) 
Storage Storage time (weeks) 
Sample Temp. 
0 2 5 10 15  20 CO C) 
2 3.22 4.62 7.71 15.21 17.24 22.51 
Refined 20 3.22 10.40 14.68 21 .52 24.97 35.42 
Oil 30 3 .22 1 1 .77 15.65 25.39 3 1 .02 40.13  
40 3.22 12.64 19.07 29.04 33.91 
2 4.28 5.94 8.80 13.52 15.59 20.53 
Unrefined 20 4.28 10.04 12.40 16.43 19.95 24.89 
Oil 30 4.28 10.36 12.80 18.47 20.36 24.66 
40 4.28 10.79 13.04 17.43 18.88 
Analysis: Refractive Index Value (25?C) 
Storage Storage Time (weeks) 
Sample Temp. 
0 2 5 10 15  20 CO C) 
2 1 .4670 1 .4670 1 .4670 1 .4672 1 .4672 1 .4681 
Refined 20 1 .4670 1 .4671 1 .4671 1 .4679 1 .4680 1 .4690 
Oil 30 1 .4670 1 .4670 1 .4673 1 .4681 1 .4682 1 .4693 
40 1 .4670 1 .4675 1 .4679 1 .4688 1 .4692 
2 1 .4720 1 .4720 1 .4720 1 .4720 1 .4721 1 .4722 
Unrefined 20 1 .4720 1 .4721 1 .4721 1.4721 1 .4726 1 .4729 
Oil 30 1 .4720 1 .4722 1 .4722 1 .4722 1 .4729 1 .4731 
40 1 .4720 1 .4722 1 .4725 1 .4728 1 .4731 
338 
Analysis: Colour absorbance value 490 nm 
Storage Storage time (weeks) 
Sample Temp. 
0 2 5 . 10 15  20 ec) 
2 1 .86 1.84 1 .80 1 .76 1 .71 1 .68 
Refined 20 1 .86 1 .75 1 .69 1 .52 1 .41 1 . 17 
Oil 30 1 .86 1 .72 1 .60 1 .36 1 .10 0.83 
40 1 .86 1 .65 1 .39 0.99 0.65 
2 2.04 2.04 1 .99 1 .96 1 .94 1 .97 
Unrefined 20 2.04 1 .96 1 .95 1 .90 1 .83 1 .73 
Oil 30 2.04 1 .96 1 .90 1 .83 1 .67 1 .54 
40 2.04 1 .92 1 .80 1 .63 1 .31  
Analysis: Odour score ? Cold sample 
Storage Storage time (weeks) 
Sample Temp. 
0 2 10 15  20 ec) 5 
2 0.93 1 .00 2.14 2.40 2.87 2.92 
Refined 20 0.93 0.93 2.57 3 .30 3.42 3.50 
- Oil 30 0.93 1 .36 2.54 4.07 4.42 4.42 
40 0.93 1 .21 2.79 4.90 5.04 
2 1 .29 1 .17 2.06 2.10 2.50 3.21 
Unrefined 20 1 .29 1 . 17 2.27 2.90 3.47 3.62 
Oil 30 1 .29 1 .08 2.10 3.65 4.30 4.21 
40 1 .29 1 .33 2.50 4.90 5.20 
339 
Analysis: Odour score ? Warm sample 
Storage Storage time (weeks) 
Sample Temp. 
0 2 5 . 10 15 20 CO C) 
2 1 .04 1 .00 2.14 2.10 2.70 2.33 
Refined 20 1 .04 1 .50 2.50 2.73 3 .20 3.33 
Oil 30 1 .04 1.64 2.29 3 .37 4.15 4.08 
40 1 .04 1.50 2.82 4.30 4.80 
2 1 .57 1 .33 2.08 1 .90 2.44 3.17 
Unrefined 20 1 .57 1 .33 2.24 2.70 3.12 3.54 
Oil 30 1 .57 1 .33 2.36 3 .43 3 .87 4.29 
40 1 .57 1 .25 2.81 4.35 4.94 
Analysis: Taste score ? Cold sample 
Storage Storage time (weeks) 
Sample Temp. 
0 2 10 20 eC) 5 15 
2 0.93 0.86 1 .86 2.20 2.54 3.00 
Refined 20 0.93 1 .00 2.57 3.17 3 .02 3.50 
Oil 30 0.93 1 .29 2.36 3.40 3.90 4.17 
40 0.93 1 .21 2.50 4.47 4.83 
2 1 .14 0.83 1 .90 2.08 1 .96 3.00 
Unrefined 20 1 .14 1 .08 2.13 2.55 2.79 3 .50 
Oil 30 1 .14 1 .25 2.09 3 .47 3 .75 4.17 
40 1 .14 1 .25 2.44 4.55 4.79 
340 
Analysis: Taste score - Warm sample 
Storage Storage time (weeks) 
Sample Temp. 
0 2 5 . 10 15 20 ec) 
2 0.86 0.79 1.71 2.10 2.40 2.75 
Refined 20 0.86 0.93 1 .79 2.63 2.70 3.17 
Oil 30 0.86 1 . 1 1  1 .89 3.10 3.30 3.50 
40 0.86 1 .00 2.39 3 .93 4.20 
2 0.71 0.87 1.74 1 .82 2.42 2.58 
Unrefined 20 0.71 0.96 1 .90 2.40 2.71 3.17 
Oil 30 0.71 1 .08 1 .96 2.98 3.42 3.71 
40 0.71 1 .33 2.19 3 .88 4.58 
341 
Appendix 7.1. Relationship between degree of vacuum and resi
dual oxygen content (CIG Ltd, 
1989) 
'2 
<1) 
Cl z >->< w 0 
" ? 
? 
0 .... 
? 
0 0 
-l c: 
< ? 
::J 
Cl en 0 
en c. w E a: 0 () 
1- ? z -w c: 
{.) c: a: w 0 
a. '0 <1) 
en 
ea 
CD -
1 5  
1 4  
1 3  
1 2 
1 l 
1 0  
9 -
8 I i 
I 
1 7 - - - - - - - - - -
6 
5 
4 
3 
2 - - - - -
29.92 28 26 
l 
I 
I 
I 
I 
I 
I 
! 
! 
l 
24 2 2  20 1 8  ?16 
l 
l 
l 
14 1 2  
VACUUM - INCHES MERCURY 
1 
I 
I 
I 
! 
l 
I 
I 
l 
I 
[ 
I 
) 
[ 
I I 
I 
lO 8 
342 
Appendix 7 .2. Permitted antioxidants to be used in Indonesian foods and drinks according to Health 
Ministry Regulation No.10178/A/SK!74 
Foods Antioxidants Maximum 
allowable level 
(pp m) 
a) propylgallate 100 
octylgallate 100 
dodecylgallate 100 
Fat, oil and foods mixture three of them 100 
containing vitamin A b) butylated hydroxyanisol (BHA) 200 
(except foods contain c) mixture a) and b) 100 a) 
vitamin A more than 200 b) 
100.000ill/gram) d) dibutyl hydroxitoluene (BHT) 200 
e) mixture of BHA and BHT 200 
f) nordihydroguaiaretic acid NDGA) 100 
g) a-tocopherol 100 
a) propylgallate 100 
octylgallate 100 
dodecylgallate 100 
mixture three of them 100 
b) BHA 200 
Margarine c) mixture a) and b) 100 a) 
200 b) 
d) BHT 200 
e) mixture of BHA and BHT 200 
f) NDGA 100 
g)  a-tocopherol 300 
a) propylgallate 80 
octylgallate 80 
dodecy I gallate 80 
Butter for food production mixture three of them 80 
b) BHA 160 
c) BHT 160 
d) mixture BHA and b) 160 
a) propylgallate 1000 
octylgallate 1000 
dodecylgallate 1000 
mixture three of them 1000 
Food flavours b) BHA 1000 
c) mixture a) and b) 1000 
d) BHT 1000 
e) mixture BHA and BHT 1000 
Materials containing vitamin a) BHA 10 
A more than b) BHT 10 
100.000ill/gram c) mixture BHA and BHT 10 
Appendix 7.3. Results of chemical and physical analysis of fish oil as the effects of various 
antioxidant addition during storage at 63?2?C 
Analysis: Peroxide Value (meqlkg sample) 
Antioxidants Storage time (days) 
added into 
refined fish 0 1 2 4 8 12 16  
oil 
Control 12.47 17.65 18.52 23.10 29.64 63.57 104.84 
0.02% BHA 1 1 .76 18.19 19.74 25.02 28.85 36.98 51 . 16  
0.02% TBHQ 9.76 14.49 14.47 14.81 22.42 3 1 .14 45.51 
0.1%  Grindox 9.60 14.99 15.97 16.60 23.09 32.61 82.83 
0.1%  Tocopherol 15 .28 20.76 26. 1 1  32.22 46. 13  57.30 66.55 
(heated prep.) 
0.1%  Tocopherol 20.20 21 .62 23.78 32.39 46.61 52.18  66.23 
(direct 
addition) 
Analysis: TBA value (f.Jmoles/g sample) 
Antioxidants Storage time (days) 
added into 
refined fish 0 1 2 4 8 1 2  16  
oil 
Control 13 .56 21 .14 21 .47 29.84 46.01 120.83 176.86 
0.02% BHA 16.46 18.75 18 .06 29.24 46.07 64.10  86.96 
0.02% TBHQ 13 . 1 1  15.57 16.06 23.18 34.89 48.85 82.69 
0.1%  Grindox 13 .25 21 .25 24.04 29.14 40.44 56.63 1 54.05 
0.1%  Tocopherol 15 .64 13 .56 17.64 29.17 42.32 62.64 106.70 
(heated prep.) 
0.1%  Tocopherol 13 .67 14.03 21 .07 29.43 43.18 64.1 1 106.75 
(direct 
addition) 
343 
344 
Analysis: Anisidine Value 
Antioxidants Storage time (days) 
added into 
refined fish 0 1 2 4 . 8 12 16 
oil 
Control 4.91 10.04 13.46 21 .67 47.76 1 15.33 181 .80 
0.02% BHA 5.43 10.03 14.24 22.06 43.98 70.89 108.51 
0.02% TBHQ 5.89 6.94 9.01 12.68 25.90 47.05 81 .73 
0.1%  Grindox 4.56 7.73 1 1 .40 19.53 36.14 61 .96 141.90 
0.1% Tocopherol 4.38 8.67 13 .98 22.39 49.19 82.85 127.25 
(heated prep.) 
0.1%  Tocopherol 4.31 10.37 14.68 24.20 50.59 91 .92 122.90 
(direct 
addition) 
Analysis: Totox Value (2 Peroxide Value + Anisidine Value) 
Antioxidants Storage time (days) 
added into 
refined fish 0 1 2 4 8 12 16 
oil 
Control 29.87 45.34 50.50 67.88 107.04 242.47 391.48 
0.02% BHA 28.96 46.42 53.71 72. 1 1  101 .68 144.85 210.48 
0.02% TBHQ 25.40 35.92 37.95 42.31 70.74 109.32 172.70 
0.1% Grindox 24.34 37.70 43.34 52.73 82.33 127.18  307.56 
0.1% Tocopherol 34.95 50.19  66.20 86.83 141 .46 197.46 260.36 
(heated prep.) 
