Journal of Soft Computing and Information Technology

Journal of Soft Computing and Information Technology

Investigating Thermal Conductivity in Carbon-Based Nanomaterials Using Molecular Dynamics (MD) and Graph Neural Network (GNN)

Document Type : Persian Original Article

Author
Hoda Ghavaminia
Department of Physics, Materials and Energy Research Center, Dez.C., Islamic Azad University, Dezful, Iran
10.22034/jscit.2026.539661.2154
Abstract
In this study, an accurate and efficient framework is developed to predict the thermal conductivity(κ) of carbon-based nanomaterials, including carbon nanotubes (CNTs) and graphene, by integrating Molecular Dynamics (MD) simulations with a Graph Neural Network (GNN). Training data are generated from MD simulations using the LAMMPS software and the AIREBO potential under various conditions, including defects, strain, temperature, and structural size. A Message Passing Neural Network (MPNN) model is then trained to predict κ with high accuracy (RMSE ≈ 0.18 W/mK) in less than 0.5 seconds. Results demonstrate that the GNN model accurately reproduces κ values compared to experimental and simulation data. Physical analysis reveals that atomic defects, increasing temperature, and reduced structural size decrease κ by up to 48%, 35%, and 36%, respectively. Furthermore, applying tensile strain up to 10% reduces κ by 39%, attributed to enhanced phonon scattering and reduced phonon group velocity. Interpretability analysis using Grad-CAM identifies atoms near defects, boundaries, and strained regions as the most influential in reducing κ, consistent with phonon scattering theory. This indicates that the GNN model not only predicts but also interprets the underlying physical mechanisms. The proposed hybrid framework offers a powerful tool for designing intelligent thermal materials, strain-based sensors, and nanoelectronic cooling systems.
Keywords
Carbon nanomaterials
Thermal conductivity
Molecular Dynamics
Graph Neural Network
[1] P. Kim, L. Shi, A. Majumdar, and P. L. McEuen, "Thermal transport measurements of individual multiwalled nanotubes," Phys. Rev. Lett.87 215502, 2001. doi.org/10.1103/PhysRevLett.87.215502
[2]-W. Jiang, J.-S. Wang, and B. Li, " Thermal expansion in single-walled carbon nanotubes and graphene: Nonequilibrium Green’s function approach ," Physical Review B, vol. 80, no. 21, p. 205429, 2009.
doi.org/10.1103/PhysRevB.80.205429
[3] S. Plimpton, "Fast parallel algorithms for short-range molecular dynamics," Journal of Computational Physics, vol. 117, no. 1, pp. 1–19, 1995. doi.org/10.1006/jcph.1995.1039
[4] N. Artrith and B. Kolb, "Understanding the Composition and Activity of Electrocatalytic Nanoalloys in Aqueous Solvents: A Combination of DFT and Accurate Neural Network Potentials, vol. 14, no. 5, pp. 2670-2676, 2013.
[5] J. Gilmer, S. S. Schoenholz, P. F. Riley, O. Vinyals, and G. E. Dahl, "Neural message passing for quantum chemistry," in Proceedings of the 34th International Conference on Machine Learning, vol. 70, pp. 1263–1272, 2017. Doi:10.48550/arXiv.1704.01212
[6] H. Ghavaminia, "Predicting dielectric constants of nanomaterials using graph neural networks," Journal of Soft Computing and Information Technology, Vol. 14, no.3, pp. 13-20, 2025 doi:10.22034/jscit.2025.512167.2119
[7] H. Ghavaminia, Predicting Hydrogen Uptake in Carbon Nanotubes Using Graph Neural Networks and Molecular Simulations. Journal of Chemical Reactivity and Synthesis: vol 15 no. pp. 50-68 2025. doi:10.82437/jcrs.2025.1221400
[8] N. Mingo and D. A. Broido, "Length Dependence of Carbon Nanotube Thermal Conductivity and the “Problem of Long Waves," Nano Letters, vol. 5, no. 7, pp. 1221–1225, 2005. doi/10.1021/nl050714d
[9] L. Lindsay, D. A. Broido, and N. Mingo, "Flexural phonons and thermal transport in graphene," Physical Review B, vol. 82, no. 11, p. 115427, 2010. doi.org/10.1103/PhysRevB.82.115427
[10] S. Klein, Z. Aksamija, and M. F. B., " Electrothermal properties of 2D materials in device applications,J. of Compt. Electr., Vol. 22  5
PP. 1310 – 1318,doi.org/10.1007/s10825-023-02091-z.
[11] Z. Guo, D. Zhang, and X. G. Gong, "Thermal conductivity of graphene nanoribbons," Applied Physics Letters, vol. 95, no. 16, p. 163103, 2009. doi.org/10.1063/1.3246155.
[1] X. Wei B. Fragneaud. C.A.Marianetti, and Nonlinear elastic behavior of graphene AB initio calculations to continuum description.Phys. Rev. B 80, 205407 Published 10 November, 2009.doi: 10.1103/PhysRevB.80.205407
[13] J. Hone, M. C. Llaguno, M. J. Biercuk, A. T. Johnson, Z. Batlogg, B. H. Zettl, and J. E. Fischer, "Thermal properties of carbon nanotubes and nanotube-based materials," Applied Physics A, vol. 74, no. 3, pp. 339–343, 2002. doi.org/10.1063/1.1469696
[14] P. Keblinski, J. M. Holt, D. R. Kuhlmann, and D. G. Cahill, "Mechanisms of heat flow in suspensions of nano-sized particles (nanofluids)," International Journal of Heat and Mass Transfer, vol. 45, no. 4, pp. 855-863.2002. https://doi.org/10.1016/S0017-9310(01)00175-2
[15] A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, and C. N. Lau, "Superior thermal conductivity of single-layer graphene," Nano Letters, vol. 8, no. 3, pp. 902–907, 2008. https://doi.org/10.48550/arXiv.0802.1367
[16] Y. Huang, J. Wu, and K. C. Hwang, "Thickness of graphene and single-wall carbon nanotubes," Phys. Rev.B Vol.74, PP.245413, 2006. doi.org/10.1103/PhysRevB.74.245413
[17] X. Xu, L. Wang, Q. Zhang, Q. Zhao, J. Jia, X. Zhai, and K. Chang, "High thermal conductivity of suspended exfoliated graphene," Nature Materials, vol. 13, no. 12, pp. 1117–1121, 2014.
[18] D. P. Kingma and J. L. Ba, "Adam: A method for stochastic optimization," in Proceedings of the 3rd International Conference on Learning Representations, 2015. https://doi.org/10.48550/arXiv.1412.6980
Journal of Soft Computing and Information Technology
Volume 15, Issue 1
Spring - April
Spring 2026
Pages 29-37