run quick, approximate experiments alongside slower, more accurate ones. BASF and Imperial researchers developed machine learning models that bridge this gap and better accommodate the particular experimental practices used in chemical R&D. The innovative nature of this research won the prestigious Computers & Chemical Engineering (CACE) Best Paper Award 2023.
Another example where innovative R&D using AI and machine learning is Flue2Chem, a first-of-its kind collaboration to design and validate a UK value chain to convert valuable industrial carbon emissions (both fossil and non-fossil) into sustainable materials for consumer products. BASF was proud to be part of this consortium of organisations and as part of this work, was able to demonstrate how captured CO2
can be converted to
ethanol - a versatile building block used in surfactants, cosmetics, pharmaceuticals and industrial applications.
At BASF’s global R&D hub in Ludwigshafen, researchers developed a new catalyst for turning captured CO2
to improve the material properties of the catalysts more efficiently. The catalysts were also tested under standard processing conditions to reflect how they would perform in real-world conditions. This is a critically important aspect of the work of Flue2Chem, as we need to embed these innovations into existing chemical value and supply chains, so they have greater potential to be scaled up quickly and efficiently. The ethanol was then further processed by Flue2Chem partners to make surfactants which will be used to make consumer goods.
Looking ahead and hydrogen into ethanol. AI and
machine learning were central to this effort. BASF optimised the catalytic conversion process to produce ethanol more sustainably and effectively. By integrating cutting-edge research and development techniques, along with AI and machine learning, we have accelerated the development of new catalysts. This innovative process enables us to conduct multiple tests simultaneously while using predictive models
Our experience at BASF demonstrates that from accelerating molecule discovery to improving plant reliability, optimising procurement, and enabling new more sustainable production processes, AI reduces time-to-market and strengthens resilience. In other words, done in the right way, AI can deliver both commercial advantage and sustainability outcomes. The challenge ahead is to ensure that new AI developments can be translated into the real world, for example, embedding them in the chemical production processes of the future. This will require sustained efforts from researchers, across companies like BASF and academic institutions, but also the right regulatory and funding environment to trial and test these in pilots and beyond.
basf.com
Figure 1: Snapshot overview of BASF’s contribution to the Flue2Chem project. For more information about Flue2Chem, please visit
https://www.soci.org/flue2chem LUBE MAGAZINE AR TIFICIAL INTELLIGENCE DECEMBER 2025 21
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28
