Imagine a tiny robot that works like a leaf, catching sunlight to turn water and air into useful fuel. Instead of a human scientist guessing the best settings, an AI brain manages the process to find the fastest way to produce chemicals. It combines light-catching lenses and special filters into one smart system that learns on its own.
Chemical production is often slow and inefficient because finding the right catalyst is a trial-and-error process. Current systems lack the ability to simultaneously capture light, trigger reactions, and separate products in one automated step.
A modular silicon-based photoreactor and 3D carbon nitride nanostructures managed by machine learning models to optimize chemical productivity.
If you are a renewable energy producer dealing with inefficient water-splitting for hydrogen — this project developed 3D carbon nitride photocatalysts that optimize the breakdown of H2O into H2. This allows for more efficient solar-to-fuel conversion.
If you are an industrial CO2 emitter dealing with high costs of carbon recycling — this project developed a reaction robot using CO2 as a substrate. It uses AI to maximize the productivity of turning waste gas into chemicals.
If you are a chemical synthesis SME dealing with slow R&D cycles for new catalysts — this project developed a modular reactor that operates up to 250 °C and 5 bar. This enables faster testing and autonomous optimization of chemical reactions.
Based on available project data, specific pricing or unit costs for the reaction robot are not provided.
The project has developed a modular reactor design and 3D networks, but based on available project data, the current focus is on the reaction robot prototype rather than full-scale industrial plants.
Based on available project data, there is no specific information regarding patents or licensing terms for the carbon nitride materials or AI models.
The system uses a modular design that allows for a non-invasive switch between reactor types, which may simplify integration into existing experimental setups.
The project period runs from 2022-09-01 to 2026-08-31, suggesting it is currently in the development and validation phase.
The consortium is heavily weighted toward research, with 5 universities and 2 research centers. However, the inclusion of 2 industrial partners (including 1 SME) and a 22% industry ratio indicates a clear intent to translate the AI-driven photocatalysis from the lab to a commercial application.
Contact the Universidad del Pais Vasco for technical specifications on the 3D quantum dot networks.
Contact SciTransfer to bridge the gap between this AI-robotics research and your chemical production line.
