Finding New Perovskite Halides via Machine Learning

Pilania, Ghanshyam; Balachandran, Prasanna V.; Kim, Chiho; Lookman, Turab

doi:10.3389/fmats.2016.00019

Cited by 159 publications

(101 citation statements)

References 36 publications

Supporting

Mentioning

100

Contrasting

Order By: Relevance

“…Approximately 220 ABX 3 halide perovskites were found to the Goldschmidt tolerance factor of 0.98–1.00. Furthermore, another investigation has shown that a certain data set of ABX 3 halide perovskites with high formability has the octahedral factor, the ratio of r B to r A , between approximately 0.45 and 0.70 . Therefore, in addition to a narrow Goldshcmidt Tolerance factor ranging from 0.98 to 1.00 for the high probability of being cubic, the perovskites were filtered using the octahedral factor of 0.45–0.70.…”

Section: Resultsmentioning

confidence: 99%

“…Advanced quantum mechanical DFT modeling offers a great opportunity for the materials science community to characterize the electronic, structural, thermodynamic, and mechanical properties of thousands of materials, especially when combined with high‐throughput screening techniques ,,. As such, intensive investigations of the immense chemical spectrum are being increasingly pursued, which enables scientists to design and discover new photovoltaic materials with optimal properties …”

Section: Introductionmentioning

confidence: 99%

“…ML is one of artificial intelligence algorithms that can achieve its predictive capability by learning from data . Through recent reports, it has been demonstrated that ML can be used in materials science in order to predict complex behaviors of materials and thereby to help explore vast chemical space for new materials discovery ,,. For instance, once a machine learning algorithm is properly trained, it may provide the immediate prediction of electrical properties of materials, which otherwise would have taken significantly longer time to obtain results from experiments or other computations such as DFT.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Density Functional Theory – Machine Learning Approach to Analyze the Bandgap of Elemental Halide Perovskites and Ruddlesden‐Popper Phases

et al. 2018

View full text Add to dashboard Cite

In this study, we have developed a protocol for exploring the vast chemical space of possible perovskites and screening promising candidates. Furthermore, we examined the factors that affect the band gap energies of perovskites. The Goldschmidt tolerance factor and octahedral factor, which range from 0.98 to 1 and from 0.45 to 0.7, respectively, are used to filter only highly cubic perovskites that are stable at room temperature. After removing rare or radioactively unstable elements, quantum mechanical density functional theory calculations are performed on the remaining perovskites to assess whether their electronic properties such as band structure are suitable for solar cell applications. Similar calculations are performed on the Ruddlesden-Popper phase. Furthermore, machine learning was utilized to assess the significance of input parameters affecting the band gap of the perovskites.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Density Functional Theory – Machine Learning Approach to Analyze the Bandgap of Elemental Halide Perovskites and Ruddlesden‐Popper Phases

et al. 2018

View full text Add to dashboard Cite

show abstract

“…To accelerate the design and realization of novel materials, a number of recent studies have screened promising candidates across a variety of categories, including light-emitting molecules, 1 perovskite compounds, [2][3][4][5] catalysts, 6,7 thermoelectrics, [8][9][10][11][12] and metal-organic frameworks. 13,14 Accordingly, the rise of virtual materials screening, along with high-throughput first-principles computations and experimentation, has resulted in the creation of numerous accessible databases for the materials science community.…”

Section: Introductionmentioning

confidence: 99%

“…We perform synthesis screening on SrTiO 3 and BaTiO 3 syntheses, since these materials systems have only hundreds of text-mining-accessible published syntheses, and thus provide an environment for examining the advantages of data volume augmentation. We also visually explore two-dimensional learned VAE latent vector spaces to investigate potential driving factors for brookite TiO 2 formation and to understand ion intercalation effects in MnO 2 phase selection.…”

