A Microcolorimetric Method for the Determination of Citric Acid Ii. a Note on the Substitution of Ferrous Sulfate for Hydrazine Sulfate as the Reducing Agent* by Hertha H. Taussky

In 1947 we (1) described a microcolorimetric method, with a range of sensitivity from 10 to 60 y, for the determination of citric acid in small samples of blood and urine. This method was based on the conversion of citric acid to pentabromoacetone and on the subsequent reaction between pentabromoacetone and alcoholic sodium iodide with the development of a yellow color complex. Hydrazine sulfate was utilized at two stages in this procedure for the reduction of free bromine and manganese dioxide. This reagent, which was originally introduced by Goldberg and Bernheim (2), was very satisfactory for this purpose. Unfortunately, the prolonged use of this reducing agent led to the development of undesirable skin reactions in several of the laboratory workers. This prompted a search for an alternative reducing agent which might be free of this objectionable feature. A number of substances were tested on both blood and urine samples, two of which were most promising. Hydroxylamine proved satisfactory for the analysis of citric acid in urine, but gave results with blood plasma or serum which were about twice as high as those obtained with hydrazine. Arsenic trioxide in 18 per cent solution gave the same result as did hydrazine, and equally good recoveries of added citric acid. However, the toxicity of this reagent militated against its adoption. We then explored the suitability of a procedure which had been recommended by Pucher, Sherman, and Vickery (3) for this purpose; viz., the addition of ferrous sulfate in excess. Goldberg and Bernheim (2), in the development of their method for the determination of citric acid, substituted hydrazine sulfate for ferrous sulfate to avoid certain difficulties encountered in the use of the latter reagent. They found it difficult to distinguish the end-point when ferrous sulfate was used, inasmuch as the color of the resulting ferric salt was indistinguish-