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 79
Partnership for Solid-State Lighting: Report of a Workshop Panel IV: Solid-State Lighting Roundtable INTRODUCTION Clark McFadden, Moderator Dewey Ballantine Dr. Wessner introduced Mr. McFadden, who has played a major role in industry-government partnerships, including SEMATECH. Mr. McFadden said that the current STEP initiative, under the leadership of Gordon Moore, had examined partnerships of different areas, sizes, scopes, and effects. He said that a partnership for solid-state lighting “seems promising,” with the potential for major savings, new applications, and new technologies. Most of the elements of a collaborative initiative are within reach, he said, and the challenge now is to find the best plan for moving forward. He noted the many years required to move the light bulb from its first demonstration to widespread use and urged a more expeditious pace in bringing OLEDs to market, with greater benefits for all participants. CAPITALIZING ON INVESTMENTS: THE INDUSTRY POTENTIAL Steve Domenik Sevin Rosen Funds Mr. Domenik characterized himself as both an entrepreneur and a supporter of ventures in technology development. He said he had come to the workshop primarily to deliver a “short message of optimism” and to suggest that the venture capital industry could assist in creating government-industry initiatives. The
OCR for page 80
Partnership for Solid-State Lighting: Report of a Workshop solid-state lighting industry, he said, had reached a technical stage where it needs outside funding to pool its resources and coordinate R&D efforts. He made the point that in the not-too-distant future, when systems are ready for commercialization, the venture capital industry would likely be interested in helping and participating. The Rapid Growth of Venture Capital He sketched a picture of the rapid growth of the venture capital industry over the last five years. While he predicted some retrenchment after the stock market decline of 2001, the industry has the capacity, even in his own small partnership, to take technologies all the way from the lab to the marketplace. In addition, the growth of the industry means that it is more competitive and individual firms have to seek promising technologies and take the great risks. This has not always been the case. Until about 1995, the venture capital industry invested a few billion dollars a year in new enterprises. The average investment required about seven years from time of investment until it was possible to sell the investment. In the last five years, he said, the size of the industry has mushroomed to a hundred billion dollars or more per year. This growth attracted many new people and firms to the business and many of the businesses they invested in had questionable or no technology content. The Stock Market Correction Is Not All Bad The 2001 stock market correction, he suggested, would be beneficial in the longer term, as people begin to correct their thinking about the venture capital industry. With the NASDAQ down by about half, there is half as much money to invest; many venture capital funds have vanished, and more will follow. Those that remain, he said, will regard their investments more soberly. The returns will not be as exciting or as quick, and investors will expect to be paid for results rather than promises. Having a smaller pool of money (perhaps $50 billion a year rather than $100 billion) will not be all bad. This amount is still an ample pool, and it will continue to be fed by those for whom venture capital is part of their allocation model: wealthy individuals, university endowment funds, and pension funds. A Partnership Between Venture Capital and Industry A key point, said Mr. Domenik, is that over the last 5 to 10 years there has evolved a partnership between venture capital and industry, especially in the United States. This bond is so strong that industry has now come to rely on venture-capital-funded enterprises for the commercialization of new technology. One reason for this dependence is that when a private firm underwrites a $50 million
OCR for page 81
Partnership for Solid-State Lighting: Report of a Workshop development project, it must find a way to record the equivalent of a $50 million loss on its financial reports. When a venture capital firm takes a loss, it is a private matter not reported on any public income statement. On the other hand, if the investment succeeds in creating a viable business, a public company can acquire that business and place it on their balance sheet at effectively no cost. Even in the current downturn, he said, his firm is approached by companies that want to bring the products of young firms to the marketplace, either through acquisition or through partnership. More Competition Between Venture Capital Firms Another consequence of the growth of the venture capital industry is that it is much more competitive. This business “used to be a bunch of middle-aged technology executives investing in deals with their friends,” said Mr. Domenik. “It has become much larger, more institutional, multinational, multiracial, broad based.” The competition is driving firms to invest in earlier-stage, higher-risk activities. Some of the projects Mr. Domenik’s firm is pursuing involve university professors in their laboratories supported by government grants. “Every one of these are early enough and risky enough,” he said, “that it keeps me awake at night.” With the growth of the venture industry, individual firms have specialized. Some invest at very early stages, like Sevin Rosen; others invest at later stages of development. Mr. Domenik’s firm, after 20 years, always invests during the early stages, specializing in technology and practicing a hands-on approach. The Example of Capstone Turbine An example of the companies Sevin Rosen has invested in is Capstone Turbine. Eight years ago this company was the risky but promising idea of a group of former rocket engineers who wanted to develop a new microturbine technology for distributed power generation and other uses. This technology, which had received early-stage support from several government initiatives, showed great promise but was far from commercial readiness. After more than $200 million of development work, the technology had proven to be reliable, versatile, and popular, and the once risky investment has paid off in a successful IPO and rapid subsequent growth.1 Mr. Domenik closed by saying he would like to see a solid-state lighting initiative succeed. It has excellent technical content, he said, as well as huge 1 Capstone Turbine Corp. went public in June 2000. Its line of MicroTurbines operate on the principle of a jet engine and can use not only natural gas, diesel, kerosene, and propane but also underutilized or waste fuels, such as oilfield gases. MicroTurbines power hybrid-electric vehicles, convert waste gases into electricity, run micro-cogeneration plants, and perform other functions.
