Silicon Valley shrinks cost of solar

Mark Pinto, Applied’s chief technology officer and head of its new energy and environmental business, speaks with the enthusiasm of an inventor freed to make his dreams come true. “What is exciting about solar for our employees is that we feel like it’s in our hands, we can play a very big role,” he says.

In his previous career as a scientist in Bell Labs, the research division of phone company AT&T, Mr Pinto was frustrated by a company culture that could not get great ideas into the marketplace. Now he is seeing the impact of combining his research efforts with business development in a fast-growing industry eager to embrace new methods and technology.

With semiconductors, he says, Applied has been dependent on oth­ers to come up with “killer applications”, such as Apple and its iPhone, in order to drive industry innovation and sustain demand for transistors and the equipment that makes them.  “Solar provides way less than one-tenth of 1 per cent of the world’s energy needs, so the potential market is huge. The killer app is already there – if you lower the cost of solar, the market is there,” he says.

Applied believes its technology is key to making solar energy cheap­er. Announcing its expansion into solar in September 2006, Mike Splinter, chief executive, promised to cut the cost per watt of generating solar power from $3-$5 to $1. “We plan to change the cost equation for solar power through adaptation of our existing technology and new innovation in order to help make solar a more meaningful contributor to the global energy supply,” he said.

In other words, in changing its own business, it is also changing the economics of a new industry. Ap­plied was in a unique position to do this because of its knowledge of handling silicon – a base material for both semiconductors and solar panels – and because of a separate foray into making equipment for flat-panel display makers.

Applied entered this business in 1991. In semiconductors, Applied’s core expertise is in making equipment that can create depositions – thin layers of insulating material on chip wafers. A similar deposition process is needed for flat-panel displays. Applied Materials and display manufacturers discovered they could be made more cost-effectively the bigger the size of glass from which the screens were cut. Applied’s machines for making the panels have grown so big they can no longer fit even on 747 jumbo jets and are mainly assembled on site. Now variations on this equipment have been adapted for solar panels.

Mr Pinto says the company initally assumed only panels small enough to be carried up ladders and fitted on residential roofs would be needed. But the industry has told it that bigger panels are re­quired for utilities to assemble solar farms and for other commercial installations. He says a kind of Moore’s Law is at work, referring to Gordon Moore of Intel’s 1965 paper that predicted the number of transistors on a chip would double around every 18 months. This was really about costs coming down with miniaturisation. With display and solar panels, the economics are also about cost per area, with the costs falling if the panels can be cut from bigger sheets. Installation is also much cheaper with fewer, larger solar panels.

Applied is focusing on thin-film solar, which exploits the photovoltaic effect of sunlight being absorbed by materials and converted directly into electricity. Silicon only 2-3 millionths of a metre thick is deposited on a glass substrate in the thin-film process. But Applied has also hedged its photovoltaic bet by buying companies that are expert in crystalline silicon, whose cells need more silicon, around 175 millionths of a metre thick. Its acquisitions in the past two years include Italy’s Baccini, which makes test systems for manufacturers of crystalline silicon, bought for $330m, and HCT Shaping Systems, a Swiss company that cost Applied $475m for technology that cuts silicon into pieces thinner than the slimmest salami slice.

Paula Mints, solar analyst with Navigant Consulting, says Applied has done the right thing, since she expects thin-film will grab no more than 50 per cent of the market. “There is always a cost/efficiency trade-off,” she says, alluding to the fact that thin-film is cheaper in needing less silicon, but it is less efficient in converting the sun’s rays to electricity.

In thin-film solar factories, Applied’s deposition equipment ac­counts for 70 per cent of the costs. It has therefore offered to equip com­plete factories for panel-makers, while integrating the other 30 per cent worth of equipment such as conveyor belts, sourced from other suppliers. The full solar production line it offers is another departure for Applied, says Mr Splinter: “We have changed the business model here too – when we started in thin-film solar there was basically no industry, so we started off a complete line. We’ve never done this before.”

Tim Arcuri, Citigroup semiconductor equipment analyst, says the Chinese order shows Applied’s stra­tegy of offering large-scale complete solutions is working. “It has been something of a holy grail to build monster gigawatt factories using these big pieces of glass to achieve a very low cost per watt,” he says.

The first Applied-fitted factories are coming on stream this month, equipment orders are expected to be worth more than $2.5bn by 2010, and Mr Arcuri says he can easily see solar accounting for a quarter of Applied’s revenues by then. But the company faces competition from rivals such as First Solar and Oerlikon Solar and the future path of solar technology has yet to be determined.

“This is a huge gamble for us,” cautions Mr Splinter. “We are betting a lot of shareholders’ money as well as our own research and development that this is going to be big and very successful. So far, so good – it’s the greatest opportunity the company has had in many years, but I think we still have a long way to go.”

The whys and hows of cutting what a watt costs

Applied Materials’ venture into providing manufacturing equipment for the solar panel industry is part of a surge of interest in Silicon Valley in the alternative energy.

Venture capitalists have funded start-ups trying new materials and methods to convert the sun’s rays into electricity, while Cypress Semiconductor span off solar-panel provider SunPower in 2005.

Google has the largest solar panel installation of any US corporate campus and has launched an initiative to explore solar thermal power – concentrating the sun’s heat to produce electricity.

With silicon currently the key material used in both semiconductors and solar panels, the Valley is in a good position to put its expertise to good use.

Mike Splinter, Applied’s chief executive and a Valley veteran, says solar reminds him of the early days of semis.

“People are jockeying for position; there are different technologies and materials and we are feeling our way,” he says.

“As the cost comes down, the solar market should scale much faster as there’s a huge market electricity, whereas then there was no computer market.”

According to Citigroup analyst Tim Arcuri: “There’s going to be a shake-out, with there being four or five different technologies and their different supply chains. But when there’s a scrum like this, I would not bet against the Valley.”

Paula Mints, solar analyst at Navigant Consulting, offers a different perspective: “Silicon Valley is a marketing term.

“I don’t think we have any more sand or processed sand than anywhere else. But we do have an area that draws interesting, capable, brilliant people to it.”