CuCrP2S6 keeps generating a zero-bias photocurrent above the temperature where its average structure turns symmetric, and the moving copper ions appear to be why.
The bulk photovoltaic effect is the less famous way to turn light into electricity. Rather than separating charges across a junction, as a silicon solar cell does, certain crystals generate a current throughout their volume with no junction at all — which matters because the junction is what imposes the Shockley–Queisser limit of roughly 33 per cent on an ideal single-junction silicon cell.
The catch has always been a symmetry requirement. The effect needs a crystal structure without inversion symmetry, and most candidate materials lose that property as they warm toward room temperature.
A study in Advanced Functional Materials from the Institute of Science Tokyo reports a material that does not behave accordingly. In CuCrP2S6, the photocurrent persists at room temperature even though the average structure there is centrosymmetric — and it is about 1.5 times larger than at low temperature.
Why that should not happen
The symmetry rule is not a rule of thumb but a consequence of how the effect works. If a crystal has a centre of inversion, then for every direction there is an equivalent opposite direction, and a photocurrent has no reason to prefer one — so the net current is zero.
CuCrP2S6 is a van der Waals material, built from layers held together weakly enough to be peeled apart, and at low temperature it is non-centrosymmetric as required. Warming it changes the average structure to centrosymmetric, at which point the effect should disappear.
Instead the team measured a zero-bias photocurrent across the whole range from 15 to 300 kelvin, appearing only along the polar a-axis, and growing rather than fading as the temperature rose.
“Surprisingly, our experiments revealed that a clear zero-bias photocurrent persists at room temperature,” said Takao Sasagawa, associate professor at Science Tokyo, who conducted the work with Ryoga Murata, a former graduate student there.
The gap between averages and instants
The explanation turns on what centrosymmetric means when it is a time average.
Between roughly 145 and 190 kelvin the material enters a dynamically disordered state, in which copper ions move between positions rather than sitting still. A structural measurement that integrates over time sees the average of those positions, which is symmetric; at any given instant the local arrangement is not.
The timescales are what make this exploitable, and the numbers are the crux of the paper. Copper ion motion happens on picosecond timescales or longer — 10-12 seconds — while the electronic response occurs within femtoseconds, 10-15 seconds.
So the electrons are a thousand times faster than the atoms they live among. From the electrons’ point of view the copper ions are stationary, and the local structure they experience is the instantaneous one: polar, non-centrosymmetric, and perfectly capable of producing a photocurrent. The symmetry that forbids the effect exists only in a measurement slower than the physics.
“Dynamically generated polar clusters can contribute to the shift-current response,” Sasagawa said — and the photocurrent pattern the team measured did match shift-current behaviour, which is the specific quantum-mechanical mechanism in which the act of absorbing a photon displaces an electron’s charge distribution in a fixed direction.
That match matters for the interpretation. Shift current is an intrinsic bulk property rather than an artefact of contacts or heating, and the team tested both silver and platinum electrodes, which addresses the most obvious alternative explanation.
A different design rule
The conclusion Sasagawa draws is about materials strategy rather than this particular crystal.
“The work suggests a new materials-design philosophy: rather than suppressing atomic motion,” he said — the implication being that disorder can be recruited instead of eliminated.
This inverts a long-standing assumption. Photovoltaic materials research generally treats atomic motion as loss: phonons scatter carriers, disorder creates trap states, and higher temperatures degrade performance, which is why real solar panels lose efficiency on hot days. A mechanism in which thermal motion generates the very asymmetry the effect requires points at a class of materials that would work better warm.
It also widens the search. If the requirement is instantaneous local polarity rather than a globally non-centrosymmetric average structure, then materials previously excluded by their room-temperature crystallography become candidates — and ion-mobile, dynamically disordered compounds are common.
The distance to a solar cell
The Shockley–Queisser limit is mentioned in the release as the thing a bulk photovoltaic device could in principle exceed, and the gap between principle and practice here is very large.
Bulk photovoltaic materials have been studied for decades and have never been competitive, for a reason the release does not mention: the currents are extremely small. The effect typically generates high voltages and minute photocurrents, which is useful for photodetection and all but useless for power generation, and nothing in the reported result changes that. A photocurrent 1.5 times a small number is a small number.
The release also reports no efficiency figure, no absolute current density, and no comparison against any other bulk photovoltaic material — so whether this crystal is good in the field of bulk photovoltaics, or merely interesting in its temperature behaviour, cannot be judged from it. CuCrP2S6 also contains chromium and copper in a layered sulphide, which is not obviously a material anyone would manufacture at scale.
Read as physics rather than as energy technology, though, the result is clean and the mechanism is specific: a measurable effect that survives the symmetry condition it supposedly requires, explained by a timescale separation that can be checked. That is worth more than another incremental efficiency number.
The paper appeared online on 9 August 2026 and in print on 10 September 2026. The research was funded by the Japan Science and Technology Agency and the Japan Society for the Promotion of Science; the release does not give the full author list or state any conflicts of interest.
Notes
- “Unusual Temperature-Enhanced Bulk Photovoltaic Effect Beyond the Centrosymmetric Phase in CuCrP2S6 van der Waals Crystals.” Advanced Functional Materials, volume 36, issue 73, 2026. DOI: 10.1002/adfm.77517.
- Institute of Science Tokyo. “Atomic motion could help push solar cells beyond conventional limits.” EurekAlert!, 5 October 2026. https://www.eurekalert.org/news-releases/1146480.
Photo by Pixabay on Pexels.com
Leave a Reply