0.1% Tocopherol 44.70 53.61 62.23 88.99 143.80 196.83 255.36 
(direct 
addition) 
345 
Analysis: Refractive Index Value (25?C) 
Antioxidants Storage time (days) 
added into 
refined fish 0 1 2 4 8 12 16 
oil 
Control 1 .4710 1 .4729 1 .4734 1 .4745 1 .4749 1 .4751 1 .4752 
0.02% BHA 1 .4720 1 .4730 1 .4740 1 .4740 1 .4744 1 .4746 1 .4745 
0.02% TBHQ 1 .4720 1 .4730 1 .4735 1 .4740 1 .4745 1 .4745 1 .4745 
0.1%  Grindox 1 .4720 1 .4729 1 .4735 1 .4740 1 .4745 1 .4746 1 .4750 
0.1%  Tocopher. 1 .4721 1 .4730 1 .4740 1 .4742 1 .4747 1 .4750 1 .4750 
(heated prep) 
0.1% Tocopher. 1 .4721 1 .4732 1 .4737 1 .4746 1 .4748 1 .4749 1 .4750 
(direct 
addition) 
Analysis: Colour absorbance value 490 nm 
Antioxidants Storage time (days) 
added into 
refined fish 0 1 2 4 8 12 16 
oil 
Control 2.20 2.15 2.05 1 .78 1 . 15  0.35 0.10  
0.02% BHA 2.25 2.22 2.02 1 .82 1 .30 0.80 0.31 
0.02% TBHQ 2.25 2.19 2.13  2.05 1 .8 1  1 .39 0.63 
0.1%  Grindox 2.24 2.22 2.06 1 .87 1 .51  0.96 0.14 
0.1%  Tocopherol 2.24 2.24 2.12 2.00 1 .69 1 .23 0.43 
(heated prep.) 
0.1% Tocopherol 2.24 2.16 2.08 1 .97 1 .63 0.98 0.49 
(direct 
addition) 
346 
Appendix 7.4. Results of chemical and physical analysis of fish oil as the effects of various levels 
of BHA addition during storage at 63?2?C 
Analysis: Peroxide Value (meqlkg) . 
BHA dosage Storage time (days) 
level added 
refined oil 0 2 5 9 1 3  17 
Unrefined oil 1 1 .40 13 .36 17.50 22.77 42.66 74.54 
(control) 
0.005% 10.57 15.61 24.03 3 1 .33  50.51 80.10  
0.010% 10.59 15.64 24.67 28.30 45.34 62.36 
0.015% 10.27 16.94 24.09 26.26 40.37 54.61 
0.020% 9.91 16.19 24.26 25.10 39.82 53.91  
Analysis: TBA Value (Jlmoles/g samples) 
BHA dosage Storage time (days) 
level added 
refined oil 0 2 5 9 13  17 
Unrefined oil 17.44 13 .73 27.21 49.97 90.23 162.61 
(control) 
0.005% 19.14 23.15 34.45 73.56 103.25 162.27 
0.010% 18.07 23.78 33.49 57.08 86.78 1 30.05 
0.015% 15.49 22.53 33.75 58.35 79.27 120.14 
0.020% 15.07 20.94 3 1 .54 52.34 76.95 107.95 
347 
Analysis: Anisidine Value 
BHA dosage Storage time (days) 
level added 
refined oil 0 2 5 9 1 3  17 
Unrefined oil 1 1 .36 18.09 3 1 .22 72.67 103 .44 466.48 
(control) 
0.005% 9.54 17.50 35.29 82.28 1 14.27 457.64 
0.010% 9.22 17.35 35.75 73.92 100.83 183.20 
0.015% 8.55 18.32 34.47 73.47 96.27 159.32 
0.020% 9.39 17.82 34.57 70.34 91 .95 142.51 
Analysis: Totox Value 
BHA dosage Storage time (days) 
level added 
refined oil 0 2 5 9 1 3  17 
Unrefined oil 34.15  44.82 66.22 1 18.21 188.77 615.57 
(control) 
0.005% 30.67 48.73 83.36 144.95 215.30 617.84 
0.010% 30.39 48.62 85.08 130.53 191 .50 307.92 
0.015% 29.10 52.21 82.65 125.98 177.01 268.55 
0.020% 29.20 50.21 83.09 120.55 171 .59 250.33 
Analysis: Refractive Index Value (25?C) 
BHA dosage Storage time (days) 
level added 
refined oil 0 2 5 9 1 3  1 7  
Unrefined oil 1 .4750 1 .4751 1 .4750 1 .4754 1 .4755 1 .4769 
(control) 
0.005% 1 .4720 1 .4742 1 .4750 1 .4753 1 .4753 1 .4760 
0.010% 1 .4720 1 .4741 1 .4750 1 .4751 1 .4752 1 .4760 
0.015% 1 .4720 1 .4742 1 .4750 1 .4750 . 1 .4751 1 .4760 
0.020% 1 .4720 1 .4741 1 .4750 1 .4750 1 .4750 1 .4759 
348 
Analysis: Colour absorbance value 490 nm 
BHA dosage Storage time (days) 
level added 
refined oil 0 2 5 9 1 3  1 7  
Unrefined oil 0.95 0.88 0.74 0.47 0.33 0.31 
(control) 
0.005% 0.70 0.64 0.50 0.27 0.16 0.12 
0.010% 0.71 0.64 0.51 0.30 0.20 0.12 
0.015% 0.71 0.65 0.52 0.30 0.20 0.13 
0.020% 0.72 0.65 0.53 0.32 0.22 0.13 
Appendix 7.5. Results of chemical and physical analysis of fish oil during storage in vacuum package at 63?2?C and 30?_2?C 
Analysis: Peroxide Value (meq/kg) 
Storage Fish Oil Packaging Storage time (days) 
Temperature Condition 
0 1 2 4 8 12 16 
non-vacuum 5.02 32.59 33. 13  26.43 16.28 12.58 9.88 
Refined 
vacuum 5.02 15.52 1 1 .41 10.06 6.95 6.33 4.13  
63  ? 2?C 
non-vacuum 7.18 23.76 28.45 22.51 12.68 8.33 6.22 
Unrefined 
vacuum 7.18 13.45 1 1 .44 9.69 5.70 4.5 1 3.92 
Storage time (weeks) 
0 1 2 4 8 12 16 
non-vacuum 6.17 32.58 66.24 48.75 40.77 38.98 38.62 
Refined 
vacuum 6.17 20.91 23.82 18.61 18.44 18.31 17.93 
30 ? 2?C non-vacuum 6.79 44.80 60.22 43.23 31 .98 29.96 27.80 
Unrefined 
vacuum 6.79 23.04 22.66 16.93 13.86 15.3 1 13.84 
w 
? 
Analysis: TBA Value (fJmoles/g) 
Storage Fish Oil Packaging Storage time (days) 
Temperature Condition 
0 1 2 
non-vacuum 9.83 62.34 59.79 
Refined 
vacuum 9.83 29. 16 19.90 
63 ? 2?C 
non-vacuum 10.29 37.90 39.82 
Unrefined 
vacuum 10.29 17.56 13.21 
Storage time (weeks) 
0 1 2 
non-vacuum 8.85 58.66 1 15.49 
Refined 
vacuum 8.85 32.09 33.8 1 
30 ? 2?C non-vacuum 10.20 65.32 79.07 
Unrefined 
vacuum 10.20 33.59 30.90 
4 8 
42.82 29. 1 1  
13.70 10.74 
23.39 14.32 
7.73 5.55 
4 8 
73.48 53.66 
24.85 19.43 
53.59 33.03 
16.82 13.51 
12 
26.52 
10.88 
13.73 
5.66 
12 
50.98 
16.83 
26.26 
10.56 
16 
21.76 
9.59 
12.27 
5.55 
16 
42.54 
17.26 
26.45 
9.71 
w 
Lll 
0 
Analysis: Anisidine Value 
Storage Fish Oil Packaging Storage time (days) 
Temperature Condition 
0 1 2 
non-vacuum 9.32 37.35 51 .29 
Refined 
vacuum 9.32 21.17 23.69 
63 ? 2?C 
non-vacuum 1 1 . 12  29.93 45.94 
Unrefmed 
vacuum 1 1 . 12  20.04 22.28 
Storage time (weeks) 
0 1 2 
non-vacuum 9.13 1 8.96 46.37 
Refined 
vacuum 9.13 15.74 18.87 
30 ? 2?C non-vacuum 10.96 21 .21 34.83 
Unrefmed 
vacuum 10.96 15.37 18.38 
4 8 
58.66 54.52 
23.75 23.20 
53. 1 1  50.72 
22.69 22.29 
4 8 
43.22 42.32 
20.03 19.62 
37.95 36.92 
18.81 18.83 
12 
56.43 
24.19 
52.76 
24.30 
12  
43.72 
19.37 
37.34 
19.58 
16 
54.28 
25.02 
51 .94 
24.50 
16 
42.23 
19.96 
40.97 
19.63 
w 
VI 
,_.. 
Analysis: Totox Value 
Storage Fish Oil Packaging Storage time (days) 
Temperature Condition 
0 1 2 4 8 12 16 
non-vacuum 19.35 102.54 1 17.54 1 1 1 .53 87.09 81.61 74.04 
Refined 
vacuum 19.35 52.20 46.51 43.72 37.09 36.84 33.28 
63 ? 2?C 
non-vacuum 25.47 77.44 102.84 98. 13  76.09 69.41 64.38 
Unrefined 
vacuum 25.47 46.93 45.16 42.07 33.69 33.32 32.34 
Storage time (weeks) I 
0 1 2 4 8 12 16 
non-vacuum 21.47 84. 14 178.86 140.71 123.86 121 .69 1 19.46 
Refined 
vacuum 21.47 57.55 66.52 57.25 56.51 55.99 55.81 
30 ? 2?C non-vacuum 24.53 1 10.80 155.27 124.41 100.88 97.26 96.56 
Unrefmed 
vacuum 24.53 61 .44 63.70 52.67 46.54 50.19 47.31 
IJ) 
? 
Analysis: Refractive Index Value 
Storage Fish Oil Packaging Storage time (days) 
Temperature Condition 
0 1 2 
non-vacuum 1 .4700 1 .471 1  1 .4715 
Refmed 
vacuum 1 .4700 1 .4711  1 .471 1  
63  ? 2?C 
non-vacuum 1.4725 1 .4730 1 .4731 
Unrefined 
vacuum 1 .4725 1.4730 1 .4730 
Storage time (weeks) 
0 1 2 
non-vacuum 1 .4690 1 .4703 1 .4730 
Refmed 
vacuum 1 .4690 1.4702 1 .4730 
30 ? 2?C non-vacuum 1 .4729 1 .4730 1 .4750 
Unrefmed 
vacuum 1.4729 1 .4730 1 .4750 
4 8 
1 .47 17 1 .4720 
1 .47 12 1 .4712 
1 .4735 1 .4735 
1 .4730 1.4730 
4 8 
1 .4740 1 .4730 
1 .4740 1 .4730 
1 .4755 1 .4750 
1 .4751 1 .4747 
12 
1 .4720 
1 .471 1  
1 .4735 
1 .4730 
12  
1 .4730 
1 .4731 
1 .4754 
1.4754 
16 
1 .4720 
1 .47 1 1  
1 .4735 
1 .473 1 
16 
1 .4726 
1 .4726 
1 .4750 
1 .4750 
w 
Vl 
w 
Analysis: Colour absorbance value 490 nm 
Storage Fish Oil Packaging Storage time (days) 
Temperature Condition 
0 1 2 4 8 12  16 
non-vacuum 0.79 0.52 0.43 0.37 0.36 0.36 0.37 
Refined 
vacuum 0.78 0.68 0.68 0.65 0.63 0.63 0.60 
63 ? 2?C 
non-vacuum 0.99 0.75 0.58 0.53 0.53 0.54 0.54 
Unrefined 
vacuum 0.99 0.9 1 0.85 0.84 0.83 0.8 1 0.79 
Storage time (weeks) 
0 1 2 4 8 12  16 
non-vacuum 0.77 0.61 0.29 0.30 0.28 0.28 0.34 
Refined 
vacuum 0.77 0.68 0.58 0.68 0.58 0.62 0.62 
30 ? 2?C non-vacuum 1 .04 0.81 0.50 0.51 0.55 0.58 0.64 
Unrefmed 
vacuum 1 .04 0.95 0.78 0.81 0.83 0.84 0.88 
w 
? 
355 
Analysis: Odour score for fish oil stored at 30 ? 2?C 
Fish oil Packaging Storage time (weeks) 
Condition 
0 2 4 8 12 16  
non-vacuum 0.83 2.93 3.67 3.62 4.50 4.37 
Refined 
vacuum 0.83 2.43 2.25 2.00 3.00 3.50 
non-vacuum 0.17 2.43 3.33 3.75 4.00 4.62 
Unrefined 
vacuum 0.17 1 .78 2.08 2.25 3.00 3.37 
356 
Appendix 8. 1 .  Questionnaire used for fish meal factory survey 
FISH MEAL FACTORY SURVEY 
PLEASE CONSIDER THE FOLLOWING IDEA 
I am going to introduce a new method for fish oil refining in order to produce fish oil for human 
consumption. In this method, the fish oil is passed through a resin packed column. The Refining 
unit is simple, easy to be designed and low in labour costs. 
1 .  Do you think that the above idea is interesting? 
( ) Yes 
( ) No 
2. Would you like to adopt the above refming method, if I can fmd the technology? 
( ) Yes 
( ) No 
If YES, please continue to question 3. 
If NO, you can stop answering this questionnaire, but please give your 
reasons, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........................... . . .................. .................. .................. ..... . . . . . . . . . .. . .  . 
3. How would you apply this refining method in your factory? 
( ) replace the existing refining unit totally 
( ) operate it together with the existing refming unit 
4. What are the types of fish oils which will be refined using this method? 
( ) all fish oil types produced 
( ) certain fish oil type, please mentioned ... . . . . . . . . . . . . . . . . . . . . . . . . . . 
==================================== 
Additional Information: 
Factory 
Location 
357 
Appendix 8.2. Results of chemical, physical and sensory analysis of Indonesian fish oil as the 
effects of resin refining process 
Analysis: Free fatty acid value (% oleic acid) 
Fish Oil 
Sample 
Fish meal oil Canning waste oil 
Untreated oil 0.20 0.27 
Fraction-1 0.17 0.19 
Fraction-2 0.31 0.33 
Analysis: Refractive Index Value 
Fish Oil 
Sample 
Fish meal oil Canning waste oil 
Untreated oil 1 .4791 1 .4791 
Fraction-1 1 .4720 1 .4750 
Fraction-2 1 .4791 1 .4780 
Analysis: Colour absorbance value 490 run 
Fish Oil 
Sample 
Fish meal oil Canning waste oil 
Untreated oil 1 .51 0.33 
Fraction-I 1 .01 0.21 
Fraction-2 2.31 0.78 
358 
Analysis: Odour score 
Fish Oil 
Sample 
Fish meal oil Canning waste oil 
Untreated oil 7.7 3 .9 
Fraction-1 4.3 1 .8 
Fraction-2 7.3 4.9 
Analysis : Fatty acid profiles 
Fish meal oil Canning waste oil 
Fatty acids 
Untreated Frac.-1 Frac.-2 Untreated Frac.-1  Frac.-2 
Oil Oil 
14:0 13.3 13.8 12.6 12.7 12.6 12.0 
15:0 0.6 0.7 0.7 0.8 0.8 0.9 
16:0 20.0 20.3 24.9 21 .2 21 .0 21.4 
16: 1 15.3 15.8 16.1 14.9 15.0 14.8 
17:0 0.6 0.6 0.8 0.8 0.9 1 .0 
17:1 4.2 4.4 4.2 4.0 4.2 4.0 
18:0 3.1 3.0 2.8 3 .5 3.4 4.0 
18:1 7.2 7.0 7.1 8.0 8.3 8.3 
18:2 1 .1  1 . 1  1 .1 1 .2 1 .3 1 .3 
18:3 1 .0 1 .0 1 .1  1 .2 1 .2 1 .2 
18:4 1 .8 2.0 1 .9 2.1 2.2 2.2 
20:1 0.9 0.9 0.8 0.8 0.9 0.9 
20:3 0.1 0.1 0.1 0.1 0.1 0.1 
20:4 2.6 2.6 2.4 2.4 2.4 2.3 
20:5 19.0 17.8 15.5 17.2 16.9 16.5 
22:1 1.8 1 .5 1 .2 1 .2 1 .2 1 .2 
22:4 0.1 0.1 0.1 0.1 0.1 0.1 
22:5 1 .0 1 .2 1 .0 1 .2 1 .2 1 .2 
22:6 5.7 5.5 4.8 6.1 6.3 6.0 
359 
Appendix 8.3. Results of chemical and physical analysis of Indonesian fish oil during storage at 
63 ? 2?C 
Analysis: Peroxide Value (meqlkg) 
Storage Time (days) 
Fish Oil Treatment 
0 2 4 7 1 1  
Unrefined 21 .45 19.80 21 .68 18.35 33.65 
Fish meal 
oil Refined 24.96 23.46 30.65 34.02 87.63 
Unrefined 28.41 25.93 32.15 37.47 71 .73 
Canning 
waste oil Refined 31 .61 34.53 51 .94 92.24 1 18.69 
Analysis: TBA Value (J.lmoles/g) 
Storage Time (days) 
Fish Oil Treatment 
0 2 4 7 1 1  
Unrefined 27.62 25.13 27.15 38.62 73.87 
Fish meal 
oil Refined 35.88 45.43 50.45 72.75 182.49 
Unrefined 53.10 46.07 50.25 88.56 163.95 
Canning 
waste oil Refined 58.33 64.74 84.60 189.35 274.22 
Analysis: Anisidine Value 
Storage Time (days) 
Fish Oil Treatment 
0 2 4 7 1 1  
Unrefined 16.08 21 .68 29.63 45. 1 1  71 .98 
Fish meal 
oil Refined 16.68 27.97 42.75 67.85 468.37 
Unrefined 20.42 31 .31  42.54 82.34 479.72 
Canning 
waste oil Refined 18.93 32. 11  54.45 160.20 471 .74 
360 
Analysis: Totox Value 
Storage Time (days) ? 
Fish Oil Treatment 
0 2 4 7 1 1  
Unrefined 58.98 61 .29 73.00 81.81  139.28 
Fish meal 
oil Refined 66.61 74.58 104.05 135.89 643.63 
Unrefined 77.23 83.17 106.84 157.28 623.18 
Canning 
waste oil Refined 82.14 101.18 158.32 344.68 709.12 
Analysis: Refractive Index Value 
Storage Time (days) 
Fish Oil Treatment 
0 2 4 7 1 1  
Unrefined 1.4770 1 .4770 1 .4770 1 .4770 1 .4770 
Fish meal 
oil Refined 1 .4731 1 .4758 1 .4763 1 .4766 1 .4773 
Unrefined 1 .4770 1 .4770 1 .4770 1 .4770 1 .4780 
Canning 
waste oil Refined 1 .4760 1.4760 1 .4765 1 .4773 1 .4784 
Analysis: Colour absorbance value 490 nm 
Storage Time (days) 
Fish Oil Treatment 
0 2 4 7 1 1  
Unrefined 1 .30 1 .31 1 .22 1 .02 0.73 
Fish meal 
oil Refined 1.14 0.99 0.83 0.60 0.21 
Unrefined 0.34 0.55 0.69 0.60 0.46 
Canning 
waste oil Refined 0.19 0.14 0.1 1 0.06 0.08 
Appendix 9.1 .  Questionnaire used for supermarket survey 
Supermarket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 
City . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
No. Trade Fish Ingredients/ 
Mark Species medium 
Weight 
(g) 
361 
Can Can Price Product 
Form Size (Rp.) Origin/ 
Producer 
& 
Super-
market 
362 
Appendix 9.2. Questionnaire used for cannery survey 
CANNERY SURVEY 
Direction: Please answer this questionnaire by giving thick mark (V) 
1 .  What fish species are processed into canned fish in your company? Please give the quantity of 
canned fish processed from each fish species per year, if possible. 
( ) ... . . ton sardine 
( ) . .  .. . ton mackerel 
( )  . . . . . ton tuna 
( ) . . . . . ton skip jack 
( ) . . . . . ton small tuna 
( ) . . . . . others, please specify ... . . . . . . . . . . . . . . 
2. What is the medium normally added canned fish in your company? Please give the amount 
of canned fish processed from each fish species per year, if possible. 
( ) .. .. . ton tomato sauce 
( ) . . . . . ton brine 
( ) ... . . ton vegetable oil 
( ) . . .. .  ton vegetable oil and brine mixture 
( ) . . .. . ton others, please specify .... . . . . . . . . . . . . . . 
3. Does your company export the canned fish product? 
( ) YES ( ) NO 
If "YES", please specify the percentage of your product to be exported .. . . . % 
4. For local market, what fish species and medium type are processed the most? ( Please give 
percentage for all products marketed locally) 
Fish Species 
( ) . . . . . % Sardine 
( ) . . . . .  % Mackerel 
( )  . . . . . % Tuna 
( ) . . . . . % Skipjack 
( ) . . . . . % Small tuna 
( ) . .. . . % Others, please specify .... . . 
