Due to its long discovery cycle and capital commitment, materials discovery has been put on the shelf, as we often see materials innovation losing the battle against financial engineering (M&As in particular). In recent years, materials informatics (MI) has been evolving into the next secret weapon for materials discovery.
As Lux closely tracks the progress in this space, we clearly see three strategies for a chemical and materials company to engage.
1. Work with a startup
Citrine Informatics is an MI startup that Lux has rated very highly. It would make a good partner for any chemical and materials company looking to strengthen its MI talent pool and standard data infrastructure. The downside, however, is having to share your data with Citrine, and its current business model requires at least a three-year commitment with a low-figure millions payment per year. Other startups in MI still operate on a contract-based model, but are, in general, limited to certain materials and applications. Uncountable, for example, mainly focuses on polymer-based formulation development, while Lumiant focuses on metal alloy development, although both companies claim their platform is material-agnostic.
2. Work with your peers
The National Institute for Materials Science (NIMS) of Japan is a good example of the consortium model in action. While there isn’t much information disclosed, we assume it involves some precompetitive agreements, and no confidential polymer data is involved at this stage. If the current model works well, the NIMS consortium might later evolve to the next level, where more data gets involved, making the most use of MI. This model works better for midsized chemical and materials companies, as they don’t individually have the core competency to take on the top largest chemical and materials companies.
3. Start an internal MI initiative
Mitsubishi Chemical Holdings (MCH) is at the forefront of this approach, establishing the Materials Informatics Center of Excellence. This is a good model for a large company that has lots of in-house data, can build its own MI team, and can standardize data infrastructure. This model keeps data always from outside companies, unlike strategy 1, and avoids the messy IP issue companies adopting strategy 2 face. However, this model may require a long learning curve before seeing significant returns and will not work well for smaller companies that do not have much data.
The three strategies presented here are by no means exhaustive – there are many others. For example, the consortium hosted by Worcester Polytechnic Institute (WPI) is similar to strategy 2, but at a smaller scale. Also, based on our conversations with clients, there are other models that involve multiple parties. While we understand that the proposed three strategies could face challenges and issues, we do think the fear of such challenges and issues cannot be the reason to not engage with MI. Accelerating materials discovery requires a multidisciplinary and international effort that brings together a wide variety of individuals working effectively across their specialties as well as cross-sector and political boundaries.
MI sits at the very center of materials innovation, and, because of its datacentric nature, whoever invests in it first will gain a first-mover advantage.
To learn more about the latest in materials informatics, join us and other experts from the National Institute of Standards and Technology (NIST), and the Center for Hierarchical Materials Design (ChiMaD), for a full day event on March 27th at Northwestern University in Chicago. Click here to register.
- Tech Page: Materials Informatics (for Lux members only)
- Press Release: Despite $6.1 Billion in Funding, China Not Leading in AI Innovation