The goal of the activities on the elucidation of the anomalous properties of liquid and amorphous metals has been two-fold. On the one hand it has been pushed towards a detailed comparison of numerical calculations and experimental data. Among other puzzles we have investigated the occurrence of a pseudo-gap in the one-particle spectrum and its consequences on transport properties. Muffin-tin EMA indeed yields such a pseudo-gap,which becomes more and more pronounced as short-ranged order increases. As a result anomalous temperature dependence of the transport coefficients is expected. Indeed this is the route to the explanation of the experimentally observed Mooij rule as shown by our numerical results,and vertex corrections augment the sensitivity of the conductivity in the temperature. This also allows for a natural explanation of the sign of Q/T (Q being the thermopower),and its amplitude,at low temperature. A publication is in preparation (U.Werner and R.Fresard).
As a related problem, metal-to-insulator transitions have also been studied. Substantial achievement on how to handle the measure in slave boson functional integral has been reached. It has been shown how to proceed with radial fields in such functional integrals thanks to a compensation lemma,and the work will be published (L.Gellhoff and R.Fresard,J.Phys. : Cond.Matt.,at press). The lemma has then been applied to more involved slave boson representations,to calculate the spin and charge structure factors of the 2-d Hubbard Model in the vicinity of the Mott transition. The charge structure factor has been found to agree quantitatively with numerical simulations,both in the weak and the strong coupling regimes (W.Zimmermann,R.Fresard and P.Wolfle,in preparation).