Rape and flaxseed oils absorb a part of the spectrum but do not affect the rest of it. The spectroscope is of little practical utility in oil analysis.[263]

311. Critical Temperature of Solution.—The study of the critical temperature of solution of oils has been made by Crismer, who finds it of value in analytical work.[264] If a fatty substance be heated under pressure, with a solvent, e. g., alcohol, it will be noticed that as the temperature rises the meniscus of separation of the two liquids tends to become a horizontal plane. If at this point the contents of the tube be thoroughly mixed by shaking and then be left at rest, a point will soon be reached at which the two liquids again separate, and this point is distinctly a function of temperature. Following is a description of a convenient method of determining the critical temperature of the solution of fats and oils for experimental purposes. Tubes are prepared for holding the reagents in such a way that, after the introduction of the fat and alcohol, they can be easily sealed. The capacity of these tubes should be about five cubic centimeters. They should be charged with about one cubic centimeter of the dry filtered fat and about twice that quantity of ninety-five per cent alcohol. Care should be exercised to avoid touching the upper sides of the tube with the reagents. When charged the tubes are sealed in the flame of a lamp and attached to the bulb of a delicate thermometer in such a manner as to have the surface of its liquid contents even with the top of the bulb. The tube is conveniently fastened to the thermometer by a platinum wire. For duplicate determinations two tubes may be fastened to the same thermometric bulb. The apparatus thus prepared is placed in a large vessel filled with strong sulfuric acid. The operator should be careful to protect himself from the danger which might arise from an explosion of the sealed tubes during heating. It is advisable in all cases to observe the reaction through a large pane of clear glass. The bath of sulfuric acid is heated by any convenient means and an even temperature throughout the mass is secured by stirring with the thermometer and its attachments. When the meniscus which separates the two liquids becomes a horizontal plane the thermometer is removed and the liquid in the tubes well mixed until it appear homogeneous. The thermometer is replaced in the bath, which is allowed to cool slowly, and the phenomena which take place in the sealed tubes are carefully noted. The critical temperature of solution is that at which the two liquids begin to separate. This moment is easily noted. It is, moreover, preceded by a similar phenomenon taking place in the capillary part of the tube which retains a drop of the mixture on shaking. In this droplet an opalescence is first noted. In the mass of the liquid this opalescence, a few seconds afterwards, is observed to permeate the whole, followed by the formation of zones and finally of the reappearance of the meniscus of separation between the two liquids. The temperature at this moment of opalescence preceding the separation of the liquid is the critical temperature of solution. With alcohol of 0.8195 specific gravity, at 15°.5 (ninety-five per cent), the observed critical temperatures for some of the more common fats and oils are as given below:

When the alcohol is not pure or if it be of a different density from that named, the numbers expressing the critical temperature of solution will vary from those given above.

312. Polarization.—The pure glycerids are generally neutral to polarized light. In oils the degree of polarization obtained is often variable, sometimes to the right and sometimes to the left. Olive oil, as a rule, shows a slight right hand polarization. Peanut, sesamé, and cottonseed oils vary in polarizing power, but in no case is it very marked. Castor oil polarizes slightly to the right.

In determining the polarizing power of an oil it should be obtained in a perfectly limpid state by filtration and observed through a tube of convenient length, as a rule, 200 millimeters. The deviation obtained may be expressed in divisions of the sugar scale of the instrument or in degrees of angular rotation.

313. Turbidity Temperature.—The turbidity temperature of a fat, when dissolved in glacial acetic acid, as suggested by Valenta, may prove of some diagnostic value.[265] The fats are dissolved, with the aid of heat, in glacial acetic acid and, on slowly cooling, the temperature at which they become turbid is observed. The following data observed by Jones are given for comparison.[266]

The numbers represent the turbidity temperature of the fat when treated with the glacial acetic acid, and allowed to cool slowly. Butter fat, from 40° to 70°, mostly from 52° to 65°; oleomargarin, 95° to 106°; rape oil, 101°; sesamé oil, 77°; linseed oil, 53° to 57°; lard oil, 96°; olive oil, 89°; peanut oil, 61° to 88°.

It is important in this test that the acetic acid be absolutely glacial. About three cubic centimeters of the glacial acetic acid, and three of the fat, should be used.

CHEMICAL PROPERTIES.

314. Solubility in Alcohol.—As has already been noted, the glycerids are freely soluble in ether, chloroform, carbon bisulfid, acetone, carbon tetrachlorid, and some other less commonly used solvents. Their solubility in absolute alcohol is variable and the determination of its degree may often be useful in analytical work.

