U.S. Department of Energy
Doug Rose, SunPower
Charlie Gay, Applied Materials
Kevin Hutchings, IBM
John Gloekler, Apogee Solar
James Moreland, SolarWorld
Panel members were asked to characterize the role of their company in the PV industry. They responded as follows:
Dr. Rose: SunPower is the leader in most PV markets in the U.S., with a worldwide workforce of about 5,000 employees.
Dr. Gay: Applied Solar, part of Applied Materials, makes equipment used to manufacture solar panels around the world; the world’s largest producer of equipment for panels; about 1,500 employees in the solar business.
Kevin Hutchings: IBM is “at the center” of several semiconductor collaborations and has supported SRC, SEMATECH, and others. With its experience, skills, and intellectual property in semiconductors, IBM has much to offer the PV industry.
John Gloekler: Apogee Solar is developing a 50-micron solar cell and has developed a fabless process to totally outsource manufacturing to existing facilities.
Jim Moreland: Solar World of Oregon, with headquarters in Bonn, Germany, is expanding and hiring in a state with 10 percent unemployment; working with silicon and multicrystalline silicon.
THE ISSUE OF A PV ROADMAP (CONT’D)
Mr. Lushetsky thanked the panel members and asked them to continue the discussion of a PV roadmap. He said that contrary to the impression of some, this discussion was “not part of a DoE plan to exert control over the industry.” In fact, he said, much of the conversation was initiated by materials suppliers who had approached DoE and asked for guidance in understanding where the industry is going. They said that they were used to dealing with roadmaps in the semiconductor industry and suggested that DoE could play a role. “What would you say to those suppliers,” he asked, “that could help them be suppliers to you, and ultimately help you to be more competitive?”
Dr. Gay clarified that despite the name of Applied Materials, it is not a materials company but an equipment company that works closely with materials companies like Dow Corning and DuPont, and with the producers of feedstocks. He said what they can do is combine the tool with the material in a way that reduces cost for the PV customer. They have met with the raw materials suppliers, he said, and described their strategic plan around the roadmaps.
A ROADMAP TO GUIDE THEIR BUSINESS
“Everybody has a roadmap that guides their business,” said Dr. Gay. “This is what allows them to receive financing and to establish their identity. We’ve brought together a lot of folks who want to help. In PV we relate to things that are in our past, so it’s exciting to have people from the IC industry here wanting to help. What I think we can do is have a helpful exchange about what is similar and what is different. One thing that is different is technology half-life. In IC, the technology half-life is about 18 months, when another node is reached—Moore’s observation. In PV, the technology half-life may last a decade. So in PV we do need to plan, but part of it is a set of ideas about how to bring in financing.”
Dr. Gay said that the two industries sometimes seemed more similar than they were. He recalled the comment by Dr. Rose that PV may use 5-, 6- or 8-inch wafers. “You know where the idea of a wafer came from?” he asked. “From the IC industry. We got from them the notion that the form factor needed to look round like a wafer, instead of like a rectangle or something else. People start to differentiate around those fundamental ideas. What we want to do is lay the foundations for the long haul, and get the time constants right. We need to have the universities link in here.” He referred to an earlier comment by Jim Sites about enabling students and faculty to engage in ways that allow innovation to emerge.
REACHING OUT TO THE BANKING COMMUNITY
Dr. Gay also noted that one of biggest barriers to the PV industry today is in the banking community. “They look for consultants to guide them on what
technologies in PV to invest in. I’ve been a consultant,” he said, “when I had no job. Banks rely on other people, when they should be calling Jim Sites. Part of the networking that could be helpful here is to be aware of the fact we’re bringing in a lot of new stakeholders, and then call the Jim Siteses. It takes the IC world, the materials supplier world, and the banking community to actually create the financing models that could allow this industry to scale. We have a chance here to aggregate a lot of stakeholders, because we want to change the situation of where we get energy.”
A ROADMAP TOWARD COMMON GROUND
Dr. Gay noted that PV stakeholders also have a lot of work to do in educating the pubic about the need for renewable energy. “We need all of you, and all the stakeholders we can possibly engage to try to get together. Whether we call it a roadmap or something else, this process of getting together and forming partnerships in setting goals on how much clean energy can be adopted and how much solar can be adopted would be a great thing for us to build on after this conference. Finding that common ground is very difficult, but it’s up to each of us to find partners with whom we can cooperate.”
