Om Patel

Bachelor's Thesis Research: Machine Learning Potentials for MOF Systems at SRM Institute of Science and Technology (2024-09 – 2026-04)

Supervisor: Dr. Arijit Sen, Research Associate Professor

Developed Python-based GAP-DFT hybrid workflow (ASE, custom CASTEP parsers) for 5 lanthanide-based MOFs; achieved force accuracy of 0.05 eV/Å (mean absolute error) with 15,000× computational speedup compared to pure hybrid DFT on HPC clusters (108 cores, OpenMPI).
Validated GAP-predicted geometries against hybrid DFT benchmarks: mean RMSD of 0.08 Å for 150-atom MOF systems across 50 ps MD trajectories; adaptive on-the-fly learning scheme reduced DFT calls by 94% while maintaining energy accuracy within 2 meV/atom.

Undergraduate Research Project: Polymer Nanocomposite Mechanical Properties at SRM Institute of Science and Technology (2024-01 – 2024-09)

Supervisor: Dr. Mithilesh K. Dikshit, Assistant Professor

Simulated elastic moduli of h-BNNS/polyurethane nanocomposites (15 configurations, 3–7 wt% loading) using Materials Studio Forcite; found 40% increase in Young's modulus at 5 wt% with uniform nanosheet dispersion; predictions within 5% of experimental benchmarks (32 vs 34 GPa).
Computed Young's moduli (22–67 GPa) for 12 polyurethane-BNNS interface configurations via DFT (CASTEP) and MD (Forcite); identified optimal nanosheet orientation (30° to loading axis) that maximised tensile strength by 2.8× compared to pure polymer baseline.
Analysed stress-strain behaviour under uniaxial tensile loading (0–10% strain); determined interfacial shear strength (45 MPa) and critical nanosheet concentration (4.5 wt%) for percolation-driven reinforcement.

Academic Coursework and Thesis Preparation at SRM Institute of Science and Technology (2024-10 – 2024-12)

Completed advanced electives: Computational Quantum Chemistry, Advanced Materials Modeling, High-Performance Computing Applications.
Conducted literature review and preliminary calculations for Bachelor's thesis on ML potentials for MOF systems.

Computational Research Intern (Remote) at A-Labs Advanced Materials R&D Division (2025-01 – 2025-03)

Project Lead: Gyurae Ryan Kim, Chief Technology Officer

Computed CO2 adsorption energies and isotherms for 8 porous materials (5 MOFs, 3 zeolites) using DFT (CASTEP); binding energies ranged from −18 to −42 kJ/mol, with zeolite 13X showing 23% higher volumetric capacity than MOF-5 at 1 bar and 298 K.
Performed isotherm fitting (Langmuir, dual-site Langmuir models) to experimental data; identified HKUST-1 as optimal for low-pressure CCUS applications based on working capacity (3.2 mmol/g between 0.1–1 bar) and regeneration efficiency.

About