A prediction-driven database to enable rapid discovery of nonlinear optical materials

Nonlinear optical (NLO) materials capable of generating coherent light by second harmonic generation (SHG) are urgently needed in modern laser technology. Despite the importance of this class of materials, only a small fraction of non-centrosymmetric (NCS) crystalline compounds with finite electronic band gaps which exhibit SHG response have been studied experimentally or theoretically, and thus an efficient exploration of NLO materials with high performance is greatly limited. In this work, we introduce to date the largest computational NLO materials database which contains calculated SHG coefficients for 2354 NCS crystal structures as well as their electronic band gaps. The computational details and validation of data are presented, along with a detailed description of the database helpful for the usage of these computational results. One of the main features of the present database is that it contains a large number of new thermodynamically stable and metastable structures discovered through evolutionary algorithm searches, providing possibilities for finding novel NLO materials with good properties.