Hamiltonian learning reveals optoelectronic mechanisms across thermodynamic state space in soft semiconductors

Abstract

Predicting optoelectronic response across thermodynamic state space requires coupling finite-temperature nuclear dynamics to electronic structure at scales where direct first-principles calculations are impractical. Machine-learning force fields and Hamiltonian-learning models provide scalable predictions, but integrating them into reliable and interpretable workflows remains challenging. Here, we introduce FLOW-OTTER, a modular, model-agnostic framework that automates molecular dynamics, Hamiltonian prediction, observable extraction, reliability assessment, and Hamiltonian-level interpretation. We demonstrate FLOW-OTTER in halide perovskites, soft semiconductors whose anharmonic fluctuations strongly modulate electronic response. Using FLOW-OTTER, we show that independently learned nuclear and electronic models remain predictive when composed end-to-end, reproducing experimental temperature- and pressure-dependent band-gap trends using models trained only on first-principles targets at zero pressure. By resolving the nonlinear evolution of Pb-s/Br-p antibonding at the valence-band maximum, it identifies the microscopic origin of the asymmetric pressure response. FLOW-OTTER thus establishes Hamiltonian learning as a general route from thermodynamic trajectories to experimentally grounded optoelectronic mechanisms.

Theoretical description of the energy flow mechanisms in photoexcited semiconductors
Raja Sen, Nathalie Vast, Jelena Sjakste
Proceedings of SPIE, the International Society for Optical Engineering26 May 2026Peer-reviewedPublisher termsFull textAtomic, molecular & optical physicsHAL
Prescreening Point Defects in Semiconductors With Machine Learning
Paul Karlsson, Joel Davidsson, Rickard Armiento
arXiv13 Sep 2026PreprintarXiv licenceFull textCondensed matterarXiv
Comment on "Supervised Learning with Quantum Hamiltonian Embeddings" (arXiv:2610.26833)
Alexander G. Abanov, Luca Candelori, Kharen Musaelian
arXiv3 Nov 2026PreprintCC BY 4.0Full textQuantum physics & quantum informationarXiv
Thermodynamic learning
F. Corberi, S. dello Russo, G. Messuti +2
arXiv10 Sep 2026PreprintarXiv licenceFull textCondensed matter Statistical & nonlinear physicsarXiv
Quantum Sensing and Hamiltonian Learning under Stochastic Parameter Evolution
Kelvin Koor, Patrick Rebentrost
arXiv30 Sep 2026PreprintarXiv licenceFull textQuantum physics & quantum informationarXiv

Cite

BibTeX RIS

BibTeX

@misc{vonhoff2026hamiltonian,
  title = {{Hamiltonian learning reveals optoelectronic mechanisms across thermodynamic state space in soft semiconductors}},
  author = {Frederik Vonhoff and Jesper R. Pedersen and Frederico P. Delgado and Martin Schwade and Peter Beck and Jonas A. Oldenstaedt and Ivano E. Castelli and David A. Egger},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.29361},
  eprint = {2609.29361},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.29361},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Hamiltonian learning reveals optoelectronic mechanisms across thermodynamic state space in soft semiconductors
AU  - Frederik Vonhoff
AU  - Jesper R. Pedersen
AU  - Frederico P. Delgado
AU  - Martin Schwade
AU  - Peter Beck
AU  - Jonas A. Oldenstaedt
AU  - Ivano E. Castelli
AU  - David A. Egger
PY  - 2026
DA  - 2026/09/24
PB  - arXiv
DO  - 10.48550/arxiv.2609.29361
UR  - https://arxiv.org/abs/2609.29361
AB  - Predicting optoelectronic response across thermodynamic state space requires coupling finite-temperature nuclear dynamics to electronic structure at scales where direct first-principles calculations are impractical. Machine-learning force fields and Hamiltonian-learning models provide scalable predictions, but integrating them into reliable and interpretable workflows remains challenging. Here, we introduce FLOW-OTTER, a modular, model-agnostic framework that automates molecular dynamics, Hamiltonian prediction, observable extraction, reliability assessment, and Hamiltonian-level interpretation. We demonstrate FLOW-OTTER in halide perovskites, soft semiconductors whose anharmonic fluctuations strongly modulate electronic response. Using FLOW-OTTER, we show that independently learned nuclear and electronic models remain predictive when composed end-to-end, reproducing experimental temperature- and pressure-dependent band-gap trends using models trained only on first-principles targets at zero pressure. By resolving the nonlinear evolution of Pb-s/Br-p antibonding at the valence-band maximum, it identifies the microscopic origin of the asymmetric pressure response. FLOW-OTTER thus establishes Hamiltonian learning as a general route from thermodynamic trajectories to experimentally grounded optoelectronic mechanisms.
ER  -

Abstract

Related

Cite