Applied’s Solar Strategy Goes Very Large

David Lammers, News Editor -- Semiconductor International, 11/20/2007 9:06:00 AM

Applied Materials Inc.’s decision to buy Baccini SpA (Treviso, Italy) for $330M, announced earlier this week, expands Applied’s capabilities in the crystalline silicon (c-Si) portion of the photovoltaic (PV) industry.

While the c-Si market is growing as more businesses and homeowners seek to connect rooftop solar panels to the electrical power grid, Applied CEO Michael Splinter has his sights set on the larger prize: thin-film PV modules.

Splinter seeks to leverage Applied’s ability to handle the very large glass substrates used in the flat panel display industry for solar power generation. By depositing thin silicon layers on glass, the thin-film PV modules sidestep the shortage of polycrystalline silicon wafers that, while easing now, hampered growth in the solar industry last year.

Applied is pushing the use of enormous sheets of glass, 5.7 m², for use in solar farms used to generate gigawatts of electricity. To improve the efficiency of large-scale electricity generation, Splinter said during the company’s recent financial results conference call, “The key is 5.7 — no one else is in the 5.7 m² space. The larger format will win.”

The plasma-enhanced deposition equipment used for thin-film solar lines can handle 5.7m2 glass substrates.

The 5.7 m² panels are aimed at solar farms, and reduce the cost of module installation, cabling and bracketing, he said. “At these major installations, we must drive the installation costs down with the larger panels.”

Asked by one securities analyst about delays in the comprehensive energy bill that has been under consideration by the U.S. Congress for much of this year, Splinter said he remains confident that the legislators will support provisions within the bill that support solar energy generation. Again, he weighed in on the need for the large-scale solar farms: “Today, there is not much of a market for solar in the United States. Most of it is in Europe, with expansions in China and India. In the United States, we see a few rooftops, but that [rooftops] is not going to be cost-effective.”

$1/W solar target

Applied’s goal is to reduce solar costs from the current $2-3 per watt now, to ~$1 per watt within the next two years and to 70 cents by about 2010. Splinter said he believes that national and state legislators in the United States will pass incentives and tax credits to make solar a competitive energy source. “We need clean energy, renewable energy,” he said. “In the end, the legislators have to find a way to do the right thing to support that goal.”

Asked during a recent interview on CNBC how large of an investment Applied is prepared to make in its solar business unit, Splinter said, “We think this market can be very, very big. Capital spending in the energy field right now is about $350B a year. If solar could be just 1% of that, it’s a giant step. If we’re talking on a five-year horizon, I would like to see solar energy take 1% of all cap-ex energy investment. Now, how much of that share we’d be able to get is a competitive issue.”

Moore’s Law for solar

Splinter came to Applied in 2003 from Intel Corp., where Moore’s Law and transistor scaling is the driving force. Splinter said Applied wants to foster a similar culture of cost-per-function reductions in solar.

Building around the 5.7 m² format, Applied has developed turnkey thin-film silicon solar module factory solutions, under the brand name of Sunfabs, that can be duplicated around the globe, much as Intel uses its “Copy Exact” strategy to share efficiencies at its various semiconductor fabs. While the key equipment would come from Applied’s own systems, the turnkey solutions also incorporate less-value-added systems from other suppliers.

Applied Materials' Sunfab layout includes critical deposition tools from Applied.

“If we can get many customers building Sunfabs, we can get thousands of engineers moving the cost down,” Splinter said during the conference call. “We believe we can be successful making 5.7 m² the solar farm standard. That is pretty much the path we are moving down. We have moved beyond horizontal [solar installations] to starting up the higher-capacity fabs for customers. We are guaranteering output, and working with customers to increase efficiencies.”

