Uneven ion spacing changes how a quantum of vibration travels

One excitation, ten possible locations

A phonon is a quantum of vibrational excitation. In this experiment, it describes a small amount of motion perpendicular to a chain of trapped ions. Electric repulsion couples the ions’ vibrations, allowing an excitation prepared at one ion to spread to others. The ions remain near their equilibrium positions; it is the excitation that travels.

Such systems can act as quantum simulators: controllable devices that reproduce the dynamics of a chosen physical model. Here, the researchers studied single-excitation transport, rather than a many-particle system or a quantum computation.

Their arXiv preprint, submitted on November 10, 2026, reports measurements on ten calcium ions held in a linear Paul trap.

Why the gaps matter

The trap confines the chain along its length with a harmonic potential. Combined with the ions’ mutual repulsion, this produces closer spacing near the center and wider gaps near the ends. The measured neighboring-ion separations ranged from approximately 16 micrometers in the center to 22 micrometers at the edges.

Those differences affect transport in two ways. They change how readily an excitation transfers between ions, and they shift each ion’s local vibrational frequency. The coupling falls with the inverse cube of separation, making nearby ions the strongest partners.

A frequency mismatch can suppress transfer even between neighbors. Near the center, coupling is strong enough relative to those mismatches to support substantial neighboring-site exchange. At the edges, the mismatch with the next ion is more restrictive. The starting position therefore selects a different balance between coupling and frequency differences.

Preparing and reading the vibration

The team used ten individually controlled laser beams, one for each ion. After cooling the radial motion close to its ground state, they applied laser pulses to prepare one local vibrational excitation at a selected site. They then switched off the lasers for an adjustable interval, allowing the coupled vibrations to evolve.

To read the result, another set of pulses mapped vibrational excitation onto the ions’ internal electronic states. Fluorescence detection then distinguished those states at each site.

Repeating the sequence at different evolution times built up a picture of the excitation’s distribution. These measurements are not a continuous recording of one phonon following a definite route through the chain.

Three starting points, different patterns

When preparation targeted Ion 1, at an end of the chain, the excitation spread across other sites and partially refocused at Ion 10 after roughly 200 microseconds. Population later refocused at the starting end. Neither concentration was complete: some excitation remained distributed elsewhere.

The authors interpret this behavior through the combination of suppressed neighboring-site exchange at the edges and matching local frequencies at the two ends. It is not evidence that the excitation bypasses the intervening system.

Starting at Ion 3 produced an intermediate pattern, with population splitting among several regions. Starting at Ion 6, near the center, distributed excitation across much of the chain relatively quickly. Simulations showed alternating regions of high and low probability characteristic of a continuous-time quantum walk. A less distinct version appeared in the experimental measurements.

The control pulses are part of the dynamics

Preparation and readout were not instantaneous. The relevant sideband pulses lasted about 35 microseconds, long enough for appreciable hopping to occur during the operations, especially near the chain’s center.

Simulations that included these finite pulse durations generally reproduced the measurements, including early-time features missed by an idealized treatment. This matters because the measured pattern reflects both free evolution and the operations used to prepare and detect it.

Residual differences remain. The authors identify imperfect cooling, preparation, laser control and readout as possible contributors. They measured heating of about 0.029 vibrational quanta per millisecond and argue that heating was not a major limitation over their longest, one-millisecond propagation interval.

The demonstrated dynamics remain within an effectively single-particle model. Multiple excitations and interacting spin–vibration systems are proposed extensions, not results of this experiment. Reaching them would require more demanding preparation and detection, alongside stable trapping, cooling and individual control of larger ion chains.