The commercialization of perovskite solar cells is currently prevented due to their poor thermal stability. One valuable tool to understand the micro- and nano-scale processes correlated to performance degradation is in situ heating in the transmission electron microscope. Here devices are produced in accordance with four common approaches in the literature. As the cells are heated within their operational range, the photovoltaic performance is characterized.
Transmission electron microscopy is then used to analyze the devices as they are heated further in situ to monitor changes in morphology and chemical composition using Energy-dispersed X-ray (EDX) elemental mapping. Mechanisms for structural and local chemical changes were identified, such as iodine and lead migration. These can be correlated to the synthesis conditions.
- Aging and heating cause perovskite solar cells to degrade. As such, it is necessary to understand the stability of the hybrid organic-inorganic perovskite and the organic hole transporting layers.
- Heat-induced degradation pathways need to be identified for the devices
- The behavior of four devices produced with different approaches needs to be compared
- Reliable FIB preparation and immediate TEM analysis after lift out are required for air-sensitive samples
- Dwell time needs to be limited due to potential beam damage under repeated EDX mapping
- The temperature during EDX analysis needs to be lowered to pause sample dynamics
Four different approaches were used to manufacture and prepare perovskite solar cells for in situ TEM heating using a focused ion beam (FEI Helios Nanolab). Sample A used a double-step conversion in vacuum, while Sample B used a double-step in the glovebox. Sample C used a double step in air. Finally, Sample D used a single-step in glovebox.
Figure 1. Thermal ramp applied to the samples.
As shown in Figure 1, increasing temperatures were used to heat the samples for a time of 30’ per step. A cooling process to 50 °C followed to freeze the degradation dynamics. EDX spectrum images were acquired after each heating step. Multivariate analysis was used to denoise the EDX data to improve the signal-to-noise ratio.
Figure 2. Changes in the HAADF signal at the perovskite / hole transporter interface and corresponding elemental maps for Pb and I after heating of sample C at different temperatures.
Figure 2 demonstrates how the degradation dynamics were different for the four samples. Samples A, B and D experienced elemental migration of lead and iodine from the perovskite layer. These aggregated into particles, which first appeared near the interface between the titania scaffold and the FTO then migrated over the surface of the lamella.
As shown in Figure 3, such aggregation was not observed with sample C. Rather, elemental mapping showed that iodine and lead infiltrated the hole transporting layer and diffused towards the metal electrode of the cell.
Figure 3. Comparison of the four samples after heating at different temperatures. Visible degradation of the perovskite layer nucleates from the perovskite/TiO2 interface; bright particles appear in A, B and D, whereas only a decomposition of the perovskite layer is visible in sample C.
This experiment demonstrated that the accuracy and fast response of MEMS-based heating holders were vital in isolating the thermal stress from the analysis phase.
References and Further Reading
- G. Divitini, S. Cacovich, F. Matteocci, L. Cinà, A. Di Carlo and C. Ducati, Nature Energy 15012 (2016)
This information has been sourced, reviewed and adapted from materials provided by DENSsolutions.
For more information on this source, please visit DENSsolutions.