Medium Type 
( ) ..... Tomato sauce 
( )  ... . . Vegetable oil 
( )  ... . . Brine 
( ) . .. .. Vegetable oil and brine 
mixture 
( ) . . .. . . Others, please specify 
363 
Please consider this product idea: 
I am going to introduce canned fish enriched with disguised fish oil. This idea is aimed to 
optimize the fish oil utilisation and to produce canned fish which is nutritionally better than existing 
products in the market. 
5. What is your comment about this idea. If the product is going to be marketed in Indonesia? 
( ) interesting ( ) not interesting 
6. What is the medium you prefer to use, if you produce this canned product? 
( ) tomato sauce 
( ) brine 
( ) vegetable oil 
( ) vegetable oil and brine mixture 
( ) others, please specify .... . . . . . . . . . . . . . . . . . 
7. Would you like to be informed, if a processing technology for this product is developed? 
( ) YES ( ) NO 
If "NO", please give the reason ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 
If "YES", please give the percentage of the product which is going to be produced, in terms of 
total production: 
( )  1 - 10% 
( )  1 1 - 20% 
( )  21 - 30% 
( )  3 1 - 40% 
( ) Over 40% 
Comments: ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
==================================== 
Please complete this section. Your information is kept confidentially. 
Name of factory . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Location . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Production capacity: ..... . . . . . . . . . . ton/day 
Annual production : .... . . . . . . . . . . . . . ton/year 
Fish price : sardine : Rp . .. . .  -... . . /kg 
mackerel : Rp . .. . . -... . . /kg 
skipjack : Rp .. . . . - ... . . /kg 
tuna : Rp . ... . - ... . . /kg 
small tuna : Rp . ... . -. .. . . /kg 
: Rp .. . . . - . . . . ./kg 
364 
Factories participating in the cannery survey: 
1 .  P.T. Bali R.aya, Benoa, Ball 
2. P.T. Blambangan R.aya, Muncar, East Java 
3 .  P.T. Karya Manunggal Prima Sukses, Muncar, East Java 
4. P.T. Sumino, Negara, Bali 
5. C. V. Harapan Lancar, Muncar, East Java 
6. C.V. Samudra R.aya, Negara, Bali 
7. P.T. Maya Muncar, Muncar, Bali 
8. P.T. Bali Maya Permai, Negara, Ball 
9. P.T. Sumber Yala Samudra, Muncar, Bali 
10. P.T. Bali R.aya, Negara, Bali 
1 1 .  P.T. Sofcol, Bitung 
12. P.T. CIP, Denpasar, Bali 
13 .  N.V. Muncar, Muncar, Ball 
14. P.T. Sinar Laut Jaya, Muncar, Ball 
15. P.T. Pengambengan R.aya, Negara, Bali 
16. P.T. Indo Bali, Negara, Bali 
365 
Appendix 9.3. Questionnaire used for consumer survey 
CONSUMER SURVEY 
Dear Madam/Sir: 
I am a PhD student in the Food Technology Department, Massey University, New Zealand. I am 
conducting a survey for a product development project on the use of fish oil in food for the 
Indonesian market. 
The purpose of this questionnaire is to obtain information from the consumer for developing the 
product. 
Please answer the questionnaire below and return it to the person distributing the questionnaire or 
to: 
Hari Eko Irianto 
cl- Ir.Giyat:ffii Irianto 
Akademi Gizi Muhammadyah 
Jln. W onodri Dalam II/22 
SEMARANG 
I would like to thank you in advance for your contribution to this project. 
Sincerely yours, 
Hari Eko Irianto 
366 
I. Fish and Fish Product Consumption 
1 .  Indicate your frequency in consuming fish or fish products in the following table 
Products frequency of consumption 
once/ twice/ >twice/ twice/ once/ 
week week week month month 
1. Fresh fish, 
including frozen and 
chilled fish 
2. Processed product: 
- dried salted fish 
- boiled salted fish 
- fermented fish/shrimp 
paste 
- pedah (moist fermented 
fish) 
- jambal (spongy fermented 
fish) 
- fish sauce 
- canned fish 
- smoked fish 
- softened bone fish 
- fish ball 
II. Fish Oil 
Would you like to consume refined fish oil, if you knew that fish oil had health benefits? 
( )  YES ( ) NO 
How would you like to consume fish oil? 
( )  capsule 
( ) table spoon 
( )  salad oil 
( )  disguised into ordinary foods 
m. Canned fish and canned fish containing rJSh oil 
1 .  When buying canned fish, do you consider the kind of medium? 
( ) YES ( ) NO 
If "YES", what kind of medium most like? 
( ) tomato sauce 
( ) vegetable oil 
( ) brine 
( ) vegetable oil and brine mixture 
( ) others, please specify .... . . . . . . . . . . . . . . . . . . . 
2. When buying canned fish, do you also consider the fish species? 
( ) YES ( ) NO 
If "YES", what kind fish species do you frequently buy? 
( ) sardine 
( ) mackerel 
( ) tuna 
( ) others, please specify ..... . . . . . . . . . . . . . . . . . . 
Please consider this idea: 
367 
I would like to introduce canned fish with disguised fish oil in order to optimise the fish oil 
utilisation and to produce canned fish nutritionally better than existing product in the market 
3. Do you fmd that product described above attractive? 
( ) YES ( ) NO 
- 4. As listed below, which medium of canned fish would you prefer the fish oil to be disguised? 
( ) tomato sauce 
( ) vegetable oil 
( ) brine 
( ) vegetable and brine mixture 
( ) others, please specify .... . . . . . . . . . . . . . . . . . . . . . . . . . . . .  
5. Would you like to buy this product? 
( ) YES ( ) NO 
6. In your opinion, what size would you like to buy? 
( )  155 g ( )  185 g ( )  215 g ( )  415g 
7. What price would you expect for this product? 
( ) Rp. 400 - 999.?
( )  Rp. 1000 - 1799.?
( ) Rp. 1800 - 2599.?
( ) Rp. 2600 - 3000.-
368 
==================================== 
Please fill in this information. It will be used only for analysis and will be kept confidential. Your 
name and address will help me if I may need to conduct another survey for this product 
Name 
Address 
Age : . .. . . . . . . . years 
Occupation . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Family income: 
( ) >Rp. 500,000/month 
( ) Rp. 150,000 - 500,000/month 
( ) <Rp. 150,000/month 
Appendix 9.4. Canned fish product being available in Indonesian market 
No. Trade Fish Ingredients/medium Weight Can form Price (Rp.-) Producer 
Mark Species (g) 
1 Yamato Sardine tomato sauce, salt, sugar, MSG 400 oval 1030 Indonesia 
2 Yamato Sardine tomato sauce 425 tube 960 - 1050 Indonesia 
3 Yamato Sardine tomato sauce 155 tube 425 - 475 Indonesia 
4 Bantan Sardine tomato sauce 400 oval 1010 - 1055 Indonesia 
5 Bantan Sardine tomato sauce 425 tube 975 - 1075 Indonesia 
6 Bantan Sardine tomato sauce 155 tube 415 - 455 Indonesia 
7 Bantan Mackerel tomato sauce 400 oval 1210 - 1700 Indonesia 
8 Del Monte Sardine water, tomato paste, salt, cellulose gum, 213 oval 1550 - 2020 Korea 
natural flavouring 
9 Del Monte Sardine water, tomato paste, salt, cellulose gum, 425 oval 2555 - 3250 Korea 
natural flavouring 
10 Gaga Mackerel tomato, garlic, chilli !55 tube 620 - 660 Indonesia 
1 1  Gaga Mackerel tomato, garlic, chilli 425 tube 1500 Indonesia 
12 Gaga Sardine tomato, garlic, chilli !55 tube 470 - 475 Indonesia 
13 Botan Mackerel tomato sauce 425 oval 2675 - 3345 Indonesia 
14 Bot an Mackerel tomato sauce 425 tube 1700 - 1880 Indonesia 
15 Botan Mackerel tomato sauce !55 tube 740 - 850 Indonesia 
16 Botan Sardine tomato sauce 425 tube 1350 - 1795 Indonesia 
17 Botan Sardine tomato sauce 155 tube 535 - 800 Indonesia 
18  Botan Sardine tomato sauce 215 oval 1825 - 1945 Indonesia 
19 Botan Sardine tomato sauce 425 oval 3 100 - 3350 Indonesia 
20 CIP Sardine tomato sauce 400 oval 1400 - 1675 Indonesia 
21 CIP Sardine tomato sauce 155 tube 615  Indonesia 
22 Three Star Sardine tomato sauce 425 tube 975 - 1055 Indonesia. 
23 Three Star Sardine tomato sauce 155 tube 400 - 415 Indonesia 
24 King's Fisher Sardine tomato sauce, salt 155 tube 410 - 440 Indonesia 
25 King's Fisher Sardine tomato sauce, salt 425 tube 1000 - 1065 Indonesia 
26 King's Fisher Tuna oil, onion, green peas, spices 185 tube 1030 - 1055 Indonesia 
-
V) 
$ 
Continuation of appendix 9.4. 
No. Trade Fish Ingredients/medium 
Mark Species 
27 Bottom Mackerel tomato sauce 
28 Bottom Mackerel tomato sauce 
29 Calmex Sardine tomato sauce, salt 
30 Pronas Tuna chunk vegetable oil, salt 
3 1  Pronas Tuna water, salt 
32 Pronas Sardine tomato sauce 
33 Kiku Sardine tomato sauce 
34 Swallow Sardine tomato sauce 
35 Kokin Sardine tomato sauce, palm oil, onion, sugar, salt 
36 Deho Chunk light Tuna vegetable oil, salt 
37 Bill tan Mackerel water, tomato paste, salt, cellulose gum, natural 
flavouring 
38 Bill tan Mackerel tomato sauce 
39 Bill tan Mackerel tomato sauce 
40 May a Sardine tomato sauce 
41 Maya Sardine tomato sauce 
42 Maya Sardine tomato sauce with cilli 
43 Katan Tuna fried chilli, spice, onion, garlic, vegetable oil, 
coconut milk, glutamat salt (sambel goreng) 
44 Fried Dace Dace black bean, vegetable oil, soy sauce, salt, spice 
45 A yam Sardine tomato sauce, salt 
46 A yam Sardine tomato sauce, salt, olein 
47 A yam Sardine tomato sauce, salt, olein 
48 A yam Sardine tomato sauce, salt olein 
49 A yam Sardine tomato sauce, salt, olein 
Weight Can form 
(g) 
!55 tube 
425 tube 
425 oval 
185 st.tube 
200 st.tube 
425 tube 
!55 tube 
!55 tube 
!55 tube 
185 st.tube 
425 oval 
425 ' tube 
155 tube 
425 tube 
155 tube 
!55 tube 
185 st.tube 
227 oval 
425 oval 
215  oval 
425 tube 
230 tube 
!55 tube 
Price (Rp.-) 
550 - 650 
1350 - 1920 
2240 - 2290 
1200 - 1995 
1215 - 1455 
1 1 50 - 1250 
395 - 425 
410 - 480 
400 - 425 
975 
1085 
1075 - 1450 
450 
1 175 
450 
525 
1850 
1925 - 2580 
41 10 - 4275 
1550 - 2050 
2950 - 3720 
1825 
1600 
Producer 
Indonesia 
Indonesia 
Mexico 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
Indonesia 
PR China 
Korea 
Korea 
Korea 
Korea 
Korea 
l>l -.1 0 
Continuation of appendix 9.4. 
--????-- ?---
No. Trade Fish Ingredients/medium 
Mark Species 
50 Heinz Greenseas sandwich tuna vegetable oil, salt, water 
5 1  Heinz tuna vegetable oil, salt, water 
52 Heinz grecnseas light meat salt, water 
chunk style tuna 
53 Star Kist solid light tuna spring water, vegetable broth, salt 
54 S & W  chunk light fancy tuna vegetable oil, salt, hydrolysed protein 
55 S & W  solid light fancy tuna soya oil/cotton seed oil, salt 
56 S & W  fancy red sockeye blue salt 
black salmon 
57 S & W  fancy red sockeye tuna salt 
58 John West chunk style tuna vegetable oil, salt 
59 John West tuna water, salt 
60 John West sandwich tuna vegetable oil, salt 
61  John West pink salmon salt 
62 John West pink salmon salt 
63 John West pink salmon salt 
64 John West tuna onion, vegetable oil, tomato puree, vinegar, sugar, 
salt, spices 
65 John West Red salmon salt, water 
66 John West herring fillets tomato puree, water, soya oil, vinegar, modified 
starch, spices 
67 John West skipper filter water, salt 
Weight Can form 
(g) 
170 st. tube 
180 st. tube 
180 st. tube 
184 st. tube 
184 st. tube 
184 st. tube 
220 st. tube 
220 st. tube 
185 st. tube 
185 st. tube 
185 st. tube 
105 st. tube 
210 st. tube 
440 tube 
1 85 st. tube 
105 st. tube 
200 oval 
200 oval 
---------- -- --
Price (Rp.-) 
4675 - 7500 
4715 
9750 
5000 
3 100 - 3265 
3375 - 3545 
4135 
7250 - 13500 
5 105 - 5280 
4840 - 5280 
4840 -5280 
4800 
6350 - 8400 
7700 - 1 1600 
4840 - 5280 
4800 
5700 
5700 
Producer 
Australia 
Australia 
Australia 
USA 
USA 
USA 
USA 
USA 
Australia 
Australia 
Australia 
Australia 
Australia 
Australia 
Australia 
Australia 
Australia 
Australia 
w 
-...! ...... 
Continuation of appendix 9.4. 
No. Trade Fish Ingredients/medium 
Mark Species 
68 John West skippers smoked vegetable oil, salt, liquid smoke 
brisling 
69 Hormel skinless & boneless water, salt 
chunk pink salmon 
70 Bumble Bee sockeye red salmon salt 
7 1  Bumble Bee sockeye red salmon salt 
72 Duchef tuna chunk vegetable (soya bean) oil 
73 Ligo sardines tomato sauce 
74 Smiling Fish fried sardine sugar, chilli soy sauce, vegetable oil 
75 Minerva sardine salt, soy oil 
76 Minerva sardine tomato, vegetable oil, salt 
77 Plumrose mackerel fillets tomato concentrate, salt, water 
78 Plumrose mackerel fillets vegetable oil, salt 
79 White Rose chunk light tuna water seasoned with vegetable broth, salt, 
pyrophosphate 
80 Gold Cup sardines tomato sauce 
8 1  Gold Cup sardines tomato sauce 
82 A yam sardines fish oil, salt 
83 Sea Gift smoked sardine sardine oil, salt 
Weight Can form 
(g) 
106 oval 
141 st.  tube 
210 st.  tube 
440 tube 
170 st. tube 
425 tube 
155 tube 
125 rectangular 
125 rectangular 
125 ' oval 
125 oval 
184 st. tube 
400 oval 
425 tube 
106 rectangular 
106 rectangular 
Price (Rp.-) 
4200 - 4375 
6960 - 8020 
8190 
15895 
1500 - 1700 
1925 - 2300 
2160 
2970 
2970 
4200 - 4395 
4200 - 4395 
1275 - 1495 
1 150 
1 150 
3000 
2970 
Producer 
Australia 
USA 
Canada 
Canada 
Australia 
Chile 
1l1ailand 
Portugal 
Portugal 
Denmark 
Denmark 
Indonesia 
Indonesia 
Indonesia 
Norway 
Norway 
? 
-..1 
N 
373 
Appendix 9.5. Dimensions of can found in the market 
Can Size (g) Height (Cm) Diameter 
Tall tube can 
155 8.8 5.3 
425 1 1 .3 7.5 
440 1 1 .3 7.5 
Short tube can 
141 4.1 8.5 
170 a 4.5 8.5 
170 b 4.2 8.5 
1 80 4.5 8.5 
184 4.2 8.5 
185 a 4.6 8.5 
185 b 4.0 8.3 
200 4.5 8.5 
210 4.8 8.5 
220 5.0 8.417.7 
Oval can 
213 3.1 8.5/12.4 
227 4.7 7.2/14.5 
400 3.9 10.8/15.7 
420 3 .3 10.9/15.7 
Rectangular can Length Wide/height 
106/125 10.6 8.7/1 .8 
Appendix 9.6. Chi-square, degree of freedom and Cramer's  V of Crosstab analysis 
results for consumer survey 
Variable 
Age*Fish oil consumption method 
Age*Fish species selection 
Age*Medium selection 
Age*Product idea 
Age*Buying trend 
Income*Fish oil consumption method 
Income*Fish species selection 
Income*Medium selection 
Income*Product idea 
Income*B uying trend 
Occupation*Fish oil consumption method 
Occupation*Fish species selection 
Occupation*Medium selection 
Occupation *Product idea 
Occupation*Buying trend 
?ote: UF = de ree of freedom g 
Cramer's V classification (Craft, 1990): 
< 0.10 
0.1 1 - 0.25 
0.26 - 0.40 
0.41 - 0.50 
> 0.50 
Chi Square DF Crariler's V 
131 .628 8 0.709 
131 .703 8 0.709 
134.072 8 0.715 
135.595 8 0.719 
135.044 8 0.718 
133.206 6 0.713  
131 .593 6 0.709 
136.198 6 0.721 
132.813 6 0.712 
136.685 6 0.722 
131 .386 4 0.708 
135.881 4 0.720 
131 .004 4 0.707 
131 .259 4 0.708 
131 .431 4 0.712 
= weak association 
= weak to moderate association 
= moderate association 
= moderate to strong association 
= strong association 
374 
Appendix 10.1 .  Sensory form used for evaluating tomato sauce acceptability 
SENSORY EVALUATION FOR TOMATO SAUCE 
USED IN CANNED FISH 
Name : ... . . . . . . . . . . . . . . . . . . . . . . . . . . 
Date : ... . . . . . . . . . . . . . . . . . . . . . . . . . . 
Instruction: Please indicate the score that best reflects your attitude about the tomato sauce 
Description Score 
Extremely very acceptable 9 
Very acceptable 8 
Acceptable 7 
Slightly acceptable 6 
Not sure 5 
Slightly unacceptable 4 
Unacceptable 3 
Very acceptable 2 
Extremely very unacceptable 1 
Attributes Sample code 
Consistency 
Odour 
Colour 
Mouth feel 
Appearance 
Overall acceptability 
Comments: . . ...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 
Have you ever eaten canned fish in tomato sauce medium? yes/no 
375 
Appendix 10.2. Sensory f01m used for evaluating sterilized fish oil 
SENSORY EVALUATION OF STERILIZED FISH OIL 
Name : ... . . . . . . . . . . . . . . . . . . . . . . . . 
Date : . .. . . . . . . . . . . . . . . .. . . . . . . . . 
Instructions: please evaluate samples in front of you in terms of taste and odour. Score the 
samples using the following scale: 
Fishy taste and odour 
1 no fishy taste/odour 
2 
3 slightly fishy taste/odour 
4 
5 moderately fishy taste/odour 
6 
7 strong fishy taste/odour 
8 
9 extremely strong fishy taste/odour 
Sample Code Taste 
Rancid taste and odour 
1 Typical and fresh fish oil 
2 Bland-indicates incipient rancidity 
3 Rancidity just noticeable 
4 Rancidity clearly noticeable 
Fishy Rancid 
Odour 
Fishy Rancid 
376 
Appendix 10.3. The example of the calculation of ingredient effects 
STUDY ON FISH OIL EFFECTS 
1. Effect on sauce consistency 
Cold observation 
high amount of fish oil 
51+56 
-------- = 54 
2 
Warm observation 
high amount of fish oil 
47+53 
--------- = 50 
2 
low amount of fish oil 
53+32+56 
--------------- = 4 7 
3 
low amount of fish oil 
50+32+56 
------------- =46 
3 
377 
Comments: high amount of fish oil was preferred in terms of sauce consistency from both cold 
and warm samples 
2. Effect of sauce odour 
Cold observation 
high amount of fish oil 
51+52 
--------- = 52 
2 
Warm observation 
high amount of fish oil 
47+48 
--------- = 48 
2 
low amount of fish oil 
54+48+54 
------------ = 52 
3 
low amount of fish oil 
48+46+51 
------------ = 48 
3 
Comment: The amount of fish oil did not affect the odour of tomato sauce from both cold and 
warm observation 
Appendix 10.4. Results of chemical, physical, sensory analysis of fish oil as tbe effect of 
sterilization treatment 
Analysis: Peroxide Value (meqlkg) 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value 
no.ooc 1 16.7?C l21.1?C 
139 min. 79 min. 64 min. 
Vacuum 4.99 0.60 0.89 0.49 
Unrefined 
Non-vacuum 4.99 1 .26 1 .23 1 .37 
Vacuum 4.25 1 .36 1 .21 1 .09 
Refined 
Non-vacuum 4.25 _ 2.35 2.06 1 .95 
Analysis: TBA value {pmoles/g) 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value 
no.ooc 1 16.7?C l21.1?C 
139 min. 79 min. 64 min. 
Vacuum 15.66 4.14 3 .42 3.81 
Unrefined 
Non-vacuum 15.66 3.54 3 .76 3.82 
Vacuum 15.40 4.65 4.42 4.12 
Refined 
Non-vacuum 15.40 6.07 6.03 6.04 
378 
379 
Analysis: Anisidine value 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value 
1 10.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 5.59 9.91 9.88 10.62 
Unrefined 
Non-vacuum 5.59 1 1 .28 1 1 .06 1 1 .64 
Vacuum 4.85 7.24 7.48 7.42 
Refined 
Non-vacuum 4.85 9.08 8.93 8.74 
Analysis: Totox value 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value 
l lO.OOC 1 16.7?C l21.1?C 
139 min. 79 min. 64 min. 
Vacuum 15.58 1 1 . 1 1  1 1 .67 1 1 .60 
Unrefined 
Non-vacuum 15.58 13 .80 13.53 14.38 
Vacuum 13 .36 9.95 9.91 9 .60 
Refined 
Non-vacuum 13.36 13.34 13.04 12.64 
Analysis: Free Fatty Acid Value (% oleic acid) 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value 
1 10.0?C 1 16.7?C 121 .1?C 
139 min. 79 min. 64 min. 
