The test for chlorinated products is an important one in the examination of bitter almond oil, cherry laurel oil and a number of chemical preparations. In the case of the two oils mentioned, the test serves as a means to detect adulteration with common, technical benzaldehyde; in the case of chemical preparations it affords a clue as to their purity. The principal chemicals of this group are benzylalcohol, benzaldehyde, phenylacet aldehyde, cinnamic aldehyde, synthetic camphor, and benzyl acetate.

Socalled Beilstein test. A piece of copper oxide fastened to a platinum wire is moistened with the oil to be tested and held in the outer zone of a Bunsen flame. As soon as the carbon has burned away, the presence of chlorine is indicated by the greenish or greenish-blue color of the flame. This color is produced by volatilizing copper chloride and its intensity and duration depends on the amount of chlorine present.

Lime test. The oil to be tested is mixed intimately with about ten times its amount of burnt marble1) free from chlorine and the mixture heated to a low red heat in a crucible for a short time. Any chlorine that may be present combines with the calcium. The powder is dissolved in nitric acid and the filtered solution tested for chlorine in the ordinary manner.

Combustion method.2) This method consists in burning the oil and testing the combustion products for hydrogen chloride. A piece of filter paper about 5x6 cm. is folded, saturated with the oil, the excess of oil thrown off, and the paper placed in a small porcelain capsule which in turn rests within a larger porcelain evaporating dish about 20 cm. in diameter. The paper is ignited and immediately covered with a beaker of 2 liter capacity the inner surface of which is moistened with water. The sizes of the implements should be so selected that the rim of the larger porcelain dish projects somewhat beyond that of the beaker. After the flame has become extinguished, the beaker is allowed to remain in its position for several minutes. The products of combustion which have condensed on the inner surface of the beaker are then washed into a filter with about 10 cc. of distilled water. The filtrate, rendered acid with a drop of nitric acid, should not be rendered turbid upon the addition of silver nitrate test solution.

1) In place of the marble, calcined soda free from chlorine may be taken. 2) Comp. Schimmel's Bericht April 1890, 99; also Report October 1904, 7.

This method has proven especially effecient in the testing of volatile oils. Even the minutest traces of chlorine compounds can thus be detected. For the sake of absolute certainty a blank test should be made with a pure oil, since deceptions may occur if the water and the implements used have not been absolutely free from chlorine compounds. Compared with the lime test it has the advantage of being more convenient, especially if a large number of oils have to be tested.

In the case of oils containing hydrocyanic acid, deceptions may occur if unoxidized hydrogen cyanide is deposited on the inner surface of the beaker and washed out with the water. In such cases silver nitrate may produce a turbidity although the oil was pure. Such turbidity is due to AgCN and not to AgCl. This turbidity, however, differs from that produced by silver chloride in its disappearance when the liquid is heated carefully not quite to the boiling point.

Whereas the combustion method is much more delicate and reliable than Beilstein's method when applied to volatile oils, the latter is better adapted, according to Stephan,1) as a test for chlorine in artificial camphor.

The quantitative determination of chlorine can be carried out according to the well known method of Carius. A definite amount of oil is heated with fuming nitric acid in the presence of silver nitrate in a sealed tube and the silver chloride thus produced is weighed.

1) Comp. Lohmann, Berichte d. deutsch. pharm. Ges. 19 (1909), 222.