Melting of aluminum, molybdenum, and the light actinides

A semiempirical model was developed in order to explain why the measured melting curves of molybdenum, and the other bcc transition metals, have an unusually low slope (dT/dPsimilar to0). The total binding energy of Mo is written as the sum of the repulsive energy of the ions and sp electrons (modeled by an inverse sixth power potential) and the d-band cohesive energy is described by the well known Friedel equation. Using literature values for the Mo band width energy, the number of d electrons and their volume dependence, we find that a small broadening of the liquid d-band width (similar to1%) leads to an increase in the stability of the liquid relative to the solid. This is sufficient to depress the melting temperature and lower the melting slope to a value in agreement with the recent diamond-anvil cell measurements. Omission of the d-band physics results in an Al-like melting curve with a much steeper melt slope. The model, when applied to the f electrons of the light actinides (Th-Am), gives agreement with the observed fall and rise in the melting temperature with increasing atomic number.