Two pillars maintain thin film presence

While 2012 was a year to forget for many thin film players, there are signs that downstream integration is paying off for the remaining large manufacturers. It certainly remains clear that Arizona-based First Solar and Japan’s Solar Frontier remain the dominant players in thin film.
The global PV market is remaining difficult for thin film producers to sell modules. The dynamic whereby thin film producers have lost their cost advantage over crystalline silicon (c-Si) rivals has been observed for some time. Even former cost leader First Solar now finds its costs equal to or even trailing some leading c-Si manufacturers, coming in at around $0.60/W. In the face of this challenge, First Solar has clearly moved away from its ambitions to be the largest volume module producer and towards being a project developer.
Solar Frontier is doing much the same and has a project development business that is set to drag the company into profit in 2013. The company had the bad timing of completing its big new manufacturing facility in Miyazaki, in Japan’s south, just as polysilicon and module prices markedly decreased. However, it has hardly been all bad luck for Solar Frontier, as it appears ideally situated to benefit from the booming Japanese PV market, on the back of generous feed-in tariffs and significant demand for new electricity generating capacity in the country.

Power plant provider

First Solar is seeing the benefits of its shift away from module supply and into project development and its performance and profits – that’s right, profits – have been greeted warmly by stock watchers in the U.S., with the share price reaching over $0.55 on June 10. While First Solar’s Q2 2013 financial results saw it miss earnings-per-share and revenue guidances, its performance overall in 2013 has been strong, allowing the firm to issue stock and increase its cash reserves. The additional cash – First Solar finished Q2 2013 with $1.29 billion – will allow the company to continue to develop its project business and continue to invest in its proprietary and new technology.
Despite shedding some manufacturing capacity and shelving plans for new facilities, by some margin First Solar remains the biggest thin film producer. After shuttering its German manufacturing operations, in Frankfurt Oder, and scaling back its operations in Malaysia in 2012, the company’s global production capacity has remained steady and the continuing development of its downstream operations means that it is benefiting from high utilization rates. IHS Solar, in its newly released thin film ranking, puts First Solar’s manufacturing capacity at 2 GW, and its full year production for 2012 at 1.875 GW.
“The company is doing extremely well,” says Principal Analyst with IHS Solar Stefan de Haan. “They are tailoring their capacity and their output to their own project pipeline, they had a brilliant Q1 2013 and they are sold out for the rest of the year.” De Haan notes that while First Solar remains the cost leader amongst the thin film producers, it may not remain locked into its cadmium telluride (CdTe) thin film technology indefinitely.
One of the worst kept secrets in thin film was that the company had maintained an interest in CIGS technology, on an R&D level, but it backed away from this when the company was making its change of tack, in terms of business strategy, in 2012. In April, First Solar announced the acquisition of TetraSun’s potentially high efficiency, low cost c-Si technology and plans to build manufacturing capabilities by mid-2014.
The TetraSun acquisition shows that First Solar is willing to investigate opportunities outside of CdTe thin film and that the company backs its ability to take into production crystalline technology, using a copper metallization process. If successful in ramping the technology, the TetraSun products would allow First Solar to deliver to Japan and the company has indicated its initial capacity in the new technology will supply the Japanese residential market. Currently First Solar is locked out of supplying Japan as products containing cadmium are forbidden in the country. Whether this is an indication that First Solar will move away from CdTe in the near future, de Haan councils caution: “It really depends on how long First Solar plans to continue with CdTe, whether it will remain competitive through to 2017 or to 2025.” Certainly First Solar continues to make progress with its CdTe technology. In February this year the company announced that it had achieved a solar cell conversion efficiency of 18.7%. The company’s strong utilization rates would also allow global cost-per-watt to continue to be reduced. Last month First Solar announced that it acquired GE’s CdTe intellectual property portfolio, which the latter had previously picked up when it acquired the Colorado-based startup PrimeStar in 2011. Since then GE has continued to develop the technology and beat First Solar’s record, achieving an efficiency of 19.6% for a research cell.
GE had previously developed plans to establish a 400 MW production facility for its technology in Colorado and it came in 10th position in the latest thin film ranking. That plan now seems dead and buried, with GE picking up 1.75 million shares in First Solar, as a part of the technology transfer deal. GE is now one of First Solar’s top ten investors, going from potential rival to partner overnight.
First Solar will produce modules for GE under the deal, said the thin film giant’s CEO James Hughes. Hughes noted GE’s role as a major renewable energy developer in the company’s Q2 earnings call: “GE, which has 34 GW of renewable energy installed globally, will also enhance its presence in solar by buying GE branded modules from First Solar at agreed upon pricing for future global GE deployments, in addition to its investment in inverters, controls, balance of plant and ownership of utility-scale systems.”

