London-based AI startup Polaron announced on February 3 it has raised $8 million USD in funding to build the “intelligence layer for materials science.” Impact-focused VC Racine2 led the round, with participation from Speedinvest and Futurepresent.
Imperial College London spun out Polaron in 2023, co-founded by CEO Isaac Squires, CTO Steve Kench, and Chief Scientist Sam Cooper. This followed seven years of academic research within AI and materials science.
The organization plans to use its latest funding to expand its engineering team, accelerating rollout of generative design tools to better support growing demand from customers in the automotive and energy sectors.
So far, Polaron claims that its technology – which uses AI to characterise advanced materials – has already been used by engineers at global manufacturing leaders, including an electric vehicle maker responsible for over a third of the world’s EV production.
Automating materials management
The funding comes just after the UK Department for Science, Innovation and Technology unveiled a strategic partnership with Google DeepMind, which would establish a research lab in the UK where AI and robotics would be used to synthesise and characterise hundreds of materials per day.
Polaron’s funding reflects a bottleneck in materials manufacturing, where processing materials typically involves the use of a researcher using microscopy to analyse structures and gain insights into strength, potential lifetime, and likelihood of failure. This is a time-consuming manual process, which slows down innovation.
To address this inefficiency, Polaron has trained AI models on real microscope images and measured properties to help interpret object microstructure, explaining why materials behave as they do. The solution automates material characterisation so that engineers can produce high-performance systems faster, offering 3D reconstructions of materials from 2D images alongside rapid identification of complex microstructural features.
“Polaron addresses a fundamental problem in materials and manufacturing: industry still struggles to understand how manufacturing processes shape material performance. Small changes in processing can lead to large differences in outcomes, particularly in energy systems like batteries,” said Squires while in conversation with 150sec.
“Today, these relationships are often explored through trial and error and expert intuition, which is slow, expensive, and difficult to scale. Polaron makes these links explicit and usable, allowing engineers to improve performance more systematically.”
Polaron’s design layer emphasises generative design. For instance, using learned process-structure-property relationships, the system explores the design space to identify optimal material configurations and the processing conditions required to achieve them for materials including metals, ceramics, polymers, and composites.
In some cases, Polaron can reconstruct 3D structural information from standard 2D images, reducing the need for the former.
The global materials market
Research published by Markets and Markets estimates that the global material informatics market will grow from $170.4 million USD in 2025 to $410.4 million USD by 2030, at a compound annual growth rate of 19.2%.
The study attributes the growth in the market to the increasing use of AI in the discovery and development of materials, with vendors such as Google DeepMind, Microsoft Azure, and Orbital Materials using the technology to characterise or discover materials.
Squires argues that Polaron is differentiated from other providers in the market due to its emphasis on microstructure, using AI to learn from real microscopy images and measured performance data rather than academic modelling; it is a solution designed for industrial use.
Off the back of its new funding, Squires says the company expects to double its current team of nine employees by the end of 2026, hiring across key areas including AI research, software engineering, and materials science.
Featured image: Courtesy of Polaron.