Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 49 of the 49 references it cites were found in a registry or an index.

Machine-learning structural and electronic properties of metal halide perovskites using a hierarchical convolutional neural network

Open access
Saidi, Wissam A., Shadid, Waseem, Castelli, Ivano E. · 2020 · npj Computational Materials

Abstract

The development of statistical tools based on machine learning (ML) and deep networks is actively sought for materials design problems. While structure-property relationships can be accurately determined using quantum mechanical methods, these first-principles calculations are computationally demanding, limiting their use in screening a large set of candidate structures. Herein, we use convolutional neural networks to develop a predictive model for the electronic properties of metal halide perovskites (MHPs) that have a billions-range materials design space. We show that a well-designed hierarchical ML approach has a higher fidelity in predicting properties of the MHPs compared to straight-forward methods. In this architecture, each neural network element has a designated role in the estimation process from predicting complex features of the perovskites such as lattice constant and octahedral till angle to narrowing down possible ranges for the values of interest. Using the hierarchical ML scheme, the obtained root-mean-square errors for the lattice constants, octahedral angle and bandgap for the MHPs are 0.01 Å, 5°, and 0.02 eV, respectively. Our study underscores the importance of a careful network design and a hierarchical approach to alleviate issues associated with imbalanced dataset distributions, which is invariably common in materials datasets.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41524-020-0307-8

Cites

49 of 49 references verified0 citation errors · 0 not checkable

Statistics

Figures

Image 1
CC-BY · From the PDF
Fig. 1 Schematic representation of the selected material design space for the MHPs. The A site of ABX 1 X 2 X 3 can be occupied by Cs in addition to 18 different organic molecules: ammonium [NH4] + , [AsH4] + , [PH4] + , [AsH4] + , [PF4] + , methylammonium [CH3NH3] + , [CH3PH3] + , [CH3AsH3] + , hydrazinium [(H3N)(NH2)] + , azetidinium [(CH2)3NH2] + , formamidinium [NH2(CH)NH2] + , [NH2(CH)PH2] + , [NH2(CH)AsH2] + , imidazolium [C3N2H5] + , dimethylammonium [(CH3)2NH2] + , acetamidinium [NH2-C-CH3-NH2] + , ethylammonium [(C2H5)NH3] + , and hydroxylammonium [H3NOH] + . The B site can be occupied by Sn or Pb, while as the X 1 , X 2 , and X 3 halogen can be Cl, Br or I.
Fig. 1 Schematic representation of the selected material design space for the MHPs. The A site of ABX 1 X 2 X 3 can be occupied by Cs in addition to 18 different organic molecules: ammonium [NH4] + , [AsH4] + , [PH4] + , [AsH4] + , [PF4] + , methylammonium [CH3NH3] + , [CH3PH3] + , [CH3AsH3] + , hydrazinium [(H3N)(NH2)] + , azetidinium [(CH2)3NH2] + , formamidinium [NH2(CH)NH2] + , [NH2(CH)PH2] + , [NH2(CH)AsH2] + , imidazolium [C3N2H5] + , dimethylammonium [(CH3)2NH2] + , acetamidinium [NH2-C-CH3-NH2] + , ethylammonium [(C2H5)NH3] + , and hydroxylammonium [H3NOH] + . The B site can be occupied by Sn or Pb, while as the X 1 , X 2 , and X 3 halogen can be Cl, Br or I.CC-BY · From the PDF
Image 3
CC-BY · From the PDF
Fig. 3 Schematic representation of the convolutional neural network architecture. For the classi fi er network the element-wise recti fi ed linear activation unit (relU) is replaced by an element-wise sigmoid unit.
Fig. 3 Schematic representation of the convolutional neural network architecture. For the classi fi er network the element-wise recti fi ed linear activation unit (relU) is replaced by an element-wise sigmoid unit.CC-BY · From the PDF
Fig. 4 Data distribution. Histogram analysis for ( a ) lattice constant, ( b ) octahedral angle, and ( c ) bandgap of the perovskites.
Fig. 4 Data distribution. Histogram analysis for ( a ) lattice constant, ( b ) octahedral angle, and ( c ) bandgap of the perovskites.CC-BY · From the PDF
Fig. 5 HCNN architecture. Schematic representation of the convolutional neural network employed for ( a ) bandgap and ( b ) structural features of the perovskites. Type 1, 2, and 3 correspond respectively to elemental, precursor-derived and complex type features. In ( a ) and ( b ) we show two design schemes by either using a direct regressor network, or a combined hierarchical network that employs a classi fi er for the range followed by a regressor network for the output value. The dotted square in ( a ) indicates that Type 3 structural features can be obtained using the network design shown in ( b ).
Fig. 5 HCNN architecture. Schematic representation of the convolutional neural network employed for ( a ) bandgap and ( b ) structural features of the perovskites. Type 1, 2, and 3 correspond respectively to elemental, precursor-derived and complex type features. In ( a ) and ( b ) we show two design schemes by either using a direct regressor network, or a combined hierarchical network that employs a classi fi er for the range followed by a regressor network for the output value. The dotted square in ( a ) indicates that Type 3 structural features can be obtained using the network design shown in ( b ).CC-BY · From the PDF
Fig. 6 HCNN performance. Representative plots for the prediction of the CNNs with 80% of the data for training and 20% for testing for ( a ) lattice constant, ( b ) octahedral angle and ( c ) bandgap.
Fig. 6 HCNN performance. Representative plots for the prediction of the CNNs with 80% of the data for training and 20% for testing for ( a ) lattice constant, ( b ) octahedral angle and ( c ) bandgap.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 49 references

