Factor of Five
By Bill Moore
Fifty years ago, a pair of Bell Lab scientists introduced to the world the silicon photovoltaic chip, a revolutionary way to convert sunlight into electricity. Eventually their efforts would power a remarkable range of applications from orbiting communications satellites to off-the-grid cabins.
And while steady progress -- and even breakthroughs -- have been made in the last half century to reduce the per watt cost of photovoltaic panels, it remains one of the most expensive methods currently available for generating electricity when compared to centralized power generation from natural gas, coal and even wind.
But that situation may be about to change...finally. A California start-up manned by some of the most experienced photovoltaic manufacturing experts in the world is about to shake up the PV industry and helping lead that effort is long-time EV World reader and occasional contributor, Bill Yerkes.
Interestingly, it wasn't the activities of his new company that compelled Yerkes, whom I regard as a good friend, though we occasionally share conflicting political views, to contact me. He emailed me to remind me of a prediction he'd made years ago that cellulosic ethanol production was just around the corner. His email alerted me to an article describing the first delivery of commercially made, cellulosic ethanol, an event that now opens the door to a promising new fuel substitute for gasoline.
But my first surprise was the size of today's PV industry. According to Yerkes, it is now a $4 billion a year business. He told me that when he sold his first company, Solar Technology International, to Atlantic Richfield in 1978, they were projecting the industry would reach this size by 2000; so things are pretty much on track.
"All the oil companies are coming back," he told me, "... all the large companies that can build this kind of massive infrastructure."
Hopefully, that's a good thing.
The next revelation was that his new company, Solaicx doesn't actually build solar panels, which was a misunderstanding on my part. Instead, they manufacture the silicon wafers that are eventually fabricated into PV chips and then assembled into modules by panel makers. Solaicx's claim to fame will be their advanced silicon crystal grower that turns out a continuous flow of silicon ingots, unlike current Chaikovsky technology, which is based on a more energy-intensive, time-consuming and wasteful batch growing process.
In a way, this brings Yerkes full circle from where he started over thirty years ago, back in 1975, when he was buying his silicon wafers from a German manufacturer. His original plant was located in a 4000 square foot warehouse in Chatsworth, California. Every Monday, he'd receive a shipment of silicon wafers that he'd manufacture into PV cells and then assemble into modules and finally into panels, which he'd ship out on Friday. His first large order went to Explorer Motor Homes, a recreational vehicle manufacturer.
Yerkes explained that while the silicon ingots used by the computer industry as the basis of their microprocessor chips are essentially the same as those used by PV makers; the computer industry requires wafers that are thicker and have a mirror finish. They can also pay more for their wafers because they can as many as 3,000 individual microprocessors out of a single wafer.
Factor of Five
PV manufacturers, confronted by competition from cheap centralized electric power plants, are very cost sensitive, but their tolerances aren't as demanding as computer chip makers because a solar cell is a single diode that can be thinner and therefore less expensive. What Solaicx hopes to do is supply the world's top 10-20 PV manufacturers with their new, less expensive wafers that Yerkes estimates will reduce the cost of photovoltaic panels by a "factor of five."
How do they plan to do this? Their new silicon ingots are generated in a continuous, uninterrupted process that is computer controlled to insure quality. The ingots are also larger in diameter, meaning it takes fewer to make a solar panel. Finally, a new type of saw will cut the ingots into thinner wafers with less damage and materials wasted. The bottom line, Yerkes projects, is $1/watt solar panels, dramatically down from the current $3-5/watt wholesale. Assuming these savings are eventually passed on to consumers, this means a basic 1 kW PV system that today costs upwards of $6-10,000 (not including installation or inverter) will cost a mere $1,000, and will last at least 25 years. Yerkes thinks they could last much longer.
An important part of these savings is tied to the efficiency of the individual solar cells, themselves. The more efficiently a cell converts the sunlight into flowing electrons, the fewer are need to generate a given amount of current. Where thin film and similar lower-cost PV technology are in the sub-10% range, modern silicon-based cells are now twice as efficient at 20-22%. As a result, it takes half as many panels to generate the same amount of electricity. And half the number of panels, means lower installation costs. Yerkes stated that his wafers will enable companies to manufacture 20% efficient cells at a much lower cost. He also noted that because Solaicx cells are thinner and easier break, the company has developed proprietary systems for handling the wafers and installing them in modules that they will share with their customers.
