Deterministic Pre-DFT Lattice Screening via Electron Network Density
First-principles computational pipelines utilizing iterative Self-Consistent Field (SCF) Density Functional Theory (DFT) impose severe computational latencies (48–72 core-hours per supercell) when triaging combinatorial structural spaces. We present a deterministic, non-iterative screening engine grounded in the Chaware Electron Network Density (END) architecture. Operating strictly on invariant macroscopic mass-balance parameters (bulk density, formula molar mass, and participating active valence count), the system extracts the real-space carrier standoff horizon R_END and intrinsic degeneracy pressure P_Fermi in sub-millisecond execution times. Incorporating domain-specific geometric transformations (covalent, metallic, and ionic), the framework reproduces empirical laboratory X-ray diffraction (XRD) interatomic spacing with a mean absolute relative deviation of 0.91% across 35 benchmark crystalline phases spanning seven symmetry classes. When deployed as an upstream gatekeeper, this closed-form engine autonomously flags interfacial shear dislocations and eliminates structurally unviable compositions prior to HPC resource allocation.
Authors
- Akash Hanumantrao CHAWARE (ORCID: https://orcid.org/0009-0004-0811-9759)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-16
- DOI
- https://doi.org/10.5281/zenodo.24267811
- Primary Topic
- Machine Learning in Materials Science
- Type
- preprint