Molecular Dynamics with Nuclear Effects on Quantum Computers

arXiv:2610.01549 · quant-ph, physics.chem-ph · Submitted 2026-10-01 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: Today's paper: "Molecular Dynamics with Nuclear Effects on Quantum Computers".

Mira: The gist Molecular Dynamics with Nuclear Effects on Quantum Computers introduces a novel hybrid quantum-classical algorithm for ab-initio molecular dynamics that incorporates nuclear quantum effects via the nuclear-electronic orbital method…

Kai: First, who's behind it and why it matters.

Title and authors: Kai: So, to recap this paper "Molecular Dynamics with Nuclear Effects on Quantum Computers" is about using a hybrid algorithm where the classical computer handles the geometry changes, but the quantum computer does the heavy lifting for calculating energy and forces when you include those nuclear quantum effects.

Mira: Exactly. They are using something called the Nuclear Electronic Orbital or NEO method to treat specific light nuclei quantum mechanically while keeping everything else classical, which they say is way more accurate than just treating everything as pure electronics.

Lev: And from my side, what’s interesting is how they have to manage that workload on current quantum hardware right now. They have to select a small active space because you simply can't fit every possible interaction onto a noisy intermediate-scale device yet.

Kai: Right, Lev? So they build this active space by scoring orbitals using things like single-orbital entropies and checking how much those orbitals change along the expected path of a chemical reaction.

Mira: That’s the crucial part for me because it shows you don't need to tackle every single quantum degree of freedom at once; you just need to focus your limited qubits on what really matters for that specific molecular movement.

Lev: And they do use parity mapping to get those fermionic Hamiltonians onto the qubits, which is a standard trick, but they also have to pick a specific variational quantum eigensolver called the real amplitudes ansatz because it’s shallower and better for devices we have now.

Kai: So they test this entire setup on simple molecules like water and hydrogen peroxide ions and show that the NEO approach actually gives them better error estimates than purely electronic methods for water's vibrations.

Mira: But there are trade-offs, though; specifically for hydrogen, the NEO method actually underestimates the frequency compared to a purely electronic calculation, which is something we have to keep in mind as we look at these results.

Lev: The numbers they report show that with a small active space—maybe five to seven orbitals—they get average errors under five percent when using the UCCSD ansatz, which is pretty solid for what’s on the table today.

Kai: That’s good news because it means we can actually use this kind of method to get reliable predictions for how molecules vibrate, even with the constraints of current quantum machines.

Mira: But they also pointed out a limitation; they said the accuracy really depends on getting those initial electronic and nuclear orbitals right, no matter which basis set you pick.

Lev: That makes sense from an engineering standpoint because if the basis set is wrong, all that complex quantum calculation won't fix it; you have to start with good input data.

Kai: So they’ve shown a path for running high-quality molecular dynamics on quantum computers by carefully balancing what you put on the quantum chip and what you keep on the classical machine.

Mira: And that leads right into how these results might change how we design new materials or understand chemical reactions in the future.

The paper's summary: Kai: So we've talked about setting up this hybrid system for molecular dynamics on quantum hardware, and now we’re looking at what the paper actually summarizes as their main findings.

Mira: Basically, "Molecular Dynamics with Nuclear Effects on Quantum Computers" showed that by mixing classical geometry updates with a quantum computer handling the nuclear forces via the NEO method, you can get pretty good results for molecules like water and ions.

Lev: They are proving that including those nuclear quantum effects actually helps improve the accuracy of vibrational spectra when compared to methods that ignore them entirely.

Kai: So they’re showing that these extra quantum effects matter for understanding how molecules move and vibrate in a dynamic way, not just static snapshots.

Mira: They also pointed out specific behaviors; for example, the NEO simulation shows a complete lack of mode coupling, but then the pz-type NEO simulation shows more structure because of those nuclear pz orbitals increasing the effective interaction range.

Lev: That means understanding those specific orbital details is key to getting meaningful physical information from the quantum calculation.

Kai: And they also discussed how they can speed up the process by looking at advanced gradient calculation techniques, like NEODDIIS, to get those forces faster from the quantum computer.

Mira: That’s a big deal for us theorists because calculating those gradients is where a lot of the complexity hides, and if you can extrapolate or use some sort of linear combination of previous calculations, it cuts down on the necessary circuit depth significantly.

Lev: Faster convergence means less time running on hardware that has limited coherence times, which is a huge hurdle for any real quantum computer we’re aiming to build.

Kai: And they also looked at basis sets—which are basically the mathematical tools we use to describe the nuclei—and they found that using bigger nuclear basis sets can actually hurt accuracy if you don't choose them right.