Another benefit to thinner cells is that, depending on the design of the finished panel, they can more easily conduct heat away from the cell. Yerkes explained that as the sun heats up a cell, the output voltage drops slightly. This used to be a problem when charging storage batteries, but today most PV installations include very stable and efficient inverters that are less susceptible to voltage fluctuations. The thinner the PV cell, the more rapidly it dissipates summer heat and the less the voltage drop.
So, it turns out that "thin" really is "in" when it comes to photovoltaics.
I asked Yerkes about what changes he's seen over the last thirty years. He recalled that when he sold his first panels in 1975, he charged the equivalent of $15 a watt for a 20 watt panel. The diameter of the cells and their efficiency has steadily improved over the last three decades. He told me that his first wafers were a mere 3 inches in diameter. After he sold his company to ARCO, he convinced them to build a new, more automated facility in Camarillo, California, one that used 4 inch wafers and turned out 35 watt panels, many of which are still in operation. Today, most 100-150 watt panels, designed around six inch diameter wafers, go for about the same price as Yerkes charged for his 20 watt panel in 1975, showing a continual drop in costs per watt over the last 30 years.
According to Yerkes, Sharp Electronics is the largest PV manufacturer in the world at the moment, using 250,000 wafers a day, making it a highly desirable customer for Solaicx. Kyocera, BP Solar and Shell Solar are all roughly at the same level in terms of volume, Yerkes said. Below them is a third tier of firms with interesting technology including Sanyo with a new, 21% efficient cell that is based on low temperature manufacturing processes.
He is also intrigued by a new company being set up in the Philippines by a long-time acquaintance of his, Dick Swanson, called Sun Power. Their cells were the first to reach 20% efficiency and are now over 21% efficient. That plant will turn out 25 megawatts annually. It was Swanson's breakthrough solar cells that powered Honda to two victories in Australia's Darwin-to-Adelaide solar car race, Yerkes said. Swanson's cells also powered the Helios solar airplane.
Yerkes stated that not only will Sun Power continue to make high efficiency cells but Swanson has now come up with a way to do it even cheaper in mass production.
This contrasts with once promising lower-cost technologies like dye-sensitive and thin film amorphous which are struggling to get to 10%, with dye-sensitive cells at only 4%. He noted that in markets with heavy tax incentives like Germany, efficiency is less important than cost. In non-subsidized markets, efficiency becomes more important.
"I guess the bottom line of my point is all of these cells have been touted at one time or another as being the great white hope, sort of the silver bullet that was going to get down to the low cost and go into big production. It turns out that none of them got over ten percent, which the number you can calculate that is the minimum you have to be in order to pay for the structure to hold you [the solar panel] up in a wind and keep you from blowing over. There is a certain minimum efficiency, (even) if the panels were free, you couldn't afford to install them," Yerkes explained. This is why most of the large, successful companies continue to rely on higher-efficiency silicon-based PV cells, the kind Solaicx will turn out at a lower cost.
Switching to Ethanol
With all these potentially dramatic cost reductions in the offing, I asked Yerkes for his advice on when I should start considering installing solar panels in a state that offers virtually no incentives and does not allow grid connection?
His reply surprised me. He told me to invest in ethanol, instead and sell it to Californians to power their cars. California has now banned the sale of MTBE, an methane-derived oxygenate linked to ground-water contamination in wide areas of America. MTBE was originally required as an additive to reduce smog-forming chemicals emitted from car exhausts. Instead of MTBE, states are now being encouraged to switch to ethanol, which fulfills a similar function, but without polluting ground water. It is also a renewable fuel made from corn sugars. It is widely available as a 10 percent blend with gasoline in the Midwest. I use it regularly in my Honda Insight, further reducing my reliance on petroleum. Millions of American vehicles can even burn 85% ethanol known as E85, though only a handful of stations sell it. There are only two in the entire state of Nebraska.
In Yerkes' view, California Governor Schwarzenegger's "hydrogen highway" and George W. Bush's hydrogen fuel cell initiative are "pie in the sky."
"It don't think people understand that hydrogen is not a source of energy like oil or ethanol," he stated. "Hydrogen has to be made from something else like water or natural gas.. It doesn't create any additional energy; it's more like a battery. It stores the energy, and it's not even a very handy way to store [it]."
