Error-Mitigation Enabled Multicomponent Quantum Simulations Beyond the Born–Oppenheimer Approximation

Delmar G. A. Cabral, Brandon Allen, Fabijan Pavošević, Sharon Hammes-Schiffer, Pablo Díez-Valle, Jack S. Baker, Gaurav Saxena, Thi Ha Kyaw, Victor S. Batista · 2025

Read Error-Mitigation Enabled Multicomponent Quantum Simulations Beyond the Born–Oppenheimer Approximation by Delmar G. A. Cabral, Brandon Allen, Fabijan Pavošević, Sharon Hammes-Schiffer, Pablo Díez-Valle, Jack S. Baker, Gaurav Saxena, Thi Ha Kyaw, Victor S. Batista (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-11941.txt.

Introduction

Delmar G. A. Cabral and eight co-authors run quantum simulations of molecules with quantum nuclei. Electrons and nuclei are both treated as quantum particles. The Born-Oppenheimer approximation, which freezes nuclei in place, is dropped. Light nuclei, above all protons, tunnel and spread out. The approximation distorts their chemistry.

A multicomponent unitary coupled cluster method is built for the task. It rests on the nuclear-electronic orbital formalism. Targets include positronium hydride and hydrogen with a quantum proton. Resource costs are mapped for distinct excitation levels. A compact variant ran on IBM's Heron hardware. Error mitigation kept energies within chemical accuracy. Chemistry beyond Born-Oppenheimer, on real quantum hardware, is the milestone.

Catalog record

CREATORDelmar G. A. Cabral, Brandon Allen, Fabijan Pavošević, Sharon Hammes-Schiffer, Pablo Díez-Valle, Jack S. Baker, Gaurav Saxena, Thi Ha Kyaw, Victor S. Batista
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-11941

