A unique space for
scientific discovery

A common goal

The Institut Courtois is a unique space for scientific discovery where the latest theoretical and experimental approaches in material properties research are brought together and leading researchers in chemistry, physics and computer science collaborate to build on recent advances in artificial intelligence (AI) in order to accelerate the discovery, development and characterization of new materials. The Institute aims to mobilize artificial intelligence at each stage of discovery: for simulation of possible molecules and of the test properties, robotic synthesis and characterization, automated analysis of the generated data, automatic feedback on simulations, and subsequent syntheses.

 

An innovative approach

A new generation of researchers is working at the Institut Courtois to meet the most complex challenges facing our society, such as discovering new materials that can help us achieve our environmental goals or migrating from existing computers to quantum computing. Our team will deploy state-of-the-art methods to accelerate the discovery process. An innovative approach is being applied to the development of new materials, including new tools for organic and inorganic synthesis such as plasma-assisted approaches and AI tools including high-speed synthesis.

The intersection of materials research and AI offers immense potential for accelerating the development and characterization of materials that our society needs.

Laboratories

The Institute’s research teams have access to several laboratories equipped for specific needs and areas of expertise, with the required technical, robotic and high-performance computing support. As high-performance computing capabilities are key, super-calculators that can process complex problems and manage high volumes of data are available to researchers at all times.

Accelerating 3D research

One of the objectives of the Institut Courtois is to accelerate research in additive manufacturing of objects with complex 3D architectures presenting functional properties. Combinations of compounds, 3D architecture and additive manufacturing conditions are studied to identify designs that endow an object with the desired properties. AI resources are essential for rapid exploration of these vast possibilities.

Exploring quantum properties of matter 

One of the Institute’s research ambitions is experimental investigation of the physical processes that govern measurement of quantum states and, ultimately, understanding them. The Institut Courtois’ alternative approach consists in using the unique properties of semiconductor materials to better understand the systems governed by quantum mechanics, and more generally to explore the quantum properties of matter. Among other things, this work will support the development of quantum materials such as superconductors.

The progress

Characterizing and modeling quantum states requires enormous resources. Progress in AI offers new approaches to addressing this problem, notably through:

Knowledge amplification: it is possible to train a network of artificial neurons with known materials and use it to predict new materials that are potentially more effective. It is also possible to simulate small-scale systems and to use this network to extrapolate larger and therefore more realistic systems. 

Learning from emergence: machine learning is applied to enable the algorithm to learn from the emerging properties of the state under study and thereby develop more effective and powerful theoretical models. 

Quantum computing: quantum computers operate through quantum entanglement and runs tasks in parallel, whereas a conventional processor must perform them in series.  

 

In addition to the properties of matter, the scientists at the Institut Courtois are striving to better understand the structures that will govern the next chapter in the development of AI. As our comprehension of complex systems and materials evolves, thanks in part to progress in AI, this new knowledge can be used to improve the AI algorithms themselves, and the improved algorithms can in turn serve to accelerate materials research.  

Understanding matter not only opens up new horizons for basic research but also facilitates the development of protocols for manipulating complex states to discover new materials.