No abstract

Double proton-transfer reactions are characterized by three primary kinetic H/D isotope effects (kie.) involving a set of rate constants kHH, kHD, kDH and kDD. We have succeeded in measuring such sets for well defined symmetrical double proton-transfer reactions where kHD = kDH for the intramolecular hydrogen migration in meso-tetraphenylporphine (TPP) and for the intermolecular 1 : 1 proton exchange between acetic acid and methanol dissolved in tetrahydrofuran. In this system we also detected a triple proton transfer involving one methanol and two acetic acid molecules.Rate constants kHHH, kHHD, kHDD and kDDD of this reaction are reported. Additionally, we also observed an intramolecular double proton-transfer reaction in 2,5-dianilino-benzoquinone-1 ,Cdianil (azophenine, AP) for which we have measured the HH/HD k.i.e. For the determination of rate constants dynamic n.m.r. spectroscopy was used.The predictions of the fractionation-factor theory, transition-state theory and different protontunnelling theories on the HH/HD/DH/DD k.i.e. of symmetrical double proton-transfer reactions are discussed with special emphasis on the rule of the geometric mean and compared with the experiments. The kinetic and the i.r. results for TPP and AP are well reproduced by the vibrational model of tunnelling as proposed previously by us. TPP is, therefore, the first proton-transfer system with NH-stretching levels not broadened by hydrogen-bond effects. The k.i. e. of the intermolecular proton exchange in the system acetic acid + methanol + tetrahydrofuran are consistent with tunnelling from the OH-stretching ground states. The energy of activation is associated with the enthalpy of formation of the cyclic complex in which the exchange takes place and the excited intermolecular vibrational states within this complex. The tunnelling theories are modified in order to take this enthalpy into account.

Chcriiicd KineticsThc 90 M H z ' H N M R spectra of '5N,'sN'-dipentadeutcrophenylformamidine (DPFA), a nitrogen analogue of formic acid, dissolved in tetrahydrofuran-d8 (THF) have been measured as a function of concentration, deuterium fraction in the 'H-I5N sites, and of temperature. The spectra show characteristic changes, from which thermodynamic and kinetic information on hindered rotation, hydrogen bond association and proton exchange of DPFA in THF are obtained by NMR lineshape analysis. DPFA forms two conformers A and B in THF, to which s-trans and s-cis structures have been assigned. At low concentrations both DPFA conformers are located in a hydrogen bond to the solvent molecules. However, as a result of the different structure, only A is able to selfassociate to any observable extent. This effect leads to concentration dependent A/B populations. Higher selfassociates or mixed AB associates are not observed. Whereas B is not able to exchange protons, A is subject to a very fast proton transfer. By measuring proton lifetimes as a function of concentration and of the deuterium fraction in the labile proton sites, it was established that two protons are transferred in every exchange process. Thermodynamic data of the association process obtained by the analysis of the chemical shifts, of the A/B populations and the proton lifetimes as a function of concentration agreed very well. These results are evidence that A forms only selfassociated hydrogen bonded dimers with a cyclic structure in which the double proton transfer takes place. The observation of a kinetic HH/HD isotope effect of 20 at 178 K establishes this transfer as the rate limiting step of the overall proton exchange. Rate constants of the double proton transfer in the cyclic dimer were obtained as a function of temperature from which an activation energy of about 19 kJ mol-' was obtained. In addition, the rates of interconversion between the two conformers were determined as a function of temperature. Details of the reaction mechanism and differences to the related carboxylic acids are discussed. Thus, it is shown that dynamic NMR spectroscopy can be a useful tool for elucidating elementary steps of complex reaction networks.

The rate constants of the intermolecular proton exchange in pure methanol, i.e. the reciprocal proton lifetimes, τ0−1, have been determined as a function of the temperature by total lineshape analysis of the 1H‐NMR spectra. Since CH3OH is an AB3 spin system of high order the quantum mechanical density matrix formalism was employed for the simulation of the spectra. The neglect of high order effects as well as the presence of impurities had led to inconsistencies in previous studies. For the first time, the primary kinetic isotope effects were determined indirectly by simulation of the 1H‐NMR spectra of CH3OD samples containing 1 vol‐% CH3OH. The results are given by for CH3OH, and for CH3OD, with a kinetic isotope effect of 3.2 ± 0.4 at 298 K. ‐ The data cannot be explained by a cyclic exchange mechanism. However, they can be quantitatively related to the autoprotolysis constant of methanol and to Grunwald's kinetic data on proton transfer in buffered methanol solutions. It is concluded that the proton lifetimes in pure methanol are determined by the natural amount of free solvated CH3OH2+ and CH3O− ions generated by autoprotolysis. The observed energy of activation is then the sum of two terms, namely the energy of activation of the proton jumps between the ions and a methanol molecule, and half the enthalpy of methanol selfdissociation. In the presence of acid or basic impurities the second term becomes negligible. We propose a method for the quantitative determination of these impurities in the 10−9 to 10−7 mol 1−1 range. From our results we derive an equilibrium isotrope effect of (KH/KD)298K = 6 ± 3 for the autoprotolysis of pure methanol, which has not been reported before. It is shown that the literature data on proton exchange in pure water and other protoc systems can be explained in a similar way. The mechanism of the neutralisation in water and methanol is discussed. For the neutralisation of solutions containing 1 mol of each ion it is found that the mean number of proton transfer steps necessary is 5.3 for water and 2.5 for methanol. These data correlate well with the molar solvent density.

No abstract