Mira: So they concluded that getting those initial electronic and nuclear orbitals correct is absolutely paramount, so you have to be meticulous about your choice of tools before you even start the quantum calculation.

The paper's improvements: Kai: Now we’re looking at what the authors suggest to make this hybrid system even better, focusing on how they can improve these simulations moving forward.

Mira: They propose making the active space selection process smarter by using single-orbital entropies and checking their variability along the reaction path, which means we can choose exactly which quantum degrees of freedom are worth spending our limited qubits on.

Lev: That’s a practical step because if you’re not careful with that active space, you just end up wasting all your qubits on stuff that doesn't actually matter for the chemical change.

Kai: Then they talk about using more advanced gradient calculation techniques, like something called NEODDIIS, to speed up how fast we can get those forces from the quantum computer.

Mira: That’s a big deal for us theorists because calculating those gradients is where a lot of the complexity hides, and if you can extrapolate or use some sort of linear combination of previous calculations, it cuts down on the necessary circuit depth significantly.

Lev: Faster convergence means less time running on hardware that has limited coherence times, which is a huge hurdle for any real quantum computer we’re aiming to build.

Kai: And they also look at basis sets—which are basically the mathematical tools we use to describe the nuclei—and they found that using bigger nuclear basis sets can actually hurt accuracy if you don't choose them right.

Mira: They concluded that getting those initial electronic and nuclear orbitals correct is paramount, so you have to be meticulous about your choice of tools before you even start the quantum calculation.

Conclusion: Kai: So we've covered how they set up this hybrid system for molecular dynamics on quantum hardware, summarizing what we found about using quantum computers for these simulations.

Mira: Basically, the paper "Molecular Dynamics with Nuclear Effects on Quantum Computers" showed that by mixing classical geometry updates with a quantum computer handling the nuclear forces via the NEO method, you can get pretty good results for molecules like water and ions.

Lev: I just see it as a proof-of-concept for how these hybrid models could work on actual hardware, but they kept emphasizing that the success depends entirely on how well you choose your active space and your VQE ansatz.

Kai: That’s right, Lev? So if you want to run this on a noisy intermediate-scale quantum computer, you have to be really careful about those choices.

Mira: They also pointed out that while the method works well for some things, like water, other systems like hydrogen might actually see an overcorrection in their frequency predictions.

Lev: And they flagged that limitation—the choice of the initial basis set matters a lot; if you pick the wrong tools for those nuclei, even a clever quantum algorithm won't save you.

Kai: It seems like the main implication here is that this hybrid approach is a viable path for getting more accurate molecular simulations when you need to account for those proton shuttling motions in complex systems.

Mira: It means we can start thinking about how this framework could be used to study things beyond just simple molecules, maybe even designing new materials where nuclear interactions are key.

Lev: For running this on real quantum devices, the focus will definitely have to be on making those active space selections and ansatz choices more automated so they don't take hours of manual tuning every time.

Kai: So we’ve seen how they set up this complex system, what the numbers show for water and ions, and where the current limitations are.

Mira: And that brings us to where we go next—we need to talk about how this kind of method might connect with other quantum approaches for many-body systems.

Lev: I think the next step is figuring out how to scale that active space selection process without losing accuracy, which is where error correction research often comes in.

Kai: Well, that’s where we head off for now as we look at what else is coming in the quantum computing space.

Lukas Haßfurth, Juliane Heitkämper, Elias Walter, Birger Horstmann

German Aerospace Center (DLR) · Helmholtz Institute Ulm

quant-ph, physics.chem-ph

Submitted: 2026-10-01

Updated: 2026-10-01

Comments: 21 pages, 14 figures

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 67/100

The gist: The gist Molecular Dynamics with Nuclear Effects on Quantum Computers introduces a novel hybrid quantum-classical algorithm for ab-initio molecular dynamics that incorporates nuclear quantum effects

Key concepts

Nuclear Electronic Orbital (NEO) Method
This framework treats selected light nuclei quantum-mechanically alongside electrons using second quantization. It establishes coupled nuclear-electronic self-consistent field equations to generate molecular orbital coefficients, transforming atomic orbitals into MOs for both electronic and nuclear descriptions.
Active Space Reduction via Hartree-Fock Embedding
To handle the limited qubits on current quantum computers (NISQ), an active space is chosen. This involves treating only a subset of electronic and nuclear MOs quantum-mechanically, while the rest are treated classically using the Hartree-Fock level. Criteria like orbital entropy and reaction path variability guide this selection.
Variational Quantum Eigensolver (VQE)
VQE is used on the quantum computer to find the instantaneous ground state energy at each timestep of classical molecular dynamics. It works by minimizing the expectation value of the Hamiltonian with respect to parameters in a chosen ansatz, which prepares an accurate quantum state.
Hardware-Efficient Ansatz
These are specific circuit designs for VQE that are tailored to be efficient on current quantum hardware. Examples include the tiled unitary product ansatz (tUPS) and real amplitudes ansatz, which are favored because they have shallow depths and nearest-neighbor topologies suitable for NISQ devices.

