Quantum Hardware Enables Complex NMR Spectroscopy Simulation

Simulating the complex behaviour of molecules remains a significant challenge for modern computing, yet understanding molecular structure is crucial for advances in materials science and drug discovery. Artemiy Burov, Julien Baglio, Clément Javerzac-Galy, and colleagues demonstrate a breakthrough in simulating one-dimensional nuclear magnetic resonance (NMR) spectra, a key technique for determining molecular structure, using noisy intermediate-scale quantum (NISQ) hardware.

The team successfully simulated the behaviour of spin systems up to 34 spins, exceeding the practical limits of classical methods, by combining advanced error mitigation techniques with commercially available quantum computers. This achievement represents a major step towards realising the potential of quantum computing for practical applications in NMR spectroscopy and opens new avenues for exploring complex molecular systems. Quantum Simulation of NMR Free Induction Decay Scientists are making significant strides in simulating nuclear magnetic resonance (NMR) spectra using quantum computers, accurately modeling systems with up to 34 nuclear spins, surpassing the limitations of conventional computational methods.

This research demonstrates the potential of quantum computing to tackle complex molecular structures, crucial for advancements in chemistry and materials science, by simulating the Free Induction Decay (FID) signal, the raw data acquired in NMR experiments.

The team developed a sophisticated pipeline incorporating error suppression and mitigation techniques to enhance the accuracy of these quantum simulations, and compared results obtained from different quantum hardware platforms. The researchers studied several molecular systems, including a simple phosphorus molecule and more complex phosphorus clusters, to benchmark the performance of their simulations. They utilized both the IonQ Forte Enterprise, which employs trapped-ion qubits, and the IBM Aachen system, based on superconducting qubits, to compare the results and assess the impact of different hardware architectures.

The team determined that 700 computational steps provided optimal accuracy for the IonQ system, demonstrating a balance between simulation depth and noise accumulation. Results show that the IonQ Forte Enterprise produced smoother spectra with fewer errors compared to the IBM Aachen system, suggesting potential advantages in qubit coherence and control. The quantum simulations, incorporating error suppression techniques, closely matched classical simulations, achieving high agreement in terms of both mean squared error and cosine similarity. This demonstrates that quantum hardware noise is becoming less of a limiting factor, and algorithmic improvements are now key to achieving higher fidelity in these simulations.

Quantum Simulation Exceeds Classical NMR Limits Scientists have achieved a breakthrough in simulating NMR spectra using quantum computers, successfully modeling systems with up to 34 nuclear spins, exceeding the practical limit of 32 spins for conventional computational methods. This work demonstrates the potential to tackle complex molecular structures, crucial for designing new materials and pharmaceuticals, using quantum approaches.

The team employed advanced error suppression techniques, developed by Q-CTRL, in conjunction with both superconducting qubits from IBM and trapped-ion qubits from IonQ, to mitigate noise and improve the accuracy of simulations. Experiments revealed a substantial improvement in simulation accuracy, with a reduction in mean square error of up to 22times on a 22-spin system when utilizing the Q-CTRL Fire Opal error suppression pipeline. This enhancement enabled the recovery of important spectral features that would otherwise be obscured by noise in near-term quantum devices. Researchers performed simulations using the 156-qubit IBM Aachen superconducting computer and the IonQ Forte Enterprise trapped-ion computer, allowing for a cross-platform comparison of performance and error metrics.

The team utilized a single computational step per time point in the simulation, achieving circuit depths of up to 250 after optimization on the IBM Heron hardware. Data shows that the simulations accurately reproduce the salient features of 1D NMR spectra, as validated by comparison with classical simulations from SPINACH. By employing a pure transverse magnetization state and a simplified Hamiltonian decomposition, they were able to model the dynamics of complex molecular systems with unprecedented accuracy. This breakthrough delivers a crucial step towards near-term utility in NMR spectroscopy, offering a powerful new tool for molecular analysis and materials discovery. Simulating 34-Spin NMR Spectra on Quantum Hardware This research demonstrates a significant advance in quantum computing’s potential for practical applications, specifically in the field of nuclear magnetic resonance (NMR) spectroscopy. Scientists successfully simulated the Hamiltonian of liquid-state one-dimensional NMR spectra for systems containing up to 34 spins, exceeding the practical limitations of classical computational methods. This achievement stems from the development of an advanced error suppression and mitigation pipeline, combined with the use of commercially available superconducting and trapped-ion quantum computers.

The team’s approach, utilizing a single computational step per time point and the Q-CTRL Fire Opal tool, substantially reduced the impact of quantum noise, achieving up to a 22-fold reduction in mean-square error compared to noiseless simulations. Results obtained on both IBM and IonQ quantum hardware demonstrate the feasibility of executing deep circuits, enabling the accurate simulation of complex molecular systems. Notably, the IonQ results suggest a further reduction in noise and the avoidance of unwanted artifacts in the resulting spectra. Future research will explore incorporating open-system interactions into the quantum simulations, potentially leveraging quantum thermodynamics concepts to further enhance accuracy and realism. This work represents a crucial step toward realizing near-term utility for quantum computing in NMR spectroscopy and potentially other areas of molecular simulation. 👉 More information 🗞 Large circuit execution for NMR spectroscopy simulation on NISQ quantum hardware 🧠 ArXiv: https://arxiv.org/abs/2512.14513 Tags: