Although the power conversion efficiency of lead halide perovskite solar cells has begun to match and even exceed silicon solar cells, reports of limited stability have impeded upscaling and remain a major challenge for commercialization. Recently, reverse bias stability, in which perovskite cells breakdown under reverse bias (as would be inevitably experienced in real world conditions like partial shading) has become a major concern receiving significant interest. Indeed, many now consider the reverse bias stability issue to be the most demanding durability issue for fielded solar modules. To date, studies of reverse bias stability of perovskite solar cells have been made, but with limited progress.  Meanwhile, compared with other types of stability under light, heat and humidity, reverse bias stability has not yet received enough attention from the perovskite community.

Figure 1. Partial shading resulting in solar panels with reverse bias instability (generated with ChatGPT)

I stepped into the reverse bias field in June 2022, at that time, I was struggling with landing on an appealing research topic for my postdoctoral fellowship application to secure my stay at University of Washington. One day, my PI, Prof. David Ginger, a very supportive and talented scientist, slacked me that he was impressed by Prof. Michael McGehee (University of Colorado Boulder)’s talks at the Gorden Research Conference, which showed that perovskite solar cells suffer instantaneous breakdown when subjected to reverse bias stressing. From there, I began my journey.

I was lucky, because, my collaborators shared with me a wealth of knowledge when I first stepped into the field. These knowledge are fundamental, yet can be easily overlooked. For instance, the current density passing a shaded cell in a serially-connected module will not, in theory, exceed the current density at maximum power point (J_mpp) of other unshaded cells. Therefore, it may not be meaningful to stress the cell with current density that is 10 or even 100 times higher than J_mpp.  In preparation for this paper, my collaborators and I spent nearly 15 months on the back-and-forth discussions within the team. I have to say, this is my best team work so far -  I truly appreciate all the support and help from my collaborators, all the scientific discussion with other researchers, and, all the comments from the four reviewers which significantly improved the integrity of this study.

Figure 2. The effect of different HTLs on V_rb in perovskite solar cells with (a-b) Ag and (c-d) Au electrode.

As for the scientific content of our paper, we clarified, at the beginning, that, from an device engineering perspective, there are two main strategies for improving the reverse bias stability: one is to fabricate solar cells that do not degrade when passing high reverse current (J_mpp). A second, more common approach, widely seen in commercial/well-established silicon photovoltaics, is to stabilize solar cells at high reverse bias, typically via improving the breakdown voltage (V_rb), so as to minimize the use of bypass diodes needed to protect a solar panel. During our experiments, we found that our perovskite solar cells, regardless of their different device architectures, suffer instantaneous performance loss when passing current J_mpp. This is consistent with various previous reports, and suggests to us that the first approach may not be promising in halide perovskites.

We then systematically investigated factors that could affect the stability of p-i-n structured perovskite solar cells at high reverse bias (the second approach), ranging from passivation of halide vacancies and inserting an additional electron transport layer, to systematically varying the hole transport layer (HTL) and metal electrode. Our results suggest that optimization of the HTLs and selection of electrochemically stable electrodes is of vital importance for preventing reverse bias-driven degradation of perovskite solar cells. Even in the presence of a more reactive Ag electrode, using a robust PTAA HTL improves the average V_rb to -7.6 V, as compared with that of -1 V for MeO-2PACz based perovskite solar cells. Further optimization on the metal electrode via replacing Ag with Au extends the average V_rb to exceed -15 V.

High V_rb is not the ultimate goal, stabilizing cells at high reverse bias for long time without performance loss is the goal. Our prior experiments suggested that some cells (primarily n-i-p structures with doped spiro-OMeTAD HTL) suffer very serious performance loss at very low reverse bias (i.e., -1 V) despite their high V_rb (-13 V). Luckily, in the current work, our optimized  PTAA-based p-i-n cells  demonstrate recoverable performance loss after stressing at -7 V for 9 hours both in the dark and under partial illumination, as well as good resilience to multiple-cycle stressing tests. These are very encouraging results for us.

Figure 3. Proposed degradation mechanism of perovskite solar cells under reverse bias.

Ultimately, we systematically examined why the choice of HTLs and electrode materials matters a great deal to cell stability at high reverse bias. We put forward the degradation model which is supported by various analytical studies in literature and our own characterizations. We think that, while previous reports have indicated the importance of oxidation events in dictating reverse bias stability, the field seems to have overlooked the fact that charge conservation requires that oxidation and reduction reactions occur in pairs (oxidation would not occur without pairing reduction). In this regard, we think the role of a good HTL is to slow down electron injection (reduction events), and then slow down necessarily-paired oxidation events. On this basis, the role of an electrochemically stable electrode is to replace severe Ag electrode oxidation with benign halide oxidation, thus delaying the device degradation to an even higher reverse bias regime.

Reverse bias stability in perovskite is becoming a hot topic. Our work highlights that the reverse bias stability challenge in perovskite solar cells is potentially solvable by installing bypass diode, just like commercial silicon solar panels.  We hope it will bring confidence to the whole perovskite photovoltaics community to accelerate the commercialization pace of this unprecedented solar technology.