Beyond Materials. Beyond Limits.
Embedded Files
Platform for AI-driven inverse design of metamaterials
Platform for AI-driven inverse design of metamaterials
The 20th century saw incredible advancement in physics, while the 21st century became the century of digital services and AI.
At Matterdyne, we aim to drive a new era of advancement in physical and materials science with AI.
We are building a platform and tooling to design metamaterials with customized physical properties on demand.
Between February and May 2025, Matterdyne took part in the Oxford University Innovation Startup Incubator Phase 1 programme
Metamaterials
Metamaterials
Microscopic structures, such as repeating units of 3D-printed cells, can possess exotic physical properties differing from their underlying material.
Well-known examples include extreme stiffness at low density, negative refractive index, and negative Poisson ratio ("auxetic" metamaterials).
Metamaterials can be built to custom designs to have unique mechanical, thermodynamic, optical, acoustic and electromagnetic properties.
Metasurfaces
Metasurfaces
Engineered surfaces with small (e.g. nanoscale) structures can have exotic physical properties.
Well-known examples optical metasurfaces for flat lenses, anti-reflection coatings, low aerodynamic drag ("sharkskin") surfaces, and even antibacterial metasurfaces.
At Matterdyne, we are developing software tools to design these structures to have customer-requested physical properties.
Physics-based "forward" simulation
Physics-based "forward" simulation
Armed with knowledge of physics and properties of base materials, we write and test physics-based simulators for metamaterials.
This is the "forward" simulation problem, and it is well-defined. The "inverse design problem", i.e. calculating the correct structure to fit the required physical properties, to a traditional physicist, is an "ill-posed" problem, since it can have one, zero, or many possible solutions.
"Classically", this is a hard problem. We can leverage AI to solve it.
Generative AI for "inverse design"
Generative AI for "inverse design"
Machine learning is an approach to AI that tunes the parameters of expressive mathematical models to fit a dataset, rather than to hard-code the logic like a conventional computer program.
Machine learning allows us to use datasets produced by forward simulation of many metamaterials to perform the reverse process (inverse design).
Generative AI allows us to rapidly iterate with many different suggestions to meet the requirements. Solutions are validated first in simulation, then in the real world, saving time and lowering R&D costs.
Page updated
Report abuse