A solar panel is a solar panel is a solar panel, right? Not exactly. They’re kind of like cars. While all cars have wheels, a steering wheel, and pedals to control it, there are a wide variety of engines, horse power, types of tires and brakes that all impact how efficient the car is and handles the road. The same is true for solar panels. There are different materials and technologies that can have a profound impact on the amount of energy generated and the cost. And the past year or so has had some really interesting advancements that will have a big impact. Plus, what if I told you it might be possible to harvest power from shadows?
A brief history
The cost of solar power has been dropping dramatically over the past decade, while at the same time solar panel efficiency has been rising. If you look back to 1977, the cost per watt of solar energy was around $77, but today it’s around $0.13. And it’s continuing to drop. It’s worth taking a quick look at how solar evolved to give the latest developments some context. There’s a much longer history than you might have realized.
The photoelectric effect was first observed in 1839 and the first patent was awarded to William Coblentz in 1913. But it wasn’t until the 1950’s that solar power started to become a real thing. In 1954 Bell Labs invented the first practical silicon solar cell1, which had an efficiency around 6%. Solar cell efficiency is how much of the collected sunlight the cell is able to convert into electricity. In 1957, Hoffman Electronics was able to increase that efficiency to 8%, and then to 10% by 1959. By the time we get into the early 1960’s solar cells had achieved about 14% efficiency.2
Today, most solar panels are somewhere between 15% and 20% efficient. With some of the higher efficiency models that you can buy being in the low 20% range. The LG panels (LG365 Q1C-A5) I had installed on my roof are 21.1% efficient. And SunPower has a panel that’s almost 23% efficient.3 4 So commercially available solar cells have gone from about 10% efficiency in 1959 to 23% efficiency today. That’s 60 years to more than double the efficiency.
Beyond efficiency of solar panels, it’s also important to understand the materials used to create them. The primary material used in solar panels today is silicon, which can be formed in three ways: mono-crystalline, polycrystalline, and thin-film panels.
Mono-crystalline solar panels have the highest efficiency with current ratings between 15-22.2%, and a lifespan around 25-30 years. To make a mono-crystalline, or single crystal solar cell, silicon is formed into bars and then cut into wafers. The single crystal structure gives the electrons more room to move and creates a better flow of electricity.
Polycrystalline solar panels have average efficiencies between 12-18% with a 23-27 year lifespan. They’re also made from silicon, but instead of cutting bars of single crystal wafers, manufacturers melt many fragments of silicon together to form the wafers. The mixture of many kinds of crystals gives the electrons less room to move, so it’s not as efficient. But the benefit is the price because they’re cheaper to produce.5
And finally there’s thin-film, which is the least efficient between 9-14%, and a lifespan closer to 20 years.6 Instead of forming thicker, rigid wafers, this is a very thin layer that can be applied to plastic to create flexible solar panels. These are typically only seen in large scale installations where space isn’t a premium, or you need to mold a cell to the shape of something, like an RV or boat.7
So are we stuck at 23% efficient solar panels? No, but it’s not far off from the theoretical maximum efficiency of a single material. It’s referred to as the Shockley-Queisser limit and for silicon panels it’s around 30%.
But, the good news is we aren’t limited to silicon. There’s been growing research around perovskite, which is a class of man-made compounds that share the same crystalline structure as the calcium titanium oxide mineral with the same name.8 9 What makes perovskite an enticing silicon alternative is that the structure makes them highly effective at converting light photons into usable electricity. They’re capable of beating traditional mono and polycrystalline silicon solar cell efficiency, and since they’re from a man-made compound, manufacturing costs should be lower. Perovskite cells can be made through a process called "solution processing," which is very similar to the printing of newspapers … you can use inkjet printers to deposit materials on plastic sheets.10 So perovskite solar cells are another form of thin-film solar, but with much higher efficiency. And unlike silicon, you don’t have to heat it to thousands of degrees to form it.
