Related Experiment Video
Updated: Oct 9, 2025
10:36
Advanced Experimental Methods for Low-temperature Magnetotransport Measurement of Novel Materials
Published on: January 21, 2016
10.7K
Machine learning-accelerated design and synthesis of polyelemental heterostructures
Carolin B Wahl1,2, Muratahan Aykol3, Jordan H Swisher2,4
1Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA.
Science Advances
|December 22, 2021
Summary
Machine learning accelerates materials discovery by guiding nanoparticle synthesis. A closed-loop system successfully created 18 complex nanomaterials, advancing materials science and applications.
Area of Science:
- Materials Science
- Nanotechnology
- Computational Chemistry
Background:
- Synthetic capabilities in materials discovery often exceed data analysis.
- Machine learning (ML) is crucial for navigating vast chemical spaces.
- Efficient methods are needed to identify novel materials with specific properties.
Purpose of the Study:
- To develop an ML-driven, closed-loop experimental process for synthesizing polyelemental nanomaterials.
- To target specific structural and interfacial properties in nanoparticle discovery.
- To bridge the gap between synthesis and data extraction in materials science.
Main Methods:
- Utilized an eight-dimensional chemical space (Au-Ag-Cu-Co-Ni-Pd-Sn-Pt) as input for ML models.
- Employed a Bayesian optimization algorithm within a closed-loop experimental framework.
- Iteratively synthesized and analyzed nanoparticle compositions, feeding results back to the ML algorithm.
Main Results:
- Successfully synthesized 18 novel heterojunction nanomaterials.
- Discovered nanoparticle compositions too complex for traditional chemical intuition.
- Included the synthesis of highly complex biphasic nanoparticles.
Conclusions:
- The developed ML-driven closed-loop system effectively guides the synthesis of complex nanomaterials.
- This approach significantly accelerates materials discovery and expands the range of accessible material compositions.
- The platform has the potential to revolutionize materials discovery across various industries.
