A Practical Treatise on the Manufacture of Perfumery. C. Deite
and may sometimes be divided into small grains. With oil of bitter almonds, cassia oil, and some oils of clove, as well as volatile oil containing an acid, the tannin test is not available. The first two oils even dissolve tannin, and large quantities of it, if they contain alcohol.
The above-mentioned oils may, however, be rendered fit for the tannin test by mixing them with double their volume of benzine or petroleum-ether, and allowing the mixture to stand for two or three days. If, however, the oils contain much alcohol, the tannin is dissolved. The use of powdered tannin is not advisable, because it generally deposits in a thin layer on the bottom, and its alteration is not so perceptible. If, for practical reasons, a content of 0.5 per cent. anhydrous alcohol might be accepted as permissible in a volatile oil, the tannin test would have to be so modified as to mix 10 drops of the oil with a piece of tannin the size of two peas, and allow the whole to stand for one hour. In this time the above-mentioned content of alcohol would yield no result.
Detection of chloroform.—An adulteration with chloroform, if moderate, cannot always be detected by the odor and taste. In most cases, chloroform will considerably increase the specific gravity of the oil. Bring into a test-tube 15 drops of the suspected oil, 45 to 90 drops of alcohol, and 30 to 40 drops of dilute sulphuric acid. After thorough shaking, add 2 or 3 shavings of zinc sheet and heat until a vigorous evolution of hydrogen takes place. After again shaking, set the whole aside, and heat again when the evolution of gas becomes weaker. This heating and gentle shaking of the fluid is several times repeated. After 20 to 25 minutes, compound the fluid with an equal volume of cold distilled water, shake vigorously and filter through a paper-filter moistened with water. Strongly acidulate the filtrate with nitric acid and compound with nitrate of silver solution. If chloroform is present, turbidity or a precipitate of chloride of silver appears.
Detection of benzine.—An adulteration with benzine can be readily detected only in oils specifically heavier than water. The separation of benzine is effected by distillation from a small glass flask in the water bath. The distillate together with an equal volume of nitric acid of 1.5 specific gravity is gently heated in a test-tube. A too vigorous reaction is modified by cooling in cold water, and a too sluggish action quickened by gentle heating (dipping in warm water). If the mixture has a yellow color, dilute it with water, shake with ether, mix the decanted ethereal solution with alcohol and hydrochloric acid, add some zinc and place the whole in a lukewarm place to convert the nitrobenzol formed into aniline. After evolution of hydrogen is done, neutralize with potash lye, shake, take off the layer of ether, let the latter evaporate and add to the residue a few drops of calcium chloride solution. If benzine is present, a blue-violet color reaction takes place.
Adulterations with alcohol, chloroform, and benzine are quantitatively determined by bringing a weighed quantity of the oil into a glass flask so that it occupies about four-fifths of the volume of the flask. Place upon the flask a cork through which has been passed a glass-tube bent at a right angle and provided with a cylindrical glass vessel serving as a receiver and heating in the water bath. If the distance from the level of the oil to the angle of the glass tube in which it inclines downwards, amounts, for instance, to 4.72 inches, and the neck of the flask up to its angle is 2.75 inches high outside of the direct effect of the heat of the water bath, only the above-mentioned adulterants distill over, while the vapor of the volatile oil condenses at a height of 2.75 inches and flows back into the flask. The distillate is weighed and examined as to its derivation. First add one cubic centimeter of it to two or three cubic centimeters of potassium acetate solution of specific gravity 1.197 and shake moderately. If a clear mixture results, alcohol alone is present. If, however, the mixture is not clear, and the distilled fluid sinks down and collects on the bottom of the test-tube, chloroform is very likely present, and if it remains floating upon the acetate solution, benzine. Next bring two to three centimeters of the distillate into a test-tube and add a piece of sodium metal, the size of a pea. If violent foaming, i.e., an evolution of gas, takes place, alcohol is certainly present, and possibly also chloroform and benzine towards which sodium is indifferent. However, in the presence of benzine, the sodium solution would be colorless, and in the presence of chloroform, yellowish and turbid. In case the sodium produces no reaction and alcohol is, therefore, not present, add an equal volume (two to three cubic centimeters) of anhydrous alcohol, and after moderately shaking allow the solution of the sodium and the evolution of gas to proceed, whereby benzine produces a nearly colorless, turbid fluid, and chloroform a yellowish, milky one. Now dilute the fluid with an equal or double volume of water, shake and allow the mixture to stand quietly. In the presence of benzine a colorless, turbid layer collects on the bottom of the fluid, while that collecting in the presence of chloroform is yellowish. In the latter case, i.e., in the presence of chloroform, the aqueous filtrate yields with lead acetate solution a white precipitate (lead chloride and lead hydroxide). The adulterant having thus been recognized, further particulars are learned from the specific gravity of the oil as well as of the distillate.
Adulterations with terpenes or terpene-like fluids, such as are gained in the preparation of concentrated or patent oils, are difficult to recognize. They may be detected by the specific gravity, the terpenes being, as a rule, specifically lighter, their specific gravity varying between 0.840 and 0.870.
The detection of adulterations with volatile oils of a lower quality is very difficult, if not led to it by the odor and taste. Many methods for establishing such adulterations have been proposed, of which the following are the most important:—
I. Test with iodine.—This test is based upon the fact that some oils violently detonate with iodine, while others develop heat and vapors, and others again remain indifferent. For this test pour upon about 0.19 gramme of dry iodine in a watch-crystal 4 to 6 drops of the oil to be examined.
1. A vigorous reaction (detonation) with considerable increase in the temperature and emission of vapors takes place with the following oils: oils of bergamot, lemon, lavender, nutmeg, orange peel, spike, turpentine, wormwood.
2. Such a reaction as mentioned under 1, does not take place with oils of bitter almonds, copaiba, calamus, clove, peppermint, rose.
3. Moderate heating and slight vapors are developed with oils of anise-seed, fennel, camomile, curly mint, marjoram, rosemary, sassafras, thyme.
When an oil of the second series becomes heated with iodine and evolves vapors, it may first of all be adulterated with cheaper oils. This may also be the case when an oil of the third series reacts violently with iodine and evolves vapors with strong heating. Formerly the iodine test was highly valued; it has, however, been shown to be unreliable since it is frequently dependent on the age of the oil.
In place of iodine, Rudolph Eck recommends a very dilute alcoholic iodine solution, which is not discolored by oils of turpentine, while other oils discolor it. Dissolve a drop of the oil to be examined in 3 cubic centimeters of 90 to 100 per cent. alcohol, and add a drop of the iodine solution. The latter is not discolored in the presence of an oil of turpentine. There are also, however, several volatile oils, which do not discolor the iodine solution. Mierzinski mentions the following: All cold-expressed oils from the Aurantiaceæ, further oils of coriander, caraway, galanga, rue, sassafras, rose, rosemary, anise-seed, fennel, calamus, neroli, angelica, wormwood. Hence, this reaction cannot be relied upon.
II. Hoppe's nitroprusside of copper test.—This test sometimes gives good results, but only with hydrocarbons absolutely free from oxygen and oxygenated oils. It is, therefore, not suitable for oils derived from the Aurantiaceæ. The process is as follows: Add to a small quantity of the oil to be examined in a perfectly dry test-tube, 2 to 5 milligrammes of pure nitroprusside of copper previously thoroughly dried and finely pulverized, shake vigorously and gradually heat to boiling. After boiling for a few seconds allow to cool. If the oil is free from oil of turpentine, or another oil containing no oxygen, the precipitate formed is brown, black, or gray, and according to the quantity of the reagent added and the original color of the oil, the supernatant oil will be differently colored and appear more or less dark. If, however, the oil is adulterated with oil of turpentine, the precipitate formed shows a handsome green or blue-green color, while the supernatant oil retains its original color or at the utmost acquires a very slightly darker one. The longer the oil is allowed to stand