Complex amorphous oxides: property prediction from high throughput DFT and AI for new material search

Setten, Michiel J. van; Dekkers, H.F.W.; Pashartis, Christopher; Chasin, Adrian; Belmonte, Attilio; Delhougne, Romain; Kar, Gouri Sankar; Pourtois, Geoffrey

doi:10.1039/d2ma00759b

Cited by 4 publications

(4 citation statements)

References 127 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This shows that the models are capable of learning the difference between a semilocal and a hybrid functional. Similar effects have been observed previously . The important lesson here is that including DFT results can be done using a cheap semilocal functional and no significant improvement can be achieved by moving to a more expensive hybrid functional.…”

Section: Resultssupporting

confidence: 73%

“…Similar effects have been observed previously. 10 The important lesson here is that including DFT results can be done using a cheap semilocal functional and no significant improvement can be achieved by moving to a more expensive hybrid functional. Moreover, it shows that the deviations from the results from Choi et al 29 are likely to originate from differences in the used dataset or pre-or postprocessing-related differences.…”

Section: ■ Results and Discussionmentioning

confidence: 99%

“…Machine learning has proven to be a powerful tool in chemistry, particularly for screening large datasets, forecasting molecular properties, − and predicting new functional molecules and materials. ,− These successes paved the way for the application of machine learning to the prediction of accurate activation energies of reactions. − The most recent achievements were able to predict kinetic barriers with an average error in the range of 5 kcal mol –1 from their measured values using simple molecular descriptors. Such predictions are especially useful in determining whether a chemical process is kinetically limited.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Improving Accuracy and Transferability of Machine Learning Chemical Activation Energies by Adding Electronic Structure Information

Marques

Gendt

Pourtois

et al. 2023

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Predicting chemical activation energies is one of the longstanding and important challenges in computational chemistry. Recent advances have shown that machine learning can be used to create tools to predict them. Such tools can significantly decrease the computational cost for these predictions compared to traditional methods, which require an optimal path search along a high-dimensional potential energy surface. To enable this new route, we need both large and accurate datasets and a compact yet complete description of the reactions. Although data for chemical reactions is becoming increasingly available, the key step of encoding the reaction as an efficient descriptor remains a big challenge. In this paper, we demonstrate that including electronic energy levels in the description of the reaction significantly improves the prediction accuracy and transferability. Feature importance analysis further demonstrates that electronic energy levels have a higher importance than some structural information and typically require less space in the reaction encoding vector. In general, we observe that the results of the feature importance analysis relate well to the domain knowledge of fundamental chemical principles. This work can help to build better chemical reaction encodings for machine learning and thus improve the predictions of machine learning models for reaction activation energies. These models could ultimately be used to recognize reaction limiting steps in large reaction systems, allowing to account for bottlenecks at the design stage.

show abstract

Section: Resultssupporting

confidence: 73%

Section: ■ Results and Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improving Accuracy and Transferability of Machine Learning Chemical Activation Energies by Adding Electronic Structure Information

Marques

Gendt

Pourtois

et al. 2023

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent years, the HTC approach has been applied in screening high‐performance semiconductors such as thermoelectric, ferromagnetic, battery materials, and so on. It brought up some new routes for discovering new semiconducting materials and improving the performances of related devices [274] …”

Section: Applications Of Htc In Materials Developmentmentioning

confidence: 99%

Advances in data‐assisted high‐throughput computations for material design

Xu,

Zhang,

Huo

et al. 2023

Materials Genome Engineering Advances

View full text Add to dashboard Cite

Extensive trial and error in the variable space is the main cause of low efficiency and high cost in material development. The experimental tasks can be reduced significantly in the case that the variable space is narrowed down by reliable computer simulations. Because of their numerous variables in material design, however, the variable space is still too large to be accessed thoroughly even with a computational approach. High‐throughput computations (HTC) make it possible to complete a material screening in a large space by replacing the conventionally manual and sequential operations with automatic, robust, and concurrent streamlines. The efficiency of HTC, which is one of the pillars of materials genome engineering, has been verified in many studies, but its applications are still limited by demanding computational costs. Introduction of data mining and artificial intelligence into HTC has become an effective approach to solve the problem. In the past years, many studies have focused on the development and application of HTC and data combined approaches, which is considered as a new paradigm in computational materials science. This review focuses on the main advances in the field of data‐assisted HTC for material research and development and provides our outlook on its future development.

show abstract

Computing elastic tensors of amorphous materials from first-principles

Pashartis,

van Setten,

Houssa

et al. 2024

Computational Materials Science

View full text Add to dashboard Cite

Complex amorphous oxides: property prediction from high throughput DFT and AI for new material search

Abstract: With decreasing dimensions and increasing complexity, semiconductor devices are getting more difficult to fabricate. Especially the allowed deposition temperature becomes lower. Amorphous materials, which do not require annealing steps, are...

Cited by 4 publications

References 127 publications

Improving Accuracy and Transferability of Machine Learning Chemical Activation Energies by Adding Electronic Structure Information

Improving Accuracy and Transferability of Machine Learning Chemical Activation Energies by Adding Electronic Structure Information

Advances in data‐assisted high‐throughput computations for material design

Computing elastic tensors of amorphous materials from first-principles

Contact Info

Product

Resources

About