Abstract
Bulk moduli appear readily accessible in electronic structure calculations, but the calculated values are often substantially greater than experimental bulk moduli. This discrepancy is the result of an unfair comparison of calculated and experimental results: many workers ignored the zero-point and finite-temperature effects that are present in experiments but absent from most calculations. These effects can alter bulk moduli by up to 20%. We show how good approximations to the required corrections may be obtained with little effort. We also deal with the statistical errors and biases in quantities derived from the noisy energy-volume curves produced by quantum Monte Carlo simulations.
- Received 8 November 2001
DOI:https://doi.org/10.1103/PhysRevB.66.052104
©2002 American Physical Society