Dr. Rose agreed with the need to gather and communicate this information. “Having the new companies learn more quickly where they should look and be able to communicate information upward is essential. But I want to caution against the idea of a roadmap that gives what some of the materials suppliers want—the roadmap they’re used to in the IC industry. Every industry is different. The makers of parts for jet engines are probably used to dealing with two or three customers, but that would not make sense for the IC industry, and it certainly would not make sense for the PV industry. We deal with some suppliers who are really uncomfortable because it is different from what they are used to, but they see that here’s an industry that could grow extremely rapidly, and they’ll deal with the complexity of it in order to participate.”
GUIDANCE FOR SPENDING LIMITED R&D FUNDS
Mr. Hutchings said that one benefit of a roadmap for the photovoltaic industry was that it could help extend the value of limited R&D money. “A company can’t do everything,” he said. “We discussed giving people the freedom to pursue all kinds of ideas, versus a narrow view. But limited money for R&D is the reality. One benefit of a roadmap is to make sure those limited R&D dollars are spent well. Whether or not you take the high-end, top-level view of cents per kWh, which makes a lot of sense, you’ll ultimately have to figure out how to get there. You’ll get into discussion with suppliers who say: I only have this much money to spend, where do I spend it? That’s coming—especially if you believe this is all about how to strengthen manufacturing in the United States. If so, you have to look at the
cost per kilowatt-hour, and say is that coming down fast enough for us to compete globally. And once you can answer that, you’ll find out whether the roadmap is adequate, because it should tell you where to spend these limited R&D dollars.” It is the same for the DoE, he added, which also has limited money, and has to decide where to spend it. “You need some guidance. I think there’s a need for roadmaps. There are details that have to be worked out that are unique to the PV industry.”
WHY COMPANIES NEED STANDARDS
Dr. Guha added an argument in favor of standards. “Companies need standards to get involved and grow the critical mass of this industry,” he said. “Without standards, how do I know [that] I can have a market for my product and my R&D dollars?” Citing his own experience, he noted a time when there was no market for Wi-Fi; “But there were six standards before there was any market. As the market came, one standard emerged: 802.11b. This was followed by ‘g,’ and then ‘n.’” Knowing those standards in advance, he said, gave companies the ability to invest in new technologies with lower market risk.
Dr. Moreland said that his company has to be able to talk to its suppliers with or without standards. “What I’ve found,” he said, “is that they have their own idea of what PV is and what PV needs. They try to help us but end up hurting us in terms of the quality of the materials. So that conversation has to happen, and maybe, to some low level, at least some standards would be useful.”
DECADES OF WORK ON STANDARDS
Dr. Gay said that he didn’t want anyone to leave the symposium thinking there were no standards. Since the early 1980s, he said, a consensus process had been at work throughout the industry to agree on standards as they were needed. The process of setting PV standards is organized by the International Electrotechnical Commission (IEC), and followed by manufacturers of modules, installers, and others in the industry. “This is always evolving and improving,” he said. “A lot of people have worked for a very long time on this, especially the technical committee on PV15 charged with assuring the quality, durability, electrical integrity, safety, and so on for the PV industry. “In addition,” he said, “NREL, Sandia, Brookhaven, and others have worked on environmental, health,
15The International Electrotechnical Commission (IEC) is a global organization that has prepared and published international standards for all electrical, electronic, and related technologies since its formation in 1906. Its technical committees are charged with preparing standards on many electrotechnical topics; e.g., TC 47 concerns semiconductor devices, and TC 82 concerns solar photovoltaic energy systems. TC 82 holds a plenary meeting every 18 months where working groups are charged with writing standards on many topics. TC 82, created in 1981, has recently considered standards for such topics as flat-plate PV modules, concentrator PV modules and assemblies, and installation and safety requirements for PV generators.
and safety standards and served as an independent third-party validator of safety standards. All of this,” he said, “is to ensure the global integrity of what is being offered and warranted for the customer.”
OPPORTUNITIES FOR PRECOMPETITIVE PARTNERSHIPS
Mr. Lushetsky added perspective on the difference between the PV industry and the semiconductor industry. “Put simply,” he said, “the IC industry is one materials set with an infinite number of circuits; the PV industry is one circuit with an infinite number of materials.” He said that that formulation helped him to frame the issue for himself. “Where the IC industry was able to collaborate on materials, we clearly run into differences in PV.” For other technical issues, however, such as metrology, material handling, and deposition tooling at a high level, he suggested that opportunities for precompetitive partnerships might be found. Another opportunity for collaboration may be installation cost, which dominates total system cost.
