Nanostructures triple organic solar cells efficiency

Dec 06, 2012
A conventional solar cell, left, reflects light off its surface and loses light that penetrates the cell. New technology, right, develop by Princeton professor Stephen Chou and colleagues in electrical engineering, prevents both types of loss and is much thinner. Credit: Illustration by Dimitri Karetnikov

Princeton researchers have found a simple and economic way to nearly triple the efficiency of organic solar cells, the cheap and flexible plastic devices that many scientists believe could be the future of solar power.

The researchers, led by electrical engineer Stephen Chou, were able to increase the 175 percent by using a nanostructured "sandwich" of metal and plastic that collects and traps light. Chou said the technology also should increase the efficiency of conventional inorganic , such as standard silicon solar panels, although he cautioned that his team has not yet completed research with inorganic devices.

Chou said the research team used nanotechnology to overcome two primary challenges that cause to lose energy: light reflecting from the cell, and the inability to fully capture light that enters the cell.

With their new metallic sandwich, the researchers were able to address both problems. The sandwich – called a subwavelength plasmonic cavity – has an extraordinary ability to dampen reflection and trap light. The new technique allowed Chou's team to create a solar cell that only reflects about 4 percent of light and absorbs as much as 96 percent. It demonstrates 52 percent higher efficiency in converting light to than a conventional solar cell.

That is for direct . The structure achieves even more efficiency for light that strikes the solar cell at large angles, which occurs on cloudy days or when the cell is not directly facing the sun. By capturing these angled rays, the new structure boosts efficiency by an additional 81 percent, leading to the 175 percent total increase.

The physics behind the innovation is formidably complex. But the device structure, in concept, is fairly simple.

The top layer, known as the window layer, of the new solar cell uses an incredibly fine : the metal is 30 nanometers thick, and each hole is 175 nanometers in diameter and 25 nanometers apart. This mesh replaces the conventional window layer typically made of a material called indium-tin-oxide (ITO).

This electron microscope image shows the gold mesh created by Chou and colleagues. Each hole is 175 nanometers in diameter, which is smaller than the wavelength of light. Credit: Image courtesy of the Chou lab

The mesh window layer is placed very close to the bottom layer of the sandwich, the same metal film used in conventional solar cells. In between the two metal sheets is a thin strip of semiconducting material used in . It can be any type – silicon, plastic or gallium arsenide – although Chou's team used an 85-nanometer-thick plastic.

The solar cell's features – the spacing of the mesh, the thickness of the sandwich, the diameter of the holes – are all smaller than the wavelength of the light being collected. This is critical because light behaves in very unusual ways in sub-wavelength structures. Chou's team discovered that using these subwavelength structures allowed them to create a trap in which light enters, with almost no reflection, and does not leave.

A key part of the new technology is a thin gold mesh, which serves as a "window" layer for the solar cell. Credit: Image courtesy of the Chou lab

"It is like a black hole for light," Chou said. "It traps it."

The team calls the system a "plasmonic cavity with subwavelength hole array" or PlaCSH. Photos of the surface of the PlaCSH solar cells demonstrate this light-absorbing effect: under sunlight, a standard cell looks tinted in color due to light reflecting from its surface, but the PlaCSH looks deep black because of the extremely low light .

The researchers expected an increase in efficiency from the technique, "but clearly the increase we found was beyond our expectations," Chou said.

Chou and graduate student Wei Ding reported their findings in the journal Optics Express, published online Nov. 28, 2012. Their work was supported in part by the Defense Advanced Research Projects Agency, the Office of Naval Research and the National Science Foundation.

The researchers said the PlaCSH solar cells can be manufactured cost-effectively in wallpaper-size sheets. Chou's lab used "nanoimprint," a low-cost nanofabrication technique Chou invented 16 years ago, which embosses nanostructures over a large area, like printing a newspaper.

Beside the innovative design, the work involved optimizing the system. Getting the structure exactly right "is critical to achieving high efficiency," said Ding, a graduate student in electrical engineering.

