Quantum simulating continuum field theories with large-spin lattice models

Gabriele Calliari, Marco Di Liberto, Hannes Pichler, Torsten V. Zache · 2024

Read Quantum simulating continuum field theories with large-spin lattice models by Gabriele Calliari, Marco Di Liberto, Hannes Pichler, Torsten V. Zache (2024) 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-2412-15325.txt.

Introduction

Gabriele Calliari and co-authors show how large-spin lattice models can stand in for smooth quantum field theories on a quantum simulator. Their 2024 paper tackles a twin cut problem: space must be chopped, and each site's state space must be trimmed too. Multi-level atoms supply the trim.

The team builds lattice Hamiltonians whose large-spin limit lands on the target field theory. Heavy numerical runs walk the extrapolation ladder for a famous wave model and land on the known answers. Lone wave packets, still and moving, were built and crashed into each other on cue. A tweak that breaks the model's special balance sparked string-like breaking and plasma ringing, echoing gauge theory behavior.

Catalog record

CREATORGabriele Calliari, Marco Di Liberto, Hannes Pichler, Torsten V. Zache
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE135 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-15325

Browse the full library · About the Tower of Babel

Full text

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

Quantum simulating continuum field theories with large-spin lattice models Gabriele Calliari, Marco Di Liberto, Hannes Pichler, Torsten V. Zache Abstract Simulating the real-time dynamics of quantum field theories (QFTs) is one of the most promising applications of quantum simulators. Regularizing a bosonic QFT for quantum simulation purposes typically involves a truncation in Hilbert space in addition to a discretization of space. Here, we discuss how to perform such a regularization of scalar QFTs using multi-level or qudit systems, and show that this enables quantitative predictions in the continuum limit by extrapolating results obtained for large-spin lattice models. With extensive matrix-product state simulations, we numerically demonstrate the sequence of extrapolations that leads to quantitative agreement of observables for the integrable sine-Gordon (sG) QFT. We further show how to prepare static and moving soliton excitations, and analyze their scattering dynamics, in agreement with a semi-classical model and analytical predictions. Finally, we illustrate how a non-integrable perturbation of the sG model gives rise to dynamics reminiscent of string breaking and plasma oscillations in gauge theories. Our methods are directly applicable in state-of-the-art analog quantum simulators, opening the door to implementing a wide variety of scalar field theories and tackling long-standing questions in non-equilibrium QFT like the fate of the false vacuum. I Introduction Quantum field theories (QFTs) play an essential role in theoretical physics, ranging from effective low-energy descriptions in condensed matter Sachdev 2023 to fundamental models of nature in particle physics Weinberg 1995 . Understanding the physics of QFTs, especially in out-of-equilibrium scenarios, is often guided by computationally challenging numerical simulations, and therefore QFTs constitute an ideal target for quantum simulation Bauer et al. 2023 . Recent quantum simulation experiments are starting to probe the boundary of classical simulations Daley et al. 2022 ; Scholl et al. 2021 ; Joshi et al. 2023 ; Andersen et al. 2024 ; Manovitz et al. 2024 , with the exciting prospect to provide genuine insights into the equilibrium and non-equilibrium properties of quantum many-body systems. However, it remains an outstanding challenge to faithfully simulate physical phenomena of QFTs, such as string-breaking in gauge theories Cochran et al. 2024 ; Gonzalez-Cuadra et al. 2024 ; De et al. 2024 ; Crippa et al. 2024 . QFTs are typically formulated in terms of continuous variables living on continuous space-time – in contrast to, e.g., many-body spin models on spatial lattices. While Lagrangian space-time discretizations are well established in numerical lattice field theory Montvay and Münster 1994 , digital quantum simulations are typically based on a Hamiltonian formulation with finite-dimensional Hilbert spaces and thus require truncations of bosonic field variables Jordan et al. 2012 ; Tong et al. 2022 . In this context, it appears natural to employ multi-level systems as realized with trapped ions Ringbauer et al. 2022 or proposed with Rydberg atoms González-Cuadra et al. 2022 ; Kruckenhauser et al. 2022 . Alternatively, QFTs can also arise as low-energy effective field theories in analog quantum simulations Tajik et al. 2023 ; Viermann et al. 2022 . However, the question how to perform controlled truncations of bosonic field variables in an analog approach has remained largely unexplored. Figure 1: Overview. (a) d -dimensional square lattice of particles with nearest-neighbor interactions. We assume each particle to be described by a large spin of length J , with \langle\hat{J}_{+}\rangle\sim\langle