1/ How AI is used: closed-loop laboratory and machine learning models
2/ Application task: high-throughput evaluation of hydrogen catalyst compositions and exploration of promising families
3/ Reported result: 2,942 compositions were evaluated and 6 families of palladium-based oxides were identified
4/ According to C&EN, Lila Sciences combined a closed-loop laboratory with machine learning models to evaluate 2,942 catalyst compositions.
5/ The process reportedly identified 6 families of palladium-based oxides that could reduce reliance on scarce iridium in electrolytic hydrogen production.
6/ From a materials research perspective, this case shows that closed-loop exploration combining automated experiments and machine learning can expand the speed and scope of catalyst-composition screening.
7/ It is especially suitable for workflows that rapidly evaluate many composition combinations and narrow promising families to the next round of experiments.
8/ However, the report noted that long-term durability in industrial electrolyzers remains a separate challenge.
9/ Therefore, a composition that looks promising in the laboratory cannot be assumed to last under actual device conditions. [1]
Materials AI · written by AI
Lila Sciences rapidly screens hydrogen catalyst candidates with an AI-guided automated laboratory
Lila Sciences reportedly evaluated 2,942 catalyst compositions using a closed-loop laboratory and machine learning models, and identified 6 families of palladium-based oxides that could reduce dependence on iridium in electrolytic hydrogen production. However, actual durability in industrial electrolyzers remains a separate challenge. [1]
How to apply this
To rapidly screen thousands of catalyst compositions with closed-loop automation and machine learning, then narrow the search to promising families.
- Design an automation-compatible protocol by dividing the experimental variables into composition, process conditions, and measurement indicators.
- Calibrate the measurement system with a small number of reference samples, then evaluate many compositions under identical conditions.
- Use a machine learning model to update the next set of experimental candidates and reduce the exploration space in a closed-loop manner.
- Resynthesize the top candidates as independent samples and verify their performance and stability.
- Data needed
- Data are needed on catalyst composition samples, reaction performance measurements, and experimental conditions that allow repeated measurements.
- Checking the result
- Resynthesize the candidate families separately and remeasure them, then check whether the initial screening ranking matches independent performance indicators.
- What can go wrong
- A composition that looks good in initial screening may perform poorly in durability or reproducibility under actual device conditions.
이 가이드는 원문의 방법을 바탕으로 AI가 제안한 적용 방향입니다. 원문에 적힌 내용이 아니며 같은 결과가 재현된다는 보장도 아닙니다.
Related concepts and practice
이 이슈를 읽는 데 필요한 개념입니다. 용어마다 Colab에서 바로 돌려볼 수 있는 실습 노트북이 있습니다.
- 폐루프 탐색Closed-loop Discovery · 순환 탐색 Practise in Colab ↗Download notebook
- 머신러닝 포텐셜Machine Learning Potential · ML Potential
- 고속 스크리닝High-throughput Screening · HTS · 다단 스크리닝 Practise in Colab ↗Download notebook
- 베이지안 최적화Bayesian Optimization · BO Practise in Colab ↗Download notebook
- 머신러닝Machine Learning · ML · 기계학습 Practise in Colab ↗Download notebook
Sources
웹 자료를 바탕으로 AI가 자동 요약한 글입니다. 연구의 세부 조건과 결과는 연결된 원문을 확인해주세요.