A STUDY OF THE INFLUENCE OF SOLVENT AND MOISTURE IN THE EXTRACTION OF LIPIDES FROM PLANT TESSUE Thesis for the Degree of M. S. MICHIGAN STATE COLLEGE Raul Antonio Zambrana {Heppe 3 194’? . a w...— .p—‘mu—od -. n‘a-n— -J—_ _- —.... W'f: Fry." _ ”a r .51 I ‘ COPE OF THIS DOCUMENT ARE AVAIIABLE Oi} Iz'iICROF 1le BY COIJTACTBQG 1."; I W... i, - w/¢t77',‘1/i7'5/t/,q D -1B a: ‘1 l l fl, 1 d / l‘ ‘ 1; . 1 I. "M This 18 to certify that the ‘« .,._ ’1‘ thesis entitled ' ' ‘ '. . a $- W % j} l . , h I I "L . l I , r Q ~ . '1' 4 . f '. . ‘ g S I z . , p . , 51% MM W * f _ . presented by V _ ‘1 , , t has been accepted towards fulfillment of the requirements for I 4 - t > I, /' A -_ ,, ,W. . _ ,‘ ' Major professor . f ‘ ’ ' A " Date 21’41 A]; (ZZZ 1 I “-795 ( "‘ s 't l ’ y f . '. ' A I t I _ ‘ l " t ‘> . _ \ I t'l "‘/ < i- T r’ ‘ ' . .V '1 \ . ‘1 I l 1' :i . Is 4‘! W) ‘ I. l i 'x‘ f\ _ fun". ‘ \ I I ‘ ‘. “If" .' r ‘ \ I“\“Lt I F ,_ . ‘ I f '1 ' / A STUDY or THE INFLUENCE or SOLVENT AND MOISTURE IN THE EXTRACTION OF LIPIDES FROM PLANT TISSUE BY RAUL ANTONIO ZAMBRANLCHOPPE) A Thesis Submitted to the Faculty of Michigan State College of Agricultural and Applied Science in partial fulfillment of the requirements of the degree of MISTER IN SCIENCE Department of Chemistry 1947 f6/ :3 5/3" “ CH; ._ ... ACKNOWLEDGEMENT The author wishes to express his gratitude to Professor 0. D. Ball, for his advice and assistance in planning and conducting the experiments and in the preparation of this manuscript. 189101 I. II. III. IV. V. CONTENTS Page Introduction .............................. 1 Historical ................................ 2 1. Determination of Lipides by Solvent ex- traction .............................. 2 2. Refractometric determination of lipides 6 5. Commercial extraction of fats and oils by the use of solvents ................ 7 Experimental .............................. ll 1. Influence of moisture in the determin- ation of Total lipides by Standard‘ procedures ............................ l4 2. Influence of moisture in the determinp ation of Total lipides using various solvents .............................. 18 3. Influence of moisture in the determin- ation of Unsaponifiable matter by the Kumagawa-Suto Method .................. 22 Discussion of results ..................... 30 Conclusions ............................... 37 Bibliography 0.00.0...000..00.00.000.000... 59 -1... INTRODUCTION Numerous organic solvents have been used for various purposes in the extraction of lipides from biological tissue. In the quantitative determination of lipides, up to date, no solvent has been found satisfactory for use as a standard or so called official solvent. The implication is that each solvent does not completely extract identical materials (I). In addition to the specificity of solvent action many other factors have been studied which affect the yield of lipide extracted. In the quantitative determination as much as in the commercial extraction of lipides it is of import- ance to know to what extent those factors affect the gross extract or some particular components of the lipide fraction. The purpose of the research connected with this problem was to compare the values obtained in the extraction of "Total Lipides" and WNon-Saponfiable Matter" by several sol- vents in materials containing different amounts of moisture, and to determine how the presence of any moisture would af- fect the amount of gross extract and Non-Saponifiable Matter. The term "Total Lipides", as used in this paper, refers to the gross extract obtained by the different solvents. The method used in the determination of Non-Saponifiable Matter is the modified Kumagawa-Suto procedure (2). -2- HISTORICAL The application of solvents to the extraction or solu- tion of lipides has been done with different aims, such as their determination; their commercial extraction and their fractionation. l. DETERMINATIONIQE LIPIDES EX EXTRACTION Biochemists have tried for many years to find a solvent, a mixture of solvents or a sequence of solvents, which could be used as a standard for the determination of lipides in bilOgical tissue. Very outstanding works were made by Schulze and Steiger (3), maxwell (4), Rather (5), Hertwig (6), and Rask and Phelps (7), and from their studies it is now well known that the Official Method of the Association of Agricultural Chem- ists does not give an accurate measure of the lipide content of a material. Many solvents have been suggested for analysis of fats and comparative studies favor special solvents, depending upon the material under consideration and the conditions under which the work was carried out. Taufel and Standgil (8) working with flax seed used tri-chlorethylene, benzine, carbon disulfide, benzene, ethyl ether, acetone and chloroform. The