Tumblr Icon RSS Icon

By 2015, we could see a new generation of photovoltaic technologies, including 3D solar cells

By Stephen Lacey  

"By 2015, we could see a new generation of photovoltaic technologies, including 3D solar cells"


google plus icon

Solar3DSolar photovoltaics (PV) is one of the hottest high-tech areas around.

For instance, 3-D technology is all the rage these days.  Now, one company thinks it can bring the concept to solar too.

Santa Barbara-based Solar3D is working on a silicon “microcell” at the nano scale that uses an optical element to direct sunlight into a walled-in structure, thus capturing more photons and increasing the amount of electrons that are discharged. If a traditional solar cell is the ceiling of a room, the 3-D solar cell would be the room itself with the optical element acting as a skylight.

Jim Nelson, the CEO of Solar3D, says the cell could theoretically be 25% efficient. (A traditional silicon-based cell is usually in the 12-15% efficiency range, meaning that 12-15% of the sunlight hitting the cell will be turned into electricity. High efficiency cells are now reaching above 19%, with record cells hitting over 24% in the lab.)

But let’s acknowledge the key word here for the 3-D solar technology: “theoretically.”

Although Solar3D says it is ahead of schedule in development of its cell, it won’t be done with its working prototype until later this year. And it won’t be until 2012 when the company actually plans to ramp up production – assuming its prototype works as expected.

Nelson, who stopped by our office this week to chat with Climate Progress about the company, is a firm believer that next-generation technologies are the key to scaling solar photovoltaics. He doesn’t see the current crop of conventional thin films and silicon-based panels as adequate. And, like the leaders of many early-stage companies, he would shift resources from the government’s project development incentives (grants and loan guarantees) to its competitive research and development programs.

His view is representative of a common debate in renewables generally and solar PV specifically: Which is better – focusing on R&D, manufacturing, or project-level deployment to bring down costs?

“In an ideal world, we’d have an appropriate amount of all three,” says GTM Research Senior Solar Analyst Shyam Mehta. “You need all three for different reasons. Spurring innovation is a combination of focusing on existing technologies and coming up with new ones – but clearly deployment-based incentives give early-stage companies a chance to scale.”

One of the world’s largest solar manufacturers, First Solar, is a great example of this. The company began working on cadmium-telluride thin film modules in 1984. But it wasn’t until the 2000′s, when Germany implemented a feed-in tariff to encourage rapid solar development, that First Solar was able to scale manufacturing to dozens of MW of capacity. A couple years ago, the company became the first ever to manufacture a module below $1 per watt and reach over 1,000 MW of production capacity.

“First Solar’s success had a lot to do with factors outside the company’s control. Had there not been robust demand for projects in Germany, they wouldn’t have been able to scale up like they did,” says Mehta.

The history of solar has been filled with all kinds of innovative concepts and technologies, many of which have never taken off because of technical problems or unforeseen capital requirements when scaling. We could go down the list of solar inks, paints, plastics and variations of thin films that have been called “revolutionary,” that are still mostly in the lab, but the list would be very long (see this story on a potentially 90%-efficient cell.)

That’s not to say many of these important innovations won’t have a material impact in the future; just that they take much longer to scale than is often thought. And competing with the current crop of silicon-based products “” which are still in the process of substantial technical progress “” is tough.

Here’s another recent example: Five years ago, a number of companies were claiming that high-efficiency thin film products called CIGS (Copper Indium Gallium Deselinide) would soon dominate solar. At that time, the world was facing a shortage of silicon and technologies that didn’t use silicon were very attractive to investors.

However, major companies that looked promising (Heliovolt, Miasole, Solyndra) had troubles controlling costs and getting their manufacturing processes right. Even with billions of dollars invested in R&D from venture capitalists and a dramatic increase in overall global solar installations, companies still struggled. Why? Because it takes more than just a “revolutionary idea” to bring a technology to market. It takes many years of market experience for a technology to mature.

That’s what’s happening in CIGS today. Learning from earlier troubles, a handful of CIGS producers are starting to crank out product. In fact, a company not in the original high-profile pack of companies, Solar Frontier, just opened a 1 GW capacity plant in Japan. Even with that development, global capacity will be about 3 GW sometime next year – still small compared with the 50 GW of total production capacity. But the CIGS sector is now starting to hit its stride.

