In Situ Photoluminescence Images of Blade-Coated Perovskite Solar Cells

1,129 solar cells were fabricated and monitored using a photoluminescence imaging setup during the drying and crystallization of the perovskite layer. The dataset contains time-resolved image series across four spectral channels, cropped to 65x56 pixel patches of each cell's active area. Felix Laufer from the Karlsruhe Institute of Technology created this dataset to enable machine learning for process understanding and performance prediction.

Use Cases

• Predicting photovoltaic performance parameters like power conversion efficiency based on in situ photoluminescence image time series.
• Identifying variations in the blade-coating fabrication process between iterations using spectral imaging data.
• Analyzing the relationship between perovskite layer thickness metrics and crystallization dynamics captured in the luminescence data.
• Developing computer vision models for quality inspection of perovskite thin-films during the vacuum quenching process.

Strengths

• Contains data from 1,129 fabricated solar cells, providing a substantial sample size for analysis.
• Includes four channels of time-series image data per cell, capturing reflectance and three parts of the photoluminescence spectrum.
• Provides multiple target variables for each cell, including four photovoltaic performance parameters and three thickness metrics.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.
• Row count is unknown, which may limit suitability assessment for certain modeling tasks.

Provenance

Source

Karlsruhe Institute of Technology

Collection Method

Data was acquired during the blade-coating deposition and vacuum quenching process of perovskite solar cells using a photoluminescence imaging setup.