Abstract
Addition of small amounts of calcium hydroxide (slaked lime) to fish, press liquor or polishing water in the fish meal and oil production process leads to a considerable reduction in the free fatty acid content of the resulting fish oil.
INTRODUCTION
In a previous paper from this Institute the development of free fatty acids (FFA) in fish oil at the different stages of production was examined Citation[1]. It was found that no formation of FFA occurs during its production, which involves a cooking, pressing, centrifuging and polishing (washing) stage. FFA are, however, generated during enzymatic spoilage of the fish prior to processing.
Spoiled fish inevitably leads to high FFA levels (up to 30%) in the oil and ultimately to financial penalties. The present paper describes a novel method for reducing the FFA content of fish oil by refining it through the addition of small amounts of slaked or unslaked lime (calcium hydroxide or calcium oxide) to either the fish (at the cooking stage), the press liquor (after the pressing stage), or the polishing water.
This procedure, which is incorporated in the manufacture of the fish oil, is referred to as in situ refining. The in situ refining procedure is particularly attractive to the producers of fish oil as little or no extra equipment is necessary for its implementation. In addition, lime can be purchased cheaply, while the calcium soaps resulting from the lime treatment end up in the fish meal. If the addition of slaked lime is kept below 0.5% (based on the wet fish) the extra calcium accumulating in the meal is only of the order of 1% and not detectable.
Finally, the in situ refining procedure has been adopted with remarkable success by several fish meal and oil factories in the Republic of South Africa.
MATERIALS AND METHODS
General
Anchovy (Engraulis japonicus), pilchard (Sardinops ocellata) and red-eye (Etrumeus whiteheadi) were collected from a nearby factory during the 1995–1998 fishing seasons and transported to the Institute where they were converted into fish oil and fish meal in the pilot plant.
Addition of Lime to Fish
The fish was divided into 3 kg portions and treated with 0.2–1% (based on the weight of the fish) slaked lime. The fish was then cooked (20 minutes) and the liquid separated from the solids in a basket centrifuge at 4000 rpm. The liquid phase (press liquor) was treated with about 300 mℓ of hexane and centrifuged at 2500 rpm. The hexane layer was removed with a pipette, filtered over anhydrous sodium sulphate and evaporated on a rotary evaporator. The FFA content of the resulting oil was then determined. A control sample, without addition of lime, was included in the experiments.
The aqueous layer (stickwater) was concentrated on a rotary evaporator and mixed with the solid phase (press cake). This press cake was dosed with the antioxidant ethoxyquin (1,2-dihydro-6-ethoxy-2,2,4-trimethylquinoline) at a level of 400 mg/kg and dried in an oven at about 70°C. After drying, the solids were ground in a laboratory hammer mill and the residual meal lipids extracted with chloroform:methanol (2:1, by vol.) as described Citation[2]. The FFA content of the residual meal lipids was then determined.
Addition of Lime to the Press Liquor
The fish was treated as described previously, but 0.05–0.1% lime (based on the weight of the fish) was added to the press liquor and thoroughly mixed into it. The oil was then extracted with 300 mℓ of hexane and treated as described above. A control without any addition of lime was included in the experiments.
Addition of Lime to the Polishing Water
At the outset the FFA content of the unpolished oil was determined and the theoretically required amount of lime to neutralise the FFA in the oil calculated (the stoichiometric amount). The oils were then polished by vigorously shaking 50 mℓ of the crude oil with 5 mℓ of water or 5 mℓ of an emulsion of the stoichiometric amount of lime in water. This was done in a centrifuge bottle immersed in boiling water for 3 minutes. The mixtures were subsequently centrifuged at 2500 rpm and the top layer of oil removed with a Pasteur pipette. The FFA content of the polished oils was then determined.
Vulnerability Towards Atmospheric Oxidation of the Oils
The oils produced by the in situ refining technique should ideally be indistinguishable from normally produced oils, particularly in respect of vulnerability towards atmospheric oxidation. To test this, four of the in situ refined oils were stored at 25°C in partially filled stoppered glass bottles and their behaviour on storage monitored by comparing their polyunsaturated fatty acid (PUFA) content at day 0 and at day 90 with that of a control oil.
ANALYTICAL PROCEDURES
Free Fatty Acid Content
The FFA content of the oils and meal lipids was determined by titrating about 0.3 g material in a mixture of hot 20 mℓ ethanol, plus 5 mℓ chloroform with a 0.02M potassium hydroxide using phenolphthalein as indicator. The results were expressed as percentage oleic acid (on a dry basis).
Polyunsaturated Fatty Acid Content
The PUFA content of the fish oils was determined by gas chromatography of the methyl esters on a capillary column Citation[3]. The identity of individual fatty acids was established by comparing the retention times of their methyl esters with standards obtained from Sigma Chemical Company, St Louis, Missouri, USA. Those fatty acids containing two and more double bonds were classified as PUFA.
RESULTS AND DISCUSSION
Addition of Lime to Fish
Table records the effect of addition of 0.2–1.0% lime to red-eye and anchovy on the FFA levels of the resulting fish oils and their residual meal lipids. This clearly shows that lime addition to fish was effective in minimising FFA levels of fish oils.
Table 1. Effect of Lime Addition to Fish on the FFA Content of the Resulting Oil and Meal Lipids
For instance, addition of 0.33% lime to anchovy gave a reduction in FFA content of anchovy oil from 13.03 to 7.00%, while the FFA content of the meal lipids remained almost identical at about 18%. Addition of 1.0% lime gave a reduction in FFA content from 12.70 to 1.84%, while the FFA content of the meal lipids was reduced from 24.06 to 11.81%.
