In our ERC funded project Defect Engineering, Advanced Modelling and Characterization for Next Generation Opto-Electronic-Ionic Devices (OptEIon) we investigate defects in metal halide perovskites, as defects are the key to understanding many processes such as efficiency losses and degradation phenomena. We are mainly interested in mobile ionic defects, which lead to an ionic conductivity. We intend to characterize them deeply and include these parameters in our numerical device models. Additionally, we use various macroscopic characterization methods (current-voltage-intensity-temperature data, impedance spectroscopy, current and voltage transients) to unravel the electro-optics of the devices. To gain microscopic insights, we use micro-spectroscopy and SPM techniques. To further advance and complement our physics-based modeling, we look into the opportunities of employing machine learning. With the gained knowledge, we want to fabricate devices exploiting the mixed ionic-electronic conductivity in memristive devices.

This huge piece of work covers various aspects towards commercialization of perovskite photovoltaics including upscaling, stability, and environmental impacts. The 24 subsections address the status, current and future challenges and advances in science and technology to meet the challenges. Firouzeh contributed a section regarding thermal stability of the perovskites.

Read the full paper (open access): Shien-Ping Feng et al 2023 J. Phys. Mater. 6 032501.

In this collaboration with Naresh Kumar from Feng Gao's group in Linköping we studied the peculiar behavior of of perovskite LEDs under pulsed illumination. Naresh observed that the electroluminescence yield depends linearly on the on-time of the rectangular voltage pulse, which we found unprecedented and at the same time intriguing. We set up a simple mathematical model that can describe this peculiar behavior. It turned out that the duty cycle is the decisive criterion and that mobile ions that modify recombination rates are the suspects behind the phenomenon. Our description of the duty-cycle dependence of the electroluminescence yield complements the overshoot phenomenon that has been observed on perovskite solar cells by our colleagues in Vilnius a few years earlier.

Read the full paper (open access): Kumawat, N. K., Tress, W. & Gao, F. Mobile ions determine the luminescence yield of perovskite light-emitting diodes under pulsed operation. Nat Commun 12, 4899 (2021).

In this collaboration with Maximilian Sirtl from the Bein group we investigated efficiency limiting processes in these double perovskite solar cells. Compared to the lead-halide perovskites, almost everything is different in this material. Absorption features are narrow instead of broad, emission features, however, are broad instead of narrow and strongly red shifted. Not even the open-circuit voltage can be easily analyzed based on the reciprocity between electroluminescence and photovoltaic quantum efficiency, an approach that works always well for the lead-based devices. Our conclusion is that beyond the material-intrinsic issues such as a much too high bandgap, the contacts are further culprits for the low efficiency. Furthermore we concluded that questions remain unresolved regarding this material.

Read more here and the full paper (open access): Sirtl, M. T., Ebadi, F., Gorkom, B. T. van, Ganswindt, P., Janssen, R. A. J., Bein, T. & Tress, W. The Bottlenecks of Cs2AgBiBr6 Solar Cells: How Contacts and Slow Transients Limit the Performance. Advanced Optical Materials 9, 2100202 (2021).

This work still started at EPFL dragged along for years and was stuck in the review process for long. Firouzeh made the discovery that during light-soaking the photoluminescence and the electroluminescence spectra do not necessarily coincide any more as they commonly do for metal halide perovskites. This triggered our interest and we studied this phenomenon in more detail with focus on mixed halide compositions with phase segregation. We were excited seing photoluminescence, electroluminescence and the expected Voc diverge during light soaking. At that time, the low-gap phase was seen as reason for the low Voc of devices with high Br-content. We found that the decrease of Voc is not directly correlated with the formation of this low-gap phase but a rather high non-radiative recombination rate in these compositions and devices already prior to phase segregation. This finding is in nice agreement with what had been published in the meantime.

Read the full paper: Ebadi, F., Yang, B., Kim, Y., Mohammadpour, R., Taghavinia, N., Hagfeldt, A. & Tress, W. When photoluminescence, electroluminescence, and open-circuit voltage diverge – light soaking and halide segregation in perovskite solar cells. J. Mater. Chem. A (2021).