Machine learning approaches for estimating formation enthalpies in energetic materials
Source abstract
Abstract
The accurate prediction of thermodynamic properties of energetic materials is crucial for advancing the development of safer and more efficient explosives, propellants, and pyrotechnics. This research presents a comprehensive machine learning approach to predict the enthalpy of formation of energetic materials using molecular descriptors and functional group information. By analyzing a dataset of 113 distinct energetic compounds, we developed and compared multiple machine learning models, including Decision Trees, Random Forest, AdaBoost, K-Nearest Neighbors, Ensemble Learning, Multilayer Perceptron Artificial Neural Networks, and Convolutional Neural Networks. The CNN model demonstrated superior generalization capabilities with an R2 value of 0.83 for the total dataset and minimal overfitting between training and testing phases. Feature importance analysis using SHAP revealed that azide functional groups have the most significant influence on enthalpy prediction, aligning with established thermochemical principles. This predictive framework offers a reliable computational alternative to expensive experimental methods for screening novel energetic materials.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Materials Science · Chemical Thermodynamics and Molecular Structure · Energetic Materials and Combustion
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Zhuhai City Polytechnic
University College of Duba, University of Tabuk
Marwadi University Research Center, Marwadi University
GLA University
School of Sciences, Jain (Deemed to be University)
Centre for Research Impact & Outcome, Chitkara University Institute of Engineering and Technology, Chitkara University
Sathyabama Institute of Science and Technology
Siksha ‘O’ Anusandhan (Deemed to be University)
National Pedagogical University of Uzbekistan
Herat University
The Islamic University
References (63)
Show source-provided references
1. 10.1007/s40846-017-0360-z10.1007/s40846-017-0360-z ↗
2. Abdul Rahim KS et al. A comprehensive overview and future frontiers of energetic materials comprising nitrocellulose. Siti Nor Ain and Ariff Hafizah and Abdul Latif Nur Shazwani a comprehensive overview and future frontiers of energetic materials comprising nitrocellulose.
3. 10.1134/S001050822005001910.1134/S0010508220050019 ↗
4. 10.1016/j.petrol.2018.11.06710.1016/j.petrol.2018.11.067 ↗
5. 10.1016/j.microc.2010.12.00710.1016/j.microc.2010.12.007 ↗
6. 10.1016/j.dajour.2022.10007110.1016/j.dajour.2022.100071 ↗
7. 10.1016/0885-064X(88)90020-910.1016/0885-064X(88)90020-9 ↗
8. 10.1016/j.jngse.2022.10442510.1016/j.jngse.2022.104425 ↗
9. 10.1134/S001050821306001410.1134/S0010508213060014 ↗
10. 10.1016/j.tafmec.2025.10509210.1016/j.tafmec.2025.105092 ↗
11. Chagas P et al. 2018. Evaluation of convolutional neural network architectures for chart image classification. 2018 International Joint Conference on Neural Networks (IJCNN); Rio de Janeiro, Brazil..
12. 10.1002/prep.20220027610.1002/prep.202200276 ↗
13. 10.1007/s10853-017-0912-110.1007/s10853-017-0912-1 ↗
14. Fayet G, Rotureau P, Adamo C. 2025. Improving safety with molecular-scale computational approaches for energetic and reactive materials. In: Kunal Roy, Arkaprava Banerjee, editors. Materials informatics III: polymers, solvents and energetic materials. Springer; pp 233–264.10.1007/978-3-031-78724-9_9 ↗
15. Feng W et al. 2020. Fsrf: An improved random forest for classification. 2020 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA); Dalian, China.
16. 10.1016/S1352-2310(97)00447-010.1016/S1352-2310(97)00447-0 ↗
17. Gorriz JM et al. 2024. Is k-fold cross validation the best model selection method for machine learning? Arxiv preprint arxiv:2401.16407.
18. 10.1021/acs.langmuir.4c0408310.1021/acs.langmuir.4c04083 ↗
19. 10.1007/s00894-020-4337-410.1007/s00894-020-4337-4 ↗
20. Hernandez AM. 2018. Computational modeling of advanced multi-material energetic materials and systems.
21. 10.1021/acs.iecr.4c0356610.1021/acs.iecr.4c03566 ↗
22. 10.1007/s44371-025-00179-y10.1007/s44371-025-00179-y ↗
23. 10.1038/natrevmats.2015.410.1038/natrevmats.2015.4 ↗
24. 10.1002/prep.20110015610.1002/prep.201100156 ↗
25. Kim P. 2017. Convolutional neural network. In: Matlab deep learning: with machine learning, neural networks and artificial intelligence. Apress; pp 121–147.10.1007/978-1-4842-2845-6_6 ↗
26. 10.1038/nrd458110.1038/nrd4581 ↗
27. 10.1007/978-3-642-38652-7_210.1007/978-3-642-38652-7_2 ↗
28. 10.1109/TNNLS.2021.308482710.1109/TNNLS.2021.3084827 ↗
29. 10.1016/j.jechem.2024.10.03510.1016/j.jechem.2024.10.035 ↗
30. 10.1016/j.fpc.2021.05.00210.1016/j.fpc.2021.05.002 ↗
31. Mader CL. 1963. Detonation properties of condensed explosives computed using the Becker-Kistiakowsky-Wilson equation of state. Los Alamos Scientific Laboratory of the University of California.
32. 10.1016/S0040-6031(01)00795-X10.1016/S0040-6031(01)00795-X ↗
33. 10.1016/S0040-6031(01)00795-X/10.1016/S0040-6031(01)00795-X/ ↗
34. 10.1021/tx200211v10.1021/tx200211v ↗
35. 10.3390/sym1407146110.3390/sym14071461 ↗
36. 10.1039/D3CP02121A10.1039/D3CP02121A ↗
37. 10.1016/0010-2180(83)90062-710.1016/0010-2180(83)90062-7 ↗
38. Mohapatra N, Shreya K, Chinmay A. 2020. Optimization of the random forest algorithm. Vol 2900. Springer Singapore.
39. 10.1007/s11760-021-02009-x10.1007/s11760-021-02009-x ↗
40. 10.1016/j.cej.2024.15041010.1016/j.cej.2024.150410 ↗
41. 10.1016/0925-2312(91)90023-510.1016/0925-2312(91)90023-5 ↗
42. Pal K, Patel BV. 2020. Data classification with k-fold cross validation and holdout accuracy estimation methods with 5 different machine learning techniques. IEEE.
43. Politzer P, Lane P, Concha MC. 2003. Computational approaches to heats of formation, in theoretical and computational chemistry pp 247–277. Elsevier.
44. 10.1016/S0166-1280(01)00533-410.1016/S0166-1280(01)00533-4 ↗
45. 10.1016/j.bspc.2021.10317010.1016/j.bspc.2021.103170 ↗
46. 10.1016/j.scriptamat.2024.11635010.1016/j.scriptamat.2024.116350 ↗
47. 10.3390/en1602067510.3390/en16020675 ↗
48. Serafim MSM et al. 2025. Experimental assays: Chemical properties, biochemical and cellular assays, and in vivo evaluations. Comput-Aided Machine Learning-Driven Drug Design. 347.
49. 10.1134/S001050822106012510.1134/S0010508221060125 ↗
50. 10.1016/j.jhazmat.2008.04.01110.1016/j.jhazmat.2008.04.011 ↗
51. 10.1016/j.jhazmat.2004.05.01610.1016/j.jhazmat.2004.05.016 ↗
52. 10.3390/electronics1103032210.3390/electronics11030322 ↗
53. 10.1109/TNNLS.2015.242499510.1109/TNNLS.2015.2424995 ↗
54. 10.1016/j.physleta.2025.13059710.1016/j.physleta.2025.130597 ↗
55. 10.3390/molecules2102023610.3390/molecules21020236 ↗
56. 10.1007/s00521-021-05995-810.1007/s00521-021-05995-8 ↗
57. 10.1109/TII.2023.329386310.1109/TII.2023.3293863 ↗
58. 10.1002/advs.20240526210.1002/advs.202405262 ↗
59. 10.1007/s10462-022-10283-510.1007/s10462-022-10283-5 ↗
60. 10.3390/molecules2801032210.3390/molecules28010322 ↗
61. 10.1016/j.combustflame.2024.11385810.1016/j.combustflame.2024.113858 ↗
62. 10.1016/j.csbj.2021.07.01310.1016/j.csbj.2021.07.013 ↗
63. 10.1002/advs.20230636010.1002/advs.202306360 ↗