Walking a thin line

Markets & Trends, 01 / 2013 | By:  Jonathan Gifford

Thin film 2013: Picking winners in this year’s PV market is difficult, to say the least. In the thin film sector, picking survivors may be near impossible. With brutal price competition continuing to make life hard, pv magazine casts its eye over the thin film field to see what strategies are emerging in this fight for survival.

First Solar’s downstream business managed to help the company turn around losses in 2012, as did dramatic moves to cut costs by shelving plans for new capacity and closing fabs.

On a bright July day, the dark wood boardroom of MiaSolé gives welcome respite from the glare outside. On one side of the room, a number of the company’s modules lean up against the windows looking out to a lush Silicon Valley lawn and garden. Lying along the table is a two meter long flexible module. It’s only a prototype, but as a young and enthusiastic executive explains, it’s a product that MiaSolé thinks its CIGS flexible thin film cells suit well.

Fast-forward four months and MiaSolé is no longer in charge of its future. China’s Hanergy Holding Group has swept in and purchased the company for a mere US$30 million. The sum seems trifling given that since 2004, MiaSolé is estimated to have attracted over US$500 million in VC funding. MiaSolé had also showed great promise, in terms of cost reduction and relatively high efficiencies.

Previously, the future for MiaSolé often seemed bright. It boasted a fab CAPEX of below US$0.50/W, notoriously a tricky figure to bring down with CIGS technology. It had a unique fab design and process, using roll-to-roll deposition for the CIGS semiconductor stack onto a stainless steel substrate.

This deposition process was employed using a horseshoe-shaped equipment layout, which itself had an extremely small footprint. Its front-end deposition could be separated from its back end, where the cells were then assembled into modules. This meant that the front end could deposit the CIGS stack onto foil, which could be assembled into cells and the lightweight product shipped to back end facilities close to market.

In terms of efficiency MiaSolé had plenty to boast about also. It had module efficiencies of 14% on the board and a road map to 18% over five years. Production costs were under US$0.80/W with a sub US$0.50/W road map also over the same period. And the flexible module: In May the company announced that it had achieved a world record 15.5% aperture area efficiency for a flex module around 17 m² in size.

But with only 50 MW of MiaSolé product in the field producing real world data, and fierce price competition across all of PV, the cash burn required to get the planned 150 MW fab ramped was brutal. MiaSolé’s potential was huge, but the technology needed time to establish itself and if there is one thing the current PV market is short on, it is breathing room. Like many players in the current period of consolidation in the PV market, MiaSolé simply ran out of puff.

In recent weeks, it’s not only been MiSolé to go that way. AQT put its CZTS technology, IP and equipment up for sale in August, and Asian buyers have purchased both HeliVolt and Ascent Solar, some would say cheaply. Schüco appointed auction houses to carry out the sale of its assets in December and Global Solar announced the closure of its U.S. manufacturing operations in the same month after winding up its German operations earlier in the year. Electronics giant GE shelved plans to build a CdTe fab for at least 18 months, and laid off around 70 employees at the PrimeStar Solar plant it acquired in 2011, focusing on R&D rather than manufacturing.

A telling example

In many ways the primary problem that faced all of these players was similar. The advantage that thin film manufactures had when polysilicon prices were high is long gone and many of the investments into thin film technologies now are proving to be money wasted (see diagram p.16 showing that polysilicon supply and demand is not about to come back into balance soon). It is often spoken of in cleantech VC circles that every fund sports a failed thin film firm of some sort or another.

Even for manufacturers with major capacity already on line, like First Solar, there simply is no cost advantage to their technology. “First Solar has lost cost leadership,” says Stefan de Haan, the principal analyst with IHS Solar Research, “because c-Si cost has come down so much faster than they had anticipated.” It is true also that with the current oversupply in the module market, production is competing on price essentially with inventory modules, which may be being shifted well below cost.

The challenges First Solar has faced in this market have been fairly well publicized. Last year the global CdTe manufacturer took a sharp change of course, shelving plans for new fabs in France and Vietnam and then earlier this year suddenly announcing the closure of its German operations – where one of the fabs had only been fully operational for a matter of months. While its German FF01 and FF02 fabs, with a combined capacity of 497 MW, were kept open a little longer than planned, in November the company confirmed that 1,050 staff are still seeking new employment, after the lights finally went out at the end of last year. One of the casualties of this required change of tack was former CEO Rob Gillette, who was quickly shifted to one side in October 2011, with replacement James Hughes being installed some eight months later. Hughes has considerable experience in the downstream component of the electricity and energy business and knowledge of emerging PV markets, the company claimed in a press release announcing the appointment.

