top of page

Vision-driven Autocharacterization of Semiconductors

Paper: [link]

Code: [link]

​

My Contributions:

  • Study design

  • Algorithm design and implementation

  • Wrote the computer vision code (Python)

  • Built multi-material printing hardware

  • Algorithm-hardware integration (C++)

  • Synthesized semiconductor materials

  • Conducted characterization (hyperspectral, band gap, XRD, stability)

  • Data analysis

  • Wrote the paper

In materials research, the task of characterizing hundreds of different materials traditionally requires equally many human hours spent measuring samples one by one. We demonstrate that with the integration of computer vision into this material research workflow, many of these tasks can be automated, significantly accelerating the throughput of the workflow for scientists. We present a framework that uses vision to address specific pain points in the characterization of perovskite semiconductors, a group of materials with the potential to form new types of solar cells. With this approach, we automate the measurement and computation of chemical and optoelectronic properties of perovskites.

Through the use of high-fidelity computer vision segmentation, we demonstrate the scalability of accurate and automatic characterization for 80 unique FA1-xMAxPbI3 perovskite samples. The developed vision segmentation method avoids edge effects and removes aberrations to output a clean segmentation of optical data. From the segmented optical data, the automatic band gap computation is demonstrated to achieve an accuracy of 89.1%, the automatic stability measurement is demonstrated to achieve an accuracy of 87.8%, and the automatic composition extraction is demonstrated to achieve an accuracy of 78.8% across all 80 perovskite samples, relative to a human evaluation reference. Furthermore, the results extracted by the developed vision-driven autocharacterization tools are validated to align with traditional x-ray diffraction (XRD) characterization. The results achieved by the methods developed in this paper support the advancement of scalable, high-throughput novel materials discovery by surmounting the obstacle of sample-by-sample characterization.

Automatic band gap extractor

Automatic stability measurement

Output band gaps

Output instability

bottom of page