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Error-Mitigation Enabled Multicomponent Quantum Simulations Beyond the Born–Oppenheimer Approximation Delmar G. A. Cabral, Brandon Allen, Fabijan Pavošević, Sharon Hammes-Schiffer, Pablo Díez-Valle, Jack S. Baker, Gaurav Saxena, Thi Ha Kyaw, Victor S. Batista Abstract We introduce a multicomponent unitary coupled cluster framework for quantum simulations of molecular systems that incorporate both electronic and nuclear quantum effects beyond the Born–Oppenheimer approximation. Using the nuclear–electronic orbital formalism, we construct mcUCC ansätze for positronium hydride and molecular hydrogen with a quantum proton, and analyze hardware requirements for different excitation truncations. To further reduce resource costs effectively, we employ the local unitary cluster Jastrow ansatz and implement it experimentally on IBM Q’s Heron superconducting hardware. With the Physics-Inspired Extrapolation error mitigation protocol, the computed ground-state energies remain within chemical accuracy, consistent with the stated uncertainty level. These results provide the first demonstration of error-mitigated multicomponent correlated simulations on quantum hardware and outline a path toward scalable algorithms unifying electronic and nuclear degrees of freedom. † † affiliation: Department of Chemistry, Yale University, New Haven, CT 06520, U.S.A. † † affiliation: Department of Chemistry, Yale University, New Haven, CT 06520, U.S.A. † † affiliation: Algorithmiq Ltd., Kanavakatu 3C, FI-00160 Helsinki, Finland † † affiliation: Department of Chemistry, Princeton University, Princeton, NJ 08540, U.S.A. † † affiliation: Instituto Tecnológico de Galicia, Cantón Grande 9, Planta 3, 15003 A Coruña, Spain † † affiliation: LG Electronics Toronto AI Lab, Toronto, Ontario M5V 1M3, Canada † † affiliation: LG Electronics Toronto AI Lab, Toronto, Ontario M5V 1M3, Canada † † affiliation: LG Electronics Toronto AI Lab, Toronto, Ontario M5V 1M3, Canada † † email: [email protected] † † affiliation: Department of Chemistry, Yale University, New Haven, CT 06520, U.S.A. † † affiliation: Yale Quantum Institute, Yale University, New Haven, CT 06511, U.S.A. † † email: [email protected] 1 Introduction Many fundamental chemical processes involve nuclear quantum effects that challenge the Born–Oppenheimer (BO) separation of electronic and nuclear motion. Phenomena such as proton tunneling, hydrogen transfer, and proton-coupled electron transfer are governed by zero-point motion and nonadiabatic couplings that significantly influence reaction thermodynamics and kinetics 47 ; 33 ; 16 ; 46 ; 24 ; 18 . Classical treatments that confine nuclei to point particles can therefore produce qualitatively incorrect predictions. The nuclear–electronic orbital (NEO) framework addresses this limitation by treating selected light nuclei (typically protons) quantum mechanically alongside electrons. By incorporating nuclear delocalization, tunneling, and electron–nucleus correlation directly into the wavefunction, NEO theory provides a unified and systematically improvable approach for systems in which electronic and nuclear degrees of freedom are strongly coupled 48 ; 17 . Despite the accuracy of NEO-based methods, solving correlated electronic–nuclear structure problems remains computationally demanding. Classical algorithms for exact solutions scale exponentially with system size, motivating the use of quantum computation as an alternative route to scalable molecular simulation 1 ; 5 . Two major paradigms have emerged: quantum phase estimation (QPE) and the variational quantum eigensolver (VQE) 7 ; 10 ; 8 . QPE can, in principle, yield exact eigenvalues efficiently, but its deep, coherent circuits remain impractical for current noisy intermediate-scale quantum (NISQ) devices 40 . VQE, in contrast, employs shallow parameterized circuits with classical optimization and has been successfully demonstrated on small molecular systems 39 ; 36 ; 23 ; 41 ; 20 . Within VQE, the unitary coupled cluster (UCC) ansatz 2 , derived from the coupled cluster formalism 11 , provides a physically grounded and systematically extensible representation of correlated wavefunctions. While classical simulations of UCC scale exponentially, its quantum implementation requires only polynomial resources, making it a natural foundation for chemically realistic quantum algorithms. Recent work has unified NEO theory with quantum algorithms via multicomponent UCC (mcUCC) ansätze, enabling beyond-BO simulations that explicitly include quantum nuclear motion in both the reference and variational spaces 37 ; 27 ; 35 ; 25 ; 26 ; 12 . This NEO–quantum computing (NEO-QC) framework provides a rigorous route to first-principles modeling of electronic–nuclear correlated systems, including positronic and protonic species, where classical