Cyclic and rotating-disc voltammetry were used to investigate the electrochemical reduction of halothane (2-bromo-2-chloro-1,1,1-trifluoroethane) on various metal electrodes (Cu, Ag, Au, Zn, Cd, Hg, Ga, In, TI, Sn, Sb, W, Ni, Pd and Pt) in several solvents (water, propylene carbonate, acetonitrile, methanol, acetone and tetrahydrofuran). In agreement with previous mechanistic studies, the overall reaction was found to involve two electrons yielding 2-chloro-1,1,1-trifluoroethane and bromide as the main products. The transfer of the first electron is likely to be the rate-determining step. Irrespective of the metal, solvent or electrolyte used, no reaction intermediates were detected by fast scan cyclic voltammetry, i.e. all electrode and/or chemical reactions following this rate-determining step are too fast. The effect of the nature of the solvent is manifested mainly in the decreasing rate of the halothane transport as the solvent viscosity increases, and in the decreasing solvation of the bromide anion with the decreasing polarity of the solvent, which makes the reduction less favourable energetically. In contrast with a previous mechanistic study, a strong effect of the nature of metal on the halothane reduction was observed. First, the halothane reduction is quite inhibited when an oxide film is formed at the metal surface. Second, in the absence of the metal oxide film, the half-wave potentials of the halothane reduction on various metals can differ by several hundreds of millivolts. A simple theory of this effect was outlined, which assumes formation of an intermediate bond between the surface metal atom and the Br moiety of the halothane molecule.