Talanta, Vol. 32, No. I, pp. 531-537, 1985 0039-9140/85 $3.00 + 0.00 Printed in Great Britain. All rights reserved Copyright 0 1985Pergamon Press Ltd zyxwvut ANIONIC INTERFERENCES WITH COPPER ION-SELECTIVE ELECTRODES CHLORIDE AND BROMIDE INTERFERENCES A. LEWENSTAM, T. SOKALSKI and A. HULANICKI Department of Chemistry, University of Warsaw, Warsaw, Poland (Received 14 June 1984. Revised 10 January 1985. Accepted 1 February 1985) Summary---The effect of halide ions on copper ion-selective electrodes is connected with complexation and redox reactions, with the formation of amorphous sulphur, which by blocking the surface causes instability of potential response. It may be eliminated by addition of sodium thiosulphate solution. The electrode behaviour has been explained on the basis of the diffusion model. An equation is proposed for linearization of the calibration curve. The parameters of the semiempirical model which describes the electrode behaviour agree well with the physical meaning presented by the diffusion model. The treatment given enables analytical measurements of copper concentration to be made even in the presence of significant concentrations of chloride or bromide. Copper-selective electrodes with membranes com- posed of copper(I1) sulphide mixed with silver sul- phide or of pure copper(I) sulphide show a rapid and Nemstian response (i.e., 29.6 mV/decade at 25”) over a broad range of copper(I1) concentration, but appli- cations of these electrodes to real analytical samples often reveal non-Nernstian behaviour.’ Such effects were observed when the Orion 94-29 electrode was used for measurements of copper concentrations in sea-water2 or lake water.3 Such facts have diminished the attraction of copper-selective electrodes as anal- ytical tools, but at the same time have attracted the attention of scientists interested in electrode-response mechanisms, as the electrode behaviour could not be explained by existing theories.4 Attempts to elucidate the super-Nemstian electrode slope, attributed to the presence of chloride ions, have been made in a number of papers. In some of theiu the source of anomalous electrode behaviour was considered to be exchange reactions occurring at the electrode surface:@ Ag,S + 2Cu2 + + 2n Cl - + CuS + ZAgCl:, -n (1) cuzs + cu*+ + 2nCl- Q=cus + 2CuCI’ -” n (2) Others”’ have attributed the behaviour to redox reactions with participation of the membrane mate- rial: Ag,S + 2Cu* + + 4nCl- e 2CuCIi-” + 2AgCl;-” + S (3) cu2s+2cu2+ +4nCl-*4cuc1’-“+s ” (4) cus+cu2+ +2nCl-G=2CuCl~-“+S (5) However, most papers do not offer a detailed explanation of the electrode response. Ross’s interpretation4 does not assume the occurrence of super-Nernstian slope, but indicates only that silver chloride is formed in the case of the CuS/Ag,S membrane [reaction(l)]. The formation of electro- active AgCl has been experimentally confirmed by Moody and co-workers. 5~‘2 Lanza’ suggests that the electrode interference is connected with the presence of Ag,S and is due to reaction (1). His conclusions are not unequivocally confirmed by his own experimental data. There is also no explanation for super- Nernstian slope in the absence of chloride, and for electrodes which do not contain Ag2S. Hepel,’ ex- tending the views of other authors, assumes reaction (2) to be the main reason for the interference. In his considerations, based only on potentiometric data, he neglects the role of redox processes in spite of some earlier remarks9*‘s15 about their importance. The influence of the redox reaction (5) has been discussed by Westall et ~1.~ Their theoretical considerations were well supported by experiments, but these cov- ered only a relatively narrow concentration range, namely pCu from 3.5 to 5.0, because according to the authors’ statement this reaction is hindered by for- mation of AgCl at the membrane surface. They did not confirm the presence of sulphur at the membrane surface, but this has been reported independently.“,” No complete explanation has so far been given, then, for the non-Nernstian electrode behaviour, especially the variation of slope between 0 and 120 mV/pCu in the pCu range from 2 to 5. In this paper we attempt to give a more general description of the interferences with copper electrodes in solu- tions containing copper-complexing anions, and pro- pose a method for elimination of this interference and for practical utilization of the super-Nemstian slope. EXPERIMENTAL Measurements of the e.m.f. were made with Radiometer pHM 64 and Radelkis OP-206 instruments, with a Radio- 531