Computational Discovery of Novel Semiconductors for Beyond-Si Photovoltaics

Abstract

Novel and efficient solar cell materials are required to balance the energy needs of the constantly growing population. Given the disadvantages of silicon solar cells, in this study, we perform computational investigation and scanning of pnictide compounds (phosphides and nitrides) to evaluate their ability to be used as prospective beyond Silicon solar cell materials. We postulate crystal structures for redox-inactive, non-toxic and relatively unexplored chemical space of ABCX2, BB’B”X2 and A4BX2 (A = Li, Na, K; B = Ca, Sr, Mg, Zn; C = Al, Ga, In; X = N, P) composition classes based on experimentally stable compounds and employ Density Functional Theory to compute their equilibrium structures, polymorph stability, compositional stability (0K convex hull diagram), electronic properties (density of states, band structure and effective mass) and intrinsic defect formation energies to filter these compounds based on criteria set on all the above properties for declaring them as suitable candidates for beyond-Si photovoltaic materials. Upon evaluating the above-mentioned properties for the eight compositions in the BB’B”X2 class, 24 compositions in the A4BX2 class and 72 compositions in the ABCX2 class, we find that 31 compositions are stable or metastable with the postulated structures, and four compounds viz. K4ZnP2, Na4ZnP2 of the R-3mH space group and NaSrInN2 and NaCaInN2 of the P21/c space group have optimal direct (or nearest direct) bandgaps and resist forming intrinsic defects. Therefore, we propose that these compounds can be prospective candidates for the utility in beyond-Si photovoltaic materials.