Although attempts have been made repeatedly to work out a general method for the determination of aldehydes and ketones, this problem has not yet been solved in a satisfactory manner. Failure has been due to the fact that not all aldehydes and ketones react equally well with a given reagent, but that in each case only a few yield quantitative results. The first attempt in this direction was made by Benedikt and Strache.1) According to their method the aldehyde or ketone content of a volatile oil is determined by means of the carbonyl oxygen it contains. The oil to be investigated is heated with a weighed amount of phenyl hydrazine, after some time the hydrazone which has formed is separated by filtration, and the unchanged phenyl hydrazine in the filtrate oxidized with boiling Fehling's solution. By this treatment all the nitrogen of the phenyl hydrazine which has not taken part in the reaction is liberated as gas. From the volume of the collected nitrogen the amount of the unused phenyl hydrazine can be calculated. From this the amount which has gone into combination is known and consequently the amount of the ketone or aldehyde present. The amount of carbonyl oxygen, expressed in V10 percents is designated as carbonyl number.

1) Monatsh. f. Chem. 14 (1893), 270. Watson Smith has tried to improve on the method by driving over the nitrogen with a current of carbon dioxide and not with steam as suggested by Benedikt and Strache. Chem. News 93 (1906), 83. According to Chem. Zentralbl. 1906, 1. 1289. Comp. also Report of Schimmel & Co. October 1906, 103.

With bitter almond oil (benzaldehyde), cumin oil (cuminic aldehyde), and rue oil (methyl nonyl ketone) this method yields fairly good results. With cassia oil, caraway oil, fennel oil and lemon oil, however, the determinations fall far too low,1) possibly because the time allowed does not suffice for the quantitative completion of the reaction.

The modification of the above method proposed by Rother2) has met with a similar fate. It likewise yields satisfactory results in a few instances only. The aldehydes and ketones are converted into phenylhydrazones by means of a definite amount of phenyl-hydrazine. It differs, however, from the original method in this that the excess of phenylhydrazine is allowed to react with iodine and the excess of iodine is titrated with thiosulphate solution.

For the assay of aldehydes and ketones we are, therefore, dependant on several methods, of which the most important in actual practice are here described. The determination is mostly volumetric or titrimetric, seldom gravimetric. The best known and most commonly used method is the non-aldehyde constituents remain undissolved. The amount of dissolved oil reveals its aldehyde content.

Bisulphite method. It was first introduced in 18903) by Schimmel & Co. for the cinnamic aldehyde assay of cassia oil. At present it plays an important role in the assay of oils containing cinnamic aldehyde or citral. The method depends on the property of both cinnamic aldehyde and citral to dissolve in hot concentrated sodium bisulphite solution with the formation of sulphonates, hence can be abstracted quantitatively from the oils, whereas the

1) Report of Schimmel & Co. October 1893, 52.

2) Die Bestimmung der Aldehyde und Ketone zurBeurteilung atherischer Ole. Inaug. Dissert. Dresden 1907.

3) Report of Schimmel & Co. October 1890, 15.

For this determination a special glass flask (cassia flask, aldehyde flask, fig. 75) is used. It has a capacity of about 100 cc, is provided with a neck 13 cc. long, the inner diameter of which is 8 mm., and which is calibrated into V10 cc. The entire neck of the flask has a capacity of about 6 cc. The zero point of the scale is placed slightly above the transition point of the flask into the neck.

By means of a pipette, 10 cc.1) of oil are transferred to the flask and an equal volume of 30 p. c. sodium acid sulphite2) solution is added. The mixture is shaken and the flask placed in a boiling water bath. After the solid mass has become liquid more acid sulphite solution is added until the flask is at least three-quarters full, the mixture being constantly heated and occasionally shaken. The solution is heated until no more solid particles are visible and the odor of aldehyde has disappeared. When the clear oil floats upon the salt solution, flask and contents are allowed to cool and sufficient acid sulphite solution is added until the oily layer is well within the neck of the flask. Should drops of oil adhere to the flask these are induced to rise into the neck by gentle tapping and rotation of the flask. The number of cc. of oil is read off, and by deducting this number from ten, the aldehyde content in percentage by volume is ascertained by multiplying the resultant with ten. In order to obtain the percentage by weight the number obtained is multiplied with the specific gravity of the aldehyde in question and the product divided by the specific gravity of the oil.

1) In the case of oils that contain less than 40 p. c. of aldehyde only 5 cc. are taken or a flask is used the neck of which has a capacity of 10 cc. and is calibrated from 1 to 10.

2) Care should be taken that the solution does not contain too much free sulphurous acid since this retards the reaction. If necessary, the solution should be neutralized somewhat by the addition of sodium carbonate.

Fig. 75.

Fig. 75.

As already pointed out, this method is used for the quantitative determination of cinnamic aldehyde and citral, which go into solution as sulphonates. It can, however, be used also in connection with those aldehydes the bisulphite addition products of which are soluble in water as such, e. g. benzaldehyde, anisic aldehyde, phenyl acetaldehyde. 10 cc. of the aldehyde in question are shaken with 40 to 50 cc. of a 30 p. c. sodiumbisulphite solution in an aldehyde flask, and the reaction product dissolved, with the aid of gentle heat, in additional water (not sodium bisulphite solution). The oil that has not taken part in the reaction is driven into the calibrated neck by adding more water. Its volume is read off as soon as the contents of the flask have assumed room temperature. Prolonged standing frequently causes the bisulphite compound to crystallize out again.