Conditional generative models enable targeted exploration of MAX phase design space

Jamie Swaine, Cyprien Bone, Prakriti Kayastha, Matthew T. Darby, Ewan Galloway, Keith T. Butler · 2026

Read Conditional generative models enable targeted exploration of MAX phase design space by Jamie Swaine, Cyprien Bone, Prakriti Kayastha, Matthew T. Darby, Ewan Galloway, Keith T. Butler (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13584821.txt.

Description

Jamie Swaine and colleagues teach a language-style AI model to invent new layered materials in this 2026 study. The materials are MAX phases, hard ceramic-metal mixes that can be peeled into thin sheets called MXenes with many uses. The possible recipes number in the thousands, so testing each one in a lab would take far too long.

Their model was tuned on more than six thousand known recipes with two metal sites. Given target settings for sheet-forming promise and bond strength, it produced new recipes at double the rate of untargeted runs. Ten fresh recipes were checked with detailed physics calculations, and five held up as stable or nearly stable. The authors present the method as a scalable way to aim invention at the parts of a design space that matter most.

Catalog record

CREATORJamie Swaine, Cyprien Bone, Prakriti Kayastha, Matthew T. Darby, Ewan Galloway, Keith T. Butler
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50647
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13584821

Browse the full library · About the Tower of Babel