Materials in the New Millennium: Responding to Society's Needs (2001)

After Senator Domenici’s keynote address, the forum moved into a series of talks designed to set the scene for the following two days. First, Thomas W.Eagar of the Massachusetts Institute of Technology (MIT) addressed the most fundamental question of all: Why are materials important to society? Dr. Eagar is the POSCO Professor of Materials Engineering at MIT, where he heads the Department of Materials Science and Engineering. He was elected to the National Academy of Engineering in 1997 for his contributions to the theory and practice of welding. Dr. Eagar is a member of the National Materials Advisory Board.

Arden L.Bement, Jr., of Purdue University examined the timeline of materials development and factors that affect successful progress from the laboratory to the marketplace. Dr. Bement is one of the Distinguished Professors of Engineering and head of the School of Nuclear Engineering at the University. He has also served as vice president of science and technology at TRW and deputy undersecretary of defense for research and engineering. Dr. Bement’s research interests are primarily in electroceramics. He is a member of the National Academy of Engineering.

James C.Williams of Ohio State University presented an overview of past successes and future opportunities for structural materials. Dr. Williams is the Honda Professor of Materials at the University. He was previously general manager of the Materials and Process Engineering Department at GE Aircraft Engines. His interests include the structure-property relationships of high-strength materials, the performance of materials in extreme environments, materials processing, government policy as it affects materials, and the management of high-technology organizations. He is a member of the National Academy of Engineering and a former chair of the National Materials Advisory Board.

James W.Mitchell of Lucent Technologies presented an overview of past successes and future opportunities for functional materials. Dr. Mitchell is director of the Materials Processing Research Laboratory at Bell Laboratories in Murray Hill, New Jersey. His research interests include the use of microwave discharges and plasmas for materials synthesis, the development of methods for ultrapurifying reagents and research chemicals, and the development of environmentally responsive methods for producing electronic chemicals. Dr. Mitchell is a member of the National Academy of Engineering.

Thomas A.Weber of the National Science Foundation addressed the government’s role. Dr. Weber has held a variety of positions at the Foundation, where he is currently director of the Division of Materials Research. His research interests are in computational chemistry and materials, particularly the use of computer simulation to study air pollution,

polymers, glasses, liquids, metals, and semiconductors. In 1994, Dr. Weber was awarded the Meritorious Executive Presidential Rank Award.

The session ended with a panel discussion. The following are summaries prepared by the editors who adapted them from the remarks made by the individual presenters.

The ages of mankind are labeled by their dominant materials: the Stone Age, the Bronze Age, the Iron Age, and now the Silicon Age. Why? Because materials define the sophistication and wealth of nations. Materials create wealth, improve our standard of living, and are key to meeting society’s needs, from national security and communications to health and housing.

Traditional materials are primarily structural—steel, aluminum, concrete, polyethylene, glass. Although they are familiar, their properties have been improved dramatically in the past two decades, while their costs have fallen and their production has become more environmentally sustainable. Novel materials are primarily functional, with optical, electronic, magnetic, or biological properties or unique environmental properties, such as resistance to high temperatures.

National leadership in manufacturing and trade requires leadership in materials. Materials are essential throughout industry, in fields such as telecommunications, health care, energy, and automobiles. In these and many other areas, materials with high added value and high functionality strengthen existing industries and preserve the jobs they provide, while creating new products that enable the development of new industries and new jobs.

The future promises both the creation of novel materials and the greater availability and affordability of traditional materials. The sustainability and quality of life of an increasing proportion of the world’s population will depend on whether these promises are achieved. The United States should be the leader in developing and applying these materials for the benefit of the entire world.

Implementing new materials is a slow process. Bringing a new consumer product from invention to widespread adoption typically takes 2 to 5 years, but doing the same for a new material may take 15 to 20 years. The problem is especially challenging for advanced materials, which are inherently distant from the market, do not yet have established proofs of principle, and thus involve greater market uncertainty and business risk. As a result, advanced materials seldom attract venture capital. According to David Morgenthaler, founder of Morgenthaler Ventures, a venture capitalist should very often invest to improve or update a product, broaden a product line, or apply a

product to a new application, often invest in using technology to develop a new product, but only rarely invest in developing an enabling technology.

Shifting commercial and military priorities have made long lead times an even greater problem. One can no longer just provide a superior material and assume someone will use it. A 20-year development time is beyond the planning horizon of most businesses. Twenty years is too long for technology push to bring a new material to market, and it leaves too much time for other alternatives to emerge.

Numerous other factors also inhibit the commercialization of new materials. Market demand for new materials is often overestimated. Their initial cost is usually high, even if overall life-cycle costs are low, and product liability insurance is expensive. Designing with new materials can be complex, and standards and statistical databases are usually absent. Manufacturers that adopt a new material may need new manufacturing tools and new inspection methods. Secondary processing and recycling are often difficult. Environmental and arms export regulations may present barriers. Government testing, evaluation, inspection, and approval may cause delays. Intellectual property issues must be addressed. All these factors magnify the problem of long lead times.