Chou said that the development could have a number of applications depending on the type of solar collector. In this series of experiments, Chou and Ding worked with solar cells made from plastic, called organic solar cells. Plastic is cheap and malleable and the technology has great promise, but it has been limited in commercial use because of organic solar cells' low efficiency.

In addition to a direct boost to the cells' efficiency, the new nanostructured metal film also replaces the current ITO electrode that is the most expensive part of most current .

"PlaCSH also is extremely bendable," Chou said. "The mechanical property of ITO is like glass; it is very brittle."

Explore further: Spider's web weaves way to advanced networks and displays

Related Stories

Nanostructures improve solar cell efficiency

May 26, 2011

To make solar cells a competitive alternative to other renewable energy sources, researchers are investigating different alternatives. A step in the right direction is through new processes that change the ...

Recommended for you

The simplest element: Turning hydrogen into 'graphene'

Dec 16, 2014

New work from Carnegie's Ivan Naumov and Russell Hemley delves into the chemistry underlying some surprising recent observations about hydrogen, and reveals remarkable parallels between hydrogen and graphene ...

Future batteries: Lithium-sulfur with a graphene wrapper

Dec 16, 2014

What do you get when you wrap a thin sheet of the "wonder material" graphene around a novel multifunctional sulfur electrode that combines an energy storage unit and electron/ion transfer networks? An extremely ...

User comments : 26

Adjust slider to filter visible comments by rank

Display comments: newest first

kcameron
5 / 5 (2) Dec 06, 2012
So what is the efficiency? Saying this cell is 52% or 175% better is meaningless unless one knows the baseline performance.
Sonhouse
not rated yet Dec 06, 2012
Well, the highest efficiency recently was about 10% so 1.75 X 10= 17% That is to say, if they knew about the 10% cells. If they think the highest efficiency was only 4% then they would be at 7%.
VendicarD
1 / 5 (3) Dec 06, 2012
"Princeton researchers have found a simple and economic way to nearly triple the efficiency of organic solar cells...

By capturing these angled rays, the new structure boosts efficiency by an additional 81 percent, leading to the 175 percent total increase." - Article

Innumeracy is rampant in the failed American state.
dav_daddy
2.3 / 5 (3) Dec 06, 2012
I thought the highest efficiency was around 30%? As far as I know those weren't in mass production.

That would put these up around 85% which is greater than an internal combustion engine so I have my doubts but if so this would be a game changer.
RealScience
5 / 5 (1) Dec 06, 2012
This looks like it replaces both the anti-reflection coating and the transparent top conductor with something that does both while blocking very little light.
As such the improvement would be highest on 'cheap' cells with mediocre AR coatings.

But even silicon cells lose a few tens of percent to reflection and a few percent in conductor shading and resistive losses; it would basically make a moderate efficiency silicon cell (~17%-20%) behave like a high-end all-black back-contact SunPower cell (23%-24%).

On CPV cells the improvement would be smaller - shading, resistive losses and reflection are each in the 3%-5% range, so a single 4% hit for all of these together would be an improvement of ~8% (still very nice).
javjav
5 / 5 (1) Dec 06, 2012
I thought the highest efficiency was around 30%? As far as I know those weren't in mass production.

No, that is for silicon cells, this article is about organic cells. Organic solar cells are much less efficient ( <10% in general) but much much cheaper.
_etabeta_
not rated yet Dec 06, 2012
"....typically made of a material called indium-tin-oxide (ITO)" phrases like this made for idiots really turn me off, but all these so-called efficiency improvement claims are totally without any meaning if the basic sunlight/power efficiency is not stated. Is it so difficult to get a figure that means something? Present top efficiency for commercial PV panels is around 18%, while PV panels used on spacecraft max at around 28% but have a prohibitive cost.
Now what I would really like to see is a $1/Watt PV panel that has a 40-50% true efficiency: Imagine a 1 meter square panel that has a 500W output!
Birger
not rated yet Dec 07, 2012
-But what is the practical lifespan of the device? If the solar cell rapidly deteriate, nothing is won.
jalmy
1 / 5 (2) Dec 07, 2012
Ok lets say the best organic cell is 10% efficient. 175% improvement would be 10x1.75&10=27.5% Im no mathematician but high school math skills tells me that much at least. So ya if you could make 28% efficient panels out of plastic it definitly would change the planet over-night. Assuming their nanopaterning technique can be scaled up to industrial sizes and timeframes.
dnatwork
not rated yet Dec 07, 2012
This looks like it replaces both the anti-reflection coating and the transparent top conductor with something that does both while blocking very little light.