Expansion plans

It’s no surprise that few firms are locking in expansion plans at present. Although First Solar’s focus on emerging markets, coupled with a possibility that it has not installed the tooling from Frankfurt Oder in its existing fabs elsewhere, means that new facilities in Middle or South America or even China may be on the cards.
“First Solar does have a good presence in Chile, in the Middle East and in China. Maybe First Solar could reuse some of the tools from Frankfurt Oder in these markets,” says de Haan. Although he does warn that at this stage such talk is largely speculative.
What is clear is that First Solar is moving forward in China and taking an almost evangelical role in promoting thin film technology there. At the SNEC trade show in Shanghai earlier this year, First Solar took out a large booth and hosted a series of well attended presentations about its technology, something the firm rarely does in the current PV market.
Attending the SNEC trade show as a part of the relatively large First Solar team was Bruce Yung, Managing Director and Vice President of First Solar in China. Yang told pv magazine that one of the keys for First Solar to be successful in China was to ensure that the strategy of the company there aligns with the strategic goals of the Chinese government. It’s certainly not a stretch to envisage that local production facilities would align with such goals and curry favor with the government and perhaps open doors in the developing of projects.
First Solar has chosen to expand, through acquisition, in South and Central America and its acquisition of Chilean PV developer Solar Chile in January was received positively by some market analysts. Solar Chile had a 1.5 GW project pipeline in northern Chile at the time of acquisition, in January this year. In Mexico, First Solar has also acquired a 1.5 GW project pipeline, through its takeover of Element Power.
Goetz Fischbeck, Managing Director of Renewable Analytics in San Francisco, said that while it would make some sense to develop local manufacturing close to projects given high transportation costs, it would depend heavily on a local project pipeline. “Overall I expect that First Solar will reduce total capacity volume,” said Fischbeck, “which means utilizing less of their equipment.” First Solar has continued to build upon its leading market position in the expansion of utility-scale project development in Australia. While the evolution of this market segment has at times been problematic Down Under, it recently signed off on a deal to develop what will become the largest project in the country. First Solar will develop the project with Australia’s AGL, and will presumably supply the 155 MW of modules from its 1.4 GW manufacturing facility in Kulim, Malaysia.

Riding the wave

Across the Pacific from First Solar’s headquarters in the U.S., Solar Frontier in Japan has been able to ramp production at its 900 MW Kunitomi fab, which opened in February 2011, and has even updated and recommenced production at its older 60 MW Miyazaki No.2 plant, which had temporarily suspended production since the end of 2012. While exact costs per watt for the CIS/CIGS producer are difficult to ascertain, the fact that it was able to ramp production at its relatively new fab has helped it reduce costs.
IHS Solar’s de Haan understands that Solar Frontier may be currently producing modules at something like $0.85/W, which would still make competing in the current marketplace difficult. Its cost reduction track record lately has been good, with IHS models predicting production costs well above $1/W only two years ago.
Solar Frontier also continues to push the case that its modules still deliver the highest kWh/kWp, and refers to independent data from trial installations in Japan, which have been established by the energy arm of telecommunications giant Softbank and Toyko Electric Power. “More and more people are gaining a third party assessment that, indeed, confirms what we have been quoting is true,” says Atsuhiko Hirano, Solar Frontier’s Director of Global Sales and Marketing. “As a result we are seeing quite a high proportion of the share in the new market we have created.” Hirano reports that Solar Frontier is currently producing modules with an efficiency of 14.7% from its scale production, which the Solar Frontier executive describes as running at “full blast.” Where Solar Frontier has indisputably been getting good returns is from its project development business, where it works with a range of EPCs and also diversified industrial players in Japan, looking to develop installations, given the high FIT returns.
Hirano also reports continuing support for PV from Japanese policy makers and says that while the FIT regime was slow to be put in place, he has every confidence that it will be maintained at least for the three year period initially set out for the program to run. “Given that there is an imminent need in the Japanese government to renew its energy policy, to diversify some of the dependency they had on nuclear and also on hydrocarbon fuelled power, definitely one way is to expand the use of renewable energies and especially solar.” Solar Frontier enjoys the support of a parent company with deep pockets, which assisted it to ride out a tricky 2011 and 2012 when the firm was not profitable. It is a wholly-owned subsidiary of Showa Shell Sekiyu and the parent company would presumably be pleased that Solar Frontier is on track to register a profit in 2013, on the back of its strong project business.
“Solar Frontier’s capacity in 2013 is 960 MW, and for the first quarter at least our information is that it was running at 95% utilization,” says de Haan. The IHS analyst says that the Japanese PV market is expected to deliver peak growth in 2013, with “way more” than 5 GW expected to be installed nationally. “Solar Frontier has a large project pipeline. It is developing not only ground-mounted projects, but also large commercial rooftops. The statement that they are running at very high utilization is reasonable.”

