Closing Remarks

William Spencer

Washington Advisory Group

A REVIEW OF SEMATECH’s ROLE

Dr. Spencer reviewed several elements of SEMATECH’s development that were relevant to the needs of solid-state lighting. He said that in 1985, when SEMATECH was starting to be advocated, the U.S. semiconductor industry was in crisis and rapidly losing market share to foreign firms. Within a decade of initiating a government-industry partnership in the form of SEMATECH, it had come to understand the economic and scientific advantages of working together.

Cooperative activities started 1987 with the decision that the participating companies would contribute $100 million in cash; in-kind contributions were not accepted. This was matched by the federal government, mandated by Congress and supplied through DARPA. By 1994, SEMATECH had decided to end government support both because it had met its goals and because U.S. industry had returned to a position of relative strength. In 1995, it decided to become international and invited the main competitors of U.S. companies to join. Every region did decide to join except Japan, which initiated its own version of SEMATECH.

SOME LESSONS LEARNED

He summarized some of the lessons learned from the SEMATECH experience:

  • Recruit the best leaders. From the beginning, the very best leaders of the semiconductor industry were involved.3 Board members included the chief executive officers or key leaders of the major participating companies.

3  

SEMATECH directors included Robert Noyce, cofounder of Intel Corp.; Charles Sporck, CEO of National Semiconductor; and Robert Galvin, CEO of Motorola.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop Closing Remarks William Spencer Washington Advisory Group A REVIEW OF SEMATECH’s ROLE Dr. Spencer reviewed several elements of SEMATECH’s development that were relevant to the needs of solid-state lighting. He said that in 1985, when SEMATECH was starting to be advocated, the U.S. semiconductor industry was in crisis and rapidly losing market share to foreign firms. Within a decade of initiating a government-industry partnership in the form of SEMATECH, it had come to understand the economic and scientific advantages of working together. Cooperative activities started 1987 with the decision that the participating companies would contribute $100 million in cash; in-kind contributions were not accepted. This was matched by the federal government, mandated by Congress and supplied through DARPA. By 1994, SEMATECH had decided to end government support both because it had met its goals and because U.S. industry had returned to a position of relative strength. In 1995, it decided to become international and invited the main competitors of U.S. companies to join. Every region did decide to join except Japan, which initiated its own version of SEMATECH. SOME LESSONS LEARNED He summarized some of the lessons learned from the SEMATECH experience: Recruit the best leaders. From the beginning, the very best leaders of the semiconductor industry were involved.3 Board members included the chief executive officers or key leaders of the major participating companies. 3   SEMATECH directors included Robert Noyce, cofounder of Intel Corp.; Charles Sporck, CEO of National Semiconductor; and Robert Galvin, CEO of Motorola.

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop Convey your message publicly. These industry representatives were sent early to explain the purpose and advantages of SEMATECH to leaders of the government and private sector. Focus the program. SEMATECH did not try to approach the entire semiconductor industry; rather, it focused its attention on the equipment industry, which was about one-fourth the size of the semiconductor industry. This stimulated a flow of money similar to the flow seen recently from the automobile industry to its suppliers. Set measurable objectives. In 1988, SEMATECH began creating roadmaps. Objectives were to advance generic and precompetitive knowledge. The consortium did not support product development or other competitive activities, which would have raised conflicts among members. By focusing on infrastructure, members could work together on methods and instrumentation that benefited all without compromising competitive positions. Set uniform requirements. Participating members were not allowed to reduce their dues by opting out of particular programs. Everyone joined every program, so that support was not fragmented. Plan first, spend later. Dr. Spencer said that SEMATECH got off to a slow start as it felt its way toward mission. In 1988, he said, the government and industry gave $200 million to “an immature organization in Austin” that was not yet prepared to spend those funds wisely; this required several years to learn. He advised all consortia to have their roadmaps in place before launching programs. Consortium Accomplishments What was accomplished? One way SEMATECH defined success was to in-crease its global market share through a stronger equipment industry. The Economist did a study after SEMATECH “declared success” in 1993, and despite the journal’s general skepticism about government-industry partnerships, it concluded that an “amazing turnaround” had occurred in U.S. industry, and that some people were giving SEMATECH credit. Dr. Spencer added that SEMATECH’s members recognized that “we were out of crisis mode,” increased their private funding by 30 percent in 1994, and agreed that working together can bring economic advantages. In addition, members found the consortium to be a valuable forum for exchanging information. At the outset, when assignees first joined SEMATECH, most of them had strict instructions not to reveal any information of their own but instead to learn as much as they could. After a short time they found that “all the participants already knew everything” about the other participants, “the locks disappeared from the filing cabinets,” and a cooperative culture evolved. “One of the major things we learned,” said Dr. Spencer, “was that even though the semiconductor industry is highly competitive, companies can work together to the benefit of all.”

