CERTAIN FACTORS INFLUENCING THE OXIDATION OF BUTTER FAT Thesis for the Degree of M. S. ' MICHIGAN STATE COLLEGE Fay Carl Ewbank I942 CERTAIN FACTORS INFLUENCING THE OXIDATION OF BUTTER FAT CERTAIN FACTORS INFLUENCING THE OXIDATION OF BUTTER FAT By Fay Carl Ewbank .943 A THESIS Submitted to the Graduate School of Michigan State College of Agriculture and Applied Science in partial fulfilment of the requirements for the degree of MASTER OF SCIENCE Department of Dairy Husbandry June, 1942 TH E515 ACKNOWLEDGMENTS The writer sincerely wishes to express his appreciation to Doctor Earl weaver, Head of the Dairy Department, for making these studies possible, and to Doctor I. A. Gould for directing and planning these investigations and for his guidance in the preparation of the manuscript. y.» gs {4-3 {:3 ’49 C3 I. II. III. IV. V. ITITROD-UCTIOTI0...0........................000..... REVIEW OF LITERATURE ..................o..... A. B. F. SCOPE 0 HeChanism Of fat OXidation 00.....000000 Accelerated methods for determining the Stghility Of fats 00000.00...0.....0.0.0 Methods of measuring fat oxidation ..... Influence of heat treatment Upon oxidation . Influence of certain metals with and without heat upon the oxidation of butter fat ....... AntiOXIdantS .0........0.............0.. 01-] IEVESTIGATION 0....0................ EXPERIMENTAL PROCEDULE 000.00.00.00000000000. RESU'LI‘S 0.0.000.0000000000DOOOOOOOOOOOOOOOOOO A. B. C. D. EU 0 F. Studies of the peroxide method ......... Acceleration methods ................... 1. Aeration method .................. 2. hot air oven method ............... Heat influence on butter and butter oil Influence of heating cream on oxidation Studies of Avenex Concentrate .......... Carotene destruction as related to perox formation .............................. DISCEISS-[Obl OOOOOOOOOOOIOI.0.00.00...00.00.... SUKIARY AND CONCLUSIONS 0.................0.. -1...“ REFELfiyJ‘L‘ECES OOOOOOOOOOOOOOOOOOOOOOOO0.0.090... new J NDIX TABLES 000........0.00....o......... de }J0 29 30 S5 71 BI 83 96 105 STUDIES ON THE OXIDATION OF BUTTER FAT INTRODUCTION During storage, substances containing fat tend to undergo chemical changes resulting in objectionable flavors. The nomenclature describ- ing these flavor defects varies to some extent with different industries. For example, in the general field of fats and oils the term "rancidity" designates these off-flavors irrespective of the chemical changes involved. In the dairy field, however, differentiation is made depend- ing upon the Specific chemical reaction. Here the term "rancidity" is used to denote only those flavors resulting from a hydrolysis of fatty glycerides with the ultimate production of free fatty acids, whereas the term9"oxidized" and “tallowy” are used to characterize those flavors resulting from the oxidation of the unsaturated fatty acids. Economic losses resulting from.chemical Spoilage of butter during storage has stimulated considerable research on fat oxidation. Inves- tigators have observed that metals, light, air, acidity of the cream, salt, storage temperature, and chemical composition of the fat are among those factors affecting chemical deterioration of the unsaturated fatty acids. The knowledge gained through these investigations has resulted in numerous changes in commercial practices. For example, replacement of c0pper equipment with stainless steel has resulted to protect the product against metal contamination. Also, particular attention has been given the protection of fats from excessive exposure to light and air. Accurate standardization of cream acidity, salt control, and the use of low storage temperatures are all practiced to improve the keeping quality of butter. -2- Even though many studies have been conducted dealing with fat oxidation, there are certain phases which are in need of further investigation. There is a necessity of further studies upon the deve10pment of a simple, reliable, acceleration method for determin- ing the stability of fats toward oxidation. Also, within recent years, certain changes in manufacturing methods have occurred due to the develOpment of new equipment as well as to an effort on the part of plant Operators to hasten the processing procedure. One such change has been in the direction of high temperature heat treatment of the cream. In some cases, these transitions of processing procedure have come about without much consideration of their possible effect upon the keeping quality of the product during storage. Because of the incompleteness of available information regarding the influence of certain of these changing trends in processing procedures upon the chemical deterioration of the fat, studies in this connection seemed eSpecially apprOpriate. -3- REVIEW OF LITERATURE l. Mechanism of Fat Oxidation Many theories have been presented concerning the chemical changes involving the oxidation of butter fat. Although there are differences of Opinion as to the first products of oxidation, all workers are in agreement that the double bonds of the unsaturated fatty acids are the points of oxidation. Peroxides: Holm (50), Powick (91), Smith and WOod (112), and Tschirch (122) believe the first reaction is a formation of fatty acid peroxides by direct combination of the active oxygen with the double bonds of the unsaturated fatty glycerides. This reaction is presented by Holm.(50) as follows: 0-0 -C- =C-C- + 02 —-) ' ' Unsaturated Oxygen -C-C-C-C- fatty glyceride Peroxide These workers assume that after their formation, the peroxides then rearrange into moloxides and ozonides with subsequent formation of aldehydes, acids, and other substances. Staudinger (113) and Miles (78) are of the Opinion that the first step in the oxidation process is a formation of a highly reactive substance referred to as a moloxide or dative peroxide. Miles (78) suggests that, in addition, there is a connection between the energy content of the molecular valence electrons and the oxidation reaction. The first step in the oxidation is thought to be a combination of the valence electrons of the fat molecule to the active oxygen molecule as follows: Unsaturated Oxygen fatty glyceride H O Dative peroxide These unstable dative peroxides are thought immediately to rearrange to form more stable peroxides or other compounds. Oxides: Browne (11) suggests a combination of oxygen with the UL) .ce with a simultaneous liberation of ‘20 CT ,1- 4 O N H. 3.. double bond forming a f an atom of active oxygen. The active oxygen is then thought to act upon the glyceride causing the formation of free fatty acids, aldehydes, carbon dioxide, water, and other by products. Ozonides: Gortner (41) suggests the possible formation of ozonides by the addition of ozone at the double bonds. This reaction is pres- ented by Gortner as follows: -CH=CH- 4- 03 '--—-5D -CH-CH- rearrangemengb -CH' CH- Unsaturated ' ' ' ' fatty Ozone O—O: : O O—O glyceride Ozonide Ozonide These ozonides, since they are highly reactive, wdll readily decompose into hydrogen peroxide, acids, and half aldehyde acids. Other workers (106) suggest the breaking of peroxides into oxides and a molecule of active oxygen, the active oxygen then combining with another peroxide forming an ozonide which rearranges into acids and aldehydes. Water formation: In a more recent study, Mattill (75) found water to be an important oxidation product, and presents the following hypo— thetical equations: -CH2-CH2-CH-CH- -|- -O- —-) -CH=CH-CH-CH‘ + H20 ' 3 I ' O-O Active O-O Peroxide oxygen Unsaturated water peroxide -5- The new double bonds formed by the above reaction are believed to undergo further oxidative changes, thus: -CH=CH—CH—CH- + 02 ———) -CH-CH-CH-CH- ——) I I I I I I Peroxide Oxygen Double Peroxide -CH-CH-CHO + -CHO n I O-O Peroxide aldehyde Aldehyde The mechanism of fat oxidation has been summarized by Mattill (75) who states: "No all-inclusive theory of autoxidation can yet be formulated. Once the obscure induction period is past, events seem to follow thick and fast in greatest confusion, but like the devastating Speed and apparent disorder of a blitzkrieg, everything nevertheless goes according to plan. The plan is flexible and is in part dictated by the events themselves. Further careful study is needed to indicate what the events are and why they take place." II. Accelerated Methods for Determining_the Stability of Fats. Even though many methods have been devised and used to deter- mine the relative stability of oils and fats toward oxidation, one is yet to be found that is entirely reliable. However, all fat stability tests that have been used to any great extent are basi- cally the same. The oxidation is accelerated under carefully controlled conditions, thus reducing the induction period from several months or weeks to a few days or even hours. The catalysis of the oxidation process is usually achieved by heat, utilizing -6- either a hot liquid bath or a hot air oven, although in some instances some catalyst other than heat may be used; i.e., light or metals. Fat stability apparatus method: One method used in commercial laboratories for determining the relative stability of different fats was devised in Swift's Laboratories by King et a1 (62). In this method the samples of oil or fat to be tested are placed in to a test tube which is held in a constant temperature oil bath. Washed air is passed through the liquid fat at a constant rate in order to maintain an air-saturated condition in the oil. Samples are tested at definite time intervals to determine the end of the induction period. This method has been used successfully by Bull (17, 18, 19), Musher (83), and Stebnetz and Sommer (114, 115% and Freyer (40) reports on the use of this apparatus in eleven commercial laboratories. The precision of the fat stability apparatus to detect the rela- tive stability of different fats toward oxidation has been a matter of considerable question. The result of a three year committee study presented by Prayer (40) indicates that the testing of c00perative samples in 11 different laboratories by this method gave inconsistent results. Even by allowing 1? per cent average tolerance of induction period time, about one-fourth of the laboratories reported results outside of this liberal range. Four of the laboratories participating presented consistently uniform data, but these laboratories were among the first to adOpt the use of the method and, therefore, had consider- able more eXperience. Two of the laboratories reported inconsistent results at first, but after using the method for about a year, reported results that were in agreement with the other laboratories. However, -7- two laboratories failed to show acceptable results even after two years experience. Prayer (40) suggests that the lack of agreement between the different laboratories was apparently due to insufficient experience with the method and feels that even though the method is not perfect and has many limitations it still is the best accelerated procedure known for determining the relative stability of fats and oils toward oxidation. Hot air oven method: Barnicoat and Palmer (4), Bird (8), Dahle and Nelson (33), and Ritter and Nussbaumer (96, 97, 98, 99, 103, 104) have made use of hot air oven methods to determine the susceptibility of fat toward oxidation. Ritter and Nussbaumer (96, 97, 98, 99, 103, 104) held fat for eight hours in an oven at 1040 C. in 10 cm. petra dishes. At the end of the eight hour incubation period the peroxide value was measured to determine the relative stability of the fats. In their work with the oxidation of butter fat Barnicoat and Palmer (4) decanted the butter oil into straight-sided crystallizing dishes. These dishes were then heated in an air-stirred water-jacketed oven thermostatically controlled at 80° C. for six hours. The alde- hyde value was then used to determine the degree of oxidation of the oil. Because of the simplicity of the method, Bird (8) made use of an oven method in his studies of the stability of vitamin A—containing fish oils. In this work he exposed small, fairly uniform samples of fish oil to the atmOSphere at temperatures above normal. Samples were taken at intervals from the series and vitamin A determinations made to determine the stability of the fish oil. Dahle and Nelson (33) used a 600 C. hot air oven in their oxida- tion studies of butter fat. Equal volumes of fat were placed into test tubes which were held in the incubator in a basket. Although - 8 - the temperature of different tubes did not vary as much as one degree, the induction period of different tubes varied as much as four days. In this connection, Lea (70) points out that temperature control is essential in all rapid oxidation methods where elevated temperatures are used and further states that l0 C. variation in temperature may introduce as much as a 10 per cent error. III. Method of Measuring Fat Oxidation. Considerable effort has been expended to establish a method for quantitatively determining the oxidation of fats. In general, it may be said that a method for detecting fat oxidation should have the following characteristics: (a) simplicity, (b) reliability, (c) Specificity in reaction, and (d) sensitivity to such a degree as to detect minute oxidative changes in the fat, if possible, before the end of the induction period. Numerous methods have been used to detect the oxidation of fats and oils but many have failed to conform, even within reasonable limits, to the qualifications listed above. Among the less satisfactory methods may be included tests for unsaturation, the acid value, and the Kreis test. Investigations of Brown and coeworkers (16), Nelson and Dahle (84), Overman and associates (87), and Swanson and Sommer (118) indi- cates that changes in iodine absorption of butter fat expressed as iodine number do not correlate with initial oxidative changes reSpon- sible for flavor changes. However, Swanson and Sommer (118) found that a lowering of the iodine number of the phOSpholipid fraction accompanies the develOpment of oxidized flavor in milk. Briggs (10) found close relationship between the iodine number decrease and oxygen - 9 - absorption of butter fat. Henderson and Roadhouse (49) ShOW’a rela- tion between iodine number and induction period of fat from cows on normal and submaintenance rations. Stebnitz and Sommer (115) show that, in general, the greater the unsaturation, as measured by iodine number or thiocyanogeneiodine number, the less stable is the fat toward oxidation. Wiley and Gill (126) point out that the thiocy- anogen number, as determined by the usual method, is more inaccurate than the iodine number to determine the degree of unsaturation. Attempts have been.made by numerous investigators to use the acid value as a means of detecting the end of the induction period (6, 10, 125). Generally, these workers have concluded that this method is not sensitive in detecting the end of the induction period. The best known and probably the widest used qualitative test for detecting oxidative changes in fats and oils is the Kreis test. The Kreis test has been investigated by Hohm and Greenbank (51), Kerr (60), Kerr and Sober (61), Mattill (75), and Powick (91) with the conclusion that this test cannot be used as an accurate quantitative method. The accuracy of the test for detecting oxidation of fats, unless the color is examined SpectroscOpically, is questioned by Taffel and Revis (120) and Mattill (75) because the reaction is not Specific for epihydrin aldehyde. Other methods have been tried but their merits have not been entirely established. Holm.(50) used a dye reduction.method in his studies of keeping quality of butter. Bickford and Markley (7) and Greenbank and Holm.(45) develOped methods for determining the rate of oxidation of vegetable oil by making use of a fading methylene blue chloride-fat system. Another method utilized by Clark and Rugg (21), in working with soy bean oil, made use of a Special drOp-Spreading -10.- pressure method to detect oxidation. It was found that the readings increased continually with the drOp-Spreading pressure method, whereas the peroxide value increased to a maximum and then decreased with increasing oxidation. Thus far only three methJ: have been found to possess at least the major desirable characteristics of a useful method for measuring fat oxidation; namely, (a) the peroxide value, (b) the fat-aldehyde value, and (c) oxygen absorption. Peroxide value: The peroxide method is based upon the measure- ment of peroxides by titration of the iodine freed by the fat in a chloroform-acid solution. Taffel and Revis (120) utilized a method for measuring the peroxides in fat based upon the reaction presented by Lea (67) and perfected by Wheeler (124) into the present commonly used peroxide method. Greenbank and Helm (47) used a modification of this method. Much use has been.made of this method to detect oxidative changes of fats and oils (3, 33, 37, 52, 53, 63, 68, 69, 70, 71, 91, 97, 99, 101, 102, 111, 115, 117, 124, 125). Several workers have found this method to be a desirable means of determining the degree of oxidation of fats (s, 10, 23, 46, 62, 72, 96, 98, 100, 116, 120). The early work of Taffel and Revis (120) and Wheeler (124) reveals that reproducible results can be obtained ‘with this method. Wheeler (124) found the deveIOpment of peroxides to proceed any appreciable change in iodine number of cottonseed or corn oil during oxidation. Briggs (10) found that during the induction period of butter fat, changes occurred which resulted in the formation of small amounts of peroxides. It was concluded that the estimation of these peroxides - 11 - gave the best indication of the extent of oxidation. Stebnitz and Sommer (116) observed that the production of peroxides ordinarily precedes the appearance of a tallowy flavor in the oxidation of butter fat; however, the tallowy flavor was not always apparent at the same peroxide value. These investigators believe the peroxide value to be an excellent criterion of the degree of oxidation, eSpecially to detect the end of the induction period. Ritter and Nussbaumer (100) concluded that the best method of determining the keeping quality of butter fat is to determine the time required for the peroxide content to reach a value from five to ten. Lowen et al (72), working with fish oil, observed a close correlation between increases in peroxide value and the develoPment of off-odors. Bird (8) found peroxides to develOp in fish liver oil at the same time the vitamin A is destroyed and concluded from this that either a peroxide or vitamin A determination.was a satisfactory index of oxidation. Others have found limitations of the peroxide method (4, 10, 21, 32, 72). Dahle and Josephson (32) found no relationship between the peroxide value of stored butter and the score of the butter. Barnicoat and Palmer (4) concluded that the peroxide test was of little value when applied to butter. Studies of Briggs (10) indicates that the development of peroxides do not correSpond well with the absorption of oxygen. Stebnitz and Sommer (116), Clark and Rugg (21), and Wheeler (124) point out that the peroxide value goes to a maximum and then, upon further oxidation, decreases. Fat-aldehyde method: Another determination of fat stability that has received considerable attention recently is the fat-aldehyde test, devised by Schibstead (108). This method is based upon the color intensity produced by rosaniline hydrochloride combining with a fatty glyceride aldehyde through the $02 linkage. Schibstead (108) found considerable correlation between the oxygen absorbed by butter fat oxidized at 100° c. and the fat aldehyde value. These findings are presented in the following table: Oxidation of butter fat at 100° c. Duration of Oxygen absorbed __:Fat heating per gram of aldehyde (hours) butter fat value 0 0 1.2 5 14 740.0 10 20 890.0 15 30 965.0 20 38 257.0 25 - 258.0 30 40 106.0 35 - 38.0 Extreme oxidation in this case resulted in decreased aldehyde values. Lea (70) states that from the theoretical standpoint the fat aldehyde test should provide the best measurement of oxidation. This method has been used by numerous workers in fat oxidation studies (4, 70, 108, 125). Wiley (125) found fair correlation between the aldehyde values and the peroxide values of stored butter but decided that the aldehyde value was the most sensitive in detecting oxidative changes. In his work with the oxidation of stored butter, this worker observed that, on the average, low oxidation values were secured from high score butter and high oxidation values for low score butter, but that individual trials showed wide variations in fat- aldehyde values. Barnicoat and Palmer (4) determined the fat-aldehyde test to be only an empirical test, accurate to within about five per cent. Oxygen absorption methods: Many investigators have used oxygen absorption methods of determining the stability of fats (10,37, 39, 44, 46, 52, 57, 74, 77, 79, 87, 108). Overman et a1 (87) in their studies of storage butter found a definite relationship between the quality of butter and the rate of oxygen absorption by the butter fat. Greenbank and Holm.