PV-Lab / dissimatrix

Dissimilarity matrix analysis for perovskite cappping-absorber pairs degradation data

Dissimilarity Matrix for Perovskite Capping-Absorber Pairs

This documentation is prepared as the workflow to accompany the following study:

"Tailoring capping-layer composition for improved stability of mixed-halide perovskites"

Noor Titan Putri Hartono (1), Marie-Hélène Tremblay (2), Sarah Wieghold (3), Benjia Dou (1), Janak Thapa (1), Armi Tiihonen (1), Vladimir Bulovic (1), Lea Nienhaus (4), Seth R. Marder (2,5-8), Tonio Buonassisi (1), Shijing Sun (1)

Affiliations:

1. Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139
2. Georgia Institute of Technology, North Avenue, Atlanta, GA 30332
3. Argonne National Laboratory, 9700 S. Cass Avenue, Lemont, IL 60439
4. Florida State University, Department of Chemistry and Biochemistry, 95 Chieftan Way Tallahassee, FL 32306
5. University of Colorado Boulder, Renewable and Sustainable Energy Institute, Boulder, CO 80303
6. University of Colorado Boulder, Department of Chemical and Biological Engineering, Boulder, CO 80303
7. University of Colorado Boulder, Department of Chemistry, Boulder, CO 80303
8. National Renewable Energy Laboratory, Chemistry and Nanoscience Center, Golden CO 80401

Installation and Requirements

To run the dissimilarity_matrix_all.py, Python needs to be installed (quickest way: install Miniconda https://docs.conda.io/en/latest/miniconda.html, and install Spyder by typing conda install -c anaconda spyder on Anaconda Prompt). The following packages also need to be installed.

1. Pandas (https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html)
2. NumPy (https://docs.scipy.org/doc/numpy/user/install.html)
3. Seaborn (https://seaborn.pydata.org/installing.html)
4. Matplotlib (https://matplotlib.org/users/installing.html)
5. SciPy (https://www.scipy.org/install.html)
6. Scikit-learn (https://scikit-learn.org/stable/install.html)

OR clone the following repository: pip install -r requirements.txt

Workflow

The raw image data (.bmp files) needs to be processed first, to extract the average red, green, and blue (RGB) values over time for each samples. The dataset result is shown in Dataset/(DegradationRun) folder.

There are several datasets available.

1. sample_r_cal.csv, sample_g_cal.csv, sample_b_cal.csv: The average RGB values over time.
2. times.csv: The degradation time.
3. Samples_cap.csv: The sample list for each batch, with the basic information on the absorber, type of capping layer materials, and processing conditions.
4. Samples_cap_merit.csv: The sample list for each batch, including the instability index results for each RGB channel.

Based on this data, we can calculate the dissimilarity matrix.

In dissimilarity_matrix_all.py, there are different inputs that you can change.

1. datapoint: How many time points you would like to include in the analysis.
2. frequency: How often the degradation images are taken.
3. MAPbBrContent_1, MAPbBrContent_2, MAPbBrContent_3: The amount of MAPbBr₃ of interest.
4. concentration_1, concentration_2, concentration_3: The concentration of capping layers of interest.
5. annealing_1, annealing_2, annealing_3: The annealing temperature for the capping layers of interest.
6. capping_2, capping_3: Capping layer materials of interest.
7. metric: The dissimilarity matrix distance measure, can be 'euclidean', 'cosine', or 'manhattan'.
8. folderToSave: Where to save the analysis results.

Run the dissimilarity_matrix_all.py file. This will generate dissimilarity matrix for bare, 9-Cl-capped, and PTEAI-capped films with different types of absorbers.

In the Results folder, there are dissimilarity matrix results for different metrics: 'cosine', 'euclidean', and 'manhattan'. There is also a folder called dissValues which contains the dissimilarity values for different capping-absorber pairs.

Authors

Author(s)	Noor Titan Putri Hartono
Version	1.0/ September 2021
E-mail(s)	noortitan at alum dot mit dot edu

Attribution

This work is under an Apache 2.0 License. Please, acknowledge use of this work with the appropriate citation to the repository and research article.

Citation

@Misc{dissismatrix2021,
  author =   {The Dissimilarity Matrix authors},
  title =    {Dissimilarity Matrix for Perovskite Capping-Absorber Pairs},
  howpublished = {\url{https://github.com/PV-Lab/dissimatrix}},
  year = {2021}
}