The micro-determination of phosphorus as phosphomolybdate

Abstract

THE oxidation of organic matter with sulphuric and nitric acids and the estimation of ammonium phosphomolybdate volumetrically was introduced into biochemical work by Neumann [1902]. Plimmer and Bayliss [1906] found that the precipitation was not certain and the filtration was slow and liable to error. They defined the conditions for precipitation, introduced a new way of rapid filtration and proved the accuracy of the method for quantities of phosphorus ranging from 10 to 20 mg. Gregersen [1907] also determined the conditions of precipitation and found the method accurate for 1 to 30 mg. of phosphorus. Woerner [1908] confirmed the accuracy of the method. Other workers, Raper [1914], Heubner [1914], Jodidi and Kellog [1915], Kleinmann [1919], Iversen [1920], Euler and Svanberg [1921], Kuhn [1923], S0rensen [1925], Macheboeuf [1926] and Flatter [1933] have maintained either that the method is inaccurate for micro-quantities, or of no value for quantities below 041 mg. Greenwald [1913] and Taylor and Miller [1914] however obtained satisfactory results with quantities of 0*015 to 0*12 mg. Pregl [1930], who weighed the precipitate, regarded the process as accurate and quotes Lieb as proving the method applicable volumetrically to 0'03 mg. Since excellent results are generally agreed to be given with from 0-1 to 10 mg., there seemed no reason for the failure with 0-01 to 0.1 mg. Having used the method for estimating 1 to 30 mg. since 1906, we were interested to ascertain the reason for the general failure with the micro-quantities. EXPERIMENTAL. It was shown by Plimmer and Bayliss [1906, p. 441] that the condition for complete precipitation of ammonium phosphomolybdate was the presence of 30 cc. of 50 % ammonium nitrate (17 g.) solution for 10 cc. of conc. sulphuric acid in a volume of about 200 cc. The same conditions were applied to the precipitation of micro-quantities of phosphorus from 0-01 to 1 mg., but with smaller quantities of reagents and in a smaller flask. In all experiments there was complete precipitation. Filtration and washing of the macro-precipitate was accomplished on a filter-paper on a perforated platinum plate fused into a glass funnel. This filter has been found unsuitable for micro-quantities, even with two papers of barium sulphate quality. The use of asbestos on this filter was rejected as a large amount of inert material would be present during the subsequent titration with the necessity of testing the filter before each estimation. Trial of an asbestos filter of the form described by Bertrand [1913] for glucose estimation showed that it