Talk of a concentrating photovoltaic (CPV) market that is about to take off will court derision from some quarters, with sceptics arguing that they’ve heard it all before. But this industry has undoubtedly matured in recent times, and there is very good reason to believe that CPV deployments will rise and rise, offering a great opportunity for makers of triple-junction cells that can fulfil the wishes of their customers. Richard Stevenson investigates.

 

In 2008 it looked like the CPV market was about to head into overdrive. Although projects in this industry, which uses mirrors and lenses to focus light onto triple-junction III-V cells, were typically for deployments of just a few hundred kilowatts, plans were in place to build far, far bigger plants. This included a staggering 154 MW installation in Mildura, Australia.

But a few months later many of these plans were in tatters, including the flagship project in Mildura. By then the credit crunch was wreaking havoc and many chastened investors were unwilling to take the risk of putting their cash behind a relatively new technology. The fledgling CPV industry was left treading water.

Since then the installed capacity of CPV across the globe has grown modestly – according to the CPV consortium, which represents more than 90 percent of the industry, in 2010 about 8 MW of CPV was installed and generating electricity. This year should be a fair bit better, and even more encouragingly, there is good reason to believe that the rocketing growth predicted at the back end of the last decade can now take place over the next few years.

Evidence for this surge in CPV deployment can be found in the press releases of the leading system manufacturers. “The amount of announcements has been huge, and the traction for CPV is not like it’s ever been before,” says Nancy Hartsoch, Chairman of the CPV Consortium and VP of Marketing at the Californian-based CPV system maker SolFocus. “Before it was push, push, push – now there is pull.”

The rafter of recent announcements for CPV deployment includes a 30 MW contract for SolFocus to install 30 MW in San Diego County. Work will start in the last quarter of this year and will be completed by the end of 2012. “What we are excited about [with this project] is that this is being funded by an independent. It’s not someone investing in us – it’s someone investing in the project,” enthuses Hartsoch. “It truly says that CPV at scale is financeable by conventional financing. That’s a big step for CPV.”

The European CPV system manufacturer Soitec has an even bigger project in the pipeline. It has signed powerpurchase agreements to supply a total of 305 MW to the San Diego Electricity and Gas company. To deliver, Soitec will build a factory in the US, before installing the CPV Systems between 2013 and 2015. The finance for this venture is in place, because the company recently raised €150 million for its CPV and LED business.

Why now?

The tremendous growth that is expected within the CPVindustry comes at a time of weak, precarious financialrecovery in the US and Europe. So the success in winning these power-purchase agreements has not been easy, and is a triumph for CPV system manufacturers, who are now able to make a far stronger case for deployment of this technology than ever before.

One of the biggest reasons why the CPV industry is in far better shape than it was two or three year’s ago is that it now can now provide evidence that this technology can succeed. “If you can show an investor, a bank and an independent engineer that you are able to a do a megawatt with a customer, a power-purchase agreement and real market pressure – and set up the plant in time, in-cost, and the plant performs in-spec – that makes the change,” explains Concentrix founder Hansjörg Lerchenmüller, who is now the Senior VP of the Customer Group of the Solar Energy Business Unit at Soitec.

Hartsoch agrees with Lerchenmüller, pointing out that some of the leading CPV system makers now have installations that have been running for up to three years: “Our oldest site is a test site that was put in the ground in late 2007.” At this location the company’s first-generation product, plus subsequent generations, has been put ‘on sun’ and data gathered for four years. In addition to testing under normal operating conditions, engineers subject the cells to very high temperatures to confirm that they can withstand the most extreme conditions and still remain within the operating temperature range of the cells.

Another factor behind today’s positive outlook for CPV is an improvement in product quality. For example, several players have certified their products, and some have introduced of new generations of modules that have led to a gain of a few percent in efficiency at the system level. CPV’s competitiveness has also increased through improvements in the efficiency of triple-junction cells, which are getting cheaper and cheaper. Today they only account for a low-double-digit percentage of the system cost – in the past it was up to 20 percent. According to Hartsoch, the lower costs stem from increased competition: “The market was dominated by one or two guys for a long time. Now there is a lot of pressure.” Cell shipments have also gone up, which leads to lower production costs through economies of scale, and there is the promise of far larger orders.

 

SolFocus announced its SF-1136SX CPV system,which has been refined to meet the growing needs of the CPV market. This 15.6 kW system is claimed to output more power per tracker, and it features a streamlined design utilising multi-panel assemblies to cut installation times and costs

Shifting landscapes

How CPV is installed is also changing. Up until now,quite a few of the CPV projects have been focused ongenerating electricity for very local consumption. Forexample, Sol Focus recently completed a 1 MWinstallation for a pistachio farmer in California’s centralvalley that provides 70 percent of the power for hisprocessing facilities. But from now on, deployments forutilities will account for a higher and higher proportionof CPV system shipments.

