Study of nuclear magnetic resonance spectra with the multi-model multi-level quantum complex exponential least squares algorithm

Antonio Márquez Romero, Josh J.M. Kirsopp, Giuseppe Buonaiuto, Michal Krompiec · 2025

Read Study of nuclear magnetic resonance spectra with the multi-model multi-level quantum complex exponential least squares algorithm by Antonio Márquez Romero, Josh J.M. Kirsopp, Giuseppe Buonaiuto, Michal Krompiec (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-24664.txt.

Introduction

Antonio Marquez Romero, Josh Kirsopp, Giuseppe Buonaiuto, and Michal Krompiec apply a quantum algorithm to nuclear magnetic resonance spectra in a 2025 paper. Magnetic resonance reads the spins inside molecules, and its spectra guide chemical analysis. Their algorithm is an advanced form of quantum phase estimation built for early fault tolerant machines. It handles several signal modes at several levels at once.

In quantum simulations of spin systems, including ones with complex coupling patterns, the method gave sharper phase resolution than standard tools. It pulled out spectral features with up to ten times fewer signal evaluations than the classic Fourier transform approach. The authors frame the work as a bridge between advanced quantum algorithm design and a practical spectroscopy task. It points toward scalable quantum simulation of magnetic resonance systems.

Catalog record

CREATORAntonio Márquez Romero, Josh J.M. Kirsopp, Giuseppe Buonaiuto, Michal Krompiec
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-24664