(44) devised a method, which was later improved by Mattill (74) and again by French, Olcott, and Mattill (39) of automatically measuring the beginning of oxygen uptake. The latter workers (39) considered samples of fat to have equal stability if a difference of not over one hour induction period time existed between the two samples. Evans (37), Johnston and Frey (57), Monagahn and Schmitt (79), and Heyer, Kass, and Burr (77) have used the Barcroft-Warburg apparatus to study the oxygen absorption of fats and oils. This method is essen- tially the method devised by Greenbank and Holm.(44) with the exception that a constant temperature water bath is used. The samples are mechan- ically agitated and the volume of oxygen absorbed is measured accurately with carefully graduated manometers at definite time intervals. Recently, Johnston and Frey (57) made a Special study of this method of measuring the induction period of fats, especially at temperatures between 50° C. and 1100 C. From their extensive investigations, these men concluded that it was possible to measure, with considerable pre- cision, the induction period of fats at any temperature within the range studied. A comparison of the induction period of sesame and cottonseed oils indicated that results obtained would agree within one to three per cent. Vibrans (123) believes that this method is useful in research work but will not be generally used as a routine method for determining the relative stability of fats and oils. IV. Influence of Heat Treatment Upon Oxidation. Butter Oil: Extensive investigations have been carried out to determine the influence of a high temperature heat treatment upon the oxidation of numerous fats and oils (43, 63, 96, 103, 109). Greenbank and Holm (43) found that heating butter fat to 100° c. reduced its keeping quality; the reduction in keeping quality being preportional to the length of time the fat was exposed to the high temperature. Schulz and Storck (109) heated butter oil to 90° C. and found that the oil thus treated kept well in storage for two years. Kochling and Taufel (63) and Ritter and Nussbaumer (96, 98) observed that the heating of butter fat from.l50° to 250° 0. reduced its induction period. Ritter and Nussbaumer (96, 98) report a decrease in initial peroxide value of stored fat when heated to 1950 to 2500 0., whereas, storage of the heated fat caused marked increases in peroxide value. They suggest that high temperatures destroy the antioxidant preperties of the fat, thereby reducing the length of the induction period. Ritter and Nussbaumer (103) revealed that butter oil filtered at 100° 0. oxidized less readily than the same oil filtered at 42° c. Butter: The influence of heating butter upon its keeping quality in storage has been investigated by Patil and Hammer (88) and Ritter and Nussbaumer (101). Patil and Hammer (88) worked with ghee, a product made by heat- ing unsalted butter over a low fire until most of the water is evapo- rated from the butter, followed by filtration to remove the curd. These workers observed that ghee made using temperatures of 130° to 1400 C. kept much better in storage than other lots heated at 1100 to 120° c. -15... Ritter and Nussbaumer (101) found that at 104° 0., butter oil oxidized to a greater extent in 8 hours than did butter in 24 hours. This difference was not due to water since the results were not materially affected by the addition of water. Butter fat from prOperly cooked butter showed slower increases of peroxide number than butter fat from non-heat treated butter. The authors believe this difference to be associated with the phOSpholipid content of the butter serum which was shown by Ritter (94) to be concentrated in the serum of the butter. Milk and Cream; At present, pasteurization temperatures from 62.90 C. for 30 minutes up to as high as 148.80 C. flash are used in commercial dairy plants to pasteurize cream for buttermaking. Numerous workers have discussed the influence of heat upon milk and cream (12, 13, 16, 30, 38, 42, 48, 58, 81, 90, 93, 95, 105, 125, 128, 129). Hunziker (55) gives a complete discus- sion of the different types of pasteurizers used for commercial pasteuriza- tion of cream. Mortensen (81) and Guthrie and co-workers (48) observed that cream pasteurized at 73.90 and 76.6° C. resulted in butter of superior keeping quality to lots of the same cream.pasteurized at 62.90 C. Wiley (125), in an extensive study of the influence of pasteurization temperature upon the oxidation of stored butter, observed that heat treatment does have a defin- ite influence upon the rate of oxidation of butter in storage. Data taken from Wiley's work (125) are presented below: -16- Influence of pasteurization temperature of cream on keeping quality of butter. Pasteurizing: Fat alde-:Grade before:Grade after: Acid value :Acid value temp. ° F.*:hyde value: storage : storage :before storage:after stor. Unsalted 150 z 0.2 : 93.5 x 89.0 : 0.31 z 0.737 175 8 0003 8 9305 8 9100 8 0020 8 0020 200 x 0.0 : 93.5 : 91.5 : 0.20 : 0.20 Salted 150 8 008 8 9300 8 9005 8 0032 8 0042 175 x 0.13 : 93.5 : 91.5 : 0.22 z 0.22 200 8 0008 8 9300 8 91.5 8 0022 8 0022 * All heat treatments were flash. These results show that the oxidation of the butter in storage, as measured organoleptically and by the fat-aldehyde value, varied inversely with the temperature of pasteurization. The marked increase in acid value of butter made from cream pasteurized at 65.50 C. led Wiley (125) to suggest that the enzyme lipase may be responsible for the greater susceptibility of the fat toward oxidation.when the lower pasteurizing temperature was used. Wilson and associates (128) later substantiated the findings of Wiley (125) in their studies with the Reid flash, Cooney flash, and vat pasteurizers. Platon and Olsson (90) observed that flash pasteurization at 90° to 96° 0. gave butter with a higher keeping quality than pasteurizing at 80° to 83° c. In a further study of the influence of pasteurization upon the oxidation of butter in storage, Wiley (125) compared the effect of pasteurizing sweet, ripened and acidified creams at different tempera- tures. The results of these experiments ShOW’that the higher tempera- ture heat treatment reduced the rate of oxidation of the butter in storage. -17- Brown (12, 13), Fabricius and Bird (38), and Wilster (129) have made studies of a vacuum processed cream using a machine known as the "Vacreator." Generally observations have been that butter made from cream pasteurized by this process, in which the cream reaches a tem- perature from 88° to 96° C., kept better in storage than did butter from the same cream vat pasteurized at a lower temperature (38). Wilster (129) reports that at the end of one month storage,the butter from vacreated cream had depreciated 14.8 per cent less in score than butter from the same cream vat pasteurized at lower temperatures. Furthermore, after four months storage this same butter from vat pasteurized cream had depreciated 37.0 per cent more in score than that from cream processed at higher temperatures in the vacreator. A summary of the data presented by Fabricius and Bird (38) shows a significant difference in flavor score of fresh butter from vacuum pasteurized cream over the vat pasteurized cream. This same differ- ence existed after storage at -17.700 C. for six months. Roberts, Coulter, and Combs (105), using a laboratory high temp perature pasteurization apparatus, pasteurized cream at 126.60 C. flash and in a vat at 71.10 C. for 30 minutes, and observed no appre- ciable difference in the keeping quality of the butter. Several workers have noted that high temperature pasteurization of milk prevents the formation of oxidized flavor in milk (16, 30, 42, 58, 95). To enable a more complete understanding of the increased stability of milk toward oxidation Gould and Sommer (42) and Josephson and Dean (58) have carried on extensive investigations of sulphide production. In their extensive study of the effect of heat upon the cooked flavor of milk, Gould and Sommer (42) revealed that sulphide -18- liberation takes place in cream held for three minutes at 660 to 680 C. It was found that as the temperature of milk was increased above 800 C. the tendency toward development of oxidized flavor was less pronounced while heating to temperatures above 84° C. almost entirely prevented the develOpment of oxidized flavor. Cooked flavor did occur at these higher temperatures, the intensity of which depended upon the tempera- ture of heat treatment. V. Influence of Certain metals With and Without Heat upon the Oxidation of Butter Fat. That metal ions in milk, cream, or butter will catalyze chemical deterioration of the products with the production of fishy or oxidized flavors has been long realized (56, 107). The exact reaction involved in this acceleration of oxidation is unknown, although one general suggestion given by Hunziker (55) is that metals may act as oxygen- carriers forming metallic oxides which come into contact with the un- saturated fat giving up oxygen. Metals in milk and cream: Davies (34, 35), Dahlberg and Carpenter (29), Krauss and washburn (65), and Rice and Miscall (92) have investi- gated the metal content of various dairy products. Davies (35) found that normal milk contained about 0.5 ppm. of cOpper whereas pasteurized milk (34) varied from 0.4 to 4.0 ppm. cepper, with most of the samples examined ranging from 0.6 to 1.3 ppm. of cepper. The iron content of milk varied from 1.7 to 3.8 ppm. Dahlberg and Carpenter (29) found normal milk to contain 0.131 ppm. cepper whereas the iron content was 0.379 ppm. Krauss and Washburn (65) studied the seasonal variation of copper and iron finding from 0.14 to 0.17 ppm. of cepper and from 0.34 to 0.53 ppm. of iron. The variations between the figures presented by different workers may be partly due to different analytical methods as well as to different sources of the products examined. Rice and Miscall (92), present data to Show that the metals do not go with the fat itself, although Davies (35) noted that with special methods of separation there were higher concentrations of metals in the cream. Davies (35) attributes this to the adsorption of metals by the complex proteinates at the fat-globule surface. Effect of heat and_metals on milk and cream: Barnicoat and Palmer (4), Brown, Thurston, and Dustman (14), Gould and Sommer (42), Hunziker (55), and McFarland and Burgwald (76) have studied the in- fluence of heat treatment of milk and cream upon metallic-induced oxidation. Brown and associates (14) and Gould and Sommer (42) have found that cepper added after pasteurization is more effective in producing an oxidized flavor in milk than adding the cepper before pasteuriza- tion. Gould and Sommer (42) found that when milk containing 1.4 ppm. cepper was heated it was necessary to use temperatures of at least 84° to 86° C. momentarily in order to have a marked retardation of oxidized flavor development, whereas normal milk showed the same in- fluence at 76° to 78° c. Brown et a1 (14) are of the opinion that the cOpper combines with the proteins and in this way is inactivated as a catalyst. Gould and Sommer (42) suggest either that the cepper may combine with the protein that liberates the heat-labile sulfur prevent- ing the liberation of the sulphides at the temperature of sulphide liberation in normal milk, or that possibly the cOpper may combine with liberated sulphides preventing the cooked flavor from being noticeable until greater amounts of sulphides are liberated. working with frozen cream, McFarland and Burgwald (76) found they could pre- vent the deve10pment of oxidized flavor during storage of frozen cream, even in the presence of cepper, by heating to 77.70 C. for 5 minutes. Metal content of butter: Several workers (35, 46, 54, 59, 127) have studied the metal content of butter. Davies (35) found that butter made in the laboratory contained about 0.5 ppm. c0pper and 1.0 ppm. iron. It was found by Davies (35) that the amount of metals entering butter from cream is dependent upon the curd-nitrogen content of the butter. Keilling (59) determined the cepper content of commercial pasteurized butter and found it to range from 0.76 to 7.0 ppm. The cepper content of first grade New Zealand butter was found by Williams (127) to vary from 0.2 to 0.25 ppm. with the iron content ranging from 0.5 to 1.0 ppm. Horrall and Epple (54) found commerical butter to range from 0.14 to 4.0 ppm. c0pper, depending upon exposed capper equipment. Effect of heat and metals on butter and butter oil: Barnicoat and Palmer (4) and Ritter and Nussbaumer (98) have investigated the influence of metals upon the oxidation of butter. Barnicoat and Palmer (4) observed that the addition of metals (0.05 to 0.25 ppm. Cu) to butter granules before working caused no significant increase in oxidation of the butter. These workers are of the Opinion that it is the oxidation of the lecithOprotein membrane surrounding the fat glob- ule that causes the accelerated oxidation in the presence of metals. The addition of cepper to oxidized butter was found by Ritter and Nussbaumer (98) to inhibit further peroxide formation eSpecially in -21.. the presence of heat and light. Considerable attention has been given the influence of metals upon the oxidation of butter oil. Barnicoat and Palmer (4), Ritter and Nussbaumer (97), and Briggs (9) have made studies of the influ- ence of metals upon oxidation. In other studies Battie et a1 (5) and Smedley-MacLean and associates (110) found catalytic influences when investigating the effect of cupric salts upon the oxidation of certain fatty acids. Battie and co-workers (5) observed that the action of cupric salts are only effective when the substance to be oxidized is an acid. In an alkaline or neutral solution of succinic acid little oxidation takes place. Briggs (9) found that metallic ions hastened oxidation of butter fat at 100° c. In their extensive study of butter, Ritter and Nussbaumer (97) observed that butter containing 1 ppm. cOpper showed greater peroxide formation than the same butter with higher concen- trations of c0pper when heated for eight hours at 104° C. Presumably, the cepper not only hastens the formation of peroxides, but also accelerates the rearrangement of peroxides into acids, aldehydes and other substances. The addition of more copper to the oxidized fat further decreased the peroxide value. In addition, these workers (97) observed that cepper accelerated the oxidation of butter fat stored in the dark whereas, the same cepper-contaminated fat in the light, showed slower peroxide formation than the control fat. Later (97), it was observed that the peroxide content of the COpper contamv inated fat decreased in Spite of light, whereas the peroxide value of the control fat continued to increase. - 22 - VI. Antioxidants. Different fats vary in their stability toward oxidation, the variation being partly due to natural impurities present in the fat which tend to inhibit oxidation. These catalytic inhibitors are generally referred to as antioxidants. Theories of antioxidant action: Maureu and Dufraisse (82) and Christiansen (20) were instrumental in deve10ping explanations of the action of antioxidants. The theory of Maureu and Dufraisse (82) involves the formation of peroxides both from the peroxides of the oxidized fat and from the antioxidant. During the same reaction the oxidized fat is converted into a lower oxide. The antioxidant per- oxide and the fat oxide then reacts to give the original unsaturated fat, the antioxidant, and oxygen. According to this theory, the antioxidant should act as a true catalyst, never being used up in the reaction. Christiansen (20) presents a theory in which the antioxidant is used up or destroyed during the reaction. French, Olcott, and Mattill (39) also indicate that the antioxidant is destroyed during the induc- tion period or at least it is not effective during the period of rapid oxidation. In line with this theory, Christiansen (20) believes there is a progressive chain reaction with the oxidizable fat reacting with the active oxygen forming a peroxide. Energy is liberated which is then passed on causing progressive reaction until thousands of mole- cules may become involved. In case an antioxidant is present, the energy liberated from the formation of the peroxide is absorbed by the antioxidant. Likely, the antioxidant is itself oxidized due to its increased energy level. Since the oxidation reaction of the anti- - 23 - oxidant does not liberate energy, the energy chain reaction ceases. This eXplanation would account for the extended induction period in the case of fats containing antioxidants. Christiansen's (20) theory indicates that an antioxidant must be a substance easily oxi- dized. Evans (37) suggests that antioxidants may combine with some of the oxidation catalysts present such as metals, thus preventing their accelerating effect upon oxidation. French et a1 (39) believe that all antioxidants act in the same manner since none are active after the end of the induction period. Acids: Numerous acids have been found by Bird (8), Dagneaux (28), Greenbank and Holm (47), and Lea (69) to Show inhibitive effects toward the oxidation of fats and oils. Extensive investigations by Bird (8) revealed that hydrochloric acid as well as dry hydrogen chloride gas greatly reduced the induc- tion period of halibut liver oil. PhOSphoric acid was found to re- duce the induction period slightly whereas acetic acid had no effect. Other acids studies were maleic, citric, tartaric, furmaroic, suc- cinic, sulfuric, and ascorbic. All these showed considerable inhibi- tive effect upon the oxidation of halibut liver oils, while benzoic acid showed little influence. Greenbank and Holm (47) have shown that some unsaturated polybasic aliphatic acids, eSpecially maleic, are excellent antioxidants. Dagneaux (28) found maleic acid to be an effective antioxidant in powdered milk. In studies with lard, Lea (69) found lactic, glycollic, and maleic acids to be moderate anti- oxidants whereas the polybasic hydroxy-acids, such as tartaric and citric, were powerful antioxidants. Malonic and phOSphoric acids -24- were also found to be effective. Salts: One would expect the action of salts to be closely allied to the catalytic effect of the metal ion of the salt. Bird (8) and Lea (69) have studied the influence of salts upon the oxida- tion of fats. The work of Bird (8) with fish viscera oil indicates salts can- not be used as antioxidants, since all salts, except sodium sulfate, Showed destructive effects on the induction period. The degree of reduction of the induction period depends upon the metal ion of the salt, with cobalt, cepper, calcium, magnesium, tin, and iron being the most destructive. Lea (69) found that sodium citrate and sodium malonate acted as powerful antioxidants in lard. .nggg: Alkalies are generally believed to cause a decreased stability of fats. Bird (8) found that an aqueous solution of alka- lies materially reduced the induction period of halibut liver oil, but that dry alkalies had no destructive effect. This author believes that dry alkalies do not destroy the natural antioxidants present in oil, thus indicating that the destruction of the antioxidant sub- stances is associated with the aqueous phase. Proteins and amino acids: Briggs (9) and Lea (69) have noted that proteins exert antioxygenic influences upon the oxidation of fats. Lea (69) has demonstrated that some of the aliphatic amino- acids (glycine, aSparatic and glutamic acids, and aSparagine) are powerful antioxidants. Corbett and Tracy (25) observed that when condensed skim.milk was added to cream to make a milk of four per cent fat, the tendency to develOp an oxidized flavor was reduced. These workers suggested that some antioxidant was liberated from the milk protein during the condensing process which acted to prevent oxidation. Later Corbett and Tracy (26) found tyrosine and the more soluble esters of tyrosine to be effective in preventing oxidized flavor in milk in concentrations of 0.02 to 0.04 per cent. Normal amyl ester of leucine was also an effective antioxidant whereas glutamic acid gave no protection. Both of these compounds produced objectionable off flavors. Phospholipids: Dagneaux (28), Dahle and Nelson (33), and Ritter and Nussbaumer (99, 101, 103) have observed that certain phospholipids exert a retarding influence upon the oxidation of fats. Lecithin was originally believed by Evans (37) to have an anti- oxygenic effect. Later, it was found by Ritter and Nussbaumer (101) that lecithin had little or no retarding effect upon oxidation; but that cephalin, an impurity in most commercial lecithin, showed strong antioxidizing action. Ritter and Nussbaumer (99) observed that butter ph03phatides from.buttermilk powder retarded oxidation of butter oil. Further investigations by these workers (103) reveakxithat filtering butter at 100° 0. gave butter oil with higher keeping qualities than filtering at 42° c. This difference was thought to be due to the regulation of the amount of phOSpholipids in the butter oil carried by the water content of the filtered butter oil. Thurston et al (121) believe lecithin to be reSponsible for the oxidized flavor of milk rather than the butter fat, and Swanson and Sommer (119) found the phOSpholipids to be oxidized when the milk possessed an oxidized flavor. In an attempt to determine the active fraction of cat and soy bean flour Dahle and Nelson (33) found that the phOSpholipid fraction - 25 - showed considerable antioxygenic prOperties. Carotene: Considerable discussion has been given the influence of carotene and vitamin A upon the oxidation of fats and oils. Bird (8), Dagneaux (28), Monagahn and Schmitt (79), and Olcovich and Mattill (86) have made studies of the influence of carotene and vitamin A upon the oxidation of fats and oils. Monagahn and Schmitt (79) found that carotene inhibited the oxygen uptake of linoleic acid during the induction period, but after oxidation began, the oxygen uptake was accelerated. Dagneaux (28) states that carotene acts as a slight inhibitor toward oxidation. Con- trary to these findings, Olcovich and Mattill (86) found carotene to act as a pro-oxidant and Bird (8) observed that vitamin A acts as an oxidation catalyst in fish oils, the accelerating effect depending upon the concentration of the vitamin A. Enzymes: Recently, enzymes have been used as a means of prevent- ing oxidation of milk. Anderson (1), Corbett and Tracy (26), Doan and Miller (36), and Nelson and Dahle (85) report the use of trypsin to prevent the development of oxidized flavor in milk. Doan and Miller (36) suggest that the enzyme acts upon the milk proteins liberating reducing substances. Contrary to this Opinion, Nelson and Dahle (85) believe the protective action comes from the action of the enzyme upon the fatty materials rather than upon the milk proteins. Avenex: More emphasis has been placed upon the use of Avenex, an oat flour product, than upon any other antioxidant used to retard the oxidation of fats and oils. Investigations by Peters and Musher (89) indicate that the oxidation of lard, cottonseed oil, castor Oil, corn oil, and soybean oil can be retarded by Avenex. Conn and Asnis (23) - 27 - Show that it is possible to inhibit oxidation of the fat on potato chips by dusting them with five per cent of their weight with Avenex. Lowen and co-workers (72) observed oxidation of fish fats to be re- tarded by oat flour, but not to as great an extent as in the case of lard or vegetable oils. During their work with frozen cream, Maack and Tracy (73) observed that an oat flour content of 1.5 per cent protected frozen cream from develOping an oxidized flavor. Corbett and Tracy (24, 26), Corbett et a1 (27), and Dahle and Josephson (32) have studied the application of Avenex to cream to prevent develOpment of oxidation in butter. These workers found that cat flour preparations retarded the onset of oxidation, but did not prevent it. Corbett, Tracy, and Hansen (27) observed that adding Avenex concentrate to sweet cream at the rate of 0.15 to 0.3 per cent of the fat retarded the develOpment of oxidized flavor, but did not prevent it entirely. They noted further, that the addition of Avenex concentrate to second grade cream.was of more advantage in improving the score than adding it to first grade cream. Combs, Coulter, and Whitman (22), Dahle and Josephson (31), and Koenig (64) have compared the use of avenized parchment with standard parchment wrappers for protection of butter prints from oxidation. These workers found that the avenized parchment exhibit protective action against oxidative changes eSpecially at the surface of the butter. Dahle and Nelson (33), Musher (83), Ritter and Nussbaumer (99), and Schulz and Storck (109) found that cat flour preparations added to butter oil retarded oxidation of butter oil in storage. The following data are taken from Musher's work (83): _ 28 - Effect of Avenex in preventing oxidation of butter oi1.* Treatment Stability (hourS) contrOI butter Oil 000000oooooooooooocooooooooo 5 Plus 1% Avenex ooooooooooooooooooooooo000000000 9 Plus 2% Avenex 0.0000000000000000.0000000000000 1005 Plus 5% Avenex cocoa.coo...00000000000000.0000. 1400 * As determined by fat stability apparatus. The butter oil used in the above trial shows an abnormally short induc- tion period as measured by the method used. However, the same general trend may be eXpected with normal butter oil. SCOPE OF INVESTIGATION Because of the need of further information regarding factors that influence the oxidation of butter fat, this study was conducted with the following intentions: 1. 2. 5. 4. 5. To investigate the methods of determining the stability of fats. To study the influence of heat treatment of (a) cream, (b) butter, and (c) butter oil upon the stability of the fat to oxidation. To determine the effect of metals with and without heat in (a) cream and (b) butter oil upon the ability of the fat to withstand oxidation. To ascertain the influence of the antioxidant, Avenex, when added to cream and to butter oil. To study the destruction of carotene with oxidative changes. Since there is little available information concerning the relia- bility of the fat stability and hot air oven methods, extensive stud- ies were made to determine the accuracy of the methods as well as a comparison of the two methods. Also, extensive investigations were carried out upon the influence of various heat treatments upon oxida- tion while only minor studies were made of Avenex and carotene. EXPERIMENTAL PROCEDURE Since many of the laboratory studies were of factors influencing the oxidation of butter oil, it was necessary to standardize methods of preparation in order to eliminate, in so far as possible, all variable influences upon oxidation. Source and general treatment of milk and cream; Raw milk used for this work was obtained either from the College Creamery or directly from the College Farm. In instances where raW'milk was separated, the milk was secured from the College Farm with particular care being taken to prevent metal contamination. The raw milk was centrifugally separa- ted into cream testing from.30 to 40 per cent fat. This cream was either immediately used for oxidation trials or stored at about 3° C. until used. Where pasteurized cream was desired and slight metal cone tamination was of no significance, sweet pasteurized cream separated from mixed-herd milk and pasteurized at the College Creamery was ob- tained. All raw cream used in any trial was pasteurized prior to churning at temperatures of at least 62.80 C. for 30 minutes. Methods of securing and_purifying_the butter oil: Cream was churned in motor-driven, two-gallon Daisy churns. These churns were well tinned, and agitation was supplied by means of a wooden paddle. After the buttermilk was drained, the granules were washed either with cold tap water or with distilled water depending upon the significance of metal contamination. When butter was desired, the granules were worked by hand with a wooden paddle into a homogeneous mass and placed into pyrex glass beak- ers. However, if only the butter oil was desired, the granules were melted in a water bath at 500 C., the water and curd were siphoned off, -U1- and the butter oil centrifuged and filtered. Filtering was attained in an incubator held between 40° and 500 C. In case the oil was not to be studied immediately it was refrigerated at 3° C. Heat treatment of cream, butter, and butter oil: Laboratory pasteurization of cream to temperatures below that of boiling water was accomplished in a round-bottomed glass flask suSpended in a water bath, and using a mechanical glass agitator to stir the cream. To attain temperatures above 1000 C., cream in 2 liter Erlenmeyer flasks, st0ppered with cotton plugs (600 ml. cream per flask), was heated in the autoclave for 30 minutes at the desired steam pressure after allowing 15 minutes for the autoclave to reach the desired temperature and pressure. Cooling was then accomplished by submerging the samples in cold water. In every case the cream was aged in the same container in which it was heated for a Specified number of hours before churning. The heat treatment of the butter and butter oil to temperatures below 1000 C. was also accomplished in a water bath. Special care was taken to protect the sample from light during the heating period. To secure temperature above 1000 C., 200 ml. samples were placed into 250 ml. Erlenmeyer flasks stOppered with cotton plugs and heated in the autoclave at the desired pressure for 30 minutes. After heating, the samples were either immediately cooled to the storage temperature or, in certain cases where butter was involved, the butter was centrifuged and the butter oil filtered. Treatment and care of glassware; Laug (66) and Stebnitz and Sommer (115) have shown the necessity of using Special care in cleaning glass- ware used in fat oxidation work. For this purpose these workers used chromic acid cleaning solution (H2804.K20r207). Hot water was found by Laug (66) to be effective in removing the cleaning solution from the glass. Stebnitz and Sommer (115), in their work with the fat stabil- ity apparatus, indicate that the best results could be obtained by washing the tubes and then soaking them in cleaning solution over night, followed by several consecutive leachings in distilled water for a period of three or four days. Laug (66) showed that several successive leachings with distilled water were necessary to remove all of the chromic acid from the glass. Special precautions were taken of the glassware used for all determinations in these studies. The glassware was first washed in a strong solution of tri-sodium phOSphate, rinsed several times with hot tap water, and then held in chromic acid cleaning solution and placed into large glass jars and rinsed six or more times with hot tap water. After this, the glassware was rinsed with six or more changes of distilled water, each lot of water being allowed to re- main over the glassware for 12 hours or more. Before use, the glass- ware was thoroughly dried in a hot air oven. It was also found neces- sary to treat the rubber stOppers used in the oxidation tubes of the fat stability apparatus with a strong alkali followed by thorough soaking and rinsing in distilled water. Methods of analysis: The peroxide determination was used to measure the degree of oxidation of fats. The method used was essen- tially that outlined by Wheeler (124). To five grams of liquid butter oil in a 250 ml. beaker was added 50 ml. of chloroformpacetic acid solution. This fat solution was allowed to react with 2 ml. of 50 per cent KI solution for one minute, after which time 100 ml. of distilled water was added. The iodine liberated from the KI was then titrated -33- with standard 0.01 N. sodium thiosulfate solution using a soluble 2 per cent starch solution as an indicator. The results are eXpressed as moles of peroxides per 1000 grams of butter oil. The free fatty acids content of the butter fat was determined according to the method outlined by The Association of Official Agricultural Chemists (2). Ten grams of butter oil are heated to boiling in 25 ml. of neutral ethyl alcohol containing phenolphthalein. This hot mixture is then titrated with standard 0.05 N. NaOH solution until a pink color persisted for at least 30 seconds. The results ob- tained are expressed as acid degrees (ml. of 1N. NaOH per 100 grams of butter oil), a value easily obtained by dividing the ml. of NaOH re- quired by 10 grams of butter by two. ‘ Carotene determinations were made by a modification of the method devised by Moore (80). To 10 grams of liquid butter oil “g: added 10 m1. of ethyl alcohol (95 per cent) and 10 ml. of benzine (distilled at 200 to 21.10 C.). After thorough mixing, the samples were placed into cold water for 10 minutes, being shaken three times during this period. Dis- tilled water (10 ml.) was then added to cause inversion of the phases and the samples were centrifuged for 4 minutes at 1900 ftp.m. The color of the benzine-fat protion was measured photometrically, using a one ml. sample in an Evelyn photometer with a 440 filter. The carotene content was expressed as micrograms per gram of fat. Methods of accelerating fat oxidation: An accelerated method of oxidation was used to enable rapid determination of the stability of fats. In these studies two such methods have been compared and used, namely: (a) Swift's fat stability apparatus (an aeration method) and (b) hot air oven. The Swift's fat stability apparatus was devised by King, Roschen, and Irwin (62). In this method, the fat to be oxidized was placed in test tubes and heated to the temperature of boiling water in an oil bath. A constant flow of air was maintained through each tube of fat being oxidized. As the fat reached the end of the induction period samples were periodically taken from the apparatus and tested quanti- tatively to determine the degree of oxidation. Another acceleration method used was a hot air oven. In this method definite amounts of fat in 100 ml. beakers were placed in a thermostatically controlled hot air oven having a mechanical air agitator. As the fat oxidized, samples were taken from the oven at definite time intervals and the degree of oxidation determined. - 35 - RESULTS Studies_gf theflPe:gx§de_Method The peroxide method in one or more of its modifications has been accepted as a standardized procedure for determining the degree of oxidation for a considerable time. However, since the quantity of chloroform, acetic acid, and potassium iodide required for the many determinations in this study was large and also relatively ex- pensive, studies on the possibility of reducing the amount of solvent and potassium iodide seemed eSpecially apprOpriated. Furthermore, the preparation of a fresh saturated potassium iodide solution for each group of titrations as well as maintaining a solution of con- stant strength during the determinations were other problems which seemed to indicate the desirability of using a weaker solution of this reagent if it were found feasible. Variation in solvent: To study the influence of varying the quantity of solvent used in the peroxide determination four samples of butter oil were oxidized in the hot air oven to different per- oxide contents. Duplicate peroxide determinations were then made upon these four samples by three procedures. In one, the standard method for the peroxide determination was used. This involved the use of 50 ml. of solvent. In the other two, lesser amounts of solvent were utilized; in one 40 ml., and in the other 30 ml. The results obtained are given in table 1. Table 1. Influence of volume of solvent upon peroxide values. Volume of solvent le.) Deviation (is? cent) 50 : ‘40 : 30 Ifi(2)from(1){(37from(l) (17' = Iler' : (3) z '(43 = (51' 406 3 402 3 301 3 807 3 3206 805 3 708 3 600 3 803 3 3000 13.0 : 11.9 : 8.8 : 8.5 : 32.4 24.4 : 20.7 : 17.1 : 15.0 : 29.9 From the above data it is apparent that decreasing the amount of solvent 20 per cent causes a decrease in peroxide value of 8.3 to 15.0 per cent whereas a decrease in solvent of 40 per cent resulted in about 30 per cent decreases in peroxide values. From these results it is evident that changing the volume of solvent definitely decreases the peroxide values obtained. Variations in congentration of potassium iodide solution: In order to determine the possibility of using a weaker potassium iodide solution than the recommended saturated solution, fat was examined having widely different peroxide values, using a saturated, a 50, and a 33 per cent potassium iodide solution. The results of these inves- tigations are given in table 2. Table 2. Influence of concentration of KI solution upon peroxide values. Trial 1 Saturated ; Thirty-three per cent solution solution : : : Deviation (per cent) (lml.) : 1 ml. : 2 m1. :*(2)from(l):(3)from(l) (11’ z (2) : (3) = (4) = (5) 508 3 405 3 4.6 3 2204 3 2007 11.1 3 805 3 805 3 2304 3 2304 17.0 3 13.0 3 1300 3 2305 3 2305 31.9 : 24.2 : 24.4 : 24.1 : 25.5 (contd.) - 37 - Table 2. (contd) Trial 2 *_ Saturated : Fifty per cent solution —_ solution : : : Deviation (per cent) (1 m1.) : 1 ml. : 2 ml. : (2)from(l):(3)from(ly (17 = T2) : (3) : (4) = (57 5.8 : 5.7 : 5.7 : 1.7 : 1.7 1101 3 10.5 3 1009 3 504 3 108 17.0 : 15.3 : 16.6 : 4.1 : 2.4 31.9 : 29.4 : 29.9 : 7.8 : 6 3 When two milliliter of 50 per cent potassium iodide solution were used, the per cent decrease from the control sample varied from one to two per cent at peroxide values ordinarily determined in oxi- dative studies. The decrease was slightly greater if one milliliter of the 50 per cent solution was used. In the case of the 33 per cent solution, the percentage deviation from the saturated solution was from 20 to 25 per cent. From these results it appears that 50 per cent potassium iodide solution can be used without introducing excessive errors into the results. For this reason, the peroxide determinations in all the following studies were made using two milliliters of 50 per cent potassium iodide solution in place of the saturated solution that is commonly recommended. Comparison of sensitivity of peroxide value and acid degree in detecting end of induction period: In conducting these studies on fat oxidation data were secured which permitted a comparison of two methods of measuring the degree of oxidation in the fat, namely; (a) the peroxide value and (b) the acid degree. It appeared desir- able to tabulate and analyse these data the results of which are presented in table 3. Table 3. Comparison of peroxide value and acid degree in detection of end of induction period. Peroxide Number Range in Average value of acid acid range samples degree degree 0 - 2 19 0.16 - 0.34 0.25 2 - 5 13 0.16 - 0.34 0.22 5 - 10 6 0.19 - 0.34 0.24 10 - 20 3 0.21 - 0.31 0.25 20 - 45 8 0.34 - 0.45 0.45 These results indicate, generally, no close dependable correla- tion between oxidation as determined by peroxide formation and by titration of free fatty acids. Samples showing excessive peroxide formation had considerable higher acid values, but often samples having a peroxide value of 5 to 10 showed normal acid degrees. Con- sequently, titrations for the free fatty acids do not appear to be a satisfactory method of detecting the end of the induction period. Acceleration Methods Two methods of accelerating fat oxidation were used in these studies. At the beginning of the experiments, an aeration method, developed in Swift's laboratories and commonly referred to in litera- ture as the "Swift's Fat Stability Apparatus," was used. However, because of the complexity, this method and the limitation of the nump ber of samples that could be oxidized at one time, this procedure was later replaced by a hot air oven method. Aeration Method: Oxidation is accelerated in this method by air, the flow of which is measured by a Sargent wet test meter. After passing through the meter, the air is washed with distilled water and then passed through a solution of potassium permanganate to remove any reducing substances present. After this treatment, the air is then measured from central chambers through small capillaries into the fat. From the fat, the air passes out into the atmOSphere through an atmos- pheric condenser that returns the condesible vapors to the fat. The fat (25 milliliters) is contained in 2 x 20 centimeter test tubes suSpended in an oil bath which is heated by a surrounding boiling water bath. In studying the aeration method as a means of oxidation accel- eration, trials were conducted to determine the influence of certain factors and techniques on irregularities in the results secured as well as to determine the precision of the method in detecting the end of the induction period. Temperature fluctuations of aeration method: Since variations in peroxide values had been observed between samples taken from dif- ferent positions of the oil bath at the same time there was some question as to the uniformity of the temperature of the oil bath. Consequently, one point considered was the possibility of errors in Speed of oxidation of the fat due to temperature fluctuations in the oil bath. To ascertain the extent of differences that might exist, the apparatus was allowed to heat for several hours and the tempera- ture taken at different positions in the oil bath. Results of these investigations are presented in table 4. Table 4. Temperature of oil bath of fat stability apparatus at different locations.* '_Tube number 1 4 6 9 12 15 18 Tgmperature ° C. 97.5 97.5 97.5 97.8 97.8 97.7 97.7 * Tubes numbered from left to right. - 40 - These results show that the center of the oil bath is slightly higher in temperature than either end which is likely due to the effect of the heater that is located in the center of the bath; however, such slight differences in temperature would doubtless have no marked influence upon the accuracy or reliability of the aeration apparatus. Reliability of aeration method based ongperoxide value of individual samples: In studies dealing with the aeration method of rapid oxidation, variation between duplicate samples was ascertained by determining the peroxide values of duplicate samples taken from the apparatus at the same time. These values were then arranged at different peroxide values according to the deviation and are present- ed in table 5. 