New connection between stacked solar cells can handle energy of 70,000 suns

Sep 06, 2013 by Matt Shipman
Stacked solar cells are currently the most efficient cells on the market, converting up to 45 percent of the solar energy they absorb into electricity. But a new discovery will allow solar cell manufacturers to create stacked solar cells that can handle high-intensity solar energies without losing voltage at the connecting junctions, potentially improving conversion efficiency to better than 45 percent. Credit: Roger W. Winstead, North Carolina State University

(Phys.org) —North Carolina State University researchers have come up with a new technique for improving the connections between stacked solar cells, which should improve the overall efficiency of solar energy devices and reduce the cost of solar energy production. The new connections can allow these cells to operate at solar concentrations of 70,000 suns worth of energy without losing much voltage as "wasted energy" or heat.

Stacked solar cells consist of several solar cells that are stacked on top of one another. Stacked cells are currently the most efficient cells on the market, converting up to 45 percent of the solar energy they absorb into electricity.

But to be effective, solar cell designers need to ensure the connecting between these stacked cells do not absorb any of the solar energy and do not siphon off the voltage the cells produce—effectively wasting that energy as heat.

"We have discovered that by inserting a very thin film of into the connecting junction of stacked cells we can virtually eliminate voltage loss without blocking any of the solar energy," says Dr. Salah Bedair, a professor of electrical engineering at NC State and senior author of a paper describing the work.

This work is important because photovoltaic energy companies are interested in using lenses to concentrate solar energy, from one sun (no lens) to 4,000 suns or more. But if the solar energy is significantly intensified—to 700 suns or more—the connecting junctions used in existing stacked cells begin losing voltage. And the more intense the solar energy, the more voltage those junctions lose—thereby reducing the .

"Now we have created a connecting junction that loses almost no voltage, even when the stacked solar cell is exposed to 70,000 suns of solar energy," Bedair says. "And that is more than sufficient for practical purposes, since concentrating lenses are unlikely to create more than 4,000 or 5,000 suns worth of energy. This discovery means that solar cell manufacturers can now create stacked cells that can handle these high-intensity solar energies without losing voltage at the connecting junctions, thus potentially improving conversion efficiency.

"This should reduce overall costs for the energy industry because, rather than creating large, expensive , you can use much smaller cells that produce just as much electricity by absorbing intensified from concentrating lenses. And concentrating lenses are relatively inexpensive," Bedair says.

Explore further: Drive system saves space and weight in electric cars

More information: The paper, "Effect of GaAs interfacial layer on the performance of high bandgap tunnel junctions for multijunction solar cells," was published online Sept. 5 in Applied Physics Letters. apl.aip.org/resource/1/applab/v103/i10/p103503_s1

Abstract: The effect of the heterojunction interface on the performance of high bandgap InxGa1-xP:Te/ Al0.6Ga0.4As:C tunnel junctions (TJs) was investigated. The insertion of 30[Angstrom] of GaAs:Te at the junction interface resulted in a peak current of 1000 A/cm2 and a voltage drop of ~3mV for 30 A/cm2 (2000x concentration). The presence of this GaAs interfacial layer also improved the uniformity across the wafer. Modeling results are consistent with experimental data and were used to explain the observed enhancement in TJ performance. This architecture could be used within multijunction solar cells to extend the range of usable solar concentration with minimal voltage drop.

Related Stories

The fluorescent future of solar cells

May 09, 2013

(Phys.org) —For some solar cells, the future may be fluorescent. Scientists at Yale have improved the ability of a promising type of solar cell to absorb light and convert it into electrical power by adding ...

Recommended for you

First-of-a-kind supercritical CO2 turbine

11 hours ago

Toshiba Corporation today announced that it will supply a first-of-a-kind supercritical CO2 turbine to a demonstration plant being built in Texas, USA. The plant will be developed by NET Power, LLC, a U.S. venture, together w ...

