Cutting “lab-to-fab” short: high throughput optimization and process assessment in roll-to-roll slot die coating of printed photovoltaics

Wagner, Michael; Distler, Andreas; Le Corre, Vincent M.; Zapf, Simon; Baydar, Burak; Schmidt, Hans-Dieter; Heyder, Madeleine; Forberich, Karen; Lüer, Larry; Brabec, Christoph J.; Egelhaaf, H.-J.

doi:10.1039/d3ee01801f

Cited by 8 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To expand the algorithm space, the ensemble-boosting tree learning algorithms LightGBM and CatBoost are incorporated, which also offer the ability to adjust sample weights and employ regularization techniques. , The remaining 20% of the data is reserved for further testing. Considering the trade-off between the time complexity in high-dimensional spaces and the global optimal solution, the Bayesian optimization (BO) algorithm is employed for hyperparameter tuning. − The bounds of the hyperparameter space are dynamically adjusted through a progressive learning process, gradually converging to a smaller region of optimal values. As shown in Figure c, the three models (with the best hyperparameters detailed in Table S21) exhibit a similar trend with relatively lower precision and higher specificity (Table S19) when considering HEDs as the class of interest.…”

Section: Resultsmentioning

confidence: 99%

Temporal Decoupling-Based Machine Learning Framework for Precise Efficiency Prediction in Perovskite Solar Cells

Yang,

Meng

et al. 2024

J. Phys. Chem. C

View full text Add to dashboard Cite

The rapid and accurate identification of potential high-efficiency design strategies for perovskite solar cells (PSCs) is of paramount importance in advancing their development and commercialization. However, the application of machine learning (ML) algorithms in this field is hindered by unstable PSC data sets (e.g., time-related noise and data imbalance). Here, we introduce a ML framework specifically tailored for temporal decoupling through feature engineering and uncertainty modeling (noise processing techniques) to accurately predict the efficiency of PSCs. In our framework, we utilize one-hot encoding and feature fusion methods to extract features from a shared data set encompassing perovskite material, processing, and PSC architecture. After temporal decoupling, our ML model shows an outstanding precision of 96.88% and a specificity of 99.3% for high-efficiency devices and is successfully applied to predict PSC efficiencies in the period between 2021 and 2023. This temporal decoupling ML framework also reveals hidden relationships between features and efficiency through cross-feature analysis. Our work demonstrates the potential of ML for predicting performance and elucidates the associated mechanisms, accelerating PSC commercialization and reducing costs.

show abstract

Section: Resultsmentioning

confidence: 99%

Temporal Decoupling-Based Machine Learning Framework for Precise Efficiency Prediction in Perovskite Solar Cells

Yang,

Meng

et al. 2024

J. Phys. Chem. C

View full text Add to dashboard Cite

show abstract

“…With the versatility of slot-die coating and the ability to coat using a range of viscosities, there have been a wide range of materials coated with this technique, including but not limited to active layer blends (i.e., P3HT:PCBM [ 90 , 165 ], PPDT2FBT:PCBM [ 67 , 88 ], PBDB-T:ITIC [ 113 ], and PM6:Y6 [ 170 , 175 ]), electron (i.e., ZnO NPs [ 161 , 162 , 170 ], SnO 2 [ 58 , 176 ], PEI [ 139 ], and AZO [ 85 , 153 ]) and hole transport layers (i.e., PEDOT:PSS [ 139 , 165 ]), and electrodes (i.e., AgNWs [ 176 , 177 ]). As for the number of layers coated via slot-die coating on a single device, a significant amount of papers have implemented this technique for two [ 155 , 162 , 178 ] or three layers [ 104 , 105 , 154 , 179 ], with a minor amount applying the technique for a single layer [ 42 , 65 , 87 ].…”

Section: Coating and Printing Techniques For Opvmentioning

confidence: 99%

Printing and Coating Techniques for Scalable Organic Photovoltaic Fabrication

Kirk,

Bjuggren,

Andersson

et al. 2024

Materials

View full text Add to dashboard Cite

Within recent years, there has been an increased interest towards organic photovoltaics (OPVs), especially with their significant device performance reaching beyond 19% since 2022. With these advances in the device performance of laboratory-scaled OPVs, there has also been more attention directed towards using printing and coating methods that are compatible with large-scale fabrication. Though large-area (>100 cm2) OPVs have reached an efficiency of 15%, this is still behind that of laboratory-scale OPVs. There also needs to be more focus on determining strategies for improving the lifetime of OPVs that are suitable for scalable manufacturing, as well as methods for reducing material and manufacturing costs. In this paper, we compare several printing and coating methods that are employed to fabricate OPVs, with the main focus towards the deposition of the active layer. This includes a comparison of performances at laboratory (<1 cm2), small (1–10 cm2), medium (10–100 cm2), and large (>100 cm2) active area fabrications, encompassing devices that use scalable printing and coating methods for only the active layer, as well as “fully printed/coated” devices. The article also compares the research focus of each of the printing and coating techniques and predicts the general direction that scalable and large-scale OPVs will head towards.

show abstract

Pathways to Upscaling Highly Efficient Organic Solar Cells Using Green Solvents: A Study on Device Photophysics in the Transition from Lab‐to‐Fab

Mazzolini,

Pacalaj,

et al. 2024

Advanced Science

View full text Add to dashboard Cite

As the rise of nonfullerene acceptors (NFA) has allowed lab‐scale organic solar cells (OSC) to reach 20% efficiency, translating these devices into roll‐to‐roll compatible fabrication still poses many challenges for researchers. Among these are the use of green solvent solubility for large‐scale manufacture, roll‐to‐roll compatible fabrication, and, not least, information on charge carrier dynamics in each upscaling step, to further understand the gap in performance. In this work, the reproducibility of champion devices using slot‐die coating with 14% power conversion efficiency (PCE) is demonstrated, under the condition that the optimal thickness is maintained. It is further shown that for the donor:acceptor (D:A) blend PM6:Y12, the processing solvent has a more significant impact on charge carrier dynamics compared to the deposition technique. It is found that the devices processed with o‐xylene feature a 40% decrease in the bimolecular recombination coefficient compared to those processed with CB, as well as a 70% increase in effective mobility. Finally, it is highlighted that blade‐coating yields devices with similar carrier dynamics to slot‐die coating, making it the optimal choice for lab‐scale optimization with no significant loss in translation toward up‐scale.

show abstract

Cutting “lab-to-fab” short: high throughput optimization and process assessment in roll-to-roll slot die coating of printed photovoltaics

Abstract: Commercialization of printed photovoltaics requires knowledge of the optimal composition and microstructure of the single layers, and the ability to control these properties over large areas under industrial conditions. While...

Cited by 8 publications

References 24 publications

Temporal Decoupling-Based Machine Learning Framework for Precise Efficiency Prediction in Perovskite Solar Cells

Temporal Decoupling-Based Machine Learning Framework for Precise Efficiency Prediction in Perovskite Solar Cells

Printing and Coating Techniques for Scalable Organic Photovoltaic Fabrication

Pathways to Upscaling Highly Efficient Organic Solar Cells Using Green Solvents: A Study on Device Photophysics in the Transition from Lab‐to‐Fab

Contact Info

Product

Resources

About