TY - JOUR
T1 - Accurate formation enthalpies of solids using reaction networks
AU - Fromsejer, Rasmus
AU - Maribo-Mogensen, Bjørn
AU - Kontogeorgis, Georgios M.
AU - Liang, Xiaodong
N1 - Publisher Copyright:
© The Author(s) 2024.
PY - 2024
Y1 - 2024
N2 - Crystalline solids play a fundamental role in a host of materials and technologies, ranging from pharmaceuticals to renewable energy. The thermodynamic properties of these solids are crucial determinants of their stability and therefore their behavior. The advent of large density functional theory databases with properties of solids has stimulated research on predictive methods for their thermodynamic properties, especially for the enthalpy of formation ΔfH. Increasingly sophisticated artificial intelligence and machine learning (ML) models have primarily driven development in this field in recent years. However, these models can suffer from lack of generalizability and poor interpretability. In this work, we explore a different route and develop and evaluate a framework for the application of reaction network (RN) theory to the prediction of ΔfH of crystalline solids. For an experimental dataset of 1550 compounds we are able to obtain a mean absolute error w.r.t ΔfH of 29.6 meV atom−1 using the RN approach. This performance is better than existing ML-based predictive methods and close to the experimental uncertainty. Moreover, we show that the RN framework allows for straightforward estimation of the uncertainty of the predictions.
AB - Crystalline solids play a fundamental role in a host of materials and technologies, ranging from pharmaceuticals to renewable energy. The thermodynamic properties of these solids are crucial determinants of their stability and therefore their behavior. The advent of large density functional theory databases with properties of solids has stimulated research on predictive methods for their thermodynamic properties, especially for the enthalpy of formation ΔfH. Increasingly sophisticated artificial intelligence and machine learning (ML) models have primarily driven development in this field in recent years. However, these models can suffer from lack of generalizability and poor interpretability. In this work, we explore a different route and develop and evaluate a framework for the application of reaction network (RN) theory to the prediction of ΔfH of crystalline solids. For an experimental dataset of 1550 compounds we are able to obtain a mean absolute error w.r.t ΔfH of 29.6 meV atom−1 using the RN approach. This performance is better than existing ML-based predictive methods and close to the experimental uncertainty. Moreover, we show that the RN framework allows for straightforward estimation of the uncertainty of the predictions.
UR - https://doi.org/10.11583/DTU.25898161.v2
U2 - 10.1038/s41524-024-01404-5
DO - 10.1038/s41524-024-01404-5
M3 - Journal article
AN - SCOPUS:85206348410
SN - 2057-3960
VL - 10
JO - npj Computational Materials
JF - npj Computational Materials
IS - 1
M1 - 244
ER -