Like "Hydrogen Hype" author Joseph Romm, whom EV World will feature in our next edition, Yerkes believes we need to give people a fuel like ethanol that can go into their gasoline tanks, rather than a highly flamable, difficult to store and distribute, energy intensive gas like hydrogen.
It may turn out, Yerkes said to me, that the most important invention to come out the National Renewable Energy Lab (NREL) in Golden, Colorado will be its development of enzymes that rapidly breakdown cellulose into fermentable sugars and then ethanol. The then head of NREL in 1995, Charlie Gay -- who took over ARCO Solar after Yerkes left -- gave him a tour of NREL's fuels division and made this statement, which even surprised Yerkes.
He sees cellulosic ethanol in much the same way he looks at PV technology, both convert sunlight into energy. While a solar cell makes electricity directly, plants convert it into sugars that can be turned into liquid fuels, either ethanol or biodiesel. He sees America's vast agriculture and forest areas as the "Saudi Arabia" of plant-derived energy, where the waste or "stover" from everything we grow is a potential fuel source.
He talked briefly about a plan to harvest rice stover in California's Sacramento River delta area to make ethanol rather than burning it as it currently the practice, contributing to the region's air pollution ills. All kinds of crop and forest waste can be converted into ethanol, a fuel not unlike what most drivers are used to buying.
Efficiency is the Name of the Game
Yerkes, who now commutes between his home in Santa Barbara and his plant in Los Gatos, drives a Honda Insight because of its high efficiency. His new silicon ingot furnace is significantly faster (up to five times) than current technology and more efficient. The wafers his plant will produce will help improve the efficiency and drive down the costs of PV panels. So, Yerkes is very much a man focused on efficiency improvements.
So, when I asked him how he would write a national energy plan, he immediately replied that he would first concentrate of finding ways to improve efficiency in every area possible.
"I believe that figuring out more clever ways to reduce the amount of energy we need is really a big pay-off. We've been a country that keeps doing things the same way until we have a crisis."
Yerkes admitted that he doesn't think we need lots of government-funded research and development, especially for hydrogen. He'd prefer to see some of that money put into the construction of cellulosic ethanol plants all over the nation. He'd encourage its use in high-efficiency, hybrid-electric vehicles. From a global warming perspective, which he is personally still uncertain of, the net effect of converting plant matter to a fuel adds no additional carbon dioxide to the atmosphere.
The key to his energy plan is to do things that are doable today, not fifteen or twenty years from now. He has an innovative idea that combines minivans, cellphones and computer technology that EV World plans to feature at some future date. We also plan to talk to him about battery technology since he's also an expert in this arena, as well.
The Second Lieutenant Theory
While Yerkes isn't opposed to some "blue sky" spending and research, he believes in what he calls the "second lieutenant" theory, that the only way you are going to take a hill is for the "second lieutenant" to get out of his fox hole and lead his squad up the hill. The generals aren't going to it. Each of us, individually, needs to take the initiative, a notion to which I also subscribe.
By way of example and returning to the PV industry, Yerkes noted that all new subdivisions being planned and built in California now must include solar panels on their homes or the subdivision won't be connected to the power grid. While the per house PV output might be just 1 kW, this is sufficient to mitigate the construction of additional power lines and the importing of power from outside the state during the summer when electricity demands spike. This has the added advantage that the cost of the installation of the PV system becomes part of the mortgage, and as Yerkes pointed out, the system lasts many times longer than the dishwashers and carpets that are part of the cost of the original house.
"The banks out here are quite happy with kind of thing now because it means the houses get a little premium," Yerkes commented. "Houses that mention anything about solar and saving energy do quite well."
While Yerkes isn't as excited about retro-fitting existing homes with PV systems because of the added installation and engineering costs, he does see integrated PV systems in new homes, where the system is engineered into the roof structure and thus doesn't require its own mounting system, as a future trend, especially in California, Florida and elsewhere where summers are hot and there is abundant sunshine. He estimates the installation cost is half that of retro-fitting an existing home.
As we neared the end of our interview, he promised to do a follow-up with me on the state of advanced battery technology and his transportation-on-demand idea. If anything, Bill Yerkes is a second lieutenant that has definitely earned a "battle field" promotion for assaulting and capturing more than his share of "hills."
Editor's Note: This article represents a summation of our 45 minute-long conversation, the entirety of which is available in MP3 format by using the MP3 player at the top of this page or by downloading it to your PC and playing on your favorite MP3 device. We encourage you to listen to the entire discussion when you have the time.