## UD-Led Team Sets Solar Cell Efficiency Record of 42.8%; Joins DuPont on $100M Project ##### 28 July 2007  The lateral solar cell architecture with a specially designed concentrator contributes to the enhanced performance. Click to enlarge. Using a novel technology that adds multiple innovations to a very high-performance crystalline silicon solar cell platform, a consortium led by the University of Delaware has achieved a record-breaking combined solar cell efficiency of 42.8% from sunlight at standard terrestrial conditions. That number is a significant advance from the current record of 40.7% announced in December and demonstrates an important milestone on the path to the 50% efficiency goal set by the Defense Advanced Research Projects Agency (DARPA). In November 2005, the UD-led consortium received approximately$13 million in funding for the initial phases of the DARPA Very High Efficiency Solar Cell (VHESC) program to develop affordable portable solar cell battery chargers.

Combined with the demonstrated efficiency performance of the very high efficiency solar cells’ spectral splitting optics, which is more than 93%, these recent results put the pieces in place for a solar cell module with a net efficiency 30% greater than any previous module efficiency and twice the efficiency of state-of-the-art silicon solar cell modules.

As a result of the consortium’s technical performance, DARPA is initiating the next phase of the program by funding the newly formed DuPont-University of Delaware VHESC Consortium to transition the lab-scale work to an engineering and manufacturing prototype model. This three-year effort could be worth as much as $100 million, including industry cost-share. Allen Barnett, principal investigator and UD professor of electrical and computer engineering, and Christiana Honsberg, co-principal investigator and associate professor of electrical and computer engineering led the research. The two direct the University’s High Performance Solar Power Program and will continue working to achieve 50% efficiency, a benchmark that when reached would mean a doubling of the efficiency of terrestrial solar cells based around a silicon platform within a 50-month span. The highly efficient VHESC solar cell uses a novel lateral optical concentrating system that splits solar light into three different energy bins of high, medium and low, and directs them onto cells of various light sensitive materials to cover the solar spectrum. The system delivers variable concentrations to the different solar cell elements. The concentrator is stationary with a wide acceptance angle optical system that captures large amounts of light and eliminates the need for complicated tracking devices. Modern solar cell systems rely on the concentration of sunlight. The previous best of 40.7% efficiency was achieved with a high concentration device that requires sophisticated tracking optics and features a concentrating lens the size of a table and more than 30 centimeters, or about 1 foot, thick. The UD consortium’s devices are potentially far thinner at less than 1 centimeter. This is a major step toward our goal of 50% efficiency. The percentage is a record under any circumstance, but it’s particularly noteworthy because it’s at low concentration, approximately 20 times magnification. The low profile and lack of moving parts translates into portability, which means these devices easily could go on a laptop computer or a rooftop. —Allen Barnett Honsberg said the advance of 2 percentage points is noteworthy in a field where gains of 0.2 percent are the norm and gains of 1 percent are seen as significant breakthroughs. Many of us have been working with programs to take us to a real photovoltaic energy future. This project is already working in that future. DARPA has leapfrogged the ‘conventional,’ demonstrating that creativity and focus can significantly accelerate revolutionary research-bench concepts toward reality, demonstrating this does not have to take decades. This is a first step—but a significant one in making sure our energy future is what we know it should look like. —Lawrence L. Kazmerski, director of the US Department of Energy’s National Center for Photovoltaics at the National Renewable Energy Laboratory Barnett and Honsberg said that reaching the 42.8% mark is a significant advance in solar cell efficiency, particularly given the unique small and portable architecture being used by the consortium and the short time—21 months—in which it was developed. During the first 21 months of the VHESC program, a diverse team of academia, government lab and industrial partners, led by UD, was focused on developing the technology basis for a new extremely high efficiency solar cell. The rapid success of that effort has enabled the present transition to a focus on prototype product development. The team’s approach provides for affordability and also flexibility in the choice of materials and the integration of new technologies as they are developed. Barnett credits the early success of the program to the team approach taken to solving the problem. Partners in the initial phase included BP Solar, Blue Square Energy, Energy Focus, Emcore and SAIC. Key research contributors included the University of Delaware, National Renewable Energy Laboratory, Georgia Institute of Technology, Purdue University, University of Rochester, Massachusetts Institute of Technology, University of California Santa Barbara, Optical Research Associates and the Australian National University. The newly formed DuPont-University of Delaware VHESC consortium will be made up of industrial partners, national laboratories and universities. (A hat-tip to Marty!) Resources: ### Comments Stan, I suggest you retake your college courses on thermodynamics. There are so many flaws and misunderstandings in your arguments thar I don't even know where to begin refuting them. Good day. Stan, your argument is overly complicated and yet all wrong. A solar panel is not much different then black road which is all over the world. The only difference is that the solar panel makes electricity that would have been generated in some other way that would probably generate more heat. Burning coal for example makes heat way beyond the electricity it makes and co2 to cause global warming. The heat generated by the electricity's use is not a factor, it is already happening and is the same no mater the source. In all a solar panel is far less dangerous for global warming then a simple asphalt road. Mr. Peterson, In what way would the heat that is generated from the output of the solar cells have less of an opportunity to radiate back into space? You are just spreading FUD among the uneducated. The only problem that I can foresee is if a large enough concentrated plot of land is covered with very efficient solar cells and the electricity is transported elsewhere at high efficiency. Such a giant solar cell farm would be enough to cool deserts and thus disrupt local climates. That is why you create many smaller localized installations - i.e. on roofs and other man-made structures. Besides it would be extremely difficult and expensive to have a power grid transport that much electricity from such a small area to hundreds of destinations each of them thousands of miles away. Distributed (non obstructive roof top mounted) high efficiency (40%) low cost ($1/watt) PVs may be a very smart and clean way to supply 50% + of the electricity required for our homes, PHEVs and BEVs. A 4KW system could produce an average of about 20 KWh/day and would require about 10 X 1 square meter, high efficiency, panels.

