It’s pretty clear that we are at an inflection point with adoption of solar power. For the last 18 years in a row, solar PV electricity capacity has increased more (as a percentage increase) than any power source. Solar now accounts for 4.5% of global power generation. Wind generation is at 7.5%, which means wind and solar combined are at 12%. By comparison nuclear is at about 10% generation globally.

Solar PV is currently the cheapest power capacity to add to the grid. Extrapolating gets more complicated as solar penetration increases because we increasingly need to consider the costs of upgrading electrical grids and adding grid storage. But wind and solar still have a long way to go. Adopting these renewable energy source as quickly as possible is helped by technological improvements that make their installation and maintenance less expensive (and material and power hungry) and increase their efficiency. Part of the reason for the steep curves of wind and solar adoption is the fact that these technologies have been steadily improving.

There are numerous research programs looking at various methods for improving on the current silicon PV cells which dominate the market. The current range of energy conversion efficiencies for the top silicon solar cells on the market range from 18.7%–22.8%. Those are great numbers – when I started following the PV solar cell industry closely in the aughts efficiencies were around 15%. The theoretical upper limit for silicon is about 29%, so the technology has some head room. Increased efficiency, of course, means more energy per dollar invested, and fewer panels needed on any specific install.

Silicon, however, is heavy, brittle, and needs high temperatures to manufacture, which has its own carbon footprint and raised the price of manufacturing. For these reasons there has been a lot of interest in discovering the material that can replace silicon for PV cells. Right now there are two leading contenders, organic solar cells and perovskite. Organics are lighter, cheaper, and more flexible, but have lower conversion efficiencies. They will likely come into commercial use, but probably for applications that supplement rather than replace rigid PV panels. Perovskite solar cells are generally considered to be the likely heirs the solar panel industry. But the technology has been tricky to perfect.

Perovskite solar cells can be much thinner than silicon, using only 1% of the raw material as silicon solar cells. They can also been manufactured at room temperature, so are much cheaper to manufacture.  But perhaps most importantly, perovskite solar cells have a higher upper limit of conversion efficiency. There is no single number for this, depending on the exact type of perovskite used (the name perovskite refers to a type of ceramic with the same crystal structure, but mostly contain calcium titanate), but already there are estimates above 40% efficiency. That would be huge, almost doubling existing commercial efficiencies – and in a thinner, lighter, and cheaper panel.

But technical hurdles remain, mainly the long term stability of perovskite. It tends to break down over time when exposed to sunlight and oxygen. The crystals are also polymorphic, and one form of the crystal structure does not convert light into electricity. A recent study, however, has found a way to stabilize the effective crystal structure and minimize conversion to the inactive structure.

Even still, for this reason it seems that the commercial PV sector is rapidly moving in the direction of silicon-perovskite tandem solar cells. This may be the solar panels of the next decade or so, until advanced organic, pure perovskite, or something else, takes their place.

One company, Oxford PV, plans to release silicon-perovskite tandem solar cells commercially this year, with a conversion efficiency of 27%. That’s a significant increase over current commercial solar cells, with an expected 15% decrease in cost. Other companies plan to make tandem solar cells starting in 2025. I have been following the perovskite story for years and this is what I have been looking for – commercial release of perovskite panels (not just another laboratory breakthrough). Hitting the market with better and cheaper panels is likely to transform the industry, and many are predicting that this will be the solar panel of the next decade. Part of the reason tandem panels are more efficient is because they are complementary – perovskite absorbs better toward the UV range while silicon is better toward the red end of the spectrum.

We can also expect continued incremental improvements from there. In the lab tandem solar cells have already exceeded 30% (I see up to 32.5%), which exceed the theoretical limit for silicon only.  The theoretical upper limit for tandem solar cells can be as high as 45%. The more the price for solar panels comes down, per energy produced, the greater demand will be. It may soon get to the point where a home owner can’t afford not to install rooftop solar and continue to pay high prices for their electricity.

As I have discussed before, there are lots of moving parts when you consider the entire energy infrastructure. You can’t just add solar panels and assume the whole thing will work. But we know what needs to be done, and we have the technology. We need to upgrade the electrical grids, make sure net metering laws favor distributed solar energy production, add grid storage, and allow electric vehicles to act as smart battery backup. We can get solar to 20-30% of capacity over the next 10-15 years. Add in wind, nuclear, hydro, and geothermal and we can phase out fossil fuel by 2050.