WHAT’S COMING FOR SOLAR THIN FILM

Thin Film 2010:Market Outlook to 2015
March 18, 2010 (Greentech Media Research)

THE POINT
The thin film materials that convert light into electricity with more flexibility than traditional solar panels represented one of the biggest New Energy success stories of the 21st century's first decade. The potential was described as “disruptive.” The sector had a 96% compounded annual growth rate from 2002 to 2008.

There are now 170 companies in the sector and more than $2 billion in venture capital has been invested. First Solar, the leading company, was the darling of the stock market from 2006 to 2008. Backed by enormous marketplace attention, First Solar grew its production capacity from 25 megawatts in 2006 to 1.2 gigawatts at the end of 2009.

And then the supply of crystalline silicon caught up with the demand.

Thin Film 2010: Market Outlook to 2015, from Shyam Mehta of Greentech Media (GTM) Research, describes what is expected to happen in the field that has had so much sound and fury and yet has only produced 2 companies with a production capacity greater than 100 megawatts per year.

click to enlarge

Thin film materials are the product of the search for alternatives to traditional crystalline silicon photovoltaic (PV) panels. Made primarily from (1) amorphous silicon (a-si), (2) cadmium telluride (CdTe), or copper indium gallium diselenide (CIGS), thin films are less efficient but also less expensive than PV panels.

They seemed to offer a real and effective alternative in the middle years of the decade when there was a shortage of refined silicon. Competition for the limited supply drove the cost of crystalline silicon up and made standard PV panels more expensive. First Solar's CdTe formula captured a progressively larger portion of the PV market.

But CIGS has failed to achieve commercial production. Refined silicon is once again highly available, thanks to Asia’s manufactories. Credit for promising but unproductive businesses started disappearing in late 2008 and has not returned.

The thin film sector now essentially belongs to First Solar. The GTM Research report recounts the fate of thin film in the last 2 years and predicts it will continue to struggle in the short term. But thin film will recapture growth in the 2011-to-2012 period, Mehta reports, especially with the most practical a-Si and CdTe formulas. And it will continue growing through the middle of the coming decade, by which time the so-far unrealized potential of CIGS thin film may be fulfilled.

click to enlarge

THE DETAILS
Thin film grew from a production capacity/market share of 17 megawatts in 2002 to nearly 1,000 megawatts (a gigawatt) in 2008.

The overall thin film production capacity grew from 349 megawatts at the end of 2006 to 4.4 gigawatts at the end of 2009.

The 2009 thin film throughput was 1,111 megawatts. Efficiency averaged 11%. Cost was as low as 83 cents per watt.

click to enlarge

First Solar is the only pure solar company to be listed on the Standard & Poor index. It is now the biggest PV module producer in the world.

Sanyo produced what may have been the first thin film, using amorphous silicon (a-si), in 1974. Other Japanese companies were making thin films by the early 1990s.

Interest was limited until silicon shortages in the early 2000s generated renewed interest. 46 companies and $1.8 billion in venture capital came in between 2004 and 2008. The 3% 2001 market share became a 12% market share in 2007. Costs fell toward grid parity.

click to enlarge

But by 2010, only First Solar’s CdTe and United Solar’s triple junction a-Si formulas had broken the $1 per watt cost level.

The variables that would affect the viability of thin film going forward:
(1) Operational characteristics (module degradation rate, temperature coefficient, spectal response);
(2) System cost variables (module price, balance-of-system costs);
(3) Price and availability of substitutes (crystallized silicon PV);
(4) Market characteristics (incentives, types of markets, environment);
(5) Potential supply (available manufacturing capacity, throughput capacity, yields); and
(6) Qualitative factors (suppliers, technologies, risks, financing).

click to enlarge

Examples of how the variables affect thin film:
(1) Thin film is more attractive when the price of polysilicon is high.
(2) Thin film becomes more attractive as its efficiency approaches that of other rooftop solar collectors.
(3)Thin film attracts more market attention as the capacities to make and deliver it grow.
(4) Thin film becomes more attractive when its ability to function in extreme climates becomes a market factor.
(5) Thin film becomes more attractive if incentives reward lower costs and less attractive if incentives reward higher efficiency.
(6) Thin film is less attractive when the risk of the delivery of ordered quantities is in doubt.

click to enlarge

The overall assessment of thin film’s future is a synthesis of all factors and potential markets.

11 major findings of the study:

(1) Total thin film capacity – which was 4.4 gigawatts at the end of 2009 – is expected to be greater than 10 gigawatts by 2012.

The rate of expansion in the sector will slow considerably from 2010. By 2012, a-si thin film will have a ~5.65 gigawatt capacity, CdTe will have a ~2.47 gigawatt capacity and CIGS will have a ~2.11 gigawatt capacity.

click to enlarge

(2) Asia will likely stay primarily with a-si and CIGS thin film. Silicon is a common substance and Asian entrepreneurs are familiar with it as a result of having long supplied it to the IT semiconductor industry. Asia and the U.S. will move toward CIGS as the many recent start-ups develop the capacity to realize the technology's potential.

(3) Producers will discover best practices in all 3 sub-sectors that bring costs down to ~80 cents per watt by 2012 but not all producers will use best practices. CdTe will likely go to ~70 cents per watt because, though Te costs may rise, thinner film will be achieved. Single junction a-si efficiency will top out at 8% but double junction with 10% efficiency will take over. CIGS will eventually cut its production costs.

(4) First Solar (CdTe) will remain dominant. Sharp (double junction a-si) will be 2nd. Showa Shell Sekiyu (CIS) will be 3rd. Solyndra (cyclindrical CIGS) will be 4th. QS Solar (double junction a-si) will be 5th. Those 5 companies were 90% of thin film production in 2008 and will be two-thirds of it in 2010 and 2012. Because China’s thin film makers are small companies, only 1 Chinese thin film maker made the top 5.

click to enlarge

(5) CIGS and a-si will not be significant players until 2012. Best practice cost reductions and efficiency improvements will take time to develop.

(6) Profits in thin film will be narrow because of the need to compete with ever cheaper crystalline silicon PV panels. This is especially true for a-si thin film makers and it is why single junction a-si will be obsolete by 2012.

(7) a-si will do best in utility scale projects in the very hot climates (for which it is best suited) of the Middle East, the U.S., India and China and will not do well in European and other feed-in tariff (FiT)- dominated markets and moderate climate rooftop markets.

click to enlarge

(8) Thin film will do better in rooftop markets as its efficiency gets to 12% and beyond, probably in the 2011-12 period. Laboratory results with CIGS suggest it could lead into the higher efficiency area.

(9) Beyond 2012, venture-backed CIGS may dominate because the technology may ultimate be more capable of low-cost large-scale production of a higher efficiency, more flexible product. Though the venture capital investment in thin film CIGS has not yet paid off, the spending on companies in this category (Nanosolar, Solyndra, Miasole) have recently demonstrated the capability to do the things that will pay off in the longer run.

(10) The new versions of CIGS are expected to begin taking market share from crystalline silicon panels and CdTe after about 2013.

(11) Consolidation of thin film companies is expected through 2012.

click to enlarge

QUOTES
- Harold McMaster, solar investor and entrepreneur, early 1980s: “…the essential cost element of large area solar arrays was glass, and [he] could treat the actual solar cell as simply a different kind of coating on glass.”

click to enlarge

- Shyam Mehta, Greentech Media Research: “…thin-film PV represented a technology that could be manufactured using glass’s high-thoughput coating process instead of the slow, cumbersome batch process of traditional silicon wafer-based PV – and one that had one-hundredth the feedstock requirement.”