Section: Introductionmentioning

confidence: 99%

Virtual screening of inorganic materials synthesis parameters with deep learning

Kim

Huang

Jegelka³

et al. 2017

npj Comput Mater

162

136

View full text Add to dashboard Cite

Virtual materials screening approaches have proliferated in the past decade, driven by rapid advances in first-principles computational techniques, and machine-learning algorithms. By comparison, computationally driven materials synthesis screening is still in its infancy, and is mired by the challenges of data sparsity and data scarcity: Synthesis routes exist in a sparse, highdimensional parameter space that is difficult to optimize over directly, and, for some materials of interest, only scarce volumes of literature-reported syntheses are available. In this article, we present a framework for suggesting quantitative synthesis parameters and potential driving factors for synthesis outcomes. We use a variational autoencoder to compress sparse synthesis representations into a lower dimensional space, which is found to improve the performance of machine-learning tasks. To realize this screening framework even in cases where there are few literature data, we devise a novel data augmentation methodology that incorporates literature synthesis data from related materials systems. We apply this variational autoencoder framework to generate potential SrTiO 3 synthesis parameter sets, propose driving factors for brookite TiO 2 formation, and identify correlations between alkali-ion intercalation and MnO 2 polymorph selection.

show abstract

Thermodynamic Stability Landscape of Halide Double Perovskites via High‐Throughput Computing and Machine Learning

Sun

et al. 2019

Adv Funct Materials

147

134

View full text Add to dashboard Cite

Formability and stability issues are of core importance and difficulty in current research and applications of perovskites. Nevertheless, over the past century, determination of the formability and stability of perovskites has relied on semi empirical models derived from physics intuition, such as the commonly used Goldschmidt tolerance factor, t. Here, through high-throughput density functional theory (DFT) calculations, a database containing the decomposition energies, considered to be closely related to the thermodynamic stability of 354 halide perovskite candidates, is established. To map the underlying relationship between the structure and chemistry features and the decomposition energies, a well-functioned machine learning (ML) model is trained over this theory-based database and further validated by experimental observations of perovskite formability (F 1 score, 95.9%) of 246 A 2 B(I)B(III)X 6 compounds that are not present in the training database; the model performs a lot better than empirical descriptors such as tolerance factor t (F 1 score, 77.5%). This work demonstrates that the experimental engineering of stable perovskites by ML could solely rely on training data derived from high-throughput DFT computing, which is much more economical and efficient than experimental attempts at materials synthesis. demonstrate enhanced stability in comparison to their single perovskite counterparts. For example, although the photoactive perovskite phases of RbPbI 3 , CsPbI 3 , and FAPbI 3 are unstable at room temperature, the proper mixing of (Rb, Cs, and FA) on the A site can result in a stable mixed-A-site perovskite. [21,23] Currently, for the precise experimental control of stability, more mixing elements, such as triple-A (Cs, MA, FA)InBiBr 6 , [26] triple-A double-X Cs x (MA 0.17 FA 0.83 ) (1-x) Pb(I 0.83 Br 0.17 ) 3 , [20] and quadruple-A double-X Rb-FA 0.75 MA 0.15 Cs 0.1 PbI 2 Br, [27] are used, making the problem more complicated. Further insight is required to understand the effect of elemental mixing and provide guidance for stability engineering, especially relating to the type of mixing elements and their concentrations.

show abstract

Finding New Perovskite Halides via Machine Learning

Cited by 159 publications

References 36 publications

Density Functional Theory – Machine Learning Approach to Analyze the Bandgap of Elemental Halide Perovskites and Ruddlesden‐Popper Phases

Density Functional Theory – Machine Learning Approach to Analyze the Bandgap of Elemental Halide Perovskites and Ruddlesden‐Popper Phases

Virtual screening of inorganic materials synthesis parameters with deep learning

Thermodynamic Stability Landscape of Halide Double Perovskites via High‐Throughput Computing and Machine Learning

Contact Info

Product

Resources

About