A thin film frontier

The thin film PV market in 2010 was worth some 3.4 gigawatts (GW). Competing in a wider PV market where mono- and poly-crystalline modules still dominate, some expect thin film manufacturers to compete for a market share of 18 to 20 percent in the near future, increasing to 20 to 25 by 2015. Essentially thin film technologies, such as CIGS (copper indium gallium (di)selenide) and CdTe (cadmium telluride), are looking like they are emerging from niche market status and various companies have invested in the development of their technology of choice and production facilities. Yet among the thin film technologies, manufacturing processes and strategies, it’s not clear where the major advantages will come from and which producers will rise to the top. Japan’s Solar Frontier is one of the contenders, and, with a new 900 megawatt (MW) facility, rising efficiencies and a major campaign to raise awareness of its modules and technology, the sleeping Japanese giant may be awakening to truly make its mark.
“I really love this guy, see the gentle touch he has when applying these stickers,” says Solar Frontier’s Brooks Herring, the Vice President overseeing the Communications and Operations Organization at the new Solar Frontier Kunitomi plant. When fully ramped, some time in 2012, it will have an annual production capacity of 900 megawatts (MW). “Here he goes again, see that?” Herring stands in front of one of the proprietary robots on the production line applying the branding stickers and module wattage barcodes to Solar Frontier’s CIGS thin film modules. With a genuinely gentle touch, it caresses the labels into place, and with a second flourish, runs its applicator down across the sticker to hold it securely in place.
While stopping short of giving actually giving the robot a name, it almost sounds like Herring is referring to one of his children. And in some ways the Kunitomi plant, located on a prominent hilltop outside of the small southern city of Miyazaki, represents the kind of investment a parent makes in a child. Solar Frontier has not only invested heavily in its CIGS thin film technology but also the proprietary plant in place at the Kunitomi plant. As CEO Shigeaki Kameda describes it, “there were not manuals for our machines, we wrote the manuals ourselves.” It’s worth noting that Solar Frontier’s modules utilize CIGS thin film technology. However, over the 18 years the company has worked with it, it has used the term CIS despite the panels containing gallium and sulfur, and it’s stuck.
In late July, Solar Frontier announced that all lines at the Kunitomi plant were now in production. The schedule for the ramp up is such that, by the end of the year, production should be about 500 MW, with the full 900 MW scale of production being reached some time in 2012. The company invested around one billion U.S. dollars (USD) into the plant. Whether the investment truly pays off is dependent on a number of factors, a decisive one being Solar Frontier’s ability to communicate to integrators and the market at large the unique points of difference the company has. These differences can occur on a technological level – in terms thin film generally, as a developer of CIGS technology and in its proprietary manufacturing processes – and also on a business strategy level.

Efficiency levels

Often perceived as low kilowatt-peak (kWp) efficiency and still dwarfed in terms of production by its older and vastly bigger PV brothers in the mono and polysilicon crystalline manufacturers, thin film has some way to go in establishing itself as a serious player in the PV industry worldwide. But as silicon becomes increasingly scarce, with some predicting shortages to kick-in around 2015 and development in mature silicon-based technologies potentially slowing, thin film has advantages in terms of production costs and energy payback periods; temperature coefficient, which is particularly decisive in the sun-belt regions; diversity of application for the flexible substrate players; and genuine “bang-for-your-buck” when it comes to kilowatt-hour (kWh) output of thin film modules, otherwise known as energy yield.
EuPD Research has recently produced its PV Thin Film Guidebook for 2011 and one of EuPD’s Consultants Veit Robert Otto, believes kWh output is a factor many installers will sit up and take notice of. “I would recommend thin film manufacturers to focus more on their strengths, that is on kWh and not only on kWp and efficiency. Today, many end customers still stick to the kWp as a decisive performance criterion. This way of thinking needs to be overcome, since a high kWp does not necessarily lead to a high energy output of the PV system. Thin film players should […] demonstrate their competitiveness under certain conditions – for instance in the sunbelt regions – even with a lower kWp.” The Solar Frontier management team is certainly well drilled when speaking publicly about the performance of their modules in this regard. With discipline that demonstrates not only extensive media training but also a sense of Japanese orderliness, they repeatedly make this argument when conducting tours of the shiny new Kunitomi plant, and also in conversation. James Plastow is Solar Frontier’s Global Project Manager and he sets out the argument. “Remember that efficiency means in standard test conditions. And for 1000 watt-per-square-meter, that’s at 23°C. Now those conditions are almost impossible to occur in real life because […] the module is going to be much higher temperature than 23°C. So […] for the real world what matters is when I put this module in the sunshine, how many kilowatts will I produce? And that’s not directly related to the efficiency because the yield is also very important. And [Solar Frontier’s CIGS thin film modules] have significantly higher yield than a crystalline module.”