The installed cost of this sype of system would be about $8k to$10K per residence. For those of us who think that this is too much to pay for clean power, it represents about the home price inflation for ONE YEAR during the last 5 years. Inflation did not contribute to the reduction of GHG but we all voluntary paid for it.

Eventually, domestic energy back-up storage units could (if desired) capture excess power and further reduce grid consumption during low or no sunshine hours.

People (10 million +) with oversized systems could sell excess power to the grid.

Multiply similar systems by 50 million and about half the current USA coal fired power generating plants could be closed.

With 100+ million PHEVs and BEVs (i.e. = 2 per home) about half the current oil refineries could be closed.

What use are very efficient panels if they are not cheap? If the watt per dollar are not low enough its not going to be economically competitive.

Right now solar systems are quite expensive. Installed by a pro $8 a watt would be low. It needs to get down below$3 a watt and then it will take off. I am thinking of putting up a system now even at the high prices.

Man, I haven't seen such a pure psychobabbling intellectual meltdown like that in ages!  Stan Peterson has just jumped the shark.

I think it's worth it to take a look at what a PV-powered world would really mean.  For the sake of argument, let's assume that the entire electric supply of the world comes from PV, that the world consumes 4x as much as the USA alone, and the USA consumes 5000 TWH/year (roughly 25% more than at present).  Let's also assume that PV supplies, as electricity, 25% of the energy pumped out of the ground as oil (8.5e7 bbl/day @ 6.1 GJ/bbl).  The PV panels are 15% efficient, absorb 100% of the incident sunlight, and replace surfaces with an average albedo of 0.30.  If the heat is distributed evenly, how much does the world heat up?

First, the average power:  20,000 TWH/yr is 2.28 TW, and 25% of 85 million bbl/day * 6.1 GJ/bbl is 1.50 TW for a total of 3.78 TW.  Harvesting this at 15% efficiency requires intercepting 25.2 TW of sunlight.  30% of this would have been reflected before, so the additional heating is 7.56 TW.

The Earth has a surface area of about 511 trillion square meters.  7.56 TW of power over this area is 14.8 milliwatts per square meter.  If the emissivity is 0.5 (allowing for the greenhouse effect), the temperature increase to radiate this extra heat is 0.0148/(4*0.5*280K^3*5.67e-8) which equals....

(drum roll)

0.006°K.

Now, I haven't checked this with a unit analysis, and my differentiation of the blackbody radiation equation might be wrong.  But I'll bet a case of beer that I'm within an order of magnitude.

Engineer-Poet, no bets from me, that looks about right!

Even if it was a problem though, remember that reflecting more light to regulate the Earth's temperature is trivial - just paint the deserts white.

Stan,

You misuse the term "Cassandra." Cassandra of Troy was given the ability to prophesy future events but with the curse that all the events would be bad and she wouldn't be believed. So she was right but no one believed her.

A certain divinity school dropout is wrong but many people believe him.

Since I attended the first Earth Day observance (1969) and learned about the coming ice age, raw material shortages, and mass starvations that would inevitably occur by the 1990's, I have concluded that people just like bad news and there are always those who will supply it. Their mantra is "all change is bad" unless it hurts those people or institutions that they don't like.

