Physicists at the University of Cambridge have experimentally created the first two-dimensional “Bose glass” — a strange state of matter in which particles become trapped rather than spreading through the system. The experiment challenges some of our familiar assumptions about how physical systems reach equilibrium.

Matter is usually introduced to us through familiar categories.

Solid. Liquid. Gas. Plasma.

But at extremely low temperatures and on quantum scales, matter can organise itself in ways that have no obvious equivalent in everyday life.

Physicists at the University of Cambridge’s Cavendish Laboratory have now experimentally realised one of those exotic states in two dimensions for the first time: the Bose glass.

A glass made from quantum particles

The name is deceptive.

A Bose glass is not glass in the conventional sense, and it is not something that could be held in your hand.

It is a quantum state predicted to occur when interacting bosonic particles are subjected to disorder.

One of its defining characteristics is localisation.

The particles become confined instead of freely spreading throughout the system. The Bose glass is therefore insulating, but unlike some other insulating quantum states, it remains compressible and lacks long-range phase coherence.

Cambridge researchers use an everyday analogy to explain the strange behaviour.

Imagine adding milk to coffee.

Normally, stirring causes the milk and coffee to mix until the liquid becomes uniformly coloured.

If the ingredients behaved like particles in a localised system, however, the intricate light and dark pattern could persist rather than gradually mixing away.

Building matter with light

To create the Bose glass, the researchers began with approximately 120,000 ultracold potassium-39 atoms.

They then used laser beams to create an artificial landscape in which those atoms could move.

Four independent optical lattices were superimposed at 45-degree angles, producing an eight-fold symmetric quasicrystalline optical lattice.

A quasicrystal occupies an intriguing territory between order and disorder.

An ordinary crystal contains a structure that repeats periodically.

A quasicrystal possesses long-range order, but its pattern does not repeat periodically.

The Cambridge researchers compare the resulting geometry to a Penrose tiling — a mathematical pattern that can continue indefinitely without becoming conventionally repetitive.

This unusual landscape provided the environment needed for the Bose glass to emerge.

From superfluid to Bose glass

The researchers could control the properties of their atomic system.

In one regime, the atoms formed a superfluid.

A superfluid is another extraordinary quantum state in which particles behave collectively and can flow without ordinary viscosity.

But as the researchers changed the strength of the optical lattice and the interactions between the atoms, the behaviour changed.

The atoms became increasingly localised.

Eventually the system entered the Bose-glass phase.

Crucially, the Cambridge team was able to directly observe the transition between the Bose glass and the superfluid and map the phase diagram experimentally.

A system that struggles to forget its past

The experiment revealed something deeper.

In conventional statistical mechanics, a system that is changed sufficiently slowly can generally remain close to equilibrium.

This idea underpins the concept of an adiabatic process.

But the Bose glass behaved differently.

The researchers attempted to transform the system slowly from the non-periodic quasicrystalline lattice into a conventional periodic lattice.

When starting from the superfluid state, the system could adapt rapidly.

When starting from the Bose glass, however, restoring coherence proved much more difficult. Significant entropy increases remained even when the transformation was performed slowly.

This behaviour is consistent with the expected non-ergodic character of the Bose glass.

In simplified terms, the system can retain information about where it has been rather than rapidly exploring all the configurations available to it.

Why this challenges statistical mechanics

Statistical mechanics is extraordinarily successful at describing how huge collections of particles behave.

One of its central ideas is that, given enough time, interacting systems tend towards thermal equilibrium.

But localisation can interfere with that process.

Particles may effectively become trapped, preventing the system from exploring all of its possible states in the conventional way.

That is why Bose glasses are scientifically interesting.

They provide physicists with a controlled environment in which to investigate where familiar assumptions about equilibrium and thermalisation begin to fail.

A quantum system too large for ordinary computers

There is another reason the experiment matters.

Large interacting quantum systems rapidly become extraordinarily difficult to simulate using classical computers.

The number of possible quantum configurations grows enormously as more particles are added.

Professor Ulrich Schneider, who led the Cambridge research, pointed to this as one of the experiment’s major advantages: researchers now have a real two-dimensional quantum system whose dynamics and statistics can be studied directly rather than relying entirely on computer modelling.

In effect, the experiment itself becomes a kind of quantum simulator.

Scientists construct a highly controlled physical system and then allow nature to perform the quantum evolution.

A bridge to many-body localisation

The Bose glass may also help researchers investigate another major problem in modern physics: many-body localisation.

In many-body systems, particles interact with one another.

Normally those interactions help energy spread through the system and eventually produce thermal equilibrium.

But physicists have investigated circumstances in which interacting particles may remain localised instead.

The Cambridge results suggest that the Bose glass provides an experimental route for studying connections between localisation, glass-like dynamics and possible many-body-localised quantum states.

Matter beyond everyday intuition

Nothing in ordinary experience prepares us particularly well for the quantum world.

A table remains where it is placed.

Milk mixes with coffee.

Heat spreads from warmer regions to colder ones.

These behaviours seem so obvious that they can appear almost inevitable.

Quantum physics repeatedly demonstrates that they are not.

Under carefully engineered conditions, matter can enter states in which particles remain localised, coherence disappears, equilibrium becomes difficult to reach and the statistical rules familiar from ordinary materials begin to require a much more subtle description.

The Bose glass is one of those states.

For decades it was largely the territory of theoretical physics and lower-dimensional or related experimental systems.

Now Cambridge physicists have produced and directly studied its two-dimensional form.

And in doing so, they have created a laboratory in which some of the deepest assumptions about how large quantum systems evolve can be tested experimentally.

Source: Based on “Discovery of a new phase of matter in 2D which defies normal statistical mechanics,” published by the University of Cambridge’s Cavendish Laboratory on 11 September 2024. The underlying open-access study, “Observing the two-dimensional Bose glass in an optical quasicrystal,” by Jr-Chiun Yu, Shaurya Bhave, Lee Reeve, Bo Song and Ulrich Schneider, was published in Nature, volume 633, pages 338–343, on the same date.