The National Growth Fund project, ‘Big Chemistry’, has provided eight consortia with €2.8 million. The projects focus on accelerating the search for chemicals with the desired properties.
Many of the project leaders are familiar faces from C2W | Mens & Molecules. You will find summaries of the projects below, along with links to our stories highlighting the project leaders or their research. Please note that these are separate from the projects listed below.
Biobased Peptide Surfactants via Borrowing Hydrogen Catalysis - Prof. B.L. Feringa (Rijksuniversiteit Groningen)
Consortium partner: Syensqo
Feel Good by Nature. This project aims at developing the novel generation surfactants for cleaning, soap, shampoos etc for personal and home care products. Bio-based acids and oils will be used to make peptide surfactants with a novel green, waste-free, catalytic process for their preparation using AI and robotics methods.
Near-infrared light drives molecular motors; Improving sustainability starts in your own lab; Functional coatings are the future; Phosphorus simply goes chiral
CASEIN: Casein Assembly Systematically Engineered with Integrated Neural networks - Prof. J.C.M. van Hest (TU Eindhoven)
Consortium partner: Arla Foods Amba
Milk Matters: Understanding Casein Assembly Milk, yoghurt, cheese and other dairy products are an important part of many people’s diet. Notably, producing these products can have a significant environmental impact and contribute to your carbon footprint. In this project, researchers at the Eindhoven University of Technology are collaborating with dairy producer Arla to understand the behaviour of an important milk protein, casein. Researchers will investigate casein protein using a combination of robotics and machine learning. This fundamental knowledge will help to develop new dairy products, which beyond milk also contain plant-based ingredients.
Polymer skeleton keeps artificial cell in shape; Artificial cell communication: protein on demand;
High-throughput measurements and deep learning of polymer properties - Prof. W.T.S. Huck (Radboud Universiteit)
Consortium partners: TNO, VLCI
The solution to polymer problems. In this project, we will use high-throughput experiments using liquid and solid handling robots to determine the solubility of a 1000 polymers in >25 different solvents and measure the viscosity of polymer solutions at many different concentrations. This data set will be used to train a deep learning model to predict the properties of polymers in solution. The polymer property data set and deep learning model will accelerate innovation in formulation science.
High-throughput characterization for improved prediction of surfactant mixture properties - Prof. W.T.S. Huck (Radboud Universiteit)
Consortium partner: Croda International
High-throughput analysis for accelerated design of complex surfactant formulations Almost every formulation we use in our daily life, such as shampoos, paints or cosmetics, contains a mixture of surfactants. However, the interactions between these surfactants are hard to predict, and hence, designing new formulations in a targeted way is difficult. In this project, we will implement methods to analyse important formulation properties, such as surface tension and foaming. We will collect a dataset from which we can machine learn the relationship between surfactant mixture composition and its formulation properties. This dataset and model will accelerate design of new formulations.
Automated navigation of bio-based building block implementation in complex resin formulations - Dr. P.A. Korevaar (Radboud Universiteit)
Consortium partners: Koninklijke van Wijhe Verf, WYDO NBD
Automated navigation of bio-based building block implementation in complex resin formulations. This proposal addresses a core challenge in formulating complex paint solutions: introducing bio-based polymer binders as substitutes for ingredients that are traditionally oil-based. Formulating a new compound into dispersions or emulsions with paint-like rheological characteristics requires navigating highly multi-dimensional parameter spaces of polymer structures and additives (e.g. dispersants, co-solvents and surfactants), concentrations and preparation protocols. Exemplifying our platform by polyhydroxyalkanoates as a bio-based paint binder, the to-be-developed high-throughput methodologies and machine learning strategies will be instrumental in accelerating the necessary transition towards “the paint of the future”, and ultimately allow for suggesting suitable molecular structures as novel bio-based paint ingredients.
Predator: the final droplet hunt
Defining the optimal bio-based surfactant; synthesis, robotics, and machine learning in concert - Prof. A.J. Minnaard (Rijksuniversiteit Groningen)
Consortium partner: Henkel AG
Bio-based detergents will replace the currently used detergents based on fossil resources. But it is very challenging to obtain a similar level of performance with bio-based surfactants, which are new kids on the block compared to the conventional surfactants that have been produced for decades. Preparing libraries of novel bio-based surfactants and cherry-pick the best ones, or the best combination of these, is an effective strategy. Artificial Intelligence will guide this cherry-picking and save a tremendous amount of experimental trial-and-error.
Intermolecular Diels-Alder is last step in super complex natural product; Bringing hidden chirality to vibrational light; The rise of electrosynthesis; Crenarchaeol, a molecule of superlatives
BioSoftCoat: Balancing Softness and Durability in Ultra-Soft Coatings: A Robotics and Machine Learning Approach - Dr G. Vantomme (TU Eindhoven)
Consortium partner: SupraPolix
Ultra-soft coatings hold great promise for biomedical applications, such as medical implants, wound dressings, and drug delivery systems, due to their ability to seamlessly integrate with biological tissues. However, achieving coatings that are both soft and durable is a major challenge, as increasing softness often compromises mechanical integrity and long-term stability. This research will leverage AI-driven formulation, robotics, and high-throughput experimentation to systematically map the structure-property relationships governing ultra-soft coatings. By uncovering how molecular assembly controls softness and durability, this project will pave the way for next-generation biomaterials that enhance patient comfort, healing, and therapeutic effectiveness.
AROMA: Accelerated Research in Olfactory Molecular AI- Dr J.M. Weber (TU Delft)
Consortium partner: DSM-Firmenich
AROMA: Accelerated Research in Olfactory Molecular AI The prediction of flavour experiences from molecular formulations remains a long-standing question in chemistry. Expert flavourists rely on years of experience to craft successful combinations. AROMA focuses on the development of novel chemical AI methods for property prediction and high-throughput iterative experimentation for formulations. We introduce semantics into representation learning schemes, develop a property predictor, and design smart iterative experiments. AROMA predicts partition coefficients (release) of flavour molecules in beverages. Our chemical AI methods accelerate reduced-alcoholic beverage design, mimicking known sensory profiles of high-alcohol equivalents.
For more information about the funding itself, please see the website of NWO.
Nog geen opmerkingen