Thursday, February 7, 2008

The "Foolish" State of the Industry

Here at the Solar Coaster, my policy has largely been to generate original content or at least original takes on the latest policy, business and technology solar developments available on the web. But every now and then, I get tempted to reproduce articles wholesale, and today is another example. The following is investor website, Motley Fool's industry survey on the state of solar today:

Innovation Series: Solar Energy, by Toby Shute, February 6, 2008

Over the next several years, industry analysts peg the growth of the solar market at around 50% per annum. In a market moving that fast, it's develop or die. Because of the pace of change, it's hard to imagine a more perfect match for the Fool's series on innovative business sectors.

To do justice to developments in the solar area, I'll break my overview into three parts. We'll begin with standard photovoltaics (PVs), before getting into the more exotic thin-film and concentrated solar stories.

Standard PV under the microscope
For a sense of where innovations in standard solar modules have and will come from, it's important to get a handle on the industry's economics and the stages of the production process.

Each management team in the space has its favorite cost metric, which invariably casts the company as the most competitive coal-slayer of the near future. First Solar (NYSE: FSLR) claims the industry's lowest manufacturing cost per watt. SunPower (Nasdaq: SPWR) focuses on minimizing the cost of an installed system. I'm not completely sold on either metric, but both point to the industry's motivating force: making solar power as cheap as possible. It bears repeating that solar's Holy Grail is "grid parity," or achieving costs per kilowatt-hour on par with fossil fuel-derived electricity.

There are as many opinions on how to achieve low costs as there are solar companies, but the many stages of the production process open multiple avenues for innovation. In the beginning, there is metallurgical silicon, supplied by Hemlock Semiconductor -- of which Corning (NYSE: GLW) is part-owner -- and others. That silicon is then gasified, refined, solidified, melted, crystallized, sliced, doped, coated, fitted with electrical contacts, tested, framed, and sealed. I'm sure I missed a few steps, but that gets us close to a solar panel.

With so many steps involved, leading companies have naturally found tons of ways to cut costs. REC Group uses a closed-loop process that eliminates feed gas inputs and slashes waste by-products. Q-Cells has automated its entire production process. SunPower uses monocrystalline rather than multicrystalline cells to achieve higher efficiency across the same surface area.

Fundamentally disruptive fare captures most of the venture capital dollars these days, but there are still major innovations occurring in the traditional silicon realm. To name but one, REC's fluidized bed reactor technology is poised to slash the company's polysilicon deposition costs. This would further REC's goal of slashing its 2005 per-watt module production cost nearly 50% by 2010, akin to SunPower's aim of halving its installed system cost by 2012, and a good reminder that the more traditional players aren't going down without a fight.

Thin-film throwdown
In the near-term, thin-film technology excites folks because it eschews polysilicon, which is really scarce these days. But there's a more disruptive element that makes me think thin-film might eventually dominate the solar market.

First Solar, which uses cadmium telluride, is the biggest and best-known thin-film player, but there is a veritable cornucopia of startups waiting in the wings.

CSG Solar is eschewing wafers by depositing a thin layer of amorphous silicon directly onto glass; Ascent Solar (Nasdaq: ASTI) and Miasole are attempting to commercialize thin-film solar solutions based on copper-indium-gallium-diselenide (CIGS).

Perhaps most promising, though, is Nanosolar, a company that prints CIGS nano-ink onto highly conductive foil. The company counts the Google (Nasdaq: GOOG) founders among its early investors, it's officially commercial (although their limited production run is sold out for 2008), and the panels are reportedly selling for less than $1 per watt.

Some people dismiss thin-film because of its lower efficiency. The panels are cheaper and less powerful than standard PVs, which sounds exactly like the sort of disruptive technology outlined by Motley Fool CAPS mentor Clayton Christensen in his work on innovation. This is why I can't shake the idea of these panels' eventual dominance.

Let's concentrate here, people
Another contender is arguably the hottest technology around. The terminology, yet to be standardized, is known alternately as concentrator, concentrated, or concentrating photovoltaics (CPV).

These systems use the most efficient cells there are -- the ones developed for satellite applications by EMCORE (Nasdaq: EMKR) and Boeing's (NYSE: BA) Spectrolab. When focused with low-cost lenses, the sun's rays can be magnified hundreds of times, resulting in industry-leading efficiency, and the amount of active semiconductor material is inversely related to the degree of magnification. Naturally, the super-efficient cells are super-expensive, but they're a tiny piece of the overall package.

A traditional problem has been offsetting the heat that results from this intense concentration (I told you this technology was hot). Both SolFocus and Concentrix, two leading start-ups in this field, indicate that heat sinking should not be a major issue.

But just as with the thin-film entrants, it would be premature to declare victory before these new technologies are battle-proven. Demonstrating efficiency in a lab is far different from decades of exposure to the elements, and it's fairly inevitable that some of these unproven applications will sputter out well in advance of their warranty.

SolFocus may actually best First Solar and SunPower's cost metrics with the more holistic concept of levelized cost of energy -- the lifetime value of energy produced divided by total costs. This means that any innovator will not only have to win in the realm of production and installation costs, but the product will have to perform at a high level for decades.

In other words, the race to grid parity is a marathon, not a sprint.