Catalyst Performance Data for Methanol Production from CO2 Hydrogenation

A database of 1,425 datapoints on catalysts for thermocatalytic CO2 hydrogenation to methanol compiled from literature. The dataset includes 12 experimental descriptors used to predict methanol space-time yield (STY) with ensemble-tree models achieving R² > 0.85. It was developed by Manu Suvarna at ETH Zurich as part of a machine learning framework for catalyst performance prediction.

Use Cases

• Predict methanol space-time yield based on catalyst descriptors like space velocity and pressure.
• Interpret feature importance for catalyst design based on the 12 experimental descriptors.
• Validate machine learning models for catalytic processes using the provided experimental data.
• Guide experimental optimization of CO2 hydrogenation catalysts using the data-driven framework.

Strengths

• Contains 1,425 datapoints compiled from literature.
• Models built on the data achieved high accuracy with R² > 0.85.
• Includes 12 descriptors for prediction, with significance of key factors like space velocity revealed.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known but the total number of unique catalysts or experiments is unspecified.
• Last update date is unknown; freshness unverified.

Provenance

Source

ETH Zurich

Collection Method

Compiled from literature sources.