To anyone who has spent some time in photovoltaic (PV) power circles, the word ‘perovskite’ probably sounds familiar. Offering arguably better bandgap properties than traditional silicon cells, perovskite-based PV panels also promise to be cheaper and (literally) more flexible, but commercialization has been elusive. This is something which Oxford PV seeks to change, with the claim that they will be shipping the first hybrid perovskite-silicon panels to a US customer.
Although Oxford PV prefers to keep the details of their technology classified, there have been decades of research on pure perovskite PV cells as well as tandem perovskite-silicon versions. The reason for the tandem (i.e. stacked) construction is to use more of the solar rays’ spectrum and total energy to increase output. The obvious disadvantage of this approach is that you need to find ways to make each layer integrate in a stable fashion, with ideally the connecting electrodes being transparent. A good primer on the topic is found in this 2021 review article by [Yuanhang Cheng] and [Liming Ding].
The primary disadvantage of perovskites has always been their lack of longevity, with humidity, UV irradiation, temperature and other environmental factors conspiring against their continued existence. In a 2022 study by [Jiang Liu] et al. in Science it was reported that a perovskite-silicon tandem solar cell lost about 5% of its initial performance after 1,000 hours. A 2024 study by [Yongbin Jin] et al. in Advanced Materials measured a loss of 2% after approximately the same timespan. At a loss of 2%/1,000 hours, the perovskite layer would be at 50% of its initial output after 25,000 hours, or a hair over 2.85 years.
A quick glance through the Oxford PV website didn’t reveal any datasheets or other technical information which might elucidate the true loss rate, so it would seem that we’ll have to wait a while longer on real data to see whether this plucky little startup has truly cracked the perovskite stability issue.
Top image: Summary of tandem perovskite-silicon solar cell workings. (Credit: Yuanhang Cheng, Liming Ding, SusMat, 2021)
I mean, there are definitely use-cases for higher efficiency panels where degradation doesn’t matter. I know because I work in one: scientific ballooning doesn’t care about the lifespan of the panels, because they’re just going to die on landing anyway.
I bet the military would be happy to replace a solar panel every 3 years in exchange for double the output per pound/square foot. In combat nothing lasts more than a month anyway.
Would the decrease of 2% per 1000hrs be linear or decaying exponential? I am guessing the latter, assuming that at any given moment while being irradiated, the probability of a molecule being destroyed is constant.
That would still mean 60% after 25khrs (0.98^25).
Of maybe double that time if half if the time it is dark. Still far from ideal though…
It’s great if you are a legislator who owns shares in a solar panel factory
They aren’t damaged due to radiation, they’re damaged due to thermal/mechanical stress (and moisture absorption). The overall issue is that it takes very little energy to fracture a perovskite crystal. I mean, OK, photon absorption can obviously do stuff too, but losing molecules doesn’t matter as much as losing a large chunk of crystal.
So the kinetics aren’t exactly easy to characterize. A lot of the really stable ones show a brief exponential decrease and then linear afterwards, and in some of them the exponential’s so quick they’re effectively just linear.
Learn what a half life is. Many decay processes, from radioactivity to medicine, follow this trend.
When taking about the degradation of cells the time often refers to hours with standard radiation (IIRC 1000W/m²). With less radiation (e.g. during the night) the degradation is much lower. So instead of 2.85 years I would expect something like 7-10 years.
The questions arc can they pay for themselves within that timeframe, and how easily can they be recycled compared to a panel with an expected lifetime of 20+ years.
Honestly, PV panel with silicon aren’t already recycled correctly. How could a dual technology be recycled since you’d have to first separate both crystals and then treat them differently, all of that in a cost efficient way?
As the silicon panel underneath should have the same sort of lifespan as any other silicon panel I’m more interested in what performance loss having the degraded layer above brings – if this is a booster you can add to a panel that gives it stronger performance for 5-10 years and in the 20-30 years after that doesn’t cause massive extra losses its not a terrible idea. As long as that extra layer is not hugely expensive anyway.
Recycling being a bit more complex in several decades time isn’t the end of the world, as no matter how complex it gets if there is enough of them reaching EOL they will be massively easier to recycle than harvesting virgin materials and all the processing required to make them that pure. So its only a problem if there are only ever a handfull of these things making it not economically viable to set up the recycling pipeline.
The primary disadvantage of perovskites has always been their ENVIRONMENTAL TOXICITY.
material scientists cannot resist the sirens call of lead
It only takes one 555 LED blinker to get you hooked on those sweet, sweet fumes.
Don’t release to the environment, problem solved.
Just go nuclear 24/7, way more environment friendlier.
Solar, wind, and EVs are junk technologies and actually add toxicity to the environment given the materials needed to manufacturer and install. Tons of fossil fuels are used to mine and create these things.
Since when Texans are worried about the environment ???
By that argument all industry, and basically every single aspect of human life is junk…
That V8 also took tones of fossil fuel to source, purify, machine and assemble the raw materials, and then it goes on to use even more to work…
A solar panel need not have any fossil fuel involved in creation at all. It almost certainly does, but it isn’t a required element anywhere in the process! Then even if the whole thing was made entirely on fossil fuels unless you take a sledge hammer to it almost immediately it will pay back that energy cost potentially many many times over though its life – an upfront cost that then prevents the need to burn more fossils overall… Now actually making use of all the energy it produces being an energy source for which the output isn’t controllable or even 100% predictable is a challenge, but not an insurmountable one.
EV’s are little less clear cut right now, but they should come good in the end – the problem now is EV’s are just too new in the mass market that the recycling and repair industries haven’t built up around them the way they have in the century+ of using ICE and the power grids likely require some investment to handle the transition.