approximations fail. However, the accuracy of hybrid quantum–classical algorithms on present-day devices remains constrained by noise and decoherence. Full quantum error correction is not yet feasible, but quantum error mitigation (QEM) methods can substantially reduce systematic bias without fault-tolerant overhead. Among the most widely used are zero-noise extrapolation (ZNE), which estimates noise-free observables by deliberate noise amplification and extrapolation 29 ; 45 , and probabilistic error cancellation (PEC), which reconstructs unbiased estimators via quasi-probability sampling 45 . Complementary strategies such as symmetry verification 9 and virtual (state) distillation 21 have further improved accuracy in quantum chemistry experiments on real hardware 4 ; 31 ; 30 . In this work, we employ a recently developed, physically motivated error mitigation approach, Physics-Inspired Extrapolation (PIE), which extends the ZNE framework by deriving its functional form from restricted quantum dynamics 13 ; 42 . PIE provides an interpretable extrapolation model, mitigates overfitting, and reduces sampling overhead relative to polynomial ZNE, enabling chemically accurate energy estimates for beyond-BO benchmarks on current NISQ hardware. The remainder of this paper is organized as follows. Section 2 introduces the NEO Hamiltonian and working equations. Section 3 describes the mcUCC ansatz used in classical VQE simulations, which serve as surrogates for near-term quantum implementations limited by circuit depth and hardware noise. Section 4 outlines the VQE framework and Hamiltonian constructions for PsH and HHq systems. Section 5 presents PIE-based, error-mitigated VQE results using the LUCJ ansatz on IBM Q’s Heron device. Finally, Section 6 summarizes the results and discusses prospects for scalable, multicomponent quantum simulations beyond the BO approximation. 2 Nuclear Electronic Orbital (NEO) Framework Building upon prior work establishing the foundations of mcUCC methods for quantum computation 37 , we perform simulations of the same systems on a quantum simulator and analyze the computational requirements for execution on real NISQ devices. Two model systems are considered: molecular hydrogen with one quantum mechanical proton ( \mathrm{HHq} ) and positronium hydride ( \mathrm{PsH} ) (see Fig. 2 ). In the NEO formalism, both electrons and selected light nuclei (e.g., protons or positrons) are treated quantum mechanically, while heavier nuclei remain classical. The total NEO Hartree–Fock (NEO-HF) wavefunction is expressed as a product of electronic and nuclear components, |\Psi_{\text{NEO-HF}}(\chi_{e},\chi_{p})\rangle=|\Phi_{e}(\chi_{e})\rangle\otimes|\Phi_{p}(\chi_{p})\rangle, (1) where \Phi_{e}(\chi_{e}) and \Phi_{p}(\chi_{p}) are the electronic and quantum-nuclear wavefunctions, each expanded in their respective molecular orbital bases, \chi_{e} and \chi_{p} . The total Hamiltonian for a system containing electrons, quantum nuclei, and classical nuclei is written as \displaystyle\hat{H}_{\text{NEO}} \displaystyle=\hat{T}_{e}+\hat{T}_{p}+\hat{V}_{eN}+\hat{V}_{pN}+\hat{V}_{ee}+\hat{V}_{pp}+\hat{V}_{ep}+V_{NN}, (2) where \hat{T}_{e} and \hat{T}_{p} are the kinetic energy operators for electrons and quantum nuclei, respectively, and the \hat{V} terms represent the corresponding Coulomb interactions. The one-particle operators are defined as \displaystyle\hat{T}_{e}+\hat{V}_{eN} \displaystyle=\sum_{i=1}^{N_{e}}\left(-\frac{1}{2m_{e}}\nabla_{i}^{2}-\sum_{a=1}^{N_{\text{nuc}}}\frac{Z_{a}}{|\mathbf{R}_{a}-\mathbf{r}_{i}^{e}|}\right), (3) \displaystyle\hat{T}_{p}+\hat{V}_{pN} \displaystyle=\sum_{I=1}^{N_{p}}\left(-\frac{1}{2m_{p}}\nabla_{I}^{2}+\sum_{a=1}^{N_{\text{nuc}}}\frac{Z_{a}}{|\mathbf{R}_{a}-\mathbf{r}_{I}^{p}|}\right), (4) while the two-particle Coulomb interactions take the form \displaystyle\hat{V}_{ee} \displaystyle=\sum_{i<j}^{N_{e}}\frac{1}{|\mathbf{r}_{i}^{e}-\mathbf{r}_{j}^{e}|}, (5) \displaystyle\hat{V}_{pp} \displaystyle=\sum_{I<J}^{N_{p}}\frac{1}{|\mathbf{r}_{I}^{p}-\mathbf{r}_{J}^{p}|}, (6) \displaystyle\hat{V}_{ep} \displaystyle=-\sum_{i=1}^{N_{e}}\sum_{I=1}^{N_{p}}\frac{1}{|\mathbf{r}_{i}^{e}-\mathbf{r}_{I}^{p}|}. (7) Here, indices i,j denote electrons, I,J denote quantum nuclei, and \mathbf{r} represents particle coordinates. Z_{a} and \mathbf{R}_{a} denote the charge and position of the a^{\text{th}} classical nucleus, \nabla is the Laplacian operator, and m_{e} and m_{p} are the electron and quantum-nuclear masses, respectively. The classical nucleus–nucleus repulsion is given by \displaystyle V_{NN}=\sum_{a<b}^{N_{\text{nuc}}}\frac{Z_{a}Z_{b}}{|\mathbf{R}_{a}-\mathbf{R}_{b}|}, (8) which is constant for a fixed nuclear geometry. For practical computations, these operators are expressed in a finite one-particle basis \{\phi\} in terms of

Read the complete text