Imperial and BASF spinout company uses AI to improve chemical manufacturing

22 July 2024 | Muriel Cozier

A spinout company launched by Imperial College London and BASF aims to use innovative processing techniques alongside machine learning models to predict the best ways to manufacture high-value chemicals, such as drugs and fertilisers. 

The company, called Solve, uses novel techniques in flow chemistry to build large data sets on chemical reactions. These data sets are used to train machine learning models to rapidly predict the best way to manufacture high-value chemicals. The company has launched with funding from BASF subsidiary Chemovator in a pre-seed round led by venture capital firm Creator Fund.

The technology could help companies to scale manufacturing of new chemicals more quickly and to optimise manufacturing processes. It could also make chemical production more sustainable by helping in the move to more benign materials, as well as reducing waste and energy use. This would lead, in many cases, to reduced set-up and operating costs of production plants. Supply chains could also be made more resilient as producers could have the ability to adapt if there are changes in the availability of their raw materials.

The partners say that Solve's technology could lead to a data-driven transformation of research and development in chemical manufacturing.

"Chemistry needs to become more data driven. Once you do an experiment the data should be stored in detail so you can make use of it later. In future there will still be room for creativity by scientists, but they will be working from a more solid base,’ said Solve CEO and founder Dr Linden Schrecker.

The company is building up experimental data sets using novel techniques in flow chemistry, in which reactions are carried out in a continuous flow rather than in batch vessels. “The data we collect in flow is more accurate because we have a more precisely controlled set of conditions and plenty of control over how those conditions change. It’s also a more efficient way to collect the data,” said Schrecker.

One area where it is anticipated that SOLVE’s technology is set to have an impact is in the choice of solvents. “Solvents often improve reactions but are not in the final product, so they are inherently wasteful and are subject to increasingly heavy regulation. Even in the last decade we’ve seen many commonly used solvents being banned. We can provide clients with extra flexibility to vary how they manufacture a product,” Schrecker added.

SOLVE technology is based on techniques developed by Shrecker as a PhD student at the UK Government backed EPSRC Centre for Doctoral Training in Next Generation Synthesis & Reaction Technology at Imperial.

The transition from laboratory to business has been rapid for Solve, which was formed less than three months after the decision was made to spinout. Typically the transition time is an average of 11 months for businesses in this sector, Solve's backers said. The new company has backing from BASF’s business incubator Chemovator. This development marks the first time in BASF’s 150 year history that it has formed a spinout in partnership with a university.

More innovation news: 

Carbon capture: Using machine learning to cross the 'Valley of Death'

Top 10 emerging technologies of 2024: from cooling solutions to carbon capture

Can the UK meet electric vehicle battery demands? Industry experts discuss

 

Show me news from
All themes
from
All categories
by
All years
search by