The problem can be boiled down to two dilemmas. First, the market will not accept a new material until its cost falls, but its cost will not fall until the market accepts it. Second, designers will not select a new material until it is evaluated in service, but a new material cannot be evaluated in service until a designer selects it.

Resolving these dilemmas requires cutting development time at least in half. Doing this will require a combination of approaches, involving technology, management, and government. Technology can contribute by anticipating application problems earlier in the R&D process, developing material models that are linked better to design objectives, and developing more efficient methods for variation reduction and probabilistic verification of product and process designs. Management approaches include encouraging collaboration between materials developers and users, putting earlier emphasis on cost reduction, focusing on demonstrations of specific components, and seeking patient capital to enable infrastructure development. Government can contribute by encouraging partnerships for infrastructure development and early technology insertion, providing patient capital at the proof-of-principle stage, and more closely assessing the economic risks of regulations, policies, and practices that could slow the development cycle.

Traditionally, structural materials have enabled new products or improved performance through lower weight, higher strength, higher temperature tolerance, longer life, or unique combinations of structural and functional capabilities. Today, structural materials are also required to make products

more affordable. They must be cheaper to produce, enhance component manufacturing by being net-shape capable, and lower the cost of ownership.

Materials enabled most of the great engineering achievements of the twentieth century. Without materials improvements, seven of the top ten accomplishments would have been impossible: the Apollo moon landing, the airplane, the transistor, the Manhattan Project, the integrated circuit, the airplane jet engine, and the communications satellite. The other three— digital computers, television, and the Panama Canal—all benefited from better materials, too, even if they did not absolutely require them.¹

Structural materials are still essential for competitive products, but today new materials must be mature at introduction. Their technical capabilities must be clearly understood. Their failure modes must be thoroughly characterized. Their cost and supplier base must be known and stable. They must add value that outweighs any increased cost. And the risk of introducing them, both real and perceived, must be low. As a result, incrementally improved materials with a history of prior use are more readily accepted than completely new materials.

This new demand for maturity at introduction requires changes in traditional materials development practices. Past practice was to introduce a new material, modify it somewhat during a prototype phase, and finalize it just prior to production. New commercial products today have a faster development cycle and usually less risk tolerance. For example, jet engines are now developed in 30 months rather than 60, largely because of improved computational design tools. Improved methods for materials development have not kept pace with this compression of the product cycle, however. The development of new materials for jet engines typically still takes at least 60 months. Materials scientists and engineers must address this mismatch in timing and learn to develop new materials in half the time and at half the cost.

Improved computational tools are the most promising approach and must become the grand challenge for materials research. This multifaceted problem includes:

• modeling the composition of matter to guide changes in alloy chemistry

• modeling processes to accelerate convergence of the process window

• modeling properties to reduce testing time and the cost of property data

• building error analysis into all models

Models that permit high-confidence interpolation would be an excellent start. Bottom-up prediction or high-confidence extrapolation will come later. The infrastructure to start this effort exists now, but meeting the challenge will take substantial time and resources, comparable in magnitude to those required for the human genome project.

The world is in the midst of a communications explosion fueled by more than 100 years of invention and innovation. Communications services are now so easily accessible that it is easy to forget how sophisticated the underlying hardware is. Functional materials are essential to that hardware. In fact, communications is a perfect example of a high-technology industry enabled by advanced functional materials and their processing.

Challenges. Advances in functional materials will be key in many future applications. Some examples:

• nanotechnology and microelectromechanical systems (MEMS) for ultrahigh-density interconnects

• membranes and other materials for small, low-cost fuel cells in portable electronics

• organic circuits for “electronic paper”

Opportunities. Some technologies, if they became available, would find immediate use in applications that already exist today. Some examples:

• mechanistic control of materials chemistry at high temperatures

• photodefinable planar optical materials

• low-cost nonmetal substrates with high thermal conductivity

• photonic bandgap materials

• materials with tunable properties

Dreams. It is easy to dream about the future of communications. Here are some dreams that would have important implications for functional materials:

• wireless Internet connectivity anywhere, anytime

• people communicating directly with computers

• systems on a chip

The National Science Foundation, like other government agencies involved in science and technology, must maintain a balance among competing demands. It must address both science and engineering, it must support both research and education, and it must balance the needs of people, instrumentation, and facilities. In doing all this it strives to meet three goals:

• a diverse, internationally competitive workforce of scientists, engineers, and citizens

• discovery in the service of society

• state-of-the-art, broadly accessible databases and shared tools for research and education

Working toward these goals, as an agent for the government’s investments in research and education, means seeking a broad portfolio of

integrated investments, providing opportunities across the spectrum of science and engineering, and ensuring excellence through competitive merit review.