If next-generation thin film technologies are beginning to run, conventional silicon-based PV has been in a full-on sprint: Helped along by lower-cost Chinese producers, those technologies represented more than 85% of total global production in 2010. This dominance came about not because governments “picked” silicon PV as the wining technology; it happened because the technology was best fit to meet the rapid deployment goals in countries around the world.

“Conventional PV was there and ready to meet demand,” says GTM Research’s Mehta. “But eventually, there will be a technology that comes in and takes its place. And that may not be as far away as people think.”

Indeed, after years of consistent cost reductions, conventional PV manufacturers are starting to “get stuck” and are finding they have fewer options, says Mehta. By 2015, he says, many of the costs may have been squeezed out of the current suite of conventional silicon-based manufacturing lines.

“Right now, silicon is dominating. But sooner or later, we might be facing more of a revolution than an evolution like we’ve seen.”

That could be good news for companies like Solar3D. If the company has a cell that actually works, it seems to have a decent plan to deploy the technology in an efficient way. Nelson says that the cells are designed to be produced on existing manufacturing lines, and can be dropped into an existing module. That would take away the need to create a whole new set of manufacturing equipment (one of the problems with CIGS) and allow them to substantially increase the efficiency of panels.

“The whole idea is to reduce costs and maximize efficiency within lines we have now,” says Nelso.

Those cells need to get developed first – highlighting the need for a solid focus on research grants, lab partnerships and competitive prizes for innovative technologies. But as history has proven, they won’t get anywhere without a reason to manufacture – which means continued incentives for deployment are important as well.

That’s why we need a combination of all three types of incentives: funds to encourage more R&D and get the next generation of technologies primed for market; incentives to scale up manufacturing and newly-commercialized technologies a chance to scale; and deployment incentives that provide the most cost-competitive products with a path to market.

“It’s not an either/or,” says Mehta. “They’re all important for different steps of the technology chain.”

Stephen Lacey

Related Post:

‹ Conservative senators parrot Big Oil’s talking points

Bill McKibben: Wildfires And Spills In The Canadian Tar Sands ›

16 Responses to By 2015, we could see a new generation of photovoltaic technologies, including 3D solar cells

  1. Mike Roddy says:

    It seems like I hear about a PV breakthrough about once a month, either in cost or efficiency. One of them is going to work some day, though they tend to say they are 5 years from deployment.

    Since better components are on the way, it’s important to design rooftop and utility scale arrays to easily accept modular retrofits. The solar industry is not cooperating in that way today, using different dimensions for both panels and components. The cost of everything else involved in building solar usually exceeds the cost of the panels themselves, so retaining existing installations as much as possible is important for economic as well as ecological reasons.

    The other issue is that utilities and customers use news of breakthroughs- but not commercial deployment- as an excuse to not order them now. This is a mistake, partly because rebates and credits are being phased out already in light of these cost reductions.

  2. Matter says:

    “the cell could theoretically be 25% efficient.”

    You’re right to point out the ‘theoretically’. Silicon solar cells can ‘theoretically’ be 31% efficient. CdTe closer to 33%.

    But FirstSolar’s modules are still stuck at around 11% because the real world is different!

  3. Robert says:

    I am thinking of installing PV on our house (in the UK) and to that end spent a couple of hours this morning with a solar company. The thing that staggered me most was the level of feed in tariffs (FIT) – these were introduced in April 2010 and is the government’s way of chasing it’s goals under the 2008 climate change act. The table of taiffs is here:


    For retrofit PV (i.e. PV on an existing building) you get paid 43.3p per KWh, plus another 3p for exporting it. The 43.3p is paid just for generating it, irrespective of whether you use it directly or export it to the grid. That is about four times the rate of buying the same unit off the grid! The FIT in index linked to inflation and guaranteed by the government for 25 years.

    Even so I need to work out the economics to see what the payback is over the 25 year expected life. The system costs £14000 for 12 panels, each peak rated at 245W. The killer is the fact that the hardware is only guaranteed for 5 years so when it goes wrong repairs may be difficult and expensive.

    I haven’t signed on the dotted yet and am doing more research…

    (I have spotted a niche in the market – attach high efficiency LED lighting powered from the grid to illuminate the solar panels and make a profit from the FIT! (just joking))

  4. Dr.A.Jagadeesh says:

    I am confused to read every week that Solar Cell efficiency is going up. Is there not an upper limit for the attainable Solar Efficiency?

