Systematics of metal-insulator interfacial energies: A new rule for wetting and strong catalyst-support interactions

This work concerns three questions: How can one decide whether a particular liquid metal will wet a given oxide substrate? What does analysis of experimental data tell one about the nature of the metal-substrate interactions? Is there correlation with strong metal-support intereactions in catalysis? Comparison of the trends predicted by the continuum results of Barrera and Duke with those observed shows that the dispersion contribution to metal-oxide interfacial energies cannot be the only important term. From the observed trends for non-reactive metal-oxide systems one deduces: (i) Wetting should be favoured by a low metal plasma frequency (i.e., low electron density). (ii) Wetting should be favoured by a small insulator bandgap. (iii) The insulator refractive index is especially useful as a classifier. It appears to be a rule that only for substrates with n2 ⪆ 4.5–5 does one find wetting by most non-reactive metals. The same rule is found for strong metal-support interaction, which should therefore correlate with wetting. (iv) Trace impurity effects are very pronounced whenever the impurity can react with the insulating substrate. Both the classification and the recognition of trace impurity effects have potential consequences in many applied problems, including substrate effects in support catalysts, liquid phase sintering and liquid metal embrittlement.