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Study of nuclear magnetic resonance spectra with the multi-model multi-level quantum complex exponential least squares algorithm Antonio Márquez Romero, Josh J.M. Kirsopp, Giuseppe Buonaiuto, Michal Krompiec Abstract We present a novel application of the multi-modal, multi-level quantum complex exponential least squares (MM-QCELS) algorithm, a state-of-the-art, early fault-tolerant quantum phase estimation (QPE) technique, to the simulation and analysis of nuclear magnetic resonance (NMR) of spin systems. By leveraging the robustness and precision of MM-QCELS, we demonstrate enhanced phase resolution in quantum simulations of spin dynamics, also in systems with complex coupling topologies. Our approach enables accurate extraction of spectral features with up to an order of magnitude fewer evaluations of the time series signal in comparison with the conventional Fourier transform, making a significant step toward scalable quantum simulations of NMR Hamiltonians. This work bridges an advanced quantum algorithm design with a practical spectroscopic application, offering a promising new approach for a quantum-based chemical analysis. I Introduction Nuclear Magnetic Resonance (NMR) is an essential spectroscopic technique widely used in chemistry and material science Levitt 2008 ; Keeler 2011 . It provides insightful information about the molecular structure, dynamics, and interactions between components. NMR is based on the probe of the local chemical environment of nuclei, most commonly single spins of hydrogen ( 1 H) or carbon ( 13 C), under the influence of a strong magnetic field, ideally achieving high spectral resolution. The response of the spin system under such electromagnetic pulses outputs the experimental spectra which we intend to compare with our theoretical descriptions. NMR enables us to determine the connectivity of atoms within a molecule, identify functional groups, and even infer three-dimensional structures. Beyond chemistry, NMR is also used in complex systems such as proteins Wüthrich 1986 ; Wagner 1993 ; Dyson and Wright 2004 , where it can reveal dynamic processes, folding patterns, and intermolecular interactions. Furthermore, its quantitative outputs allows for precise concentration measurements in a mixed solution. Classical simulations of NMR are computationally challenging due to the exponential scaling of the spin Hamiltonian with the system size. Quantum computers are naturally suited for this kind of calculations, with a one-to-one correspondence between the number of spins in the system and the number of qubits in the quantum register. Notably, NMR analysis is considered one of the most promising candidates to demonstrate practical quantum advantage on this early stage of quantum hardware development Knill and Laflamme 1998 ; Seetharam et al. 2023 and interest in the field has grown significantly in recent years Sels et al. 2020 ; Khedri et al. 2024 ; Fratus et al. 2025 ; Ossorio-Castillo and Rodríguez-Coello 2024 ; Burov et al. 2024 ; Das and Merz Jr 2025 . In this work, we propose an application of a recently developed fault-tolerant method based on a single-ancilla quantum phase estimation routine for the simulation of spin-based NMR Hamiltonians. Similar to classical approaches, this application of the method relies on the evaluation of the time evolution of the system and computing the expectation value of spin operators. However, instead of an ad hoc evaluation until the NMR spectra has the desired resolution, our application employs well-defined hyperparameters that provide an accurate estimate of the computational resources needed to resolve the NMR spectrum peaks within the requested resolution. We validate our implementation with molecular examples drawn from previous studies in the field, which also allow for classical-methods benchmarking, and we assess its viability in realistic settings. This paper is organized as follows: in Section II , we introduce the fundamentals of NMR spectroscopy and discuss its treatment using classical methods. Section III presents a summary of the original quantum algorithm and outlines its adaptation to the NMR spectroscopy problem. In Section IV , we detail the quantum circuit implementation of our proposed application. Section V showcases the results of our simulations, and finally, in Section VI we provide our concluding remarks and future prospects. II The NMR spectroscopy Simulation of the NMR technique consists of the description of the local spin system of the molecule as the contribution of one-body and two-body terms using the Heisenberg-like Hamiltonian in one dimension 1 1 1 2D Simpson 2011 and even 3D Spiess 1991 studies of NMR spectroscopy can also be performed \begin{split}H&=\sum_{i}^{N_{\rm{spins}}}\delta_{i}S_{i}^{(z)}+\sum_{i<j}^{N_{\rm{spins}}}J_{ij}(S_{i}^{(x)}S_{j}^{(x)}+S_{i}^{(y)}S_{j}^{(y)}+S_{i}^{(z)}S_{j}^{(z)})\\ &=\frac{1}{2}\sum_{i}^{N_{\rm{spins}}}\delta_{i}Z_{i}+\frac{1}{4}\sum_{i<j}^{N_{\rm{spins}}}J_{ij}(X_{i}X_{j}+Y_{i}Y_{j}+Z_{i}Z_{j}),\end{split} (1) where S_{i}^{(x,y,z)} are the spin operators acting on spin i , which are related to the usual Pauli operators (X,Y,Z) as S^{(x)}=\frac{1}{2}X,\,\,\,\,S^{(y)}=\frac{1}{2}Y,\,\,\,\,S^{(z)}=\frac{1}{2}Z. (2) The parameters of the NMR Hamiltonian are the chemical shifts \delta_{i} , which take into account an averaged effect of the surrounding cloud of the molecular electrons; and the coupling constants J_{ij} , which model the interaction between the different spins at play Smith et al. 1992a ; Smith et al. 1992b . While other interaction terms may be included for a more accurate description, the isotropic Hamiltonian as described in Eq. ( 1 ) is commonly used in most liquid-state NMR experiments. Under the action of a static magnetic field in the Z -axis direction, the individual spins experience a Zeeman effect and the chemical shifts are increased proportionally with the amplitude of such magnetic field. Thus, it is conventional to see in the literature the strength of the magnetic field B multiplying the chemical shifts components of the NMR Hamiltonian ( 1 ). Once the sample is aligned along the Z -axis, a perpendicular electromagnetic pulse is applied, causing the different spins to precess around the Z -axis. Consequently, the collective motion creates a magnetization signal which can be measured. The magnetization M on the perpendicular plane to the Z -axis can be computed as the expectation value of the X- and Y- spin chains with respect to the state of the system |\Psi\rangle : \begin{split}\langle\Psi|M|\Psi\rangle&=\langle\Psi|M_{x}+iM_{y}|\Psi\rangle\\ &=\frac{1}{2}\sum_{k}^{N_{\rm{spins}}}\left(\langle\Psi|X_{k}|\Psi\rangle+i\langle\Psi|Y_{k}|\Psi\rangle\right).\end{split} (3) Finally, if we consider a time evolution of the spin system given by the Hamiltonian ( 1 ) as |\Psi(t)\rangle=e^{-iHt}|\Psi_{0}\rangle, (4) where |\Psi_{0}\rangle is the initial ansatz, the magnetization ( 3 ) is a time-dependent quantity and in the frequency domain corresponds to the NMR spectrum of the molecule. That is, \begin{split}\textrm{NMR}(f)&=\mathcal{F}\left[\langle\Psi(t)|M|\Psi(t)\rangle\right]\\ &=\mathcal{F}\left[\langle\Psi_{0}|e^{iHt}Me^{-iHt}|\Psi_{0}\rangle\right]\end{split} (5) where \mathcal{F} corresponds to the standard Fourier transformation over the positive domain \mathcal{F}[f(t)]=\int_{0}^{\infty}dtf(t)e^{2\pi ift}. (6) In a realistic experimental setting, the interaction of the sample with the environment needs to be taken into account. While a precise description needs the Lindblad formalism, phenomenologically we make the crude approximation of inserting a damping exponential factor in the magnetization signal, evolving in time as \langle\Psi(t)|M|\Psi(t)\rangle=e^{-\eta t}\langle\Psi_{0}|e^{iH_{\rm{NMR}}t}Me^{-iH_{\rm{NMR}}t}|\Psi_{0}\rangle. (7) We refer to this quantity as the Free Induction Decay (FID) signal Keeler 2011 . Figure 1: Standard plot of the magnetization signal. a) Case without attenuation, theoretical description of magnetization according to Eq. ( 3 ) b) Case with an attenuation factor included, corresponding to the FID signal in Eq. ( 7 ) Measured experimental NMR spectra have the form as shown in subplot b) of Fig. 2 , where the peaks follow a Lorentzian lineshape, that is, L(f)=A\frac{\eta}{\eta^{2}-(f-f_{0})^{2}}, (8) where A and f_{0} are the amplitude (height) and the center (position of the height) of the peak, respectively. Thus, the peak amplitudes and centers are obtained by means of lineshape analysis Roy and Kailath 2002 ; Hua and Sarkar 2002 . The centers are directly related to the chemical

Read the complete text