Vacuum 2.71 2.66 2.58 2.61 
Unrefined 
Non-vacuum 2.71 2.67 2.67 2.54 
Vacuum 2.33 2.29 2.27 2.25 
Refined 
Non-vacuum 2.33 2.29 2.31 2.31 
380 
Analysis: Refractive Index value 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value no.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 1 .4691 1 .4691 1 .4692 1.4692 
Unrefined 
Non-vacuum 1 .4691 1 .4692 1 .4693 1 .4692 
Vacuum 1 .4660 1 .4661 1 .4662 1 .4661 
Refined 
Non-vacuum 1 .4660 1 .4662 1 .4662 1 .4662 
Analysis: Colour value ( Photometric method) 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value no.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 21 .01 25.77 25.68 26.51 
Unrefined 
Non-vacuum 21.01 29. 11  27.84 28.33 
Vacuum 16.41 18.37 18.75 17.92 
Refined 
Non-vacuum 16.41 18.45 18.51 18.61 
Analysis: Fishy Odour 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value no.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 6.33 4.08 4.50 4.17 
Unrefined 
Non-vacuum 6.33 4.83 4.50 4.33 
Vacuum 4.55 4.89 3.61 3.54 
Refined 
Non-vacuum 4.55 4.00 3.86 3 .93 
381 
Analysis: Rancid Odour 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value no.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 1 .37 1 .92 1 .92 2.08 
Unrefined 
Non-vacuum 1 .37 1 .75 1 .71 1 .83 
Vacuum 1 .30 2.04 2.00 2.00 
Refined 
Non-vacuum 1 .30 1 .86 1 .79 1.71 
Analysis: Fishy Taste 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value no.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 5.70 4.25 4.42 4.33 
Unrefined 
Non-vacuum 5.70 4.83 4.75 4.17 
Vacuum 4.55 3.93 3.75 3.68 
Refined 
Non-vacuum 4.55 4.18 3 .75 3.89 
Analysis: Rancid Taste 
Temperature and Period of 
Fish oil Sterilization Initial Sterilization 
condition value no.ooc 1 16.7?C 121.1?C 
139 min. 79 min. 64 min. 
Vacuum 1 .37 1 .92 1 .92 2.00 
Unrefined 
Non-vacuum 1 .37 1 .75 1 .50 1 .83 
Vacuum 1 .23 1 .64 1 .64 1 .86 
Refined 
Non-vacuum 1 .23 1 .64 1 .64 1 .64 
Analysis: Fatty acid profiles (% fatty acids) 
Unsterilized Sterilillllion Condition 
oil 
Fatty 
acids 1 10'C for 139 minutes 1 16.7'C for 79 minutes 121 . 1  'C for 64 minutes 
UR"1 R?> Vacuum Non-vacuum Vacuum Non-vacuum Vacuum Non-vacuum 
UR R UR R UR R UR R UR R UR R 
14:0 5.8 5.8 5.1  5.8  6.0 5.9 6.1 5.9 6.1  6.2 5.8 5.7 6.2 4.5 
15:0 0.7 0.4 0.6 0.6 0.7 0.7 0.7 0.5 0.4 0.7 0.4 0.6 0.7 0.3 
16:0 12.8 12.9 12.9 12.8 12.9 12.9 13.1  12.8 13.2 13.3 12.7 12.5 13.4 13.4 
16 : 1  8.0 8.1 8.2 8.1  8.3 8.2 8.4 8.3 8.5 8.5 8.1  8.0 8.4 8.6 
17:0 0.3 0.2 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 
17:1  2.3 2.2 2.4 2.0 2.4 2.4 2.4 2.3 2.4 2.7 2.4 2.3 2.4 2.5 
1 8:0 2.5 2.6 2.5 2.5 2.0 2.0 2.2 2.0 2.5 2.0 2.5 2.2 2.6 2.3 
18 : 1  28.6 29.1 28.9 29.1 29.0 29.2 28.9 29.1 29.1 29.5 28.8 28.6 29.0 29.7 
1 8:2 1 .7 1 .8  1 .8  1 .8  1 .8 1 .8  1 .8 1 .8 1 .8  1 .8 1 .8 1 .8 1 .8 1 .8 
1 8:3 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.9 0.7 0.8 0.8 0.8 
1 8:4 2 .1  2.1  3.0 2.9 2.9 2.6 3.0 3.0 2.9 1 .9 2.4 3.0 2.9 2.1 
20:1 7.8 7.8 7.8 7.8 7.7 7.8 7.5 7.8 7.6 7.7 7.9 7.8 7.6 7.8 
20:3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.3 0.3 0.3 0.3 
20:4 0.7 0.7 0.8 0.7 0.8 0.7 0.7 0.7 0.7 0.7 0.8 0.8 0.7 0.7 
20:5 1 1 .0 10.6 10.9 10.7 10.7 10.5 1 1 .0 10.7 10.2 10.2 1 1 . 1  1 1 .0 10.2 10.8 
22:1 3.7 3.7 3.5 3.5 3.5 7.6 3.4 3.6 3.4 3.5 3.6 3.7 3.4 3.6 
22:4 0 . 1  0 . 1  0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0. 1 0.1 
22:5 4.8 4.4 3.9 4.4 4.4 4.3 4.1 4.4 4.1  3.9 4.8 4.7 3.6 4.4 
22:6 6.0 5.4 5.7 5.7 5.5 5.7 5.1 5.8 5.5 5.6 5.7 6.0 5.6 5.9 
?ote: "1 UK - unreunea ou 
**) R = refined oil 
...., 
?3 
Appendix 1 1 .1 .  Sensory sheet for Plackett and Burman experiment of canned fish 
PLACKETT AND BURMAN EXPERIMENT OF 
CANNED FISH 
Name : ... . . . . . . . . . . . . . . . . . . . . . . .. . 
Date : ... . . . . . . . . . . . . . . . . . . . . . . . . . 
383 
Please evaluate each sample for the following attributes by placing a vertical line on the scale at 
the point which you think best describes the product and put the sample code at the top of this line. 
Please also make a line where your IDEAL canned fish would be and label this line with "I" 
Example: 
FISH 
1. Appearance 
2. Flesh texture 
3. Bone softness 
4. Sourness 
5. Saltiness 
6. Overall spiciness 
342 123 421 
I 
not broken 
tender 
not soft 
not sour 
not salty 
not spicy 
very broken 
tough 
very soft 
very sour 
very salty 
very spicy 
384 
7. Fishiness 
not fishy very fishy
? 
TOMATO SAUCE 
1. Colour 
bright red not bright red/dark 
2. Mouth feel of sauce 
not oily very oily 
3. Sourness 
not sour very sour 
4. Saltiness 
not salty very salty 
5. Overall spiciness 
not spicy very spicy 
6. Fishiness 
not fishy very fishy 
OVERALL ACCEPT ABILITY 
not acceptable very acceptable 
Thank you for your time 
385 
Appendix 1 1 .2. Sensory sheet for experiment on canning process optimization 
FISH CANNING PROCESS OPTIMIZATION 
Name: .. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 
Date: ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
Please indicate the score that best reflects your attitude about each sample in tetms of product 
characteristics and acceptability 
Score Product Characteristics 
9 
8 
7 
6 
5 
4 
3 
2 
1 
Extremely very .... . . . . . . 
Very .... . . . . . . . . . . . . . . . . 
Slightly .. . . . .. . . . . . . . . . 
Not sure 
Slightly not ........ . . . . 
Not ... . . . . . . . . . . . . . . . . . . 
Very not ..... . . . . . . . . . . . 
Extremely very not .. . . .  . 
Score Product Acceptability 
9 
8 
7 
6 
5 
4 
3 
2 
1 
Extremely very acceptable 
Very acceptable 
Acceptable 
Slightly acceptable 
Not sure 
Slightly unacceptable 
Unacceptable 
Very unacceptable 
Extremely very unacceptable 
Note: . . . . .  means followed by description placed under evaluated characteristics 
Attributes 
FISH 
1 .  Flesh texture 
(tender-not tender or 
tough 
2. Softness of bone 
(soft-not soft) 
3.  Sourness 
(sour-not sour) 
Product characteristic/Product acceptability 
S A M P L E  C O D E S  
4. Saltiness 
(salty-not salty) 
5. Overall spiciness 
(spicy-not spicy) 
6. Fishiness 
(fishy-not fishy) 
TOMATO SAUCE 
1. Sauce colour 
(bright red-not bright 
red or dark red) 
2. Mouth feel of sauce 
(oily-not oily) 
3. Sourness 
(sour-not sour) 
4. Saltiness 
(salty-not salty) 
5. Overall spiciness 
(spicy-not spicy) 
6. Fishiness 
(fishy-not fishy) 
OVERALL ACCEPTABILITY 
COMMENTS: ...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 
Thank you for your time 
386 
387 
Appendix 12.1 .  Questionnaire form used for consumer product testing 
CONSUMER TEST 
Dear Sir/Madam 
Attached to this questionnaire is a sample of canned sardine. The product has been developed in 
the Food Technology Department, Massey University, New Zealand for the Indonesian market. 
I would like you to taste the sample and answer the questions below. 
Thank you for your time and cooperation. 
Hari Eko Irianto 
PhD Student 
Food Technology Department 
Massey University 
Palmerston North 
New Zealand 
INSTRUCTION: 1 .  Taste this product the way you would normally eat canned fish, but do not 
add anything, e.g. spices, water, salt, etc. 
2. Please answer the questions by ticking the appropriate blank 
I. How did you eat the canned fish? 
(a) treatment: 
. ... the samples was heated (warmed) 
.. . . no treatment at all 
(b) with other foods 
. ... with rice and/or other foods, e.g. vegetables .... 
. ... with rice only 
. ... the sample only 
. ... other, please specify .. . . .. . . . . . . . . . . . . . . ........ . 
ll. Question about the FISH 
1 .  How is the flesh texture when you bite it? 
. . .. very tender 
. ... slightly tender 
. ... just right 
.. . . slightly tough 
. ... very tough 
2. How is the bone softness when you bite it? 
.... very soft 
.... slightly soft 
.... just right 
.. . . slightly hard 
.... very hard 
3 .  How is the sourness of the fish? 
.... very sour 
.... slightly sour 
.... just right 
. .. .  slightly lacking in sourness 
. ... very lacking in sourness 
4. How is the saltiness of the fish? 
.. . . very salty 
.... slightly salty 
.... just right 
.... slightly lacking in saltiness 
.... very lacking in saltiness 
5. How is the overall spiciness of the fish? 
.... very spicy 
. ... slightly spicy 
.... just right 
. ... slightly lacking in spiciness 
. ... very lacking in spiciness 
6. How is fishiness of fish? 
. ... very fishy 
. .. .  slightly flShy 
.... just right 
. ... slightly non-fishy 
. ... very non-fishy 
m. Questions for the TOMATO SAUCE 
1 .  What is the colour of tomato sauce? 
