Search for: All records

Abstract Data-driven approaches to materials exploration and discovery are building momentum due to emerging advances in machine learning. However, parsimonious representations of crystals for navigating the vast materials search space remain limited. To address this limitation, we introduce a materials discovery framework that utilizes natural language embeddings from language models as representations of compositional and structural features. The contextual knowledge encoded in these language representations conveys information about material properties and structures, enabling both similarity analysis to recall relevant candidates based on a query material and multi-task learning to share information across related properties. Applying this framework to thermoelectrics, we demonstrate diversified recommendations of prototype crystal structures and identify under-studied material spaces. Validation through first-principles calculations and experiments confirms the potential of the recommended materials as high-performance thermoelectrics. Language-based frameworks offer versatile and adaptable embedding structures for effective materials exploration and discovery, applicable across diverse material systems. 

Successful dopability in AgInTe₂requires careful navigation of the compensating intrinsic defects to maximize dopant solubility and efficiency. 

We used the 138Baðd; αÞ reaction to carry out an in-depth study of states in 136Cs, up to around 2.5 MeV. In this Letter, we place emphasis on hitherto unobserved states below the first 1þ level, which are important in the context of solar neutrino and fermionic dark matter (FDM) detection in large-scale xenon-based experiments. We identify for the first time candidate metastable states in 136Cs, which would allow a realtime detection of solar neutrino and FDM events in xenon detectors, with high background suppression. Our results are also compared with shell-model calculations performed with three Hamiltonians that were previously used to evaluate the nuclear matrix element (NME) for 136Xe neutrinoless double beta decay.We find that one of these Hamiltonians, which also systematically underestimates the NME compared with the others, dramatically fails to describe the observed low-energy 136Cs spectrum, while the other two show reasonably good agreement.