Quantum simulator watches a string snap, revealing a new way particles are born

Deep inside every proton, quarks are held together by a force so strange that pulling them apart does not set them free. Instead, the energy stored in the bond between them grows and grows until it becomes cheaper for the vacuum itself to spawn a new quark-antiquark pair, snapping the connection in two. This phenomenon, known as string breaking, is thought to govern how matter fragments in high-energy particle collisions and even how the early universe cooled after the Big Bang. Yet despite its fundamental importance, the real-time dynamics of this process have remained stubbornly beyond the reach of classical computers. Now, a team of physicists has watched string breaking unfold frame by frame in a quantum simulator, and in doing so they have uncovered a mechanism that nobody had expected.

The experiment, published in Nature Physics by Arinjoy De, Alessio Lerose, Christopher Monroe and colleagues at institutions including Duke University, the University of Maryland, KU Leuven and Caltech, used a programmable trapped-ion quantum simulator to emulate a simplified one-dimensional gauge theory. Rather than tackling the full complexity of quantum chromodynamics, the theory of the strong force that binds quarks, the researchers modeled a (1+1)-dimensional lattice gauge theory with a discrete Z2 symmetry. This stripped-down version preserves the essential physics of confinement while remaining tractable in the laboratory, and it can be mapped exactly onto a chain of quantum spins whose collective behavior encodes the interplay between charges and the strings that bind them.

The mapping is elegant. In the dual spin description, a charge in the original gauge theory appears as a kink, a pair of neighboring spins pointing in opposite directions. The electric field of the gauge theory corresponds to the local spin polarization, and an electric-field string is simply a domain of down-pointing spins. The couplings between spins control the particle mass and short-range interactions, a longitudinal magnetic field sets the string tension, the energy stored per unit length of the string, and a transverse field introduces the quantum fluctuations that allow strings to break and charges to be created. By tuning these parameters independently, the team could dial the system through regimes where confinement dominates and where quantum fluctuations tear strings apart.

The experimental platform consisted of thirteen ytterbium-171 ions confined in a microfabricated surface trap, with each ion encoding one spin in two internal energy levels. What set this experiment apart was its unprecedented degree of control. Two overlapping arrays of tightly focused laser beams, each with independently programmable amplitude, phase and frequency, allowed the researchers to engineer long-range spin-spin interactions that decay exponentially with distance, while simultaneously applying a different magnetic field to every individual ion. This site-resolved control proved to be the crucial ingredient, because it enabled a clever trick: the finite chain of thirteen ions could be made to behave as if it were embedded in an infinite system, with semi-infinite static regions extending to either side.

These virtual environments were essential for posing physically meaningful questions. To study an isolated charge, something that can never be observed directly in a particle collider because isolated quarks do not exist in nature, the team emulated a semi-infinite static string on one side of the simulated region and a semi-infinite vacuum on the other. The charge trapped between them could then be watched as it evolved. With no string tension, the charge spread ballistically across the lattice at a velocity that matched theoretical predictions precisely. But when the researchers turned up the string tension, something remarkable happened: the charge stopped spreading and began to oscillate coherently around its starting position, a phenomenon analogous to Bloch oscillations, in which an electron in a crystal lattice subjected to a constant electric field bounces back and forth rather than accelerating indefinitely.

This localization arises because the string tension imparts a constant acceleration to the charge, but on a lattice the momentum cannot grow without bound. The momentum cycles periodically, and the charge oscillates with a spatial amplitude and temporal period that the team found agreed with theoretical predictions even well outside the perturbative regime where those predictions were derived. The result demonstrates that the primary effect of confinement is to halt the spreading of charges, pinning string endpoints in place, a picture that had been proposed theoretically but never before observed with full spatiotemporal resolution.

The centerpiece of the study, however, was the direct observation of string breaking itself. The researchers prepared a classical string stretched between two static charges, then abruptly increased both the string tension and the quantum fluctuations. Conventional wisdom, dating back to Julian Schwinger’s work in the 1950s, holds that string breaking proceeds through the uniform, spontaneous creation of charge pairs throughout the bulk of the string, with a rate that depends exponentially and extremely sensitively on the system parameters. What the team observed instead was strikingly different. Charge pairs appeared preferentially at the edges of the string, near the static charges, and only then spread inward toward the middle.

This edge-facilitated mechanism makes intuitive sense once one considers the energy landscape. Creating a charge pair costs the least energy at the string edges, where the adjacent vacuum partially offsets the energetic penalty. The researchers developed a perturbative theory showing that, after the sudden quench, the dynamics reduce to a two-body quantum problem in which a single charge pair, born at one edge or the other, propagates through an effective potential landscape. As the string tension increases, equipotential channels open up in this landscape, allowing the pair to quantum-mechanically tunnel and diffuse from the edges into the bulk. Remarkably, this simple picture captured the essential features of the experimental data even at parameter values far beyond where the theory should be quantitatively reliable, and additional numerical simulations confirmed that the mechanism persists when the surrounding vacuum is allowed to fluctuate dynamically rather than being held static.

The team also verified that the observed dynamics were genuinely transient, preceding the eventual thermalization of the system. By comparing the measured electric-field profiles with those expected in thermal equilibrium, computed from a Gibbs ensemble at the temperature fixed by the injected energy, they showed that the string character of the state persisted throughout the observation window, dissolving faster when quantum fluctuations were stronger but never fully equilibrating during the experiment. A refined observable, the time-dependent probability distribution of the largest vacuum bubble within the chain, provided further non-perturbative confirmation of the edge-driven breaking picture across the full range of parameters explored.

The implications extend well beyond this particular model. The strategy of emulating probe charges and virtual environments through site-dependent control should be portable to other quantum simulation platforms, including two-dimensional ion crystals and neutral-atom arrays, and could be combined with existing proposals for simulating non-Abelian and higher-dimensional gauge theories that are closer to the Standard Model. Future experiments with receding pairs of probe charges, mimicking the fragmentation of a stretched string in a heavy-ion collision, could determine whether the edge-facilitated mechanism observed here plays a role in real particle physics. For now, the work marks a substantial step toward the long-sought goal of first-principles quantum simulations of string fragmentation and hadronization, processes that classical computers may never fully unravel but that quantum machines are increasingly poised to reveal.

Subject of Research: Real-time observation of string breaking in a one-dimensional lattice gauge theory using a programmable trapped-ion quantum simulator

Article Title: String-breaking dynamics in a quantum simulator

Article References: De, A., Lerose, A., Luo, D., Surace, F. M., Schuckert, A., Bennewitz, E. R., Ware, B., Morong, W., Collins, K. S., Davoudi, Z., Gorshkov, A. V., Katz, O., & Monroe, C. (2026). String-breaking dynamics in a quantum simulator. Nature Physics. https://doi.org/10.1038/s41567-026-03422-0

Image Credits: AI Generated

DOI: 10.1038/s41567-026-03422-0

Keywords: string breaking, quantum simulation, trapped ions, lattice gauge theory, confinement, quantum chromodynamics, Schwinger mechanism, charge dynamics, quarks, gauge theories, quantum many-body physics, Nature Physics