Screening of Metal Catalysts for CO2 Conversion via Machine Learning and Molecular Simulations

Document Type : Original Article

Authors

1 Department of Physical Science, Jaramogi Oginga Odinga University of Science and Technology, P.O. Box 210

2 Alupe University P.O. Box 845 Busia- Kenya 50400

Abstract

This study's primary objective is to improve catalyst discovery by assessing earth-abundant metal catalysts for the conversion of CO2 to methane through the use of machine learning (ML) and molecular dynamics (MD) simulations. The highest CO2 binding energy on 61 metals was determined to be -9.75 eV for nickel (Ni), -8.7 eV for copper (Cu), and -7.75 eV for carbon (C). Various ML models were developed to predict binding energies on the metallic surfaces. Easily accessible properties of the metals and features obtained from molecular simulations were used as input features. RANSACRegressor, LinearSVR, HuberRegressor, OrthogonalMatchingPursuit CV, and LarsCV models exhibited high prediction accuracy with R-squared values of 0.99 and RMSE ranging from 0.18 to 0.40. Feature significance analysis revealed that density (D) is among the most significant structural features affecting binding energy. This work offers a dependable, high-throughput method for identifying efficient CO2 conversion catalysts, advancing sustainable technologies.

Keywords

Main Subjects


[1]    Hidalgo, D. and Martín-Marroquín, J.M., 2020. Power-to-methane, coupling CO2 capture with fuel production: An overview. Renewable and Sustainable Energy Reviews132, p.110057.
[2]    Bianchini, M., Wang, J., Clément, R.J., Ouyang, B., Xiao, P., Kitchaev, D., Shi, T., Zhang, Y., Wang, Y., Kim, H. and Zhang, M., 2020. The interplay between thermodynamics and kinetics in the solid-state synthesis of layered oxides. Nature materials19(10), pp.1088-1095.
[3]    Zhang, Q., Gao, S. and Yu, J., 2022. Metal sites in zeolites: synthesis, characterization, and catalysis. Chemical reviews123(9), pp.6039-6106.
[4]    Bathena, T., Phung, T., Murugesan, V., Goulas, K.A., Karakoti, A.S. and Ramasamy, K., 2024. Transition metal oxides in CO2 driven oxidative dehydrogenation: Uncovering their redox properties. Journal of CO2 Utilization84, p.102848.
[5]    Fu, N., Liang, X., Li, Z. and Li, Y., 2022. Single-atom site catalysts based on high specific surface area supports. Physical Chemistry Chemical Physics24(29), pp.17417-17438.
[6]    Tang, Q., Ma, Y. and Wang, J., 2021. The active sites engineering of catalysts for CO2 activation and conversion. Solar RRL5(2), p.2000443.
[7]    Wang, J., Dou, S. and Wang, X., 2021. Structural tuning of heterogeneous molecular catalysts for electrochemical energy conversion. Science Advances7(13), p. eabf3989.
[8]    Lou, B., Shakoor, N., Adeel, M., Zhang, P., Huang, L., Zhao, Y., Zhao, W., Jiang, Y. and Rui, Y., 2022. Catalytic oxidation of volatile organic compounds by non-noble metal catalyst: Current advancement and future prospectives. Journal of Cleaner Production363, p.132523.
[9]    Rittiruam, M., Khamloet, P., Ektarawong, A., Atthapak, C., Saelee, T., Khajondetchairit, P., Alling, B., Praserthdam, S. and Praserthdam, P., 2024. Screening of Cu-Mn-Ni-Zn high-entropy alloy catalysts for CO2 reduction reaction by machine-learning-accelerated density functional theory. Applied Surface Science652, p.159297.
[10]  Gao, J., Huang, Q., Wu, Y., Lan, Y.Q. and Chen, B., 2021. Metal–organic frameworks for photo/electrocatalysis. Advanced Energy and Sustainability Research2(8), p.2100033.
[11]  Huang, Y. and Xin, H., 2021. Ab initio machine learning for accelerating catalytic materials discovery.
[12]  Wu, G., Zhou, H., Zhang, J., Tian, Z.Y., Liu, X., Wang, S., Coley, C.W. and Lu, H., 2023. A high-throughput platform for efficient exploration of functional polypeptide chemical space. Nature Synthesis2(6), pp.515-526.
[13]  Abraham, B.M., Jyothirmai, M.V., Sinha, P., Viñes, F., Singh, J.K. and Illas, F., 2024. Catalysis in the digital age: Unlocking the power of data with machine learning. Wiley Interdisciplinary Reviews: Computational Molecular Science14(5), p.e1730.
[14]  Chen, Y.Y., Kunz, M.R., He, X. and Fushimi, R., 2022. Recent progress toward catalyst properties, performance, and prediction with data-driven methods. Current Opinion in Chemical Engineering37, p.100843.
[15]  Motagamwala, A.H. and Dumesic, J.A., 2020. Microkinetic modeling: a tool for rational catalyst design. Chemical Reviews121(2), pp.1049-1076.
[16]  Pavese, N., Tai, Y.F., Yousif, N., Nandi, D. and Bain, P.G., 2020. Traditional trial and error versus neuroanatomic 3-dimensional image software-assisted deep brain stimulation programming in patients with Parkinson disease. World Neurosurgery134, pp. e98-e102.
[17]  Allen, K.R., Smith, K.A. and Tenenbaum, J.B., 2020. Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning. Proceedings of the National Academy of Sciences117(47), pp.29302-29310.
[18]  Choung, S., Park, W., Moon, J. and Han, J.W., 2024. Rise of machine learning potentials in heterogeneous catalysis: Developments, applications, and prospects. Chemical Engineering Journal, p.152757.
[19]  Lantz, B., 2019. Machine learning with R: expert techniques for predictive modeling. Packt publishing ltd.
[20]  Harvey, J.N., Himo, F., Maseras, F. and Perrin, L., 2019. Scope and challenge of computational methods for studying mechanism and reactivity in homogeneous catalysis. Acs Catalysis9(8), pp.6803-6813.
[21]  Shankar, U., Gogoi, R., Sethi, S.K. and Verma, A., 2022. Introduction to materials studio software for the atomistic-scale simulations. In Forcefields for atomistic-scale simulations: materials and applications (pp. 299-313). Singapore: Springer Nature Singapore.
[22]  Sun, H., Ren, P. and Fried, J.R., 1998. The COMPASS force field: parameterization and validation for phosphazenes. Computational and Theoretical Polymer Science8(1-2), pp.229-246.
[23]  Sharma, S., Kumar, P. and Chandra, R., 2019. Applications of BIOVIA materials studio, LAMMPS, and GROMACS in various fields of science and engineering. Molecular dynamics simulation of nanocomposites using BIOVIA materials studio, Lammps and Gromacs, pp.329-341.
[24]  Barrionuevo, G.O., Rios, S., Williams, S.W. and Ramos-Grez, J.A., 2021, May. Comparative evaluation of machine learning regressors for the layer geometry prediction in wire arc additive manufacturing. In 2021 IEEE 12th International Conference on Mechanical and Intelligent Manufacturing Technologies (ICMIMT) (pp. 186-190). IEEE.
[25]  Yang, F., Deng, D., Pan, X., Fu, Q. and Bao, X., 2015. Understanding nano effects in catalysis. National Science Review2(2), pp.183-201.
[26]  Irfan, A. and Mahmood, A., 2018. Designing of efficient acceptors for organic solar cells: molecular modelling at DFT level. Journal of Cluster Science29, pp.359-365.
[27]  Tahir, M.H., Mubashir, T., Shah, T.U.H. and Mahmood, A., 2019. Impact of electron‐withdrawing and electron‐donating substituents on the electrochemical and charge transport properties of indacenodithiophene‐based small molecule acceptors for organic solar cells. Journal of Physical Organic Chemistry32(3), p.e3909.
[28]  Kluger, F., Brachmann, E., Ackermann, H., Rother, C., Yang, M.Y. and Rosenhahn, B., 2020. Consac: Robust multi-model fitting by conditional sample consensus. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 4634-4643).
[29]  Fotouhi, M., Hekmatian, H., Kashani-Nezhad, M.A. and Kasaei, S., 2019. SC-RANSAC: spatial consistency on RANSAC. Multimedia Tools and Applications78, pp.9429-9461.
[30]  Mustaqim, A.Z., Adi, S., Pristyanto, Y. and Astuti, Y., 2021, June. The effect of recursive feature elimination with cross-validation (RFECV) feature selection algorithm toward classifier performance on credit card fraud detection. In 2021 International conference on artificial intelligence and computer science technology (ICAICST) (pp. 270-275). IEEE.