Terminology

Summary

The gist Molecular Dynamics with Nuclear Effects on Quantum Computers introduces a novel hybrid quantum-classical algorithm for ab-initio molecular dynamics that incorporates nuclear quantum effects via the nuclear-electronic orbital method to evaluate ground state energies and forces on a quantum computer while updating molecular geometries classically

How it works

The proposed hybrid quantum-classical ab-initio molecular dynamics algorithm follows a second quantization approach and is based on the classical Born-Oppenheimer dynamics It evaluates the energy and the forces acting on the nuclei on the quantum computer, while molecular geometries are updated classically where selected nuclei are treated quantum-mechanically according to the nuclear electronic orbital (NEO) method At every timestep, an instantaneous ground state is found using the variational quantum eigensolver (VQE) on the quantum computer The total force on each nuclear coordinate is calculated by measuring the expectation values of the force operators and adding the nuclear repulsion force The nuclear coordinates are then updated using the velocity Verlet algorithm

Nuclear-electronic structure calculations

The NEO theoretical framework treats selected light nuclei quantum-mechanically in second quantization on the same footing as electrons, leading to coupled nuclear-electronic selfconsistent field (SCF) equations This approach yields molecular orbital (MO) coefficients for electronic and nuclear orbitals, transforming atomic orbitals into MOs The NEO Hamiltonian in second quantization is given by HNEO The one-body electronic and nuclear integrals are defined as h e,n pq, and the two-body integrals are g e,n pqrs

Active Space Reduction

To manage the limited number of qubits on noisy intermediate-scale quantum computers (NISQ), an active space is selected by implementing Hartree-Fock embedding The active space NEO Hamiltonian HA NEO is constructed by treating a subset of electronic and nuclear MOs on the quantum computer while the remaining orbitals are treated at the Hartree-Fock level Active space selection involves combining three criteria: scoring orbitals by their single-orbital entropies, variability along the expected reaction path, and enforcing symmetry through mirror image inclusion for physical consistency A reference geometry is also defined to uniquely fix the active space by maximizing overlap integrals with the active space on that geometry

Quantum Chemistry on Quantum Computers

The fermionic representation of the Hamiltonian is mapped to qubits using a parity mapping scheme, which can be extended to NEO by tapering two additional qubits for the number of quantum nuclei The instantaneous ground state is prepared using VQE, minimizing the expectation value of the Hamiltonian with respect to parameters of an ansatz U(θ) Hardware-efficient ansatzes, such as the tiled unitary product ansatz (tUPS) and real amplitudes ansatz, are employed because they are designed to be applicable on current quantum devices

Results and Discussion

Simulations of H2, H2O, and the Zundel ion H5O2+ were performed comparing simulated vibrational spectra with experimental data For the hydrogen molecule, NEO results underestimate the frequency while the purely electronic method overestimates it For water, the NEO method achieves better errors than the purely electronic method for all normal modes, but a systematic underestimation of frequencies is observed For the Zundel ion, including nuclear quantum effects makes the proton shuttling movement effectively barrierless and improves errors in simulated frequencies The NEO s-type simulation shows a complete lack of mode coupling, whereas the pz-type NEO simulation exhibits more structure due to nuclear pz orbitals increasing effective interaction range The hybrid MD algorithm produced the highest-quality results in conjunction with the UCCSD ansatz, with an average error of less than 5 % while applying a relatively small active space of five to seven orbitals

Conclusion

The hybrid MD algorithm successfully simulated vibrational modes for H2, H2O and the Zundel ion H5O2+ by incorporating nuclear quantum effects via the NEO framework The results demonstrate that employing NEO improves normal mode simulations for water and the Zundel ion, while the H2 stretch mode frequency gets overcorrected The most accurate simulations employed the (NEO)UCCSD ansatz, achieving an average error of less than 5 % with a small active space of five to seven orbitals The real amplitudes ansatz is suggested as a promising candidate for running the algorithm on NISQ hardware due to its shallow depth and nearest-neighbor topology The correct determination of active electronic and nuclear orbitals is essential for achieving reliably accurate predictions, irrespective of the concrete chosen orbital basis

How it works

The paper details a hybrid quantum-classical ab-initio molecular dynamics algorithm that uses the quantum computer as an efficient energy and force evaluator to update molecular geometries classically The algorithm follows a second quantization approach based on classical Born-Oppenheimer dynamics, where the instantaneous ground state is found via VQE at every timestep

How it works

The method for simulating ab-initio molecular dynamics on a quantum computer involves propagating classical point masses according to Hamiltonian mechanics, and obtaining the force operators from the ground state expectation values The total force is calculated as FI (ti) = − ⟨ψ0(ti)∇IH(ti) ψ0(ti)⟩ − ∇IVNN(R)

How it works

The calculation of the nuclear gradient of the NEO-Hamiltonian dHNEO/dR is performed using analytical expressions derived from AO integral derivatives or through Coupled Perturbed Hartree-Fock (CPHF) methods The complexity analysis shows that the most expensive part of the analytical derivative is the transformation of two-body integrals to the MO basis, which can be written as four O(M5) steps

How it works

The paper discusses various ansatz types for VQE, including UCCSD and hardware-efficient ansatzes like tUPS and real amplitudes ansatz The choice of ansatz greatly influences the efficacy of the VQE algorithm, with hardware-efficient ansätze being more suitable for near-term quantum devices due to their shallower circuits

How it works

The paper investigates basis set dependence by using different nuclear basis sets like DZSNB and DZSPNB for the Zundel ion, showing that larger nuclear basis sets severely underestimate simulated frequencies The study concludes that the correct determination of active electronic and nuclear orbitals is essential for achieving reliably accurate predictions irrespective of the concrete chosen orbital basis

How it works

The paper evaluates anharmonicities by considering ab-initio potential energy surfaces, which automatically accounts for these effects in simulations like H2 The simulation of the Zundel ion with the reduced NEOUCCSD ansatz shows that coupling between the proton transfer mode and both water wagging and water-water stretch modes is visible, improving upon its purely electronic counterpart The hybrid MD algorithm produced the highest-quality results in conjunction with the UCCSD ansatz, with an average error of less than 5 % while applying a relatively small active space of five to seven orbitals

How it works

The study concludes that there is a relatively free choice of initial displacement as long as it is sufficiently small, with an upper bound for the displacement around 0.

Improvements for AI systems

  1. We can improve ab-initio molecular dynamics simulations by implementing a hybrid quantum-classical algorithm that evaluates ground state energies and forces on the quantum computer using a variational quantum eigensolver while updating geometries classically, allowing for the simulation of proton shuttling movement in the Zundel ion [to become] effectively barrierless.

  2. The improved system can accurately predict vibrational spectra for molecules like H2, H2O, and H5O2+, achieving results comparable to the more accurate UCCSD ansatzes while using hardware-efficient ansatzes that require significantly less quantum resources.

  3. The algorithm can be specialized for specific chemical processes by selecting an active space based on three criteria: scoring orbitals by single-orbital entropies, checking their variability along the expected reaction path, and ensuring the active space is invariant under the chemical reaction’s geometric transformations.

  4. For proton transfer studies, a system can be optimized to predict energy barriers by comparing results from purely electronic methods with NEO, noting that the zero-point motion of the shared proton exerts a notable influence even in room-temperature systems, which changes the potential energy surface from a double-well to a single-well potential.

  5. The system can incorporate advanced gradient calculation techniques, such as NEODDIIS, which uses extrapolation to improve convergence speed by solving systems where the extrapolated Fock matrix derivatives are then constructed by linear combination of the previous matrix derivatives.

Abstract

Nuclear quantum effects are critical for describing proton transfer and hydrogen bonding, but their incorporation into quantum chemistry calculations is often computationally prohibitive on classical hardware. A promising alternative are quantum computers due to their linear scaling in the space requirements with system size. We introduce a novel hybrid quantum-classical algorithm for ab-initio molecular dynamics that incorporates nuclear quantum effects via the nuclear-electronic orbital method. The proposed algorithm evaluates ground state energies and forces on the quantum computer using a variational quantum eigensolver, while the molecular geometries are updated classically. We validate our approach through simulations of H 2, H 2 O and the Zundel ion H 5 O 2+, comparing the simulated vibrational spectra with experimental data. Upon inclusion of nuclear quantum effects, the proton shuttling movement in the Zundel ion becomes effectively barrierless, and errors in the simulated frequencies improve significantly. Employing compact hardware-efficient ansatz circuits we achieve results comparable to the more accurate UCCSD ansatzes, which hints towards the feasibility of executing our algorithm on near-term quantum devices.

Sources

Related papers