Dr. Rose said that he, too, could see areas for productive collaboration, and there are some ongoing now. For inverters there is a standard now in place that addresses anti-islanding, interconnections, and safety, but there is also intense development and standards activity to integrate inverters with the smart grid, providing information and voltage support to the grid while increasing power point matching within the array and decreasing installation time. In addition, there are “areas that the industry has poked at but hasn’t closed on.” This included the module energy rating. “This is one I’d love to see good collaboration on,” he said, “so consumers can know that if they buy a particular module, and put it in a particular climate, they can expect a certain amount of energy production.” He suggested that many areas naturally invite collaboration, but he proposed letting those areas “develop organically while the major emphasis goes to the big picture roadmap that we communicate up to the Congress and other constituents.”
A ROADMAP AROUND A CORE OF INSTITUTIONS
Dr. Gay suggested building a roadmap around a core of institutions, such as the partnership that currently exists among NREL, the Colorado School of Mines, Colorado State University, and the University of Colorado at Boulder. “There’s a model that works well,” he said, “and we could replicate it around other national labs.” He said that one area of emphasis for a roadmap should be integrating PV into the grid, and within that area, to improve the communication of grid information to the utilities. “The United States lags China by a decade in this,” he said. “We get grid information late because our systems gather and aggregate data only every 15 minutes. It’s one reason we’ve had blackouts in the Northeast and Northwest.” He said that the electrical grid in China has real-time detectors that display the waveform constantly so that operators can act immediately on current information. He also said a roadmap should plan how to integrate
renewable power into the grid, moving PV-generated power from the Southwest throughout the nation, just as wind generation companies are planning to move power “from North Dakota to places where it’s needed. It’s the same paradigm the rest of the world is already working on,” he said, “and they are ahead of us. So I’d vote for pulling in the national labs more closely, building the relationships to the universities, and expanding how we think about real-time use and where we link solar with the grid.”
Mr. Hutchings raised the issue of whether individual PV companies would be willing to raise the money needed to achieve a roadmap. The industry would first have to decide whether the rate of improvement is now adequate for the companies and the United States. If it is not considered adequate, U.S. companies may decide a roadmap is desirable. This may cause them first to address the issue of what is precompetitive and what is not, he said. Then they may have to pool resources, “which is what happened in semiconductor industry for different reasons.”
A LACK OF TRAINED PEOPLE
Dr. Gloekler said that one of biggest challenges to the U.S. solar industry is that it has been unable to commercialize technology rapidly enough, especially when compared to Chinese firms. This, he said, was due to a lack of people skilled in commercializing technologies. And it is true for many industries, he said, where the United States has advanced technologies, but the transfer from lab to full-scale production proves to be more difficult and time-consuming than estimated. He said that it was not a question of more spending, but of bringing more talent from the semiconductor industry. As an example, he said that Dick Swanson, the founder of SunPower, credited the six months he spent in Austin at SEMATECH with grounding him in essential principles of silicon manufacturing. “There’s a very large challenge in doing baseline processes and developing reliability around process development that’s not in the industry today,” he said. “We need the talent pool in here from experienced bases that can help us drive these technologies a lot faster to market.”
Dr. Moreland said the industry also needs to expand the pool of young talent. One way, he said, was to work through a consortium like the Silicon Solar Consortium (SiSoC), where industry members form partnerships, determine what is competitive and what is precompetitive, and develop their own working relationships. At the same time, they can stimulate and help students gain experience in PV so there are more knowledgeable people to hire.
START-UPS UNDER PRESSURE
Mr. Lushetsky asked Dr. Gloekler, because he represented an early-stage start-up, to describe the current challenges to his firm and compare them with the challenges of 12 months ago when financing and other issues were more favorable. Dr. Gloekler said, “I can tell you there’s almost no investment out there
today. VCs consider solar a commodity, which is the death knell any time you’re trying to raise money.” He said that it was very hard to get VC interest “even if you have innovative technologies that can shift the game.” He said that the VCs “really want capex16 efficiency. One, we have to show that there’s essentially no technology risk. Two, our time to market has to be one to two years. Three, we need to run it on a lean model overall. You can only achieve that as a start-up if you have the ability to use an SVTC or some other type of prototyping facility that allows you to take your one area of technology, prove it, and sell it with the entire supply chain intact.” He also said that “the standardization of equipment interfaces is going to be pretty fundamental to us, so we can sell to many manufacturers. Start-ups are under a lot of pressure right now.”