Drive system saves space and weight in electric cars

Oct 17, 2014

Siemens has developed a solution for integrating an electric car's motor and inverter in a single housing. Until now, the motor and the inverter, which converts the battery's direct current into alternating ...

Dispelling a misconception about Mg-ion batteries

Oct 16, 2014

Lithium (Li)-ion batteries serve us well, powering our laptops, tablets, cell phones and a host of other gadgets and devices. However, for future automotive applications, we will need rechargeable batteries ...

Turning humble seaweed into biofuel

Oct 16, 2014

The sea has long been a source of Norway's riches, whether from cod, farmed salmon or oil. Now one researcher from the Norwegian University of Science and Technology (NTNU) researcher hopes to add seaweed ...

User comments : 20

Adjust slider to filter visible comments by rank

Display comments: newest first

shavera
5 / 5 (1) Sep 06, 2013
Instead of lenses, why not mirrors? I know a lot of the reflective solar plant designs are for heating and the ol' steam turbine method of generation, but couldn't one also create a nice parabolic mirror to reflect 70,000 suns into a solar cell?
massau
5 / 5 (1) Sep 06, 2013
Instead of lenses, why not mirrors? I know a lot of the reflective solar plant designs are for heating and the ol' steam turbine method of generation, but couldn't one also create a nice parabolic mirror to reflect 70,000 suns into a solar cell?


I think the efficiency of a lens is higher than the parabolic mirror. The lenses are also more compact and could be used on a roof. the mirrors might be better for high power applications.
NickFun
1.7 / 5 (6) Sep 06, 2013
Should I invest now or wait until this is proven to be more than smoke and mirrors? ')
italba
3 / 5 (4) Sep 06, 2013
...I think the efficiency of a lens is higher than the parabolic mirror...


Why? We use mirrors for large telescopes, not lenses! Optical property of a parabolic mirror are better than lenses, glass is not perfectly transparent.
MikeBowler
5 / 5 (1) Sep 06, 2013
...I think the efficiency of a lens is higher than the parabolic mirror...


Why? We use mirrors for large telescopes, not lenses! Optical property of a parabolic mirror are better than lenses, glass is not perfectly transparent.

mirrors simplify the design, i don't know if they're anymore efficient than lenses though, mirrors use two materials both of which i presume would absorb some light, that's glass and silver, but lenses use one material, glass, fewer types of materials might mean lower absorbtion throughout the spectrum
Q-Star
3.5 / 5 (2) Sep 06, 2013
I think the efficiency of a lens is higher than the parabolic mirror. The lenses are also more compact and could be used on a roof. the mirrors might be better for high power applications.


Lenses are not more efficient than mirrors, they are not more compact, the same amount of light (energy) is gathered by equal areas. Mirrors are equal or better in all respects than lenses in all parameters germane to gathering energy.

But in the most important way they far surpass lenses,, weight and cost of production. For the same size collecting area mirrors can weigh as little as 1/100th what a lens would. And mirrors are about that amount cheaper to manufacture. A sharper or shorter focus requires a VERY thick (HEAVY) lens, all ya need to sharpen or shorten the focus of a mirror is alter the manufactured shape.

And that is not praising the optical qualities where mirrors really leave lenses way behind.

That's why they don't & won't use lenses. In solar power or telescopes.
Humpty
1.3 / 5 (9) Sep 06, 2013
I have made parabolic solar concentrators....

Oooooooooooooooooo intense. I think my first one was about 600mm in diameter, and concentrated onto an area of about 40-50mm in diameter....

Counting: ONE and = cardboard in flames...

I think the ratio was about 180:1

70,000 : 1 ----- Oooooooooooooooooooooo

Even 2000:1

Or 700:1 - Oooooooooooooooooooo

That is intense.
italba
not rated yet Sep 07, 2013
... mirrors use two materials both of which i presume would absorb some light, that's glass and silver...

