Catalysis & Chemicals
Description
Screen catalyst compositions and reaction routes by prediction. MatterSeek predicts catalytic material properties from composition, then validates them in an autonomous lab.
Catalysis & Chemicals — Material System / Predictable Performance / Validation Pathway
AI for Catalyst Discovery — Screen Active Phases and Supports Before You Run a Reactor
It predicts catalytic material properties from chemical composition, so active phases, promoters and support materials can be ranked before a reactor run or an electrochemical test is committed.
Activity, selectivity and stability are coupled to composition in ways that are not linear, and each experimental evaluation consumes reactor time, feedstock and analytical capacity. Screening by intuition means that large regions of the composition space are never examined — including the ones that hold the best trade-off between activity and durability.
MatterSeek predicts electrochemical and thermodynamic property families from composition, including oxygen evolution onset potential and oxygen evolution Tafel slope — the two quantities that most directly rank electrocatalysts for oxygen evolution — alongside melting point, Seebeck coefficient, power factor and thermoelectric figure of merit. Composition handling supports one to four base elements with doping-level precision of 0.02, so promoter and dopant studies are modelled at the granularity they are actually specified. A prediction round takes about one minute for two base elements, two for three, and seven for four.
not just a list. Results are returned as multi-property radar charts for comparing candidates across several metrics at once, 2D and 3D heat maps that show how activity-related properties shift across a composition range, and exportable tables for design review.
Shortlisted compositions move into MatterSeek's 1,200 m² autonomous laboratory. Two of its six process clusters map directly onto catalyst work: the wet-chemistry cluster covers precursor preparation, hydrothermal and solvothermal synthesis, calcination and a leaching test step; the sol–gel cluster covers sol hydrolysis, gel ageing, solvent exchange and supercritical CO₂ drying — the standard route to high-surface-area supports and aerogel catalyst carriers. Automated characterization includes BET surface area analysis, XRD, Raman and simultaneous thermal analysis, with a stated target of 300 to 500 samples per day depending on route.
Catalyst material families addressed include supported metal and oxide catalysts, zeolite and porous supports, electrocatalysts for oxygen evolution, photocatalysts and thermoelectric materials.
FAQ
Q: Which catalytic properties can be predicted?
A: Electrochemical and thermodynamic families, including oxygen evolution onset potential, oxygen evolution Tafel slope, melting point and thermoelectric parameters such as Seebeck coefficient and figure of merit.
Q: Can MatterSeek design catalyst supports as well as active phases?
A: Yes. Support materials are part of the composition space, and the sol–gel and supercritical drying route in the autonomous laboratory is specifically suited to high-surface-area carriers.
Q: How is catalyst surface area measured?
A: Automated BET analysis runs inside the autonomous laboratory as part of the characterization chain.
Yuyao (Beijing) Technology Co., Ltd. is a professional company specializing in the intersection of AI and advanced materials. The company has independently developed MatterSeek—a foundational large model tailored to the materials sector—and pioneered the HNN (Hierarchical Neural Network) algorithm, while building a proprietary, controllable materials experiment database.