Cited by

See all 170

Metadata

Authors
Saidi, W. A., Shadid, W., & Castelli, I. E.
Affiliations
  • University of Pittsburgh (US)
  • University of North Carolina at Charlotte (US)
  • Technical University of Denmark (DK)
Corresponding author
Saidi, Wissam A.
Venue
npj Computational Materials
Published
2020-04-14
Volume
6
Issue
1
ISSN
2057-3960 (ISSN-L)
OpenAlex
W3016987575
Cited by
164
References
49
Citations by year
2026: 19 · 2025: 27 · 2024: 26 · 2023: 30 · 2022: 31 · 2021: 23 · 2020: 8
FWCI
8.95
Type
journal article
Language
en
Open access
Licence
Locations
  • npj Computational Materials (publishedVersion, cc-by)
  • Technical University of Denmark, DTU Orbit (Technical University of Denmark, DTU) (submittedVersion, other-oa)
Funders
  • National Science Foundation (1809085, DE-AC02–06CH11357)
  • U.S. Department of Energy (DE‐ AC02‐06CH11357, AC02‐06CH11357, 06CH11357, DE-AC02-)
  • University of Pittsburgh
  • Danmarks Tekniske Universitet
  • Office of Science (AC02‐06CH11357, DE-AC02, 06CH11357, DE-AC02-06CH11357)
  • Center for Research Computing, University of Pittsburgh
  • Division of Materials Research (DE-AC02-06CH11357, DMR-1809085, 1809085)
  • Argonne National Laboratory (DE- AC02-06-CH11357, DE-AC02, AC02–06CH11357, Contract DE-AC02-06CH11357, 06CH11357)
Topics
  • Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Perovskite Materials and Applications
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Advanced Chemical Physics Studies
Keywords
halide perovskites, convolutional neural networks, hierarchical neural networks, band gap prediction, structure-property relationships, materials design, class imbalance, network architecture design

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Machine-learning structural and electronic properties of metal halide perovskites using a hierarchical convolutional neural network. https://app.cochat.ai/papers/10.1038/s41524-020-0307-8
The paper
Saidi, W. A., Shadid, W., & Castelli, I. E. (2020). Machine-learning structural and electronic properties of metal halide perovskites using a hierarchical convolutional neural network. npj Computational Materials. https://doi.org/10.1038/s41524-020-0307-8

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.