Infrared intensities and Raman-scattering activities within density-functional theory

Abstract

We show that the computational complexity associated with the density-functional-based determination of infrared intensities and nonresonant Raman scattering activities is the same as that required for vibrational modes. Further, we use extremely large basis sets to determine the intrinsic accuracy for calculating such phenomena within the density-functional theory. We present benchmark calculations on ${\mathrm{CH}}_{4}$, ${\mathrm{H}}_{2}$O, ${\mathrm{C}}_{2}$${\mathrm{H}}_{2}$, ${\mathrm{C}}_{2}$${\mathrm{H}}_{4}$, and ${\mathrm{C}}_{2}$${\mathrm{H}}_{6}$ within both the local-density approximation (LDA) and the generalized gradient approximation (GGA). Tests of the reliability and numerical stability of the theoretical scheme are presented. We show that in order to obtain reliable results, appropriate polarization basis functions and well-converged wave functions are necessary. While most of the Raman spectra predicted by LDA agree very well with experimental data, some of the infrared intensities show substantial errors. The GGA functional overcomes most of these deficiencies, leading to an overall good agreement with experiment. \textcopyright{} 1996 The American Physical Society.