Large-scale NMR simulation on a trapped-ion quantum computer

Nuclear Magnetic Resonance (NMR) spectroscopy is a vital tool for chemical analysis, yet simulating the complex spin dynamics of large molecules remains a challenge for classical computers. This paper investigates whether current trapped-ion quantum hardware can accurately simulate the NMR spectrum of 1,2-di-tert-butyl-diphosphane, a 22-spin molecule known to be classically demanding. To make the simulation feasible on the Quantinuum System Model H2, the authors implemented a hardware-efficient reduction of the nuclear-spin Hamiltonian, projecting the phosphorus-spin pair into an effective single-qubit subspace. This reduced the system to 21 effective spins, allowing for Trotterized real-time evolution using circuits with a two-qubit gate depth of up to 1442.

The researchers successfully reconstructed the liquid-state proton NMR spectrum of the molecule. A critical achievement of this work is the reproduction of a distinctive double-peak structure in the 3.9–4.3 ppm range, which was absent in previous quantum hardware attempts. The authors employed a suite of error suppression techniques—including Pauli twirling, dynamical decoupling, and leakage detection—to mitigate device noise, particularly memory errors. Comparison with noiseless emulator simulations confirms that these techniques were essential for preserving the spectroscopic features of interest.

This study represents a significant step toward practical quantum utility in chemical research. By demonstrating that large-scale NMR simulations can be performed with sufficient fidelity to capture non-trivial spectral structures, the authors highlight the potential of quantum computers to eventually outperform classical methods in regimes where spin-spin couplings dominate. The work also provides a clear roadmap for managing hardware-specific noise in complex, long-duration quantum simulations.