 |
Center for Nonlinear Studies |
Hedin1 showed that if we could determine only four functions – the Green’s function, the self-energy, the polarization, and the vertex correction – we could exactly solve the many-body perturbation series. This is an extremely difficult task, as all these functions must be determined self-consistently. However, a second issue bedeviled early attempts at a solution: how does one determine the bare hopping and interaction parameters when the experimental parameters are strongly renormalized by the interactions? In recent years a hypothesis has arisen: that an appropriate density-functional theory (DFT) calculation can be used to derive the bare parameters. This idea lies at the basis of the many recent ‘DFT+X’ approaches to correlated materials, particularly when X includes self-energy effects, via some GW calculation, or some vertex corrections. In this talk, I discuss the particular version of the model that our group has developed. This involves two self-consistent steps. (1) First, we extend the commonly-used G0W0 self energy to a one-parameter self-consistent quasipqrticle-GW (QPGW) approach, more appropriate when the renormalization parameter Z is significantly less than one.2 (2) While this works well at low temperatures, the phase transitions are treated as mean-field. To improve on this, we incorporate vertex corrections to include the effects of fluctuations3. The resulting model satisfies the Mermin-Wagner theorem, and more importantly includes strong mode-coupling effects. The resulting frustration due to competition between conventional Fermi-surface nesting and Van Hove nesting has many characteristics of the pseudogap phase. 1. L. Hedin, J. Phys. Condens. Matter 11, R489-R528 (1999). 2. Tanmoy Das, R.S. Markiewicz, and A. Bansil, Adv. Phys. 63, 151-266 (2014). 3. R.S. Markiewicz, I.G. Buda, P. Mistark, and A. Bansil, arXiv:1505.04770.
Host: Filip Ronning