In January 2024, a team led by Prof. Alex Greilich at TU Dortmund University reported something that sounded like a contradiction: a time crystal that kept ticking inside an ordinary piece of semiconductor for hours, with no external clock driving it. Time crystals are not supposed to need winding. They repeat in a regular rhythm on their own, a property so odd that the phrase itself sounds like a physicist’s joke. Now, in a study published in Nature Communications, Greilich and his colleagues have gone further. They have shown that more than one of these crystals can form inside the same chip and, under the right conditions, fall into step with each other, even when they sit tens of micrometers apart with nothing obvious connecting them.
That distance matters. Forty micrometers is roughly one thousand times larger than the size of a single oscillator in this system, according to the findings reported by www.sciencedaily.com. For two independent quantum-scale clocks to notice each other and adjust across a gap that large is not a trivial result. It suggests a form of coupling that does not depend on the oscillators sitting right next to each other, which is usually how synchronization works.
The chip has been quietly running a clock without a battery
The material at the center of the experiment is gallium arsenide doped with small amounts of indium and silicon. Those dopants trap individual electrons at specific points in the lattice, and each trapped electron sits surrounded by roughly one million nuclear spins, the tiny magnetic moments belonging to the atomic nuclei nearby. Cooled to near absolute zero, about minus 270 degrees Celsius, the material becomes quiet enough that these spin interactions stop being noise and start being signal.
The researchers use a pump laser to align the electron spins in one direction. Those electrons then hand off their polarization to the surrounding nuclear spins, like a current passing through a crowd. A weak magnetic field is applied next, and the nuclear spins begin to precess, rotating steadily around that field. Left alone, that rotation would decay quickly. But feedback between the electron spins and the nuclear spins they polarized keeps reinforcing the motion, so the oscillation persists far longer than it has any right to. A second laser lets the team watch the whole process unfold without disturbing it.
A clock that keeps ticking without being wound is not supposed to exist.
Every patch of the crystal ticks slightly differently, until it doesn’t
No real semiconductor is perfectly uniform. Strain, impurity density, and local composition all vary slightly from one microscopic region to the next, and those variations mean that a time crystal forming in one patch of the material will naturally oscillate at a slightly different frequency than one forming in a neighboring patch. Each region, left to its own devices, keeps its own time.
The Dortmund team found that this independence breaks down under a broad illumination pattern. When the pump laser is widened to cover many regions of the crystal at once, rather than pinpointing a single spot, the separate oscillations stop drifting apart. They lock together and settle into one shared frequency, as if the whole illuminated area had agreed on a single beat.
Distant clocks in the same crystal started keeping the same time.
Huygens saw this in 1665, just not in a semiconductor
The phenomenon has a well-known ancestor. In 1665, Christiaan Huygens noticed that two pendulum clocks mounted on the same wooden beam would gradually synchronize, their swings falling into a shared rhythm because the beam itself transmitted faint mechanical vibrations between them. It was one of the earliest documented cases of coupled oscillators finding common ground through an indirect, almost accidental channel.
The semiconductor version has no beam and no mechanical vibration. According to the researchers, the link between separated time crystals runs through the movement of spin-polarized electrons drifting through the material, carrying information about one region’s oscillation phase into another. It is a electronic echo of Huygens’s shared support, occurring at a scale and temperature that would have been unimaginable in the 17th century.
Where the coupling breaks, and why that boundary matters
Synchronization in this system is not unlimited. The Dortmund group found that once two oscillating regions are separated by more than roughly 40 micrometers, they stop locking together and drift back into independent rhythms. That cutoff gives physicists a concrete number to work with, a boundary that any theory of the underlying coupling mechanism will eventually need to explain.
The practical interest here has less to do with clocks and more to do with control. A material that can host multiple independent spin oscillators, and can be coaxed into synchronizing or desynchronizing them on demand, starts to look like the raw material for a network. Spin-based devices already underpin parts of quantum information research, and a demonstrated, tunable form of long-range coupling between spin ensembles gives that field a new lever to pull. www.sciencedaily.com reports that the team frames the result as a step toward controllable networks of spin oscillators, though turning a cooled gallium arsenide chip into a practical device remains a separate and much harder problem.
What stays open
Nobody in this study claims to fully explain why electron-mediated coupling reaches as far as it does, or why 40 micrometers is the point where it gives out. The result documents a real, measured effect: distant, independently formed time crystals inside one semiconductor can be made to tick in unison, and then, past a certain distance, they simply stop listening to each other. That gap between “it happens” and “here is exactly how” is where the next round of experiments at TU Dortmund will likely aim.
Source: www.sciencedaily.com
Credit: User Untitled on Unsplash