I think the plasmonic guide does more than simply reduce blocking of light. It sounds like it bends light coming in at an angle and actually draws more light into the cell. A lot of work has gone into concentrating light and trying to keep panels pointed at the sun, but this thing might obviate much of the need for that.
RealScience
1 / 5 (1) Dec 07, 2012
It sounds like it bends light coming in at an angle and actually draws more light into the cell.

@dnatwork - yes, that is basically what an anti-reflection coating does. The plasmonic guide sounds like a simpler way to do this than the multiple sub-wavelength coatings or the tall sub-wavelength textures used today. And it gives a conductive top coating at the same time.

Great-sounding improvements are typically limited by cost, but this does not sound expensive (particularly for CPV cells), so there is hope that is will actually be a useable improvement.
totinen
5 / 5 (2) Dec 08, 2012
If we read the abstract of study, we see "a 4.4% power conversion efficiency (PCE) at standard-solar-irradiation, which is 52% higher than the reference ITO-SC"

So, total efficiency is about 8%.
totinen
not rated yet Dec 08, 2012
Edited
indio007
1.5 / 5 (2) Dec 08, 2012

At least once a month there is some new geometry or material that improves solar efficiency.
Researchers need to start combining techniques that improve efficiency.

A few articles comes to mind.
http://phys.org/n...ing.html

http://phys.org/n...lls.html

http://phys.org/n...lar.html
DavidW
1 / 5 (2) Dec 09, 2012
Regardless of what totinen says, because I don't understand that part…

175% over 8% (Typical Organic) for: "efficiency in converting light to electrical energy" = 14%

8% original organic efficiency + additional 14% efficiency = 22% total efficiency for Organic (this is better than typical silicon cells today)

If they get similar results with silicon cells then…

175% over 18% (Typical Silicon) for: "efficiency in converting light to electrical energy" = 31.5%

18% original silicon efficiency + additional 31.5% efficiency = 49.5% total efficiency for silicon

This is what the article implies. It further implies that the early and late day sun light which causes 18% efficient collectors to see a real world 13% does not have this amount of negative effect.
indio007
3.3 / 5 (4) Dec 09, 2012
Extraordinary light transmission through
opaque thin metal film with subwavelength holes
blocked by metal disks
Wen-Di Li, Jonathan Hu, and Stephen Y. Chou*
NanoStructures Laboratory, Department of Electrical Engineering, Princeton University, Princeton, NJ 08544, USA

http://www.prince...2011.pdf
totinen
not rated yet Dec 09, 2012
Regardless of what totinen says, because I don't understand that part…


I calculated the original efficiency of their organic solar cell without metal mesh (~2,9%) and then with metal mesh (~8%).
RealScience
not rated yet Dec 09, 2012
@DavidW: It is much easier to improve the efficiency of something with low efficiency than something with high efficiency.

Silicon cells only reflect roughly 1/3 of the incoming light, so a perfect anti-reflection coating could only give back-contact silicon cells a 50% boost Add in the few-percent conductor resistance and shading and the maximum possible for cheaper front-contact silicon cells would be a ~70% boost.

40%-efficient CPV cells only reflect ~4% of the useable wavelengths and have 4% conductor shading and resistive losses, so even a perfect coating could only give a ~30% boost.

Those are theoretical maxima for a perfect coating with zero reflection and zero resistance, so the actual coating won't meet these maxima (see prev post).