OCR for page 82
Partnership for Solid-State Lighting: Report of a Workshop demand and social benefit. Even though his firm does not normally invest in pure science projects, he sees an exciting potential for commercializing solid-state lighting. A PARTNERSHIP OPPORTUNITY? Arpad Bergh Optoelectronics Industry Development Association Dr. Bergh asserted that government involvement in partnerships usually brings a variety of benefits that are not well recognized. In Japan, the government-industry relationship in electronics has been very successful; a decade ago, Japanese industry was fully networked while American industry was not. He suggested that this was at least part of the reason why Japan’s market share in optoelectronics was 70 percent and U.S. market share was 10 percent. Government’s Role in Networking and Setting Direction One important function of government in such a partnership, he said, is to set strategic directions and to promote industrial networks and collaborations. When private firms are not networked, each will move in separate directions toward niche applications. For the industry as a whole to move toward a common goal, such as the use of LEDs and OLEDs for general lighting, there must be some mechanism to coordinate objectives and agree on directions. He referred to several successful instances of this. DARPA and the National Science Foundation had played major roles in providing a vision for the U.S. communications industry, which thrived under government guidance; the U.S. share of the world market in optoelectronics has grown from 10 percent to about 35 percent. Another compelling reason to network industry is to build an intellectual property base. Finally, as the experience with SEMATECH showed, it is necessary to build a manufacturing infrastructure, and this requires collaboration. The Role of OIDA In addition to a government partnership, Dr. Bergh explained the need for an industry association such as OIDA. That association now has 70 members; in addition to the voting members and associate members (including the major lighting companies), university centers have joined as affiliate members. Its mission is to promote optoelectronics worldwide and to advance the competitiveness of its members. OIDA has four functions aimed at building up competitiveness: (1) understanding the markets that pull industrial activity; (2) identifying the technologies that are needed to reach the markets; (3) understanding the infrastructure needed to expedite this technology; and (4) providing a unified voice for the
OCR for page 83
Partnership for Solid-State Lighting: Report of a Workshop optoelectronics industry in communicating with the government and coordinating funding activities. He added that OIDA could not have progressed on solid state lighting without the close collaboration of the Department of Energy. A Totally New Industry He emphasized that the mission of OIDA is not to replace the light bulb; rather, it is to replace conventional lighting with a totally new industry and a new way of understanding lighting. He compared this change to the transitions from vacuum tubes to semiconductors and from cathode ray tubes to flat panel displays used in portable computers. In the “new lighting” there will be two kinds of light sources: inorganic LEDs, which will provide replacements for most incandescent lamps and other point sources of light, and OLEDs, which will replace fluorescent bulbs as area sources. To develop the technologies needed for both efforts, OIDA has performed extensive roadmapping with DOE and has identified milestones that must be reached to penetrate these markets. In inorganic LEDs, the organization has received major support from Sandia National Laboratories, which has been exploring LEDs for many years. Reasons for Optimism Dr. Bergh saw reason for optimism because of considerable efficiency improvements lately on both fronts. These improvements allow OIDA to project a level of lighting performance that far outstrips conventional light sources. He agreed with Dr. Haitz, a pioneer in LED applications at Agilent, that inorganic LEDs would someday reach 200 lumens per watt (lm/W), “although some people think the final figure will be closer to 150.” “People always underestimate new technology,” he added, but warned that the higher target would be achievable only with a partnership that included government support as well as government laboratory and university researchers. “Without this, industry will go to niche markets and will not penetrate any of these domains.” Changing Lighting Altogether A national program, he said, would “change lighting altogether.” One major change is that solid-state lighting for the first time separates the source of energy from the radiation itself. For example, an ultraviolet source and a phosphor can be coupled through space, making the light appear at a location different from that of the source. In addition, solid-state lighting would bring many new applications, environmental benefits, and a new lighting industry. These reasons have already prompted other countries to step ahead of the United States with major government-sponsored activity, especially Korea, Japan, Taiwan, and parts of Europe.