CIGS or CdTe

One of the keys to these purported yield results is the energy coefficient; that a module continues to effectively produce electricity when hot. In this respect, EuPD Research’s Otto explains that CdTe thin film modules also perform very well. “CdTe have very high temperature coefficient, which means suitable also for large scale applications in the desert.” First Solar, far and away the biggest player in the thin film market, claims to produce unites with a very good temperature coefficient. The EuPD Research table of page 80 shows just how stark the contrast is between First Solar and all other comers in the thin film field when it comes to market share. Boasting sales of 533 million USD in its quarter two results, with a profit of 61 million USD, First Solar has recently expanded production facilities in Germany and Malaysia. It’s also worth noting that crystalline thin film players, amorphous silicon (a-Si and a-Si, tandem/triple) still capture a major slice of the thin film market, in total over 40 percent in 2010.
The Phoenix-based CdTe player has 2014 goals for balance-of-systems costs to be reduced to $0.91 to $0.98 per watt, with planned module efficiencies of 13.5 percent to 14.5 percent. First Solar has also reported total systems costs falling by 30 percent with goals to reduce that by another 19 percent. In high irradiant zones, where the CdTe temperature coefficient kicks in – such as U.S. Southwest, India, China, Australia, the Middle East, Spain and Italy – this gives First Solar very competitive kWh installed costs. Solar Frontier is less willing to engage in debate over balance-of-systems cost per watt of module production, other than saying it is “very competitive”.

Green in the “green space”

When first formulating its report into the thin film business, EuPD Research asked manufacturers and those in the industry what exactly they would like the report to focus on. One of the resounding responses was carbon footprint. Thin-film manufacturers, in part because the active electricity generating layer is literally very thin, requires temperatures in the region of 500°C in deposition. As there are also fewer stages to the manufacturing process, in comparison to crystalline module production, thin film producers believe less energy is required. The carbon footprint during manufacture is therefore considered lesser. Brooks sets out the specific figures, “Solar Frontier’s modules have an energy payback time of about 0.8 to 0.9 years as determined by third parties. Crystalline modules have a payback time of a year-and-a-half or more.” EuPD Research believes that, while energy payback shouldn’t be a decisive factor in specific thin film manufacturers prevailing over others, or even thin film generally taking market share from crystalline producers, it can be a factor to installers of all sizes if many other factors between manufacturers seem equal. Otto says, “[Energy payback] could also be an additional input for the marketing strategies and this is being done by some players.” However, being “green in the green space,” a term Brooks likes to use to describe companies adopting environmentally friendly practices while working in “green industries,” doesn’t stop with energy usage alone. End-of-life factors, the ease with which modules can be recycled after they have stopped being useful, also comes into play. Solar Frontier is very proud that its modules are cadmium and lead free. It uses a zinc oxide process in manufacture, circumventing the need for cadmium. In Japan, the use of cadmium is forbidden and Brooks believes this makes the recycling of Solar Frontier’s cells easier and greener. “Our modules […] just have no special recycling requirements. So typically you can pull the frames off and take the glass apart and acid wash the thin film from it. Then crush the glass and reuse it all over again.” He continues that therefore less energy is required in the recycling process also.
By contrast, vastly bigger thin film producer First Solar uses a CdTe based technology and therefore rely on cadmium for its modules. A rather unsavory and unofficial publicity battle has raged for some years over the subject of cadmium usage between an anti-cadmium lobbying group – the Non-Toxic Solar Alliance, who some believed are linked to crystalline manufacturers – and First Solar. In July, a German public relations regulatory body stepped in to give the anti-cadmium lobby a public dressing down due to a lack of transparency, however it is true to say that cadmium’s less-than-ideal environmental credentials still effect the perception of CdTe thin film technology to some degree and in some markets.
To further emphasize that its panels are cadmium and lead free, Solar Frontier voluntarily submitted them for testing by TÜV Rheinland for official “Restriction of Hazardous Substances” (RoHS) compliance. It requires that cadmium concentration is limited to 0.01 percent and lead 0.1 percent. In August Solar Frontier received such compliance verification from TÜV. However, it is somewhat of a moot point in Europe as photovoltaic modules are exempt from RoHS compliance under European Union legislation and in a number of other parts of the world Despite the publicity battle, EuPD Research does not believe that cadmium, or the lack thereof, will heavily contribute to any successes of CIGS over CdTe. While taking it into account in its market predictions, the PV researchers considered the effect not to be a major one. “There remains enough market volume in the States for example, where this issue isn’t tackled at all,” says Otto, “and in some parts of northern Africa and in the Middle East there are very good opportunities for CdTe technologies.”

