The ideal next-generation thin film PV absorbing materials need to satisfy a multitude of desired material properties to reach high efficiency. Considering only the band gap size (e.g., Schockley-Queisser formula) has proven to be inefficient to the screening of good PV absorbers. In this talk, first, I will present a simple selection metric called “Spectroscopic Limited Maximum Efficiency (SLME)”  for the more effective initial screening. It considers not only the band gap, but also the absorption spectra and optical-type dependent radiative/nonradiative recombination losses. By high-throughput first-principles quasiparticle calculations of SLME for ~1000 group I-III-VI and I-V-VI materials, we have identified tens of new potential PV absorber materials. Second, I will discuss some other physical factors that may determine the solar cell device efficiency except those considered in SLME. As an example, I will talk about the problem of iron pyrite (FeS2) as solar cell, which has high theoretical SLME, but low real efficiency in experiment due to too low open circuit voltage. We find that this problem is related to phase separation near surface/interface region at moderate temperature . Based this understanding, we come up with a new design principles for Fe-based PV absorber material, according to which, two new potential earth-abundant ternary materials (Fe2SiS4 and Fe2GeS4) have been proposed . Finally, I will talk about some of our ongoing and future work on the interfacial physics of metal oxides. Particularly, I will discuss the role of different defects playing in the observed metal-like conducting behavior at the interface between two band insulators (LaAlO3 and SrTiO3).
 Liping Yu and Alex Zunger, Phys. Rev. Lett. 108, 068701 (2012)
 Liping Yu et al., Adv. Energy Mater. 1, 748–753 (2011)