Iterative learning guides discovery of a cobalt-doped nitride battery catalyst
Source check: Source checked. Not independent scientific confirmation.
Finding: Iteratively updated regression models screened transition-metal compositions and guided synthesis of three candidate battery catalysts. Cobalt-doped molybdenum nitride, Co 0.1 Mo 0.9 N, showed the best reported performance of these candidates in laboratory lithium-carbon-dioxide and lithium-air cells. Reported low overpotentials were measured at 50 milliamperes per gram; they do not establish commercial cell performance or broad model accuracy.
Technology: Iterative supervised regression; random forest; Gaussian-process regression; XGBoost; nearest-neighbor regression
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Regression models learned catalyst-descriptor/overpotential relationships, with measurements added across five iterations before screening 15012 compositions and selecting three for synthesis.
- Test and result
- Three selected catalyst compositions were synthesized and characterized; Co 0.1 Mo 0.9 N was evaluated against MoN and carbon-nanotube cathode comparators in laboratory battery tests.
- Limitation / still unknown
- Generalization is not established by the reported 0.106-volt mean absolute error on only three selected candidates. Practical battery durability, manufacturing and device-level benefit are unestablished; reported overpotential depends on current density and cell conditions.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1021/jacs.5c15395 · https://doi.org/10.1021/jacs.5c15395.s001
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 21 Nov 2026. Review scope: Primary publisher article-front metadata and abstract, complete official author supporting information Methods/model tables/experimental characterizations and result captions, and current PubMed notice metadata. Paywalled main article body was not read; no independent replication performed..
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.