e^{i\hat{\varphi}}\rangle and \langle\hat{J}_{z}\rangle\sim\langle\hat{\pi}\rangle identifying field theory and lattice operators, as described in Sec. II . Continuum physics is extracted by a sequence of extrapolations in spin length J , system size N and lattice spacing a [see Fig. 2 ]. (b) Sketch of real-time scattering dynamics from large-spin models. We show how to faithfully implement 1. the vacuum (Sec. III ), 2. quasi-particle wave-packets and 3. their propagation (Sec. IV ), as well as 4. collisions (Sec. V resp. VI ), resulting in phase shifts resp. particle production for the example of an integrable resp. non-integrable 1D QFT. In this work, we address this question and show how to quantitatively recover the continuum limit of scalar quantum field theories for an explicit truncation of the bosonic field variables. Motivated by a recent proposal to employ large-spin Rydberg atom arrays Kruckenhauser et al. 2022 , our approach is based on a generalized large-spin Heisenberg Hamiltonian, which makes it immediately applicable to analog quantum simulations. We show how appropriate extrapolations [illustrated in Fig. 1 (a)] enable the accurate implementation of general scalar field theories, including variants of the paradigmatic sine-Gordon (sG) model. To demonstrate our proposal, we perform extensive benchmark simulations for the 1D sG model employing tensor-network methods. In equilibrium, we find quantitative agreement with analytical predictions for ground and excited state correlation functions in the continuum limit. Out of equilibrium, we discuss how to create solitonic excitations and find that their spreading induced by quantum fluctuations is well described by a phenomenological semi-classical model. We further simulate the non-equilibrium dynamics of scattering pairs of soliton wave-packets, in agreement with predictions from the known S-matrix. Finally, we consider a perturbed sG model, where a quantum simulator would provide access to a spatially and temporally resolved picture of scattering dynamics in regimes that potentially lie beyond the reach of traditional methods. Before delving into the details, the following Section II provides an overview of our approach, including a discussion of its broad applicability and relevance for emulating condensed matter and high-energy physics. We also provide a brief summary of our main results, together with the structure of the rest of this paper, which is also illustrated in Fig. 1 . II Scalar field theories from large-spin models Our target are d -dimensional QFTs described by the Hamiltonian \displaystyle\hat{H}_{V}=\int{\rm d}^{d}x\left\{\frac{1}{2}\left[\hat{\pi}(x)\right]^{2}+\frac{1}{2}\left[\boldsymbol{\nabla}\hat{\varphi}(x)\right]^{2}+V(\hat{\varphi}(x))\right\}\;, (1) defined in terms of the continuum field operators \hat{\varphi}(x) and their conjugate momentum \hat{\pi}(x) which fulfill the canonical commutation relation [\hat{\varphi}(x),\hat{\pi}(y)]=i\delta(x-y) . Different models are distinguished by the form of self-interactions as described by the potential V . Below in Sec. II.1 , we introduce our prototypical large-spin lattice model that will be used as a regularization of \hat{H}_{V} . We then discuss the theoretical sequence of limits that leads from this microscopic model to scalar quantum field theories in Sec. II.2 . Sec. II.3 elaborates on the range of accessible QFTs and interactions, where we also comment on their relevance. In Sec. II.4 , we summarize our main results, pointing the interested reader to the corresponding subsequent sections. II.1 Large-spin lattice models Throughout this manuscript, we study quantum many-body models whose degrees of freedom are large spins of length J . More precisely, our microscopic regularization of scalar QFTs is based on a generalized Heisenberg lattice Hamiltonian of the form \displaystyle\hat{H}_{\text{latt}} \displaystyle=\sum_{i}\left[\Omega\hat{J}_{x}^{(i)}-\Delta\hat{J}_{z}^{(i)}+\chi\left(\hat{J}_{z}^{(i)}\right)^{2}\right] \displaystyle\qquad-\sum_{i,\kappa}\left[\lambda_{\kappa}\left(\hat{J}_{+}^{(i)}\right)^{\kappa}+\mathrm{H.c.}\right] (2) \displaystyle\qquad+\sum_{\langle ij\rangle}V_{\text{nn}}\left[\hat{J}_{z}^{(i)}\hat{J}_{z}^{(j)}-\frac{1}{4}\left(\hat{J}_{+}^{(i)}\hat{J}_{-}^{(j)}+\mathrm{H.c.}\right)\right]\;, with \hat{J}_{x/y/z}^{(i)} spin- J operators and \hat{J}_{\pm}^{(i)}=\hat{J}_{x}^{(i)}\pm i\hat{J}_{y}^{(i)} the corresponding ladder operators, i.e., they obey SU(2) commutation relations \left[\hat{J}_{x}^{(i)},\hat{J}_{y}^{(i)}\right]=i\hat{J}_{z}^{(i)} and have a fixed length J(J+1)\hat{1}^{(i)}=\sum_{\alpha=x,y,z}\left(\hat{J}_{\alpha}^{(i)}\right)^{2} . We assume the spins to be located on the sites i of a regular (square) lattice in d spatial dimensions. In Eq. ( II.1 ), we combine several terms that arise naturally for spin- J models, including (i) terms linear in single spins, such as transverse (\propto\Omega) and longitudinal fields (\propto\Delta) , as well as direct raising/lowering operators (\propto\lambda_{\kappa} for \kappa=1) ; (ii) single-spin non-linearities, in particular one-axis twisting

Read the complete text