About this Abstract |
| Meeting |
2027 TMS Annual Meeting & Exhibition
|
| Symposium
|
Computational Thermodynamics and Kinetics
|
| Presentation Title |
BLADE: Batch Lattice Analysis and Discovery Engine for High-Throughput Materials Exploration |
| Author(s) |
Chase R. Katz, Siya Zhu, Doguhan Sariturk, Raymundo Arroyave |
| On-Site Speaker (Planned) |
Chase R. Katz |
| Abstract Scope |
High-throughput materials exploration requires tools that can automatically generate, evaluate, and organize structures across large compositional spaces. We present BLADE, the Batch Lattice Analysis and Discovery Engine, an open-source Python framework for automated structure generation. BLADE accepts user-defined phases through fractional-coordinate inputs, using site labels to distinguish variable and fixed-occupancy crystallographic sites. Users provide phase prototypes, an element pool, and per-site occupancy constraints. From these inputs, BLADE generates special quasirandom structures, performs MLIP-based structural relaxation, and organizes the resulting structures and energies for downstream analysis. BLADE supports both single-sublattice and multi-sublattice phases by independently enumerating allowed species on each variable site while preserving fixed-site constraints. These outputs support phase stability analysis, CALPHAD database generation, machine-learning dataset construction, and high-throughput screening, reducing manual setup for complex phase and composition spaces. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Computational Materials Science & Engineering, Modeling and Simulation, Machine Learning |