Modified Diode Model

NEWS: 🔥Modified Diode Model for Perovskite Solar Cells is Accepted by Solar RRL!🔥

Quantifying Nonradiative Recombination and Resistive Losses in Perovskite Photovoltaics: A Modified Diode Model Approach

Minshen Lin, Xuehui Xu, Hong Tian, Yang (Michael) Yang, Wei E. I. Sha, Wenxing Zhong

The source code for the modified diode (MD) model written in MATLAB (version: R2022b).

You can use the MD model to discern and quantify the electrical losses in a complete perovskite photovoltaic device under operational conditions. The code in this repository, along with the drift-diffusion simulation data, can reproduce the results in our paper, which validates the effectiveness of this approach.

If you require any further information or if we can be of any assistance, feel free to contact us by creating an issue or send us emails.

Info

Repository Structure

├── Classes and Functions
│   ├── Models
│   │   ├── DataLoader.m               <- DataLoader class for loading the simulation dataset
│   │   ├── DataPreconditioner.m       <- DataPreconditioner class for the preconditioning of the simulation/experimental data
│   │   ├── Device.m                   <- Device class for defining perovskite PV parameters
│   └── Utils
│   │   ├── costFunction.m             <- Function for computing the cost function
│   │   ├── fittingMD.m                <- Function for curve fitting with the MD model
│   │   ├── lossAnalysis.m             <- Function for evaluating the normalized PCE gains
│   │   ├── plotJrecV.m                <- Function for plotting the J(recombination)-V curves
│   │   ├── plotJV.m                   <- Function for plotting the J-V curves
│   │   ├── plotNormPCEGain.m          <- Function for graphing the normalized PCE gains
│   │   ├── solver.m                   <- Function for solving the JV curves with the retrieved parameters from the MD model
├── Experimental Data                  <- Experimental JV data, demonstrating how the data should be formatted
├── Simulation Data                    <- SCAPS-1D simulated JV data
│   ├── Figure1
│   └── Figure4
├── single_dataset_fitting.m           <- Script for applying the MD model to the simulated single dataset
├── batch_dataset_fitting.m            <- Script for applying the MD model to the simulated batch dataset
└── main.m                             <- Script for applying the MD model to the experimental JV data

Content summary

Files in the root of this repository:

• main.m is a script where the MD model is applied to the experimental JV data for loss quantification.
• single_dataset_fitting.m is a script where the MD model is applied to one set of simulated JV curve, reproducing the results in Fig. 1 of our paper.
• batch_dataset_fitting.m is a script where the MD model is applied to a batch of simulated JV curves, reproducing the results in Fig. 4 of our paper.

Files in the Simulation Data folder:

• Figure1 is a folder that contains the JV curve (simulated by SCAPS-1D) under both 1 Sun and 50 Suns illumination. Using the data in this folder with single_dataset_fitting.m script, you can reproduce the results in Fig. 1 of our paper.
• Figure4 is a folder that contains batches of JV curves (simulated by SCAPS-1D). These data are simulated by varing the parameters related bulk and interface SRH recombination simultaneously, in order to study their impact on the performace of the MD model. Using the data in this folder with batch_dataset_fitting.m script, you can reproduce the results in Fig. 4 of our paper.

Files in the Experimental Data folder contains the experimental JV data, intended to show how the data should be formatted to interface with the main.m script.

Usage

To get started, you can run the main.m script with the your experimental JV data (the format of the data should comply with those in the Experimental Data folder). The workflow of the script can be summarized as the following steps:

• First, initialize a Device object in which you can specify the parameters of your photovoltaic device.
• Second, initilize a DataPreconditioner object that can parse the JV data and store it into the expected format.
• Third, solve the nonlinear least squares fitting problem using the fittingMD function.
• After retrieveing the model parameters, numerically solve the JV curve with solver function and plot it with plotJV.
• Finally, compute the normalized PCE gains with lossanalysis function and create a bar graph with plotNormPCEGain.

You can refer to these steps to write your own script, and follow the guidelines in our paper to apply the MD model and interpret the results for your own device.

Besides, you can directly run the single_dataset_fitting.m or the batch_dataset_fitting.m to reproduce the results in our paper, and if you want to further explore the MD model with SCAPS-1D, you can use the DataSet and the DataPreconditioner classes to parse the simulated data.

Citation

If you find this work useful in your research, please cite:

@article{lin2024quantifying,
  title={Quantifying Nonradiative Recombination and Resistive Losses in Perovskite Photovoltaics: A Modified Diode Model Approach},
  author={Lin, Minshen and Xu, Xuehui and Tian, Hong and Yang, Yang and Sha, Wei EI and Zhong, Wenxing},
  journal={Solar RRL},
  volume={8},
  number={1},
  pages={2300722},
  year={2024},
  publisher={Wiley Online Library}
}

Other Info

Repository Author: Minshen Lin

Email: [email protected]

Institution: Zhejiang University