unsaponifiable matter, the saponification number and iodine number were fairly constant in all portions. Trichloroethylene gave the highest yield. -3- Walker (9) determined fat in 30 different types of feed- ing stuffs. These were run according to the Official method, and both anhydrous ethyl ether and dichloromethane were used. He found four results lower with dichlorethane; twenty five were higher; and thirteen were within 0.10 percent of each other. On the average there were 0.23 percent more fat dis- solved by the chlorinated hydrocarbon. After this work of Walker, directions were sent to 17 chemists for collaborative work. About 400 tests were made on 27 types of feeding stuffs. The results showed that ethyl ether did not remove all the fat, and dichloromethane was not recommended because of difficulty of removal and its expense (10). Vizem and Guillot (ll) compared carbon disulfide, ethyl ether and petroleum ether. Carbon disulfide was found the best solvent of the three. Ethyl ether and petroleum ether are relatively poor fat solvents, the solubility of fats be- ing approximately the same in both. Solid glycerides are less soluble in ethyl ether than in petroleum ether. Dustman (12) made a study on the effect of certain sol- vents and of sequence of extraction in the removal of fats from bones of chicks. He concluded that: l) Acetone and ethyl ether removed approximately equivalent quantities of extractive, but the asetone extracted less ash; 2) absolute ethanol and 95 percent ethanol removed more ash than did either acetone or ethyl ether; 3) when absolute ethanol preceded acetone and ethyl ether as the first solvent the total extract was somewhat less than whhn the sequence was -4- reversed; 4) ether or acetone used after 95 percent ethanol removed only very small quantities of extractives, but 95 percent ethanol used after ether or acetone removed much more additional material. Igarashi and Joshitoyo (13) showed that some additional agents in the extraction of soybean, containing 3 to 5 per- cent of water, had a favorable effect. They added sodium chloride, sodium acetate, clacium acetate, acetic acid, sodium hydroxide, calcium hydroxide, calcium chloride, soybean oil fatty acids, and the sodium, calcium and magnesium soaps of the soybean fatty acids. Extraction with carbon tetrachloride, chloroform, car- bon disulfide, ethyl ether and petroleum ether of egg yolk, cheese rind and chromed leather showed that in general more extract is obtained and more real fat with carbon tetrachlor- ide than with any of the others (14). Harrison (15) working with fish meal used 12 different solvents, and extraction periods ranging from 4 to 52 hours. The purpose of these determinations was to find the rate at which the various solvents approached a maximum extract value, the gross extract value of the solvents on the meals as re- ceived, and the gross extract value of the solvent after the meals had been subjected to storage conditions that would lead to oxidation of the fat. All the solvents, except ace- tone gave very little if any increase in extraction value beyond 4 hours. In the case of acetone 16-24 hours were re- quired to reach a maximum value. From the standpoint of gross extraction the supposedly aliphatic petroleum hydrocarbons -5... gave the lowest values, and these in general increased with boiling temperature. Ethyl ether, carbon disulfide, and cyclhhexane gave quite similar values, slightly above petroleum ether, hexane, and heptane. It is of interest that cyclohexane, which is an alicyclic compound gives a value intermediate between the Open chain solvents and benzene, which is definitely an aromatic com- pound. Likewise, the chlorinated aliphatic hydrocarbon sol- vents were found together on basis of extraction value and give results higher than the unsubstituted chain or ring hydrocarbons. The cyclic ether, 1, 4-dioxane, gave the high- est values of the solvents tested. The oxidation of the meal decreased the extract value of both ethyl ether and pe- troleum ether. Kaye, Leibner and Connor (16) recommended isopropyl ether for the extraction of fat from feces. It has proven more efficient than other common fat solvents. Hieserman (17) found the Koch method to be the most efficient in the removal of the lipide material for the ini- tial extraction. However, he concluded that a large percent- age of non-lipide compounds, especially when cholophyll pigments are present, are included in the value obtained by this method. In the extraction of cottonseed with several organic solvents, Frampton and Webber (1) found evidence that the composition of the extract obtained with a given solvent will vary from seed specimen to seed specimen, and the composition -6- of the extract obtained with a given seed specimen is depend- ent on the nature of the solvent used. The quantity of ma- terial extracted with a given specimen is also dependent on the nature of the solvent, and it is suggested that the se- lection of any particular solvent as the standard or official solvent to be used in the assay of oil with cottonseed is arbitrary. Heat treatment of cottonseed reduced the quanti- ty of lipids which could be extracted with petroleum ether. The influence of temperature in several extractions was found not to be of importance. It is probable, however, that the most important factor contributing to the differences may be specified as biolOgical. Eash (18), nevertheless, work- ing with the same material said that the amount of extract obtained increases with the temperature of extraction. He used a mixed solvent of the pentane type, containing a mini- mum of isopentane, isohexane and hexane, and with a boiling range between 35-4000. 2. REFRACTOMETRIC DETERMINATION Q: LIPIDES For a rapid determination of fats the Refractometric method has been largely used. Mbnochloronaphthalene (19), and (20) and Bromonaph- thalene (21), (22), (23), and (24) are considered the best solvents for refractometric methods. Bromonaphthalene has the advantages over chloronaphthalene as a solvent as it is less volatile. Other solvents as bromobenzene, amyl acetate, isobutyrate, chloroform, and benzine have bien also used with good results. - -7... 3. COMMERCIAL EXTRACTION'QE FATS AND OILS B T US 0 SOLVENTS The commercial extraction of fats and oils from plant and animal sources started at the beginning of the present century, particularly in England, France, and Germany. Its application was the natural result of the requirements of Germany and the Low Countries for supplies of edible oils and of protein feeds for their livestock. In this country, which normally produces a surplus of edible fats and oils, the oilseed processing industry grew up around machines suitable for a decentralized system of relatively small mills located ithin the crop-producing areas. Only in a very recent years have American manufac- turers entered the solvent extraction field. The list of solvents used and suggested is very long. From the standpoint of commercial extraction the choice of a solvent for a particular extraction depends upon a number of factors. First, it must be a good solvent, i.e., it must penetrate into the cell structure of the solid material and dissolve out the soluble substances. It must be easily removable from both the oil and from the residue, leaving both in a marketable condition. Second, it should not cor- rode the equipment, thus avoiding tendencies toward contam- inating the oil with traces of certain metals. Third, it should be cheap. This factor, almost alone, accounts for the rather general use of relatively low-priced hydrocarbons where at all possible, despite hazards connected with their use (25). Fourth, inflammability, an fifth, toxicity. -8- Carbon disulfide has been discarded as a solvent (26) for common extractions on account offire hazards and danger to health. Chlorinated hydrocarbons are good solvents but they extract undesireable resins, albuminoids, coloring matter and oxidized oils. Trichloroethylene is the most efficient chlorinated solvent but years of experience with it led to its abandonment, because of inferior products and its bad effects upon the health of the workers. Benzene is well suited for extraction of castosroml press cakes, but its vapors cause poisoning. Well purified and odorless pe- troleum benzine, 42-105°, 1: beyond doubt the best solvent for common extraction. The ordinary grades require careful purification with 660 sulfuric acid to remove aromatic hydro- carbons. When thus purified, it extracts only oils and fats, free from impurities, and it may be completely recovered. Ehrlich (27) gave many advantages of o-dichlorobenzene such as stability, low fire hazard, easy volatility, great avidity for fats, and its ability to extract them completely even from substances containing an appreciable quantity of water. Dichloroethane (28), and Dichloroethylene (29) and (30) have been used with success, giving an extraction residue suitable for animal feeding. Many solvents are used in the extraction and simultaneous purification of oils. Seiyu (31), and (32) used a mixture of carbon disulfide and ethanol, or methanol and a mixture of chlorinated hydrocarbons, such as trichloreothylene or dichloroethylene, with ethanol. sifter extraction the solvent -9- is separated into two layers; the lower one contains pure oils, while the upper one contains impurities. The use of solvents to obtain from semidrying oils a fraction more highly unsaturated then the original oil has been investigated. Furfural was found to be the most suit- able selective solvent of those examined (32)-Aniline, polyethylene glycol, furfuryl alcohol, and monoethyl ether of ethylene glycol have also been used in this respect. Within recent years, petroleum companies have stepped up their production of special solvent naphthas having.par- ticularly narrow boiling ranges, low evaporation residues, excellent stability, and other properties required of good solvents. In the soybean industry the commonly used solvent is essentially a normal hexane fraction boiling from 146 to 1580?. In most EurOpean plants, a gasoline boiling from 160 to 194°]? is need. Apparently there is reported only one commercial soybean extraction plant in the world using any solvent other than a petroleum cut. The exception is the Manchurian Soybean Industry Co., in Dairen. The solvent is 99.8 percent ethanol, and it is used at 80°C under pressure. Considerable research has been directed toward the use of methanolbenzene and ethanol-benzene mixtures for soybean extraction in cases where phosphatides recovery is of impor- tance. Still other solvents have been considered. Benzene was used in some of the early installations. Methanol and ethanol show partial miscibility with glyceride type of fats and -10- oils at low temperatures. The oil dissolved in the solvent- rich layer is more highly unsaturated. This has been suggest- ed as a possible method for separating the drying and non- drying constituents present in semidrying oils, soybeans and corn oils (25). In using solvents for extracting lipides, whatever the purpose is, it is necessary to consider factors such as se- lective solvent action, temperature and rate of extraction, moisture content, age and oxidation of the material to be extracted. -11.. EXPERIMENTAL The experimental work was divided into three parts. First, the determination of total lipides in several soybean meal varieties, containing different amounts of moisture, by a) the Official Method of the Association of Agricultural Chemists, b) A.0.A.C. methods followed by the Koch procedure, and c) the MOdified Koch procedure. Second, the study of various single solvents in comparison to the two above stand— ard procedures. Third, the determination of the unsaponifi- able matter by the Kumagawa-Suto method, using the modified Koch procedure and various single solvents in the initial extraction of the same material. The materials used for the experimental work were Man- darin and Manloxi soybean varieties, generic name 303a max, and the common or Navy beans, Phaseolus vulgaris. The soybean meal was prepared from the above soybeans by finely grinding in a Hobart electric mill. Using a number 20 sieve, the above meal was sieved and the residue ground again. This ame operation was repeated four times and then the total mixed meal was ground together. This procedure gave an homogenous sample. The meal was allowed to remain in contact with the atmosphere for two days and then was - placed in an air tight bottle to be stored until needed during the course of the experiment. MOisture determinations were first run on each material by weighing out duplicate 5.0 g. samples in calibrated alumi- num dishes in a Brabenderis balance and drying the weighed -12- samples in a Brabender's MOisture Tester set at 100°C. To avoid deterioration the samples which were to contain some moisture were dried at 80°C. The rate of drying at 80°C. was first determined by plotting the time of drying against the percentage moisture read on the scale of the instrument. The curves shown in Figure I were obtained. is may be ob- served those curves gave an idea regarding the approximate time necessary to remove a desired amount of moisture. The rate of water absorption from the atmosphere by the dried and partially dried meals was observed byleaving the material in contact with the atmosphere for several min- utes and redetermining its percentage msisture. shown in Table I. TABLE I Rate of moisture absorption of soybean meal in contact with the atmosphere. 1 V The data is me in contact Percentage of moisture re- laterial with the atmos- moved (air-$§y basis here. (minutes) *100 marin O 7e88 leg? 5e91- ' 1 7.87 1.96 5.90 " 2 7.86 1.96 5.90 ' 5 7.79 1.94 5.87 n 10 7.67 1.92 5.76 Manloxi 0 8.25 2.06 6.19 ' 2 8.13 1.99 ' 5 8.11 1.96 6.16 " 10 8.10 1.95 6.10 Navy 0 11.55 2.84 8.50 n 1 11.34 2.84 8.50 7 2 11.30 2.85 8.49 ' 5 11.21 2.79 8.43 9 10 11.18 2.78 8.40 .--».d . __ .1 l oa—d ’1 -70-“ o 0 :1 L_- ..e....1 e 0 e o v ~——c-oo.—.—¢ -- ._ 0 e ! , -4,_-- ? l- 'r . c: -.. ...-1..-- I l ‘ .7??? ... .t . , 1: Tall ... _ filis‘r. " .i * - to C rt-| iir - - ti 1 -1 . n u as . ., e . fl . a . 9 v.... . . . o - es. a e e a 0190 e « . —‘Iv|-I'L e o o ewe 0 e . . . A C .. Q t l u _ 6 e o .0 A. e o .- c ' o e. .0 e a e , . u a . .L . . e. a e. . i In. 1.4 * e . Q. ¢