About this Abstract |
| Meeting |
2026 TMS Annual Meeting & Exhibition
|
| Symposium
|
Artificial Intelligence Applications in Integrated Computational Materials Engineering (AI-ICME)
|
| Presentation Title |
Machine Learning Prediction of Antioxidant Additive Performance from Atomistic Simulations |
| Author(s) |
Shihab Ahmed, Stefan J Eder, Nicole Dörr, Mohamed Musthafa Iqbal, Ashlie Martini |
| On-Site Speaker (Planned) |
Shihab Ahmed |
| Abstract Scope |
Antioxidant additives play a vital role in enhancing the oxidative stability of lubricants, especially under extreme temperature and pressure. In this work, we present a machine learning (ML)-driven framework to evaluate and predict the effectiveness of antioxidants in scavenging radical species using high-throughput molecular simulations. Using classical molecular dynamics simulations with the REACTER package implemented in the open-source LAMMPS, we model scavenging reactions between polyalphaolefin radicals and various antioxidant additives to investigate their inhibiting performance. Thermodynamic and kinetic descriptors extracted from the simulations are combined with graph-based molecular features to form an integrated feature set, which is used to train ML models that predict and rank antioxidant effectiveness. This approach enables rapid evaluation of potential additives without the need for extensive experimentation, offering a data-driven route to designing more effective lubricant formulations. Our findings illustrate how simulation-integrated ML workflows can accelerate additive discovery and support smarter decision-making in materials engineering. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Machine Learning, Computational Materials Science & Engineering, |