A locality bound for quantum fermions moving in continuous space

Why locality is difficult without a lattice

To predict what happens in one part of a quantum system, must physicists track everything everywhere? Locality bounds help explain when distant regions can be neglected, with a controlled error.

A Lieb–Robinson bound limits the spreading of initially local quantum operations. In familiar lattice settings, it can define an approximate light cone: outside an expanding region, influence is small, though not necessarily zero. This is a property of the many-body model, not the light-speed limit of relativity.

Continuous space presents a particular obstacle. With ordinary nonrelativistic kinetic energy, particles can have arbitrarily large momentum and group velocity. Methods that exploit bounded propagation speeds on a lattice therefore do not transfer directly.

In a preprint submitted on 8 November 2026, Marius Lemm, Hailun Lu and Harman Preet Singh present a mathematical proof addressing this obstacle. They describe it as the first Lieb–Robinson bound for a standard one-dimensional continuum fermion system. Earlier continuum results, which they discuss, required other restrictions or different kinetic models. This is theoretical work, not an experiment.

The system and its conditions

The particles move along a one-dimensional ring, a geometry that avoids physical endpoints. They are fermions, whose quantum states obey the Pauli exclusion principle: each one-particle mode can be occupied at most once.

Their Hamiltonian combines ordinary kinetic energy with pair interactions. The interaction potential must be smooth, nonnegative, symmetric under reversing separation, and zero beyond a fixed distance. The theorem therefore does not directly cover arbitrary interactions, including long-range Coulomb forces or singular contact potentials.

The initial state also matters. The bound requires finite moments of local energy and particle number. Moments describe more than averages: higher moments constrain the weight of unusually large local values. The authors give examples satisfying suitable uniform bounds, including free-fermion thermal states. These are admissible starting states, not a claim that every interacting equilibrium state meets the assumptions.

What the bound actually says

The theorem controls a mathematical measure of the connection between two initially separated local operations after one evolves in time. For fermion creation and annihilation operators, this involves an anticommutator; the extension to polynomial expressions uses the corresponding parity-dependent commutator.

Crucially, the size is measured with a norm weighted by the initial quantum state. It is not a worst-case bound independent of the state.

For any chosen positive integer power, the estimate decreases as that inverse power of separation and increases exponentially with elapsed time. Stronger spatial decay requires higher initial moments and smoother localized functions. One cannot increase the decay power without also strengthening those requirements.

The constants are independent of the ring's size. When the required initial local moments remain uniformly bounded as the system grows, the estimate remains useful without acquiring a growing system-size penalty. That is the relevant sense of thermodynamic stability here.

However, this is not a fixed-speed light cone. Keeping the error estimate small generally requires a separation growing exponentially with time. This expanding region is a guarantee supplied by the proof, not evidence that actual disturbances spread exponentially fast.

Turning continuous space into a proof tool

The authors divide the ring into overlapping intervals called pseudosites. Smooth localization maps the continuum problem into an auxiliary lattice-like description while preserving the relevant physical dynamics. This is a mathematical construction, not an assumption that particles really occupy discrete sites.

Each pseudosite still contains infinitely many modes. The central challenge is controlling the high-energy ones rather than simply discarding them.

The proof introduces momentum-adapted localization observables: quantities that monitor energy scales differently depending on distance. The authors show that these quantities grow at most exponentially in time. They then control the error from truncating high modes and removing distant couplings, balancing the two approximations to obtain the locality bound.

What remains open

A decisive estimate uses a particularly favorable one-dimensional consequence of Pauli exclusion. Extending the framework to higher dimensions or bosons remains an ambition, not an established result.

The authors also postpone applications to future work. The preprint does not demonstrate a new simulation algorithm or establish new results about quantum phases. Its contribution is a locality theorem that could support such investigations under its stated assumptions.