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop The roadmap developed in 1987-88 was expanded in 1991-92. It has proven to be a major asset for the entire industry. SEMATECH shared it with the world, and today the organization has over 1,000 engineers and scientists and an “evergreen” document maintained on the Web site at a cost of about half a million dollars a year. DISCUSSION I Mark Ginsberg Department of Energy Roadmapping Praise Dr. Ginsberg, whose primary area of responsibility is energy efficiency, said that energy efficiency is not the only potential advantage to an LED industry. He praised the roadmapping process, and said that his department’s Vision 2020 roadmap was an effort to bring an industrial perspective to an agenda that would allow government and industry to collaborate, as they have in other industries. A roadmap should include short-, medium-, and longer-term perspectives, he said. Many people believe the department’s role is to represent the high-risk, long-term, precompetitive perspective, but he suggested a variety of roles, including the reduction of information and policy barriers. A Growing List of Applications He described a “windows roadmap” for solid-state lighting, with the objective of creating active windows that are appliances within the wall. These active windows would provide all forms of information and entertainment, including educational material, news, and stock prices; a flick of a switch could change a representation of a cloudy day to a sunny day. “What I sense today,” he said, “is appreciation for a technology that is going to have more and more applications as we look more closely. We’re just touching on the edge of it.” He recalled the earlier discussion of Thomas Edison’s role in developing new technologies. Although Edison “may have been fortieth on a list of inventors for the original incandescent,” he had the genius of seeing its potential for lighting and the imagination to implement the concept in ways that no one else did. In the same way, he helped to invent the film industry. DOE’s Role Dr. Ginsberg said that DOE has a key role to play in implementing a solidstate lighting industry. The industry’s roadmap fits his own division’s mission

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop (making buildings more efficient and affordable) and the mission of DOE (research, energy savings, and industry partnerships). The development of solid-state lighting and white-light applications can be perceived as precompetitive, longer-term objectives that help accelerate applications. The national laboratories have a long history in this area, familiar roles to play, and a track record of excellence and accomplishment. DOE’s investment in solid-state lighting is still limited, but within the core capabilities of Sandia, Lawrence Livermore, and Lawrence Berkeley are people who can add to the effort significantly, even without a congressional budget line item for LEDs/OLEDs. Furthermore, he said, the department can channel R&D resources to move the effort forward. How Industry Can Work with Government Dr. Ginsberg read a quotation from Alan Bromley, former science advisor to President G.W.H. Bush: “Technological innovation depends on a steady flow of discovery by trained workers generated by federal science investments in universities and national laboratories. These discoveries feed directly into the industries that drive the economy. It’s a straightforward relationship: industry is attentive to immediate market pressures, and the federal government makes the investments that ensure long-term competitiveness.” He said that this assertion applies to the optoelectronics industry, which was why he chose to participate in the symposium. “I’m absolutely confident,” he said, “that working together we can advance the science and accelerate the applications of this very vital technology.” DISCUSSION II Charles Becker General Electric Corporate R&D Industry’s Need for Government Funding Dr. Becker said that he spoke for General Electric, the country’s largest lighting producer, in strongly supporting a U.S. lighting initiative. His company is pursuing both inorganic and organic LEDs and sees them both as large potential markets. At present they carry high-technology risk and require longterm R&D, for which “we can certainly use government funding to accelerate the work.” He said that GE’s initial-market view, which is similar to that of the other major companies, is that they will include mostly specialty and monochromatic applications, “and we have a clear roadmap of how to get there.” What is less clear, he said, is how to make the leap to the system level: how can the industry

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop move from small lamps and demonstrations to practical systems that are part of the real world? This is where industry needs help. The Need to Drive the Entire System His company also sees this market as highly substitutable, that is, there are options other than LEDs for advanced lighting, but no others that can produce light as efficiently as LEDs. To penetrate the general lighting market and realize potential energy savings it is necessary to drive the entire system and promote the acceptance of LEDs not only by the lighting industry but also by the architectural and building industries. An Informal Symposium Summary He said he would offer a consensus of what had been said by symposium participants. National opportunities. Solid-state lighting has the potential to bring about a 50- to 90-percent reduction in lighting energy usage in the United States, with possibly $90 billion in savings and large reductions in emissions. New system opportunities can also be expected, such as integration with advanced building systems and controls. End users can anticipate lower life-cycle costs and drastically reduced maintenance. The national technology base would be strengthened by synergies for wide band-gap materials, an expanded R&D base, foreign investment in the United States, and synergies with the defense effort. Economic opportunities. For industry, LEDs have the potential of being a multibillion-dollar domestic industry, both complementary with and disruptive to existing incandescent and fluorescent systems. Lighting manufacturers can anticipate high-growth products, which holds great interest in an industry that typically grows by 3 percent annually. The nation would benefit by increased high-technology employment and global competitiveness. Challenges. Development of an LED lighting industry would require technological breakthroughs on several fronts: a reduction in costs by a factor of about 50; a 10-fold improvement in efficiencies; a 50- to 500-fold improvement in total lumen output. Broader challenges include the development of standards (sockets, measurement techniques and standards, integration with building systems), a mass manufacturing infrastructure, and the education of end users. He diagnosed the current R&D base as healthy and characterized by great progress in the national labs and universities. Because of this foundation, he said,

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop industry could probably commercialize solid-state lighting on its own, but foreign firms that are being actively supported by their governments would almost surely surpass it. Costs Must Be Lower Initial costs must be low to drive acceptance, a lesson learned with great difficulty in the compact fluorescent market. He reemphasized the key importance of cost: “At 50 cents a lumen, we’re not playing in the market with any other lighting technology. Fluorescents are well under half a cent per lumen, so we have a long way to go.” Areas for Collaboration He described several areas for collaboration, all of which require long-term, precompetitive R&D: Standards. Metrology, socket definition, electrical supplies, control interfaces, color, and consumer education. Materials. Epitaxial growth with fewer defects; chip fabrication and structure for light extraction; heat management; phosphors and phosphor deposition; organic materials for OLEDs and device design for efficient light extraction from the OLEDs; encapsulance, both in OLEDs and inorganic LEDs; light-management materials and systems; and thermal management materials and systems. Manufacturing. Reactors for epitaxial growth; high-volume organic sheet manufacturing for large areas; lamps; high-performance, low-cost packaging, an area that is not a particular strength in this country; and low-cost electronics to convert LED lamps to standard 110-volt wiring. Applications. Collaboration with the architectural design and construction industries; efficient overall optical design to deliver light where it is needed; human factors; and fixtures. A “Straw Man” for the Future Finally, he suggested a “straw man” picture of the future. He began with the need to evolve and expand existing OIDA working groups and to build more cooperation with the national labs (especially DOE and NIST) and universities, as well as with other industry bodies, such as the National Electrical Manufacturers Association, Electric Power Research Institute, and international consortia. This plan recognizes the reality that all major lighting manufacturers are now global companies. Beyond that, he sees the need to clearly identify areas of precompetitive technology, formulate a roadmap and a formal budget, and iden-

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop tify opportunities to drive initial acceptance of LED products. “They’re not likely to fly off the shelves if we just put them out there,” he concluded. “We need to create the market and then publicize it.” Why Should LEDs Do Better Than Compact Fluorescents? Leslie Levine said that his company, Fusion Lighting, was supported by DOE to develop an electrode-less light source with long life and high efficacy. He said that many of the challenges described during the symposium related to his company as well. Compact fluorescents were introduced 20 years ago and have met with only moderate success despite their high efficacy; he asked why LEDs might be expected to do any better. Dr. Becker said that General Electric believed that price was the most important issue; for example, it is possible to convince a large retailer or hotel chain to buy efficient lighting, but the residential market is a difficult sell. It is essential to drive the unit costs of LEDs down to a level comparable to those of existing technologies. Dr. Bergh called Mr. Levine’s reasoning “a trap”: The objective, he said, should not be to develop an application that provides only a one-to-one replacement of the light bulb. Instead, the goal should be to usher in a wholly new lighting industry. He suggested two thrusts: (1) gradual replacement of lighting functions, and (2) simultaneous emergence of a new lighting paradigm that is recognized as different and attractive because of its low cost, flexible adjustment of the light spectrum, long lifetime, and minimal space requirements. This paradigm might be applied initially to new buildings and architectural installations but would gradually create a new industry that could move beyond the replacement of streetlights and signs toward the general lighting market. Leveraging Ongoing Work John Zolper of the Office of Naval Research said that his organization shared some of the interests described in the symposium, especially materials for the wide band-gap area, and suggested leveraging R&D investments already made, particularly in forming substrates and the growth of active device layers. Although the Navy effort might have “different flavors,” he said, many of the challenges would be similar and it made sense to coordinate research activities. SEMATECH’s Focus on Infrastructure A questioner asked how much of SEMATECH’s success could be attributed to supported infrastructure rather than process or design work. “One hundred percent,” answered Dr. Spencer. “Focusing on infrastructure was the key.” He said it was also the reason current members had stayed. Today a manufacturing

