Authors: Robert Q. Topper(#), Gregory J. Tawa(#), and Donald G. Truhlar(#)
An optimized integration scheme for calculating vibrational-rotational partition functions by the Fourier path-integral method, as presented in the previous paper [R.Q. Topper and D.G. Truhlar, J. Chem. Phys. 97, 3647 (1992)] is applied to a three-dimensional test case involving the coupled vibrational and rotational motions of a diatomic HCl molecule in Cartesian coordinates. Converged partition functions are calculated by the new Fourier path-integral Monte Carlo scheme and by standard variational methods, and the two sets of results are compared. We obtain good agreement (~2%) between the two methods over a range of a factor of 20 in temperature.
(#)Department of Chemistry and Supercomputer Institute, University of Minnesota, Minneapolis, MN.
A reprint of this article may be requested from The Minnesota Supercomputer Institute.
July 13, 2000
The following ERRATUM has been accepted for publication in the Journal of Chemical Physics (2000):
Erratum: "Quantum free-energy calculations: A three-dimensional test case" [J. Chem. Phys. 97, 3668 (1992)]
Equation (6) for the Dunham coefficieent A_0 should read A_0 = (omega_e)^2 / 4 B_e .
In Table I, the following changes are to be made: Value of D_0 in atomic units is 1.6293169 x 10^{-1}; value of B_e in atomic units is 4.826752286 x 10^{-5}; value of alpha_e is 0.307181; and value of r_e in atomic units is 2.408555. Footnote b has an error: 1 cm^{-1} = 4.55637 x 10^{-6} hartree.
Line 5 in column 1 on page 3673 should read "dimensional triatomic.^{16}"
Reference 15 should read "F. James, Rep. Prog. Phys. 43, 1145 (1980)."
These are all typographic errors and do not affect the calculations or conclusions. For the convenience of those who wish to do further work on this system, we note that b = 11.156 and c = 6.895 x 10^{-3}.
The authors are grateful to Steven Mielke and Jay Srinivasan for assistance in identifying the errors.
Back to Prof. Topper's Home Page