Not every mirror needs a glass front-plane. A steel foil plated with chrome, silver or aluminium it's a mirror too. You should protect the reflective layer with some waterproof layer, but it can be very thin and very transparent. The better mirrors built that way could reflect more than 99% of the light.
antialias_physorg
5 / 5 (1) Sep 07, 2013
A sharper or shorter focus requires a VERY thick (HEAVY) lens,

Fresnel lenses can be quite thin. While they add distortion that is something that is not relevant to solar cells.

In the end the amount of maintenance and the size of the tracking mechanism will decide what is better - as they already cost more than the materials for mirrors, lenses or the PV cells themselves over time. It seems that lenses may have an advantage here, as they allow for a more compact system (down to microdistributed systems if need be).
mbzastava
2.8 / 5 (5) Sep 07, 2013
EE here. There is a reason solar panels do not use mirrors. The physics of energy conversion, and in fact, the energy being converted, is different from the heat collectors. A solar cell down at the molecular structure it is a crystal lattice spread over an area. It is built to absorb EM waves over that surface area, not heat collected at a singular point. Heat is nothing but detrimental to the functioning of a solar cell. Mirrors do nothing but produce heat. The lenses ensure a uniform and directional layer of light to hit the solar cells. You can take it from there.

However, if this new tech could allow for the invention of say, a 1,000W stacked solar cell no larger than 1 cubic inch, and has the ability to convert waste heat into electricity as well, THEN a combination of mirrors and lenses will produce the absolute best results possible. With the way solar cells are designed today, mirrors have no place.
italba
5 / 5 (1) Sep 08, 2013
@mbzastava: "Heat" is EM energy too, you can concentrate it with lenses (made of special glass) or with mirrors. You can block it with infrared filter, or use it at will. And, from an optical "point of view", a mirror can focus light in exactly the same way a lens do. Better, a lens can focus only one "color" at a given distance, a mirror focus point is independent from the light's wavelength. Is not true that concentrate solar photovoltaics cannot be made with mirrors:
See http://phys.org/n...uns.html
CapitalismPrevails
1 / 5 (7) Sep 08, 2013
Instead of lenses, why not mirrors? I know a lot of the reflective solar plant designs are for heating and the ol' steam turbine method of generation, but couldn't one also create a nice parabolic mirror to reflect 70,000 suns into a solar cell?


Because mirrors permanently lose their efficiency when too much sand strikes them in the desert. Protective layers just take away efficiency.
antialias_physorg
5 / 5 (2) Sep 09, 2013
Because mirrors permanently lose their efficiency when too much sand strikes them in the desert.

There have been studies done on that and it actually doesn't seem to be a problem at all. The reduction in reflected sunlight is in the single digit percent range. So low, that it mostly isn't even economically sensible to invest in cleaning apparatus/services for small installations.

That said: If it WERE an issue then sand on mirrors is a lot easier to clean than replacing abraded lenses.
The few installations that are actually in the way of any kind of sand hazard all have protective shielding in case of sandstorms - or they just get turned downwards. (Sand hazard is also in deserts. There are no consumers in deserts, so it makes little sense to place solar farms there. Even the planned African solar farms for DESERTEC are situated near the coast, where it is NOT sandy.).

triplehelix
1.3 / 5 (10) Sep 09, 2013
Yay solar panels, less CO2! ...Gallium arsenide...

Great. We stop that nasty 0.5°C increase, but the entire world is covered in arsenic, which in tiny quantities is extremely fatal and toxic.

Hooray for the environmental movement!

Gotta love the mercury in those energy saving lightbulbs too!
antialias_physorg
3.7 / 5 (3) Sep 09, 2013
We stop that nasty 0.5°C increase, but the entire world is covered in arsenic

The entire world? What kind of calculation leads you to that cnoclusion?

Wait, lemme guess: The "gut calculation" kind, right?
RealScience
3 / 5 (2) Sep 09, 2013
@triple: get a grip, time for a reality check with a bit of math!