0f the 163 duplicate determinations, 6.7 per cent showed no variation, a little less than one third (30.7 per cent) had a dif- ference between duplicate samples of from 0.1 to 0.5 peroxide units, whereas 28.2 per cent of the determinations showed variations between 0.6 and 2.0 peroxide units. About one-fifth of the comparisons varied by a peroxide value of 2.1 to 5.0. In all, 65.6 per cent of the dupli- cate determinations varied not over 2.0 peroxide units. However, con- sideration must be given to the fact that 65 per cent of the samples had a peroxide value ranging below 10.0 and there appears to be a tendency for the deviation between duplicate samples to increase di- rectly with the peroxide value of the samples. In order to determine if the peroxide numbers agree more closely when a relatively short induction period is obtained for the fat, the data in table 1 of the appendix were rearranged placing all duplicate .Hchomgq .H oHnme aH cmpqmmmng spec mpmHmaoo .maoHpquenmpoe omeHHase noH so Hmpoe* o.OOH u m.o u m.m " $.OH u m.om " m.mm u 5.09 « v.0 "Hmpop we R u u u u n u u u oponw b.¢H u H a m u m n m u m u H u u was H.0m o.Hm u u u v u m u mH u m n u o.omuH.0H . ¢.mH " u u m u m ” ¢H u e n H u 0.0HuH.m 11 u u u u u u u n J m.eH u n H u H u a “ HH u a u H " o.m-H.m «.mm a u H u H u H u e H mm u m u o.muo prOp Mo *wmbonm cumuo.om.H.0Hn 0.0HaH.m“ o.mcH.m " o.mao.o " nuH " o " ous> u H.om u u u u u u " muonaom mmHmSwm onOHHmso mo qupmH>oQ " *.oaHp oawm map as mzpmhommw doprnow on» Home some» monEwm opmoHHQSQ nooseop ous> oonohon no qupmeoQ .m chmB .Mchwmma .H oHan :H oopaommnm meme mponaoo .mqupqusnopoc oponHmac up Mo HmpoB * c.00H " o.o " n.H " a.» " e.HH “ n.5m ” e.m¢ " H.a “prop so a u u u u u u u m opens m.e m m m m a M H m H m m use H.om m m W ,. m N m m m H . .3: a”. m m W m m m n . .. W N 3:. . o.nH m m m H w H m n W n W W o.n-H.m m.m¢ m m H m m m e W mm W n W o.m-o Hm»0p ac mmmponm enemo.om-H.ono.OH-H.m m o.m-H.m m o.w-w.o m m.o-H.o w o m ousp " H.0N u . . u u u " mononom monEwm mpmoHHmSU mo qupmemQ *.mn:on mH asap mmoH MOM wnHNHoHNo Hopmw mnpwhmqmw qupmnow Scam some» monawm mewOHHmso ammspon mnpr ocHHonom Mo noHpmeoQ .w oHnt .HHcdomm¢ .H oHnt qH connommhm spec oponBoo .mqupmaHsnoumc opmeHHmsc am no Hence * 0.00H u 0.0 u n.m “ ¢.>H " H.mm " H.mm u «.mH n p.¢ «Have» no R u u u u u u u u obonw o.mH " u H " OH " e u H u u " one H.om ¢.¢m " u u m u e u b u n u n o.omuH.0H ¢.¢m u u u m u m u HH “ H H H 0.0HuH.n o.HN u u H u n o u o u e u H " o.n uH.m w.HH u u u H u H u u m u n u o.m :0 pro» Mo &uo>onw newno.omuH.0Huo.0HuH.m " o.muH.w " o.muo.o “ w.ouH.o " o " mfipr " H.0m u u H u u u " mconnom mmHmem omeHHmdu ho qupwaom u *.onoa no maaon mH quNHcHHo noemw mapwnwmmw noprhow Eon“ comm» monewm onOHHmsu noosuon manb ouHHonom Mo nOHpmeon .s oHpma -44- taken from the apparatus before 18 hours in one group and those taken from the apparatus after 18 hours in another group. The devi- ation between duplicate samples in each group were then tabulated at various peroxide values. These results are presented in tables 6 and 7. These two tables show that 73.1 per cent of the duplicate samples taken from the apparatus before 18 hours varied from each other by not more than 2.0 peroxide units, whereas only 51.2 per cent of these sam- ples taken from the apparatus after 18 hours oxidation showed varia- tions under 2.0 peroxide units. However, because the samples taken from the apparatus after 18 hours oxidation had, on the average, oxidized to a greater extent than those samples taken before 18 hours, it is possible that the above data would be more fairly represented by expressing them as per- centage deviation. Reliability of aeration method based on induction period: Additional studies were made to determine the precision of the devia- tion method in ascertaining the end of the induction period. For this purpose, a sample of butter oil was oxidized in the aeration apparatus and the end of the induction period determined in triplicate. Through interpolation from the oxidation—time curves, the induction period of each triplicate determination was secured. The results are shown in table 8. Table 8. Hours induction period of triplicate determinations of the same fat as measured by the aeration method.* Maximumi;nfl Trial Triplicates error number a b EL (per cent) 1 5.5 5.3 5.4 5.8 2 10.0 9.3 9.0 11.1 3 11.0 10.7 11.2 4.7 4 21.7 21.9 21.0 4.3 5 _22.4 21.1 22.6 7.1 * Peroxide value of 5 was used as end of induction period. ** Per cent error calculated from minimum induction period. Complete data presented in table II, Appendix. These trials were made with fat having an induction period range from 5.4 to 22.3 hours in length, consequently, covering any induction period normally encountered with butter fat. These data present a maximum error for the five trials ranging from 3.8 to 11.1 per cent, with an average maximum error of 6.2 per cent. From the average maxi- mum error obtained, one may expect an average variation in induction period of 0.3, 0.6, and 1.2 hours in fat having induction periods of 5, 10, and 20 hours resPectively. Hot air oven method: Different workers have made use of hot air ovens as a means of accelerating fat oxidation. In many cases however, consideration has not been given to the possibility that the accuracy of a hot air oven method may be greatly affected by various factors. Consequently, it appeared desirable to ascertain the influence of certain factors upon the rate of oxidation of fat in the hot air oven so that the results secured would be accurate, uniform, and reliable. For these studies use was made of a Cenco-DeKhotinsky triple wall, thermostatically controlled oven having temperature control to within t 10 C. The heating elements are shielded with ducts leading to the sides of the oven where the hot air enters at about six and 12 inches from the bottom of the oven. Two shelves were used, the bottom shelf being six inches and the tOp shelf ten inches from the floor of the oven. A centrigrade thermometer graduated to 1° C. was inserted through the tOp, the bulb of which extended to the tOp shelf of the oven. To insure positive air agitation, a Cenco air agitator was in- stalled in the tOp of the oven. Preliminary studies on temperature fluctuations: To determine how temperature influences may effect oxidation of fat, three samples of fat (50 ml. fat per beaker) were placed upon the floor, the bottom shelf, and upon the t0p shelf reSpectively. The oven was then regu- lated at 1000 C. without the air agitator running. At the end of 24 hours oxidation time the temperature of all nine samples was taken and the peroxide content determined. The results of these prelimin- ary investigations are given in table 9. Table 9. Preliminary trials with oven method to determine the temperature and correSponding peroxide values at different positions in the oven.* Position Sample Temperature Peroxide in oven number (0 C.) number 1 83 1.3 Floor 2 105 39.8 3 126 49.6 1 78 1.6 Bottom shelf 2 86 6.4 3 91 10.3 1 73 1.5 TOp shelf 2 77 1.7 3 80 1.8 * No air agitation - 47 - These results show that the temperature of the oven varies widely and, consequently, influences the rate of oxidation at different posi- tions in the oven. The floor of the oven is considerably higher in temperature and varies to a greater extent than any other position in the oven. Also, the bottom shelf is higher in temperature than the shelf four inches above it. Even individual samples on each shelf show several degrees variation in temperature between each other. This pre- lininary study indicates that if a hot air oven method is to be used some consideration should be given to temperature control. Wide distribution of samples on shelf: After finding such wide temperature variations in the oven, consideration was given to the arrangement of samples on the shelves to obtain maximum temperature control. In the first series of trials, beakers of fat were Spaced equal distances apart on the bottom shelf, the outside rows being against the walls of the oven as is illustrated in figure 1, diagram I. Thermometers were placed into the fat and arranged so as to permit reading of the temperature through the glass window of the inside door. Because of lack of Space, thermometers could not be placed into beakers of fat on the tOp shelf. However, empty beakers were placed on the t0p shelf in order to simulate normal Operating conditions. After allowe ing several hours for an equilibrium temperature to be established, at least three temperature readings were taken at two hour intervals. Trials were conducted with and without air agitation. The average of these readings are presented in table 10. - 4a - Table 10. Temperature of fat samples distributed widely over the bottom shelf of the hot air oven. No air Air Beaker agitation agitation number (0 C.) (O C.) 1 99.1 97.8 2 102.5 99.1 3 105.4 100.3 4 105.2 99.1 5 100.3 97.9 6 103.5 99.2 7 105.0 99.5 8 104.4 99.0 9 100.0 97.6 10 102.3 98.6 11 103.8 99.3 12 103.9 99.3 13 99.0 99.3 14 101.8 99.8 15 101.7 99.9 16 100.5 99.0 When there was no air agitation a maximum temperature difference of 6.40 C. was observed. Samples having the highest temperatures were beside the left side wall near the heating vent and beside the left side of the back wall, whereas the lowest temperatures were found near the door and along the front part of the right wall. Those tempera- tures taken while the air agitator was running indicate much closer ‘ temperature control. In these determinations, a maximum temperature variation of 3° C. was obtained as compared to over twice that value in the trials with no air circulation. Again, as before, the highest temperatures were obtained along the left and back walls whereas the lower temperatures were observed along the front and right walls. Close distribution of samples on shelf: Since the preceding study showed the samples placed around the sides of the oven varied appreciably in temperature, additional trials were conducted in which the samples were grouped compactly in the center of the shelf as is shown in figure 1, diagram II. Furthermore, since there was some question as to the influence of the presence of beakers on the t0p shelf upon the temperature of samples on the bottom shelf, this ang‘é was also given consideration. Consequently, as before, the temperature was taken with and without agitation and, in addition, with and without empty beakers on the tOp shelf. The results of these trials are presented in table 11. Table 11. Temperature of samples grouped compactly in the center of the bottom shelf. Top shelf empty Beakers on top_§helf No air Air No air Air Beaker agitation agitation agitation agitation number (° C.) (O C.) (0 C.) (o C.) 1 100.5 100.2 101.7 99.0 2 104.1 100.9 105.2 99.1 3 105.6 101.7 107.2 100.2 4 105.7 102.1 106.8 100.2 5 101.2 100.0 102.1 100.2 6 104.2 101.2 105.4 99.7 7 105.6 101.4 107.0 100.6 8 106.0 101.8 107.1 99.9 9 102.4 100.3 102.8 98.7 10 104.9 101.5 105.8 99.8 11 105.7 101.7 107.2 100.6 12 104.9 101.5 105.9 99.2 13 100.3 99.7 101.7 98.7 14 102.9 99.2 103.7 98.8 15 104.2 100.8 104.6 99.5 16 103.9 101.2 104.5 99.8 When the beakers are placed in the center of the shelf with no beakers on the t0p shelf, a maximum.temperature variation of 5.70 C. was observed without air agitation. This value is 0.70 C. less than was observed when the samples were distributed widely on the shelf. Again, however, the samples with the highest temperatures were located around the left side and back, whereas the lowest temperatures were found along the front and right side. -50- Under the same conditions with air agitation, a 20 C. maximum variation was obtained. This is l0 C. less variation than was secured when the samples were widely dispersed over the entire shelf. When beakers were on the tOp shelf there was an increase in tem- perature of approximately 10 C. throughout, both with and without air agitation. However, the temperature variation was not appreciably influenced as there was 5.50 C. maximum variation with neither beakers nor agitation as compared to 5.70 without the beakers. When the air was agitated, a 1.90 C. maximum variation in temperature was found when beakers were present as compared to 2.00 C. variation when there were no beakers on the t0p shelf. The slight tendency toward a lower variation in each instance is insignificant. From these studies it was concluded that having the beakers in the center of the shelf as in diagram II gave the most accurate results. Consequently, all studies made with the oven method utilized this ar- rangement. Temperature of samples on the tgp and bottom shelves: All tem- perature measurements presented so far were taken from the bottom shelf, because of lack of Space between the tOp Shelf and the oven ceiling. However, a brief study was made of the tOp shelf temperatures by plac- ing samples of fat into petri dishes and then inserting the thermo- meters at an angle. Samples were Spaced on both the tOp and bottom shelves as shown in figure 1, diagram III. The temperature relation- ship found in these studies is presented in table 12. -51- Table 12. Temperature of fat samples in petri dishes on tOp and bottom shelves of hot air oven. Petri dish No air Air number agitation agitation Bottom shelf (0 c.) (o c.) 1 103.7 99.6 2 105.8 99.9 3 102.5 99.9 4 105.3 99.8 5 103.3 99.5 6 103.1 99.2 Top shelf (0 0? (0W 7 98.7 98.3 8 101.4 99.2 9 97.7 97.6 10 101.3 99.5 11 98.1 97.9 12 98.5 98.2 “-m .9”‘ “-mv—--=—‘ A This data shows conclusively that the bottom shelf is normally higher in temperature than the tOp shelf. With the air agitator run- ning the bottom shelf averaged 1.20 C. higher in temperature than the tOp shelf. The same relationship existed when there was no air agita- tion excepting there was 4.70 C. difference in temperature. These additional data indicate, therefore, that when the hot air oven is to be used, consideration should be given to the fact that a temperature difference may exist between shelves which may appreciably affect the results. Difference in rate of peroxide formation on different Shelves: Additional trials were conducted to find the relation that exists be- tween the rate of peroxide formation on the different shelves and the correSponding temperature. In this study, 25 mls. of fat in 100 ml. beakers were oxidized on both shelves. After definite oxidation had taken place peroxide determinations were made on all samples. DIAGRAM I. .4 0 ®¢3C>6 C 0 ® © ® '06 66 w BOTTOA LF SE n, 61333 949C>0 43C>0 6<3C1 ‘1 O ‘b (a 3, h w, 0M GRA A! ll 3:3? ::. BOTTOM SHELF :::: :::: @@@6 TOP SHELF DIAGRAM ZZZ ®®® ®®® 307703 SHELF ”GD ©®® TOP SHEL F Figure 1. Diagrammatic arrangement of beakers and petri dishes in hot air oven. - 53 - Because the temperature determinatiomsof the fat samples on the tOp shelf were made in petri dishes and the oxidation of the fat was carried out in 100 ml. beakers, the temperatures of the samples are comparably but not exactly accurate. This results since the petri dishes were not at exactly the same position as the beakers of fat. The results of these investigations are presented in table 13. Table 13. Peroxide formation and relative temperatures at different locations in the hot air oven. ”‘1. “‘ _*m~‘ ‘— W‘B‘assmmfi‘ Tap—meg: - Beaker ‘mngmperature Peroxide Beakgr Temperature Peroxide number (0 C.) value number (0 C.) value 1 3 9906 3 505 3 17 3 9803 3 708 2 : 99.6 : 7.4 : 18 : 98.3 : 4.7 3 : 99.9 : 4.4 : 19 : 99.2 : 5.0 4 : 99.9 : 6.1 : 20 : 99.2 : 3.5 5 3 99.9 3 4.9 3 21 3 9706 3 306 6 : 99.9 : 3.6 : 22 : 97.6 : 2.0 7 3 9908 3 406 3 23 3 9905 3 205 8 . 99.8 : 5.3 : 24 : 99.5 : 3.0 9 : 99.9 : 4.8 : 25 : 97.6 : 3.3 10 : 99.9 : 3.8 : 26 : 97.6 : 2.7 11 3 9908 3 405 3 27 3 9905 3 600 12 : 99.8 : 9.8 : 28 : 99.5 : 7.1 13 : 99.5 : 3.4 : 29 : 97.9 : 2.7 14 : 99.5 : 4.4 : 30 : 97.9 : 3.1 15 : 99.2 : 5.0 : 31 : 98.2 : 3.5 16 : 99.2 : 6.9 : 32 : 98.2 : 9.4 Average : 99.6 : 5.27 : : 98.9 A: 4.37 The average of these relative temperatures Showsthe bottom shelf to be 0.70 C. higher in temperature than the t0p shelf. CorreSponding to this temperature difference, there is an average difference of 0.9 peroxide units: the bottom shelf has an average peroxide value of 5.27, whereas that of the t0p shelf is 4.37. That there is a difference between the tOp and bottom Shelves is further substantiated by the data presented in table III of the appendix. These data constitute those secured on 784 samples in which one lot was oxidized on the t0p and the other on the botton shelf of the hot air oven. The results Show the average peroxide value of all samples taken from the bottom shelf to be 8.10 whereas that of the tOp shelf is 7.16, an average difference of 0.94 peroxide value. Further treatment of these data shows 66.5 per cent of the samples from the bottom shelf oxidized sooner than those from the tOp Shelf, 27.3 per cent of the samples from the tOp Shelf oxidized before those from the bottom, whereas 6.2 per cent of the samples oxidized equally fast on either Shelf. Length of induction period of top and bottom Shelves: Further studies of the results secured with the tOp and bottom shelves of the hot air oven were carried out by utilizing the induction periods in- volved. From the induction period curve, assuming a peroxide value of five as the end of the induction period, the induction periods of the samples of fat on each Shelf were determined. A summary of these data is presented in table 14. Table 14. Hours difference in induction period of the same fat when oxidized on the tOp and bottom Shelves of the oven.* Induction Number Hours difference in induction period of _period between top and bottom shelves (hOIXI‘S) trials 0 "' 100 191 " 200 201 " 3075 10.0-15.0 10 6 3 1 15.1-17.5 17 8 9 O 17.6-20.0 21 9 9 2 20.1-22.5 11 5 4 2 22.6-above 7 4 3 0 Percent 50.0 42.5 7.5 * Complete data presented in table IV, Appendix. -55- These data Show that 50.0 per cent of the trials on different shelves varied from each other by not more than 1 hour induction period time, 42.5 per cent from 1.1 to 2.0 hours, and 7.5 per cent had an induction period variation greater than two hours. Further analysis of data from the original 66 trials, shows that 85.0 per cent of the trials from the bottom shelf had the shortest induction period whereas only 10.5 per cent of the trials from the tOp shelf oxidized first. Thus, there were 4.5 per cent of the trials oxidiz- ing with equal rapidity irreSpective of shelf. The average length of the induction period of the tOp shelf was 19.07 hours whereas the bottom shelf had an induction period of 18.06 hours, an average dif- ference of 1.01 hours. The normal difference in induction period between the two shelves is about 5.2 per cent. Even though the error between the shelves is not great, it may materially influence the results in oxidative studies. Influence of temperature upon rate and extent of oxidation: Little consideration has been given the importance of temperature in the formation and destruction of peroxides. Possibly some temperature exists at which destruction of the peroxides will be at a minimum and formation at a maximum, thus increasing the sensitivity of rapid oxi- dation methods. To determine the relative accuracy of accelerating oxidation at different temperatures, two identical Cenco-Dekhotinsky thermostatically controlled hot air ovens were obtained, each having a mechanical air agitator. The same sample of butter oil was oxidized in quadruplicate, first at 70° and 100° C., and then at 100° and 130° c. A summary of the data obtained is presented in table 15. - 56 - Table 15. Influence of temperature of hot air oven upon the rate of oxidation and upon the uniformity of results.* Average Maximum Minimum Deviation between Oven induction induction induction minimum and maximum temperature pperiod* _period* pperiod* Hours ‘rer cent 70 142.0 152.0 138.0 14.0 9.9 100 26.5 28.0 25.5 2.5 9.4 130 5.5 3.7 3.4 0.3 11.6 *Peroxide value of 5 considered to be the end of induction period. Complete data presented in table V, Appendix. As would be expected, the lower temperature considerably extended the length of the induction period. However, samples oxidized at 130° C. showed a somewhat greater percentage error (about 1.5 per cent more) than those oxidized at lower temperatures. However, the percentage deviations of induction period of the quadruplicate determinations at 700 and 1000 C. are about the same, indicating that these lower tem- peratures of oxidation had little influence upon the precision of the method in detecting the end of the induction period. Inasmuch as the lower temperatures used have no great influence upon the accuracy of the oven method, a temperature should be used with- in this range that is best adaptable to the laboratory work. However, too low a temperature extends the induction period to considerable length, consequently the purpose of the rapid oxidation method would be invalidated if too low a temperature is used. Therefore, a tempera- ture sufficiently high to bring about rapid oxidation should be used. Consequently, 1000 C. was considered best suited for the purpose of oxidizing butter fat because at this temperature oxidation normally occurs within 12 to 24 hours. Therefore, all studies with this method were made at 100°C. -57.. Effect of sample size upon rate of oxidation of butter oil in the hot air oven: Studies of the rate of oxidation of samples of different sizes were deemed desirable to determine the consideration that should be given this factor when using the hot air acceleration method. For this purpose, 10, 20, 30, and 40 ml. samples of butter oil were placed into 100 m1. beakers and oxidized in the hot air oven. Same ples were taken from the oven at definite time intervals and the per- oxide content determined. Two trials were conducted with duplicate samples being used in each trial. The results are presented in table 16 and are shown graphically in figure 2. Table 16. Influence of sample size upon rate of fat oxidation in hot air oven.* Hours Size of sample (m1.j oxidation 10 20 3‘0‘ 40 14 m 6.0 1.7 2.0 1:? ‘ 16 7.5 5.6 3.6 2.9 17 15.4 4.4 3.8 2.9 18 18.0 7.6 6.1 3.7 19 22.6 11.6 11.5 6.0 20 26.3 11.9 9.2 9.6 Average 16.1 701 600 405 * Results are reported in peroxide values. ** Values in peroxide number. These results show that the size of the sample definitely influ- ences the rate of oxidation, with the smaller samples oxidizing more rapidly than the larger ones. Obviously therefore, the size of same ple must be considered and maintained uniform in any oxidation work with the hot air oven. Consequently, in all studies to follow, 25ml. samples of fat were used. Effect of surface area upon rate of oxidation in hot air oven: Another factor, the surface area, was given consideration as a possi- -58- bility of influencing oxidation. To determine the effect of surface area upon the rate of oxidation of butter oil in the hot air oven, 25 m1. samples of butter oil were placed into glass containers having 62.5, 33.0, 23.7, 18.1, and 11.3 square centimeters surface area re- Spectively. These samples were oxidized in the hot air oven at 100°C. and the end of the induction period determined. The average results of two trials are given in table 17 and are presented graphically on figure 2. Table 17. Relationship of surface area to the rate of oxidation in the hot air oven. * Hours Surface area in square centimeters oxidation 6205 5300 2307 1801“ 11.3 13 3.3** 1.5 1.7 1.4 1.5 14 7.4 1.6 2.0 1.9 1.7 15 14.4 3.6 4.0 1.6 2.1 16 18.5 8.1 7.3 4.6 5.0 17 25.0 18.4 5.7 3.0 3.9 18 28.8 18.4 6.8 8.5 6.3 Average 16.2 8.6 4.6 3.5 3.4 * Results are reported in peroxide values. ** Values in peroxide number. As may be observed from these data, the rate of oxidation is defi- nitely dependent upon the surface area: the greater the surface area exposed, the more rapid the oxidation of the fat. From this, it may be concluded that the surface area must be kept constant when oxidizing fat in hot air ovens if uniform results are to be obtained. Therefore, in the studies to follow, 100 ml. beakers were used exclusively. - 5g - l T EFFECT OF SAMPLE SIZE Lu 3 , 0N RA 7'E 0F OX/DAT/Olv § §‘ :‘é 8 3.3 —O 3 x0 20 30 40 SIZE OF SAMPLE (MES) m EFFECT OF SURFACE AREA 3 ON RATE OF OX/DA r/o/v Y I 3 Lu 3 S a 33 a to 60 50 40 30 20 /0 SURFACE AREA (50. CMS.) Figure 2. Effect of sample size and surface area on rate of oxidation of butter 011. Reliability of oven method on basis of_peroxide value of indi- vidual samples: Studies were made utilizing the oven method to ascertain the degree of accuracy which may be expected under care- fully controlled conditions. To determine the accuracy of the method based on the peroxide number, duplicate determinations of the same fat taken from the same shelf were made. The results were recorded and the deviations between the duplicates secured. The data are shown in table 18. Of the 328 duplicate samples studied 79.5 per cent had a per- oxide value below ten. Of this group, 80.8 per cent showed a devia- tion between duplicates of less than 2.0 peroxide units, 16.1 per cent varied between 2.1 and 5.0 peroxide units, and 3.1 per cent of the same ples showed a variation between 5.1 and 10.0 peroxide units. Sixty seven of the samples, or 20.4 per cent, had a peroxide value of 10.1 or above. Of these, 52.3 per cent had a variation be- tween duplicates of less than 2.0 peroxide units. Of the total of all comparisons, 75.1 per cent showed a deviation of less than 2.0 peroxide units, 18.0 per cent varied between 2.1 and 5.0 units, 4.9 per cent varied between 5.1 and 10.0 units, and 2.1 per cent varied to a greater extent than 10.0 peroxide units. .maonesn ocwxoaem ea moaaw>** .aficnmmqq .HP manee ea pmpnommhm muse opmHQSOO .mnofipwqaanopoc opeoaamnp mm» mo proe * 0.00H m n.o W m.H m m.e m o.ma m e.me m m.nm m m.n masses we a m m m m M r M M m obopw n.¢ a a a e u e a n a m a a a one H.0m «.0H W W m m s W «a m an m n m H W o.om-a.oa m.an m m m m m an m as m ma m e m o.oaua.n m m m m a W m a W a W «.ma m m m m a m a m mm m cu m o.m -o Hapoa ac massage eaano.om-H.OH“ o.oa-H.nu o.n-mww," o.m-m.o " n-H " o a mafia» fin H.0m " . . u u u u oefixoamm _amuwmaem mpmewamdc no qupwH>om *.ne>o has won Mo gamma case 209% nexmp moamfiwm mpwoaague ummspon madam» mofixohom mo nofipww>mm .mH magma - 62 - Reliability of oven method on basis of induction period: Further information was secured dealing with the hot air oven method in detect- ing the end of the induction period. In this connection, several trials were run in which the induction period of a butter fat was de- termined in quadruplicate by placing four samples upon each the bottom and tOp shelves of the hot air oven. By assuming a peroxide value of five as the end of the induction period, the induction periods of all determinations were interpolated from the oxidation-time curves and are presented in table 19. Table 19. Hours induction period of quadruplicate determinations of the same fat as measured by the hot air oven method.* : Hours induction period : Trial : :Maximum difference number : a b c d : Hours Per cent 1 g 16.75 18.00 17.75 17.25 , 1.25 7.2 2 : 24.50 24.00 25.50 25.50 : 1.50 6.0 3 : 34.00 32.50 31.00 30.75 : 3.25 10.1 4 8 33000 33050 35.50 31000 3 2050 706 5 : 28.25 27.75 30.50 31.00 : 3.25 11.1 6 : 30050 30.00 31000 30.50 3 1000 303 7 : 26.00 25.25 27.25 27.25 : 2.00 7.5 8 : 27.25 27.75 28.00 26.25 : 1.75 6.4 9 3 25050 26050 27.00 25025 3 1075 607 10 : 27.00 27.50 29.25 25.25 : 4.00 14.7 11 : 27.75 26.50 27.50 27.00 : 1.25 4.6 12 : 25.25 26.00 27.50 26.50 : 2.25 8.5 Average 2.15 7.8 * Complete data in Table VII, Appendix These data ShOW'a maximum difference between quadruplicate induc- tion periods ranging from 3.3 to 14.7 per cent with an average maximum difference of 7.8 per cent. In terms of hours as based on the average results of this experiment, this method could be expected to vary 0.78, 1.56, and 2.34 hours reapectively for trials having induction periods of 10, 20, and 30 hours. ation than this: difference of 10.1, 11.1, and 14. 7 per cent reSpectively. However, certain trials did show greater vari- for instance, trials, 3, 5, and 10 showed a maximum Comparison of aeration and hot air oven methods: Because the studies so far have involved the aeration and hot air oven methods individually at different times and with different fats, it seemed important that a direct comparison of the two methods be made to permit the drawing of more definite conclusions. Conse- quently, the following experiments were conducted with this in view; Rate and extent of oxidation of fat in hot air oven and aeration apparatus at 1000 C.: In order to determine the relative Speed with which fat is oxidized by the hot air oven and aeration methods at the same temperature, samples of butter fat were oxidized by the two methods using the regular standardized procedures. The averages are presented in table 20. Table 20. Relative rate of oxidation of fat by hot air oven and aeration methods. * Hours induction period** Trial Oven method Aeration method Difference (1) from (2) number (1) (2) Hours Per cent 1 19.50 22.00 2.50 12.0 2 18.00 18.00 0.00 0.0 3 22.00 23.00 1.00 4.4 4 11.25 16.00 4.75 34.8 5 10.75 15.00 4.25 33.0 6 15.75 21.00 5.25 28.6 7 14.00 17.00 3.00 19.4 8 13.75 21.00 7.25 41.7 9 11.75 18.25 6.50 45.3 10 17.50 24.25 6.75 32.3 11 13.75 19.25 5.50 33.3 * The average of columns (1) and (2) was used to calculate the per cent difference. ** Peroxide value of 5 used to determine and of induction period Complete data in Table VIII, Appendix. These data show the induction period of fat in the hot air oven to be an average of 4.25 hours shorter than that of the same fat oxi- dized by the aeration method, an average of 25.7 per cent more oxida- -64... tion. As may be observed, this increased rate of oxidation in the case of the hot air oven is not constant and ranged from 0.0 to 43.3 per cent. However, seven of the eleven trials gave from 19.4 to 43.3 per cent faster oxidation in favor of the hot air oven. One reason for the aeration method having a longer induction period may be due to the use of saturated air to aSpirate the fat. To study this possibility, trials were conducted in which a portion of the fat-containing tubes in the apparatus was aSpirated with air pre- viously dried by passing it through sulfuric acid. The remaining tubes were aerated in the usual fashion by the use of saturated air. The average results of a typical trial (trial 1) are presented in table 21. Table 21. Influence of drying the air upon the rate of oxidation of fat by aeration method. * Peroxide values Hours Saturated Dry oxidation air air 12.0 1.1 1.5 16.5 1.5 2.8 17.0 1.7 8.1 17.5 --- 14.2 19.5 5.8 --- * Complete data presented in table IX, Appendix These results indicate that the rate of oxidation is acceler- ated from 10 to 15 per cent by drying the air used for aSpirating the fat. One additional observation made was that the drying of the air possibly may present a means of obtaining greater precision with this method. More data are needed, however, before definite conclusions may be drawn in this connection. - 65 - To study further the rate as well as the extent of oxidation, a sample of fat was oxidized both by the oven and aeration methods, samples being taken at definite time intervals to determine the degree of oxidation. The results of two trials are presented in table x of the Appendix, and data of a typical trial (trial 1) are protrayed in figure 3. These curves show, again that fat in the hot air oven oxidizes before that in the aeration apparatus. However, once the oxidation starts in the aeration apparatus it proceeds rapidly while the oxi- dation in the hot air oven shows a gradual progressive rate of per- oxide formation. At the end of 48 hours there were no more samples in the hot air oven so the extent of oxidation in this case could not be studied further. However, in the case of the aeration method samples were taken for a period of 144 hours. In this case the per- oxide content increased to a maximum at about 70 to 90 hours and then, upon further oxidation, showed marked decreases. Precision of hot air oven and aeration method in measuring the end of the induction period: In order to obtain more definite infor- mation as to the precision of each method in detecting the end of the induction period, curves were plotted of the induction periods of 11 duplicate determinations obtained with each method. From these curves, the length of the induction periods of each duplicate determination, to the nearest half hour, was secured. The results are presented in table 22. -55- Om. .33on nope can dogged hp Go .3359 .8 8300.33 mo 9838 one opwm 3.0.1» NSC \O§hDF‘ O501¢¥OJRDFJ O>U1¢DOQRDFJ 0)U1#>OQhDFJ O>UI#>OJBDFJ ‘1.8 10.6 1.2 127.0°C.-l5 min. 3.8 8.6 9.5 9.9 18.7 19.2 127.00C.-30 min. 7.3 13.9 14.1 ’14.? 18.5 23.1 .. .HHo hoppsp mo neapmcaxo moms agape Mm doHpmNszovmmm mo manpsHQQEMp we adamrwanH .n oasmfim MUZNDOMW 20R YQxXO 0 m V m. N \ Q \0 <1 d \ \ 7.. < ”a . a h. / 7 U _ 3 A s W n 3 thb>§¥ mx-.bom.m0\ §$~ Om. n .U o®.NW I .39 30/XOUJd I Q \ 3/77VA cl IQ V ole Mo alluN \ 0|. \ -79.- holding temperature and the c0pper was added and thoroughly mixed with the cream. The cream was then aged for 12 hours before churning. The average values of four trials are given in table 29, and are illustra- ted by figure 8. Table 29. Influence of heat treatment of cream upon metallic induced oxidation when metal contamination occurs after pasteuriza- tion. : Peroxide value Oxidation : Control : 5 ppm. copper segpence :62.8°-3Omin.:62.8U-30min.:85U - flash:90.6U-flash l : 2.8 : 7.3 : 3.4 : 3.2 2 3 301 3 900 3 402 3 308 5 3 409 3 9.9 3 703 3 507 4 3 609 3 1300 3 8.7 3 709 5 3 1207 3 2103 3 1006 3 909 * Complete data presented in table XV, Appendix. These data show the addition of metals to cream after pasteuriza- tion to be effective in accelerating oxidative changes regardless of the temperature used for pasteurization. However, the cream heated to 85.o°c. (185.0°F.) and 90.600. (195.o°F.) show greater stability against metallic induced oxidation than the samples of cream heated only to 62.800. (145.00F.). Nevertheless, none of the samples having metal contamination showed as much stability as the control samples. The samples pasteurized at 90.60 C. (195.00F.) showed the greatest stabil- ity of those samples having metal contamination and was only slightly less resistant to oxidation than the fat from the control lot. - 89 - .aOapouHadopmoq moans eoaasooo nofipmeflEwpnoo Hopes neg: Hfio happen mo newpmpfixo non: Somme Mo pdospmonp pee: mo monogamnH .m oaswfih mo 2g 8 m. to R was. 0 Q\ l , z - 6.0.00. .39 <§ QM. I.U oQNb I 5U Olluo\ QN ‘Q J/77V/l 30/X0é7’3d \. I'm; . Studies on Avenex Concentrate The antioxidant power of Avenex Concentrate was studieJ'when the Concentrate was added to filtered butter oil and to the cream. In studies dealing with butter oil, cream from raw milk was pas- teurized, cooled, and aged six hours before churning. After filter- ing the butter oil from the butter, five ppm. c0pper in the form of cOpper chloride solution was added to half of the butter oil. Both the control and metal containing samples of butter oils were divided into equal portions and 0.015 per cent Avenex Concentrate was added to one portion of each. To be certain of thorough mixing, 300 ml. of all samples were placed into 500 ml Erlenmeyer flasks and shaken in a mechanical shaker for 15 minutes. The butter oil was then stored for 12 hours after which time it was oxidized in the hot air oven. Aver- age results of three trials are presented in table 30. Since there was a difference of several hours between the induction periods of the normal and COpper containing samples, such a comparison cannot be made from the average data in table 30 but will have to be made from individual trials in table XVI of the Appendix. Table 30. Influence upon oxidation of adding Avenex Concentrate to butter oil. * Oxidation Normal butter oil 5 ppm. copper sequence No Avenex Avenex No Avenex Avenex 1 4.2 3.1 4.2 4.2 2 4.0 4.5 4.6 5.1 3 6.0 4.6 7.1 5.3 4 7.7 6.4 7.7 6.6 5 11.7 9.7 10.8 8.3 *Complete data presented in table XVI, Appendix. **Va1ues in peroxide number. These averages, as well as the individual trials, indicate a slight retarding influence on oxidation by Avenex Concentrate when added to both normal and cOpper containing butter oil. However, this inhibitive influence is not too definite, the samples contain- ing concentrate having about one hour shorter induction period than those with no antioxidant. Further studies were conducted in which 0.05 per cent Avenex Concentrate was added to normal cream and to cream containing five ppm. added cooper. A control sample was included in each trial. The cream was aged 12 hours after adding the COpper and antioxidant, then churned, and the oil filtered from the butter. The induction periods of these samples of butter oil were determined and the aver- age of three trials are given in table 31. Table 31. Influence upon oxidation of adding VAvenex Concentrate" to cream after pasteurization.* Oxidation __ NOrmal cream 5 ppm. copper sequence No Avenex Avenex No Avenex Avenex 1 2.0** 1.7 4.9 3.3 2 3.6 3.4 6.2 3.9 3 4.6 4.7 8.6 5.8 4 6.4 5.3 9.4 6.1 5 6.8 5.4 11.6 8.6 * Complete data presented in table XVII, Appendix. ** Values in peroxide numbers. These results show no appreciable influence of Avenex Concentrate upon oxidation when added to normal cream. However, in the presence of a metallic catalyst, the Concentrate did show an appreciable antioxi- dant action. This tendency was uniform in all three individual trials, with the Avenex-containing samples showing about two hours longer in- duction period than similar samples without the antioxidant. Carotene Destruction as Related to Peroxide Formation ‘n‘---—s---"muum~— — Since the destruction of vitamin A in fish oils has been shown to coincide closely to the peroxide formation, the possibility exists of a relationship between carotene destruction in butter fat and per- oxide formation. If a close relationship does exist, determination of the carotene content could possibly be used as a method of deter- mining the end of the induction period. Studies of the carotene con- tent of butter oil oxidized to different peroxide contents at differ- ent temperatures in the hot air oven were made. The results of these studies are given in table 32 an average of which is shown graphically in figure 9. Table 32. Comparison of carotene destruction and increases in perox- ide content of butter fat.* 100°—C. Peroxide Peroxide Peroxide value carotene value carotene4_ value carotene_ 0.0 8.14 0.0 ‘3 8.14 0.0 8:14‘__ 1.1 4.42 0.8 5.11 3.1 2.36 1.9 3.37 1.4 4.14 4.1 1.32 2.5 2.52 2.0 3.99 5.3 0.90 3.5 1.81 2.9 3.16 6.5 1.26 4.5 2.11 5.0 1.81 7.5 1.22 7.8 0.72 7.3 1.09 8.5 0.86 19.6 1.05 10.0 0.63 9.5 0.63 37.3 0.55 13.4 0.74 2.0 0.48 * Carotene expressed as micrograms per gm. fat. _— - -v' I...“ These figures and the curve show a close correlation between the destruction of carotene and the formation of peroxides, eSpecially at the lower peroxide values. resulted before a peroxide value of three was reached. Beyond this peroxide value the carotene content decreased steadily but slowly. The most rapid changes in carotene content ll..." .3... -84- No significant difference was observed between the samples oxidized at various oven temperatures. From the data obtained, it seems entirely feasible to use the carotene determination as an index for quantitatively measuring oxi- dative changes of butter fat. The carotene determination is eSpecial- ly sensitive in detecting initial changes of oxidation; for example, about 50 per cent of the carotene had been destroyed before a peroxide value of 1.5 had been reached. The oxidation, as measured by the per- oxide value, takes place slowly up to a peroxide value of 1.5. Above this value,the peroxide formation increases rapidly. Therefore, the carotene determination offers a more sensitive index of initial oxi- dative changes of butter oil oxidized in the hot air oven than does the peroxide value. However, this same relationship between initial oxidative changes and carotene destruction may not exist if the butter fat is allowed to oxidize under different conditions, such as the oxi- dation of cream or butter in storage. In all trials, the carotene content of the fat did not drOp below 0.48 micro-grams per gram of fat even in extreme cases of oxidation. This discrepancy may be due to the presence of small amounts of unoxi- dized carotene or possibly other substances that may interfere with the photometric measurement. -85- ON 2 .aofipwany ocaxoaog spas mcopoamo mo dofiposppmon Mbq v3 NQ\XQQMQ. Q\ m .m 8&3 Q Q 0/ ‘0 V N / 0 no JNJJOHVJ .1’0 SW V8.9 -0b’3//’V DISCUSSION Studies of the aeration method of accelerating oxidation of butter fat indicate that if prOper technique is exercised this method will give reliable results. Triplicate determinations of the induc- tion period of five fats having induction periods ranging from 5.4 to 22.3 hours in length revealed maximum errors in induction periods from 3.8 to 11.1 per cent. These results show even greater accuracy than reported by Freyer (40) who found that by allowing 1? per cent error in induction period, some of 11 different laboratories reported results outside this liberal range. The oven method, when variable factors were controlled, also permitted accurate determination of the induction period. Studies involving quadruplicate determinations of the same fat showed a maxi- mum difference in induction periods from 3.3 to 14.7 per cent, results comparable to those secured with the aeration method. Addi- tional data obtained by making a direct comparison of the two methods, indicate no significant difference in their accuracy. The oven method used in this study gave much closer agreement between duplicate samples than the method used by Dahle and Nelson (33). These workers oxidized butter fat in an oven at 600 C. in 50 ml. stOppered test tubes. The samples were allowed to oxidize for 15 days and samples were taken at definite intervals for peroxide determina- tions. The induction period of identical tubes of the same fat varied as much as four days; a 26.7 per cent variation. Control of factors which may contribute to uneven oxidation is necessary if an oven method is to be satisfactorily used. The factor - 87 - of temperature control is eSpecially important. The position of the samples in the oven, and the use of positive air circulation are other necessary considerations as shown by the results secured in this study. The temperature at which fat should be oxidized in an accelera- tion method is a question of importance. Too low an oxidation tempera- ture will prolong the induction period, thereby invalidating the ad- vantage of the rapid oxidation method. Although 1000 C. was found to be a desirable temperature to oxidize butter fat, different tempera- tures would necessarily be used with other fats having greater or less stability toward oxidation than butter fat. Furthermore, it may be advantageous to vary this temperature in the case of butter fat for different size samples or for dif-erences in surface area of the fat. Since the rate of oxidation of fat in the hot air oven is influ- enced by the surface area per volume of fat, the Splashing of fat on the sides of the beakers must be carefully avoided. Such a practice increases the surface area thereby increasing the Speed of oxidation. Even though the oven method is not perfect in many reSpects, it does offer certain advantages over the aeration method. The oven method appears to be as accurate as the aeration method and in certain commercial laboratories would likely be even more accurate. It requires fewer pieces of apparatus to cause contamination and the Operation pro- cedure is more simple. Furthermore, the apparatus necessary for the oven method is relatively ineXpensive and may be used in other labora- tory Operations. However, before adOpting an oven method, preliminary studies should be made by each technician to determine the control measures necessary to insure accurate results under his own conditions -88- and with the apparatus available. Individual ovens may vary greatly as to air agitation and temperature uniformity, and consideration should be given to the possibility of such variations. One point that should be mentioned in connection with any method used for the acceleration of oxidation is that of cleaning and rinsing the glassware coming into contact with the fat. The test tubes, air tubes, and condenser tubes of the aeration apparatus, and the beakers or other fat containers used in the hot air oven, should be given a thorough treatment so as to be certain that every trace of impurity is removed. That the treatment of glassware used in oxidation studies must be thorough and rigorous is indicated by the work of Laug (66) and Stebnitz and Sommer (115) who found that several rinsings with distilled water were necessary to eliminate cleaning compounds from the glass. The fact that extremely high temperatures may effect destruction of the natural antioxidant prOperties of butter fat or possibly pro- duce slight oxidation, is indicated by the results of the studies on heat treatment of butter oil. This may be an important consideration in the preservation of butter oil by heat sterilization. Heating the cream to temperatures above the temperature of boil- ing water caused considerable decrease in stability of the resulting butter fat. Practical application of these findings may be made in the case of the cream pasteurization processes involving extremely high temperatures. However, in practice, when temperatures above boiling are used, the exposure is instantaneous. Thus, the actual treatment would be less drastic than in these trials in which the steam autoclave was used and the samples were held at the designated -89- temperature for 30 minutes. The studies of the effect of heat treatment of cream upon metallic induced oxidation were particularly interesting. In this connection temperatures of 85.00 C. (185.00 F.) and 90.6° C. (195.o° F.) were of advantage in reducing oxidative changes as compared to 62.80 C. (145.00 F.) both when the metal was added before and after pasteuriza- tion. These observations are in line with the findings of Ritter (98) 'who found that cream pasteurized at 90°C. did not develOp fishy flavors whereas that processed at lower temperatures was conducive to this flavor change. From the findings of Gould and Sommer (42) and Josephson and Doan (58) the preservation action of temperatures of 85.00 C. (185.00F.) and 90.60 C. (195.00F.) may be eXpected. These investigators found that heating milk momentarily to 76.00 to 78.000. caused a formation of highly-reducing sulphur-compounds resulting in a lower oxidation- reduction potential and a prevention of oxidized flavor develOpment. Therefore, pasteurization of cream to temperatures from 85.00 C. (185.00 F.) to 90.60 c. (195.o° F.) should form sulphides, some of which would remain with the butter after churning causing the butter and butter oil to be more stable toward oxidative changes during stor- age. The sulphur compounds also would prevent oxidation by inactiva- tion of metallic catalysts which may be present. This fact is eSpe- cially pertinent to many processing plants practicing high temperature pasteurization of cream. The results in this study show that tempera- ture of 85.00 C. (185.00 F.) and 90.6° C. (195.o° F.) have an inhibit- ing influence upon metallic induced oxidation, eSpecially when metal -90- contamination occurs before pasteurization. The findings indicate the extreme necessity of preventing metallic contamination after pasteurization in order to enable maximum stability of butter against oxidation during storage. Results secured with Avenex Concentrate agree'with those obtained by Dahle and Nelson (33) but do not agree in all phases with those secured by Musher (83). The data herein presented show that adding Avenex Concentrate to butter oil with and without metal contamination created no great increase in stability of the oil toward oxidation. Contrary to this, Musher (83) found the addition of from one to five per cent Avenex to butter oil increased its stability against oxida- tion from 180 to 280 per cent. The reason for this difference may have been (a) Musher's use of extremely unstable butter oil, (b) Avenex was used by Musher whereas Avenex Concentrate was used in these studies, and (c) Musher used much higher concentration of the oat flour product than was used in these investigations. -91- SUMMARY AND CONCLUSIONS Preliminary studies were conducted of the peroxide method to ascertain the possibility of decreasing the amount of solvent and potassium iodide used. Data which were secured indicate that the amount of solvent cannot be reduced appreciably without decreasing considerably the maximum peroxide value obtained. However, the concentration of the potassium iodide solution can be altered from the recommended saturated solution to a 50 per cent solution with only a slight decrease in peroxide value. Comparison of the acid degree and peroxide value indicates that the acid degree is not sufficiently sensitive to be a desirable method for detecting initial oxidative changes in butter oil. Investigations were made to compare the aeration and hot air oven methods of accelerating fat oxidation at 1000 C. Results showed these two methods to be equally reliable when careful technique was exercised and consideration was given the influence of certain variable factors. Under carefully controlled conditions, the differ- ences in the induction periods of the same fat should agree within 6.4 per cent. This would amount to 0.64, 1.28, and 1.92 hours difference at 10, 20, and 30 hours induction periods reSpectively. Air agitation, position of the samples in the oven, size of sam- ples, and the surface area of the samples are factors influencing the accuracy of the oven method. In the oven used, it was found necessary to use positive air circulation to enable minimum temperature varia- tion between different samples on the same shelf. Also grouping of samples in the middle of the shelf reduced the maximum temperature variation. The bottom shelf of the oven studies had a consistently -92.. higher temperature than the tOp shelf. The induction period of butter fat as determined in the oven method is influenced by the surface area and the size of sample. The rate of oxidation increases directly with the surface area and inverse- ly with the size Of sample. Studies were made of the oxidation of butter fat in the hot air oven at different temperatures by using two identical ovens regulated at 7o.o° and 100.00 c. and then at 100.00 and 130.00 c. The tempera- ture of oxidation was found to have little influence upon the accuracy of the method, the results showing 9,9, and 9.4, and 11.6 per cent maximum deviation reSpectively between 70.00, 100.00, and 130.00 C. Consequently, the temperature which should be used, in one which per- mits rapid acceleration of oxidation without undue destruction of per- oxides. At the same time, the rate Of oxidation should be so governed that samples may be removed for examination at intervals convenient to the laboratory Operator. The heating of butter oil to 127.o° c. (262.00 F.) for 30 minutes in the autoclave markedly decreased its stabilitv d, whereas heating to 109.80 C. (229.60 F.) for 30 minutes did not appreciably influence its rate of oxidation as compared to non-heated butter oil. Butter, as such, appears to be much more stable against oxidative changes than the filtered butter Oil. Pasteurization temperatures Of the cream up to and including 90.60 C. (195.00 F.) did not decrease the stability of the resulting butter Oil. In fact, longer exposures of the cream to 90.60 C. (195.00 F.) appeared to increase the stability of the butter Oil slightly, although the data were somewhat variable. -95- The addition Of five ppm. COpper to cream before pasteurization did not increase the rate of oxidation of the butter oil when pas- teurization temperatures of 85.00 c. (185.00F.) and 90.60 c. (195.00 F.) were used. However, the fat from similarly treated cream that was pasteurized at 62.80 C. (145.00 F.) oxidized considerably sooner than the control. When metal contamination occurred after pasteurization, the 00p- per containing samples oxidized more rapidly than the control at all temperatures studied. However, in the case of samples heated to 85.00 c. (185.00 F.) and 90.60 c. (195.o° F.) this increase was only slight whereas at 62.80 C. (145.00 F.) the increase was considerable. ‘ Although heating cream to 85.00 C. (185.00 F.) and 90.60 C. (195.o° F.) had no harmful influence on the stability of the butter oil, temperatures of 109.80 c. (229.e° F.) and 127.00 c. (262.0°F.) decreased the stability of the butter Oil. The higher temperature was eSpecially effective in this connection. On the basis Of the results of this study, pasteurization tempera- tures for cream of approximately 85.00C. (185.0o F.) to 90.60 C. (195.00 F.) should be used from the standpoint of reducing the cata- lytic influence of metallic contamination. The addition of 0.015 per cent Avenex Concentrate to butter oil retarded only slightly the rate of oxidation of the butter Oil. When 0.05 per cent Avenex Concentrate was added to cream after pasteuriza- tion, the retarding influence on oxidation was only slight when normal cream was used. However, when metal contaminated cream was used, the influence of the antioxidant was more pronounced. Comparisons were made of the carotene destruction and peroxide -94.. formation by determining the carotene content of butter fat at various peroxide values. Average results showed that by the time the peroxide value had increased to 2.0, the carotene content decreased from 8.1 to 3.4 micro-grams per gram Of fat, a decrease of 58.0 per cent of the total carotene. This indicates the possibility that the carotene determination may be a more sensitive index of the initial oxidation of butter Oil than the peroxide value. 1. 2. 4. 5. 6. 7. 8. 9. 10. 11. 12. REFERENCES Anderson, E.O. 1939 Preventing DeveIOpment of Oxidized Flavor in Milk Through the Addition of Small Amounts of Pancreatic Enzyme. Milk Dealer 29(5): 32. Association of Official Agricultural Chemists. 1940 Official and Tentative Methods of Analysis. 5th Ed: Washington, D.C. 757 pp. Barnicoat, C.R. 1931 Rancidity Changes and the Flavour of Fats. Jour. Soc. Chem. Ind. 50: 361-365T. and Palmer, L.S. 1939 A Study of the Chemistry of Incipient Oxidation Defects in Butter. Minn. Agr. Expt. Sta. Tech. Bul. 134, 63 pp., illuS. Battie, M.A. and Smedley-MacLean, I. 1929 Catalytic Action of Cupric Salts in Promoting the Oxidation of Fatty Acids by Hydrogen Peroxide. Bio. Chem. Jour. 23: 593-599. Bendixen, H.A. 1940 Acid Values and Acid Ratios as Related to the Keeping Quality of Salted Butter. Jour. Dairy Sci. 23: 275-284. Bickford, W.G. and Markley, K.S. 1940 The Stability of Vegetable Oils. II. Apparatus for Determination of the Rate of Fading of Methylene Blue Fat System. Oil and Soap 17: 232-240. Bird, O.D. 1941 A Study of Antioxidants With ResPect to Vitamin A in Fish Oils. Thesis for Degree of Ph. D., Mich. State College, 116 pp., illus. Briggs, L.H. 1931 The Autoxidation of Butter Fat. I. Factors Influencing the Reaction. Jour. Dairy Res. 3: 61-69. 1931 The Autoxidation of Butter Fat. II. Comparison of Tests for Detecting Oxidation Changes. Jour. Dairy Res. 3: 70-79. Browne, C.A. 1925 The Spontaneous Decomposition of Butter Fat. Indus. and Engin. Chem. 17: 44-47. BI'OWTI, R.W. 1940 Vacreating Cream. Canad. Dairy and Ice Cream Jour. 19(3): 21. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. - 96 - 1940 Vacreated Cream for Buttermaking. Canad. Dairy and Ice Cream Jour. 19(9)L 26. Brown, W.C., Thurston, L.H., and Dustman, R.B. 1936 Oxidized Flavor in Milk. III. Time of COpper Contamin- ation During Production and Processing, and Aeration Versus no Aeration as Related to Oxidized Flavor Deve10p- ment. Jour. Dairy Sci. 19: 753-760. , Dustman, R.B., and Thurston, L.M. 1937 Oxidized Flavor in Milk. V. The Effect of Metal- Deve10ped Oxidized Flavor on the Iodine Number of the Milk Fat. Jour. Dairy Sci. 20: 599-604. Bruechner, H.J. 1939 Short Time High Temperature Pasteurization. Cornell Agr. Expt. Sta. Ann. Prto P. 1240 B1111, So 1937 Rancidity and Mold in Cured Bacon Retarded. Natl. Provisioner 97(13): 13. 1938 Retarding Rancidity in Lard and Pork Fat. Natl. Provisioner 98(4): 15. 1939 Rancidity of Pork Fats. Natl. Provisioner. 100 (6): ll. Christiansen, J.A. 1924 Note on Negative Catalysis. Jour. Phys. Chem. 28: 145-148. Clark, G.L. and Rugg, F.M. 1941 Measuring Oxidation of a Vegetable Oil. Indus. and Engin. Chemo Allaly’to Ed. 13: 243.244. Combs, W.B., Coulter, S.T., and Whitman, D.W. 1941 Avenized Versus Standard Parchment for'Wrapping Print Butter. Jour. Dairy Sci. 24: 117-125. Conn, R.C. and Asnis, R.E. 1937 Oat Flour as an Antioxidant. Indus. and Engin. Chem. 29: 951-9520 Corbett, W.J. and Tracy, P.H. 1937 Use of Antioxidants to Prevent Tallowiness in Butter. Natl. Butter and Cheese, Jour. 28(Dec. 25): 10-14. 1940 Effect of Condensing on the DeveIOpment of Oxidized Flavor. Jour. Dairy Sci. 23: 209-214. 1941 Experiments on the Use of Certain Antioxidants for Control of Oxidized Flavor in Dairy Products. Food Res. 6: 445-459. 27. 28. 29. 30. 51. 132. £53. £34. 355. 365. 31?. 3E3. 393. -97.. , and Hansen, C.N. : in. 1941 Cereal Antioxidant Prevents Tallowiness in Butter. Food Indus. 13: 34-36. Dagneaux, E.L.K. 1939 The rancidification of Oils and Fats. Chem Weekly. 36: 179-190 (Dairy Sci. Abs. 1: 353, 1940). Dahlberg, A.C. and Carpenter, D.C. 1936 The Influence of Methods of Sterilizing Equipment Upon the Development of Oxidized Flavor in Milk. Jour. Dairy Sci. 19: 491. Dahle, C.D. 1935 Tallowy Flavor in Milk. Penn. Agr. EXpt. Sta. Bul. 320: 22 pp. and Josephson, D.V. 1937 Improving the Keeping Quality of Butter with Treated Parchment. Natl. Butter and Cheese Jour. 28 (July 28): 6-7. 1937 Delaying Oxidative Changes in Butter. Natl. Butter and Cheese Jour. 28(Sept. 25): 18:21. and Nelson, D.H. 1941 Antioxygenic Fractions of Oat and Soya Bean Flour. Jour. Dairy SCio 24: 29-300 Davies, W.L. 1931 Observations on the COpper and Iron Content of Milk and Other Dairy Products. Jour. Dairy Res. 3: 86-92. 1933 The Mode of Combination and Distribution of Traces of Heavy Metals in Dairy Products. Jour. Dairy Res. 4: 255-2650 Doan, FOJ. 311d 17111191., Gayle 1940 Trypsin - An Anti-oxygen for Milk. Milk Plant Monthly. 24 (9): 42-47. Evans, E.I. 1935 Antioxidant PrOperties of Vegetable Lecithin. Indus. and Engin. Chem. 27: 329-331. Fabricius, N.E. and Bird, E{W. 1940 The Quality of Butter Made from Vacuum-Pasteurized and Vat-Pasteurized Lots of the Same Cream. Iowa Agr. Expt. Sta. Res. Bul. 284: 20 pp., illus. French, 8.8., Olcott, H.S., and Mattill, H.A. 1935 Antioxidants and the Autoxidation of Fats. Indus. and Engin. Chem. 27: 724-728. 1 , . , v . . . . . I . o I ' ‘ . - I < . . e l , a . A v ,, . . . s ' ' _ n u ' n . . I . ’ 3 | l - n a ' 1 7 n r ‘ u i ‘ n ‘ l ' . 1- I Y 1 § * 3 . _ . 41. 42. 43a 4&4. 445. 4&3. 4'7. 453. 45). 5].. -98- Freyer, E. 1939 How Accurate Must an Official Fat Stability Test Be? Oil and Soap 16: 191-1940 Gor‘tnel‘ , R 04 o 1938 Outlines of Biochemistry. 2nd. Ed. John Wiley and Sons, Inc a New York 0 l!) 1 7 pp 0 , 111113 0 Gould, I.A. and Sommer, R.H. 1939 Effect of Heat on Milk with Depecial Reference to Cooked Flavor. Mich. Agr. Expt. Std. Tech. Bul. 163: 48 pp., 11 1113 o Greenbank, G.R. and Holm, 0.3. 1924 Some Factors Concerned in the Autoxidation of Fats With ESpecial Reference to Butter Fat. Indus. and Engin. Chem. 16: 598-601. 1925 Measurement of Susceptibility of Fats to Oxidation. Indus. and Engin. Chem. 17: 625. 1930 A Photo-Chemical Method for Measuring Susceptibility of Fats and Oils to Oxidation. Indus. and Engin. Chem. Analyt. Ed. 2: 9-10- 1933 Photo-Chemical Oxidation of Cottonseed Oil. Indus. and Engin. Chem. 25: 167-168. 1934 Antioxidants for Fats and Oils. Indus. and Engin. Chem. 26: 243-245. Guthrie, E.S., Scheib, B.J., and Stark, C.N. 1936 The Effect of Certain Factors Upon Keeping Quality of Butter. Jour. Dairy Sic. 19: 267-278. Henderson, J.L. and Roadhouse, C.L. 1934 Factors Influencing the Initial Induction Period in the Oxidation of Milk Fat. Jour. Dairy Sci. 17: 321-330. Holm, G.E. 1937 Chemical Studies of Keeping Quality. Proceedings 29th Ann. Meeting Amer. Butter Institute (Symposium). Chicago, Ill., 15 pp. and Greenbank, G.R. 1923 Quantitative ASpects of the Kreis Test. Indus. and Engin. Chem. 15: 1051-1053. 52. 53. 54. 57. 58. 59. 60. 61. 62. 63. , and Deysher, E.F. 1927 Susceptibility of Fats to Autoxidation. Indus. and Engin. Chem. 19: 156-158. , Wright, P.A., White, W., and Deysher, E.F. 1938 The Keeping Quality of Butter. I. The Rate of Deterioration of Butter Made From Creams of Different Acidities and Stored at Different Temperatures. Jour. Dairy Sci. 21: 385-398. Horrall, B. E. and Epple, W.F. 1930 A Modification of the Ritter Test for COpper in Butter. Jouro Dairy SCI. 23: 53-600 Hunziker, O.F. 1940 The Butter Industry. 3rd. Ed. Pub, by Author, LaGrande, 111. 821 pp., illus. and Hosman, D.F. 1917 Tallowy Butter. Its Causes and Prevention. Jour. Dairy Sci. 1: 320-346. Johnston, W.R. and Frey, C.N. 1941 Autoxidation Measurements of Fatty Oils Using Barcroft- Warburg Apparatus. Indus. and Engin. Chem. 13: 479-481. Josephson, D.V. and Doan, F.J. 1939 Observations on the Cooked Flavor in Milk. Its Source and Significance. Milk Dealer 29(2): 35-36, 54-62. Keilling, J. 1939 The Problem of COpper in Pasteurized Creams and Butters. Compt. Rend. Acad. Agr. Grance 25: 382-387 (Chem. Abs. 33: 4688, 1939). Kerr, RoHo 1918 Chemical Tests for the Detection of Rancidity. Indus. and Engin. Chem. 10: 471-475. and Sorber, D.C. 1923 The Analytical Detection of Rancidity. Indus. and Engin. Chem. 15: 383-385. King, A.E., Roschen, H.L., and Irwin, W.H. 1933 An Accelerated Stability Test Using the Peroxide Value as an Index. Oil and Soap 10: 105 (Stebnitz and Sommer (115), Loc. Cit.). Kochling, J. and Taufel, K. 1939 Spoilage of Fats. VII. Effect of Heating on Storage- Stability of Fats. Fette U. Seifen. 46: 206-209. (British Chem. and Physiol. Abs. B. Applied Chem. p. 742, 1939). 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. - 100 - Koenig, V.L. 1937 Stabilization of Butter Against Oxidized Flavor. Natl. Butter and Cheese Jour. 28(Aug.): 26-28. Krauss, W.E. and Washburn, R.G. 1936 The Iron and COpper Content of Milk Throughout the Season as Related to Anemia Deve10pment in Rats. Jour. Biol. Chem. 114: 247-252. Laug, E.P. 1934 Retention of Dichromate by Glassware After Exposure to Potassium Dichromate Cleaning Solution. Indus. and Engin. Chem. Analyt. Ed. 6: 111-112. Lea, COH. 1931 Effect of Light on the Oxidation of Fats. Proc. Roy. Soc. (London) B 108: 175-159. (Vibrans (123), Loc. Cit.). 1933 The Action of Light on Fat. Jour. Soc. Chem. Indus. 52: 146T. 1936 Antioxidants and the Preservation of Edible Fats. Jour. Soc. Chem. Indus. 55: 293T. 1939 Rancidity in Edible Fats. Food Investigation Spec. Rpt. 46, 1st Amer. Ed. Chemical Publishing Co. N.Y. 23) Pp. Loftus-Hills, G., Scharp, L.R., and Belair, T.S. 1934 A Study of Factors Influencing the Keeping Quality of Some Victorian Salted Butter in Cold Storage. Jour. Dairy Res. 5: 124-136. Lowen, L., Anderson, L., and Harrison, R.W} 1937 Cereal Flours as Antioxidants for Fishery Products. Indus. and Engin. Chem. 29: 151-156. Maack, A.C. and Tracy, P.H. 1939 Controlling Oxidized Flavors. Ice Cream Trade Jour. 35(3): 32. Mattill, H.A. 1931 Antioxidants and the Autoxidation of Fats. Jour. Biol. Chem. 90: 141-151. 1941 The Mechanism of the Autoxidation of Fat. Oil and Soap. 18: 73-76. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. - 101 - McFarland, G.C. and Burgwald, L.H. 1940 Prevention of Oxidized Flavor in Frozen Cream. Jour. Dairy Sci. 23: 494-496. Meyers, J.E., Kass, J.P., and Burr, 0.0. 