OCR for page 84
Partnership for Solid-State Lighting: Report of a Workshop For example, the Japanese NEDO project, Light for the 21st Century, includes well-organized industry participants who focus on substrate to packaging issues in cooperation with a number of universities. The project is operational and has a target of 100 lm/W lighting. Taiwan, the latest country to move into the field, has launched a major activity to set up a white LED lighting industry; a technology promotion office with a budget of $50 million has been set up under the prime minister to coordinate activities. He reported seeing at a Taiwanese trade show a small house on the display floor that was lit entirely with LEDs. Progress Toward a White-Light Market The LED industry the United States does exist, with monochromatic applications. The most active markets at present include exit signs, stoplights, and brake lights.2 Another segment includes active matrix OLED displays. This can only progress to a white-light market, said Dr. Bergh, with a coalition of industry, government, and academia, guided by clear planning and funding. He advocated a 10-year plan leading to significant penetration of the general indoor and outdoor illumination markets. Such a program could help the inorganic LED industry move first into low-flux white lighting, then into high-demand illumination, and finally into general indoor and outdoor lighting. A parallel path could be drawn for OLEDs, moving from display, to decorative, to low-flux white light, and eventually into general illumination. Industry will not take this path by itself, he said, because it is not profitable for companies to make large technological jumps. Instead, they move one step at a time, diverging in random directions wherever the opportunities arise. National Competitive Needs He recapped some of the suggestions about what the country requires to move into a competitive position vis-à-vis other countries: 2 LED lighting is rapidly becoming the standard in exit sign lighting due to its energy efficiency and long life, but its dimmer light can be a concern in some applications. Exit signs are required by law in all commercial and institutional buildings and must operate continuously. Today, over 100 million exit signs are used throughout the United States and consume 22-35 billion kilowatt-hours of electricity annually. The majority of these signs use incandescent lamps for illumination; so significant savings can be achieved by taking advantage of a more energy-efficient lighting technology. According to the Pacific Northwest National Laboratory, incandescent exit signs cost $42 per year to operate versus $5 per year for LED exit signs. According to the Department of Energy, LED traffic signal lights use 85 percent less electricity than incandescent bulbs and last about 12 times longer. With about 11 million signals controlling 275,000 U.S. intersections, replacing incandescent signals with LEDs would reduce energy usage by 2.7 billion kWh per year and save U.S. taxpayers an estimated $225 million per year. Half of these savings would occur during peak load times.