First, multi-junction-cell systems already reach 33% efficiency, so the total (not just electricity) energy needs of 8 billion people at a European per-capita energy use would take only 500,000 km2 of collector area, While that is big, it is only ~1/1000 of the surface of the earth, and ~1/300 of just the land surface.

Second, at 2000x concentration the cell area would be 2000x smaller, or only 250 km2, so you are exaggerating the area by a factor of 600,000.

Third, GaAs is nowhere near as toxic as arsenic itself (see monographs.iarc.fr/ENG/Monographs/vol86/mono86-8.pdf).

Fourth, the layer this adds is only 30 angstroms thick, or 0.75 cubic meters per 250 km2 layer = 1.5 m3 or ~10 tons that this would add to power the whole world at European levels. The cell phone industry alone uses roughly more GaAs than that every single month!
ryan_pickering
not rated yet Sep 10, 2013
Neat trick, but will likely never see the light of day on an actual house. Stacked solar cells (the super efficient ones discussed in the article) are incredibly expensive devices made out of a number of rare earth metals like gallium and indium. These materials are, as I mentioned, rare, and thus insufficient supply for the millions of solar panels this industry aims to install. Given the cost (about 300X the price of conventional silicon solar which you see on rooftops today), it is unlikely that they will be disrupting the market any time soon. This new connection solves a problem for a niche solar panel that will only be used in special circumstances like satellites, where money is no obstacle and solar capture per square inch is paramount. We only have 1 sun, why do we need a solar cell that can handle 70,000 anyways?
antialias_physorg
5 / 5 (2) Sep 10, 2013
are incredibly expensive devices made out of a number of rare earth metals like gallium and indium.

The point is that you need only 1/70000 the material (at best. Mor elikely 1/2000 or thereabouts). So you can use more expensive/rare materials (if the mirrors/lenses are comparatively cheap to ordinary PV cells of the same size). And research on these matrials isn't at an end, I'm willing to wager.

We only have 1 sun, why do we need a solar cell that can handle 70,000 anyways

It makes a lot of sense if you think in larger dimensions (concentrating PV powerplants, floating powerplants...and for space exploration anything that moves closer to the sun, as "one sun" is an intensity measure at 1AE distance)
RealScience
1 / 5 (1) Sep 10, 2013
@ryan - If stacked cells are ~300x the cost and 2x the efficiency, then 2000x concentration of sunlight produces ~2*2000 / ~300 = > ten times the power per dollar of cell. Thus the cells themselves cost LESS per Watt.

However high concentration requires tracking the sun in two dimensions. Although roof-top trackers have been built, they lose much of the morning/evening advantage of tracking, so you are unlikely to see these cells on the roof of a house. Instead these cells are designed for large-scale installations in high-sun areas, where the morning/evening advantages of tracking offset the loss of diffuse light.

As the cost of the semiconductors in solar decreases over time, the cost of the glass, steel and installation labor become more important, and systems twice as efficient have half the area of glass and steel. Whether this is enough to offset the cost and complexity of tracking is an open question.
gotkb
not rated yet Sep 25, 2013
I continue to be amazed about the lack of understanding shown by even the EE comment although I do very much appreciate those who point out the obvious- that Sunlight is nearly free, and concentrating it is also really cheap especially if for example on a large manmade island that can using off the shelf oil rig stabilising systems follow the sun, literally, extending the days by chasing sunset's then turning around and racing back all night long to shorten the night and maximise the harvest of rather Day's light.

Ocean mounted 'vessel's surface like this allow for water cooling as well and won't concentrate heat in one place like coastal nuke's are rumored to even if they exist too few. A solar powered city can be miles across and instead of satellite dishes remember the raTio is squared, yes a pure integral exponent! (8 inch ;fresnel'-etc. 'disc'-'dish' has 64 time's as much light as one inch). Not outerspace, but on ocean's do we best of our Sun pass the global cooling test.