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop facility costs approximately $2 billion, with equipment costs representing 75 percent of that. If a company’s engineers have the opportunity to work with that equipment, they are less likely to make a mistake when they buy their own etching or deposition or lithography equipment. This can bring them a return worth many times their dues. He reiterated that much of SEMATECH’s success came from directing money to its suppliers for precompetitive work that did not provoke antitrust problems. Another participant noted that SEMATECH had a clearer challenge in designing a consortium because the research path was more sharply defined. In the optoelectronics industry much of the basic science is not yet understood. Dr. Bergh replied that the semiconductor experience provides a good lesson for the optoelectronics industry, which is about 30 years behind. He said that the only difference was one of timing. “It takes a new idea about 20 years to become an overnight success,” he said. Optoelectronics is perhaps 5 to 10 years from the point when SEMATECH was formed. When it reaches that point, it will be able to use what SEMATECH has taught. Integrated Circuits Versus LEDs Dr. Haitz noted a crucial difference between the integrated circuit industry and the LED industry. In integrated circuit work there is a common infrastructure: all companies buy the same equipment and use the same feature size. Differentiation comes in product design. For LEDs the epitaxial process is one crucial step where companies try to differentiate themselves; another is the LED substrate and the gas supply that goes into the epitaxial process. This, he said, makes it more complicated to follow the SEMATECH model. He was hopeful, however, that there would be common areas that could be adapted to the model. An important step is to find precompetitive areas and define them so that there is also room to compete. Dr. Spencer agreed with that point, but added that the semiconductor industry is divided as well. Some companies work on what is called the “bleeding edge,” the risky arena of buying new equipment and building large, expensive fabrication plants. One and two decades ago the memory manufacturers populated the bleeding edge. Over the last 10 years the microprocessor manufacturers moved into this space. A large part of the industry resembles the LED industry: the part that is buying equipment that has resolution capability of only 1 to 2 microns, rather than the 150 nm discussed in the EUV program. He suggested that one way to avoid the problem of different interests is to adopt the model of the Semiconductor Research Corporation (SRC), which is a much smaller activity that preceded SEMATECH by five years and funds only university-based research. At the outset the SRC’s main purpose was to ensure a supply of trained graduates, but it now supports some research aimed at the barriers that the semiconductor industry is going to face when it begins to encounter

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop quantum effects on very narrow gates and when it goes beyond the use of copper as an interconnect. In other words, it now funds longer-range, mission-oriented research in universities aimed at solving particular R&D problems. Concluding Points on a National Initiative . . . Dr. Wessner recalled Dr. Spencer’s reference to “selling” SEMATECH when the industry, despite its military importance, was under great threat. Members of the optoelectronics industry also face the challenge of trying to sell an opportunity and a somewhat distant threat. A second issue is the desire to form quickly a U.S. consortium to help U.S firms become competitive in the global market. In today’s global economy the major lighting companies are international, and consortia tend to be international as well. SEMATECH solved that in two stages, moving from a national to an international stance. One could argue in favor of forming an international consortium once a position of strength has been achieved. He reemphasized the importance of industry leadership and persistence. During the mid-1980s, the government was opposed to the idea of SEMATECH, and a sustained effort by Clark McFadden and many others from the private and public sector was required to convince government leaders to support a partnership. A final challenge is how to create a new industry for general lighting in parallel with a second effort to transform building and architectural lighting. Again, the initiative of private industry in providing funds for the Semiconductor Research Corporation may be relevant. Although the private sector should not be alone in taking such initiatives, it may be that innovations must begin there. . . . and a Final Note of Optimism Dr. Spencer echoed a point made earlier by Dr. Trimble. A governmentindustry partnership is best shaped by not restricting the natural innovative energy of private firms. Valuable collaborations can be created between ongoing research programs in government labs and federally funded, university-based R&D. He suggested a strong effort to leverage programs that now exist and to include private funding mechanisms, including venture capital firms. He concluded the symposium on a note of optimism, suggesting that both the private and government sectors had showed the ability and willingness to create the productive collaborations needed to bring solid-state lighting to a far larger marketplace.

OCR for page 89
Partnership for Solid-State Lighting: Report of a Workshop This page in the original is blank.