The method used by Milliau for determining the degree of solubility is as follows:[267] The fatty matter is deprived of its free acids by shaking for half an hour with twice its volume of ninety-five per cent alcohol. After standing until the liquids are separated, the oil or fat is drawn off and washed three times with distilled water. The sample is deprived of water by filtering through a hot jacket filter and a given weight of the dry sample is well shaken with twice its weight of absolute alcohol. A weighed portion of the alcoholic solution obtained is evaporated to remove the alcohol and the weight of the residual fat determined. From the data obtained the percentage of solubility is calculated. Olive oils, when treated as described above, show a solubility of about forty-three parts per thousand of absolute alcohol, cotton oil sixty-two parts, sesamé forty-one parts, peanut sixty-six parts, colza twenty parts, and flaxseed seventy parts per thousand.

315. Coloration Produced by Oxidants.—When oils and fats are mixed with oxidizing reagents, such as sulfuric and nitric acids, the glycerids are partly decomposed with the production of colors which have some analytical significance. The most simple method of applying these tests is by the use of a thick porcelain plate provided with small cup-shaped depressions for holding the few drops of material required. Two or three drops of the oil under examination are placed in each of the cups, a like quantity of the oxidizing reagent added, and the mixture stirred with a small glass rod. The colors produced are carefully noted and the mixture is allowed to remain at room temperature for at least twelve hours in order that the final tint may be observed. The sulfuric acid used for this reaction should have a specific gravity of one and seven-tenths and the nitric acid should have the usual commercial strength of the strongest acid. Pure lard, when treated with sulfuric acid, as above described, shows but little change of color while the vegetable oils mostly turn brown or black. In addition to the reagents mentioned many others, including sulfuric and nitric acids, sulfuric acid and potassium bichromate, chlorin, ammonia, hydrogen peroxid, sodium hydroxid and aqua regia are used. Only a few of these tests seem to have sufficient analytical importance to merit any detailed description.[268]

316. Coloration in Large Masses.—Instead of applying the color test in the small way just described, larger quantities of the fat may be used, either in the natural state or after solution in petroleum or other solvent. For this purpose about ten cubic centimeters of the oil are shaken with a few drops of sulfuric acid or sulfuric and nitric acids. Lard, when thus treated (five drops of sulfuric acid to ten cubic centimeters of lard) shows practically no coloration. When treated with an equal volume of sulfuric acid and shaken, the lard on separating has a brown-red tint.[269]

Olive oil, with a few drops of sulfuric acid, gives a green color, while cottonseed, peanut and other vegetable oils, when thus treated with sulfuric and nitric acids, show brown to black coloration. The delicacy of the reaction may be increased by first dissolving the fat or oil in petroleum ether.

In the use of the coloration test with solvents, a convenient method is to dissolve about one cubic centimeter of the fat in a test tube in petroleum ether, add one drop of strong sulfuric acid and shake.

In the case of lard, the color does not change or becomes yellow or red. Cottonseed oil, similarly treated, shows a brown or black color.[270]

317. Special Nitric Acid Test.—A special nitric acid test for cottonseed oil is made with nitric acid of exactly 1.375 specific gravity at 15°. This test is especially valuable in detecting cottonseed in olive oil. The operation is conveniently conducted by shaking together equal volumes of the oil and acid in a test tube until an intimate mixture or emulsion is secured. When any considerable quantity of cottonseed oil is present an immediate brown coloration is produced, from the intensity of which the relative proportion of cottonseed oil in the case of a mixture may be roughly approximated. When only a little cottonseed oil is present in the mixture, the test tube containing the reagents should be set aside for several hours before the final observation is made.

318. Coloration with Phosphomolybdic Acid.—Among the color tests, one which we have found of use is the coloration produced in certain oils, mostly of a vegetable origin, by phosphomolybdic acid.[271]

The method of applying the test is extremely simple. A few cubic centimeters of the oil or melted lard are dissolved in an equal volume of chloroform, and a third volume of ten per cent phosphomolybdic acid added. The mouth of the test tube is closed with the thumb, and the whole is violently shaken. On being left in repose, the phosphomolybdic acid gathers at the top, and the coloration produced therein is easily observed. Cottonseed oil and peanut oil both give a beautiful green when treated in this way, which is turned to a blue on the addition of ammonia. Linseed oil gives a green color, but forms a kind of emulsion which obscures the color to some extent. The pure lards rendered in the laboratory give no coloration whatever to the reagent, but it retains its beautiful amber color in every case. Mixtures containing as little as ten per cent cottonseed oil and ninety per cent lard, show a distinct greenish tint, while twenty per cent cottonseed oil gives a distinct green. This reaction, therefore, may be considered of great value, and on account of its easy application it should come into wide use. But it is probable that different samples of cottonseed oil, refined to different degrees or in different ways, vary in their deportment with phosphomolybdic acid as they do with silver nitrate. In other words, there may be some samples of cottonseed oil which will not give the green color when treated as above, or so faintly as to have no diagnostic value in mixtures.