Splinter said he recently went to India, where a large-capacity module production facility is under construction with Applied as the turnkey supplier. Obviously pleased at what he saw there, Splinter said, “When I was in India, there were CVD and PVD systems already on the floor, the automation systems were set to go, they were finishing up the building. There were a ton of people there, working to complete it. The work is scheduled down to the hour on these kinds of projects. So we are excited about how fast we are going to learn as we build more of these Sunfabs around the world. The key is thin-film solar, and the 5.7 m² format.”

Charles Gay, general manager of Applied’s solar business unit, said the larger-capacity solar farms leverage Applied’s equipment used to process the large glass substrates used by the flat panel industry. “The LCDs are using the same-sized panel as the solar equipment, which is one of the reasons we can spin up our solar business so quickly. It is the same core platform. The only difference is that the piece of glass used in a solar panel is thicker, so the arms of the robot must be stronger.”

The slightly thicker glass for solar panels is needed for the panel to withstand the hail stones and seismic shifts in a solar farm. Beyond that, Gay said the gas and temperature controls, process enclosures, power distribution and plasma CVD steps are “immediately adaptable from what’s needed to build the transistors on the display side to building the solar cells on the photovoltaic side.”

Also, the energy industry’s support for larger solar farms supports the engineering work done to develop the large-capacity tools. “Five years ago, a display coater would have been 10 times too big for the throughput that the solar manufacturers were seeking,” Gay said. “As the market has grown, the size of a solar manufacturing line has grown as well, so a new size of tool is becoming relevant. Most of those tools come from the glass on the display side. What drove Applied Materials to buy Applied Films was that it was very relevant to the processing of the contacts, the plus and minus conductors needed on a solar panel.”

Exceeding financial goals

The uptake of Applied’s solar equipment has exceeded expectations, supporting Applied’s stock price at a time of relatively weak demand from the semiconductor industry. George Davis, chief financial officer at Applied, said 2007 was “a year where we began making substantial investments in our solar business; it was the year of entry into the solar market for our Sunfab product. We revised our initial estimate for solar orders from $200M up to $600M, and now we have exceeded that as well with $700M in contracts.”

Although it will be early 2009 before most of those orders convert into revenues, Davis said, “We are off to a good start.”

Baccini acquisition

After buying Applied Films, maker of  the Aton solar sputtering tools, and HCT (the latter for $463M, getting its precision wafer shaping equipment), Applied this week bolstered its offering on the c-Si end of the market by buying Baccini. The family-owned company supplies automated metallization, edge isolation, inspection and test, and the integrated handling systems required for the back-end manufacturing of the c-Si photovoltaic cells.

Mark Pinto, Applied’s corporate chief technology officer who went to Italy to announce the acquisition on Monday, said Baccini supplies a “fully automated screen printing approach that, looking forward, will support the ultrathin wafers, which are 30% thinner that the conventional wafers,” at <150 µm.

Pinto said Applied’s plan is to take the well-established Baccini integrated platform, combine it with certain Applied metallization skills, and expand the Baccini equipment capabilities to thinner linewidths. “The plan is to add process sequences onto a very similar platform” to the Baccini integrated offerings, which he said eliminates the extraneous wafer handling required of less integrated approaches. “We can support metallization all the way through to test and sort,” he added. The combined Baccini-Applied solution will address ~50% of the total cost of c-Si cell manufacturing, outside of the wafer preparation steps.

The acquisition is evidence that while the lowest cost per watt may come from large thin-film solar farms, the solar industry will continue to see many entry-level PV module makers targeting c-Si facilities that can generate 40-50 MW of electricity, he said.

In its 2007 fiscal year conference call, Applied had bookings of ~$98M for c-Si processing solar equipment. Splinter said there are substantial efficiency gains to be had in c-Si processing, mentioning improved color uniformity and thinner wafers as ways that can contribute to lower costs per watt.

“We have to get improved productivity if we are to get to 50 MW and above,” Splinter said. “We need to improve on wafer thinning, which is where the HCT acquisition comes in. Silicon is expensive, and the industry needs to cut the number of grams of silicon per watt in half. Then we need to keep moving on our roadmap to cut it in half again."