These deployments initially targeted the sunny parts of Europe, such as Greece and Spain, where developers could exploit very attractive feed-in tariffs. But those great incentives have now been withdrawn, and the South West US is the new hotbed for CPV. “If we look at 2011 and 2012, probably about half our market will be the US; maybe a quarter Europe, the Middle-East and Africa; and a quarter the rest of the world,” says Hartsoch.

Lerchenmüller offers several reasons why California, in particular, is a very attractive place for deploying CPV systems: a strong demand for summer, daytime electricity due to widespread use of air-conditioning; incredibly sunny sites that are 100 miles or so from areas with a very high population density; and a daytime premium on electricity. Although there are no feedin tariffs in operation, CPV installations are able to apply for investment tax credits (ITC).

Hartsoch believes that CPV deployments in California have the potential to generate electricity at a cost of 11 or 12 cents per kWhr, even without incentives. And Lerchenmüller agrees, saying that he expects Soitec’s near term projects to produce electricity at a cost of 12-15 cents per kWhr, a figure that should fall to 10-12 cents per kWhr without the ITC in three-to-four-years’ time. “That, for me, is really grid parity on a power plant level,” says Lerchenmüller. “Below 10 cents is clearly another hurdle.” The case for CPV is so strong in the sunny climes of California that it has helped SolFocus to win two or three contracts against two of the biggest names in the PV sector: The trailblazer of CdTe-on-glass panels, First Solar; and SunPower, a massive player in the silicon sector with annual sales of a more than a billion dollars. SolFocus won the contracts, says Hartsoch,  because although it had slightly higher capital costs, its system would generate 20 to 30 percent more energy.

She believes that within a few years, CPV systems will be able to start making a stronger impact in ‘rest-of the-world’ areas where there are no incentives. In many of these regions, energy prices are currently unpredictable, and a switch to CPV could quash these fluctuations. In addition, this form of power generation could bring electricity to regions that are currently without this valuable resource. “CPV can be a big asset there when you partner with somebody who has the storage equation,” says Hartsoch.

Opportunities for cell makers

The big contracts being signed by CPV makers will begreatly welcomed by developers and manufacturers oftriple-junction solar cells. Both Hartsoch andLerchenmüller offer some insights into what thesechipmakers must do if they are to be successful, suchas excelling in several areas of device performance.“The cells we get today are reliable and proven in spaceapplications. Cell reliability is not a problem today, and for new entrants in the cell market, reliability is your ticket to entry,” says Hartsoch. Decisions by SolFocus on the selection of cell suppliers are heavily influenced by the cost of the device and its performance. “You can have really low cost and lower efficiency, and that’s OK, or you can have really high efficiency and high cost, and that’s OK. The real issue is how the cost of the cells and the related performance affects the overall cost of energy from the system,” says Hartsoch.

Lerchenmüller hold similar views. However, he believes that efficiency is the key metric, arguing that one of the benefits of higher efficiency at the system level is a reduction in the number of cells and modules needed to deliver a given output power. He also thinks that it doesn’t necessarily follow that a cell must be more expensive, if it is more efficient.

Any company that is trying to win orders with CPV system manufacturers may have to begin with a qualification period, which can take several months. SolFocus and Soitec both have IEC-certified product. So if they are to adopt different cells, they have to build modules with these cells and then send them out to test labs to pass IEC certification. “It can take six months,” says Lerchenmüller. CPV system makers will only go with a start-up if it can scale its manufacturing capacity. “This industry is going to grow rapidly,” says Hartsoch, “so you want to partner with the guys that have a technology that can go to volume manufacturing and increase quickly, and investors that are willing to take the companies there.” She points out that the capital costs for CPV system makers to build a 100 MW factory can be as low as just 12 cents per Watt, and depend on the CPV system design. “ For the cell guys, to increase their capacity, it’s a lot of capital investment.”

If a start-up increases its capacity, it will then be competing with the incumbents that are unlikely to need any in-house investment to cater for the predicted hike in orders over the next few years. “We have at least four cell manufacturers that are readily available to increase product at the right quality and the right cost,” says Lerchenmüller. He explains that these suppliers have a “huge buffer zone” in their manufacturing capacity, which is needed to accommodate the volatility of orders for cells for space applications. It is imperative to hit the deadlines for these aerospace contracts, as the knockon effects of a delay would be disastrous. “Just converting this overcapacity [into production of cells for CPV] will serve the business for the next two to three years.”

