UC San Diego Shaped a Laser Beam to Switch Magnetism Nine Layers Deep

Researchers at the University of California San Diego have demonstrated optical switching of magnetic states in a nine-layer platinum-cobalt structure, a material three times thicker than anything previous experiments could handle. The team claims the technique could operate more than 1,000 times faster than conventional magnetic-field writing. The work was published in Nature Communications on October 27, 2026 (DOI: 10.1038/s41467-026-77572-3).

What did the engineers actually build?

The team, led by Abdoulaye Ndao, professor of electrical and computer engineering at the UC San Diego Jacobs School of Engineering, did not design a new magnetic material. Instead, they reshaped and shrank the beam of a specialized ultrafast laser. Earlier experiments, by the researchers' own account, worked only on structures with no more than three magnetic layers and required light with a specific polarization.

"Past experiments had found that increasing the material's thickness beyond three layers suppressed optical switching," Ndao said. "That put a constraint on the overall thickness of the magnetic material and thus, its ability to retain memory in the long term."

The new result removes both limits. Switching worked in nine alternating platinum and cobalt layers, and it worked regardless of polarization.

How does the switching work?

The method concentrates laser energy into an extremely small region and fires a sequence of ultrafast pulses. The first pulses briefly heat a localized spot, causing a small region to reverse its magnetic orientation. Later pulses enlarge that reversed region until the new state stabilizes.

Study first author Muhammad Waleed Khalid, a Ph.D. student in Ndao's group, framed the advance as an exercise in optical engineering. "We've optically engineered the light to change the physics that's happening in the material at the micro- and nanoscale," Khalid said.

The beam reduction itself is dramatic. The researchers describe a decrease of tens of orders of magnitude relative to earlier experiments. That claim covers the beam-shaping demonstration, not a shipping memory device, and the paper offers no write-speed measurement against commercial flash or MRAM.

"The smaller the size of the beam, the smaller and more dense the optical memory," Ndao said.

Why did peer acceptance take effort?

The results deviated sharply from prior literature. The group needed repeated, extensive verification before the optics community accepted the findings.

"The effects we were seeing were so new and unusual that we had a difficult time convincing others in the field that our discovery was not a one-time fluke," Khalid said. "We spent a lot of time and effort repeating and verifying our experiments to substantiate our work to the optics community."

The project paired Ndao's optics team with Eric Fullerton, an expert in thin-film magnetic materials. Fullerton holds appointments in electrical and computer engineering and in chemical and nano engineering, and he serves as Endowed Chair Professor of UC San Diego's Center for Memory and Recording Research. The collaboration attacked an established memory problem from the light side rather than the material side.

What stands between the lab and a product?

Two obstacles are explicit in the announcement. First, the specialized ultrafast laser used in the experiments does not integrate into computer chips in its current form. Second, the beam still needs to shrink further; the team's stated target is a few hundred nanometers, using optical structures that concentrate light into progressively smaller regions.

The researchers are also screening for magnetic materials that respond to laser systems more suitable for on-chip integration. No timeline, power figure, or cost estimate accompanies any of this. The paper's co-authors include Sheena K. K. Patel, Koffi-Emanuel Sadzi, Mohammed Salah El Hadri, Sergio A. Montoya, Stéphane Mangin, Eric E. Fullerton, and Abdoulaye Ndao.

The 1,000x speed claim is a projection, not a measured benchmark. But if beam-shaping can push switching into thicker, retention-friendly structures at a few hundred nanometers, the technique could reposition optical memory from a physics curiosity into a credible path toward compact, high-density, low-energy storage.