Back to search

Cornell University

SEM Image Reduction and Analysis for Machine Learning Optimization of Metal Halide Perovskites

Abstract

dc:description.abstract

Solar energy resources are poised as the biggest renewable competitor of fossil fuels. Solar perovskite thin films are an emerging technology with efficiencies close to established photovoltaic technologies. However, structural and thermalinstability and an exponentially large number of possible compositions and processing parameters have led to reproducibility and scalability issues even at the lab scale. A DOE-funded collaboration across universities and national labs is creating a multi-scale statistical machine learning (ML) model to optimize the entire process from perovskite synthesis to device performance. The collaboration is called SPIRALs, which stands for Science and Processing Informed by Rational Algorithmic Learning (SPIRALs). A key challenge is integrating data from diverse data sources, such as experimental data and SEM images, as inputs to the ML algorithms. This work reports the Python code developed to reduce SEM images using spatial functions such as power spectral densities and autocorrelation functions. Quantitative analysis of these functions is done to derive spatial parameters such as the correlation length, grain size, fractal dimension, and Hurst exponent that describe the surface morphology of thin films. These parameters can tie synthesis methods to device performance and act as a check for ML framework predictions. The developed coding methodology and subsequent analysis can be applied to any thin film image and extended to different instruments to facilitate the SPIRALs project.

Degree

thesis:*
Name thesis:degree_name
M.S., Chemical Engineering
Level thesis:degree_level
Master of Science
Discipline thesis:degree_discipline
Chemical Engineering
Grantor
Cornell University
Year dc:date.issued
2023

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Riaz, Shanza
Chair dc:contributor.chair
  • Hanrath, Tobias
Committee members dc:contributor.committeemember
  • Estroff, Lara
  • Marohn, John

Subjects

dc:subject × 6

Rights

dc:rights
Statement dc:rights
  • Attribution 4.0 International
Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Bibid
16219170
Dc Identifier Other
ProQuest Submission ID: 11950
ProQuest Publication ID: 30631549

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Riaz, Shanza. SEM Image Reduction and Analysis for Machine Learning Optimization of Metal Halide Perovskites. Master of Science thesis, Cornell University, 2023. https://hdl.handle.net/1813/114502