(The author is a Reuters market analyst. The views expressed are his own.)
By Gerard Wynn
LONDON, July 5 The bankruptcy this week of United States-based Abound Solar rams home the victory of silicon-based modules in the latest round of a bitter technology race that has hurt investors, including the U.S. Department of Energy, and is far from over.
The two main photovoltaic (PV) technologies are based on crystalline silicon and thinner ("thin-film") cadmium-based semiconductors.
It was China's entry into the market five years ago that tipped the scales in favour of silicon after a price spike had made it less competitive. Chinese manufacturers chose the lower-tech crystalline silicon technology and immediately ramped up production, leading to a halving of solar panel prices last year.
That has demolished the share price of the world's leading thin-film module maker, U.S.-based First Solar (FSLR.O). Its shares have tumbled by 88 percent in the past two years.
Other victims include U.S. government-backed Abound Solar and the infamous bankruptcy of Solyndra, a thin-film company which defaulted on about $500 million of U.S. government loans. [ID:nL3E8I265Z]
China has now established a massive silicon PV industry, doubling its market share between 2007 and 2010, while creating a formidable barrier for competing technologies.
But the picture is more complicated than that. Silicon PV producers are also going bankrupt and the price plunge will run out of steam. This offers thin-film businesses a route back by fine-tuning module efficiency and field performance to bring down the total installed cost, only half of which is the price of the module itself.
And thin-film still has big backers, exemplified by General Electric's (GE.N) acquisition of U.S.-based PrimeStar Solar last year. <^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Market share by technology, NREL, page 29:
Market share by country, NREL, page 27:
Potential for efficiency improvements, IEA, page 22:
Solar PV technologies work by absorbing light, displacing electrons and causing a current to flow, captured in an electric circuit.
One problem for thin-film technology, as it tries to mount a fightback, is that the two main performance metrics are working against it: module cost and efficiency.
A more sensible measure is the full cost of power generation by the finished installation per kilowatt hour (kWh).
Thin-film and silicon PV producers say that they can now produce power at scale at optimum sites (such as Spain or in the Middle East) at about $0.12 per kWh, competitive with industrial but not wholesale prices for power from fossil fuels.
But that so-called levelised cost of electricity (LCOE) is project-specific and useless as an industry benchmark.
The LCOE of a solar park depends on local factors such as the weather and geographical latitude, plus installation and financing costs and the ingenuity of deployment.
That's a lot to take in, and the industry prefers the percentage efficiency of solar modules (the amount of light converted into electricity) and their cost per watt.
That doesn't make thin-film look good, and no wonder its producers are now reaching for LCOE.
First Solar is emphasising options to reduce LCOE, such as using trackers to follow the sun through the day and better management of power volatility. That makes sense, but also confirms its deterioration on conventional metrics.
Crystalline silicon (c-Si) modules are more efficient and may soon be cheaper than First Solar's leading thin-film technology, cadmium telluride (CdTe).
Commercial c-Si modules achieve about 14.5 percent efficiency, compared with CdTe's 12.5 percent. Meanwhile c-Si makers are manufacturing at about $0.75 to $0.90 per watt, targeting $0.55 to $0.70 by the end of the year.
First Solar says that its "core costs" are about $0.70 now and it is targeting $0.50 per watt and 15 percent efficiency by 2015.
Thin-film still has advantages. It has a higher theoretical maximum efficiency than conventional multi-crystalline silicon PV (about 30 percent, against 23 percent) because it captures a broader range of light wavelengths. It also performs better at higher temperatures and its manufacture uses less energy and raw materials.
Thin-film will survive, but it faces huge challenges to regain the initiative.
The c-Si, CdTe race, meanwhile, appears to have left others trailing in their wake.
An alternative thin-film technology, called CIGS (copper indium gallium selenide), is more expensive while achieving small efficiency gains.
Third-generation organic and dye-based solar cells are thin, lightweight and flexible, enabling cheap printing press-style production.
But they are far less efficient outdoors and are limited to valuable but niche indoor applications competing with batteries, such as in TV remote controls and wireless keyboards.
Meanwhile, concentrating solar power (CSP) uses a different approach to generating electricity, heating fluids to drive a steam turbine, though this has been cut short by the c-Si price plunge.
(Editing by David Goodman)
((Gerard.Wynn@thomsonreuters.com)(+44 207 542 2302)(Reuters Messaging: email@example.com)) Keywords: COLUMN WYNN/SOLAR
(C) Reuters 2012. All rights reserved. Republication or redistribution of Reuters content, including by caching, framing, or similar means, is expressly prohibited without the prior written consent of Reuters. Reuters and the Reuters sphere logo are registered trademarks and trademarks of the Reuters group of companies around the world.
Our top photos from the past week.