Time-dependent coupled Hartree–Fock calculations of the frequency-dependent mean polarizability α(ω) and its anisotropy Δα(ω) are presented for a wide range of molecules containing first and second row atoms. The Cauchy moments S(−2) and S(−4), which describe the dispersion in α(ω), are systematically underestimated by 20% and 40%, respectively, with the 6-31G(+sd+sp) basis set. The dispersion in Δα(ω) is also underestimated, typically by 50%. Many exceptions to this behavior are observed, most of which appear to be an underestimate of the experimental dispersion in Δα(ω) by a greater amount. This seems to be the result of systematic experimental overestimates of dispersion in Δα(ω) for molecules with very small Rayleigh depolarization ratios (ρ0<10−2), suggesting that the vibrational Raman contribution in these experiments is not only important, but significantly frequency dependent.
No takes yet. Share an insight, caveat, or question.
Mark A. Spackman (1991) studied this question.