Considering that the "normal" climate of the last few 100,000 years is "ice age" and that we are currently enjoying a little break from massive ice sheets in our backyards, I have to think that if we can warm the earth and keep it warm, then that's a good thing. Let's keep that CO2 coming; plants love it. During past warm periods, plants (including agricultural plants) thrived farther north than they do today and equatorial jungles spread, too. Cold is death; warmth is life.

Meanwhile, let's keep looking for ways to get off the crude that isn't endless and that funds people who want us dead. No nation or state has ever conserved their way out of shortages (rationing is institutionalized shortages) and I'm not expecting that to work now. New energy sources are needed to avoid a return of the Dark Ages when the middle eastern oil fields play out.

Solar cells? Great idea. Where I live there are 300 sunny days a year. Just get the cost down to something that starts coming out ahead in a few years and not long after I'm dead. While we're at it, I want a car I can plug in and use for local driving without burning fuel. But you will have a very hard time powering industry with solar cells until heavy industry is willing to relocate to wide open spaces.

That's why I come to Green Car Congress; I'm looking for good news that will enhance my quality of life and give us the ability to improve the quality of lives in the third world. It is better to invent the future than to surrender to it.

@ HealthyBreeze -

PV panels that rely on heliostat arrays aren't operated at high temperatures, they are thin structures whose rear faces are aggressively cooled. You do have a point, however, in that the there has to be a temperature gradient in the thickness direction that drives the heat transfer. That means the exposed and rear faces experience differential expansion. If the panels are too stiff and the coefficient of thermal expansion too high, thermal stresses could break the panels.

Yeah, I'd love to have these on my roof, but high-efficiency solar cells are currently so expensive only NASA can afford them.

At today's prices, it's more like $10,000 per installed kW of PV for us regular folk. Given those prices you also look heavily at wind charts and local height ordinances. :) When, The article states they're moving into prototype phase soon. This design paradigm change for production purposes is 3yrs off according to the article. "The consortium's goal is to create solar cells that operate at 50 percent in production, Barnett said. With the fresh funding and cooperative efforts of the DuPont-UD consortium, he said it is expected new high efficiency solar cells could be in production by 2010." “By integrating the optical design with the solar cell design, we have entered previously unoccupied design space leading to a new paradigm about how to make solar cells, how to use them, and what they can do.” also, anticipating future cost... "The team's novel approach provides for affordability and also flexibility in the choice of materials and the integration of new technologies as they are developed." Since a heavyweight like Dupont is involved now, along with Defense backing, the cost should go down as they ramp up production in the future. Duponts involvement will promote an eventual lowering of cost due to their manufacturing conglomerate worldwide. They'll want to push this technology downline to all solar manufacturers. This shows competition is heating up for solar dollars. The goal is 50%, but if you read the article, the scientist believe they can achieve higher results in the future due to a paradigm shift in design techniques. Question for Stan or other engineers, How much heat increase does dark roofing material add? Does not roofing absorb heat? This new technology will actually decrease the size required for power generation. Whats the difference between putting these on a roof and the roof itself for absorbing heat? Well just about everybody is hitting the nail on the head (with the exception of Stan, yikes...) but not coming right out and saying it. As far as I'm concerned efficiency levels of modern day solar cells are wonderful. Besides, most people will never be able to go out and buy THE most efficient cells on the market, they just aren't cost effective. That's why most commercial cells for home use are still at about 25% efficiency. Kinda works like buying a graphics card for your computer; hey if you can afford to go out and spend$500 dollars for an Nvidia 8800GTX, but I digress...

The real effort needs to be made in the area of industrial processes; i.e reducing the COST of solar cells, not increasing efficiency. I'm not too impressed with news reports that someone increased efficiency by an extra few percent. I'll probably never even consider using those arrays. But if someone finds a way to make 20% efficient cells for half the price, that would have an immediate and significant impact on the consumer market for solar cells.

Once again, just my two cents.

Peace,
Cosmo

Saw a Discover or Science channel show last week about processess using compounds applied like ink to plastics or glass and mass producable with a goal of ten cents per Watt within 18 months.
This is in MIT country.

There is a company now formed to proceed. It has a name like Kornack or something similar.

Anyone tracking that? If so, please let me know at
bobsomm@aol.com
Peace and enjoy.

Better not let Dick Cheney know we can get such efficiencies from solar or he'll send a rocket into the sun to blow it up.

The comments to this entry are closed.