Altrove Uses AI and Lab Automation to Create New Materials

In recent years, the development of new materials has been accelerating, and a new French startup named Altrove is ready to make a significant impact in this innovation cycle. This deep tech startup has already raised €3.7 million (around $4 million at current exchange rates) to fuel its ambitions.

A Slow Journey to Material Breakthroughs

For decades, finding new materials has been a slow and challenging process. Thibaud Martin, the co-founder and CEO of Altrove, explained to TechCrunch that there have been many bottlenecks in research and development over the past 50 years. One major issue has been predicting whether new materials made from a combination of elements can theoretically exist.

When combining two different chemical elements, there are tens of thousands of possibilities. Adding a third element increases the combinations to hundreds of thousands, and with four elements, the possibilities reach into the millions.

The Role of AI in Material Prediction

Teams from companies like DeepMind, Microsoft, Meta, and Orbital Materials have been using artificial intelligence models to overcome these calculation constraints. These AI models help predict new materials that could potentially exist in a stable state. “More stable materials have been predicted in the last nine months than in the previous 49 years,” Martin said.

However, knowing that new materials can exist isn’t enough. Creating these materials requires a precise recipe, which involves not just the elements but also their proportions, the temperature, the order of mixing, and the duration of the process. There are many factors and variables involved in creating new materials.

Focusing on Rare Earth Elements

Altrove is currently focusing on inorganic materials, particularly rare earth elements. These elements are challenging to source, their prices vary greatly, and they often come from China. Many companies are trying to reduce their reliance on China to avoid regulatory uncertainties, creating a market opportunity for Altrove.

Automating the Material Creation Process

Altrove does not invent new materials from scratch. Instead, it selects interesting candidates from the new materials predicted by AI models and uses its own AI models to generate potential recipes for these materials. The company tests these recipes one by one, producing a tiny sample of each material. They then use their proprietary characterization technology, which involves an X-ray diffractometer, to understand if the material performs as expected.

“It sounds simple, but checking what you’ve made and understanding why is very complicated,” Martin said. Often, the result is not exactly what was initially intended.

The Importance of Characterization

This is where Altrove shines. The company’s co-founder and CTO, Joonatan Laulainen, has a PhD in materials science and expertise in characterization. The startup owns intellectual property related to characterization, which is crucial for improving recipes and developing new materials. Altrove aims to automate its lab to test more recipes simultaneously and speed up the feedback loop.

“We want to build the first high throughput methodology,” Martin said. Pure prediction only takes you 30% of the way to creating a material that can be used industrially. The other 70% involves real-life iteration, which is why an automated lab is so important. It increases throughput and allows for parallel experiments.

Looking to the Future

Altrove defines itself as a hardware-enabled AI company. It plans to sell licenses for its newly produced materials or manufacture those materials with third-party partners. The company raised €3.7 million in a funding round led by Contrarian Ventures, with participation from Emblem and several business angels, including Thomas Clozel (CEO of Owkin), Julien Chaumond (CTO of Hugging Face), and Nikolaj Deichmann (founder of 3Shape).

The startup draws inspiration from biotech companies that use AI to discover new drugs and treatments, but Altrove applies this approach to new materials. The company plans to build its automated lab by the end of the year and sell its first asset within 18 months.