Introduction
With the exception of the triple uranyl acetate salts there are few satisfactory weighing forms for sodium. Other inorganic reagents have been recommended, but these have never been able to compete with the uranyl acetate reagents. In the field of organic reagents, dihydroxytartaric acid has received some attention, but the reagent is unstable and its sodium salt is rather too soluble for general application as a weighing form for sodium. Recently, ~-methoxyphenylacetic acidl, 2 and the magnesium salt of 8-aminonaphthalene-l-sulphonic acidaA have been recommended for the gravimetric determination of sodium. As these communications were published during the course of the present study 5, which was concerned with the development of organic precipitants for the alkali metals, particularly sodium and potassium, we have felt it appropriate to include our own observations on these reagents.
o~-Methoxyphenylacetic acid is not in our experience as sensitive for sodium as is claimed I. The lowest concentration of sodium to give a positive test is 3.3 mg of sodium per ml of test solution. This compares very unfavourably with the uranyl acetate reagents. It is, however, more selective than zinc or magnesium uranyl acetate in that quite large concentrations of lithium do not interfere with its use. The precipitate given by this reagent is apparently an acid double salt of composition C6HsCH(OCHa)-COOH.C6HsCH(OCHs)COONa. Our results on 5o-mg amounts of sodium show a negative error of 5-7 %, the magnitude of which increases with a decrease in the amount of sodium present. The published procedure 2 involves the use of several different empirical correction factors which are applied in different circumstances; this seriously detracts from the use of the reagent for precise analytical work. We have also found that substitution in the active centre of the molecule of o~-methoxyphenylacetic acid leading to compounds such as ~-ethoxyphenylacetic acid, ~-methoxydiphenylacetic acid, and o~-methoxyfurylacetic acid destroys completely the property associated with the parent compound of forming a sparingly soluble sodium salt 5.
The work of DRANITSKAYA 3 on l-aminonaphthalene-8-sulphonic acid is of greater significance. The magnesium salt of this acid is sufficiently soluble in water to allow the use of a IO~/o solution as the reagent. A three-fold excess of the reagent is necessary and the sodium salt is precipitated from a solution containing about 5 ยฐ ยฐ/o of ethanol. No details are given of the amounts of sodium determined, although this reagent is