Duke Team's 13-Ion Quantum Simulator Recreates String-Breaking

A team led by the Duke Quantum Center encoded a model of 'string-breaking' — the process by which stretched connections between quarks snap and release energy as new particles — into a chain of 13 trapped ytterbium ions, according to research published in Nature Physics on September 23, 2026.⁵⁶³

Why quarks are hard to study

According to the researchers, quarks serve as matter's most basic components, sized at roughly one-billionth the scale of an atom, and remain permanently hidden inside larger particles, making direct observation impossible.⁵²⁶³

String-breaking, the phenomenon in which two connected quarks stretch apart and build up energy before snapping to create new particles, normally requires energy levels found only in extreme environments such as the Large Hadron Collider or immediately after the big bang, according to the dossier sources.⁵²⁶

According to NBC News, quantum chromodynamics is the theory that explains this behaviour through color confinement, and isolated quarks have never been observed; calculating the dynamics of quantum chromodynamics directly is described as extremely difficult for classical computers.¹

Inside the experiment

According to La Brújula Verde and Quantum Zeitgeist, the Duke team built a programmable array of 13 trapped ytterbium-171 ions, flanked by two extra ions used for trapping purposes, and used finely tuned laser pulses to adjust how the ions interacted with one another, replicating the way a string stretches and eventually snaps.³⁴

The system was prepared in an out-of-equilibrium state and its evolution was tracked, with researchers observing effective charges appearing at the edges of the simulated string and spreading inward, a pattern the sources describe as differing from what the Schwinger mechanism would predict.⁵¹³

NBC News reported that forming a charge pair near an edge requires less energy given the vacuum properties there, a phenomenon the outlet characterized as an atypical edge-driven route to particle pair creation.¹

The researchers additionally replicated the process using a conventional computer, verifying the accuracy of the quantum outcomes, although NBC News notes that today's classical machines remain capable of duplicating this specific 13-ion setup.⁵²⁶¹

A benchmark across platforms

Google and QuEra Computing independently recreated string-breaking using different quantum computing platforms — Google with superconducting circuits and QuEra with neutral atoms — providing a comparison across what Christopher Monroe called the three leading platforms in quantum computing.⁵²¹

La Brújula Verde reported that Google's team relied on superconducting qubits arranged in two dimensions to model a gauge theory spanning two spatial dimensions plus time, whereas QuEra positioned dozens of rubidium atoms in a kagome-lattice configuration; Duke's ion-chain setup, meanwhile, operated within just one spatial dimension.³

Per La Brújula Verde and NBC News, Duke's model represents a streamlined gauge theory confined to a single spatial dimension plus time, falling short of the complete three-dimensional, non-Abelian quantum chromodynamics that actually governs quarks; more elaborate theories incorporating extra dimensions remain beyond current simulation capabilities.³¹

Looking ahead

Monroe said quantum computer simulations offer the best available platform for probing questions such as matter formation short of directly witnessing the big bang, adding that the findings mark a development in quantum science that opens new avenues for understanding string-breaking dynamics.²

As problem size grows, the sources state, only quantum computers will be able to solve these kinds of problems, since the computational demands exceed the capabilities of conventional machines; NBC News reports that quantum machines could eventually be used to tackle real-time high-energy physics problems.⁵²⁶⁴

Partners on the project included the University of Maryland, Oxford University, Caltech, Cornell University and KU Leuven, with funding provided by the Department of Energy, the National Science Foundation, the Air Force Office of Scientific Research, DARPA and Amazon Web Services.⁵