journal article·2024-08Open access

Interpretable machine learning boosting the discovery of targeted organometallic compounds with optimal bandgap

Taehyun Park, JunHo Song, Jinyoung Jeong, Seungpyo Kang, Joonchul Kim, Joonghee Won, Jungim Han, Kyoungmin Min

Abstract

Organometallic compounds (OMCs) have attracted tremendous attention in various fields, such as photovoltaic cell and high-k dielectric application, due to their beneficial properties. Despite their potential, the progression of OMCs into industrial applications is hindered by the limited databases available for their properties and the absence of efficient surrogate models. To address this, in this study, optimally selected feature-based surrogate models for predicting the electronic properties of OMCs are constructed via various multiscale features and extensive database. To this end, high-throughput calculation was performed to obtain electronic properties of more than 18k materials generally known as organometallics, augmenting around 12k organic materials obtained from the public open data set, OMDB-GAP1. For generating features closely related to OMCs, descriptors encapsulating the information ranging local to global, also other widely-used composition-, structure-based features (more than 3.5k in total) were employed. Among these descriptors, we identified 48 critical features that elucidates the physicochemical underpinnings of OMCs’ properties, suggesting their impact on the properties of OMCs. The light gradient boosting machine model achieved high-accuracy predictions across the entire database with just 1 % of the total descriptors, sufficiently compared to the entire sets (decreased of around 0.01 by R2 score and 0.01 eV by MAE). Furthermore, the efficacy of active learning process was demonstrated to find OMCs with optimal properties rapidly. As a result, expected improvement outperforms other methods by identifying 69 % of the target materials only searching 46 % of the total search space. Our constructed platform with a high-throughput calculated database can pave the way for the rapid screening of OMCs for the targeted industrial application, and suggest a comprehensive grasp of the intrinsic properties of OMCs and related compounds.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Computational Drug Discovery Methods · X-ray Diffraction in Crystallography

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (100+)

Show source-provided references
1. Organometallic compounds in drug discovery: past, present and future10.1016/j.ddtec.2019.06.001 ↗
2. Selected organometallic compounds for third order nonlinear optical application10.3390/nano9020254 ↗
3. Organometallic-based hybrid perovskite piezoelectrics with a narrow band gap10.1021/jacs.0c09288 ↗
4. Multiferroic behavior associated with an order-disorder hydrogen bonding transition in metal-organic frameworks (MOFs) with the perovskite ABX 3 architecture10.1021/ja904156s ↗
5. Why is ferrocene so exceptional?10.1002/ejic.201600983 ↗
6. Exploring organo-palladium(II) complexes as novel organometallic materials for Li-ion batteries10.1016/j.electacta.2020.135659 ↗
7. Development and future prospects of selective organometallic compounds as anticancer drug candidates exhibiting novel modes of action10.1016/j.ejmech.2019.04.062 ↗
8. Romero · 2023
9. Recent advances of organometallic complexes for rechargeable batteries10.1016/j.ccr.2020.213650 ↗
10. Organic and organometallic fluorinated materials for electronics and optoelectronics: a survey on recent research10.1002/ejoc.201800657 ↗
11. Functional organic materials for optoelectronic applications10.1039/C9TC90084E ↗
12. Emerging role of organometallic compounds for drug delivery, renewable energy, and wastewater treatment10.1002/9783527840946.ch17 ↗
13. A biologically inspired organometallic fuel cell (OMFC) that converts renewable alcohols into energy and chemicals10.1002/anie.201002234 ↗
14. Research progress of high dielectric constant zirconia-based materials for gate dielectric application10.3390/coatings10070698 ↗
15. Flexible Al-Ti-Zn-O MIM capacitors fabricated by room temperature atomic layer deposition and their electrical performances10.1016/j.jallcom.2021.159391 ↗
16. Organometallic precursors for atomic layer deposition10.1007/b136145 ↗
17. Tuning an electrode work function using organometallic complexes in inverted perovskite solar cells10.1021/jacs.1c02118 ↗
18. Recent advances of organometallic complexes in emerging photovoltaics10.1002/pol.20210592 ↗
19. Organometallic-functionalized interfaces for highly efficient inverted perovskite solar cells10.1126/science.abm8566 ↗
20. A global inventory of photovoltaic solar energy generating units10.1038/s41586-021-03957-7 ↗
21. Single-junction organic photovoltaic cell with 19% efficiency10.1002/adma.202102420 ↗
22. Recent progress in chlorinated organic photovoltaic materials10.1021/acs.accounts.0c00009 ↗
23. Detailed balance limit of efficiency of P-n junction solar cells10.1063/1.1736034 ↗
24. Tabulated values of the shockley-queisser limit for single junction solar cells10.1016/j.solener.2016.02.015 ↗
25. Shockley-queisser detailed balance efficiency limit for nanowire solar cells10.1021/ph5004835 ↗
26. Investigation on structures, band gaps, and electronic structures of lead free La2NiMnO6 double perovskite materials for potential application of solar cell10.1016/j.jallcom.2015.09.187 ↗
27. Strain regulating mechanical stability and photoelectric properties of CH3NH3PbI3 containing the asymmetric CH3NH3 cations
28. Potassium-containing triple-cation mixed-halide perovskite materials: toward efficient and stable solar cells10.1016/j.jallcom.2020.158335 ↗
29. A comprehensive guide to bifacial perovskite solar cells: simulation and optimization
30. Are perovskite solar cells reaching the efficiency and voltage limits?10.1021/acsenergylett.3c01649 ↗
31. Bandgap prediction on small thermoelectric material dataset via instance-based transfer learning
32. Machine learning aided design of polymer with targeted band gap based on DFT computation10.1021/acs.jpcb.0c08674 ↗
33. Small data machine learning in materials science10.1038/s41524-023-01000-z ↗
34. Interpretable and explainable machine learning for materials science and chemistry10.1021/accountsmr.1c00244 ↗
35. A machine learning-assisted approach to a rapid and reliable screening for mechanically stable perovskite-based materials10.1002/adfm.202210374 ↗
36. Evaluation of principal features for predicting bulk and shear modulus of inorganic solids with machine learning
37. Prediction and classification of formation energies of binary compounds by machine learning: an approach without crystal structure information10.1021/acsomega.1c01517 ↗
38. Synthesizable double perovskite oxide search via machine learning and high-throughput computational screening10.1002/adts.202100263 ↗
39. Accurate prediction of band gap of materials using stacking machine learning model
40. Accurate bandgap predictions of solids assisted by machine learning
41. Exploring the large chemical space in search of thermodynamically stable and mechanically robust MXenes via machine learning10.1039/D3CP06337B ↗
42. Data mining and graph network deep learning for band gap prediction in crystalline borate materials10.1021/acs.inorgchem.3c00233 ↗
43. Enhancing property prediction and process optimization in building materials through machine learning: a review
44. Resolving transition metal chemical space: feature selection for machine learning and structure-property relationships10.1021/acs.jpca.7b08750 ↗
45. Recent advances and applications of machine learning in solid-state materials science10.1038/s41524-019-0221-0 ↗
46. A strategy to apply machine learning to small datasets in materials science10.1038/s41524-018-0081-z ↗
47. Application of machine learning to reveal relationship between processing-structure-property for polypropylene injection molding10.1016/j.polymer.2023.125736 ↗
48. Towards understanding structure–property relations in materials with interpretable deep learning10.1038/s41524-023-01163-9 ↗
49. Commentary: the materials Project: a materials genome approach to accelerating materials innovation10.1063/1.4812323 ↗
50. AFLOW: an automatic framework for high-throughput materials discovery10.1016/j.commatsci.2012.02.005 ↗
51. Materials design and discovery with high-throughput density functional theory: the open Quantum materials database (OQMD)
52. The joint automated repository for various integrated simulations (JARVIS) for data-driven materials design10.1038/s41524-020-00440-1 ↗
53. A band-gap database for semiconducting inorganic materials calculated with hybrid functional10.1038/s41597-020-00723-8 ↗
54. Band gap prediction for large organic crystal structures with machine learning
55. Organic materials database: an open-access online database for data mining10.1371/journal.pone.0171501 ↗
56. Machine learning-based model for predicting the material properties of nanostructured aerogels10.1002/pls2.10082 ↗
57. Big data of materials science: critical role of the descriptor10.1103/PhysRevLett.114.105503 ↗
58. Prediction of organic material band gaps using graph attention network
59. Interpretable machine learning for materials design10.1557/s43578-023-01164-w ↗
60. Wide bandgap oxide semiconductors: from materials physics to optoelectronic devices10.1002/adma.202006230 ↗
61. Assessing graph-based deep learning models for predicting flash point10.1002/minf.201900101 ↗
62. Crystallography open database – an open-access collection of crystal structures
63. A workflow for deriving chemical entities from crystallographic data and its application to the Crystallography open database10.1186/s13321-023-00780-2 ↗
64. Generalized gradient approximation made simple10.1103/PhysRevLett.77.3865 ↗
65. Predicting the band gaps of inorganic solids by machine learning10.1021/acs.jpclett.8b00124 ↗
66. Band gap predictions of double perovskite oxides using machine learning10.1038/s43246-023-00373-4 ↗
67. Tuning of band gap by variation of halide ions in K2CuSbX6 (X = Cl, Br, I) for solar cells and thermoelectric applications
68. Approximation to density functional theory for the calculation of band gaps of semiconductors10.1103/PhysRevB.78.125116 ↗
69. Hybrid functionals based on a screened coulomb potential10.1063/1.1564060 ↗
70. Liu · 2024 · 1
71. Accurate band gaps of semiconductors and insulators with a semilocal exchange-correlation potential10.1103/PhysRevLett.102.226401 ↗
72. The open Quantum materials database (OQMD): assessing the accuracy of DFT formation energies10.1038/npjcompumats.2015.10 ↗
73. Improving band gap prediction in density functional theory from molecules to solids10.1103/PhysRevLett.107.026403 ↗
74. Quasiparticle effects in the bulk and surface-state bands of Bi 2Se 3 and Bi 2Te 3 topological insulators10.1103/PhysRevB.85.161101 ↗
75. Universal fragment descriptors for predicting properties of inorganic crystals10.1038/ncomms15679 ↗
76. The inorganic crystal structure database (ICSD) - present and future10.1080/08893110410001664882 ↗
77. Multi-fidelity machine learning models for accurate bandgap predictions of solids10.1016/j.commatsci.2016.12.004 ↗
78. Accurate bandgap predictions of solids assisted by machine learning
79. Bandgap prediction on small thermoelectric material dataset via instance-based transfer learning
80. Predicting band gaps of MOFs on small data by deep transfer learning with data augmentation strategies10.1039/D3RA02142D ↗
81. Accurate bandgap predictions of solids assisted by machine learning
82. An empirical evaluation of the T-SNE algorithm for data visualization in structural engineering
83. Role of length-scale in machine learning based image analysis of ductile fracture surfaces
84. Machine learning for the classification of macroscale fracture surfaces
85. A fault diagnosis method for nuclear power plant rotating machinery based on adaptive deep feature extraction and multiple support vector machines
86. De Breuck, P. P.; Hautier, G.; Rignanese, G. M. Materials Property Prediction for Limited Datasets Enabled by Feature Selection and Joint Learning with MODNet. npj Comput. Mater. 2021 71 2021, 7 (1), 1–8. https://doi.org/10.1038/s41524-021-00552-2.10.1038/s41524-021-00552-2 ↗
87. Matminer: an open source toolkit for materials data mining10.1016/j.commatsci.2018.05.018 ↗
88. Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations10.1103/PhysRevB.96.024104 ↗
89. Local-density-functional calculations of the energy of atoms10.1103/PhysRevA.55.191 ↗
90. A general-purpose machine learning framework for predicting properties of inorganic materials10.1038/npjcompumats.2016.28 ↗
91. Python materials genomics (pymatgen): a robust, open-source Python library for materials analysis10.1016/j.commatsci.2012.10.028 ↗
92. Machine learning with force-field-inspired descriptors for materials: fast screening and mapping energy landscape
93. A transferable machine-learning framework linking interstice distribution and plastic heterogeneity in metallic glasses10.1038/s41467-019-13511-9 ↗
94. Molecular structures: perception, autocorrelation descriptor and sar studies: system of atomic contributions for the calculation of the n-octanol/water partition coefficients
95. Prediction of partition coefficients (LOG P oct) using autocorrelation descriptors10.1080/10629369708039129 ↗
96. Leveraging cheminformatics strategies for inorganic discovery: application to redox potential design10.1021/acs.iecr.7b00808 ↗
97. Low-code AutoML-augmented data pipeline – a review and experiments
98. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties10.1103/PhysRevLett.120.145301 ↗
99. SchNet: a continuous-filter convolutional neural network for modeling Quantum interactions
100. Graph networks as a universal machine learning framework for molecules and crystals10.1021/acs.chemmater.9b01294 ↗