One problem bugging the U.S.-headquartered manufacturer has been fallout from its low power module (LPM) issue. Modules manufactured between June 2008 and June 2009 seem to be the affected batch and while there has been criticism of First Solar’s handling of the issue, particularly from smaller customers, it did manage to satisfy its major EPCs and integrators – at least from what has been said publicly by these customers.

First Solar has also stayed the course with its plans to develop its downstream business. A pipeline of projects are emerging, some of which are developed in joint ventures (JV), in countries including Saudi Arabia, Australia, Indonesia and Mexico. The development of its U.S. pipeline of projects continues also. IHS Solar Research’s de Haan says that he believes the strategy will pay dividends in the short to middle term and has allowed First Solar to turn big losses around. “Of course at some point, not too far in the future, all these module producers integrating downstream will have to make a decision: Do I want to be in a low margin, large scale commodity business (module production) or in a high margin, high overhead cost, diversified project business?” The analyst adds, “both models will work separately, but the combination in one company will soon become problematic.” First Solar’s stock price has rebounded after hitting a bottom in June of this year.

Rising sun

What is really the only other thin film manufacturer operating at this scale is Japan’s Solar Frontier. It is difficult to truly know how its costs compare to its c-Si rivals, as the company refuses to release its cost per watt. What certainly seems true is that theCIGS manufacturer benefits from being a local in Japan’s booming market.

The Solar Frontier team is keen to emphasize that it is the only manufacturer with a product entirely “made in Japan.” And with many businesses from outside of PV looking to develop “mega solar” projects or rooftop installations, from telecommunications giant Softbank to convenience store chain Lawson, Solar Frontier’s pitch is that it can draw on its long history in the energy business. “For the customer they get to work with someone who has been in the energy business in Japan for more than 100 years,” says Vice President of communications and operations Brooks Herring, “so this is our backyard – we understand all the processes that we need to work with here.” Solar Frontier’s parent company is the Japanese oil company Showa Shell Sekiyu.

In this environment Solar Frontier can offer a “one stop shop” style service, where it can provide assistance to firms looking to generate a healthy return on investment under Japan’s revamped FITs, which came into effect on July 1. “We can help people work through the entire development process, meaning the acquisition of land, layout of the project, securing permits that are required, working through the grid connection process and finally building the facility as well as doing the operations and maintenance,” says Herring. Solar Frontier works with EPC partners to deliver this service.

One of the advantages of being in the position to do this, according to Solar Frontier, is that it is able to introduce CIGS thin film technology, which it refers to as CIS, and espouse its advantages. VP Herring explains, this means explaining the advantages of its thin film technology, as opposed to c-Si performance claims, which are based on module efficiency. “I think when people understand that we start to generate electricity earlier in the day, we finish generating electricity later in the day, that our temperature coefficient is better – they begin to understand a little deeper the characteristics of our panels,” says Herring, “they understand why we generate more kilowatt hours per kilowatt peak.”

Thin film pillars

It is of interest to note that the two major thin film manufacturers have chosen to develop their downstream business model in the face of stiff price competition. First Solar’s strategy is firmly focused on emerging PV markets, while continuing to serve its U.S. business, and Solar Frontier remains primarily focused on Japan and the MENA region, and Saudia Arabia in particular.

The two companies very much form the two pillars of the thin film sector and Stefan de Haan remarks that if either was to topple, thin film’s PV market share would go with it. “In 2012 we see a thin film market share of 13.5%, and that will not grow in the coming years,” he says. “We don’t see it decreasing massively either, but as of today it seems it really is the decision of two players.”

Slow and steady

While Atlas-like First Solar and Solar Frontier are holding up the market share of thin film, another way to visualize their role is that of a leaders’ group of two in a cycling road race, out in front of a dwindling but determined peloton. Looking over the field of surviving thin film manufacturers, still operating under their own steam, the red and black of TSMC Solar catches the eye.

Like Solar Frontier the Asian-based manufacturer employs CIGS technology, in TSMC’s case licensed from Silicon Valley’s Stion. An advantage that TSMC has is that its parent TSMC Semiconductor has deep pockets, stemming from its US$88.5 billion market capitalization and one would hope patience in its solar investment. TSMC is tooled for 100 MW of production but it is not being ramped and, according to management, there are no plans to do so until supply and demand begins to balance out. This, Paula Mints observes, is probably a good thing. Mints is the Founder and Chief Market Research Analyst with Solar PV Market Research in partnership with Strategies Unlimited. “Companies need to plan for surviving the consolidation, so ramping is probably not a good idea.” Stion itself is somewhat of an unknown entity, however it has been assisted by a US$75 million state-government loan to establish its 100 MW fab in Hattiesburg, Mississippi. But as the example of Colorado-based Abound Solar shows (the now defunct thin film company was assisted by US$400 million in government grants), this is not enough in the face of the kind of cash burn that is possible in the current market. The Denver Post observed in the lead up to the U.S. Presidential election that Abound only exists now as a political football, generating far more publicity than it did while an operating business.

Stion still operates an R&D line in its original base of Silicon Valley and when pv magazine toured operations in mid 2012, the team were reporting that the new facility in the Deep South was operating at only a couple of MW per quarter. The fab is sized for 400-500 MW of production. Stion is currently producing modules with average efficiencies in the mid 13% range, with a champion of around 13.4%. It’s also developing a tandem junction technology that has the potential, the companyclaims, to deliver module efficiencies rivaling c-Si manufacturers. The company reports achieving an efficiency of 18.8% on a prototype scale of 20 x 20 cm. In November it picked up a US$2 million U.S. Department of Energy “SunShot” grant to continue to develop its “ultra high-efficiency thin film” technology.

Regardless of potential, as MiaSolé and others have demonstrated, a key to survival in the current market is stable and ongoing access to funds. A little over 12 months ago, Stion raised US$130 million in funding from Korean flat panel display and thin film equipment maker Avaco. Marty Finkbeiner, the EVP of global operations at Stion, said that through this arrangement the question of funding has been answered. “The money and the funding hasn’t really been coming from within the U.S., that’s what Stion has recently found, and I think we’ve got access to the money we need and we’ll continue to get access to the money we need with the performance of what we’re doing here with the product and the production efficiencies.”

More MENA

As demonstrated by Solar Frontier and First Solar, the MENA region is shaping up to be a potential hot spot for thin film development. Saint Gobain’s solar subsidiary Avancis is joining the list of thin film firms increasingly active in the region.

Avancis has been going about the business of expanding its German production from 20 MW to 120 MW in recent times. In October its French parent signed a memorandum of understanding with Saudi authorities to construct a fab in Saudi Arabia. The project is a part of a program to supplement oil fired electricity generation in the country with solar and nuclear. While details surrounding the plan are scarce, the deal indicates that the potential for thin film technologies in the MENA region is being recognized, perhaps assisted by a favorable temperature coefficient exhibited by the modules.

Death pool

Almost exactly one year ago Ed Gunther, of the website Gunther Portfolio, published a CIGS startup death pool, which allowed readers to vote which company in this category was most likely not to make it through the year. While the 210 voters who took part were not entirely accurate in their predictions, the fact that four of the 10 companies listed have not survived as manufacturers, without takeover, demonstrates just how hard it is to be operating with CIGS technology in the current market. For CdTe manufacturers the evidence is that things are certainly no easier either.

The cliché of necessity being the mother of invention does however appear to apply to many of the surviving players within thin film, with a range of revenue raising and investor sourcing techniques delivering varying levels of success. The market for all thin film companies is likely to remain tough this year, but there will be positive developments, not least a relaunch of MiaSolé sometime soon. Market analyst Paula Mints has been closely observing the fierce price competition affecting all market participants over the past few years and has a longer-term view on the situation thin film manufacturers find themselves in. “In the mid-2000s (when the FIT model was expanding) prices increased for c-Si technology and many people said that c-Si was dead. It was not true,” says Mints. “Now that prices for PV technology are too low and competition is brutal, many people say that thin films are dead. It is also not true.”


First Solar’s downstream business managed to help the company turn around losses in 2012, as did dramatic moves to cut costs by shelving plans for new capacity and closing fabs.