14 examples of how LLMs can transform materials science and chemistry: a reflection on a large language model hackathon.

Large-language models (LLMs) such as GPT-4 caught the interest of many scientists. Recent studies suggested that these models could be useful in chemistry and materials science. To explore these possibilities, we organized a hackathon. This article chronicles the projects built as part of this hackathon. Participants employed LLMs for various applications, including predicting properties of molecules and materials, designing novel interfaces for tools, extracting knowledge from unstructured data, and developing new educational applications. The diverse topics and the fact that working prototypes could be generated in less than two days highlight that LLMs will profoundly impact the future of our fields. The rich collection of ideas and projects also indicates that the applications of LLMs are not limited to materials science and chemistry but offer potential benefits to a wide range of scientific disciplines.

Machine Learning-Accelerated Discovery of A2BC2 Ternary Electrides with Diverse Anionic Electron Densities.

This study combines machine learning (ML) and high-throughput calculations to uncover new ternary electrides in the A2BC2 family of compounds with the P4/mbm space group. Starting from a library of 214 known A2BC2 phases, density functional theory calculations were used to compute the maximum value of the electron localization function, indicating that 42 are potential electrides. A model was then trained on this data set and used to predict the electride behavior of 14,437 hypothetical compounds generated by structural prototyping. Then, the stability and electride features of the 1254 electride candidates predicted by the model were carefully checked by high-throughput calculations. Through this tiered approach, 41 stable and 104 metastable new A2BC2 electrides were predicted. Interestingly, all three kinds of electrides, i.e., electron-deficient, electron-neutral, and electron-rich electrides, are present in the set of predicted compounds. Three of the most promising new electrides (two electron-rich, Nd2ScSi2 and La2YbGe2, and one electron-deficient Y2LiSi2) were then successfully synthesized and characterized experimentally. Furthermore, the synthesized electrides were found to exhibit high catalytic activities for NH3 synthesis under mild conditions when Ru-loaded. The electron-deficient Y2LiSi2, in particular, was seen to exhibit a good balance of catalytic activity and chemical stability, suggesting its future application in catalysis.

Phase Transformations and Phase Segregation during Potassiation of Sn x P y Anodes.

K-ion batteries (KIBs) have the potential to offer a cheaper alternative to Li-ion batteries (LIBs) using widely abundant materials. Conversion/alloying anodes have high theoretical capacities in KIBs, but it is believed that electrode damage from volume expansion and phase segregation by the accommodation of large K-ions leads to capacity loss during electrochemical cycling. To date, the exact phase transformations that occur during potassiation and depotassiation of conversion/alloying anodes are relatively unexplored. In this work, we synthesize two distinct compositions of tin phosphides, Sn4P3 and SnP3, and compare their conversion/alloying mechanisms with solid-state nuclear magnetic resonance (SSNMR) spectroscopy, powder X-ray diffraction (XRD), and density functional theory (DFT) calculations. Ex situ 31P and 119Sn SSNMR analyses reveal that while both Sn4P3 and SnP3 exhibit phase separation of elemental P and the formation of KSnP-type environments (which are predicted to be stable based on DFT calculations) during potassiation, only Sn4P3 produces metallic Sn as a byproduct. In both anode materials, K reacts with elemental P to form K-rich compounds containing isolated P sites that resemble K3P but K does not alloy with Sn during potassiation of Sn4P3. During charge, K is only fully removed from the K3P-type structures, suggesting that the formation of ternary regions in the anode and phase separation contribute to capacity loss upon reaction of K with tin phosphides.

OPTIMADE, an API for exchanging materials data.

The Open Databases Integration for Materials Design (OPTIMADE) consortium has designed a universal application programming interface (API) to make materials databases accessible and interoperable. We outline the first stable release of the specification, v1.0, which is already supported by many leading databases and several software packages. We illustrate the advantages of the OPTIMADE API through worked examples on each of the public materials databases that support the full API specification.

Robust model benchmarking and bias-imbalance in data-driven materials science: a case study on MODNet.

As the number of novel data-driven approaches to material science continues to grow, it is crucial to perform consistent quality, reliability and applicability assessments of model performance. In this paper, we benchmark the Materials Optimal Descriptor Network (MODNet) method and architecture against the recently released MatBench v0.1, a curated test suite of materials datasets. MODNet is shown to outperform current leaders on 6 of the 13 tasks, while closely matching the current leaders on a further 2 tasks; MODNet performs particularly well when the number of samples is below 10 000. Attention is paid to two topics of concern when benchmarking models. First, we encourage the reporting of a more diverse set of metrics as it leads to a more comprehensive and holistic comparison of model performance. Second, an equally important task is the uncertainty assessment of a model towards a target domain. Significant variations in validation errors can be observed, depending on the imbalance and bias in the training set (i.e., similarity between training and application space). By using an ensemble MODNet model, confidence intervals can be built and the uncertainty on individual predictions can be quantified. Imbalance and bias issues are often overlooked, and yet are important for successful real-world applications of machine learning in materials science and condensed matter.

Computational Investigation of Copper Phosphides as Conversion Anodes for Lithium-Ion Batteries.

Using first-principles structure searching with density-functional theory (DFT), we identify a novel Fm3̅m phase of Cu2P and two low-lying metastable structures, an I4̅3d-Cu3P phase and a Cm-Cu3P11 phase. The computed pair distribution function of the novel Cm-Cu3P11 phase shows its structural similarity to the experimentally identified Cm-Cu2P7 phase. The relative stability of all Cu-P phases at finite temperatures is determined by calculating the Gibbs free energy using vibrational effects from phonon modes at 0 K. From this, a finite-temperature convex hull is created, on which Fm3̅m-Cu2P is dynamically stable and the Cu3-x P (x < 1) defect phase Cmc21-Cu8P3 remains metastable (within 20 meV/atom of the convex hull) across a temperature range from 0 to 600 K. Both CuP2 and Cu3P exhibit theoretical gravimetric capacities higher than contemporary graphite anodes for Li-ion batteries; the predicted Cu2P phase has a theoretical gravimetric capacity of 508 mAh/g as a Li-ion battery electrode, greater than both Cu3P (363 mAh/g) and graphite (372 mAh/g). Cu2P is also predicted to be both nonmagnetic and metallic, which should promote efficient electron transfer in the anode. Cu2P's favorable properties as a metallic, high-capacity material suggest its use as a future conversion anode for Li-ion batteries; with a volume expansion of 99% during complete cycling, Cu2P anodes could be more durable than other conversion anodes in the Cu-P system, with volume expansions greater than 150%. The structures and figures presented in this paper, and the code used to generate them, can be interactively explored online using Binder.

Sodiation and Desodiation via Helical Phosphorus Intermediates in High-Capacity Anodes for Sodium-Ion Batteries.

Na-ion batteries are promising alternatives to Li-ion systems for electrochemical energy storage because of the higher natural abundance and widespread distribution of Na compared to Li. High capacity anode materials, such as phosphorus, have been explored to realize Na-ion battery technologies that offer comparable performances to their Li-ion counterparts. While P anodes provide unparalleled capacities, the mechanism of sodiation and desodiation is not well-understood, limiting further optimization. Here, we use a combined experimental and theoretical approach to provide molecular-level insight into the (de)sodiation pathways in black P anodes for sodium-ion batteries. A determination of the P binding in these materials was achieved by comparing to structure models created via species swapping, ab initio random structure searching, and a genetic algorithm. During sodiation, analysis of 31P chemical shift anisotropies in NMR data reveals P helices and P at the end of chains as the primary structural components in amorphous Na xP phases. X-ray diffraction data in conjunction with variable field 23Na magic-angle spinning NMR support the formation of a new Na3P crystal structure (predicted using density-functional theory) on sodiation. During desodiation, P helices are re-formed in the amorphous intermediates, albeit with increased disorder, yet emphasizing the pervasive nature of this motif. The pristine material is not re-formed at the end of desodiation and may be linked to the irreversibility observed in the Na-P system.