New Japan fab?

Given this high utilization rate, Solar Frontier may be the only other thin film producer, other than First Solar, looking to add some capacity in the near future. The company has qualified for a subsidy program which aims to revitalize the T?hoku region, which was badly affected by the Fukushima Daiichi nuclear plant disaster in March, 2011. Funding under the program could be employed by Solar Frontier to build a new fab in the region. Solar Frontier’s Hirano confirms that while the subsidy program is still valid, no corporate decision has been made. “We are studying the case very carefully.” Stefan de Haan says that such a move would not be entirely unexpected. “If Solar Frontier wants to survive at all then it has to expand,” says de Haan. “It fits into the picture because the company does believe in its technology.” However he is reluctant to draw a wider case for thin film or CIGS on the back of any capacity expansion plans. “Whether it will ultimately help CIGS to survive against c-Si is still an open question.” Renewable Analytics’ Morbitzer says that while the outlook remains very strong for the Japanese market, he is skeptical that the timing is right for Solar Frontier to make such a move. “Adding capacity in the next 12 months might help, but also runs the risk of completing the expansion at a point in time when government support in Japan decreases.” The booming Japanese market is consuming much of Solar Frontier’s attention at present, although it is continuing to evaluate opportunities in economic and emerging markets. Wolfgang Lange, the firm’s European head, says that if a tension exists between serving the Japanese market and looking elsewhere, it is a “healthy tension.” He adds, “It is very important that we discuss and come up with optimum solutions as to how we can use the Japanese market but also how we can make sure that we have a good connection with the international business.” Lange reports that Solar Frontier has retained its business unit in Europe and that it is largely concerned with evaluating future strategies and opportunities for the company.

Challenges remain

The apparent challenge in surviving in the current PV market for thin film players may have been sidestepped by the two pillars of thin film technology, however for amorphous silicon (a-Si) producers, the future still looks rather bleak. Leading the a-Si pack in previous years was Japanese giant Sharp. However facing falling ASPs for modules globally and fierce cost reduction from c-Si producers, Sharp vastly scaled back its production from 220 MW in 2012 to less than 30 MW, meaning that it has dropped out of the IHS Solar top ten ranking altogether.
Italy’s 3Sun appears to be one of the stronger remaining a-Si producers, and is producing modules for Sharp, among others. De Haan says that the temperature coefficient in its modules helps them compete with c-Si product, and that the company has been making strong progress in terms of efficiency. “3Sun is still improving its efficiency. Last year it was around 10%, this year they are around 12%,” says de Haan. This progress for amorphous silicon technologies is quite remarkable.
IHS figures have 3Sun operating at full capacity, which amounts to 160 MW. Given the favorable temperature coefficient, 3Sun targets markets where heat is a factor, and reports show the firm is selling strongly in Italy, India, Australia and it has activities in Greece and Turkey also.
Taiwan’s NexPower has 130 MW of a-Si capacity and is therefore next on the IHS ranking, however Stefan de Haan reports that the firm is struggling. In 2012, NexPower produced 80 MW of modules, but it halved its workforce. Renewable Analytics Morbitzer has his doubts as to the ongoing competitiveness of a-Si producers such as 3Sun and NexPower. “Pure a-Si manufacturers will find it hard to continue,” said Morbitzer, “unless they find a customer base in markets that are as of yet very price conscious, such as Africa.” Trony Solar is the third a-Si manufacturer to appear in the top ten ranking, however the Chinese producer is utilizing only a fraction of its 265 MW nameplate capacity. Ominously the firm stopped financial reporting more than one year ago. It is supplying a 15 MW project under the Chinese Government’s Golden Sun scheme, but demonstration projects look to be insufficient to save the firm.

Building-integrated PV

Like a number of other smaller thin film producers, NexPower is looking toward BIPV applications as providing a market for its modules. Equipment supplier Manz has also been doing the same. Manz acquired Würth Solar’s operations in Germany’s south and has been attempting to sell its CIGS Fab turnkey line for some years, although it has no sales as yet. pv magazine understands that Manz is focusing on areas such as the MENA region and South Africa as promising sites for one of its fabs. Manz has also produced specialty modules to demonstrate the flexibility, in terms of form, color and design, to architects. “Architects are artists in the end, they want to design their building,” says Bernd Sprecher, the Managing Director of Manz’s CIGS technology. “That’s also an advantage for our technology, we can design modules in each color, we can make transparent modules and in almost every shape the architect wants.” CIGS manufacturer Avancis has also been pushing BIPV applications for its modules and has come in at fifth position in the thin film ranking. It has 200 MW of production capacityin Germany, and IHS Solar understands it produced 80 MW of modules in 2012. “Avancis’ future depends a lot on Saint Gobain,” said IHS’ de Haan. “We know that they scaled down their utilization in 2012. They were at around 50%.” He believes a focus on applications where its CIGS modules are not in competition with c-Si players, such as in BIPV or lightweight applications, may hold promise.

The mystery of Hanergy

Flexible modules is also something Chinese dam builder and renewable energy developer Hanergy is looking at, although to say that a clear strategy is emerging from Hanergy’s acquisitions is misleading. Hanergy has acquired former Q Cells subsidiary Solibro, roll-to-roll CIGS producer MiaSolé and flex producer Global Solar all within a relatively short space of time. MiaSolé and Solibro were two CIGS producers who had achieved significant scale, yet since acquisition Solibro has dropped out of the top 10 thin film ranking and de Haan suspects that the vast majority of its 135 MW of manufacturing capacity is lying idle. MiaSolé has fared slightly better, supplying projects in India, however its output of 49 MW in 2012, from its Silicon Valley fab, still only represents a 50% utilization rate.
While the acquisitions have all come at a relatively low cost, Hanergy’s intentions and strategy remains opaque at best. “For Hanergy, you get the impression that they need to spend money and they are acquiring potentially beneficial technologies, but it’s a very risky game,” says de Haan. Renewable Analytics Morbitzer believes Hanergy’s strategy makes sense only if it believes it can develop the technology it has acquired to scale: “Currently it is too early to say, we are still waiting to see the higher efficiency, lower cost new products.” MiaSolé trumpeted its achievement of a flex module with an efficiency of more than 14% in 2012. Its process of cutting cells from its stainless steel roll after deposition and sorting them into higher efficiency batches may be able to maximize efficiency of modules, especially given that the lamination process into a flexible module remains a high cost manufacturing process. “I don’t know if there is a chance to get lamination costs reasonably down,” says de Haan. “It would make sense to try because then you really would have a new product that would open up new markets.”

CIGS hopes

To say that all middle-sized CIGS players are aiming only for niche applications would be untrue. Stion, which is based in California and has a 100 MW facility in Hattiesburg, Mississippi, is claiming that it is making good progress with ramping up its operations and developing sales channels. Stion is leveraging semiconductor expertise to optimize its operations, and Global Operations Vice President Marty Finkbeiner says the startup is looking to drive out costs. “We have to manically pay attention to every detail from a cost per watt perspective, because at the end of the day, that’s what’s going to make or break any business.” De Haan says whether the company can stay the course depends on the continued backing of its investors. Stion was able to access a low interest loan of $75 million to develop its Mississippi fab. The company has already attracted $130 million in equity investment from a range of partners, including equipment supplier Avaco. A consortium of Korean private equity investors are also backing the firm, and part of the deal would see Stion establishing a subsidiary manufacturing operation in South Korea.
The Californian startup has also licensed its technology to Taiwanese semiconductor giant TSMC. TSMC remains at small-scale production, which company representatives claim is an advantage at present, while it waits for market conditions to improve. In the meantime TSMC Solar has been registering some impressive efficiency gains. In June, TSMC announced a champion module efficiency of 15.7%, which was certified by TÜV SÜD. Efficiencies like these, along with modules such as TSMC’s 140 W – 155 W C1, are certainly closing the efficiency gap with crystalline modules. Although it is worth noting that the champion module efficiency was not achieved at scale production. Some CIGS insiders have also told pv magazine that they remain skeptical of the result and believe that the TSMC solar division is under pressure from its semiconductor parent to demonstrate results.
A straight comparison with thin film production against c-Si shows that across CdTe, CIGS and a-Si, thin film is still a comparative minnow, with something like a 12% market share in 2013. This is primarily driven by the two pillars of First Solar and Solar Frontier, who themselves are heavily involved in project development. While some predictions, made as far back as 2008, that thin film would become the predominant technology in PV have certainly proven wrong, some firms are pursuing technologies and approaches that still may hold promise for the future.