Crunch time

The ease at which global triple-junction cell capacity can increase over the next few years will restrict the success of start-ups. “Of those 15-20 start-ups, there will be two or three winners,” says Hartsoch, who believes that those that succeed will be those that can win investment, scale up capacity and partner with CPV firms to secure initial business.

“It’s an exciting part of the business, with no other technology can you be so flexible at adopting an advancement as you can with CPV systems.” Lerchenmüller believes that cell producers can gain the edge over their rivals by adopting the mindset of companies operating in the silicon industry. “You have to provide a better product next year at a lower cost – full stop.” If companies of any size can take on this attitude and execute of this front, they will stand a far better chance of qualifying product with Soitec.

Meanwhile, SolFocus’ bugbear with cell suppliers has been a lack of commercial processes such as outgoing test, but these chipmakers are making progress in this direction. “Historically, because they were cells used in space, 100 percent incoming inspection was standard,” explains Hartsoch. One way that some cell start-ups are trying to stand out from the crowd is by developing triple-junction devices with novel architectures that offer spectral tuning.

Lerchenmüller believes that this is an important feature, but does not see a big advantage at site-specific tuning at present, because there is a high degree of similarity in the spectral profile of sunlight at many of the sites where CPV will be deployed over the next few years: “A typical site for large installations is California, in areas that are elevated, but not too highly elevate.” He adds that most sites with a high value of direct normal irradiance have a similar spectrum, and says that a choice of cells is only needed once the market has grown substantially to several GWs or more. “We believe that the CPV market, by 2015, can be as high as 1.8 GW.” If he’s right, the ramp in CPV deployments during the next few years is going to be breathtakingly fast.

 

CPV is providing power to the City Coachella Water Treatment Facility in Coachella,CA. This allows the city to purchase less power, particularly during the high-value post noon hours, including into the early evening peak period

© 2011 Angel Business Communications. Permission required.

 

Soitec and SolFocus: Adopting different approaches

The CPV system manufacturers Soitec and SolFocus have adopted different approaches to focusing sunlight onto triple-junction cells: the former uses a single lens, while the latter employs reflecting mirrors.

Soitec prides itself on the simplicity and robustness of its module. Its top part consists of a glass sheet, which underneath it has a 70 μm-thick silicone film that is embossed and includes the Fresnel lens. The bottom section also features a sheet of glass, which is the foundation for mounting metal heat distributors and the III-V cell.

Engineers at the European outfit have extensively tested silicone for its stability under UV radiation. “We started with just normal UV bulbs, and we didn’t see anything,” says Hansjörg Lerchenmüller, Senior VP of the Customer Group of the Solar Energy Business Unit at Soitec. So they added more lamps, but were still unable to detect any changes to the silicone. “So we then went down the brute force method and found a guy with a UV laser, and saw the first signs of degradation at UV doses equivalent to something like 3000 years.”

The modules that Soitec makes operate at concentrations of 500 suns. This factor could be increased to 800 or even 1000, but higher concentration requires the addition of a secondary focusing element.

According to Lerchenmüller, it is not clear whether savings resulting from using less semiconductor material outweigh the combination of the additional cost of a secondary optical element; reductions in overall efficiency that stem from additional reflections; increased risk to reliability; and lower manufacturing yields that result from a more complex production process. Modules made by Soitec use just one Fresnel lens made from silicone to focus the sun’s rays onto triple-junction cells.



With SolFocus’ reflector-based systems, which currently operate at 650 suns, the case for going to higher concentrations is much stronger. Regardless of the focusing technology, increases in concentration pay the penalty of a reduction in the acceptance angle of the incoming light.

However, according to SolFocus, the acceptance angle for its reflective system – which is over 1° for today’s modules – is two to four times higher than that for most Fresnel-based equivalents.

The synergy of Soitec and Concentrix

The French firm Soitec is renowned for its Smart Cut technology thatenables the separation of incredibly thin, single-crystal layers from asubstrate, which can be used over and over again. This technologypromises to deliver major benefits to triple-junction solar cell production,which is the reason behind Soitec’s acquisition of Concentrix in late2009.

One advantage of the Smart Cut approach is that it increases the freedom of choice for the materials in the cell, because these layers do not have to be lattice-matched to one another. The upshot is that compositions can be optimised for spectral absorption and current output, enabling a higher efficiency for the device. It is also possible to use materials with greater purity, which have been formed in ingots rather than epi-chambers.

Soitec is developing its novel technology for triple-junction cells in collaboration with: CEA-Leti, Grenoble, which is focusing on Smart Cut technology; and the Fraunhofer Institute for Solar Energy Systems in Freiburg, Germany, which is concentrating on epitaxy and cell manufacturing. Cell production is scheduled to begin in 2014 or 2015. The cells will only be used internally, and Soitec is yet to decide whether these devices will be produced in-house or with an external partner.