Room for various technologies

EuPD Research also concludes that while references to a technology race in the thin film sector may attract attention, there is room in the PV market for various technologies to take their share. Solar Frontier’s Kameda isn’t so convinced and draws on experience from other technology industries when assessing how serious the competition between thin film technologies really is. “For technologies such as laptops and mobile phones and other electronics [the market] becomes standardized into one type of product. I’m concerned about that and am looking for partners to see how Solar Frontier can expand CIGS.” Otto explains the analyst’s differing outlook, at least in the near term, “There’s space enough and room enough for each technology [so] rather it’s a manufacturers’ race that we have going on.” So when assessing manufacturers, much of it comes down to points of difference that manufacturers can develop and enunciate, both to installers and various sectors of PV market. However, a decisive factor, particularly for larger rooftop or ground-mounted installations, is always cost.

Keeping costs low

While Solar Frontier are evasive when questioned as the cost-per-watt of their production facilities, thin film rival First Solar are clear in its aims of reducing balance-of-system costs to the mid -0.90 USD range by 2014. The key to cost reductions, EuPD Research believes, is scaling of operations. In its mid-term projections, a kind of scaling race may occur between the larger thin film manufacturers. Solar Frontier chief Kaneda agrees with this observation, “Solar Frontier is just one, there are many companies that are announcing very large capacities and we realize than even one GW is not enough capacity.” He continues though that a measured time frame, established with the first two, much smaller Solar Frontier facilities would be followed. “So we’ve consistently, over the last five years, increased our factories over a span of about 18 months each time. We will confirm that the factory is working before we move to the next one. And we will continue in that way.” Wages are also a cost factor and Solar Frontier admit that wages in Japan are higher than in other parts of Asia, where thin film competitors operate facilities. Q-Cells S.E., which has seen its profits turn into losses of 318.3 million euros (449.6 USD) in the first six months of 2011, announced in August a shift in strategy which will see its manufacturing in high-wage Germany slashed by 50 percent. The company plans to focus on its lower wage Malaysian factory. Solar Frontier however believes the high level of automation, in its new plant, avoids this problem. Says Brooks; “We have about 900 MW of capability but we only have 700 to 800 employees so when you look at polycrystalline manufacturing facilities, the current set-up for those facilities has about ten-times the number of employees for the same production capacity that we have here.” PrimeStar Solar is another thin film manufacturer taking part in the “race to scale” its operations. Now with the full backing of GE, which announced it had completed its acquisition of PrimeStar in April of this year, PrimeStar plans to build the biggest thin film plant in the U.S., with a capacity of 400 MW. PrimeStar uses CdTe thin film technology developed by the U.S. National Renewable Energy Laboratory and claims to be making industry leading gains to its efficiency levels. Significantly, the backing of GE is most significant and provides mid-term financial security and the ability to pursue production expansion plans in a tight short-term market.

Financials

As a wholly owned subsidiary of Showa Shell Sekiyu K.K., significant financial resources underpin Solar Frontier’s operations. CEO Kaneda hints at this when indicating that the company is more interested in the mid-to-long term. While thin film industry analyst Otto indicates that some manufacturers may not make it through a tight market in the short term, he believes this is not the case with Solar Frontier. “They have a very strong financial background, enough financial flexibility to survive the turbulences ahead in the industry.” Solar Frontier previously went by its parent’s name, “Shell Solar Japan,” and was only rebranded in 2010.

Installation costs

Costs for installers of all sizes do not simply stop with the module themselves. Installation and equipment costs can also contribute and part of the Solar Frontier strategy is to bring these down where possible. The new Kunitomi plant sports a rooftop installation of two MW (pictured) which has been mounted directly onto the roof, circumventing the need for additional mountings. While this may not optimize output of the modules, Solar Frontier’s Brooks believes that the cost reduction outweighs this. “We try to optimize the racking system both in terms of reducing the costs of that racking system but also reducing the costs of the labor.”

Outlook

Looking long-term, predictions are difficult to make. Otto, while having faith in the financials of Solar Frontier’s backers, is unclear on whether its strategy will be successful. It’s clear that the Kunitomi plant is a significant wager on this strategy, CIGS technology and on thin film generally. Yet a lot depends on Solar Frontier’s ability to spread its energy yield message and build its reputation. Otto suggests that thin film, if costs can be reduced and, “problems with the flexible players can be solved,” could capture 25 percent or even 50 percent of the global PV market. How much of this Solar Frontier will claim for its own, depends on successful communication and whether the Kunitomi gamble comes off.