A Physics-Chemistry-Informed Neural Network (PCINN) for Real-Time Spatial-ALD Coverage Prediction and Reliable Kinetics Inversion

Researchers introduced a physics-chemistry-informed neural network (PCINN) that predicts surface coverage in spatial atomic layer deposition (SALD) in about 7 milliseconds, roughly 50,000 times faster than CFD, with a test R²_log of 0.998 from only 30 training cases. The hybrid surrogate encodes known surface kinetics as a trainable layer, enabling reliable inversion of adsorption energy and desorption rate, while identifying a degeneracy valley for prefactor and activation energy. The study, based on simulated data, verifies pipeline self-consistency and identifiability boundaries rather than real parameters.

arXiv:2608.00212v1 Announce Type: new Abstract: Spatial atomic layer deposition SALD is a leading atmospheric-pressure, high-throughput route to industrial ALD, but design and control are limited by the cost of predicting surface coverage: high-fidelity CFD is far too slow for operating-window scans, while analytic models miss transport modulation such as the gas curtain. We present a physics-chemistry-informed neural network PCINN , a hybrid surrogate with CFD-level accuracy at real-time speed: a query returns coverage in about 7 ms, roughly 5x10^4 times faster than a CFD solve, reaching a test R^2 log = 0.998 leave-one-out R^2 raw = 0.974 from only 30 training cases spanning four orders of magnitude in coverage. The architecture is not a black box: a small network learns only the operating-condition to near-wall concentration closure, while the known surface kinetics is a hard-coded, trainable chemistry layer integrated along the substrate trajectory. This single-scalar bottleneck keeps it accurate under sparse data, interpretable and invertible. We add a full identifiability analysis Fisher information, profile likelihood . The adsorption energy E ads and desorption rate k des are robustly identifiable; k ads is not separately identifiable at a single temperature only k ads c wall is . Across four temperatures the prefactor nu and E ads bind along a weakly identifiable degeneracy valley of slope 0.065 eV/decade, derived analytically as k B T eff ln 10 and turned into a reliability diagnostic: a seven-chemistry mismatch matrix shows it is invariant under any single-Arrhenius mismatch and shifts only when a second thermally activated process appears, so a slope departure flags unmodelled site heterogeneity. Data come from simulation with known ground truth inverted by the same kinetic form, so the study verifies pipeline self-consistency and the identifiability boundary, not real parameters.