But any boost is welcome if cost effective. A 50% fill of 30 nm gold is ~0.3 grams/m2, at $50/g this is $15/m2 which would be ~~$0.33/W for boosting silicon (plus 'low-cost' patterning). And for CPV the gold cost would be near zero.
Sonhouse
not rated yet Dec 10, 2012
"....typically made of a material called indium-tin-oxide (ITO)" phrases like this made for idiots really turn me off, but all these so-called efficiency improvement claims are totally without any meaning if the basic sunlight/power efficiency is not stated. Is it so difficult to get a figure that means something? Present top efficiency for commercial PV panels is around 18%, while PV panels used on spacecraft max at around 28% but have a prohibitive cost.
Now what I would really like to see is a $1/Watt PV panel that has a 40-50% true efficiency: Imagine a 1 meter square panel that has a 500W output!
You would not get 500 watts per square meter unless you were on top of the atmosphere, 100 km or so up. On the surface of the Earth you lose at least half the incoming energy so you could get MAYBE 250 watts with 50% cells. And THAT would only be during the peak hours, since it drops off radically in the afternoon, even if you have aiming structures.
RealScience
not rated yet Dec 10, 2012
@Sonhouse - The solar constant is 1367 W/m2 at the top of the atmosphere, and a really clear day can reach 1000W/m2 here on earth, so a 50% module could produce 500W/m2 peak output.
Output would indeed drop in the morning and afternoon (or early morning and late afternoon even with tracking).
sirchick
not rated yet Dec 11, 2012
What are the complications that make it hard to get a good percentage in efficiency ?
vike
not rated yet Dec 14, 2012
@VendicarD:
"Princeton researchers have found a simple and economic way to nearly triple the efficiency of organic solar cells...

By capturing these angled rays, the new structure boosts efficiency by an additional 81 percent, leading to the 175 percent total increase." - Article

Innumeracy is rampant in the failed American state.

Yes, but on whose part? You left out the part about the initial 52 percent efficiency gain, before the ADDITIONAL 81% gain ABOVE THAT. Those two gains multiply, not add. So:

1.52 * 1.81 = 2.7512

. . . which is a 175.12 % increase over the implied baseline of 1.00, so what's wrong with that?

Unless you're saying that calling x2.7512 "nearly triple" is evidence of innumeracy (as opposed to rough numbers and/or a wee bit o' hype), I'm not sure what your point is.
Sonhouse
not rated yet Dec 14, 2012
@Sonhouse - The solar constant is 1367 W/m2 at the top of the atmosphere, and a really clear day can reach 1000W/m2 here on earth, so a 50% module could produce 500W/m2 peak output.
Output would indeed drop in the morning and afternoon (or early morning and late afternoon even with tracking).

Well good luck getting 1/2kw/m^2 here in Pennsylvania where the weather is more like England:) Maybe in the desert, used to live in Scottsdale so maybe but I thought a lot less made it through.
RealScience
not rated yet Dec 14, 2012
@Sonhouse:
Agreed on the AVERAGE that makes it through in the North Eastern U.S., but the PEAK (~ noon on a clear day) is quite good.

In the deserts even the average is not too bad:
http://www.nrel.g...-res.jpg

For direct sunlight the difference is even more pronounced:
http://www.nrel.g...-res.jpg
dpaige
not rated yet Dec 29, 2012
I have a much simpler question than the efficiency %.
How is it that if the holes are smaller than the wave length of light so the light is trapped - ever lets light through in the first place. In one comment it is even described that it bends the rays in toward the surface in low or indirect light conditions. This sounds like a one way mesh: lets light in, but can't escape.

to follow up, has anything like this been observed on the surface of leaves? Maybe where efficiency is more critical i.e. ferns.
RealScience
not rated yet Dec 29, 2012
@dpaige: A simple explanation is that the photons couple with electrons in the metal, and the electrons shuttle the photons through the holes. This light-through-tiny-holes effect is part of a field called 'plasmonics', and it surprised a lot of people when it first came out a few years ago.

Regarding the surface of leaves, plasmonics requires free electrons to couple with the photons, so it generally involves metals. In leaves the closest thing that I know of would be that on a molecular level chlorophyll captures photons and uses the energy to shuttle electrons around, which is somewhat similar.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.