The 4,4′-dimethoxytrityl carbenium ion by deamination of 4,4′-dimethoxytritylamine in acetonitrile–aqueous perchloric acid: kinetics, equilibria, deuterium isotope effects, and possible ion-pair formation

Abstract

The rate of deamination of 4,4′-dimethoxytritylammonium cation, generated from the corresponding amine in aqueous perchloric acid containing a low proportion of acetonitrile at constant total perchlorate concentration, is first order in both dimethoxytritylammonium and hydronium ions. The specificity of the hydronium ion kinetic effect indicates acid catalysis with second-order rate constant, k_H. There is also appreciable reactivity at [H₃O⁺] = 0, i.e. an uncatalysed (solvent-induced) reaction channel from the substituted tritylammonium cation with first-order rate constant, k_o. The deuterium solvent kinetic isotope effects upon uncatalysed and catalysed reaction channels have been measured [k_o^H/k_o^D = 1.51 (±0.03), k_H^H/k_H^D = 1.2 (±0.1), 25 °C and ionic strength = 1 mol dm^–3]. The deuterium solvent isotope effect upon the subsequent equilibrium between 4,4′-dimethoxytrityl trityl cation and water (to give the substituted alcohol and hydronium ion) extrapolated to zero perchlorate concentration has also been determined, and the value (K_R⁺)^H/(K_R⁺)^D = 3.4 (±0.4) leads to a fractionation factor of ca. unity for the alcohol. The deamination is accelerated by increasing perchlorate concentrations at constant acidity; perchlorate also appears to have an appreciable effect upon the subsequent equilibrium involving the substituted trityl cation, water, hydronium ion, and the alcohol. Acetic acid, monochloroacetic acid, and trifluoroethanol also enhance the rate but not, apparently, by general acid catalysis, whereas the rate and final equilibrium are relatively insensitive to the proportion of the acetonitrile cosolvent. A mechanism for deamination is proposed which involves initial reversible heterolysis of the carbon–nitrogen bond of the tritylammonium cation followed by parallel diffusional separation and hydronium ion assisted separation. On the basis of the isotope effects, we conclude that the latter reaction channel involves either step-wise protonation of the ammonia molecule followed by rate-limiting separation of the cation–cation pair, or a strongly asynchronous (uncoupled) concerted process. Perchlorate may exert its appreciable kinetic and equilibrium effects simply by modifying the nature of the medium, i.e. by specific kinetic salt effects upon the various elementary steps in the overall reaction. Alternatively, the effect of perchlorate upon the final equilibrium may be explained by the reversible formation of specific ion-pairs between the extensively delocalised substituted trityl cation and the non-polarising perchlorate anion; we found no evidence of covalent dimethoxytrityl perchlorate. Correspondingly, the effect of perchlorate upon the observed pseudo first-order deamination rate constant could be due to its direct involvement in additional solvent-induced and hydronium ion-catalysed routes.