Skip to content
DraconiticResearch
Search Sources API
206,690 studies31 sourcesUpdated 22 min ago
PreprintCC BYChemRxiv

Finite-Field Second Hyperpolarizability via Adaptive Bayesian Quantum Phase Estimation for Early Fault-Tolerant Quantum Computers: Ground-State Energies Beyond Chemical Accuracy

  1. 1Research Center for Computational Science and Informatics, Resonac Corporation, 8, Ebisu-cho, Yokohama, Kanagawa, Japan, 221-8517

Abstract

The static second hyperpolarizability governs third-order nonlinear-optical responses but is difficult to compute by first-principles methods because finite-field extraction requires a fourth-order energy derivative. The resulting error amplification demands sub-µHa ground-state energies, far beyond conventional chemical accuracy. Standard multi-qubit QPE can in principle meet this precision but requires a wide ancilla register, while single-ancilla iterative QPE (IQPE) is limited by a binary-grid quantization floor. We develop an adaptive Bayesian QPE (BQPE) workflow that uses a single ancilla, terminal measurements, and real-time classical feedback to estimate the eigenphase through a continuous posterior rather than a fixed binary grid. For a linear H4 chain in the STO-3G basis, represented by a six-qubit parity Hamiltonian with one ancilla, BQPE reaches a ground-state energy error of ≈ 10−7 Ha at controlled-U depth mmax ≈ 215, about two orders of magnitude below the IQPE floor. Propagated through the finite-field stencil and Richardson extrapolation, these energies recover the second hyperpolarizability within ±10% of the FCI reference over the stable field-strength plateau, whereas IQPE at the same depth fails to do so. These results indicate that grid-free Bayesian phase estimation can make nonlinear-optical properties accessible on early fault-tolerant quantum computers.

Related

  1. Real-Space Chemistry on Quantum Computers: A Fault-Tolerant Algorithm with Adaptive Grids and Transcorrelated Extension
    C. Feniou, C. Cherfan, J. Zylberman +3
    Digital DiscoveryPeer-reviewedCC BY 4.0Full textTheoretical & computational chemistryQuantum physics & quantum informationHAL
  2. Damped polarizability and first hyperpolarizability of large systems with hybrid functionals: self-consistent and Lanczos solvers
    A. García-Blázquez, A. Erba, A. Boccuni +1
    arXivPreprintarXiv licenceFull textPhysical chemistryCondensed matterarXiv
  3. First hyperpolarizability of dithienylethene-based photoswitches: a structure-property relationship study.
    Kanchan Sarkar, Ramprasad Misra, Axel Groß +1
    RSC advancesPeer-reviewedCC BY-NCFull textMaterials & polymersEurope PMCPMC
  4. Multi-Stage State-of-Health Estimation for Lithium-Ion Batteries Based on an Adaptive SHAP-Guided Dual-Attention LSTM
    Pengxiang Li, Zixi Wang, Nan Zhou +1
    BatteriesPeer-reviewedCC BY 4.0Electrochemistry & energyCrossref
  5. BRACE: Bayesian Residual-covariance Adaptive Calibration for Multi-Endpoint ADMET Prediction
    Tianchi Chen, Kamala Dadashova, Bingjia Yang +1
    ChemRxivPreprintCC BYFull textGeneral chemistryChemRxiv
BibTeX
@misc{okuno2026finitefield,
  title = {{Finite-Field Second Hyperpolarizability via Adaptive Bayesian Quantum Phase Estimation for Early Fault-Tolerant Quantum Computers: Ground-State Energies Beyond Chemical Accuracy}},
  author = {Katsuki Okuno},
  howpublished = {ChemRxiv},
  year = {2026},
  month = sep,
  doi = {10.26434/chemrxiv.15008602/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008602/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Finite-Field Second Hyperpolarizability via Adaptive Bayesian Quantum Phase Estimation for Early Fault-Tolerant Quantum Computers: Ground-State Energies Beyond Chemical Accuracy
AU  - Katsuki Okuno
PY  - 2026
DA  - 2026/09/09
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15008602/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008602/v1
AB  - The static second hyperpolarizability governs third-order nonlinear-optical responses but is difficult to compute by first-principles methods because finite-field extraction requires a fourth-order energy derivative. The resulting error amplification demands sub-µHa ground-state energies, far beyond conventional chemical accuracy. Standard multi-qubit QPE can in principle meet this precision but requires a wide ancilla register, while single-ancilla iterative QPE (IQPE) is limited by a binary-grid quantization floor. We develop an adaptive Bayesian QPE (BQPE) workflow that uses a single ancilla, terminal measurements, and real-time classical feedback to estimate the eigenphase through a continuous posterior rather than a fixed binary grid. For a linear H4 chain in the STO-3G basis, represented by a six-qubit parity Hamiltonian with one ancilla, BQPE reaches a ground-state energy error of ≈ 10−7 Ha at controlled-U depth mmax ≈ 215, about two orders of magnitude below the IQPE floor. Propagated through the finite-field stencil and Richardson extrapolation, these energies recover the second hyperpolarizability within ±10% of the FCI reference over the stable field-strength plateau, whereas IQPE at the same depth fails to do so. These results indicate that grid-free Bayesian phase estimation can make nonlinear-optical properties accessible on early fault-tolerant quantum computers.
ER  -