TY - JOUR
T1 - Reorganization free energies and quantum corrections for a model electron self-exchange reaction
T2 - Comparison of polarizable and non-polarizable solvent models
AU - Blumberger, Jochen
AU - Lamoureux, Guillaume
N1 - Funding Information:
We want to thank Dr. P. Ren and Dr. J.W. Ponder for providing a modified version of the analyze program of the AMOEBA simulation package. J.B. acknowledges financial support from EPSRC and The Royal Society for a University Research Fellowship, and access to a compute cluster at the Center of Computational Chemistry, University of Cambridge.
PY - 2008/6
Y1 - 2008/6
N2 - The solvent contribution to the reorganization free energy for electron self-exchange in aqueous Ru(II)-Ru(III) is computed for two recently developed polarizable water models, AMOEBA [J. Phys. Chem. B 107, 5933 (2003)] and SWM4-NDP [Chem. Phys. Lett. 418, 241 (2005)], and for the earlier POL3 model [J. Phys. Chem. 99, 6208 (1995)], and compared with the reorganization free energy of non-polarizable water models. The 'solute', defined as the two ions and their first hydration shells, is treated as non-polarizable. We find that the solvent ('outer sphere') reorganization free energy is reduced by 22% for SWM4-NDP and by 11% for POL3 relative to the non-polarizable TIP3P water, but increased by 5% for AMOEBA water. This is less than the ≈ 38% reduction suggested by standard continuum theory and confirms the view that continuum theory predicts a stronger dependence of solvent reorganization on the optical dielectric constant than what is obtained from atomistic simulation. The varying degree of reduction in reorganization free energy for polarizable water models is the consequence of two opposing effects: (i) reduction in reorganization free energy due to decreased electronic response; and (ii) increase in reorganization free energy due to increased nuclear response. The first effect gives a consistent decrease of about 30%, while the second effect strongly depends on the polarizable water model used and is largest for AMOEBA water. Rate enhancements due to quantum corrections are computed in the harmonic bath approximation and range between 3.8 and 10.9, in good agreement with the estimate obtained from experimental dispersion data of liquid water, 7.7. The rigid non-polarizable water models overestimate the quantum correction in the libration modes, which effectively compensates for the neglect of quantum corrections in the absent stretching modes. About 85% of the solvent reorganization is due to the second and third solvation shell of the ion pair. Size effects caused by the finite number of solvent molecules are minor and much smaller than for oxidation of a single ion.
AB - The solvent contribution to the reorganization free energy for electron self-exchange in aqueous Ru(II)-Ru(III) is computed for two recently developed polarizable water models, AMOEBA [J. Phys. Chem. B 107, 5933 (2003)] and SWM4-NDP [Chem. Phys. Lett. 418, 241 (2005)], and for the earlier POL3 model [J. Phys. Chem. 99, 6208 (1995)], and compared with the reorganization free energy of non-polarizable water models. The 'solute', defined as the two ions and their first hydration shells, is treated as non-polarizable. We find that the solvent ('outer sphere') reorganization free energy is reduced by 22% for SWM4-NDP and by 11% for POL3 relative to the non-polarizable TIP3P water, but increased by 5% for AMOEBA water. This is less than the ≈ 38% reduction suggested by standard continuum theory and confirms the view that continuum theory predicts a stronger dependence of solvent reorganization on the optical dielectric constant than what is obtained from atomistic simulation. The varying degree of reduction in reorganization free energy for polarizable water models is the consequence of two opposing effects: (i) reduction in reorganization free energy due to decreased electronic response; and (ii) increase in reorganization free energy due to increased nuclear response. The first effect gives a consistent decrease of about 30%, while the second effect strongly depends on the polarizable water model used and is largest for AMOEBA water. Rate enhancements due to quantum corrections are computed in the harmonic bath approximation and range between 3.8 and 10.9, in good agreement with the estimate obtained from experimental dispersion data of liquid water, 7.7. The rigid non-polarizable water models overestimate the quantum correction in the libration modes, which effectively compensates for the neglect of quantum corrections in the absent stretching modes. About 85% of the solvent reorganization is due to the second and third solvation shell of the ion pair. Size effects caused by the finite number of solvent molecules are minor and much smaller than for oxidation of a single ion.
KW - Intermolecular forces
KW - Organic crystal structure
KW - Polymorphism
UR - http://www.scopus.com/inward/record.url?scp=52149087947&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=52149087947&partnerID=8YFLogxK
U2 - 10.1080/00268970802220112
DO - 10.1080/00268970802220112
M3 - Article
AN - SCOPUS:52149087947
SN - 0026-8976
VL - 106
SP - 1597
EP - 1611
JO - Molecular Physics
JF - Molecular Physics
IS - 12-13
ER -