Pressing a polymer composite between a roller and an ultrasonic horn vibrating 15,000 times a second produced a paper-thin sheet conducting 12.67 watts per metre-kelvin in plane.

Heat is the constraint on folding electronics. A rigid device can shed heat through a metal chassis or a vapour chamber; a device meant to bend cannot use either, and polymers — which bend well — conduct heat roughly a thousand times worse than aluminium.

The standard fix is to fill the polymer with flakes of something conductive, such as boron nitride or graphite, and the standard problem is that the flakes land in random orientations. Heat travels along a flake far better than across it, so a random mesh wastes most of the conductivity it contains.

A study in the International Journal of Extreme Manufacturing reports a way to align them in about a second, using ultrasound and a roller.

The authors are Daiwei Hu, Yucheng Duan, Lingxiao Cao, Haoxuan Dong, Tian Shen, Xindong Ni, Luanxin Zhang, Xudong Zhang and Zhizhu He, with He of China Agricultural University listed as the expert contact. The release does not state the other authors’ affiliations or define its author-note symbols.

The process

The method is called rolled sonication-enabled microstructure orientation, abbreviated RSMO, and it couples ultrasonic vibration with roller pressing. An ultrasonic horn vibrating 15,000 times per second acts on the composite as a roller compresses it, and the film forms in about one second.

Why the combination works is reasonably clear from the materials physics. Roller pressing alone applies compression, which flattens a film and partially aligns flakes by geometry; ultrasonic vibration at the same time fluidises the polymer locally and repeatedly breaks the friction holding flakes in place, so they can rotate into the plane rather than being locked wherever they started. Pressing gives the direction, sonication gives the mobility.

The throughput figure is the one with industrial meaning: roughly 36 square metres of film per hour, which the release puts at about 100 times faster than standard industrial hot-pressing.

Hot pressing achieves alignment by holding a composite under pressure and heat for minutes to hours, and the time is the cost — both in energy and in how much capital a production line ties up per square metre. A one-second step changes the economics of the operation rather than its physics.

What the film does

The resulting film conducts heat at 12.67 watts per metre-kelvin along its surface and is about four times thinner than a human hair, which puts it in the low tens of micrometres.

For context, an unfilled polymer sits around 0.2 watts per metre-kelvin and copper around 400. A figure of 12.67 is therefore a large multiple of the base polymer and a small fraction of a metal — which is the right comparison for a heat spreader, whose job is to move heat sideways away from a hot spot rather than to act as a heat sink.

The mechanical result is the one that justifies the whole exercise: the film can be folded in half or bent over 5,000 times without cracking. A conductive filler loading high enough to carry heat usually makes a composite brittle, and surviving 5,000 folds is what distinguishes this from a stiff thermal pad.

Two device-level demonstrations are reported. In a 24-watt LED, operating temperatures dropped by over 84 kelvin. On a flexible thermoelectric generator, the film doubled the electrical power harvested from body heat.

The thermoelectric result is the more elegant of the two, because it shows the same property working in the opposite direction. A thermoelectric generator produces power from a temperature difference across it, and getting body heat efficiently to one side is precisely a heat-spreading problem — so a better spreader doubles the gradient available to convert.

The 84-kelvin drop on an LED is a striking number and depends entirely on what it was measured against. Eighty-four kelvin is an enormous reduction, which implies the baseline was a 24-watt LED with very poor heat management indeed; the release gives no description of the comparison case.

The reporting problems

Two issues in the release deserve flagging, and the first is a direct inconsistency.

The release’s metadata description states that the material is 19 times stronger and conducts heat 171 per cent more effectively. Neither figure appears in the body of the release, and neither has a stated comparison. Nineteen times stronger than what, and 171 per cent better than what, are unanswerable from the page — and a metadata claim that the article text does not support is the kind of discrepancy a reader should know about rather than reconcile.

The second is that the release contains no direct quotations from any named researcher. Its descriptive passages — that the acoustic treatment “turns a loose fibrous mesh into an ultra-dense and paper-thin sheet that conducts heat” — are the release’s own words, so the framing here follows its account rather than anyone’s stated view.

Nor does it state funding sources or any conflicts of interest, and it gives affiliations for only one of nine authors.

What is untested

The through-plane conductivity is not reported, and for a film cooling a device it matters as much as the in-plane figure. Aligning flakes in the plane necessarily makes the film worse at conducting heat through its thickness — the two trade against each other — and heat has to get into the film from the chip before it can be spread. A heat spreader optimised entirely in plane may need something else to bridge the interface.

Filler identity and loading fraction are also unstated, which makes the result hard to place against the substantial literature on aligned boron nitride and graphite composites. And 5,000 folds, while a real demonstration of durability, is a modest number for a device meant to be opened and closed several times a day over years.

What the work offers is a fast, equipment-light route to an alignment that is otherwise slow to achieve, demonstrated in two devices. The paper was published on 10 August 2026.

---

Notes

1. Hu, D., Duan, Y., et al. “Rolled-sonicating enabled orientation of microstructures in flexible nanocomposites.” International Journal of Extreme Manufacturing, 2026. DOI: 10.1088/2631-7990/ae8a91.
2. International Journal of Extreme Manufacturing. “A single blast of sound solves foldable electronics’ biggest heat problem.” EurekAlert!, 7 October 2026. https://www.eurekalert.org/news-releases/1146988.

Photo by Tofros.com on Pexels.com