    One news item reads:”A researcher at the University of Missouri is making some eyebrow-raising claims, including that a new technology he has developed could lead to solar panels that convert 90 percent of the sun’s energy to electricity”.

    In Wind there is Betz Coefficient:

    Betz’s law is a theory about the maximum possible energy to be derived from a “hydraulic wind engine”, or a wind turbine such as the Éolienne Bollée (patented in 1868), the Eclipse Windmill (developed in 1867), and the Aermotor (first appeared in 1888 to pump water for cattle, and is still in production). Decades before the advent of the modern 3-blade wind turbine that generates electricity, Betz’s law was developed in 1919 by the German physicist Albert Betz. According to Betz’s law, no turbine can capture more than 59.3 percent of the kinetic energy in wind. The ideal or maximum theoretical efficiency n max (also called power coefficient) of a wind turbine is the ratio of maximum power obtained from the wind to the total power available in the wind. The factor 0.593 is known as Betz’s coefficient (from the name of the man who first derived it). It is the maximum fraction of the power in a wind stream that can be extracted.
    power coefficient = Cp =( power output from wind machine) / (power available in wind ).

    I am sure there is a limit for Solar Cell Efficiency too.

    What is needed is 3D approach:

    Design, Demonstrate and Disseminate.

    Dr.A.Jagadeesh Nellore (AP), India

  5. malcreado says:

    >Since better components are on the way, it’s important to design rooftop and utility scale arrays to easily accept modular retrofits.

    I don’t know about that. These aren’t like cell phones that people like to change out every year or so. These thing are pretty static in that once installed they are left in place for many years. It would have to be a big leap forward in performance to make it cost effective to swap out panels. Why go to all the cost for only a few percentage points of efficiency boost. A homeowner probably wouldn’t and a utility would probably be better off just adding on more capacity than swapping out panels. Sure standardization / modularization would be nice and handy but far from a show stopper.

  6. ferro413 says:

    Sunpower is already selling a 20% PV cell and Suniva is not far behind. Both are in production now and have impressive track records of improving efficiencies. What’s the big deal about a cell the “could theoretically be 25% efficient” and is years away from commercialization?

  7. Mark Shapiro says:

    What Mike Roddy said. Your DOE NREL graphic shows that the cost of PV panels already hit a low $2/Watt or so.

    But installation costs another $2-3/Watt, and the inverters etc cost another $2-3/Watt. Let’s attack those costs with:

    1) Building integration (BIPV) with standard sizes, shapes, and installation processes; and

    2) Electrical integration (a DC standard) connecting DC PV panels directly to DC household uses — all electronics, and now LED lighting.

    PV cell costs are still dropping, but they are already lower than the “boring” stuff — installation and inversion. To address those, we don’t need any new technology, but we do need standards. Standards speed deployment.

    We all remember how important deployment is, don’t we?

  8. Mike # 22 says:

    Good summary of 2010 installed costs ($3.40), 2016 installed costs goals ($2.20) and the dollar a watt ($1) initiative http://www1.eere.energy.gov/solar/sunshot/pdfs/dpw_white_paper.pdf (page 10)

  9. Mark Shapiro says:

    Mike # 22 –

    Thanks for pointing out DOE’s excellent $1/Watt initiative. It’s terrific – a must read.

  10. Mike Roddy says:


    You’re correct, the panels probably won’t be replace in 5 or even 10 years due to improved efficiency or anything else. It’s more an issue of marketing. People like to know that they can upgrade without too much trouble.

    Mark Shapiro, you must also have construction experience. I worked in steel framing for many years. The steel component manufacturers had a lot of trouble approaching wood framing costs until they made their products commodities. This is even more true for the fastening and electrical components that you describe for distributed solar.

    The solar trade association needs to get on the ball here, as the steel and wood industry did for construction products. It’s not an issue of each company making identical products, only similar dimensions and installation designs. Then, we will really see some price reductions.

  11. malcreado says:

    >Building integration (BIPV) with standard sizes, shapes, and installation processes; and . . .

    . . .and standardized roofs and we all live at the same latitude too I suppose. So barring that there is going to be some manual setup to the process. Which means that to drive down costs further there needs to be more than one or two installers in your area so they have to compete for business, and the best way to get that is as Joe says deployment, deployment, deployment. The solutions are easy; they are just hard to do; but don’t let that stop you.

  12. Barry says:

    I want to second all the comments about deployment and standardization.

