The use of experimental data in constraining the tight-binding band parameters of quasi-two-dimensional organic molecular metals: application to -(BEDT-TTF)2KHg(SCN)4

Abstract

Whilst tight-binding bandstructure calculations are very successful in describing the Fermi-surface configuration in many quasi-two-dimensional organic molecular metals, the detailed topology of the predicted Fermi surface often differs from that measured in experiments. This is very significant when, for example, the formation of a density-wave state depends critically on details of the nesting of Fermi-surface sheets. These differences between theory and experiment probably result from the limited accuracy to which the -orbitals of the component molecules (which give rise to the transfer integrals of the tight-binding bandstructure) are known. In order to surmount this problem, we have derived a method whereby the transfer integrals within a tight-binding bandstructure model are adjusted until the detailed Fermi-surface topology is in good agreement with a wide variety of experimental data. The method is applied to the charge-transfer salt -(BEDT-TTF)2KHg(SCN)4, the Fermi surface of which has been the source of much speculation in recent years. The Fermi surface obtained differs in detail from previous bandstructure calculation findings. In particular, the quasi-one-dimensional component of the Fermi surface is more strongly warped. This implies that upon nesting of these sheets, significant parts of the quasi-one-dimensional sheets remain, leading to a complicated Fermi-surface topology within the low-temperature, low-magnetic-field phase. In contrast to previous models of this phase, the model for the reconstructed Fermi surface in this work can explain virtually all of the current experimental observations in a consistent manner.