OCR for page 85
Partnership for Solid-State Lighting: Report of a Workshop R&D in fundamental studies (deposition chemistry, device modeling, light extraction, and an optimum high-brightness capability of 200 lm/W); New manufacturing techniques, including better packaging, and lower costs; and A new lighting infrastructure, including sockets and fixtures, to light offices and whole metropolitan areas. The Cost of a National Initiative Dr. Bergh agreed with Dr. Haitz that reaching these goals would cost about $50 million per year for LEDs. To add OLEDs and to improve the lighting infrastructure would bring this figure to perhaps $80 million per year, for 10 years. Participants would be academia, industry, and government labs, with DOE as the lead agency. This plan also assumed that industry would provide matching funds—some in cash, some in kind. Among the benefits of such a program would be reduced energy consumption, better light, and a strong U.S. position in a major new industry. In closing, Dr. Bergh reminded his audience that solid-state lighting is only part of the larger industry of optoelectronics, which includes optical communication, display, storage, solar cells, and lighting. An important theme is that all these growing industries are fed by the same technologies. Investment in a lighting program helps not only the lighting industry but it also provides indirect help for all the others. From the viewpoint of OIDA, the best investments are those with overlaps, synergies, and economies that benefit the spectrum of this rapidly emerging field. LESSONS FROM EXTREME ULTRAVIOLET AND SEMATECH David Attwood Lawrence Berkeley National Laboratory Dr. Attwood said that the light being discussed at the symposium embraced a broad range of wavelengths: from about 700 nanometers in the red to about 400 nanometers in the violet; computer chips that were made in ultraviolet at wavelengths of 248 nm and 193 nm; and extreme ultraviolet (EUV) lithography with wavelengths down to 13 nm. Topics had also included both refractive optics and reflective optics. The New Extreme Ultraviolet Initiative He described how a new initiative in EUV had emerged. It began with government support for basic research in several agencies, which made major contributions to early enabling technology; at the same time, two Stanford researchers
OCR for page 86
Partnership for Solid-State Lighting: Report of a Workshop developed a multilayer mirror. The lithography system works by shining a 13-nm light onto the mirror, which is made of multiple layers of molybdenum-silicon. The light reflects through a pattern of absorbers, an optical system reduces it by about 4:1, and the reflected light prints the final pattern. An Industry-Driven Process That effort evolved into a partnership consisting of three national labs (DOE, DOD through DARPA, and NIST through the Advanced Technology Program), universities, and some industries. The partnership developed further into a consortium involving primarily Lawrence Livermore, Sandia, and ATT Lucent, with help from the Advanced Technology Program. That consortium took the form of an industry-driven CRADA, funded for 5 years at $50 million a year. The industrial groups include Intel, Motorola, and AMD, later joined by Micron, Infineon, and a few weeks before the symposium by IBM. The national labs included Lawrence Livermore, Lawrence Berkeley, and Sandia. The companies were concentrating on the mask issue, with the rest of the work done by the three national labs. One major effort was to reduce the original quarter-micron technology by which the original Pentium chips were produced, to 70-60-50 nm patterns. These are still five years from production. Competing research programs are reviewed by an industry-led and -dominated process run by SEMATECH for evaluating and comparing various next-generation lithography techniques. For the last six years, the coalition has had an annual meeting and semiannual smaller meetings. Role of the National Labs In the partnership, Sandia is responsible for overall integration, for the EUV source, and for resist recording material development. Lawrence Livermore is responsible for optics, coating, and making the mask blank. Lawrence Berkeley is responsible for EUV metrology and student training. Dr. Attwood said that this partnership was now beginning to work more smoothly as a coordinated activity. Accelerating Some Goals The semiconductor industry regularly updates its roadmap, which describes the anticipated capabilities for feature size, clock frequency (computing speed), etc. For example, some time ago the roadmap called for printing the first Pentium chip at 250 nm and 400 megahertz. Then it mapped anticipated progress to the year 2014, which called for chip frequencies in the gigahertz range, and examined the technologies, such as krypton fluoride and argon fluoride lasers, that might help the industry to get there. The industry is now accelerating some of its goals. In the latest roadmap,
OCR for page 87
Partnership for Solid-State Lighting: Report of a Workshop 2014 has become 2013, and it further contemplates moving from a 3-year cycle to a 2-year cycle, thus potentially moving 2013 goals up to 2007. Suppliers have to promise the delivery of tools by 2007 to meet the new product requirements. The next stage is commercialization. Now that the basic technology is developed and companies have chosen EUV lithography to etch the next generation of chips, the subsequent challenges are how to build the appropriate infrastructure and accelerate the training.
OCR for page 88
Partnership for Solid-State Lighting: Report of a Workshop This page in the original is blank.
Representative terms from entire chapter: