Extraction and spectrophotometric determination of cobalt as pyridine-azide mixed ligand complex

Abstract

Clem and Huffman [1] tested the efficiencies of extraction into chloroform of pyridine-azide complexes of eleven metals and suggested spectrophotometric methods for copper, nickel, and palladium in this system. The extracts of cobalt and gold were not sufficiently stable to warrant further investigations. A new spectrophotometric method for cobalt presented here is based on the extraction of pyridine-azide mixed complex of cobalt with methylisobutylketone (abbreviated as MIBK) in the presence of salting-out agent (calcium chloride). Solvents other than MIBK, including chloroform, benzene, nitrobenzene, cyclohexane, dichloroethane, n-butanol etc., resulted in very poor sensitivities; chloroform yielded only an apparent molar absorption coefficient of 501 tool 1 cm1 at 485 nm. The mixed ligand complex of cobalt was able to be extracted into MIBK over pH range of 5 6 . 5 in the presence of 2 M calcium chloride. Without the salting-out agent the maximum extraction was attained in the range pH 6 . 5 8, but the rate of extraction was sluggish; it took 1.5 h for the system to reach the extraction equilibrium. In the presence of 2 M calcium chloride the extraction became prompted, taking 30 rain to reach the maximum absorbance. The mixed complex showed an absorption maximum at 370 nm, its absorbance being held constant for at least 24 h. Azide concentration in the aqueous phase should be greater than 0.4M, and 1.5 ~ 2.5ml of pyridine was necessary for every 50 ml of the aqueous phase to get the highest absorbance. The volume of the aqueous phase should be held approximately constant to obtain reproducible results. Calcium chloride and sodium azide can be purified very effectively by a chelating resin Dowex A-1 column. Beer's law was obeyed in the range 0.1 4 p.g Co per ml of MIBK. Apparent molar absorption coefficient was 1.2 x 1041 tool~ cm~ (370 nm), the photometric sensitivity according to Sandell 0.005 ~tg cm -z. The sensitivity is somewhat better than that of the well known thiocyanate method (isoamyl alcohol) 0.055 (620 nm); 0.009 (310 nm) ~tg cm-2], and inferior to that of the Nitroso-R method [0.0019 ~tg cm -2] [2]. Effects of foreign ions were tested for the determination of 10.0 gg Co and a range of 10.3 _+ 0.5pg Co came out as the 95% confidence limit for the mean of population. The presence of 0.4 M nitrate, 0.4 M acetate, 0.04 M tartrate, and 0.04 M citrate did not interfere with the determination of 10.0 gg Co. A list of interference study is given in Table 1. Tartrate can act as a useful masking agent for aluminum(III), titanium(IV), and tin(IV), which otherwise form insoluble hydroxide precipitates. Iron(III), copper(II), and palladium(II) interfere seriously.