Abstract
In this paper we discuss the Morse potential on a quantum computer. The Morse potential is useful to describe diatomic molecules and has a finite number of bound states which can be measured through spectroscopy. It is also a example of an exactly soluble potential using supersymmetric quantum mechanics. Using the the supersymmetric quantum mechanics formalism one can derive a heirachy of Hamiltonians such that the ground state of the next rung on the heirarchy yeids the first excited state of the hamiltonian below it. Using this method one can determine all the states of the Morse potential by calculating all the ground states of the sequence of Hamiltonians in the heirarchy. We use the IBM QISKit software together with the Variational Quantum Eiegensolver (VQE) algorithm to calculate the ground state and first excited state energy of the Morse potential and find agreement with the exact expression for the bound state energies of the Morse Potential. We analyze different optimizers to study the numerical effect on the calculations. Finally we perform quantum computations for diatomic and triatomic molecules to illustrate the application of these techniques on near term quantum computers and find excellent agreement with experimental data.
Introduction
In quantum chemistry, the harmonic oscillator is the standard model for molecular vibrations, but it is fundamentally limited as it assumes bonds never break. The Morse potential provides a significantly more realistic model by accounting for anharmonicity and the eventual dissociation of molecules at high energy levels. However, calculating the energy levels—particularly excited states—of these potentials becomes computationally expensive as molecular complexity increases.
While quantum computers offer a potential solution, current Noisy Intermediate-Scale Quantum (NISQ) devices struggle with the mathematical overhead of standard algorithms. This research explores an efficient mapping of the Morse potential onto quantum architectures, utilizing a novel approach to find excited states without the typical computational burden.
Methodology
The researchers employed a hybrid approach combining classical optimization with quantum state preparation:
- Variational Quantum Eigensolver (VQE): A NISQ-friendly algorithm where a quantum computer calculates the energy of a parameterized "guess" state (ansatz), and a classical optimizer iteratively tweaks parameters to find the minimum energy (the ground state).
- Discretization Bases: To map the continuous potential to qubits, three frameworks were compared: the Oscillator Basis (using harmonic oscillator math), the Position Basis (a grid-like approach), and the Finite Difference Basis. The Oscillator basis proved most effective for representing operators as sparse matrices, simplifying the mapping to Pauli strings.
- Implementation: Simulations were performed using IBM’s QISKit framework, typically utilizing 4 qubits for diatomic molecules and 8 qubits for triatomic models.
Supersymmetric Quantum Mechanics (SUSYQM)
A key innovation of this work is the use of Supersymmetric Quantum Mechanics to solve for excited states. In SUSYQM, every Hamiltonian has a "partner." The ground state of a "plus" partner Hamiltonian corresponds exactly to the first excited state of the "minus" partner.
By calculating a hierarchy of these partner Hamiltonians, the authors can determine multiple excited states using only ground-state VQE searches. This bypasses the difficulty of maintaining orthogonality between states, a common hurdle in quantum excited-state algorithms.
Key Results and Discussion
The team applied these techniques to a suite of diatomic molecules, including $H_2, HCl, LiH, CO, O_2,$ and $N_2$, using physical parameters derived from experimental spectroscopic data.
- High Precision: For the $H_2$ molecule, the VQE calculated a ground state energy of -4.468 eV, showing excellent agreement with the experimental value of -4.476 eV.
- Hierarchy Validation: In test cases, the VQE successfully recovered the first two energy levels of the Morse potential using the partner Hamiltonian method, confirming the viability of the SUSYQM approach.
- Triatomic Scaling: The method successfully scaled to a 2D potential for triatomic molecules using 8 qubits. Despite the complexity—the triatomic Hamiltonian required 8,808 Pauli terms—the VQE converged to within 0.5% of the exact theoretical energy.
- Optimizer Sensitivity: The choice of classical optimizer was found to be critical. While L-BFGS-B and SLSQP provided high accuracy, others like SPSA (often preferred for noisy hardware) struggled with the precision required for these simulations.
Conclusion
This work demonstrates that combining VQE with the SUSYQM hierarchy provides a powerful path for simulating realistic molecular potentials on near-term quantum hardware. While current results rely on ideal simulators, the transition to real, noisy hardware will require advanced error mitigation to handle the "Pauli explosion" seen in polyatomic models. Future research may integrate this effective Hamiltonian approach with ab-initio electronic structure calculations to create holistic simulations of complex molecular dynamics.
Summary updated