Brown engineers erase 'dead layer' in multiferroic films, eyeing low-power AI chips
The technique keeps bismuth ferrite magnetic and ferroelectric down to a few angstroms, researchers report in Nature Materials.
Brown University engineers say they have overcome a long-standing obstacle that has kept a promising class of materials out of computer chips, a step that could help curb the rising energy demands of artificial intelligence.
The team developed a fabrication method for ultra-thin films of bismuth ferrite, a multiferroic material whose electrical and magnetic properties can be controlled together with a very small voltage. The work is published in the journal Nature Materials.
The 'dead layer' problem
Multiferroics have drawn interest because each unit of their atomic lattice is electrically polarized, meaning positive and negative charges align on opposite sides. A small voltage can flip that polarization, offering a way to encode binary data with little electricity — and because the magnetic state is linked to the electrical state, the data can be read back magnetically.
The catch has been the so-called dead layer effect. To fit inside a chip, the material must be made exceptionally thin, but at low thicknesses its atomic lattice distorts and internal electric fields overpower its polarization, causing it to lose both its ferroelectric and magnetic behavior.
The Brown team's fix grows the bismuth ferrite on a substrate of dysprosium scandate, a rare-earth-derived material with a narrower crystalline structure. The resulting lattice mismatch puts the film under compressive strain that helps preserve its structure, while a second metallic layer absorbs stray electric fields.
Thin as a few angstroms
"We've established that we can get rid of the dead layer," said Lucas Caretta, an assistant professor of engineering at Brown who led the work. "We show that you can make this material basically as thin as you want — down to a few angstroms — and it's still ferroelectric, and it's still magnetic."
The researchers reported that the stacked structure not only retains the material's multiferroic properties but also produces enhanced magnetic characteristics that could make it more useful for future electronics. The advance points toward memory and logic components that could operate at far lower power than today's devices, an aim that has gained urgency as AI workloads drive up computing energy use.