Universal machine learning framework for defect predictions in zinc blende semiconductors

Mannodi-Kanakkithodi, Arun; Xiang, Xiaofeng; Jacoby, Laura; Biegaj, Robert; Dunham, Scott T.; Gamelin, Daniel R.; Chan, Maria K. Y.

doi:10.1016/j.patter.2022.100450

Cited by 32 publications

(31 citation statements)

References 85 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Eqn (3) will yield the Fermi level where V A displays a (0/À1) charge transition or where V X displays a (+1/0) transition. 15,16,101,102 Based on these equations, four types of defect properties are thus computed, namely the two vacancy charge transition levels, the equilibrium defect formation energy (D.F.E. ), and the equilibrium Fermi level E F ; the latter two are estimated based on the charge neutrality condition.…”

Section: Papermentioning

confidence: 99%

Data-driven design of novel halide perovskite alloys

Mannodi-Kanakkithodi

Chan

2022

Energy Environ. Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Papermentioning

confidence: 99%

Data-driven design of novel halide perovskite alloys

Mannodi-Kanakkithodi

Chan

2022

Energy Environ. Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, machine learning techniques have been proposed as a faster method for predicting defect energetics, but so far they still require time-consuming defect data generation for model training and limit the applicability and generalizability of the defect energetic predictions. [18][19][20][21][22][23] Especially, graph neural network based deep-learning models [24][25][26][27] have become very popular for predicting materials properties and have been used for several bulk property predictions and their applicability needs to be tested for defect property predictions. Two key ingredients needed for accomplishing this task are: 1) a pretrained deep-learning model that can directly predict the total energy of perfect and defect structures, 2) a test DFT dataset of vacancy formation energies on which the DL model could be applied.…”

mentioning

confidence: 99%

A Deep-learning Model for Fast Prediction of Vacancy Formation in Diverse Materials

Choudhary¹,

Sumpter²

2022

Preprint

View full text Add to dashboard Cite

The presence of point defects such as vacancies plays an important role in material design. Here, we demonstrate that a graph neural network (GNN) model trained only on perfect materials can also be used to predict vacancy formation energies (E vac ) of defect structures without the need for additional training data. Such GNN-based predictions are considerably faster than density functional theory (DFT) calculations with reasonable accuracy and show the potential that GNNs are able to capture a functional form for energy predictions. To test this strategy, we developed a DFT dataset of 508 E vac consisting of 3D elemental solids, alloys, oxides, nitrides, and 2D monolayer materials. We analyzed and discussed the applicability of such direct and fast predictions. We applied the model to predict 192494 E vac for 55723 materials in the JARVIS-DFT database.Defects play an important role in our pursuit to engineer performance of a material. Vacancies are a type of defects which are ubiquitous and their presence can significantly alter 1

show abstract

“…Although the Heyd-Scuseria-Ernzerhof (HSE) functional may sometimes offer a bandgap that is near to the experimental one, it is not universal for every semiconductor, also the associated computational cost is higher for HSE. 56,[92][93][94][95][96] For example, the bandgap of ZnO is also underestimated by the hybrid HSE06 (∼2.70 eV) functional. 91 However, the prior report shows that GGA+U could be used to accurately report the bandgap (∼3.3 eV) of ZnO which is almost close to the experimental one.…”

Section: Resultsmentioning

confidence: 99%