Antioxidants and the autoxidation of fats

Summary Experiments on the oxidation of purified methyl oleate support the view that its induction period, and probably that of natural oils, is due to the presence of inhibitors and that purified unsaturated compounds have no induction period, other than the time required for gaseous oxygen to diffuse into the liquid. Experiments with antioxidants indicate that phenolic inhibitors and inhibitols cause no change subsequent to the end of the induction period, that they exert their effect solely by inhibiting the formation of the initial active moloxide, and that they are entirely destroyed before the start of rapid oxidation which characterizes the end of the induction period. The mode of action of several different pro-oxidants is analyzed. Perbenzoic acid, and presumably other peracids, and copper oleate decrease the induction period by virtue of their destruction of natural inhibitors.     

Further observations on the dicarbonyl compounds formed via autoxidation of methyl linoleate

a-Dicarbonyls isolated from oxidized methyl linoleate and conclusively identified as DNP-osazones³ were glyoxal, methyl glyoxal,a-keto hexanal,a-keto heptanal, anda-keto octanal. Technical Paper No. 2012, Oregon Agricultural Experiment Station.     

On the kinetics of the autoxidation of fats

The mechanism of fat autoxidation is elucidated from the rate data. All the data treated here and in an earlier publication follow the same basic rate equation, including the time function f(t) empirically derived for heterogeneous oxidation. Metals and glass (the wall of the reaction vessel) are catalysts. Depending on the state of the catalyst, f(t)=t² or f(t)=t. When f(t)=t, the kinetics are first-order as found for monolayer autoxidation, but in bulk phase they are complicated by a transient stage caused by the solubilization of O₂ into the hydroperoxide micelles produced in the exponential (“autocatalytic”) part of the oxidation. Certain additives, such as inhibitors, affect the catalyst and thereby f(t). The kinetics, as determined by O₂ consumption or by analysis of the remaining unreacted substrate, show the first oxidation step. It is unaffected by further chemical changes of the primary oxidation products,e.g., decomposition of hydroperoxides and trimerization in the autoxidation of 9,11-octadecadienoic acid methyl ester.     

Reactions of phosphoric acid in the autoxidation of fats

At the elevated temperatures used in accelerated stability tests phosphoric acid was shown to react with fatty peroxides in such a manner as to prevent their estimation by the usual iodometric methods. A dark, phosphorus-containing precipitate formed, which was apparently polymeric in nature. Phosphoric acid also reacted with oxidized fat to form an undisassociated fat-soluble complex but did not react with fresh unsaturated fatty esters. In addition to the well known sparing action of phosphoric acid ona-tocopherol, it was shown thata-tocopherol and quinone also exerted a marked sparing action on the phosphoric acid.     

Isolation and identification of carbonyl compounds formed by autoxidation of ammonium linoleate

1. The isolation and identification of carbonyl compounds formed by autoxidation of pure ammonium linoleate (0.02 M. in water) has been carried out.
2. The autoxidation has been performed according to a method described by Tappel (23). A solution of 0.02 M. ammonium linoleate in 0.1 M. phosphate buffer of pH 6.9, containing 10⁻⁴ M. CuSO₄, was subjected to autoxidation at 37.2°C.
3. The isolation and identification of the D.N.P.-ones of the carbonyl compounds have been carried out, using a method of partition and adsorption chromatography published by van Duin (10, 11).
4. It was demonstrated that the principal compounds formed were n-hexanal, 2-octenal, and 2,4-decadienal.
5. The formation of these aldehydes is fully in line with the theories of autoxidation of unsaturated fatty acids (1, 3, 13) and of the dismutation of the hydroperoxides formed (2).

     

Comparitive study of monocarbonyl compounds formed during deep frying in different fats

Fresh commercial corn oil, fresh commercial lard, and hydrogenated vegetable shortening were analyzed for carbonyl compounds before and after deep frying. The frying was carried out in an apparatus with a capacity for 2000 g of oil designed to quantitatively trap the volatile materials evolved during frying and which would ordinarily escape into the atmosphere. The trapped distillate was also subjected to carbonyl analysis. Analysis of the fats and distillates showed a carbonyl pattern in essential agreement with the classical autoxidation mechanism for the different fats, i.e., the typical alkanals, alk-2-enals, and alk-2,4-dienals. The pattern correlated generally with the fatty acid composition of the fats. Comparison of the concentrations of the monocarbonyl compounds in the fats before and after frying, and in their distillates, indicated that the deodorization process which accompanies frying is effective in preventing the accumulation of the more volatile compounds formed. The less volatile products, mainly deca-2,4-dienal, were not efficiently removed. Accordingly, it was observed that the oils containing higher proportions of linoleic acid contained more residual monocarbonyl compounds after frying. Scientific Article No. A-1124, Contribution No. 3570 of the Maryland Agricultural Experiment Station.     

Identification of the most intense odor compounds formed during autoxidation of methyl linolenate at room temperature

The volatile compounds formed during autoxidation of methyl linolenate (MLe) at 22–24°C were analyzed by a technique which reveals the flavor compounds having the highest flavor units (ratio of the concentration to the odor treshold). After a reaction time of 48 hr, when ca. 20 mol percent of the MLe had been converted into hydroperoxides,trans, cis-2,6-nonadienal followed by 1,cis-5-octadien-3-one,rans,cis-3-5-octadien-2-one andcis-3-hexenal showed the highest flavor units. After 102 hr 1,cis-5-octadien-3-one was by far the most important odor compound, followed bycis-3-hexenal andtrans,cis-2,6-nonadienal. These results were presented at the AOCS Annual Meeting in New Orleans in May 1987.     

A study of the flavor stability and autoxidation of beef fats

Summary The odoriferous compounds isolated from a fresh edible tallow were found to be very complex in nature and could not be diminished by treatment with carbonyl reagents. When these compounds were removed by deodorization, the bland tallow which was obtained developed on autoxidation another type of odoriferous compounds. The latter contained various carbonyl compounds of the type which have been associated with flavor reversion in edible oils. The characteristic odor of oleo oil and edible tallow could be removed by slight hydrogenation of high selectivity under atmospheric pressure. This hydrogenation process raised the melting point by less than 2°C. but substantially decreased the linolenic and arachiidonic acid content. The hydrogenated products, after deodorization, were not only bland in odor and flavor but also had flavor stabilities better than those of some of the best commercial shortenings. Funds for this study were furnished by a grant-in-aid from Swift and Company, Chicago, Ill. Presented at the American Oil Chemists’ Society meeting, Minneapolis, Minn., October 11–14, 1954.     

Prostaglandin-like substances formed during autoxidation of methyl linolenate

Prostaglandin-like substances other than 3-(2-ethyl-5-hydroxy-3-oxo)-cyclopentanyl-2-propenal (compound I), which upon reacting with amino acids form reddish pigments, were purified, and their chemical structures were determined. Three red pigment-forming substances (RPS) in autoxidized methyl linolenate were purified successively by gel chromatography on Sephadex LH-20, column chromatography and thin layer chromatography on Silica Gel 60, and by high performance liquid chromatography on μ-Porasil. IR spectra of the RPS showed the absorption of an ester group (νC=O 1730 cm⁻¹ in CHCl₃) and of a longer branched chain (νC-H 2800–3100 cm⁻¹), in addition to the absorption bands observed in compound I. Electron impact mass spectra of the trimethylsilyl ethers obtained after reduction with NaBH₄ and NaBD₄ were analyzed in detail and compared with those of compound I. The structures of the RPS were identified as stereoisomeric methyl 8-[2-(2-formylvinyl)-3-hydroxy-5-oxo-cyclopentanyl]-octanoates derived from methyl 12-hydroperoxy-linolenate. Presented at the general meeting of the Japanese Society of Scientific Fisheries, held at Tohoku University, Sendai, Japan, October 1984.     

Antioxidants and the autoxidation of fats IV. Lecithin as an antioxidant

The commercial preparations known as “lecithin” have moderate antixoygenic action on refined cottonseed oil, little effect on lard, and none at all on lard-cod liver oil mixtures. Commercial lecithins contain only small amounts of true lecithin. The antioxygenic agent in these preparations is cephalin. Purified lecithin is not an oxidation inhibitor, but purified cephalin is. There is no quantitative relation between the amount of cephalin used and the prolongation of the induction period. The particular portion of the cephalin molecule responsible for its antioxygenic action is probably the mono-basic phosphoric acid radicle.     

On the kinetics of the autoxidation of fats. II. Monounsaturated substrates

The autoxidation of methyl oleate and oleic acid shows some differences as compared to the autoxidation of linoleate,e.g., the formation of water at an early stage. Linearization of experimental data on the autoxidation to high oxidation degrees of methyl oleate and other monounsaturated substrates shows that the rate equations previously derived for methyl linoleate in the range of 1–25% oxidation are valid, provided the correct expression for the remaining unreacted substrate is used. With monounsaturated substrates, part of the oxygen is consumed by a secondary oxidation reaction almost from the beginning, and only a certain constant fraction α of the total O₂ consumption is consumed in hydroperoxide formation. The fraction α is different for methyl oleate, oleyl alcohol, oleic acid andcis 9-octadecene, but the rate constant for the hydroperoxide formation is the same for all of them when experimental conditions are the same. The main difference between oleate and linoleate autoxidation is the much faster decomposition of the oleate hydroperoxides relative to their slow formation.     

Application of near infrared spectrophotometry to the study of the autoxidation products of fats

Summary The usefulness and limitations of near infrared spectroscopy as applied to the oxygenated products of fat oxidation has been studied. Hydroperoxides, acids, alcohols, esters, aldehydes, and ketones all absorb in the narrow range from 2,700 mμ to 3,000 mμ. Alcohols may be distinguished from hydroperoxides and acids, but the latter are too close together to be resolved. In mixtures all the C=O absorptions attributable to esters, aldehydes, and ketones combine to form one band. In autoxidizing methyl oleate, the increase in alcohol, carbonyl, and combined acid-hydroperoxide can be followed in the near infrared spectra of CCl₄ solutions. In the absence of acid the increase in hydroperoxide absorbance is linear with peroxide value. The changes in the spectra of a reduced and bicarbonate-washed sample of oxidized methyl oleate correlate well with the expected chemical changes. Presented at the 31st Fall Meeting of the American Oil Chemists’ Society, Cincinnati, O., Sept.30–Oct. 2, 1957. American Meat Institute Foundation Journal Paper Number 156.     

Microstructures formed by spray freezing of food fats

The spray-freezing of two food fats, tripalmitin (PPP) and cocoa butter (CB) and mixtures thereof, has been modeled experimentally using a novel single droplet freezing apparatus configured so that temperature profiles or samples for microstructure analysis can be obtained. For 2 mm diameter droplets suspended in a cold air flow at temperatures around 2–15°C, initial cooling rates were on the order of 10 K s⁻¹ and the temperature profiles could be correlated directly to DSC data collected at 20 K min⁻¹, indicating that minimal supercooling of the materials occurred in the droplet form. Microstructure analysis confirmed that PPP crystallized preferentially in mixtures, and that the surface structure was very sensitive to storage conditions. The bulk structure was much less sensitive, and the internal microstructure of the PPP droplets revealed distinct nucleation sites, which were absent from the CB: These persisted in the mixtures up to 50 wt%. X-ray analysis indicated that the fats crystallized in their more stable forms, namely, β for PPP and Form V/V1 in CB.     

Determination of pentane formed during autoxidation of oils contained in solid samples

A convenient and practical technique for the measurement of pentane formed during peroxidation of solid materials containing unsaturated oils was developed. The test has a sensitivity below. 1. parts per million (ppm) and can be used to monitor shelf-life of a variety of products. A special glassware was designed to perform a single extraction-concentration step. The pentane is recovered in 2 ml of hexane and analyzed by gas chromatography. This method correlates well with taste panel scores.     

Prooxidant effects of inorganic chromium compounds in the autoxidation of methyl linoleate

The prooxidant property of inorganic chromium compounds was determined in methyl linoleate free from natural antioxidants and metals. Prooxidant properties of inorganic chromium compounds appeared in order of sodium chromate > chromium (VI)-oxide > chromium chloride > potassium chromate > chromium (III)-oxide > potassium dichomate. In comparison with the control, additions of chromium compounds induced different amounts of autoxidation products derived from methyl linoleate, such as small amounts of hydroperoxides and conjugated dienes and large amounts of hydroxy groups,α,β,γ,δ-unsaturated carbonyls, isolatedtrans double bonds, polymers, and free radicals. From these analytical data, the catalysis of chromium compounds in the autoxidation of methyl linoleate seemed to be based on their abilities of abstracting a hydrogen from methyl linoleate and decomposing hydroperoxides derived from the autoxidation of methyl linoleate.     

The isolation and characterization of the polymers formed during the autoxidation of ethyl linolenate

Summary Pure ethyl linolenate was prepared and autoxidized; and dimeric and trimeric fractions were separated by solvent fractionation. These fractions were subjected to quantitative, functional group-analysis. On a basis of oxygen known to be present but not detected in the functional groups the possibility that polymerization had taken place through an oxygen linkage was considered. The mild nature of the reagent causing depolymerization suggested that this linkage was of the peroxide type rather than the dioxane ring. A report of work done under contract with the U.S. Department of Agriculture and authorized by the Research and Marketing Act of 1946. The contract was supervised by the Northern Regional Research Laboratory of the Bureau of Agricultural and Industrial Chemistry, Peoria, Ill. Presented at the American Oil Chemists' Society Meeting, October 12–14, 1954, Minneapolis, Minn.     

Autoxidation of tissue lipids. II. Monocarbonyl compounds formed by the autoxidation of methyl eicosapentaenoate,methyl docosahexaenoate,and cod-liver oil

Fatty acid analysis of autoxidized cod-liver oil with a peroxide value of 192 showed significant degradation of only eicosapentaenoic, docosapentaenoic, and docosahexaenoic acids of the linolenate family. Purified, mildly autoxidized cod-liver oil with a peroxide value of 28, methyl eicosapentaenoate, and methyl docosahexaenoate produced carbonyl patterns in agreement with the accepted mechanism for olefinic autoxidation. In all cases the major products were propanal andn-pent-2-enal as predicted, andn-hex-2-enal andn-hept-2-enal as reported in the literature for linolenate. In addition, the same cod-liver oil, which had been heated to 188C in vacuum for 1 hr to decompose completely the hydroperoxides before carbonyl analysis, showed the presence ofn-hepta-2,4-dienal as predicted.