Search

• Investigates a solar cell with SnO2 (ETL), fullerene (C60), KGeCl3, and Me4PACz layers. • SCAPS 1d generated 1000 datasets: VOC: 1.13V, JSC: 23.47 mA/cm², FF: 73.34%, PCE: 19.62%. • Uses ML algorithms, especially XGBoost (XGB), to predict PCE with high accuracy. • XGB shows highest R² (99.60%) and lowest MSE (0.150) in predictions. The existing designs of perovskite based solar cells frequently encounter issues with poor photovoltaic (PV) performance and stability issues. In this study, to address above issues, we investigate a novel solar cell device featuring layers of tin dioxide (SnO 2), fullerene (C 60), potassium germanide chloride (KGeCl 3), and Me4PACz. SnO 2 , serving as the electron transport layer (ETL), offers high electron mobility and compatibility with solution processing techniques. Further, different machine learning (ML) algorithms have been utilized for the prediction of accurate PCE prediction in order to simplify computations and improve accuracy. The complex interactions between these parameters and their combined impact on device stability and efficiency are shown through systematic experimentation with SCAPS 1d and 1000 datasets have been generated. Impressive PV parameters are obtained by optimization: V OC : 1.13 V, J SC : 23.47 mA/cm2, FF: 73.34 %, and PCE: 19.62 %. Ensemble machine learning techniques, such as XGBoost (XGB), perform better than individual models, exhibiting increased accuracy and resilience across a range of machine learning performance metrics. When compared to individual SVR, RF and stacked SVR & RF the MSE value produced by the XGB (0.150) algorithm is significantly lower. However, out of all the ML algorithms in this study, the R2 value found in XGB (99.60 %) is the best. The boosting strategy used by XGB is notable for how well it handles complicated datasets and enhances PCE prediction in solar cells. After finding the best suited models (RF & XGB), mean and standard deviation have also been calculated for the different ML performance matrices like MSE, R2 and CVS for 10 iterations. The development of innovative perovskite base PSCs without the need for tedious and lengthy simulations may be facilitated by this approach. [ABSTRACT FROM AUTHOR]

• Cs₂CuBiCl₆ based perovskite solar cell explored. • Optimization of thickness, density and doping performed. • Machine learning models (XGBoost, RF) implemented for PCE prediction. • XGB outperforms RF (R-squared: 99.97 %, low MSE: 0.0014). Conventional perovskite materials offer reasonable power conversion efficiency (PCE) but are plagued by stability issues. To address this, researchers have explored alternative materials, such as MA and FA-based lead or lead-free perovskites, known for their improved PCE and stability. In this study, we focused on a cesium (Cs)-based cell, Cs₂CuBiCl₆, which not only exhibits acceptable PCE but also demonstrates superior stability compared to traditional perovskite materials. To optimize the Cs₂CuBiCl₆-based cell, we conducted simulations varying parameters like thickness, bulk defect density (BDD), and doping. The optimized PCE for the Cs₂CuBiCl₆-based cell reached 10.30 %. We generated 1000 datasets through simulations, which serve as input for machine learning (ML) algorithms, including random forest (RF) and XGboost (XGB). The XGB model outperformed the RF model, achieving a higher R-squared value (R2: 99.97 %) and a lower mean squared error (MSE: 0.0014). [ABSTRACT FROM AUTHOR]

Conventional perovskite-based solar cells (PSCs) have emerged as promising candidates for next-generation solar energy due to their remarkable features, including a high absorption coefficient, tunable bandgaps, high mobility, low maintenance cost, and high power conversion efficiency (PCE). However, the major bottleneck in commercialization of conventional PSCs is their poor stability (of few days), and toxicity concerns (due to lead content). To address these challenges cesium-based perovskites are widely adopted by researchers. However, detailed understanding of these devices considering several device parameters and their connection with overall PCE is not comprehensively disclosed in previous findings. Therefore, in this study, the PV performance of six (6) different PSCs with Cs-based absorber layer (CAL) viz. CsPbI 3 , CsPbBr 3 , CsSnCl 3 , CsSnI 3 , Cs 2 AgBiBr 6 and CsSn 0.5 Ge 0.5 I 3 has been investigated through SCAPS simulator, followed by developing few machine learning models to forecast the efficiency. Total 2160 dataset has been obtained by varying the absorber layer, thickness, and doping and defect density for training and testing the five different machine learning algorithms such as linear regression (LR), support vector regression (SVR), neural network (NN), random forest (RF), and XGBoost (XGB). The XGB algorithm outperforms other approaches, achieving an impressive R2 of 99.99 % and low MSE of 0.0006. Impact of each input variable on the efficiency is also obtained by generating SHAP plot for each model which revealed that absorber layer and it thickness variation greatly affected the PCE and least impact of doping is observed on PCE. Among all the absorbers, CsPbI 3 shows promising performance by delivering a maximum PCE of 14.00 %. Results reported in this work along with developed ML models may pave the way in the development of Cs based PSCs without the need of complex device simulations. • Investigated six different PSCs with Cs-based materials CsPbI 3 , CsPbBr 3 , CsSnCl 3 , CsSnI 3 , Cs 2 AgBiBr 6 and CsSn 0.5 Ge 0.5 I 3. • Five different supervised machine learning models are employed on 2160 datasets using Python. • SHAP plots are generated to visualize the impact of each feature on the conversion efficiency. • XGB outperformed other algorithms, achieving an impressive R2 value of 99.99 % and a low MSE of 0.0006. [ABSTRACT FROM AUTHOR]