Can two-step vapor-deposition processes make efficient single junction perovskite and perovskite/silicon tandem solar cells? Last August we found the answer is yes, published today in ACS Energy Letters!
https://lnkd.in/eCawJ_Kp
Here, we use a PVD-CSS process to demonstrate >29% stabilized efficiency planar perovskite-silicon tandem solar cells with low hysteresis and good shelf stability. During the review process we used the same process on textured silicon bottom cells and also demonstrated ~28% efficiency for the same absorbers on textured silicon bottom cells even before additional optimization.
This article is meant to be a showcase of the capabilities of what PVD-CSS is capable of, as at the time of submission all available reports of tandems with vapor deposited perovskite absorbers were <28%. Since then, excellent work from the group of Yi Hou and a great preprint from the group of Steve Albrecht (and possibly more since I left academia last month!) have shown coevaporation processes are similarly capable of making highly efficient 2T tandems. The craziest thing is that all this work was completed in the last three months of my postdoc, as I challenged myself to make efficient tandems before leaving the world of CSS perovskites. I wish we had more time to expand on some very interesting observations along the way, but this fact shows the potential of these kinds of processes, and that there is still a lot to learn and achieve here. I want to sincerely thank all of the excellent collaborators on this work from EPFL PV-lab, Fraunhofer ISE, and CSEM, without whom I could not have finished this in the limited time I had left, especially Kerem Artuk who showed me all his tricks to make efficient and reproducible tandems.
When I started my PhD in December 2016 on two-step close space sublimation for perovskite cells, we were unsure if CSS was a viable method for making perovskite solar cells at all. Only two examples existed in the literature at the time and it took a little over two years to solve at least 100 small problems until Alex Harding made our group’s first working CSS devices with ~10% efficiency in early 2019, which then took some time to understand how to reliably replicate and improve. Since then, we as a community (shout out to Zhilliang Ku and Henk Bolink’s groups who have been producing a lot of very strong and interesting work on CSS lately) have made tremendous strides to understand how to reliably deposit films with reasonable performance and tunable band gaps and the performance gap with solution deposition techniques is slowly shrinking. Solution deposition may in the end prove the best way to make perovskites but vapor deposition seems to have a fighting chance. If you are struggling with a process like this feel free to reach out, I can probably help trouble shoot a bit after making almost every possible mistake in the past 9 years (and thousands of increasingly reliable devices).