.. . . very bright red 
. ... slightly bright red 
. ... just right 
. . .. slightly dark red 
.... very dark red 
2. How does the mouth feel when eating the tomato sauce? 
. ... very oily 
. .. .  slightly oily 
.... just right 
.. . . slightly non-oily 
. ... very non-oily 
388 
3. How is the sourness of the tomato sauce? 
.... very sour 
.... slightly sour 
.... just right 
. .. .  slightly lacking in sourness 
. ... very lacking in sourness 
4. How is the saltiness of the tomato sauce? 
.... very salty 
.. . . slightly salty 
.... just right 
. ... slightly lacking in saltiness 
. .. . very lacking in saltiness 
5. How is the overall spiciness of the tomato sauce? 
.... very spicy 
.. . . slightly spicy 
.... just right 
. ... slightly lacking in spiciness 
. ... very lacking in spiciness 
6. How is the fishiness of tomato sauce? 
. ... very fishy 
. .. . slightly fishy 
.... just right 
.... slightly non-fishy 
.... very non-fishy 
IV. Overall acceptability 
. ... like it very much 
... . like it slightly 
.. . . neither like nor dislike 
. ... dislike it slightly 
. ... dislike it very much 
V. Personal Data 
1 .  Sex : . . . . male 
2. Age (years): Under 20 
20 - 30 
3 1 - 40 
41 - 50 
Over 50 
3. Carrier : - school pupil 
.... female 
- college or university student 
- government officer 
- working at company 
- private work 
- others, specify .... . . . . . . . . . . . . . . . 
389 
4. Family income: Under Rp.150,000 
Rp.l50,000 - Rp.299,999 
Rp.300,000 - Rp.500,000 
Over Rp.500,000 
5. City: . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 
390 
QUESTIONNAIRE FOR HOUSEWIFE OR PERSON BEING RESPONSIBLE 
FOR SHOPPING 
391 
INSTRUCTION: 1 .  Taste this product the way you would normally eat canned fish, but do not 
add with anything, e.g. spices, water, salt, etc 
2. Please answer the questions by ticking the appropriate blank 
I. How did you eat the canned fish? 
(a) treatments: 
. . . .  the sample was heated (warmed) 
.. . .  no treatment at all 
(b) with other foods: 
.... with rice and/or other foods, e.g. vegetables .... 
.... with rice only 
.... the sample only 
. .. . others, please specify ... . . . . ............. . 
II. Question about the FISH 
1 .  How is the flesh texture when you bite it? 
. .. . very tender 
. ... slightly tender 
.. . .  just right 
.... slightly tough 
... .  very tough 
2. How is the bone softness when you bite it? 
.... very soft 
. ... slightly soft 
.. . .  just right 
.. . .  slightly hard 
.... very hard 
3. How is the sourness of fish? 
.... very sour 
.... slightly sour 
. .. .  just right 
.. . .  slightly lacking in sourness 
.... very lacking in sourness 
4. How is the saltiness of fish? 
.. . .  very salty 
. ... slightly salty 
.. . .  just right 
.. . .  slightly lacking in saltiness 
.. . .  very lacking in saltiness 
5. How is overall spiciness of fish? 
. ... very spicy 
.. . .  slightly spicy 
.. . .  just right 
.. . .  slightly lacking in spiciness 
. ... very lacking in spiciness 
6. How is fishiness of fish? 
.... very fishy 
... . slightly fishy 
... . just right 
. ... slightly no fishy 
. .. .  very no fishy 
lll. Questions for the TOMATO SAUCE 
1 .  What is the colour of tomato sauce? 
. ... very bright red 
... . slightly bright red 
.... just right 
.... slightly dark red 
... . very dark red 
2. How is the mouth feel when eating the tomato sauce? 
. ... very oily 
.... slightly oily 
.... just right 
. . .. slightly not oily 
. ... very not oily 
3. How is the sourness of the tomato sauce? 
.... very sour 
. ... slightly sour 
. ... just right 
.. . . slightly lacking in sourness 
. ... very lacking in sourness 
4. How is the saltiness of the tomato sauce? 
.. . . very salty 
. ... slightly salty 
.. . . just right 
.... slightly lacking in saltiness 
. ... very lacking in saltiness 
5. How is the overall spiciness of the tomato sauce? 
.. . . very spicy 
. ... slightly spicy 
.. . . just right 
. ... slightly lacking in spiciness 
.. . . very lacking in spiciness 
6. How is the fishiness of tomato sauce? 
.... very fishy 
. ... slightly fishy 
. ... just right 
. .. .  slightly not fishy 
... . very not fishy 
392 
IV. Overall acceptability 
. . .. like it very much 
... . like it slightly 
. ... neither like nor dislike 
. ... dislike it slightly 
. ... dislike it very much 
V. Buying trend Survey 
1 .  Would you buy this product? 
.... yes 
. .. .  no, please give a reason . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 
If "no", you do not need to continue to further questions. 
2. How often would you buy it? 
. ... everyday 
.... twice a week 
.... once a week 
.... once every two weeks 
.. . . once a month 
.. . . occasionally 
3 .  What do you think would be a reasonable price for this product? 
Rp .... . . . . . . . . . . . . . . . . 
4. When you inspect the label, what information do you prefer to retain? 
. . . .  healthy information 
. .. .  nutritional information 
. ... ingredients 
.... net weight 
. ... trade mark 
. ... name and address of processor 
. ... other information needed to be added, specify .. . .. . . . . . . 
5. Where would prefer to buy this product? 
. ... drugstore 
. ... supermarket 
.. . . food shop 
. ... retailer 
. ... others, please specify .... . . . . . . . . . . . . . . . . . . . . . . 
6. What is your reason for buying this product? 
. . .. healthy reason 
. ... nutritional reason 
. . .. family preference 
. .. . convenience, easy to serve 
. ... reasonable price 
. ... you like to eat it 
. . . . others, please specify .... . . . . . . . . . . . . . . . . . . . . . . . 
393 
VI. If you have any comments about this product, please write them here! 
V. Personal Data 
1 .  Sex : .... male 
2. Age (years): Under 20 
20 - 30 
3 1 - 40 
41 - 50 
Over 50 
3. Carrier : - school pupil 
.. . . female 
- college or university student 
- government officer 
- working at company 
- private work 
- others, specify .... . . . . . . . . . . . . . . . 
4. Family income: Under Rp.150,000 
Rp.150,000 - Rp.299,999 
Rp.300,000 - Rp.500,000 
Over Rp.500,000 
5.  City: ... . . . . . . . . . . . . . . . . . . . . . . . . . 
394 
395 
Appendix 12.2. Questionnaire form for medical doctor survey 
QUESTIONNAIRE FOR DOCTOR? SURVEY 
Dear Sir/madam, 
I am a post-graduate student in the Food Technology Department, Massey University, New 
Zealand, conducting research into the development of a new fish product for the Indonesian market. 
The aim of this questionnaire is to obtain the opinion of the medical fraternity for my research into 
canned fish with fish oil added. 
Hopefully, you can help me by completing this questionnaire. I appreciate your help and 
cooperation. 
Thank you. 
Hari Eko Irianto 
Post Graduate Student 
cl-Food Technology Department 
Massey University 
NEW ZEALAND 
(1) A lot of experiments on fish oil have been conducted in order to show the benefit to our health, 
especially the reduction of coronary disease risk. 
(a) Have you ever suggested your patients consume fish oil for their health? 
( ) yes 
. ( ) no 
(b) Have you ever suggested your patients eat more fish for their health? 
( ) yes 
( ) no 
(2) If you think that the fish oil is good for our health, in which 
form you find that your patients should consume fish oil. 
( ) capsule ( ) table spoon 
( ) salad oil ( ) disguised in ordinary foods 
( ) others, specify ... . . . . . . . . . . . . . . 
396 
Please consider this idea: 
I would like to introduce canned fish which contains fish oil disguised in the medium. The rational 
behind this idea is to optimise the fish oil utilization and to produce canned fish nutritionally better 
than the existing in the market. 
(3) Do you think that the product described above is a good way to increase fish oil intake of 
consumer? 
( ) yes 
( ) no 
( ) maybe 
(4) Do you think that this product has a good prospect in the market from the medical point of 
view? 
( ) yes 
( ) no 
( ) maybe 
(5) If this product is available in the marlcet, would you advice your patients to buy for health 
reason? 
( ) yes 
( ) no 
( ) maybe 
===================================================================== 
Please ftll in information below: 
1 .  How long have you been a doctor ? ......... years 
2. How many patients do you have in average each month? ......... . 
people 
3.  Where are you from originally? ............... . 
- 4. Speciality ................................... . 
397 
Appendix 12.3. Infonnation on the label of the developed canned fish product distributed during 
consumer testing 
"OMEGITA" - lkan sardine dalam saus tomat 
* diperkaya dengan minyak ikan 
(asam lemak omega-3) 
* tanpa bahan pengawet 
Bahan: 
ikan pilchard, air, minyak ikan, pasta tomat, 
gula, bawang merah, bawang putih, garam dan 
cuka 
Komposisi: 
Air 68.2%, protein 14.6%, lemak 10.1%, 
karbohidrat 4.3%, abu 2.8% 
Departemen Teknologi Pangan 
Massey University, SELANDIA BARU 
English translation of the label: 
"OMEGITA" Sardine in tomato sauce 
* enriched with fish oil (omega-3) 
* no preservatives 
Ingredients: 
pilchard, tomato paste, water, fish oil,sugar, 
shallot, garlic and vinegar 
Nutritive values: 
Moisture 68.2%, protein 14.6%, fat 10. 1  %, 
carbohydrate 4.3%, ash 2.8% 
Food Technology Department 
Massey University, NEW ZEALAND 
Appendix 12.4. Medical doctor's comments on developed canned fish 
----?-?????------ -???????-??---- ------- -- ---
Can canned fish be used for fish oil consumption 
Demographic improvement? 
characteristic 
YES NO MAYBE 
Sp?lallly: (N/%) (N/%) (NI%) 
Pathologist 0? 11100 0? 
Paediatrist 8/38 1/5 12/57 
Internist 14174 311 6 2110 
Obstetrician- 19190 2/10 0? 
gynaecologist 
Gen.prac. 73n2 27127 Ill 
Experience 
(years): 
<5 58m 18123 3/4 
5 - 10  20/83 4/17 0? 
1 1  - !5 15/48 8126 8/26 
16 - 20 8m 2/1 8 119 
>20 1 3n2 2/1 1 3/17 
Number of total 
patient/month: 
<100 9156 6138 1/6 
100 - 300 38/62 20133 3/5 
301 - 500 41/82 6/12 3/6 
50! - 700 !5nt 0? 6129 
>700 1 1m 2/13 2/13 
Number of patient 
having heart 
problem/month 
<10 75168 27/24 9/8 
10 - 30 24/73 5115 4/12 
31 - 50 1 !185 1/8 1/8 
>50 4/67 1/17 1/17 
Does canned fish have 
a good prospect in Indonesia? 
