ADULTERATIONS AND the nature of the ordinary impurities contained in this acid is well known, and most books on analysis profess to give methods for their detection, Alfred H. Allen has found some of the processes very unreliable, and others to require special precautions not always observed. The following remarks from his valuable paper communicated at a recent meeting of the Society of Public Analysts, will be of great interest.
The principal accidental impurities of tartaric and citric acids are salts of potassium and calcium, together with iron, lead, and copper, derived from the vessels used for the evaporation or crystallization of the acid liquids. The presence of all these impurities is indicated by the pro portion of ash left on igniting the specimen. A number of samples of commercial tartaric and citric acids recently examined gave an amount of ash varying from • 05 to • 25 per cent.
The ignition is readily effected in a porcelain crucible over a Bunsen burner. Platinum vessels should he avoided, le-t lead be present. 5 to 10 grm. should be taken for ignition. When the pro p"rtion of ash ia small it is of no interest to examine it further, except for poisonoua metals. Very sensible quantities of lead and copper are sometimes present Of course their existence will be easily indicated on dissolving the ash in a few drops of nitric acid, diluting largely, and paasing sulphuretted hydrogen. A very fair appmximative estimation of the lead or copper present may be obtained by placing the solution of the ash in a tall glass cylinder, and comparing t,he depth of tint produced by aulphuretted hydrogen with the tint obtained by treating an equal bulk of a very weak standard solution of lead or copper, in a similar manner. The plan is identical with that recommended by Wanklyn for estimating the lead in water, except that sulphuretted hydrogen is aubstituted for ammonium sulphide. Experience haa shown this to be necessary, owing to the frequent presence of iron which of course gives a dark colour in an alkaline solution. Some Arose is laid by him (Allen) on this well-known fact, as some recently examined samples of aerated waters gave a deep brow-n coloration with ammonium sulphide, apps.rently indicating the
pre,,euee of poisonous quantities of lead, but which further inquiry proved to be merely due to a conaiderable quantity of iron. The author prefers to examine the ash for poisonous metals, instead of using the original sample of acid. The presence of copper is indicated on treatment of the ash with nitric acid in the crucible by the production of a blue colour. Of course the presence of poi sonous metals in tartaric and citric acid is always accidental; but as they are occasionally present in dangcroua amounts, it behovea manufacturers to take every precaution to avoid their introduc tion, as the product contoining them must be regarded as adulterated.
Many samples of citric acid contain free sulphuric acid, which may be known by the highly deliqueaccnt character of the crystals. In testing citric aeid for sulphates with chloride of barium, it is desirable to acidify rather atrongly with hydrochloric acid.
The moat common adulterant of citric acid, and almost the only subatance plirposely mixed with it, is tartaric acid, which is frequently found in tho powdered samples sold in the shops. Many planet of detecting tartaric acid in admixture with citric acid have been proposed, but some seem to have been tried merely in the separs.te acids, and not in mixtures containing a compara tively small proportion of the The ordinary method described in text booka of precipitating the tartaric acid from a cold neutral solution by addition of calcium chloride, Allen has found far from delicate, 10 per cent. of tartaric acid in a sample escaping certain detection.
1VIesars. Ohapman and Smith found that a citrate when boiled with a very alkaline solution of potassium permanganate (auch as is used for the estimation of albutninoid ammonia) merely gave a green solution of alkaline manganate ; but a tartrate when similarly treated caused a precipita tion of brown manganese dioxide. Allen says, however, that lie has been unable to verify these results, having failed to find any decided difference in the behaviour of the two acids.