(1) Examination for Mineral and Rosin Oils.— (a) Warm the oil, and smell it. If it gives off the odour of paraffin, it contains ordinary illuminating mineral oil. In this case, carefully counterpoise a watch-glass, and having weighed on to it about 1 grm. of the oil, keep it in an air bath heated to 110° (230° F.) until the weight is constant ; note the lose of weight, and calculate to percentage of paraffin-oil.
(6) Place 10 grm. of the oil in a basin with 20 c.c. of the alcoholic caustic soda (No. 3 reagent), heat to boiling on a water-bath, and evaporate nearly to dryness. Then add 75 c.c. of distilled water, and boil for half-an-hour. Observe the nature of the resulting mass or liquid, and note one or other of the following cases.
Case 1. An emulsion only is formed. Probably there is only mineral or rosin oil. Add water, and warm, when, if a clear oil separates, mineral oil is present, but if brown, then rosin-oil may be present. Now draw off the aqueous liquid from beneath, and make it acid with sulphuric acid. If no precipitate forms, the oil was all mineral ; but if a precipitate be produced, collecting on warming in brown Viscous drops, then the presence of rosin-oil is confirmed.
Case 2. A semi-pasty mass is obtained. Probably it is an ordinary fatty oil mixed with mineral or rosin-oil. Add water, and warm, when the mineral oil, if present, will float to the surface. Now draw off the aqueous liquid, and shake it up with amylic alcohol ; if the alcohol separates in a brown layer, rosin-oil was present. Once more draw off the aqueous liquid, acidulate with hydro chloric acid, and warm, when the separation of oleic acid, with its characteristic odour of fat, will show the presence of the fatty oil. This would also apply in the case of testing a solid fat for paraffin-wax.
Case 3. A perfectly pasty soap is formed, which dissolves in warm water without any separa tion of oil. The sample is an ordinary fatty oil, containing possibly (though not likely) a little rosin-oil. Shake up with amylic alcohol, when, if no brown colour be produced in the alcohol, no rosin-oil is present. Once more draw off the aqueous liquid, add an excess of sodium chloride (which will precipitate the fatty soap) and filter. If the oil be a pure fatty one, the warmed filtrate will only give a slight turbidity on acidulating with hydrochloric acid, and will smell of fat ; but if the solution give a copious precipitate, and, when heated, has a resinous odour, the oil contains rosin. If, on setting free the fatty acids from the soap with hydrochloric acid, and cooling to 15° (60° F.), they partly solidify, animal oil, such as lard or neats'-foot, may be suspected, although certain vegetable oils (such as cotton-seed), especially when crude, give tolerably solid acids. (The fatty acids of coco-nut-oil, and palm-nut-oil, when liberated and heated, smell of volatile fatty acids, just like butter, and the insoluble portion is very low).
Having thus got a fair preliminary idea of the constitution of the sample, it may be confirmed by taking the "actual density" at 38° (100° F.), as already directed for butter (p. 1465). If the
" actual density " of the article be under 0.900, it is all mineral oil ; if between and it may be either all fatty oil, or a complex mixture ; while if over 0.960, it is all rosin-oil.
(2) Estimation of a Mixture of heavy Lubricating Mineral Oil and Rosin-oil.—There is no known method of chemically separating these, and so an approximation must be made from the "actual density." Mineral oils for lubricating have an "actual density" not exceeding 0'880, while rosin-oil is generally about 1.000 ; therefore the following table may be taken as approximate, which, however, is only good in the ensured absence of fatty oils.
The reader is warned that this is only approximate, as there are samples of rosin-oil as low in "actual density" as 0.9800.
(3) Estimation of Mineral Oil in Fatty Oils.-20 grm. of the sample are saponified with 35 c.c. of the alcoholic caustic soda (reagent No. 3) in a deep basin on the water-bath ; 20 c.c. of redistilled methylated spirit, or sufficient to perfectly dissolve the soap, are added. The whole being still kept boiling on the bath, 9 grm. of sodium bicarbonate are added little by little, well stirring after each addition, so that all the excess of alkali may become carbonate ; 50 grm. of sand (No. 5 reagent) are stirred in so as to thoroughly mix the whole, and the evaporation is con tinued until a perfectly dry residue is obtained. This residue is now packed into a stoppered percolator, and covered with petroleum-spirit (No. 2 reagent), and the whole is allowed to macerate for an hour ; the stopper is then opened, and percolation is commenced into a 40-oz. flask, until the spirit has run off. More petroleum is then added, and percolation is continued until a few drops of the liquid evaporated on a watch-glass cease to leave any residue. This takes in all a considerable quantity of petroleum, so that sometimes the whole percolate measures nearly 1 litre. The flask containing the petroleum is then attached to a condenser, and the bulk of the petroleum is reduced by distillation to something under 100 c.c. The residue is transferred to an " olein tube" (see p.1463), and the flask is rinsed with warm petroleum, so that the contents of the tube measure, say 150 ex. A platinum capsule is carefully tared, and a measured small aliquot part of the fluid in the tube is run into it, from the stopcock, and evaporated to dryness at a temperature not exceeding 104i. (220° F.). The residue is weighed, and calculated to the whole bulk of the fluid ; that result, multiplied by 5, gives the percentage of mineral oil. As a rule, the tendency of the process is to come out about 0.5 per cent. too high, so that any fraction of a per cent. under or over 0.5 may be disregarded, i.e., if 20.4 per cent., to simply call it 20 per cent., but if 20'S per cent , then report 20.5 per cent. To check the weighing, it is desirable to run off more than one small aliquot part, and evaporate and weigh. This process is equally applicable to the detection of paraffin-wax in solid fats.