1941 Comparative Rates of Oxidation of Isomeric Linolenic Acids and Their Esters. Oil and Soap 18: 107-109. Miles, N.A. 1932 Auto-Oxidation. Chem. Rev. 10: 295-364. Monagahn, B.R. and Schmitt, F.O. 1932 The Effect of Carotene and of Vitamin A. on the Oxidation of Linoleic Acid. Jour. Biol. Chem. 96: 387-395. Moore, L.A. 1941 Method of Determining Carotene in Cream. Mich. Agr. Expt. Sta. Unpublished. Mortensen, M. 1924 The Improvement of Flavor and Keeping Quality of Hand Separated Cream Butter. Jour. Dairy Sci. 7: 460-467. Maureu, C. and Dufraisse, C. 1926 Catalysis and Auto-Oxidation. Anti-Oxygenic and Pro- Oxygenic Activities. Chem. Rev. 3: 113-162. Musher, S. 1937 Oat Flour as an Antioxidant. Ice Cream Field. 31(7): 31-32. Nelson, D.H. and Dahle, C.D. 1940 The Relation of Iodine and Peroxide Number to Oxidized Flavor in Milk. Jour. Dairy Sci. 23: 391-397. 1940 Enzymes and Other Substances as Antioxidants in Milk. The Milk Dealer 29(10): 41-55. Olcovich, H.S. and Mattill, H.A. 1936 The Unsaponifiable Lipids of Lettuce. I. Carotene. Jour. Biol. Chem. 91: 105-117. Overman, OCR. , Garrett, O.F. , and Ruehe, 5011. 1938 Studies on the Keeping Quality of Butter in Cold Storage. Ill. hgro Expt. Sta. BUIO 446: P0 47‘900 Patil, VOHO and hammer, Bod. 1928 The Keeping Quality of Ghee. Jour. Dairy Sci. 11: 143-154. Peters, F.N. and Musher, S. 1937 Cat Flour as an Antioxidant. Indus. and Engin. Chem. 29: 146-151. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. Platon, 1938 Powick, 1923 - 102 - B. and Olsson, T. The Effect of Various Methods of Flash Pasteurization and of Various Foodstuffs on the Quality of Butter as Shown by Experiments at a Southern Swedish Creamery. (Pasteurizer of "Kettle",or Danish, Type Compared with A.P.V. Plate Pasteurizer). Medd. States, Mejeriforsok No. 1. (Dairy Sci. Abs. 1: 220, 1940). W.C. Compounds Developed in Rancid Fats. 'With Observations on the Mechanism of their Formation. Jour. Agr. Res. 26: 323-362. Rice, F.E. and Miscall, J. 1923 COpper in Dairy Products and Its Solution in Milk Under Various Conditions. Jour. Dairy Sci. 6:261-277. Ritter, v]. 1937 The Importance of Pasteurization Temperature for Butter Making. Alpenlandische Molkerei und Kaserie Zeitung. II: 159-151. (Jour. Dairy Sci. Abs. 22: A101, 1939). 1939 Cooking Residue of Butter. Schweiz, Milch. No. 47. (Jour. Dairy Sci. Abs. 22: A95, 1939). and Christen, M. Fishy Flavor in Butter (Influence of Adding Salts of Iron and COpper to Milk and Cream). Schweiz, Milch. No. 7. (Jour. Dairy Sci. Abs. 19: A218, 1936). and Nussbaumer, T. The Formation of Peroxides in Butter Fat. (Oxidation of Butter Fat. Schweiz, Milch. No. 12. (Jour. Dairy Sci. Abs. 22: A97, 1939). 1938 Oxidation of Butter Fat. II. The Influence of COpper on the Peroxide Number of the Butter Fat. Schweiz. Milch. No. 38. (Jour. Dairy sci. Abs. 22: A 98, 1939). 1938 The Oxidation of Butter Fat. I. Peroxide Formation in Butter Fat. Schweiz, Milch. No. 64: 59-61. (Chem. Abs. 32: 7148, 1938). 1938 The Oxidation of Butter Fat. III. The Effect of Various Substances on the Formation of Peroxides in Butter Fat. Schweiz, Milch. 54: 465-466. (Dairy sci. Abs.. 1: 95, 1939). 100. 101. 102. 1C33. 104. 105. 1065. 107. 108. 109. 110. - 103 - 1938 The Oxidation of Butter Fat. IV. Influence of Fat Peroxides on Keeping Quality of Butter Fat. Schweiz. Milch. 54: 525-528 (Dairy Sci. Abs. 1: 293,1940). 1938 Oxidation of Butter Fat. III. The Influence of Different Substances on Peroxide Formation in the Butter Fat. Schweiz. Milch. No. 82. (Jour. Dairy Sci. Abs. 22: A98, 1939). 1938 Oxidation of Butter Fat. IV. Influence of Fat Peroxides on Keeping Quality of Butter Fat. Schweiz. Milch. No. 92. (Jour. Dairy Sci. Abs. 22: A99, 1939). 1939 The Oxidation of Butter Fat in the Preparation of Melted Butter. Schweiz, Milch. 65: pp. 53-54 and 61-62. (Dairy Sci. Abs. 1: 293, 1940}. 1939 ’TEeHUEIEEEISE of Butter Fat. VI. The Influence of Diacetyl on Peroxide Formation in Butter Fat. Schweiz. Milch. 85: 193. (Dairy sci. Abs. 1: 293, 1940). Roberts, W.M., Combs, W.B., and Coulter, S.T. 1940 High-Temperature (Steam-Injection) Pasteurization of Cream for Buttermaking. Jour Dairy Sci. 23: 315-323. Rogers, L.A., Associates of 193 Fundamentals of Dairy Science. Reinhold Pub. Co. N.Y. 2 nd. Ed. 616 pp., illus. , Berg, W.N., Potteiger, 0.2., and Davies, B.J. 1913 Factors Influencing the Change in Flavor in Storage Butter. U.S. Dept. Agr. Bur. Animal Indus. Bul. 162: 69pp. Schibstead, H. 1932 New Test for Fat Aldehydes Resulting from Oxidation of Fats and Oils. Indus. and Engin. Chem. Analyt. Ed. 4: 204*2060 Schulz, M. and Storck. W. 1940 Experiments Dealing With the Manufacture and Keeping Quality of Butter Oil. Deut. Molkerei Ztg. 9: 29; 10: 143-145; 11: 166-167. (Jour. Dairy Sci. Abs. 24: A201, 1941). iSmedley-MacLean, I. and Pearce. M.S.B. 1931 The Oxidation of Oleic Acid by Means of Hydrogen Peroxide With and Without the Adddition of COpper Sulfate. Biochem. Jour. 25: 1252-1266. 111. 112. 116. 117. 118. 119. 120. 122. Smith, J.A.B. 1939 The Rapid Determination of Peroxide Values for the Fat in Milk Powder. Jour. Dairy Res. 10: 294-299. Smith, O.M. and Hood, R.E. 1926 Inhibiting Agents in the Oxidation of Unsaturated Organic Compounds. Indus. and Engin. Chem. 18: 691-294. Staudinger, H. 1925 Ber. Dtsch. Chem. Ges. 58B: 1075 (Lea (70), Loc. Cit.). Stebnitz, V.C. and Sommer, H.H. 1935 An Improvement on the Swift Fat Stability Apparatus for Approximating the End of the Induction Period. 0i1 and Soap 12: 201. (Stebnitz and Sommer (115), Loc. Cit.). 1937 The Oxidation of Butter Fat. II. The Composition of Fat in Relation to its Susceptibility Toward Oxidation. Jour. Dairy Sci. 20: 265-280. 1937 The Oxidation of Butter Fat. I. The Catalytic Effect of Light. Jour. Dairy Sci. 20: 181-196. Stephens, H.N. 1933 Studies in Auto-Oxidation. III. The Initial Act in Atto-Oxidation. Jour. Phys. Chem. 37: 209-217. Swanson, A.M. and Sommer, H.H. 1940 Oxidized Flavor in Milk. I. Effect of the DeveIOpment of Oxidized Flavor on the Iodine Number of the Phos- pholipid Fraction of Milk. Jour. Dairy Sci. 23:201-208. 1940 OxidizedfiFlavor in Milk. II. The Relation of Oxidation- Reduction Potential to its DeveIOpment. Jour. Dairy Sci. 23: 597-614. Taffel, A. and Revis, C. 1931 The Determination of Rancidity in Oils and Fats. Jour. Soc. Chem. Indus. 50: 87T. Thurston, L.H., Brown, W.C., and Dustman, R.B. 1935 The Probably Relation of Lecithin to Oxidized Flavor. Jour. Dairy Sci. 18: 301-306. Tschirch, A. 1925 Autoxidation in Fats, Resins, Terpenes, and Tannins. Chem. Umschau. 52: 29-51 (Chem. Abs. 19: 1557, 1925). 123. 125. 126. 127. 128. - 105 - Vibrans, F.C. 1941 Methods of Measuring the Rate and Extent of Oxidation of Fats. Oil and Soap 18: 109-112. Wheeler, D.H. 932 Peroxide Formation as a Measure of Autoxidative Deterioration. Oil and Soap 9: 89-97. Wiley, W.J. 1939 Oxidation of the Fat of Butter During Cold Storage. Jour. Hairy Res. 10: 300-309. and G111, A011. 934 Thiocyanogen Number. Indus. and Engin. Chem. Analyt. Ed. 6: 298. "Williams, W. 1931 Determination of COpper and Iron in Dairy Products. Jour. Dairy Res. 3: 93-100. Wilson, C.A., Tuckey, S.L., and Ruehe, H.A. 1940 A Comparison of Butter Made from Cream Pasteurized by Three Different Methods. Natl. Butter and Cheese Jour. 31(12): 12. Wilster, G.H. 1940 Vacuum Pasteurization of Cream for Butter. Ore. Agr. Fxpt. Sta. Bul. 368, 5lpp., illus. .“ -LL .J_, 3 ‘9.- 11,1")... .g...L - 106 - 37 .- 5 Variation between oxidation of duplicate samples b Table I. aeration method. Difference Difference : between Peroxide value Hours :cxidation Peroxide value between Hours oxidation a and b a b a and b 8. r3. 1 41 0 0 0 O A; 2 4. 1 0 o O 0 8 8 O 1 70 OJ 2 6 O O 0 ml. 2 7 2 2 5 7 9 «(C OJ 6. I o 0 O O O 9 8 n. 0 0 0 3 3 we 1 2 3 7. n0 0 0 o 00 A. A. 1 2 4.. 6 8 0.7 4.2 4.0 8.6 18.0 7.9 22.2 3.6 0.3 13.4 3.3 14.7 6 31.6 27.6 0.7 0.3 24.6 7.4 19.5 7.7 14.0 3.7 11 6.1 Odo 11. 0.6 4.9 16.3 13 12 5.1 12.1 3.7 0.1 0.7 0.6 4.9 4.4 0.6 0.1 7.7 14.0 11 3.5 8.4 10 12 14.6 14.6 1.7 13 16.3 1.1 1.2 4.4 8.6 1.1 0.1 3.3 8.5 10 0.2 05 n. 6.0 11 r0 11 1.2 2.6 1.2 4.4 8.6 0.2 1.0 1.5 5.1 10 12 012 000 13 11 0.2 1.3 0.6 11 5 1 OOO 1.5 ' 0.1 9.7 1.6 8.5 15 23 1.2 24 3.1 8.1 5.0 623 000 871 Brag coo UL59 12 875 000 028 27v 684.. 222 00 920 000 195 030 000 190 111 950 000 285 12 5770 222 10.8 10. 20.8 26 28 21 12.9 21.2 25 27 20 22 24 20 22 28.3 4. 32.7 5.5 1.8 34.0 28.5 2.8 15.5 3.2 4.8 3.0 6.3 25.2 1.0 6.5 1.7 r0 .0. 23 25 21 5.3 45.2 22.0 28.5 20.0 4.9 17.8 3.2 14.5 3.1 0.1 3.0 3.3 23 . . A1670 ‘1 9 220 00000.000 a34111119fqunu11 . m 7.2u11nlnox915.8 o o o o on. o o nondnonbn8u5.1no 2o itii Sllzwzgnog 00000000 844151111 2 22 55 00 502911630 222129911 0000000000000000 2004.29.10 00000000 104160300 4.3266719... o O o O O o O O 8511112011_o.1.1 2 1111 22u228829~ 00.00000 au1.7.7.1.11111. 2 11 49130022 21222211 1.4 1.3 0.1 0.0 12 19 1.3 -" c-o‘--m““m- 1.3 00 00 12.7 17 15.0 13.5 17.5 Difference between a and b 4.0 6.4 0.2 0.7 1.7 000 306 1.0 1.9 801 4.4 0.2 4.5 3.8 0.1 7.0 1.3 003 0.0 003 .6 b 11.2 13.8 11.6 15.7 1.9 9.9 12.3 1.7 "18:5 6.4 13.2 28.7 21.1 138 3.2 19.1 2.5 1 8.6 1.4 1.7 —WV‘—fi"‘ FL 1.7 12.9 7.4 11.3 2.0 2.0 14.6 6.3 1.5 2.7 1.4 6.2 7.2 8.1 10.6 Peroxide value 11.8 11.8 10.6 36.8 Hours :oxidation 23 20 20 22 18 20 18 20 19 16 16 15 17 19 15 17 - 107 - fferenoe : 1.0 1.4 2.6 0.9 1.5 0.2 1.3 2.0 1.3 1.9 1.2 1.5 0.0 -‘i- m‘“.‘.. .“m-.-‘—-- .‘ -'-.--.-.—— u. ‘ o ‘ ‘ - - -‘u- - 9‘...“ ‘3 I 0 40 -J between a and b Di v\ _ '7;" 3.1 6.7 18.6 9.0 7.6 18.4 17.3 14.9 18.1 13.7 1.0 .50 16.1 1.5 4.3 a 5.3 21.2 8.1 8.1 18.6 2.7 1.5 4.1 Peroxide value 4.1 18.6 18.6 12.9 16.8 11.8 Hours oxidehion 22““ 22 21 21 24 24 24 20 18 21 18 20 18 13 13 19 17 Table I. (continued.) 2.7 no 21 9.4. 1.4 1.6 16 15 1.7 1.7 1.2 3.6 0.0 0.1 1305 1.8 10.1 4.1 18.7 7.3 1.9 1.8 18.1 - -.- .f‘.—-—— .7 4.8 10.4 1.5 87Z“"“‘ 2.4 16.4 5.7 6.0 3.7 1.9 20.5 18 19 18.5 : ‘ZT“‘“"'"”‘ 24 22 18 19.5 19 ~u. - ---‘—' 0.4 3.3 0.2 0.1 2.3 0.2 4.4 004 6.1 11.1 2.3 109 11.7 5.5 14.1 1.7 1.6 4.1 4.3 12.8 7.1 1.4 2.4 1.5 804 5.7 4.8 109 6.0 3.7 1004' 1.7 4.8 21 18.5 18 22 20 24 19 18 19.5 Table II. Preliminary trials with aeration method. Hours Peroxide values of dqplicates oxidation a b Trial 1 4 3.3 3.9 3.6 5 7.9 8.4 8.6 6 14.7 13.8 13.4 7 22.2 18.1 18.0 8 24.6 25.0 25.3 9 27.6 30.0 31.6 Trial 6 0.6 0.9 0.7 9 3.7 4.2! 5.1 10 4.9 11.0 8.4 11 7.7 7.4 12.1 12 16.3 16.5 14.6 13 14.0 19.5 14.6 Trial 7 1.0 1.0 1.2 8 1.2 1.1 1.2 9 1.5 1.3 1.3 10 4.4 3.3 2.6 11 5.1 6.0 4.5 12 8.6 8.5 8.6 Trial 20 3.0 4.7 3.1 21 3.2 2.8 4.9 22 6.3 5.3 9.2 23 14.5 15.5 17.8 24 25.2 45.2 26.4 25 28.5 22.0 33.7 Trial 19 1.3 1.3 22 4.1 8.3 3.1 23 7.2 11.2 13. - 109 - ples taken L 1. Variation in oeroxide value of duplicate sam from too and bottom shelves of hot air oven at same Table III. time. L TOp She1f Bottom * Ton shelf Bottom :nifference she1f :Uifference * shelf 9w 4.1 1.9 -l.4 9.4 6.8 16.7 8.0 8.3 19.4 1. 1.5 1.6 1.5 2.3 2.0 5.1 9.6 2.7 2.4 4.5 2.7 5.1 24.0 28.6 0.0 1.8 1.8 2.3 17.3 19.6 0.3 5.7 8.3 32.4 * 1.4 6.9 10.6 2.2 1.1 10.8 1302 2.7 5.4 -10.3 5.4 23.5 -1.9 0.9 fiu9~ ZUZU 3. 0 1” R01. 1.4 FD 0. 1.7 6.1 2.4 2.2 8.5 7.0 -2.9 8.3 5.1 10.1 1.8 0.3 2.1 7.9 1.8 2.0 5.8 0.7 3.4 1.8 O. 2.1 2.6 3.9 3.3 3.5 9“ my '003 “109 803 o 1.8 1.6 1.9 -3.1 10.5 4.8 9w 9~ 4.1 11.8 8.5 20.3 :57. O O 0.0 . . .bno . O 8.4. 0.9 O O no.0 * * a.9. O O 3O . . RJRV O O no.1 9H 0. O. 9.4. O O 231. .2 0,RV9. O. 0. 7.4.7. 6.7 7.9 7.9 0.1 -12.6 0.1 “108 3.6 24.7 1.6 4.9 7.1.7. 0 o O 1u9u1. 0.7 8.2 9.7 8.9 10.4 0.7 1.3 5.0 6.3 11.8 0.0 1.4 1.4 1.4.8, .0. 929.4. .111. O. O. .0 :31.0, .0. 1.9. 13% O. O C. 4.:37. O O O Q. .23 ms t t * 9.4.13 .0. 0.4.1 . . .0 O. .0 :37.3 .0. 1.RUA. 1.1. C. O. .0 7.1.0 .0. 1.1.1v 1.1. 0.0 0.9 0.9 “905 23.8 14.3 0.0 1.0 18.0 17.3 0.1 0.2 -O.2 1.0 1.4.9. 1.1.1. 2.7 3.7 43.4 19.3 13.3 6.3 17.4 4.1 -8.7 -29.6 29.1 35.4 -0.1 1.2 1.1 49.1 40.4 0.1 0.1 1.4 1.5 75.8 1.6 10.3 39.3 49.6 7--.... n. O 1. .d .e mu Q :?1 .C n O C 8T 0 1: R, 0 9w * ZV O flu . O. _b O 1. 0. 9w 0 1. -1.3 ‘008 9.7 0.0 0.1 :Difference TOp shelf 3.9 10.0 Bottom shelf 2.5 * - 110 - :Difference * -0.3 -nued) TOp shelf 1.6 I 1 Bottom 1.3 1.9 shelf Table III. (cont 0.3 -002 2:. 15.1 24.8 2.0 1.8 4.6 0.7 . 1.4 1.3 0.6 3.8 -0.1 1.5 1.4 4.9 “100 -0.8 13.2 12.2 2.4 3.9 -2.8 -4.1 8.5 10.3 7.7 11.6 0 O. 3.8 17.6 7.7 14.8 1.3 “109 -0.3 24.6 20.5 17.5 11.9 10.0 17.2 “0.1 0.1 0.0 0.2 0.8 1.1 1.1 1.0 21.5 18.5 1.2 -206 ‘1008 -1309 25.1 22.5 1.3 1.7 1.5 1.3 1.9 34.3 23.5 40.8 26.9 2.3 -0.1 39.2 39.1 5.8 1.5 7.3 '802 8.4.0. .0. 4.7.0 . O... 620 00. 161 2 9.9.0. .00 681 11. ** R.1.1. .00 800 1.. _ 3.4.4. 0.0 4.1.1. 3 133 0.. 2u1.1. 5 “001 1.5 1.4 0.8 1.8 3.0 1.9 4.4 13.3 4.9 0.6 7.2 5.0 11.7 7.8 6.8 11.5 3.5 -601 7.9 1.8 “0.2 7.2 16.2 -004 7.5 0.1 6.6 14.5 21.1 1.2 1.3 7.1 10.7 1.5 -0.1 4.8 12.3 2.0 4.8 O 1 '305 ’60]. -606 '004 0.5 11.0 16.1 21.3 26.9 4.8 9.4 13.6 13.2 15.2 20.3 4.4 1.0 1.3 7.6 7.1 '00?) 0.1 6.8 6.7 11.3 1.0 1.4 8.1 14.3 18.4 0.7 1.5 3.6 7.3 9.0 1.8 -406 1.8 8.2 8.3 3.9 ”606 -4.3 .0 O 17.0 10.4 9.2 12.7 15.0 16.1 2.5 5.9 -1.1 11.8 14.3 6.0 0.0 0.4 9.6 15.6 12.2 18.1 1.3 1.3 1.8 . 6.7 (Continueif 5.6 1.4 - 111 - (continued) Table III. :Difference 01 pl. 08 Th 5 Mn. Of «61 t e Oh BS ** e C n e P. 8 P. f .1 D f 01.1 OR Twn... S m Of t1 t6 Oh BS 19.2 17.1 -003 9.. 00 -9.4 -9.4 -1.2 16.9 18.6 7.5 -400 -l.2 11.3 7.3 11.3 8.8 7.0 9.7 9.9 11.5 15.8 0.5 1.3 1.8 2.6 14.1 5 2 1.8 2.0 14.0 7 5 7.7 1.0 1.5 21.2 1.7 0.0 0.0 0.1 O. O. I. 1.0 0.2 0.1 1.9 2.1 1.1 1.1 4.2 3.5 3.4 «.06 37. 1.6 1.4 1.8 3.4 4.6 4.9 6.0 8.3 10.6 1.5 623 000 1.7 1.6 3.1 3.3 8.8 2.8 4.9 0.1 1.9 8.2 1.5 1.7 13.1 1.6 2.9 3.9 8.2 10.2 11.1 1.0 1.8 3.8 3.0 1.1 4.1 4.2 6.1 6.8 2.1 5.9 8.0 9.1 12.5 14.1 10.8 12.4 * a: :1: 22.2 14.1 3.0 16.0 19.0 5.7 3.1.1.00 O... 92nu1.9~ . O. O. O. 9014. 0... 5.1.1.1 9. O. O. O. O. 6924. O... 7.0V922u 92 * * * * 3.4.9.0. 000. "Inbncnu . O. O. O. .0 4.54.3 O... .Dacnlno 1.1.1.1. .0 O. O. .0 1.Q.R.zu O... RvRvouzu 1.1.1. 5.0 3.8 8.8 21.4 13.3 4.7 6.7 27.5 -11.8 11.4 9.1 15.7 6.5 15.6 1.5 17.2 18.7 0.2 1.6 4.9 1.2 3.8 8.1 1.4 5.4 13.0 0.0 0.0 0.3 012 111 19.9 .0 12.5 0.3 1.4 19.6 1.7 24.6 18.5 28.3 5.0 8.4 0.9 -0.8 0.9 0.9 19.7 28.1 1.5 1.4 5.6 18.0 0.4 1.5 1.9 6.5 17.2 2.4 2.1 7.8 4.5 11.4 14.8 4.3 21.6 7.0 6.8 0.1 0.0 O. 2.3 '304: 16.2 18.2 .0. 0.9 .0 1.0 7.2 0.1 21.0 28.2 1.3 1.3 0.7 0.8 - 112 - (continued) Table III. nothm shelf * _Bottom Top shelf tron shelf :Difference shelf :Difference * 0.0 1.5 1.3 0.2 0.1 0.1 0.4 0.9 0.9 5.6 1.3 1.8 1.2 1.3 2.1 6.6 13.4 7.7 9.0 19.4 1.3 1.4 2.4 6.0 1.8 2 9” 6.3 1.6 7.9 11.1 -20.7 3.5 -408 16.7 3 30.2 27.8 23.0 9.4 “-001 0 1.6 1.0 0.9 -703 31.4 26.6 -0.1 1.3 1.2 1.8 33.9 23.6 0.3 1.5 8.7 0.2 56.7 33.7 4.3 2.0 5.9 11.0 6.3 11.3 40.2 0 O 48.9 * .4 4.2 0.1 0.9 5 0.8 1.0 2.8 .0 0 1.8 0.0 1.0 1.2 15.2 1.3 1.2 1.2 1.5 2.4 4.6 01 1.7 2.2 2.1 3.8 19.8 0.6 -002 6.1 11.6 6.7 11.4 OJ 6.0 4.9 -7.0 24.3 17.3 6.4 8.4 7.7 19.2 19.4 16.1 0.2 0.7 0.9 “003 0.1 1.1 1.0 0.8 1.1 0.0 0.2 1.1 0.1 0.0 1.6 1.8 1.6 5.2 1.8 1.7 1.9 3.5 11.6 9.7 17.8 4.1 6.0 4. 0 3 3.5 10.1 6.9 14.6 10.1 9.. ho 26.1 1.1 11.7 7.2 21.8 1.0 1.1 0.0 1.2 1.1 1.3 10.1 0.8 4.5 17.2 18.0 15.8 20.3 0.1 1.5 1.8 2.0 9.3 1.6 2.0 * 26.3 9. o -11.9 39.0 27.1 79¢ 00.1.. .0 O. 5.7 10.5 2.5 0.1 0.3 43.1 0.6 0.8 45.6 0 0.2 0.9 1.1 1.1 1.2 1.1 1.0 1.1 0.0 0.1 0.7 1.1 1.1 1.6 1.5 1.4 1.5 2.2 0.0 2.0 1.6 1.7 O. O. 3.3 1.6 3.7 11.2 1.7 1.9 (Continuedy :Difference -001 -O.l 0.1 '102 0.8 __. _._~L~_- 0 Too shelf 8.2 1.3 2.8 1.6 Bottom shelf 8.1 1.2 1.6 7.4 * * :Difference * 0.2 0.0 —0.1 1.2 5.3 -6.8 ‘ ‘TBE‘"”:“"'" shelf 1.0 1.2 1.7 1.8 6.2 19.1 (continued) A Bottom Sheli: 1.2 1.2 1.6 3.0 8.8 8.6 12.3 Table III. 1‘ d A.Rw . O 2.0 12.7 3 16.5 12.7 “002 19.6 1.0 21.6 0 0.0 1.0 0.0 1.1 1.1 1. 9“ 0.0 1.2 1.2 0.2 4.0 -0.9 1.3 1.7 3.7 12.7 1.6 1.9 11.8 * 0.3 1.6 8.4 1.7 1.8 5.8 6.7 1.8 14.2 7.6 ”001 12.4 O 00 1.0 0.0 1.1 1.0 1.2 1.4 1. 0.1 1.2 1.6 1.7 4.7 1.7 “4.2 7.4 23. 1.6 5.7 2.2 10.6 15.2 A. -002 1.1 00 fly 0V 0. “001 n6 1. 1. n. flu 1.4 1.4 .6 0L r3 0.5 1.4 O. O 1.5 3.6 3.] 1.7 5.0 16.7 3.2 8.6 12.8 5.1 -0.9 O. 2.6 16.6 1.0 1.1 7.7 15.7 1.0 9.0 0.1 14.1 009 23.1 1.0 001 1.4 1.5 -002 1.3 1.1 0.0 1.4 .9~3.1. 5.7 O O. QVQV1. o o o Rvnu1. RVQVRV o o 0 9unv1¢ 1+ 2.7 0.1 9uRv . O 0.2 2.3 2.9 5.5. Qufiu 0.8 10.8 3. QC 6.6 9.6 18.3 ’001 a. n0 1U r0 A. (continued) A. 1.. (continued) Table III. Top Bottom shelf * shelf :Differehce * Shelf shelf 8 2 8 1 10.9 13.3 0 O 14.7 13.6 A. 9.. 6.8 20.5 2.3 4.0 5.1 5.8 15.4 0.1 1.0 2.7 3.7 0 8.0 10.6 1.4 01.]. 000 14.2 53 32 5. A. 2 0 5.7 5.5 0.0 -009 1.8 4.1 n/uqsa 881 A" 3.2 9.6 A. 00 3.3 1.3 3.3 6.6 ‘00 2.2 2.1 7.8 09 00 12 .0 0. "DA. .0 22 O. O 03 o. 005 *t. 12 o. 30 . O. .0 8R. .0 71 O. OcnO o. 01 l 3.1 2.6 5.7 0.3 2.0 434.1 0... 3000 O. O .0 0. 704.3. 0... Gu1.1.1. O. .0 O. 1584. O... Q51.1.l * *** 7.5.4fl .00. 007.4.fl O. 00 O. O 24.70 .00. 33.4.1. 0. O. .0 O. 9915 0000 3091 1L -001 0.8 3.3 1.7 5.0 0.2 -O.1 "Ool 1.4 1.3 1.8 10.0 1.3 we A... 1.2 4.2 0.3 0.2 2 5 6.0 0.0 16.0 A. 0.8 1.7 61112 3 3.6 1.8 4. p0 0071... 01 -O.7 0.0 1.6 1.7 3.0 6.8 1.6 0.1 0.2 0.2 0.8 3084.. 0000 1112 r5... 9“. 5.9 10.7 4.7 0.1 o 1 3 9.1 * o 1 5 o 13.5 0.1 A. 611 -007 26.7 20.0 * 1.4 3.7 4.6 “00?. 1.5 1.3 2.9 1.7 504.22 .00.. 020.40 0. O. O. C. .0 82733 .0... 12681.. O. O. O. O. 0. 18355. 0.00. 24.621 1 ***** 4.30300 0.... 30000 .0 0. O. O. 0 0.7327 0.... 1.nu1.1.1. 0. 0. O. O. O. 4.0.0aunc .0... 51112 fi(continued) 115 - (continued) Table III. TOP shelf * Bottom : Difference * TOP shelf 8 Bottom :Difference : shelf : shelf 9~A_:UAV 0 0 0 0 nVQHAIAV O 00 0 7.9uzu9. 0 0 0 . 1.9szu4. 1.9 4.6 7.8 10.8 ~9.4 0.0 0.3 0.8 16.6 0 l 1.6 00 1.3 2.1 2.0 4.1 1.9 “801 20.5 4.0 00 ~01 . 00 0. 37 oo 54. 1 0. 0. 0c? 00 95 ** DVD 00 24. Q90.» 0. 25 O. 80... 00 57 80 00 10 83 00 68 00 O. 62.... .0 88 ** 01 O. 60 . 0. .0 08 00 61 O. 07 00 21 1 -l.O —5.1 11.2 20.1 10.2 O;Z 15.0 2.0 2.9 1.8 2.0 3.4 0.0 1.8 1.8 6V1. 00 10 . 0. 0. menu 00 naps O. O. 64. 00 zucu * * :uau 00 10 . .0 1+9“ 00 nvzu 1. 5.4 7.8 8.6 3.7 7.0 3.7 00 00 .07 00 46 00 00 r34; 00 10 ll ** :02 no 01 - 0. 17 oo 65 .0 00 60. oo 56 48 000 001.. O. .0 821 on. 134 0. O. 00 869 0.. 135 *** 8.71 no. 000 .0 O 727 0.0 851.. 1 O. O O. reocnd 900 931 1 2.5 6.2 00 6.8 13.2 2.0 2.1 4.1 6.8 9.2 7.0 2.4 2.1 2.1 6.7 11 oo 02 — 0. 2A... 00 22 .00. 15 o. 2.4. ** 71 o. 60 - 00 0. 76 00 5.]. 00 0. 4.5 o. 21!. l Ru 00 4.0 . 60 .0 9u1l 1. .1.0 00 70 1 * * .b_b 00 nv1l O. 00 8AI 00 1.2u zoos 00 24. 6.3 A. 4.5 5.8 66 oo 33 0. .. 1I7. .- 80 1 0. C 75 .. IIAI llll ** 91 oo 20 .0 0. flange 00 51 .0 26 o. 81 zoos . . 00 RVQV 00 Red. 1‘9“ 00 00 Pens 00 4.9” 12 * * 1.:u 00 02 00 00 nvac 00 9s?“ 0. 00 1.1“ 00 “4.3 00 01 00 00 7C0 on 17 00 0. 11 oo 29 ** 5.0 o. 21 0. 0. r30 .0 4.2 0. 0. 70 00 63 1.8 11.7 2.4 1.8 2.3 3.3 6.2 2.? 14.0 82 o. 80 1|; 52 O. 23 CO 00 3A. 00 13. 1.1l * * 34. 00 nv1l 83 00 1‘9” 4.. 7.0 (continued) 11.4 * 5.2 2.1 3.7 8.1 116 - (continued) Table III. * Bottom Difference * T0p shelf Bottom ifferenoe sheIf shelf 805 .0 O4. .0 0. 3.2 0. 90 1 1|. 1L 0. 05 11 ** nod. 00 63 1 4.2 .0 4O 1 4.6 o. 03 21 2.1 12.2 14.3 19 1. 5.1 7 6 o o O O _ O. 0 n5. Ru 0 o r/ 5 1 2 O 9 O 0 no A... 1.9“ .x. t. 00 9 o 0 AU sl‘ Ru 0. O O 6 7 O. .0 2 Q. o o 6 8 24.5 000 nV1Izu C. .0 O. 8300 000 123 ... .. .. O. 078 00. 236 *** 998 000 143 .- OO .0 83.4.. 0.0 1|+ 1|. 06 121 O. O. 0. Ave A“. 04 000 11 2.7 5.6 4.2 11.8 13.2 12.4 —O.1 8.9 12.3 1.8 2.1 1.8 3.0 5.0 9.0 15.3 4.2 10.5 17.6 13.4 25.0 35.5 Au. 10.4 Go 7.8 10.6 7.6 8.7 13.4 2.2 2.8 -297 4‘ 4. O. 1.7 4.7 9.8 13.4 12.7 1.5 2 3 13.2 4.0 .0 06 O. 13 O. 85 O. O. 5.00 O. 81 1 ** "CO... C. Ronu .. O. O. 09 .0 06 1 7O O. 4.6 5.8 7.2 15.0 2.2 5.8 8.0 781 co. n/u AU 1|; - O. .0 9205 000 676 1 9.6 8.0 15.8 -004 1. -208 7.3 11.0 6.9 12.4 9 3.6 10.6 7.8 1.6 6.2 4.4 11.1 17.3 263.2 0000 1110 .0 .9 0a a 2006 0 O O O 41qunuqi 11. oo oo .- 4.8.RVRV O... 5611 1L **** 9~11quno O O o O nuzu9~41 o. oo C0772 .000 1123 1.7 4.8 5.2 7.2 6.4 2.4 5.3 7.7 5.8 .00. 0301 0. 00 O. O. .0 3,8371 .0... 22613 14*..881 00.00 63914. *** * .x. 0290.9 0.... 010v?~9~9~ 68959 O O O O 0 11133 0. O. O. O. .0 650595418. .0000 12.4.66 5.7 3.6 7.7 11.5 5.9 3. 6.2 4.5 '10]. 8.0 12.7 -001 301 400 . 2.2 {continued}, -11.6 12.6 5.5 26.5 14.9 — 117 - (continued) Table III. TOP Bottom shelf a: : TOp : Bottom shelf :Difference shelf :Difference * shelf 1nd“ 0 o 12 8.1 7.5 1.9 2.1 15.0 13.4 16.9 15.5 Ge 3 5.8 9.7 ”408 17.2 12.4 98 o o 01 O. O OVA. 0 o 12 8.2 O. QHAI ** 23. o 0 ~06 . . O. 0. ~02... O. 00,. n42 .0 O. 1.7. o o 72 1.2 414. 12 6.7 5.3 5.9 ”509 0. 2m 7.1 4.8 10.7 4.8 O. O. O Ru no 2.0 l l 1.9 3.0 4.6 2.9 2.0 1.8 4.9 1.1 3.5 2.9 7.7.11 000 QNAITI . . O O. O. 300» .00 Qunvv. 1 O O. .0 0,3.Re coo 55n0 *** GATE .0. 021 . O. O. O. 10.0—H... 0.0 626 1 O. O. 7Cyl 000 698 10.1 9.8 “004: 2.0 3.1 1.6 11.8 2.6 5.7 1.9 -300 —l.5 13 O. 315 1 03 0. 52 l. ** 7.05 O. 18 . 0. 81 0. 73 O. O. l.no 0. 62 1 5.5 3.8 5.1 4.6 3.8 6.1 6.9 3.6 7.2 5?» 37 2.7 “0.1 6.7 0.7 -O.8 8.3 11.1 9.0 10.3 1.7 1.6 RuA, . . l 3 C. C. no 2 O . 2 2 8 6 O O 70 5 * * no 8 . . 0 1L - oo .- QC 6 O C 3 no 1 O. O. 6 8 . . 3 l 1 4.0 2.4 2.8 2.5 1.6 2.0 5.7 5.4 1.7 A; 9.8 12.8 70 8.0 201- 2.1 1.9 -002 2.1 1.7 4.0 0.7 2.8 3.9 2.1 2.1 4.2 9~l.7.2u o . o . flvl.OvO .- O. O O. 14.19 0... 2222 .0 O. 0. O A5582 0... 112u9~2u **** Alzunvqu O... 1.41nv1. .0 O. O. O. 411.8v8 O... 413vl_3v O. O. O. 00 84.1001... 0... fiu7.1+:u 5.80. .00 011 O. O. O. A..12 00. 346 3.9 3.0 0.0 -008 4.8 O. 7.8 5.9 7.2 1.6 1.8 0.3 9.6 14.1 2.0 2.1 0.5 2.2 (continued) - 118 - (continued) Table III. Too shelf ottom shelf 1,: 1. * Top shelf Bottom :Difference :Difference * shelf 1.8 0.8 -O.5 2.5 4. 2.2 1.9 4.1 3.8 4.6 7.1 6.7 10.4 mi no -l.2 mg 4.8 -O.3 9.2 11.2 "C «I. 4.8 0.3 11.5 0.9 4.0 4.9 —1.6 18.0 16.4 6.8 0.5 5.6 6.1 3.2 6.3 10.3 1.4 “400 6.0 11.1 2.9 * -l.3 9.8 -106 5.5 1.6 1.2 6.0 9.1 15.2 15.0 606 -002 600 5.5 6.4 7.2 8.6 3.2 8.7 13.4 Cu O “603 7.8 2.7 9.4 -4.5 -3.9 8.1 PC n0 13.4 16.9 6.7 5.7 3.4 -8.6 ’005 7.2 8.1 -O.9 8.0 4.6 14.‘ 9.8 11.9 ).~ 5.6 2.3 7.5 5.8 2.6 9.3 3.0 7.0 2.8 11.6 14.4 7.9 -O.7 6.1 1.5 -105 -006 5.5 4.5 5.3 -O.8 5.3 -O.4 5.7 -203 5.5 4.9 11.9 7.0 3.4 7.7 4.9 -2.3 -006 7.8 5 1.3 3.6 2.5 5.5 0.7 6.7 7.4 4.4 4.9 4.7 2.6 1.6 3.5 2.0 3.0 2.1 4.6 4.8 1.5 2.3 O. O. O. O 7.4 6.1 5.6 5.3 6.0 1.1 2.7 5.8 0.7 2.7 5.4 2.7 7.1 8 5 O 1 5 o 3 I. O 0 who * 35 00 1nd . 1L4. 00 009 4.9, 00 46 v—w— - 119 - Table IV. Relative length of hours induction periods of t0p and bottom shelves of hot air oven. * T655; 505Mm1 : * TOp'V_W‘m555&;;W . shelf shelf :Difference * shelf shelf :Difference 18.0 : 17.0 : 1.0 * 17.0 15.5 : 1.5 17.0 : 17.5 -0.5 * 19.0 17.0 : 2.0 19.0 : 18.0 : 1.0 * 21.0 19.0 : 2.0 16.75 : 15.25 : 1.5 * 22.0 20.25 - 1.75 15.5 : 15.0 : 0.5 * 14.75 16.0 -1.25 15.25 : 14.75 : 0.5 * 22.75 19.0 : 3.75 15.25 : 14.50 ' 0.75 * 20.75 19.5 : 1.25 1605 3 15025 t 1.25 * 18.75 17.25 3 105 15.5 : 15.25 : 0.25 * 18.5 16.5 : 2.0 16.5 ° 15.0 - 1.50 * 18.0 17.25 : 0.75 15.25 : 15.0 : 0.25 * 16.25 16.0 : 0.25 15.75 : 12.75 : 1.0 * 16.75 15.75 : 1.0 25.0 : 19.50 : 3.5 * 17.0 16.75 : 0.25 14.5 : 12.0 : 2.5 * 20.25 19.75 : 0.5 11.5 - 11.0 : 0.5 * 19.50 20.0 ° -0.5 20.0 : 19.0 : 1.0 * 21.0 20.75 : 0.25 14.0 - 12.75 : 1.25 * 20.25 20.75 : -0.5 16.0 14.0 : 2.0 * 25.0 21.5 - 5.5 18.0 16.25 : 1.75 f 18.5 18.0 0.5 17.5 16.0 : 1.5 * 22.0 19.0 5.0 17.0 15.25 : 1.75 * 22.0 20.0 : 2.0 20.5 : 19.0 : 1.5 * 20.0 19.0 : 1.0 19.0 : 20.75 : -1.75 * 20.0 18.75 : 1.25 20 90 3 2000 3 OOO * 20 00 19 90 100 19.0 - 19.0 : 0.0 * 22.0 22.0 0.0 24.0 : 22.0 : 2.0 * 23.5 22.0 : 1.5 21.5 : 20.0 : 1.5 * 22.0 21.9 : 1.0 20.5 - 19.0 : 0.5 * 21.25 20.0 : 1.25 25.0 23.25 : 1.75 * 23.75 25.25 : 0.50 25.0 21.75 : 1.25 * 26.0 25.0 : 1.0 26.0 : 25.25 : 0.75 * 25.5 25.25 : 0.25 25.0 : 25.25 : 1.75 * 19.0 20.95 - -1.25 20.5 : 20.75 -0.25 * * Peroxide value of 2 considered end induction period. - 120 - Table V. Influence of temperature of hot air oven upon the rate of oxidation and upon the uniformity of results. Hours : Quadruplicate determinations* oxidation : a : b : c : d Trial 1 - 1006 C. 25 : 4.5 : 5.8 : 5.8 : 5.9 28 : 6.8 : 5.8 : 5.0 : 9.7 30 : 12.4 : 7.4 : 7.0 : 12.5 52 ° 17.8 : 15.5 : 14.7 : 15.0 . 700 C. 100 3 102 3 102 3 102 3 103 150 : 9.5 ° 5.7 - 2.2 - 14.8 155 : 20.5 7.5 5.7 : 15.2 160 : 42.8 15.7 19.0 56.9 Trial 2 - 1000 c. 22 : -- 1.9 : 1.6 : 1.7 24 : 1.9 2.6 3 2.7 o 5.5 26 3 508 4:07 3 306 3 308 28 : 9.5 8.2 : 5.6 : 7.0 135v C. 5 : 5.8 : 4.4 : 4.4 4.1 4: 3 508 3 609 3 602 . 7.3 5 3 804: 3 807 3 804 3 803 6 : 11.7 : 10.4 : 11.2 : 12.4 *Values reported in peroxide number. - 121 - L tion between oxidation of duplicate Trials determining varia Table VI. ken from the same shelf of the hot air oven. samples of fat ta A. Difference between a and b Difference Peroxide value our 3 H :oxidation between a and b Peroxide value Hours oxidation a 8.5 8. 2.3 8.9 22 0.1 8.9.05 .0. 04.0 2592R. .0. Rvnvzu 054.7. .0. Rvnu9~ 1 .0n0.8 929211 0. O. 0. 8.4.85 0.. nénupb £5257. .00 624A 1 157.1. .0. Qu?~ou nono.5 921192 9.5 8.4 17.7 8.7 24 0.5 4.1 0.4 5.2 12.8 4.6 22 3 O 8.7 15.9 8.7 24 22 8 1 7.3 4.2 16.9 7.7 7.0 15.9 91 31 11.6 14.8 7.7 2.8 1.0 18 24 14 14 15.3 4.5 19.8 14 14 18.1 1.7 19.8 28.5 8.7 19.8 1.5 18.5 4.4.0.5, .000 nonv1111 0.859211 0... RVRUOV7. 11119211 4.929265 0... Raouounu 1.1.1.11 4.4.4.4 11111.11 0. O. O. 1.0.7.R, O... nancnéno 154.:892 O... nonononc 11111.92 4.426 O... F5P5n9.o 1.1.1.1. 4.44.4. 1.1.1.11 15.6 14.1 1.5 14 15.6 15.6 14 15 1.2 0.5 6 5.8 .nJR492 7.8 7.5 0.7 0.9 4.4 5.6 15 0.4 0.0 0 431 8.1 0.5 14.0 8.4 15 7.5 6.5 1.4 0.2 5.5 4.1 11.0 11 11 5.9 8.4 1.4 1.7 4.5 4.5 14 17 17 14 12 2.6 1.0 15 2.7 1.7 1.5 1.7 14 5.5 2.1 4.8 7.6 4.0 0.5 954. nv1111 5.7 9.7 14.5 4.6 8.2 15.7 20 24 56 0.0 7.7 2.8 7.6 5.7 11.5 7.6 15.4 8.7 25.8 14 14 18 10.2 5.8 15.0 25.2 7.6 57.5 18.0 22 49.1 20.0 29.1 3508 4:304 26 48 6.7 2.8 10.4 5.9 8.9 11.1 59.5 76.8 0.0 7.9 9.4 17.5 7.9 11.7 14 15.0 20 24 56 2.5 5.0 0.1 14.4 14.5 55.4 19.4 19.5 26 48 5.4 0.4 49.6 46.2 5.9 10.5 1.6 2.0 15.5 17 11.9 25 5.0 4.0 21.5 27 9.2 14.7 54.5 25:1 49.0 55 21.5 17.5 54.5 45 59.2 1.6 40.8 59 48 1.5 5.9 10.0 1.2 4.6 11.6 15 19 25 2.8 0.1 40.9 58.1 1.9 12.2 17 4.5 7.7 21.5 _‘(continued) Difference between a and b 12.5 0.5 1.8 0.0 0.5 2.0 1.5 25.5 55.1 0.7 1.8 5.6 4.4 5.6 6.5 7.5 a 22.5 5.6 5.6 5.9 7.6 1.6 5.7 5.7 Peroxide value 40.8 Hours :oxidation 55 45 12 12 10 12 12 15 122 - Difference : between a and b 1.5 5.2 0.2 0.2 1.9 1.4 5.2 1.5 0.9 18.5 59.1 55.7 1.4 1.8 1.7 5.8 7.6 5.4 2.5 5.7 8.1 (continued) a 17.2 1.6 1.6 5.6 4.4 4.4 7.6 2.5 2.5 Peroxide value 26.9 50.5 Hours oxidation 27 59 48 11 11 12 11 11 12 15 15 Table VI. 54. 00 11 7.8 4.9 6.5 6.5 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 2.0 1.8 5.7 6.5 16 24 24 24 0.7 5.5 7.0 6.5 6.5 1.7 0.5 0.7 8.0 5.8 7.0 5.4 6.5 0.9 0.8 7.2 6.5 .20 no 24 24 24 1.8 1.0 6.2 1.9 1.8 2.6 5.5 4.5 5.3 14.1 9.8 5.0 7.9 6.1 5.5 14.2 6.5 6.3 6.5 7.9 7.9 7.9 7.9 7.9 7.9 7.9 2.9 1.4 1.4 4.7 5.5 1.6 2.4 4.0 0.7 7;7 7.4 12.5 4.6 6.5 5.5 11.9 6.3 6.5 6.5 7.9 7.9 7.9 7.9 7.9 7.9 7.9 24 24 24 24 24 5.7 7.9 2 0.9 0.4 0.2 5.5 4.4 4.9 4.6 4.8 5.3 5.5 5.5 5.5 5.5 5.5 2.1 1.7 0.0 0.8 6.7 7.4 6.1 5.6 7.9 5.5 5.5 5.5 24 24 24 24 24 rm; 0. 24 24 5.5 5.8 9.8 8 O 4.5 1.5 5.5 5.5 1.9 5.4 0.6 0.8 1.9 1.1 1.6 1.7 0.9 5.4 5.0 7.8 5.0 5.6 2.5 6.0 2.7 5.5 5.5 4.4 4.4 4.4 4.4 4.4 4.4 24 24 24 24 24 24 0.9 1.6 0.3 0.9 2.4 1.4 1.7 2.7 1.5 5.0 6.9 4.7 5.5 2.0 5.0 2.7 7.1 5.1 9.4 5.5 5.5 4.4 4.4 4.4 4.4 4.4 4.4 4.4 24 24 24 1.5 0.2 5.4 1.9 «I... 18 18 (continuedy 2.1 1.2 2.1 18 - 125 - (continued) Table VI. Difference Difference : between Peroxide value Hours :oxidation between a and b Peroxide value Hours oxidation a and b a 1.5 a 1.4 18 0.5 0.2 0.0 0.1 1.8 18 18 18 O O 1.5 1.5 18 1.7 2 O 2.4 2.6 2.6 8.7 1.5 0.5 9~ 16 2.7 6.1 16 18 18 3.26 1. s 5.1 14.2 18 2.6 E34. 1 o h‘— L — -.—_-1--.—- -— ~ .- —.-—-—-‘—m-—--~‘---b.—w—. ‘L-‘—-. 12.9 2.1 20 20 22 10.8 8.7 12.9 2 9.7 10.5 12.9 1.5 11.4 20 22 22 16 1.6 0.4 11.9 1.0 1.4 16 16 18 22 22 25 26 26 29 1.5 1.4 10.5 11.9 20.0 8.1 11.9 : 2 1.4 0.2 0.9 1.7 0.1 5.2 2.1 2.5 2.5 6.6 0.1 1.6 1.8 8.0 2 2.5 6.6 6.6 22 25 25 2.7 9 1.4 1.5 0.4 2.7 26 29 0.2 5.0 2.8 1.9 2.8 7.5 00 11 5.6 0 1.1 5.7 5.2 1.— 8 AU 0 . 4:08 1 6 A" O O 1 1 1 8 A...h O O 6 1|.“ __ _ _ 4. n0 .0 I. 7 1 o o l 5 1 Cr. C . 5 l 1 8 8 o 0 6 6 41 4. 3 3 0.0 1.4 26 29 1.6 1.6 2.0 0.2 1.8 5:9 o. oo .0 1.5 26 29 29 51 54 54 26 28 28 1|. no 0.2 0.1 1.8 1.6 5.0 0.9 0.1 (3 0:1 2.0 7.1 2.0 2.0 26 26 28 1.9 2.0 6.0 1.1 12.8 7.1 5.7 0.2 1!" 2.1 2.2 2.0 4.5 2.8 1.7 2.8 1.7 A; 4.4 1.9 9.0 6.5 10.5 1.5 0.0 0.0 52 2.2 1.2 0.2 0.1 8.1 11.5 10.5 52 1.7 1.7 26 26 28 10.5 52 1.7 1.7 1.5 1.7 26 28 1.8 2.1 2.2 2.1 0000 0101 33F1x0v 0000 4.2700 0000 5372 O 241 woman/N O. 00271 O C O O 0012 80.53 0000 1459 1L722 0000 20077 8. 22 2&55 0.5 1.8 2.1 24 2.0 1.7 5.8 8.0 n0 n6 9.7 28 7.9 1.7 5.0 2.1 5.8 5.8 9,7 26 26 .lw .l. 10.8 '(continued) - 124 - (continued) Table VI. Difference between Difference : Hours Peroxide value :oxidation between Peroxide value Hours oxifiation a and b a a and b B. 1.9 A. n/w 1.5 5.2 7 5 1 5 . . . . O 1 no 2 8 5 Q,» 7 O . . O 2 9 A: 9 CU O 0 2 . . . O 5 8 8 2 1 6 8 8. AU 2 2 2 "av 00 A~ 2 6 6 0 O . . 1 l 2 1 1L 7 A“ 8 O O . 0 2 A; 5 GO 1 .nronvnn . . . O 3 3 8 2 l 6 6 8 O 2 2 2 3 Cu 2.5 26 12.8 50 55 21 Ores 54. 48 55. 26 28 v 9.5 4.4 1.6 11.. 6.7 6.7 50 59 52 5.6 10.8 11.0 5.8 6.7 0.8 0.5 0.6 6.2 42. 26 15.8 1.2 15.0 1.2 4.2 5.9 5.4 26 28 28 50 5.8 8.7 15.9 6.4 9.0 9.0 14.4 50 5O Ow 8. 9.0 14.4 4.5 0.6 0.6 52 4.6 1.8 0.2 9.8 16.2 9» 2.5 2.5 24 14.4 1.7 4.9 15.4 2.5 7.5 2.9 2.5 4.5 4.5 3 26 28 5.0 5.4 10.0 26 O8 00 20 O3 00 82 019" .0 Ono 1 84. 22 79 o. 30 002 00 62 01.. 00 05 l 84. 22 4.6 3 26 1.1 4.2 5.] 24 26 28 28 4.5 5.2 1.8 28 22 22 1.2 0.8 5.2 6.0 1.6 5.2 1.8 1.7 0.2 0.0 22 22 Total samples 1.7 1.7 527 Average 7.95 7.84 2.6 8.7 9.7 10.3 2.4 6.1 11.4 1000' uadrnplicate* n "i 2.7 5.1 12.9 - 125 - 2.9 10.8 14.? 20.0 L 16 Hours 18 Accuracy of oven method in detecting the end of oxidation : induction neriod. Trial number Table VII 0 . . . . m * . 414111QVSUV.R\QVAiflcncRc117.2,317.8,7.3;819‘7.5 Q 7.0,R;9.419n4nv . O O O O O O O O O O O O O O 0.. O O 0.. O O O O 0 O 0’. O O O“. 11119~1u112unu111111Ru9u9~11acnfi1111zuoi119~0v111.4.812:badnonopa 1 l .I. u 1 1“ 11. H 1 O O O O. 0 O O 0 CI 0. .0“ O. C O .- O .0 O.“ O. C. O. O. O. O . . _ 1 M n m nonunoq,ounc9~8ca chc9w9.7.41:uR,Rq3,6;7.R,QVR,4.R.Q 2.9cRVO.A;4 oooooooooooo~oo0040000000000coococo 1.11112o112u:u1_1.11918;9~11A.11111.9c7.113.7.121.9~419~Aiquocngzo 1 v 1“ 1 1 1.. fl . 1 .. . M O. O. .0 O. .0 O. O. .0 O. O. O. O. .0" O. O. O. 0... O .0 O O. O O. O O. O O 0 09m .0 . 1 w . . . 262m48184690_9430w7995988521474929r3 .00.000.000...ooo.oooo.oooooooo_ooco“. 111iZJQJZUQNA1RV1111118311RURVQU11?~4.S43.5.OVOHZVONQUQLRUQUQuna%~ _ 111. . o 1 u 1 1 O O O O. O .0 . .0 0. 0 C. 0. ..~ 0. O. O. O. O. O. 0 fi 0. O. O. . . F . A __ h - . o a a 573640c415581”05217.139..”100755585cé785 . O O O O O C O C C 0 .¢. 0 O O O O O .n. O C O O C C 0.. O O O O 11119uRc111192:u1111117J9HA19c91119.41719~RuRc0V11KUQVRZRVB.anné _ 1. .1 . n _ . O O O O. O O a. O O O .0 O O. 0.... O O O O. O O. O O ‘ _ ._ 1 RVRV9~:.R.QV1141RVQV1LA.RVRVOV9~AVQU 9LALRVRVWM4.6.RWOVG.RVOL226 119~9~?~9~2u2c9~9~2o2u9~9czuzuouoc .ZLQMQHOHQuncnanKnoncncnonohc m _ . C O .0 O. O O 0 00—0. 0 O. .0 O 1 . . . . . 9” no A. ,b no 7. 16 To “0 1 _ . -‘ <0 -.--—--‘—a 11.0 11.1 3.4 6.0 1.8 4.3 8.0 5.6 3.2 4.3 5.0 2.3 2.9 4.3 15.0 5.8 6.7 2.0 2.3 4.6 2.9 4.9 10.0 0 w 4 O ‘--M——“. —— ‘_.___ 0 r3. ‘ A‘-..' - 7.3 25.9 2.2 1.7 7.3 13.4 J. 28 30 22 26 m"«-1..--=..---_2.§--_-_-i 26 28 J. 12 ll * Values renorted in oeroxide numbe - 126 - Table VIII. Precision of the oven and aeration methods of detecting the end of the induction period. Oven method* : Aeration method* HELif's'" ”g" u' H" ' n 7.1-“... ‘ : hour 3 "'7“"“ m“? “"0"" O“ oxidation : a : b : oxidation : a : b "I" ' ' " “""""" "m““* '17ng III ““"'m"""" -......-.._..._... 11 . 0.8 : --~ : 19 : 1.3 : 1.3 16 : 2.3 ' 2.2 : 22 z 4.1 : 8;3 22 : 8.5 8.9 : 23 : 7.2 : 11.2 23 : 9.3 6.5 : --- ' --- ° --- 25 ° 8.9 8.3 : -—- --- --- Trial IT 11 : 0.7 --- : 19 5.3 . 6.7 16 : 4.7 : 2.3 . 20 . 11.8 ° 11.6 22 : 3.8 : 9.0 : 21 21 9 ° 18.6 23 : 10.4 ° 6.2 --- --- --- 25 : 9.1 14.7 : --- --- --- Trial III 18 : 2.7 : 3.6 : 22 . 1.6 - 1.9 22 z 4.6 : 5.2 : 24 : 8.1 9.0 24 : 8.7 - 9.3 : --- ° --- --- Trial IV 10 : 1.2 : --- : 16 . 2.0 : 1.8 11 : 1.6 : 1.7 : l7 : 4.5 : --- 12 ° 3.6 - 3.6 : 18 11.8 13.7 Trial V 10 : 3.9 4.4 : 13 1.7 : 1.0 11 : 4.4 5.8 : 15 6.3 : 2.5 12 - 7.6 5.6 : 16 - 14.6 19.1 Trial VI 13 : 1.6 : 1.6 ° 17 1.5 1.6 15 : 2.5 : 3.4 : 19 : 1.4 2.5 16 : 5.7 ° 7.3 : 21 . 4.3 8.6 Trial VII 13 3.8 2.6 : 15 . 1.4 1.7 14 5.1 5.5 : 16 : 1.6 1.4 15 7.3 6.7 : 17 6.2 6.2 Trial VIII 11 1.3 1.1 : l9 . 1.5 1.8 12 4.5 4.1 : 20.5 : 2.4— 2.4 13 3.7 . 4.4 : 21 8.4 10.1 Trial IX 11 3.9 5.3 : 18 : 4.8 3.6 12 : 5.3 : v.5 . 18.5 : 5.7 ' 5.5 13 8.4 : 8.1 : 19 : 16.4 : 18.7 Trial X 14: 3 1 o 7 3 1 o 4: 3 20 3 6 00 3 1 o 6 16 : 1.7 : 2.7 : 22 : 1.9 : 1.7 17 : 4.3 : 3.8 : 24 : 3.7 : 7.3 Trial XI 11 : 1.3 : 1.5 : 18 : 1.7 1.8 12 : 2.1 : 4.0 : 19 : 10.4 : 4.3 14 ~ 7.6 : 5.5 : 19.5 4.8 8.1 * Values reported in peroxide numbers. - 127 - Table IX. Influence of drying the air upon the rate of oxidation of fat by aeration method. Hours : Control* Air dried with H2§0ig oxidation : a b" a c d Trial I 12 1.2 : 1.1 1.2 : l 2 1.4 : 1.3 16.5 1.6 : 1.5 1.9 : 3 8 : 17 1.7 12.7 : 3 5 : 17.5 15.0 : 3.5 19.5 5.8 : Trial II 19.0 : 1.0 : : 5.0 : 2.3 : 19.5 : : : 1.3 : 1.8 : 20.0 2 : : 1.0 : 5.6 2 20.5 : : : 1.0 : 4.7 x 21.0 : : : 1.1 : 6.7 : 22.0 : 1.3 : : 34.1 : 5.9 : 24.0 : 1.7 : : : : 26.0 : 1.3 : : ° : 28.0 1.5 : : : 30.0 4.4 : : Trial III 20 : 1.0 : : 1.2 : : 1.1 : 21 : : : 1.2 : : 1.6 : 23 : 1.3 : : 1.4 : : 1.6 : 25 : z : 1.4 : : 1.5 : 27 : 1.5 : : 2.2 : z 2.3 : 29 : 3.1 : : 7.0 : : 7.0 : 31 - 2.1 : r *Values reported in peroxide numbers. Table X. Comparison of rate and extent of oxidation with hot air oven and aeration methods. Hours Trial I* Trial 11* oxidation Oven Aeration Oven Aeration 12.0 4.7 --- --- -__ 13.0 --- 2.5 --- -_- 13.5 --- 8.1 -—- --- 14:00 8.7 --- --- ——-. 14.5 -—— 25.8 —-- --_ 15.0 —-~ --- 1.5 1.5 15.5 --- 3.0 -_— --_ 16.0 9.3 32.8 --- --- 17.0 --- 57.7 1.8 1.9 18.0 10.3 76.2 --— --- 19.0 -" 4905 3.9 --- 19.5 --- --- --- 4.3 20.0 13.5 87.4 --— --- 21.0 --- 113.2 --— --- 21.5 --- --- 10.4 30.2 22.0 18.4 103.4 --- --- 23.0 -—- 130.1 10.9 48.0 24.0 17.5 156.2 --— --- 26.0 29.4 153.9 --— --- 27.0 --— --— 18.6 82.3 35.0 --- --- 26.3 143.9 36.0 38.5 162.5 --— --- 39.0 --- --- 36.5 168.3 43.0 --- ~~~ 44.3 182.2 48.0 53.5 196.4 46.3 217.6 60.0 --- --- --- 174.5 72.0 --- --- --- 184.1 84.0 --- --- -—- 242.5 9600 --" "" "" 13904 108.0 --- --- --- 124.5 120.0 --- --- -..- 109.5 132.0 --- --— --- 59.3 144.0 --- --- --- 64.1 * Values reported in peroxide numbers. 129 - Table XI. Influence of heat treatment of butter oil upon its stability toward oxidation. Trial Hours : Peroxide value number : oxidation : Control (”6&430 min. 22§¥30 min. 1 : 22 : 2.3 : 5.8 : : 23 ° 4.9 : 7.1 : z 24 4.3 - 10.5 : : 25 6.8 : 12.8 : : 26 - 13.0 1047 : 2 : 26 z 1.3 : 1.4 : 7.9 : 28 : 1.4 : 1.5 - 12.3 : 3O : 6.7 42.3 39.5 : 31 : 2.9 : 2.9 44.2 : 32 : 2.2 : 2.9 55.8 : 33 : 12.9 : «_8.0 72.8 3 : 32 : 1.2 : 1.5 : 6.3 : 33 : 1.5 : 1.7 : 14.6 : 34 : 2.3 : 1.7 : 19.4 : 35 : 2.1 - 4.8 : 20.8 : 36 : 3.9 2.6 29.1 : 37 - 14.4 7.9 36.5 4: 26 3 103 ”" 302 27 z 1.3 --- 16.0 28 z 1.3 12.7 30.0 29 - 1.3 1.3 19.1 30 1.2 1.4 ~-- 31 1.4 a 1.5 83.4 32 1.7 1.8 --- 33 1.7 2.? 67.3 - 130 - Table XII. Influence of high temperature pasteurization of cream upon oxidation of butter oil. Trial : Hours : Peroxide value number : oxidation : 62.856.—30 min. : 8507C.-5 min. 1 : 12 : 0.9 : 1.3 ‘ 14: 3 105 ‘ 106 16 : 2.7 : 3.3 18 : 6.1 : 9.3 20 : 10.9 - 11.9 22 : 12.7 - 12.3 24 - 12.6 22.5 2 : 13 : 6.0 4.6 ° 15 : 11.5 7.5 16 : 10.2 8.6 17 : 14.9 13.7 18 : 15.0 12.8 19 : 18.2 : 13.0 20 : 23.6 : 15.1 3 : 12 : 1.0 : 1.0 : l4 : 1.1 : 1.1 ' 16 3 103 3 103 17 3 105 3 105 . 19 3 105 3 107 3 21 ° 301 3 205 23 4.8 : 5.2 4 : 10 : 0.9 1.3 - 12 : 1.6 4.6 14 z 2.4 10.6 15 3 603 3 1602 16 : 8.0 22.1 17 : 12.3 23.9 5 : 12 1.6 : 1.5 - 13 2.0 : 2.6 14' 3.6 3 106 15 5.5 - 5.0 16 6.6 6.1 17 8.3 7.6 18 10.6 9.6 6 : 11 0.7 0.9 - 13 0.9 1.0 3 1.6 3 10?. 103 z 18 : 1.3 : 1.6 ° 20 ° 2.0 4.2 22 4.8 8 6 23 6.3 6 l <1) i-J V. . Rout inu C. Peroxide value --M ‘_*-_-_.‘ k; 0 0 Hour s : oxidation Table XII (continued) Trial an «I. P. 9006000- 6208000'30 min. 0.8 12. 21.1 A. 7. 7.... 70 8 0.9 1.9 1.2 19.3 041.1.3V1.4.1 10 12 14 16 18 20 22 10 12 14 16 18 21 10 12 14 16 18 20 22 number A. 9w _b 1. 11 13 15 17 19 21 7.1. 9H 1.0 1.0 1.2 5.7 12.3 (continue677 1.1 1.2 10.0 2.0 5.1 8.0 12.6 23 ll 13 16 18 20 22 24 11 13 15 17 19 21 23 Table XII. (continued) Trial : Hours : Peroxifie value number : oxidation;_jA_63.8UC.¥30 min. : 6%7- 15 min. 1 : 11 : 1.0 : 1.3 : 13 : 1.? : 1.4 : 15 : 1.2 : 2.2 : 17 : 1.5 : 5.7 ° 19 : 1.9 : 7.4 21 : 3.9 : 15.7 23 : 9,9 20.6 2 3 11 3 009 3 100 z 13 : 1.1 : 1.2 ° 16 z 1.2 : 2.0 18 : 1.6 ° 2.5 20 : 2.5 6.8 22 : 4.2 14.8 24 - 10.0 18.6 3 : 11 0.9 : 1.2 3 13 lo? 3 1.94: : 15 : 1.4 : 1.7 : 17 z 2.0 ° 4.1 : 19 : 5.1 8.8 ° 21 - 8.0 - 15.1 23 12.6 21.4 2271:: - 15 1111110 1 : 10 0.9 : 6.7 : 12 1.9 - 16.0 : 14 : 1.2 17.6 - 16 : 1.9 . 18.0 18 3.0 : 33.4 21 12.3 : 33.1 23 19.3 : 44.4 2 : 10 : 0.7 : 0.9 ° 12 z 0.9 : 1.1 14 : 1.1 : 1.4 16 : 1.5 : 1.8 18 : 1.9 ° 3.9 20 : 3.3 5.2 22 : 7.5 12.4 (continued) Table XII. (continued) Trial : Hours : Peroxide 32146L111-- number : oxidation : 62.800.-30 min._‘: 22%;30 min. 1 : 12 : 0.9 : 1.0 : 14 : 1.5 : 1.4 : 16 : 2.7 : 3.4 - 18 : 6.1 - 6.2 20 - 10.9 : 10.3 22 : 12.7 . 16.4 24 : 12.6 17.8 2 z 13 : 6.0 : 6.1 3 15 3 1105 3 1802 : 16 : 10.2 : 12.2 : 17 : 14.9 : 13.9 3 l8 3 1500 3 908 : 19 : 18.2 : 12.8 : 20 : 23.6 : 17.6 3 : 12 z 1.0 : 2.9 ' 14 : 1.1 : 9.6 16 ' 1.3 - 12.2 17 . 1.5 11.6 19 3 106 180? 21 3 301 . 1705 23 ° 4.8 . 26.8 4 : 12 : 1.6 . 11.7 ' 15 3 200 . 15.7 : 14 : 3.6 ° 18.5 : 15 : 5.5 13.3 ° 16 : 6.6 17.4 17 3 803 . 2106 18 ' 1006 . 1800 5 : 10 : 0.9 : 6.0 : 12 . 1.6 : 17.6 : 14 : 2.4 : 18.9 : 15 : 6.3 : 17.3 ° 16 8.0 : 25.8 17 12.3 : 24.6 6 3 11 3 007 1.2 : 13 : 0.9 2.0 3 16 3 102 604 : 18 : 1.3 11.5 ° 20 : 2.0 20.8 22 z 4.8 23.4 23 ° 6.3 21.1 (continued? Table XII. (continued) Trial : Hours : Peroxide value number : oxidation : 62.8OC.—30 min. : 223—30 min. 7 : 10 . 0.9 : 22.2 3 12 3 104. 3 53.3 : 14 : 1.7 : 31.9 : 16 : 3.3 : 29.0 : 18 : 11.3 : 27.2 : 20 : 14.8 : 45.2 - 22 - 31.9 : 44.5 Table XIII. Influence of pasteurization holding time of cream on the oxidation of butter oil. Hours : Peroxide value " oxidation:90.6°C.-flash:90.6OC.-5 min.:90.6°C.—l5 min.:90.6°C.-30 min. Trial I 18 . 1.8 1.8 - 1.7 : 1.6 21 3 30? 306 2.] 3 207 22 : 8.6 : 4.9 : 3.6 : 3.0 23 3 70] 701 70? 3 307 24 : 12.5 9.4 6.9 ' 4.6 Trial 11 24 : 2.4 2.4 : 3.0 2.1 25 3 402 307 3 404 205 26 : 4.7 4.9 : 4.6 3.8 27 : 6.6 5.0 6.7 4.1 28 3 12.] 903 1105 704: Trial III 19 : 1.1 : 1.2 : 1.3 : 1.2 21 3 103 3 104' ° 105 3 104 22 : 1.6 : 1.5 1.6 ° 2.6 24 : 3.3 : 2.1 2.0 2.1 25 ° 4.2 : 3.6 3.9 3.7 Table XIV. Influence of heat treatment of cream upon metallic induced oxidation. Metal added before pasteurization. Hours : PF‘I‘OVi (19 V81 :29 oxidation : Control :62.800.-30 min.: 85”C.—f1ash :90.6UC.—flash :6?.8°C.-30 min.: 5 p.p.m. Cu. : 5 p.p.m. Cu.: 5 p.p.m. Cu. Trial I 20 : 2.0 : 9.7 : 1.9 : 1.8 22 : 2.3 : 12.6 : 3.0 : 2.5 23 3 609 3 13.2 3 501 3 400 24 z 8.3 : 17.7 : 5.5 : 5.7 25 : 9.2 : 25.2 : 6.1 : 8.8 Trial II 18 : 3.1 7.1 : 1.6 : 1.6 20 : 3.9 : 11.7 : 3.3 : 1.8 21 : 8.3 : 10.7 : 3.9 : 3.3 22 : 8.3 : 14.1 : 5.? : 5.0 23 : 10.1 ° 14.2 : 6.5 : 5.3 Trial III 15 : 9.5 : 15.9 : 10.1 : 7.1 16 : 5.7 : 14.0 : 4.8 . 4.1 17 : 7.5 : 14.8 : 8.3 : 4.8 l8 : 12.8 : 18.7 : 7.6 5.8 19 3 2007 3 2600 3 1603 1099 Table XV. Influence of heat treatment of cream upon metallic induced oxidation. Metal added after pasteurization. : Peroxide value Hours : Control :62.8”C.—30 min.: 85”C.-flash :90.69C.-flash oxidation:62.800.-30 min.: 5 p.p.m. Cu. : 5 p.p.m. Cu.: 5 p.p.m. Cu. Trial I 16 : 1.7 : 3.4 : 1.7 : 1.6 29 : 1.9 : 5.9 : 3.1 : 2.1 21 : 4.4 : 6.3 - 5.6 : 4.2 22 - 6.4 - 9.1 7.9 : 5.1 23 9.3 13.5 9.0 - 6.7 Trial II 17 : 1.5 : 5.2 : 1.8 : 1.8 20 ° 2.1 : 7.7 : 2.8 : 3.4 22 : 3.9 ° 8.3 : 5.5 : 5.1 23 : 5.6 . 10.7 : 8.0 : 5.6 24 : 11.6 17.5 : 8.6 : 10.1 Trial III 22 3 201 3 60A 3 109 3 109 24 3 202 3 909 3 205 3 300 26 z 3.5 : 12.5 : 6.1 : 6.5 27 : 4.8 : 16.7 : 7.8 : 12.4 28 : 10.8 : 23.9 : 11.6 : 11.9 Trial IV 20 : 5.7 14.3 : 9.0 : 7.4 22 : 6.3 . 12.5 . 8.3 : 6.5 23 : 7.9 : 12.3 . 12.0 : 6.9 24 : 10.8 : 15.5 : 11.2 : 8.3 25 : 18.9 - 30.2 o 13.3 ° 10.8 Table XVI. Influence upon oxidation of adding "Avenex Concentrate" to butter oi1.* Hours : Normal butter oil 2 5 ppm. copper oxidation : No Avenex : Avenex : No lvenex : Avenex Trial I 19 3 107 3 106 3 200 3 106 21 : --- : --~ : 2.4 : 2.9 22 : --- : --- : 2.5 : 2.0 24 : --- : --- : 4.4 : 2.7 25 : 3.1 : 2.6 : 6.4 3.0 26 : 3.9 : 2.3 : -~- : --- 27 : 3.2 : 2.4 . z 26 : 4.1 4.1 : ° Trial II 20 : 7.8 : 5.8 : 6.5 6.2 21 : 5.4 : 7.6 ' 7.1 : 7.6 22 : 9.1 : 7.2 : 12.1 ° 8.6 23 : 11.3 : 10.4 ° 10.8 ° 10.6 24 : 17.2 : 15.1 o 15.1 13.0 Trial III 22 . 3.1 . 2.0 : 4.2 : 4.9 23 : 3.6 : 3.2 : 4.4 : 4.7 24 : 5.0 ° 4.2 : 6.7 : 5.2 25 : 6.7 6.5 : 7.9 : 6.6 26 : 11.9 . 10.0 ° 11.0 8.8 * Values reported in peroxide numbers, Table XVII. Influence upon oxidation of adding "Avenex Concentrate" to cream after pasteurization. * ‘50urs ‘Norma1_o£e1m 2U£ZETIQQRQQQ--::II:: oxidation : No Avenex : Avenex : No Avenex : fivenex " "“"‘"“""“'fl"“fEIEI—I‘I 7“ 22 : 1.7 : 1.7 : 3.5 : 2.4 23 : -_- : --- : 3.9 : 3.3 24 : --- : --~ : 7.4 : 6.0 25 : 4.2 : 3.6 : 6.8 : 4.7 26 : 4.2 : 4.7 : 7.8 : 7.5 27 : 3.1 : 4.3 : --- : ~-- 28 - 7.9 : 5.4 : --- ° --- Trial II 21 : 2.4 : 1.8 : 3.6 : 3.7 22 : --- : --- : 7.2 : 4.0 23 : 3.5 : 4.4 : 8.6 : 6.1 24 : 6.4 : 6.9 : 10.7 : 8.1 25 11.1 : 7.5 13.5 2 10.4 26 7.3 : 7,2 --— --- Trial III 22 : 1.8 : 1.6 : 7.7 : 3.8 23 : --— : --- : 7.4 : 4.5 24.5 : --- : --- : 9.8 : 5.2 25 : --~ : --- - 10.8 : 5.4 26 : 3.0 : 2.2 13.5 : 8.0 27 ° 3.2 - 2.4 --- - --- 28 z 5.1 4.1 --- --- 29 : 5.3 : 3.6 --- --- _—-.—---«.----.—-.o—.—o~~.a-o-o»—o .0 - 4....- u...‘ a. u...— .4 a .a‘ .-'¢-a owo-u q—.n.—I-.....w arm-A...“ “—4. * Values reported in peroxide numbers. .. Stab: . . 1:: 1.2.4111. 1.1-25.1. 12.61.. «s.-- -1133: 9!... . . :11... 1.33:. .. , . .. - . £1... .-..,-.....w.a:..:.- . avian)... - .1 «.7. . 3.3.1.1.. ,. 1....-.1! VI L a...“ o E . S U M “U 8 R . . . . v u n u . . . l . _ .3“.-..\._L.v. i . , « . (. . 2. I . u 1 a . . I . . 4 . ‘ 4| - I ,‘. .Iflcl \ v. VI \ . a - . a t . . . .. . . h .1 u .. . . . x l C. .. I . .. . . v n n ‘ . l 1 ~ I . .. - . J a . . . . . . . . I ... H. t . t v u w 7 o . . . ‘ . . w. A 7 .4 . - : \ I o ., . . _ .u a . .A . . I . . , ‘ . . . u . .a I. o . . r . . s . . . . . I . . . . . . . _ . .. . . . . . \ _.... ._ . . .o . . a .. . . . . _ . . p . , x .1 . . . 1 . 0.! .. 1 . o s . . . . .. . ~ . u . . c y I x h. . e _ . . . . s 1 on. . . u . ». 3 . . . r . . ..- _ . . . . .m a r .. . Ml IGAN STATE UNIVERSITY LIBRARIES 0 3 1293 3056 2783 CH