Learned catalyst search guides oxygen-evolution catalysts made from meteorites
Source check: Source checked. Not independent scientific confirmation.
Finding: A robotic chemistry system combined a neural-network performance model with theoretical calculations and experimental calibration to select a catalyst formulation from five Martian meteorite feedstocks. Laboratory testing of the selected formulation measured an oxygen-evolution overpotential of 445.1 mV at 10 mA per square centimetre, lower than the compared theory-only and experiment-only search formulations. The work demonstrates a terrestrial laboratory catalyst result; oxygen production on Mars and a globally optimal composition were not independently established.
Technology: Neural-network oxygen-evolution performance prediction; model-guided composition search
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- A learned neural-network model combines first-principles and robot-generated experimental information to predict oxygen-evolution performance and guide which mixed-meteorite catalyst formulation to synthesize and test.
- Test and result
- The primary journal supplement compares measured overpotentials of 445.1 mV for the combined model-guided formulation, 467.4 mV for experiment-guided optimization and 542.7 mV for theory-guided selection. The original preprint abstract already reports the 445.1 mV result and more than 550000 seconds of operation.
- Limitation / still unknown
- The first located preprint was posted March 23, 2023; the journal version appeared November 13, 2023, with a March 2024 issue date. The preprint body was inaccessible; version-specific result presence is established from its primary deposited abstract, while detailed comparator measurements come from the journal supplement. Search-space coverage does not prove global optimality. Mars deployment, resource extraction at scale and integrated life-support performance remain untested.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.21203/rs.3.rs-2681831/v1 · https://www.nature.com/articles/s44160-023-00424-1 · https://static-content.springer.com/esm/art%3A10.1038%2Fs44160-023-00424-1/MediaObjects/44160_2023_424_MOESM1_ESM.pdf
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
Review date: 21 Nov 2026. Review scope: Primary preprint metadata and deposited abstract; journal front/abstract/code-availability and full author supplement, including model-guided experimental comparison. Main journal body and preprint body were not read..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.