Addition of Lime to Press Liquor
In Table are recorded the FFA contents of four fish oils and their corresponding meal lipids resulting from the addition of 0.05–0.1% lime to press liquor. These results show that this treatment was also very effective in reducing the FFA levels of the oils. For instance, the oil produced on treating the red-eye press liquor with 0.05% lime was 5.34%, compared to a control value of 10.00%, while treatment with 0.1% lime of pilchard press liquor gave an oil with FFA content of 1.75%, compared to a control value of 7.96%. In this case there was no reduction in the FFA content of the meal lipids, so the required amount of lime was much less than when it was added to the fish.
Table 2. Effect of Lime Addition to Press Liquor on the FFA Content of the Resulting Oil and Meal Lipids
Addition of Lime to Polishing Water
Table records the effect that stoichiometric lime addition to polishing water had on the FFA contents of four unpolished fish oils. For instance, the FFA content of oil 1 was reduced from 9.84 to 3.02% by polishing the oil with 10% of a 12.9% aqueous emulsion of lime, while 10% water only reduced the FFA content to 9.50%. Not all the FFA in the oil was neutralised, despite using sufficient lime. Therefore, in order to completely remove the FFA from the oil, an excess of lime has to be used. However, this is not recommended as in situ refining is not intended to exhaustively refine a fish oil, but merely to reduce its FFA content, thereby reducing the penalty levied by fish oil refiners. Completely neutralising all FFA in an oil can only be accomplished by using an excess of a concentrated solution of sodium or potassium hydroxide. This treatment, however, also removes all the natural antioxidants and a fully refined fish oil is therefore extremely vulnerable to atmospheric oxidation and can develop rancidity in a matter of days Citation[4]. In contrast, in situ refining of fish oils does not render the oils more vulnerable to atmospheric oxidation, as is shown in the following results.
Table 3. Effect of Lime Addition to Polishing Water on the FFA Content of the Resulting Polished Oil
Vulnerability of the Oils Towards Atmospheric Oxidation
The four in situ refined fish oils resulting from the addition of lime to polishing water were tested for their vulnerability towards oxidation in air at 25°C. In Table are recorded the PUFA levels of the in situ refined fish oils versus the controls at day 0 and day 90.
Table 4. Polyunsaturated Fatty Acid Levels (PUFA) of in Situ Refined Fish Oils During Storage at 25°C
The results indicate that all the oils, including the controls, showed some loss of PUFA during storage. This resulted from the oils being exposed to abnormally harsh conditions, i.e. partially filled bottles exposed to air. For example, in the worst case, in situ refined oil number 2 suffered a loss of 14.3 percentage points in PUFA during the 90 days storage, while its control did not perform much better by losing 11.4 percentage points. Viewed as a whole, however, the in situ refined oils compared very favourably with their respective controls and there seems no reason to discard the in situ refining procedure on the grounds of increased vulnerability towards atmospheric oxidation.
Calcium Soaps
The in situ refining technique seems very attractive to fish oil producers as little or no investment is required in new equipment and no waste products are produced.
The calcium soaps generated from the FFA in the oil and the lime end up in the fish meal. This inevitably raises the calcium content of the meal. Normally the calcium content of fish meal varies from 2.2–8.7% with a mean value of 4.0% (n = 64 s.d. = 1.0%) Citation[5]. If an arbitrary limit is set of not raising the calcium content of a meal by more than 1.0%, this means that not more than 0.25% calcium or 0.5% slaked lime should be added to the fish (assuming a conversion factor of fish to fish meal of roughly 4). Results of our tests clearly show that 0.5% lime addition to raw fish will markedly reduce the FFA content of the fish oil (see Table ). Far less lime was required to reduce the FFA content of the oils when this was added to the press liquor, i.e. 05–0.1%. In that case, similar reductions in FFA contents of fish oils were achieved with approximately 1/25th of the amount of lime needed for raw fish (see Table ). This is to be expected, as no lime is necessary for the reduction of the FFA in the residual meal lipids. This procedure eliminates concern about excess lime addition.
The consequences of lime addition to the polishing water of fish oils are similar to those of lime addition to the press liquor. For instance, if the oil in fish containing 5% oil is polished with 10% of a 20% aqueous lime emulsion, calculation shows this to be equivalent to 0.1% lime addition to fish, which is well below the suggested limit of 0.5% (see Table ). The addition of the stoichiometric amount of lime to the polishing water is therefore also unlikely to raise the calcium content of a fish meal by more than 1%.
Acknowledgments
REFERENCES
- De Koning , A. J. , Milkovitch , S. , Fick , M. and Wessels , J. P.H. 1986 . The Free Fatty Acid Content of Fish Oil. An analysis of anchovy lipids at different stages in the manufacture of anchovy meal and oil . Fette Seifen Anstrichmittel , 88 : 404 – 406 .
- De Koning , A. J. and McMullan , K. B. 1966 . Phospholipids of Marine Origin III. The Pilchard (Sardine ocellata, Jenyns) with Particular Reference to Oxidation in Pilchard Meal Manufacture . Journal of the Science of Food and Agriculture , 17 : 385 – 388 .
- Christie , W. W. 1982 . Lipid Analysis , Second Edition 53 – 67 . Oxford : Pergamon Press .
- Hamilton , R. J. , Kalu , C. , McNeill , G. P. , Padly , F. B. and Piece , J. H. 1998 . Effects of Tocopherols, Ascorbyl Palmitate and Lecithin on Autoxidation of Fish Oil . Journal of the American Oil Chemists' Society , 75 : 813 – 822 .
- De Koning , A. J. 1999 . The Calcium Content of South African Fish Meals . unpublished