WORKING WITH UNIVERSITIES
Mr. Lushetsky asked how companies could work more closely with universities, and Dr. Moreland said that in Oregon he had significant relationships with a consortium of universities, some of which were good in characterization and some in engineering. One university sends interns at the undergraduate level to work with the firm for about nine months, and the students use that experience to write papers that help them get their degree. Another university sends masters students to work in the factory, and they write their thesis based on the work they do. They are considered employees, and the company has a realistic setting in which to evaluate them. “If they’re good,” he said, “we hire them. Win-win.”
Dr. Gay said that Applied Materials worked with about a dozen universities on specialized projects, typically electronic or optical modeling tasks, and is also part of the SRC program and the consortium headed by North Carolina State University. “We participate in working with universities either directly or through these consortia. We see that as a critical part of building up the talent pool.” He said that many of the faculty at partner universities send their graduate students to his company when they complete their studies. “Those labs and those students are how we’ve been able to go from five employees three and a half years ago to 1,500 employees today. Five or 10 years ago that would have been impossible to do. We need those students to be in the pipeline.”
More Partnering Between Agencies
George Rozgonyi of North Carolina State University said that he would like to see more government partnering between agencies. Referring to the request by Elaine Ulrich of Congresswoman Gifford’s office for some action items to present
16Abbreviation for capital expenditure.
to the Congress, he suggested that participants forward a more complete list of needs. “I have my personal list,” he said. “If I look at my colleagues in the states who need graduate students, I would say we need a competitive program to support applicants to graduate school and make it attractive enough for people to apply.” The universities also need equipment, he said—specifically enhancements to existing university equipment so it is more useful to industry, including diagnostic and processing equipment (NC State), computer simulations (Texas Tech), and equipment in the DoE center at Georgia Tech. “We need to have a focused action item for enhancing these PV-oriented university programs. I think the government and NSF program should be recognized for what it is—an industry/university cooperative research consortium—and better coordinated with DoE and DoD and NIST and the national labs.” Ms. Ulrich thanked Dr. Rozgonyi for this request, and asked the group for more, especially “what you all feel you need collectively.”
PV Will Be Successful; Will the United States?
Dr. Rose agreed that the need for more student funding for PV is a good example of the bigger picture referred to by Dr. Rozgonyi. “When I was at NREL, and when I was at First Solar,” he said, “I used to worry that if we didn’t reduce costs, PV may never fulfill its potential; it would always be the technology of the future. The good news is that that’s no longer the case. PV in the rest of world is going to be very successful. And a few U.S. companies will be part of that success. The challenge before us is what can we do in the United States to have large-scale manufacturing and use in this country? It will take actions on a big scale, because that is what other countries are doing. If we compare the level of U.S. funding on PV to the importance of it and to what other countries are spending, the gap is evident. We need to better coordinate the labs, pull in more labs that are working on other things, and build the downstream channels, for instance by aiding demand with a green bank or other mechanisms. When we put that whole list together we’ll have the action plan that will make the United States the leader in clean technology—instead of wondering why those other countries were so successful.”
Richard Bendis said he would like to make a “closing comment that’s positive.” He reported that he had just heard of a $75 million award from a consortium of VC firms to a new company formed to manufacture PV panels. He expressed the hope that this news heralds a positive note for the solar industry at large, especially since many companies have had a tough time securing VC funding during the height of the recent recession.
A Closing Word on the Value of Roadmaps
David King, who said he had worked both with NIST and a range of private firms, offered a closing comment on the value of roadmaps, especially of the type
developed by SEMATECH. He focused on three points: One, roadmaps are powerful instruments for federal agencies to use in planning disbursement of federal funds for programs viewed as germane and with commercial potential. Two, they are powerful tools for companies inventing and developing technology—not only new materials but also new manufacturing equipment. Because the roadmap goes to companies beyond a particular sphere of influence, it can bring in partners who find the technology germane and judge it a timely fit with market conditions. And three, a roadmap is a powerful tool for people to study if they’re in graduate school or even high school and planning a career. “When will this technology be ready?” he said. “Clearly, if it’s going to be 15 years before you are creating new jobs, that’s not a place for me to be. If job will be there in three to five years, the roadmap will tell me—and I can get myself ready.”