Introduction
What is Digital Catalysis?
Digital catalysis is changing the way we discover new materials. Instead of relying on slow trial-and-error experiments, it combines advanced computer models, machine learning, and automated lab systems to find and improve catalysts much faster.
By creating “digital twins” of chemical reactions, we can test millions of potential catalyst designs on a computer in the same time it would take to make just one in a traditional lab. This approach helps speed up the development of sustainable energy solutions.
Image AI generated
Automation
The Self-Driving Lab
At the heart of our research is the establishment of self-driving labs (SDLs). An SDL combines AI with robotics to accelerate catalyst discovery. The AI plans experiments in iterative loops, using data from previous tests to focus only on the most informative syntheses and analyses. Automation boosts throughput, reproducibility, and safety, while new testing procedures, designed specifically for SDLs, allow long-term monitoring of material performance, speeding up sustainable catalyst development far beyond traditional methods.
- Autonomous decision-making: AI continuously analyzes real-time data to plan the most informative next experiments.
- Robotic synthesis: Automated systems precisely control reactions, handling fluids and complex conditions without human intervention.
- In-line characterization: Integrated spectroscopy and chromatography provide instant feedback, closing the loop for faster, smarter experimentation.
For Researchers
Technical Methodology
Our robust approach relies on a tightly integrated stack of computational software and
physical hardware, creating a closed-loop system for accelerated discovery.
Density Functional Theory (DFT)
Uses quantum mechanics to model atoms and molecules and to predict which catalyst structures are most promising. This reduces trial-and-error experiments and speeds up discovery.
Machine Learning Models
AI analyzes experimental and simulation data to find patterns, thus predicting the most effective catalyst designs under specific conditions. This guides experiments, making research faster and more efficient.
Automated Synthesis
Robotic systems carry out chemical reactions with precision and consistency, handling complex procedures that would take humans much longer. Automation increases throughput, accelerating material development.
In-situ Characterization
Real-time monitoring observes reactions as they happen in the lab. Techniques like spectroscopy and chromatography provide immediate feedback, enabling rapid optimization and understanding of catalyst performance.
Applications
Key Use Cases
Green Hydrogen Production
Green hydrogen, made from water using renewable electricity, is a cornerstone for decarbonizing heavy industries like steel and cement, powering fuel cells, and producing sustainable chemicals. It enables a transition away from fossil fuels while complementing renewable energy systems.
CO₂ Valorization
Captured CO₂ can be converted into carbon-based chemicals and fuels, helping close the carbon loop. This approach reduces greenhouse gas emissions, creates sustainable feedstocks, and supports hard-to-abate sectors where direct electrification is not feasible.
Sustainable Materials
Transforming chemical value chains to renewable or circular feedstocks—like bio-based carbon, recycled plastics, or CO₂-derived intermediates—ensures low-emission production while leveraging existing infrastructure. These materials enable drop-in solutions for fuels, plastics, fertilizers, and industrial chemicals.
Process
The Research Workflow
Our closed-loop approach enables continuous learning and optimization across humans and machines.
1. Design
AI suggests promising candidate materials based on existing data.
2. Create
Robotic platforms synthesize the chemical catalysts automatically.
3. Test
Automated in-situ evaluation of the catalysts’ physical performance.
4. Learn
Extracted data immediately feeds back to improve the ML models.