Harnessing Quantum Chemical Bonding Analysis Descriptors for Material Property Predictions

Examining the bonding between their constituent atoms in crystalline materials has played a vital role in understanding material properties.^[1–4] For instance, low thermal conductivity in materials is typically attributed to its anharmonicity, which has been reported to arise from strong antibonding interactions and local environment distortions.^[5–7] The bonds in the material are often quantified in terms of bond strength and can be extracted from crystalline materials using density-based^[8], energy-based^[9], and orbital-based^[10] methods. LOBSTER^[11] is a program that relies on an orbital-based method to extract such bonding information by projecting the plane wave-based wave functions of modern density functional theory computations (DFT) onto a local atomic orbital basis. Since our goal was to use bonding analysis descriptors for material property predictions, we needed to first systematically generate large quantities of bonding analysis data. To streamline this process, we have developed a user-friendly workflow^[12], which is now also part of the atomate2^[13] package that can generate bonding information data extracted using the LOBSTER program for crystalline materials. This workflow requires only the structure as input from the user. Employing this workflow, we have generated for ~13000 crystalline compounds such bonding analysis data. To create new descriptors from these data, we use our package LobsterPy.^[14] The curated descriptors span different types, including statistical representations of bonding characteristics for traditional ML algorithms (e.g., random forests), textual descriptions for large language models (LLMs), and structure graphs for graph neural networks (GNNs). These descriptors are then tested by employing them in several state-of-the-art ML algorithms and architectures to predict the mechanical, vibrational, and thermal properties of crystalline materials. Through this work, we are not only able to demonstrate how one can enhance the model’s predictive accuracy^[15] by incorporating quantum chemical bonding-based descriptors alongside typical composition and structure-based descriptors, but it also aids in uncovering relationships between bonding and materials properties on a larger scale, which was not possible before.

References:

[1] R. Hoffmann, Angew. Chem. Int. Ed. Engl. 1987, 26, 846–878.
[2] R. Nelson, C. Ertural, P. C. Müller, R. Dronskowski, in Comprehensive Inorganic Chemistry III, Elsevier, 2023, pp. 141–201.
[3] C. Persch, M. J. Müller, A. Yadav, J. Pries, N. Honné, P. Kerres, S. Wei, H. Tanaka, P. Fantini, E. Varesi, F. Pellizzer, M. Wuttig, Nat Commun 2021, 12, 4978.
[4] C.-F. Schön, S. van Bergerem, C. Mattes, A. Yadav, M. Grohe, L. Kobbelt, M. Wuttig, Sci. Adv. 2022, 8, eade0828.
[5] R. Pathak, S. Paul, S. Das, A. Das, S. K. Pati, K. Biswas, Angewandte Chemie n.d., n/a, e202408908.
[6] M. Dutta, K. Pal, U. V. Waghmare, K. Biswas, Chem. Sci. 2019, 10, 4905–4913.
[7] A. Das, K. Pal, P. Acharyya, S. Das, K. Maji, K. Biswas, J. Am. Chem. Soc. 2023, jacs.2c11908.
[8] R. F. W. Bader, T. T. Nguyen-Dang, in Advances in Quantum Chemistry (Ed.: P.-O. Löwdin), Academic Press, 1981, pp. 63–124.
[9] M. Raupach, R. Tonner, The Journal of Chemical Physics 2015, 142, 194105.
[10] R. S. Mulliken, The Journal of Chemical Physics 1955, 23, 1833–1840.
[11] R. Nelson, C. Ertural, J. George, V. L. Deringer, G. Hautier, R. Dronskowski, J Comput Chem 2020, 41, 1931–1940.
[12] J. George, G. Petretto, A. Naik, M. Esters, A. J. Jackson, R. Nelson, R. Dronskowski, G. Rignanese, G. Hautier, ChemPlusChem 2022, 87, DOI 10.1002/cplu.202200123.
[13] A. Ganose, et al., 2024, DOI 10.5281/zenodo.10677081.
[14] A. A. Naik, K. Ueltzen, C. Ertural, A. J. Jackson, J. George, Journal of Open Source Software 2024, 9, 6286.
[15] A. A. Naik, C. Ertural, N. Dhamrait, P. Benner, J. George, Sci Data 2023, 10, 610.