Inverse Design of High-Performance Thermoelectric Materials via a Generative Model Combined with Experimental Verification

The emergence of inverse design approaches leveraging generative models offers a promising avenue for thermoelectric material design. However, these models heavily depend on diverse training data, and current thermoelectric data sets are limited, primarily encompassing group IV-VI materials operating within moderate temperature ranges. This constraint poses a significant challenge in the pursuit of materials with high thermoelectric figure of merit (<i>zT</i>) through generative modeling. Our study introduces an inverse design model tailored for the constrained thermoelectric materials data set. By augmenting the data with 2000 entries from the experimental literature and incorporating a generative model featuring a diversity loss function and residual network (ResNet) architecture to enhance complexity, our approach has been trained to systematically generate high-<i>zT</i> thermoelectric materials across various temperature ranges. Under predefined high-<i>zT</i> criteria, our deep generative model successfully predicted 100 doped materials with <i>zT</i> values exceeding 1.0. Furthermore, this research analyzes density of states (DOS) plots for the generated materials, identifying 25 unreported previously potential thermoelectric candidates in the material database. Notably, we experimentally validated the synthesis of Mg_3.1Sb_0.5Bi_1.497Te_0.003, a representative thermoelectric material from the Mg₃(Sb, Bi)₂ family suitable for room temperature applications. This validation underscores the efficacy of our model in exploring and discovering novel thermoelectric materials.