Erratum: Solving Quantum Ground-State Problems with Nuclear Magnetic Resonance

Zhaokai Li, Man‐Hong Yung, Hongwei Chen, Dawei Lu, James Whitfield, Xinhua Peng, Alán Aspuru‐Guzik, Jiangfeng DuScientific 2012 · DOI · OpenAlex

Abstract

Quantum ground-state problems represent a significant computational challenge for many-body Hamiltonians, with no known efficient classical or quantum algorithms for general cases. However, when a trial wavefunction approximating the ground state is available—a common occurrence in physics and chemistry—quantum computers can utilize the Phase Estimation Algorithm (PEA) to project the exact ground state. This study presents an experimental realization of this concept using a variational-wavefunction approach on a Nuclear Magnetic Resonance (NMR) quantum simulator to solve the ground-state problem of the Heisenberg spin model. The iterative PEA achieved an energy accuracy of $10^{-5}$, a ground-state fidelity exceeding 80%, and accurately captured singlet-to-triplet switching near the critical field. These results demonstrate that quantum simulators can effectively leverage classical trial wavefunctions to outperform classical computational limits.

Introduction

Determining the ground state—the configuration of lowest energy—of a many-body quantum system is fundamental to understanding chemical reactions and material properties. As systems grow in complexity, the computational resources required to solve these problems classically scale exponentially. While researchers often possess "best guess" trial wavefunctions, these approximations are frequently insufficient for high-precision applications. This research explores a hybrid quantum-classical methodology to refine these approximations into precise physical states.

Methodology: The NMR Quantum Simulator

The researchers implemented a hybrid approach combining classical preparation with quantum refinement on an NMR platform.

1. Classical State Preparation

The process begins with a variational trial wavefunction, a mathematical approximation of the ground state for the Heisenberg spin model. This model is a standard benchmark in condensed matter physics for studying magnetism.

2. Quantum Projection via Phase Estimation

To refine the trial state, the team employed an iterative version of the Phase Estimation Algorithm (PEA). PEA is a critical quantum algorithm designed to measure the eigenvalues (energies) of a system's Hamiltonian. By applying this to the trial wavefunction, the simulator "distills" the exact ground state and its corresponding energy level.

3. Experimental Hardware

The experiment utilized an NMR quantum simulator, where the nuclear spins of molecules serve as qubits. These qubits are precisely manipulated using radio-frequency pulses, allowing for high-fidelity control over the quantum state.

Key Results and Physical Insights

The experimental results validated the efficacy of the hybrid approach across several metrics:

  • Exceptional Precision: The iterative PEA reached an energy accuracy of $10^{-5}$. This level of precision is vital for quantum chemistry, where "chemical accuracy" is required to predict reaction pathways accurately.
  • High Fidelity: The distilled ground state achieved a fidelity of over 80% compared to the theoretical ideal.
  • Modeling Phase Transitions: The simulator successfully captured the singlet-to-triplet switching behavior near critical magnetic fields. This ability to model physical phase transitions demonstrates the simulator's utility in exploring complex regimes of condensed matter physics.

Discussion and Future Outlook

The study highlights a significant synergy between classical and quantum computing. Rather than requiring a quantum computer to solve a problem from a blank slate, leveraging classical trial wavefunctions allows quantum simulators to bypass initial state-preparation complexities.

However, certain limitations remain. The efficiency of the algorithm is dependent on the "overlap" between the initial classical guess and the true ground state; a poor initial guess significantly hinders the distillation process. Furthermore, while NMR provides high precision for small qubit counts, scaling to the hundreds or thousands of qubits required for broad quantum advantage remains a technical challenge due to signal-to-noise constraints.

Despite these hurdles, this work provides a robust blueprint for future quantum simulators—such as those based on trapped ions or superconducting circuits—to tackle increasingly complex Hamiltonians in physics and chemistry.

Summary updated

Abstract

Quantum ground-state problems are computationally hard problems for general many-body Hamiltonians; there is no classical or quantum algorithm known to be able to solve them efficiently. Nevertheless, if a trial wavefunction approximating the ground state is available, as often happens for many problems in physics and chemistry, a quantum computer could employ this trial wavefunction to project the ground state by means of the phase estimation algorithm (PEA). We performed an experimental realization of this idea by implementing a variational-wavefunction approach to solve the ground-state problem of the Heisenberg spin model with an NMR quantum simulator. Our iterative phase estimation procedure yields a high accuracy for the eigenenergies (to the 10−5 decimal digit). The ground-state fidelity was distilled to be more than 80% and the singlet-to-triplet switching near the critical field is reliably captured. This result shows that quantum simulators can better leverage classical trial wave functions than classical computers
BibTeX
@inproceedings{li2012erratum,
  author = {Zhaokai Li and Man‐Hong Yung and Hongwei Chen and Dawei Lu and James Whitfield and Xinhua Peng and Alán Aspuru‐Guzik and Jiangfeng Du},
  title = {{Erratum: Solving Quantum Ground-State Problems with Nuclear Magnetic Resonance}},
  booktitle = {Scientific},
  year = {2012},
  doi = {10.1038/srep00911},
}

Loading related content...

Figure