Manuel Kober-Czerny

Abstract:

Metal halide perovskites (MHPs) are emerging semiconducting materials with the potential to replace established silicon-based technologies in sustainable energy applications. These materials show exceptional optoelectronic properties, including high absorption coefficients, long carrier diffusion lengths, and tunable bandgaps. Despite these advantages, their long-term operational reliability remains a key challenge. Therefore their development requires costly and time-consuming accelerated aging experiments to probe the evolution of important performance metrics under application relevant environments. To address these limitations, time series forecasting approaches have emerged as a solution to accelerate these studies. In this work, we benchmark five different forecasting methods, including conventional machine learning models, such as Light Gradient Boosting Machine (LightGBM) and Random Forest, and specialised time series models, including Temporal Fusion Transformers (TFT), Neural Hierarchical Interpolation for Time Series Forecasting (N-HiTS), and a hybrid architecture combining convolutional neural networks and long short-term memory neural networks (CNNLSTM). The results show that tree-based models consistently achieve strong performance across all materials, including those with high variability and extreme values, whereas neural network models fail when data are highly variable or autocorrelation is low. These findings highlight that model robustness and data characteristics are more important than model complexity for MHP degradation forecasting, suggesting that simpler models are often more reliable than complex state-of-the-art architectures. Future work should focus on increasing the dataset, improving data quality, and optimising model hyperparameters.

Abstract:

Vacuum-based deposition is a scalable, solvent-free industrial method ideal for uniform coatings on complex substrates. However, all vacuum-deposited perovskite solar cells fabricated by thermal evaporation trail solution-processed counterparts in efficiency and stability due to film quality challenges, necessitating advancement and improved understanding. Here, we report a co-evaporation route for 1.67-eV wide-bandgap perovskites by introducing a PbCl2 co-source to optimize film quality. We promote perovskite formation with pronounced (100) “face-up” orientation and deliver a certified all vacuum-deposited solar cell with 18.35% efficiency (19.3% in the lab) for 0.25-cm2 devices (18.5% for 1-cm2 cells). These cells retain 80% of peak efficiency after 1,080 hours under the ISOS-L-2 protocol. Leveraging operando hyperspectral imaging, we provide spatiotemporal spectral insight into halide segregation and trap-mediated recombination, correlating microscopic luminescence features with macroscopic device performance while distinguishing radiative from non-ideal recombination channels. We further demonstrate 27.2%-efficient 1-cm2 evaporated perovskite-on-silicon tandems and outdoor stability of all vacuum-deposited tandems in Italy, retaining ~80% initial performance after 8 months.

Abstract:

Performance losses in positive–intrinsic–negative architecture perovskite solar cells are dominated by nonradiative recombination at the perovskite/organic electron transport layer interface, which is particularly problematic for wider bandgap perovskites. Large endeavours have been dedicated to the replacement of fullerenes, which are the most commonly used class of electron transport layers, with limited success thus far. In this work, we demonstrate blending the fullerene derivatives [6,6]-phenyl C61 butyric acid methyl ester (PCBM) and indene-C60 bis-adduct (ICBA) as a thin interlayer between 1.77 eV bandgap perovskite and an evaporated C60 layer. By tuning the fullerene blend to a trace 2% by mass of PCBM in ICBA, we remarkably form an interlayer which features improved energetic alignment with the perovskite and the PCBM : ICBA fullerene mixture, together with a stronger molecular ordering and an order of magnitude higher electron mobility than either neat PCBM or ICBA. Additional molecular surface passivation approaches are found to be beneficial in conjunction with this approach, resulting in devices with 19.5% steady state efficiency, a fill factor of 0.85 and an open-circuit voltage of 1.33 V, which is within 10% of the radiative limit of the latter two device parameters for this bandgap. This work highlights the complex nonlinear energetic behaviour with fullerene mixing, and how control of the energetics and crystallinity of these materials is crucial in overcoming the detrimental recombination losses that have historically limited perovskite solar cells.