This reaction shows itself with nearly all vegetable oils but those which have been chemically treated either for the purpose of bleaching, or for the removal of the acidity, do not respond to the test at all, or else in a feeble manner, and that only after standing some time. Lard, goose fat, tallow, deer fat, butter fat, etc., show no change in color on being treated with this reagent, either with or without the addition of alkali. The presence of a small quantity of vegetable oil betrays itself by the appearance of the above mentioned coloration, the intensity of which forms an approximate measure of the amount of vegetable oil present in the sample. In experiments with suspected lards, which deviated in their iodin absorption numbers from those of genuine lard, the results were concordant, the color deepening as the iodin figure rose. The mineral fats (paraffin, vaselin) are without action on this reagent, and the only animal fat which reduces it is codliver oil.

In like manner some samples of lard may be found which exhibit a deportment with this reagent similar to that shown with vegetable oils, and tallow and lard oil have been shown to give more distinct reactions than some of the vegetable oils.[272]

The phosphomolybdic acid may be prepared by precipitating a solution of ammonium molybdate with sodium phosphate and dissolving the washed precipitate in a warm solution of sodium carbonate. The solution is evaporated to dryness and the dry residue subjected to heat. If a blue coloration be produced it may be discharged by adding a little nitric acid and reheating. The residue is dissolved in water, acidified with nitric and made of such a strength as to contain about ten per cent of the substance.

319. Coloration with Picric Acid.—If to ten cubic centimeters of oil a cold saturated solution of picric acid in ether be added and the latter be allowed to evaporate slowly, the acid remains dissolved in the oil, to which it communicates a brown color.

Pure lard, after the evaporation of the ether, appears of a citron-yellow color; if cottonseed oil be present, however, the mixture assumes a brown-red color.[273]

320. Coloration with Silver Nitrate.—A modification of Bechi’s method of reducing silver nitrate, given further on, has been proposed by Brullé.[274] The reagent employed consists of twenty-five parts of silver nitrate in 1,000 parts of alcohol of ninety-five per cent strength. Twelve cubic centimeters of the oil to be examined and five of the reagent are placed in a test tube, held in a vessel containing boiling water, and the ebullition continued for about twenty minutes. At the end of this time an olive oil, even if it be an impure one, will show a beautiful green tint. With seed oils the results are quite different. Cotton oil submitted to this treatment becomes completely black. Peanut oil shows at first a brown-red coloration and finally a somewhat green tint, losing its transparency. Sesamé oil is distinguished by a red-brown tint very pronounced and remaining red. Colza oil takes on a yellowish green coloration, becomes turbid and is easily distinguished in its reaction from olive oil. In mixtures of olive oil with the other oils, any notable proportion of the seed oils can be easily determined by the above reactions. Natural butter treated with this reagent retains its primitive color. That containing margarin becomes a brick-red and as little as five per cent of margarin in butter can be detected by this test. With ten per cent the tint is very pronounced.

321. Coloration with Stannic Bromid.—This reagent is prepared by adding dry bromin, drop by drop, to powdered or granulated tin held in a flask immersed in ice water, until a persistent red color indicates that the bromin is in excess. In the application of this reagent three or four drops of it are added successively to a little less than that quantity of the oil, the mixture well stirred and set aside for a few minutes. The unsaponifiable matters of castor oil give a green color when thus treated, sandal wood oil a blood-red color and cedar oil a purplish color.[275]

322. Coloration with Auric Chlorid.—The use of auric chlorid for producing colorations in oils and fats was first proposed by Hirschsohn.[276] One gram of auric chlorid is dissolved in 200 cubic centimeters of chloroform and about six drops of this reagent added to five cubic centimeters of the oil to be tested. In the case of cottonseed oil a beautiful red color is produced.

I have found that even pure lards give a trace of color sometimes with this reagent, and therefore the production of a slight red tint cannot in all cases be regarded as conclusive of the presence of cottonseed oil.[277]

In general, it may be said that the color reactions with fats and oils have a certain qualitive and sorting value, and in any doubtful case they should not be omitted. Their value can only be established by comparison under identical conditions with a large number of fats and oils of known purity. The analyst must not depend too confidingly on the data found in books, but must patiently work out these reactions for himself.

323. Thermal Reactions.—The measurement of the heat produced by mixing glycerids with reagents which decompose them or excite other speedy chemical reactions, gives valuable analytical data. These measurements may be made in any convenient form of calorimeter. The containing vessel for the reagents should be made of platinum or some other good conducting metal not affected by them.

The heat produced is measured in the usual way by the increment in temperature noted in the mass of water surrounding the containing vessel. The determination of the heat produced in chemical reactions is a tedious and delicate operation requiring special forms of apparatus for different substances. The time element in these operations is a matter of importance, since it is necessary to work in rooms subject to slight changes of temperature and to leave the apparatus for some time at rest, in order to bring it and its contents to a uniform temperature. For these reasons the more elaborate methods of calorimetric examination are not well suited to ordinary analytical work, and the reader is referred to standard works on thermal chemistry for the details of such operations.[278] For our purpose here a description of two simple thermal processes, easily and quickly conducted, will be sufficient, while a description of the method of determining the heat of combustion of foods will be given in another place.

324. Heat of Sulfuric Saponification.—Maumené was the first to utilize the production of heat caused by mixing sulfuric acid with a fat as an analytical process.[279] In conducting the process a sulfuric acid of constant strength should be employed inasmuch as the rise of temperature produced by a strong acid is much greater than when a weaker acid is employed. The process is at best only comparative and it is evident that the total rise of temperature in any given case depends on the strength of the acid, the character, and purity of the fat or oil, the nature of the apparatus and its degree of insulation, the method of mixing and the initial temperature. For this reason the data given by different analysts vary greatly.[280] For some of the methods of conducting the operation the reader may consult the work of Allen, cited above, or other authorities.[281]

In this laboratory the process is conducted as follows:[282] The initial temperature of the reagents should be a constant one. For oils 20° is a convenient starting point and for fats about 35°, at which temperature most of them are soft enough to be easily mixed with the reagent. The acid employed should be the pure monohydrated form, specific gravity at 20°, 1.845.

The apparatus used is shown in Fig. 98.

The test tube which holds the reagents is twenty-four centimeters in length and five in diameter. It is provided with a stopper having three perforations, one for a delicate thermometer, one for a bulb funnel for delivering the sulfuric acid, and one to guide a stirring rod bent into a spiral as shown. The thermometer is read with a magnifying glass. Fifty cubic centimeters of the fat are placed in the test tube and ten of sulfuric acid in the funnel and the apparatus is exposed at the temperature desired until all parts of it, together with the reagents, have reached the same degree. The test tube holding the oil should be placed in a vacuum-jacket tube, such as will be described in paragraph 316. The oil is allowed to run in as rapidly as possible from the funnel and the stirring rod is moved up and down two or three times until the oil and acid are well mixed. Care must be exercised to stir no more than is necessary for good mixing. The mercury is observed as it ascends in the tube of the thermometer and its maximum height is noted. With the glass it is easy to read to tenths, when the thermometer is graduated in fifths of a degree. When oils are tested which produce a rise of temperature approaching 100°, in the above circumstances, (cottonseed, linseed and some others) either smaller quantities should be used or the oil diluted with some inert substance or dissolved in some inert solvent of high boiling point. For a study of the variations produced in the rise of temperature when varying proportions of oil and acid are used, the work of Munroe may be consulted.[283]

The thermélaeometer described by Jean is a somewhat complicated piece of apparatus and does not possess any advantage over the simple form described above.[284]

Instead of expressing the data obtained in thermal degrees showing the rise of temperature, Thompson and Ballentyne refer them to the heat produced in mixing sulfuric acid and water.[285]

The observed thermal degree of the oil and acid divided by that of the water and acid is termed the specific temperature reaction. For convenience in writing, this quotient is multiplied by 100. The respective quantities of acid and water are ten and fifty cubic centimeters. This method of calculation has the advantage of eliminating to a certain degree the variations which arise in the use of sulfuric acid of differing specific gravities. In the following table are given the comparative data obtained for some common oils.[286]

325. Method of Richmond.—The rise of temperature produced by mixing an oil and sulfuric acid is determined by Richmond in a simple calorimeter, which is constructed by fitting a small deep beaker inside a larger one with a packing of cotton. The heat capacity of the system is determined by adding to ten grams of water, in the inner beaker, at room temperature, twenty-five grams of water of a noted higher temperature and observing the temperature of the mixture. The cooling of the system, during the time required for one determination of heat of sulfuric saponification, does not exceed one per cent of the whole number of calories produced.[287] Between the limits of ninety-two per cent and one hundred per cent the rise of temperature observed is directly proportional to the strength of the acid.

Relative Maumené Figure.—The total heat evolved per mean molecule is called by Richmond the relative maumené figure. It is calculated as follows:

• Let x = percentage of sulfuric acid in the acid employed;
• h = heat capacity of calorimeter in grams of water;
• R = observed rise of temperature (twenty-five grams of
• oil, five cubic centimeters sulfuric acid);
• K = potash absorbed for saponification (19.5 per cent of
• potassium hydroxid, standard of comparison);
• M = relative maumené figure:

326. Heat of Bromination.—The rise of temperature caused by mixing fats with sulfuric acid has long been used to discriminate between different fats and oils. Hehner and Mitchell propose a similar reaction based upon the rise of temperature produced by mixing bromin with the sample.[288] The action of bromin on unsaturated fatty bodies is instantaneous and is attended with a considerable evolution of heat. Since the action of bromin on many of the oils is very violent it is necessary to dilute the reagent with chloroform or glacial acetic acid. Owing to its high boiling point the acetic acid has some advantage over chloroform for this purpose. The tests are conveniently made in a vacuum-jacket tube. In such a tube there is no loss of heat by radiation. The bromin is measured in a pipette having at its upper end a tube filled with caustic lime held between plugs of asbestos. The bromin sample to be tested and the diluent employed are kept at the same temperature before beginning the trial. They are quickly mixed and the rise of temperature noted. The oil is first dissolved in the chloroform and the bromin then added.

A somewhat constant relation is noticed between the rise of temperature and the iodin number when one gram of oil, ten cubic centimeters of chloroform and one cubic centimeter of bromin are used.

If the rise in temperature in degrees be multiplied by 5.5 the product is approximately the iodin number of the sample. Thus a sample of lard gave a rise in temperature of 10°.6 and an iodin number of 57.15. The number obtained by multiplying 10.6 by 5.5 is 58.3.

In like manner the numbers obtained for some common oils are as follows:

327. Modification of the Heat of Bromination Method.—The method described above by Hehner and Mitchell presents many grave difficulties in manipulation, on account of the inconvenience of handling liquid bromin. The process is made practicable by dissolving both the oil or fat and the bromin in chloroform, or better in carbon tetrachlorid, in which condition the bromin solution is easily handled by means of a special pipette.[289]

In order to make a number of analyses of the same sample ten grams of the fat may be dissolved in chloroform or carbon tetrachlorid and the volume completed with the same solvent to fifty cubic centimeters. In like manner twenty cubic centimeters of the bromin are dissolved in one of the solvents named and the volume completed to 100 cubic centimeters therewith.

For convenience of manipulation the solutions are thus made of such a strength that five cubic centimeters of each represent one gram of the fat and one cubic centimeter of the liquid bromin respectively.

The apparatus used for the work is shown in the accompanying figure. The pipette for handling the bromin solution is so arranged as to be filled by the pressure of a rubber bulb, thus avoiding the danger of sucking the bromin vapor into the mouth. The filling is secured by keeping the bromin solution in a heavy erlenmeyer with a side tubulure such as is used for filtering under pressure. The solutions are mixed in a long tube, held in a larger vessel, from which the air is exhausted to secure a minimum radiation of heat. A delicate thermometer graduated in tenths serves to register the rise of temperature. The fat solution is first placed in the test tube, with care not to pour it down the sides of the tube but to add it by means of a pipette reaching nearly to the bottom. The whole apparatus having been allowed to come to a standard temperature the bromin solution is allowed to run in quickly from the pipette. No stirring is required as the liquids are sufficiently mixed by the addition of the bromin solution. The mercury in the thermometer rapidly rises and is read at its maximum point by means of a magnifying glass. With a thermometer graduated in tenths, it is easy to read to twentieths of a degree.

It is evident that the rise of temperature obtained depends on similar conditions to those mentioned in connection with sulfuric saponification. Each system of apparatus must be carefully calibrated under standard conditions and when this is done the comparative rise of temperature obtained with various oils and fats will prove of great analytical use. It is evident that the ratio of this rise of temperature to the iodin number must be determined for every system of apparatus and for every method of manipulation employed, and no fixed factor can be given that will apply in every case.

With the apparatus above described and with the method of manipulation given the following data were obtained for the oils mentioned:

Bromin and chloroform, when mixed together, give off heat, due to the chemical reaction resulting from the substitution of bromin for hydrogen in the chloroform molecule and the formation of hydrobromic acid. For this reason the data obtained, when chloroform is used as a solvent, are slightly higher than with carbon tetrachlorid. The use of the latter reagent is therefore to be preferred.

328. Haloid Addition Numbers.—Many of the glycerids possess the property of combining directly with the haloids and forming thereby compounds in which the haloid, by simple addition, has become a part of the molecule. Olein is a type of this class of unsaturated glycerids. The process may take place promptly as in the case of bromin or move slowly as with iodin. The quantity of the haloid absorbed is best determined in the residual matter and not by an examination of the fat compound. By reason of the ease with which the amount of free iodin in solution can be determined, this substance is the one which is commonly employed in analytical operation on fats.

In general, the principle of the operation depends on bringing the fat and haloid together in a proper solution and allowing the addition to take place by simple contact. The quantity of the haloid in the original solution being known, the amount which remains in solution after the absorption is complete, deducted from that originally present, will give the quantity which has entered into combination with the glycerid.

329. Hübl’s Process.—In determining the quantity of iodin which will combine with a fat, the method first proposed by Hübl, or some modification thereof, is universally employed.[290] In the determination of the iodin number of a glycerid it is important that it be accomplished under set conditions and that iodid be always present in large excess. It is only when data are obtained in the way noted that they can be regarded as useful for comparison and determination. Many modifications of Hübl’s process have been proposed, but it is manifestly impracticable to give even a summary of them here. As practiced in the chemical laboratory of the Agricultural Department and by the Association of Official Agricultural Chemists, it is carried out as follows:[291]

(1) PREPARATION OF REAGENTS.

(a). Iodin Solution.—Dissolve twenty-five grams of pure iodin in 500 cubic centimeters of ninety-five per cent alcohol. Dissolve thirty grams of mercuric chlorid in 500 cubic centimeters of ninety-five per cent alcohol. The latter solution, if necessary, is filtered, and then the two solutions mixed. The mixed solution should be allowed to stand twelve hours before using.

(b). Decinormal Sodium Thiosulfate Solution.—Dissolve 24.8 grams of chemically pure sodium thiosulfate, freshly pulverized as finely as possible and dried between filter or blotting paper, and dilute with water to one liter, at the temperature at which the titrations are to be made.

(c). Starch Paste.—One gram of starch is boiled in 200 cubic centimeters of distilled water for ten minutes and cooled to room temperature.

(d). Solution of Potassium Iodid.—One hundred and fifty grams of potassium iodid are dissolved in water and the volume made up to one liter.

(e). Solution of Potassium Bichromate.—Dissolve 3.874 grams of chemically pure potassium bichromate in distilled water and make the volume up to one liter at the temperature at which the titrations are to be made.

(2). DETERMINATION.

(a). Standardizing the Sodium Thiosulfate Solution.—Place twenty cubic centimeters of the potassium bichromate solution, to which have been added ten cubic centimeters of the solution of potassium iodid, in a glass stopper flask. Add to this mixture five cubic centimeters of strong hydrochloric acid. Allow the solution of sodium thiosulfate to flow slowly into the flask until the yellow color of the liquid has almost disappeared. Add a few drops of the starch paste, and with constant shaking continue to add the sodium thiosulfate solution until the blue color just disappears. The number of cubic centimeters of thiosulfate solution used multiplied by five is equivalent to one gram of iodin.

Example.—Twenty cubic centimeters of potassium bichromate solution required 16.2 sodium thiosulfate; then 16.2 × 5 = 81 = number cubic centimeters of thiosulfate solution equivalent to one gram of iodin. Then one cubic centimeter thiosulfate solution = 0.0124 gram of iodin: (Theory for decinormal solution of sodium thiosulfate, one cubic centimeter = 0.0127 gram of iodin.)

(b). Weighing the Sample.—About one gram of butter fat is placed in a glass stopper flask, holding about 300 cubic centimeters, with the precautions to be mentioned for weighing the fat for determining volatile acids.

(c). Absorption of Iodin.—The fat in the flask is dissolved in ten cubic centimeters of chloroform. After complete solution has taken place thirty cubic centimeters of the iodin solution (1) (a) are added. The flask is now placed in a dark place and allowed to stand, with occasional shaking, for three hours.

(d). Titration of the Unabsorbed Iodin.—One hundred cubic centimeters of distilled water are added to the contents of the flask, together with twenty cubic centimeters of the potassium iodid solution. Any iodin which may be noticed upon the stopper of the flask should be washed back into the flask with the potassium iodid solution. The excess of iodin is taken up with the sodium thiosulfate solution, which is run in gradually, with constant shaking, until the yellow color of the solution has almost disappeared. A few drops of starch paste are added, and the titration continued until the blue color has entirely disappeared. Toward the end of the reaction the flask should be stoppered and violently shaken, so that any iodin remaining in solution in the chloroform may be taken up by the potassium iodid solution in the water. A sufficient quantity of sodium thiosulfate solution should be added to prevent a reappearance of any blue color in the flask for five minutes.

(e). Setting the Value of the Iodin Solution by the Thiosulfate Solution.—At the time of adding the iodin solution to the fat, two flasks of the same size as those used for the determination should be employed for conducting the operation described above, but without the presence of any fat. In every other respect the performance of the blank experiments should be just as described. These blank experiments must be made each time the iodin solution is used.

Example of Blank Determinations.—Thirty cubic centimeters of iodin solution required 46.4 cubic centimeters of sodium thiosulfate solution: Thirty cubic centimeters of iodin solution required 46.8 cubic centimeters of sodium thiosulfate solution: Mean, 46.6 cubic centimeters.

330. Character of Chemical Reaction.—The exact nature of the chemical process which takes place in this reaction is not definitely known. Hübl supposed that the products formed were chloro-iodid-additive compounds, and he obtained a greasy product from oleic acid, to which he ascribed the formula C₁₈H₃₄IClO₂. By others it is thought that chlorin alone may be added to the molecule.[292]

In general, it may be said that none of the glycerids capable of absorbing halogens is able to take on a quantity equivalent to theory.[293] While the saturated fatty acids (stearic series) theoretically are not able to absorb iodin some of them are found to do so to a small degree. It is evident, therefore, that it is not possible to calculate the percentage of unsaturated glycerids in a fat from their iodin number alone. According to the data worked out by Schweitzer and Lungwitz both addition and substitution of iodin take place during the reaction.[294] This fact they determined by titration with potassium iodate and iodid according to the formula 5HI + HIO₃ = 6I + 6H₂O. The authors confess that whenever free hydriodic acid is found in the mixture that iodin substitution has taken place and that for each atom of hydrogen eliminated from the fat molecule two atoms of iodin disappear, one as the substitute and the other in the form of hydriodic acid. When carbon bisulfid or tetrachlorid is used as a solvent no substitution takes place and pure additive compounds are formed.

The following process is recommended to secure a pure iodin addition to a glycerid: About one gram or a little less of the oil or fat is placed in a glass stopper flask, to which are added about seven-tenths of a gram of powdered mercuric chlorid and twenty-five cubic centimeters of a solution of iodin in carbon bisulfid. The stopper is made tight by smearing it with powdered potassium iodid, tied down, and the mixture is heated for some time under pressure. By this method it is found that no hydriodic acid is formed, and hence all the iodin which disappears is added to the molecule of the glycerid. The additive numbers obtained for some oils are appended:

331. Solution in Carbon Tetrachlorid.—Gantter has called attention to the fact that the amount of iodin absorbed by fat does not depend alone upon the proportion of iodin present but also upon the amount of mercuric chlorid in the solution.[295] Increasing amounts of mercuric chlorid cause uniformly a much greater absorption of the iodin. For this reason he proposes to discard the use of mercuric chlorid altogether for the hübl test and to use a solvent which will at the same time dissolve both the iodin and the fat. For this purpose he uses carbon tetrachlorid. The solutions are prepared as follows:

Iodin Solution.—Ten grams of iodin are dissolved in one liter of carbon tetrachlorid.

In the preparation of this solution the iodin must not be thrown directly into the flask before the addition of the tetrachlorid. Iodin dissolves very slowly in carbon tetrachlorid and the solution is made by placing it in a sufficiently large weighing glass and adding a portion of the carbon tetrachlorid thereto. The solution is facilitated by stirring with a glass rod until the added tetrachlorid is apparently charged with the dissolved iodin. The dissolved portion is then poured into a liter flask, new portions added to the iodin and this process continued until the iodin is completely dissolved, and then sufficient additional quantities of the tetrachlorid are added to fill the flask up to the mark.

332. Sodium Thiosulfate Solution.—Dissolve 19.528 grams of pure sodium thiosulfate in 1000 cubic centimeters of water. For determining the strength of the solution by titration, the solution of iodin in carbon tetrachlorid and a solution of sodium thiosulfate in water are each placed in a burette. A given volume of the iodin solution is first run into a flask with a glass stopper and afterward the sodium thiosulfate added little by little until, after a vigorous shaking, the liquid has only a little color. Some solution of starch is then added and shaken until the mixture becomes deep blue. The sodium thiosulfate solution is added drop by drop, with vigorous shaking after each addition, until the solution is completely decolorized. If both solutions have been correctly made with pure materials they will be of equal strength; that is, ten cubic centimeters of the iodin solution will be exactly decolorized by ten cubic centimeters of the sodium thiosulfate solution.

333. Method of Conducting the Absorption.—The quantity of the fat or oil employed should range from 100 to 200 milligrams, according to the absorption equivalent. These quantities should be placed in flasks with glass stoppers in the ordinary way. In the flasks are placed exactly fifty cubic centimeters of the iodin solution equivalent to 500 milligrams of iodin, and the flask is then stoppered and shaken until the fat or oil is completely dissolved. In order to avoid the volatilization of the iodin finally, sufficient water is poured into the flask to form a layer about one millimeter in thickness over the solution containing the iodin and fat. The stopper should be carefully inserted and the flask allowed to stand at rest for fifty hours.

334. Estimation of the Iodin Number.—This is determined in the usual way by titration of the amount of iodin left in excess after the absorption as above described. The iodin number is to be expressed by the number of milligrams of iodin which are absorbed by each 100 milligrams of fat.

Example.—One hundred and one milligrams of flaxseed oil were dissolved in fifty cubic centimeters of the carbon tetrachlorid solution of iodin and allowed to stand as above described for fifty hours. At the end of this time, 42.3 cubic centimeters of the sodium thiosulfate solution were required to decolorize the excess of iodin remaining.

Statement of Results.—Fifty cubic centimeters of the sodium thiosulfate equal 500 milligrams of iodin; therefore, 42.3 cubic centimeters of the thiosulfate solution equal 423 milligrams of iodin. The difference equals seventy-seven milligrams of iodin absorbed by 101 milligrams of the flaxseed oil. Therefore, the iodin number equivalent and the milligrams of iodin absorbed by 100 milligrams of flaxseed oil equal 76.2.

It is evident from the above determination that the iodin number of the oil, when obtained in the manner described, is less than half that secured by the usual hübl process. Since the solvent employed, however, is more stable than chloroform when in contact with iodin or bromin, the proposed variation is one worthy of the careful attention of analysts.

McIlhiney has called especial attention to the low numbers given by the method of Gantter, and from a study of the data obtained concludes that iodin alone will not saturate glycerids, no matter what the solvents may be.[296]

It is clear, therefore, that the process of Gantter cannot give numbers which are comparable with those obtained by the usual iodin method. Any comparative value possessed by the data given by the process of Gantter must be derived by confining it to the numbers secured by the carbon tetrachlorid process alone.

335. Substitution of Iodin Monochlorid for Hübl’s Reagent.—Ephraim has shown that iodin monochlorid may be conveniently substituted for the hübl reagent with the advantage that it can be safely used at once, while the hübl reagent undergoes somewhat rapid changes when first prepared. The present disadvantage of the process is found in the fact that the iodin monochlorid of commerce is not quite pure and each new lot requires to be titrated for the determination of its purity.

The reagent is prepared of such a strength as to contain 16.25 grams of iodin monochlorid per liter. The solvent used is alcohol. The operation is carried out precisely as in the hübl method, substituting the alcoholic solution of iodin monochlorid for the iodin reagent proposed by Hübl.[297] If the iodin monochlorid solution, after acting on the oil, be titrated without previous addition of potassium iodid a new value is obtained, the chloriodin number. In titrating, the sodium thiosulfate is added until the liquid, which is made brown by the separated iodin, becomes yellow. At this point the solution is diluted, starch paste added, and the titration completed.

336. Preservation of the Hübl Reagent.—To avoid the trouble due to changes in the strength of Hübl’s reagent, Mahle adds hydrochloric acid to it at the time of its preparation.[298] The reagent is prepared as follows: Twenty-five grams of iodin dissolved in a quarter of a liter of ninety-five per cent alcohol are mixed with the same quantity of mercuric chlorid in 200 cubic centimeters of alcohol, the same weight of hydrochloric acid of 1.19 specific gravity added and the volume of the mixture completed to half a liter with alcohol. After five days such a solution gave, on titration, 49.18 instead of 49.31 grams per liter of iodin.

It will be observed that this solution is double the usual strength, but this does not influence the accuracy of the analytical data obtained. It appears that the hübl number is not, therefore, an iodin number, but expresses the total quantity of iodin, chlorin and oxygen absorbed by the fat during the progress of the reaction.

337. Bromin Addition Number.—In the process of Hübl and others an attempt is made to determine the quantity of a halogen, e.g., iodin, which the oil, fat or resin will absorb under certain conditions. The numbers obtained, however, represent this absorption only approximately, because the halogen may disappear through substitution as well as absorption. Whether or not a halogen is added, i. e., absorbed or substituted, may be determined experimentally.

The principle on which the determination depends rests on the fact that a halogen, e. g., bromin, forms a molecule of hydrobromic acid for every atom of bromin substituted, while in a simple absorption of the halogen no such action takes place. If, therefore, bromin be brought into contact with a fat, oil or resin, the determination of the quantity of hydrobromic acid formed will rigidly determine the quantity of bromin substituted during the reaction. If this quantity be deducted from the total bromin which has disappeared, the relative quantities of the halogen added and substituted are at once determined. In the method of McIlhiney[299] bromin is used instead of iodin because the addition figures of iodin are in general much too low.

The Reagents.—The following solutions are employed: