The interstitial electron model for metals-a critical appraisal

Abstract

The interstitial electron model (IEM) of Li and Goddard (1989, 1993) has reproduced with good success the phonon dispersion curves of many face-centred cubic (FCC) metals. The key feature of the model is the location of light interstitial particles in the tetrahedral holes of the structure. In this paper the validity of the model is further tested: (i) by applying it to three additional FCC metals, beta -Co, La and Th; (ii) by using it to calculate anharmonic properties; (iii) by considering extensions to hexagonal close-packed (HCP) and body-centred cubic (BCC) metals. These demanding tests reveal several shortcomings. There is good agreement for the experimental dispersion curves of beta -Co and La, but only if additional interactions and input data are used in the fitting; the agreement for Th is only fair. Applied to anharmonic properties of FCC metals, the IEM cannot explain simultaneously both the temperature dependence of the Gruneisen function and the pressure dependence of the second-order elastic stiffnesses. Other than for Mg and alpha -Co, it cannot reproduce the frequency dispersion curves of HCP metals. Its use for BCC systems is impractical due to the large number of parameters required. Even where the model is successful in predicting dispersion curves, equally satisfactory agreement can be obtained with a valence force-field, sometimes with fewer adjustable parameters.