YES NO MAYBE 
(N/%) (N/%) (NI%) 
1/100 0? 0? 
20195 0? 115 
19/100 0? 0? 
17/81 115 3/14 
46/46 3/3 52151 
41/52 3/4 35/44 
13/54 0? 1 1146 
22n! 216 7123 
10191 0? . 119 
16/89 010 2/1 1 
8/50 2/13 6137 
30/49 315 28/46 
34/68 0? 16132 
18/86 0? 3/14 
12/80 0? 3/20 
62/56 514 44/40 
22/67 0? 1 1133 
13/100 0? 010 
5/83 0? 1/17 
-- -?-????--?????-?--???-???????-------------------
Will you suggest your patient to consume this 
product? 
YES NO MAYBE 
(N/%) (N/%) (N/%) 
0? 1/100 0? 
20195 0? 115 
18195 010 115 
21/100 010 0? 
88/87 13/13 010 
70/89 9/1 1 010 
22192 2/8 0? 
27/87 3/10 1/3 
1 11100 010 0? 
17/94 0? 1/6 
13/81 3/19 0? 
50/82 10/16 1/2 
49/98 112  0? 
21/100 0? 0? 
14/93 0? 1n 
97/87 13/12 l/1 
32/97 1/3 0? 
12/92 0? 1/8 
61100 0? 0? 
I 
V) 
\0 
00 
Appendix 12.5. Fatty acid profile changes in fish oil 'and canned fish product during production trial 
-?????------ -
Tomato sauce 
Fatty acids Fish oil 
Unsterilized Sterilized 
14:0 4.4 4.2 4.3 
15:0 0.8 0.8 0.8 
16:0 17.0 17.3 17.6 
16: 1 5.5 5.2 5.4 
17:0 0.7 0.7 0.8 
17: 1 1 .4 1 .3 1 .3 
18:0 5.5 5.9 5.9 
18: 1  33.9 32.7 32.2 
18:2 1 .8 2.0 1 .9 
18:3  1 .7 1 .9 1.8 
18:4 2.4 2.6 2.6 
20: 1 5 .1  4.1 4.0 
20:3 0.2 0.2 0.2 
20:4 0.8 0.8 0.8 
20:5 6.6 7.7 7.8 
22: 1 2.1 1 .7 1 .6 
22:4 0.2 0.1 0.2 
22:5 1 .6 1 .8 1 .8 
22:6 8.3 8.8 8.9 
-??? ???-? ---------?? ??????------- ----??????-???---?---
Fish Canned. t1sh product 
Unsterilized Sterilized Unsterilized Sterilized 
6.7 6.3 5.6 6.1 
1 .8 1 .3 0.9 0.9 
27.4 26.2 20. 1 19.6 
3.7 5.0 4.9 5.6 
1.9 1.2 0.9 0.9 
1 .4 1 .5 1 .5 1 .3 
7.1 6.1 6.1 5.6 
15.0 15.5 28.6 28.9 
3.0 2.0 2.0 2.0 
1 .7 1 .7 1 .8 1 .7 
1 .2 1 .5 2.3 2.3 
3.3 2.9 3.5 3.6 
0.3 0.3 0.2 0.2 
1.6 1.2 0.9 0.8 
5.7 7.1 7.5 7.6 
2.1 1 .3 1 .1 1 .3 
0.2 0.2 0.1 0.1 
0.7 1 .2 1 .6 1 .4 
15.2 17.2 10.1 9.9 
Vl 
? 
-400 
Appendix 12.6. Chi square, degree of freedom and Cramer' s V of crosstab analysis results from 
Consumer product testing 
Variables Chi-Square DF Cramer's V 
Age*Fish texture 419.528 20 0.510 
Age*Bone softness 421 .842 20 0.512 
Age*Fish sourness 452.120 25 0.470 
Age*Fish saltiness 413.287 25 0.453 
Age*Fish spiciness 433.078 25 0.464 
Age*Fish fishiness 424.518 25 0.459 
Age*Sauce saltiness 416.768 25 0.455 
Age*Sauce sourness 420.914 25 0.457 
Age*Sauce fishiness 444.056 25 0.469 
Age*Sauce spiciness 418.504 25 0.456 
Age*Sauce mouth feel 440.45 25 0.468 
Age*Sauce colour 432.182 25 0.463 
Age*Overall product acceptability 427.18 25 0.460 
Carrier*Fish texture 428.809 32 0.516 
Carrier*Bone softness 426.178 32 0.514 
Carrier*Fish sourness 478.524 40 0.487 
Carrier*Fish saltiness 428.680 40 0.461 
Carrier*Fish spiciness 442.877 40 0.469 
Carrier*Fish fishiness 441 .814 40 0.468 
Carrier*Sauce saltiness 442.583 40 0.469 
Carrier*Sauce sourness 450.692 40 0.473 
Carrier*Sauce fishiness 432.255 40 0.463 
Carrier*Sauce spiciness 435.783 40 0.465 
Carrier*Sauce mouth feel 442.048 40 0.468 
Carrier* Sauce colour 434.724 40 0.464 
Carrier*Overall product acceptability 446.308 40 0.471 
City*Fish texture 445.671 20 0.526 
City*Bone softness 426.886 20 0.515 
City*Fish sourness 450.078 25 0.473 
City*Fish saltiness 440.099 25 0.467 
City*Fish spiciness 449.642 25 0.472 
City*Fish fishiness 479.457 25 0.488 
City* Sauce saltiness 443.173 25 0.469 
City*Sauce sourness 435.087 25 0.465 
City*Sauce fishiness 491 .1 12 25 0.494 
City*Sauce mouth feel 450.611  25 0.473 
City*Sauce colour 443.263 25 0.469 
City*Overall product acceptability 483.294 25 0.490 
Income*Fish texture 433.829 16 0.519  
Income*Bone softness 436.183 16 0.520 
Income*Fish sourness 432.871 20 0.518 
Income*Fish saltiness 435.713 20 0.520 
Income*Fish spiciness 430.547 20 0.517 
Income*Fish fishiness 434.469 20 0.519 
Continuation of appendix 12.6 
Variables 
Income*Sauce saltiness 
Income*Sauce sourness 
Income*Sauce fishiness 
Income*Sauce spiciness 
Income*Sauce mouth feel 
Income*Sauce colour 
Income*Overall product acceptability 
Sex*Fish texture 
Sex*Bone softness 
Sex*Fish sourness 
Sex*Fish saltiness 
Sex* Fish Spiciness 
Sex*Fish fishiness 
Sex*Sauce saltiness 
Sex*Sauce sourness 
Sex*Sauce fishiness 
Sex*Sauce spiciness 
Sex*Sauce mouth feel 
SeX*Sauce colour 
Sex*Overall product acceptability 
Age*Buying trend 
Age*Buying frequency 
Carrier*Buying trend 
Carrier*Buying frequency 
City*Buying trend 
City*Buying frequency 
Income*Buying trend 
Income*Buying frequency 
Sex*Buying trend 
Sex*Buying frequency 
Canned fish customer*Buying trend 
Canned fish customer*Buying frequency 
Fish species selection*Buying trend 
Fish species selection*Buying frequency 
Medium selection*Buying trend 
Medium selection*Buying frequency 
Overall prod.accept*Buying trend 
Overall prod.accept*Buying frequency 
?Jote: TIF - oe ree of treedom g 
Cramer's V classification (Craft, 1990): 
< 0.10 
0.1 1  - 0.25 
0.26 - 0.40 
0.41 - 0.50 
> 0.50 
Chi-Square DF Cramer's V 
421.648 20 0.51 1  
429.599 20 0.516 
415.842 20 0.508 
428.009 20 0.515  
420.234 20 0.5 1 1  
418.506 20 0.510 
431 .055 20 0.517 
405.297 8 0.709 
417.379 8 0.720 
405.240 10 0.709 
404.864 10 0.709 
412.524 10 0.715 
410.556 10 0.714 
407.731  10 0.7 1 1  
405.117 10 0.709 
406.215  10  0.710 
410.026 10 0.713 
407.081 10 0.7 1 1  
406.246 10 0.710 
41 1.316 10 0.714 
148.382 10 0.718 
158.558 30 0.469 
150.610 14 0.723 
171 .705 42 0.446 
153.624 10 0.730 
184.264 30 0.506 
148.002 8 0.717 
163.210 24 0.532 
145.290 4 0.710 
148.157 12 0.717 
150.27 4 0.722 
152.43 12 0.728 
145.633 4 0.7 1 1  
152.023 12 0.727 
145.856 4 0.712 
148.334 12 0.718 
189.896 10 0.812 
200.731 30 0.528 
= weak association 
= weak to moderate association 
= moderate association 
= moderate to strong association 
= strong association 
401 
402 
Appendix 12.7. Chi-square, degree of freedom and Cramer's V of crosstab analysis results from 
medical doctor survey 
Variables 
Speciality*Fish oil consumption suggestion 
Speciality*Fish consumption suggestion 
Speciality*Fish oil consumption method 
Speciality*Product idea 
Speciality*Product prospect 
Speciality*Product consumption suggestion 
Num.of HP*Fish oil consumption suggestion 
Num.of HP*Fish consumption suggestion 
Num.of HP*Fish oil consumption method 
Num.of HP*Product idea 
Num.of HP*Product prospect 
Num.of HP*Product consumption suggestion 
Num.of GP*Fish oil consumption suggestion 
Num.of GP*Fish consumption suggestion 
Num.of GP*Fish oil consumption method 
Num.of GP*Product idea 
Num.of GP*Product prospect 
Num.of GP*Product consumption suggestion 
Work exp.*Fish oil consumption suggestion 
Work exp. *Fish consumption suggestion 
Work exp.*Fish oil consumption method 
Work exp. *Product idea 
Work exp.*Product prospect 
Work exp.*Product consumption suggestion 
?ote: HP """ Patients havm ? heart roblem g p 
GP = General patients 
DF = degree of freedom 
Cramer's V classification (Craft, 1990): 
< 0.10 
0.1 1 - 0.25 
0.26 - 0.40 
0.41 - 0.50 
> 0.50 
Chi-square DF Cramer's v 
172.258 10 0.725 
171 .319  10 0.723 
195.1 14 20 0.545 
240.312 15  0.699 
234.070 15  0.690 
188.780 15 0.619 
169.314 8 0.718 
166.284 8 0.712 
176.591 16 0.519 
167.689 12 0.584 
176.556 12 0.599 
173.476 12 0.594 
168.439 10 0.717 
166.208 10 0.712 
188.007 20 0.535 
189.886 15  0.612 
182.844 15 0.610 
181 .925 1 5  0.608 
172.869 10 0.726 
168.972 10 0.718 
176.626 20 0.519 
183 .614 15  0.611  
181 .813 15 0.608 
172.682 15  0.592 
= weak association 
= weak to moderate association 
= moderate association 
= moderate to strong association 
= strong association