    Our family wanted to do grid-tie PV on our home in British Columbia. It took 2 years to find someone willing to do it and to get the electric utility to sign off on it. Crazy. Name one other product that you have to wait two years to buy after you have the money and the desire.

    What it highlighted is that it takes a loooooong time to build an install industry that is anywhere close to cost-effective and big enough to handle installing solar PV on the scale needed.

    You could have the best lab breakthrough and not be able to produce and get it on the roofs for years and years if you have neglected the deployment and standards push.

    Germans are real heroes on my book for doing the heavy lifting for years on the critical deployment task. Nice to see at least some people in the world willing to do “hard” when it comes to climate.

  13. Mulga Mumblebrain says:

    Increasing research, mass production, subsidies and feed-in tariffs equals rapid uptake. Denialists here have a subset of their moron arguments where they maintain that solar PV, or any other renewable technology, is ‘too expensive’. They leave it at that, as if it must stay expensive forever as the result of some self-evident axiom of the Rightwing pseudo-religion of market absolutism, when one need only think of any electronic innovation to know that it starts out dear and grows cheaper and cheaper. Of course the denialists have a lot of egomaniacal energy invested in their years of arguing against humanity’s best interests, the laws of physics, rationality and truth, so they are fanatically determined not to admit error, ever. It would hurt their gargantuan, but completely unmerited, self-regard, and be a victory for those ‘Leftists’, ‘progressives’ and ‘nanny-state do-gooders’ that they so passionately hate (like so many other groups of Others). Let us just pray that they have not denied, delayed and derailed progress so successfully that it is too late.

  14. Robert says:

    Mulga Mumblebrain

    Re #3 and #13 – we have very heavy subsidies here in the UK and I still cannot make it sound like an attractive investment – breakeven maybe. The feed in tariff is 43.3p/KWh (about 70 US cents) which is far far more than anyone pays for electricity.

    The system I looked at yesterday costs £14K ($22.5K) and generates about 6.6 KWh / day on average. That is about a quarter of what we actually consume in our house. On that basis it doesn’t work for me in either CO2 terms or financial terms.

    The UK Government’s scheme has some very odd features. First, it doesn’t matter what you do with the electricity you generate – you get paid anyway – so where’s the incentive to reduce consumption? Second, the subsidy is paid for by other customers of the utility not the goverenment, so basically I’m being asked to rob my next door neigbours to support a scheme that makes little sense.

    I have invested in things which definitely DO cut CO2 and save money – cavity wall insulation, loft insulation, energy saving bulbs and appliances, conservation and these have made a big difference – in our house we consume less than half the electricity than a few years ago. And we heat the house with two log burning stoves so have not needed to use central heating for over four years. (being in the UK we have never heard of air con either!).

    Summary – we have made huge CO2 cuts and big savings but I just can’t make the PV equations balance. I am also a bit suspicious that the whole mechanism is doing no more that circulating money and exporting emissions to PV factories in China.

  15. Mulga Mumblebrain says:

    Robert #14, things sound crook. We got ours, which produces all our electricity needs, for $3300, which is, at 65p to the AU$, about 2150, I suppose, in real money. But we got in on the last day of some subsidy scheme, and the feed-in tariffs are in the sights of all the authorities. I suspect that the terminal confusion is just more of that good old ‘market magic’, we have all come to love so much.

  16. spiritkas says:


    Let’s not forget the resale value of panels when you upgrade, there is going to be a big market in second hand panels…boy-o-boy does that drop the price a bit. I already have a few business ideas and uses for lower cost panels like that, and having them modular would help the second use market so you don’t need different bracketing and installation equipment for all the units. Modular is always better, to argue against it in any way is rather dull. How could a country operate if it had different electrical outlets or voltages? How would cars operate today is the gas tanks all had different fittings for putting gasoline in, how would just about anything go without ISO? Modular is green and smart.

    If economics are definitely there at some point. Over 10 years, existing panels wouldn’t have paid for themselves yet, but you can resell them and newer panels would up the sale value of a home as well. There are multiple price factors that will make more moderate increases in panel efficiency a worthwhile upgrade, certainly at the industrial scale. It might be the industrial first usage to the home usage second to keep those panels going for their full lifetimes. Modularity would make all of that happen at reduced cost. I’d wager almost any industry benefits from standards, certainly end-users benefit in terms of cost.

    Food for thought hopefully!