But there are challenges around perovskite, which includes shorter lifespan, durability and toxicity. Perovskites are more sensitive to air and moisture, which can dramatically shorten their lifespan. But this may not be a showstopper since solar cells are already sealed inside of plastic and glass for protection. However if these are going to catch on they’ll need to match the 20-25 year warranty that most silicon based solar panels come with today. As for toxicity, many of the formulations include lead, which could become problematic if not handled and recycled properly. All challenges that can be dealt with, including different formulations, but it’s worth making a note of.
In the time that scientists have been researching perovskite solar cells, the efficiency has gone from 3.8% in 2009 to 25.2% this year in single-junction, or single layer architectures. If you compare that to silicon’s efficiency increase since the 1970’s, when the National Renewable Energy Laboratory (NREL) started tracking this data, it’s a dramatic achievement.11
Today’s most efficient panels, like my LG panels, are where researchers were in the lab almost 20-30 years ago. The fact that perovskite has advanced so far so quickly is promising.
But it doesn’t stop there. There’s been a lot of work around layering multiple solar technologies to do more together than they can on their own. This is called multi-junction solar. Each layer is designed to absorb a different wavelength of the incoming sunlight, so collectively they can capture more energy.12 There are a couple of companies combining silicon and perovskite tandem layers to do just that.
A San Francisco start-up, Swift Solar, and Oxford PV are both using a thin layer of perovskite film along with a more standard silicon solar cell with promising results. The silicon absorbs the red band of the visible light spectrum and the perovskite absorbs the blue spectrum.13 Oxford PV has reached a 28% efficiency and thinks they’ll be able to break the 30% milestone. They aren’t available on the market yet, but they are actively setting up a mass production line with the help from Meyer Burger, one of the largest suppliers of photovoltaic manufacturing equipment, and they’re expecting to have that complete by the end of this year.14 15 At launch they’re expecting to have a 400 watt, 60-cell module available, with a 500 watt version down the line.16 For comparison my solar panels are 365 watts.
But the biggest breakthrough is from the National Renewable Energy Laboratory (NREL). They’ve fabricated a solar cell in the lab with an efficiency of 47.1%, which set a record this year. Now, this was in a lab and used concentrated illumination, but even under “one-sun illumination,” which simulates more real world conditions, it achieved 39.2% efficiency.17
How they did it is pretty clever. This is another multi-junction cell, but instead of two tandem layers, it’s a six junction solar cell, which basically means they’re layering six different solar technology layers. In total there are 140 layers of the six different solar materials … and all combined are still less than 1/3 the thickness of a human hair.18 19 As amazing as that is, it’s still in the lab and not ready for mass manufacturing. This is helping to establish what’s possible on the high end and shows a path forward for companies and other researchers, and it proves that we should be able to go well beyond the 30% efficiency limit of a single material.
And finally a little bit of crazy science fiction level technology, and one that I wouldn’t expect to see anytime soon, but if researchers can build on this … it’s kind of crazy. What if you could generate power from shadows and not just light? Researchers from National University of Singapore have developed a prototype of a device called the Shadow Effect Energy Generator (SEG), which generates power from … you guessed it … shadows.20 The way the technology works is by generating and harvesting a small amount of electricity from the difference in contrast between the shadow and illuminated sections of the device. If the device is in full shadow or full light, it’s not generating a voltage, but the closer you get to 50% coverage the more voltage it produces. The working prototype generates about 1.2 V, which is enough energy to power a digital watch in their demonstration.
This type of technology could take advantage of passing shadows like trees or clouds on a solar panel. Today if a solar panel gets partially obstructed, it stops producing energy. A solar panel equipped with this could turn that tree’s shadow into power. It could also be put to use as collectors inside a house, which are full of passing shadows all the time. They can also double as a sensor and log shadows passing over it, which could be used in applications with smart home devices. Again, this one is a long way off from anything practical — it’s still very much in the lab — but it’s a very unique concept worth thinking about.21