One cannot predict the future, but three research areas—the terascale, the nanoscale, and complexity—will be key starting points for the coming decades. As computers become capable of processing terabits of data at teraflop speeds, terascale computational science will provide new insights in areas from climate prediction to economics. Materials science is one of the areas with the most to gain from these new capabilities.

Ideas such as molecular wiring and quantum corrals have already made clear the potential impact of the nanoscale on materials science. The president’s recently announced nanotechnology initiative focuses on five areas: biosystems ($20 million), structures and quantum control ($45 million), device and system architectures ($27 million), environmental processes ($15 million), and modeling and simulation ($15 million). Materials are a key component of all five.

The idea of complexity reflects the wide range of scales in our physical environment, from the molecular to the global. Complexity is becoming a guiding principle in materials science, as it is in many other fields.

Education will be as important as research to our future success. The National Science Foundation seeks both to provide the twenty-first-century workforce with knowledge, people, and tools and to enable citizens of all ages to become literate in science, mathematics, and technology. Its approaches to meeting these goals include research on the science of learning, better education of teachers, and increasing diversity by widening access and opportunities.

Reflecting proudly on its first 50 years and forward to its next 50, the NSF seeks above all to follow the lead of Wayne Gretzky, who said, “I skate to where the puck is going, not to where it’s been.”

Question: It has been suggested that Moore’s law will run out in about 2010. Photonics may also hit the wall about then. If it takes 20 years to introduce new materials, is the planned level of investment enough to meet future challenges?

Weber: I’m glad to get this level of investment! Unfortunately, many people assume that materials will “just be there.” This indicates a lack of understanding. The nanotechnology initiative is a good start. An important point is that this initiative comes from the White House, which sees it as an economic issue.

Q: We have heard discussions of structural materials and how they differ from electronic materials, for which the value added is higher. Are we focusing too much on materials instead of systems? How do we encourage entrepreneurial behavior in large companies or divisions and make it successful? The role of a top-level champion is crucial in technology

implementation, but it is difficult to find in U.S. industry. If fuel economy and emissions requirements get much tougher in the next 10 to 15 years, for example, then the emphasis will be on alternate energy systems.

Weber: The issue requires a better understanding of the value to the customer of the new product or improvement, as well as the size of the market that may be affected. We are stressing more joint ventures at the precompetitive stage.

Q: Can you provide an example of the new paradigm for getting new materials implemented and into production more quickly? Williams: An example that gets part way there could be the use of forging-casting models. Both GE and Pratt & Whitney require their suppliers to have some level of process modeling and variation control to ensure product consistency.

Mitchell: Another approach may be to use combinatorial methods to screen a wide range of possible material compositions and then to select just a few for further evaluation and optimization.

Eagar: In lithium-ion batteries, the conventional thinking was that performance was controlled by the cation lattice. Computer simulations showed the anion lattice to be the controlling aspect. The theory and model predicted better performance from aluminum-lithium-cobalt systems, and this was verified by tests. This may be the first time a first-principles model preceded the actual development of a material.

Bement: Investment casting may be a good example, too. Defense Advanced Research Projects Agency (DARPA) programs have promoted best practices in industry, including greatly advanced capabilities in mold design and process control.

Q: Materials with magnetic properties have seen a resurgence of interest recently after a dearth of support earlier. Why this renewed interest?

Eagar: Sales of magnetic materials are actually greater than sales of electronic materials, although they are usually less visible. Most sales go to motors, transformers, and so on, for power systems. More and more businesses are investing, so market size is the driver.

Q: We have heard where nanotechnology is going. Are we ahead of or behind the rest of the world in this area?

Weber: A recent study indicates that our level of investment is about the same as that in Western Europe and Japan. As far as getting advances to market, the United States is doing better than Western Europe but worse than Japan.

Q: There may be a different view of the materials dilemma described in Arden Bement’s talk. An alternative approach is to increase risk tolerance by using smaller companies, which, if successful, are then assimilated into larger companies. Universities feed their ideas and students into these smaller companies, thereby linking fundamental research with product development. What is government’s role in promoting this model? What practices should the government adopt?

Bement: This is a food chain model. Current models favor more teaming with big companies—the Advanced Technology Program, for example. We

have a different paradigm for advanced materials to do better than the food chain model.

Mitchell: In very competitive technologies, if you do not incorporate new ideas into products and go to market in 18 months, you miss the window of opportunity. You have to bring the business unit and the market along at the same time as you do the development work. Successful companies are able to keep their focus on pioneering research and get their results into products.

Williams: The window of opportunity depends on when you decide to start the clock, of course. Regarding small companies, all federal agencies have Small Business Innovation Research (SBIR) set-asides. If an innovation requires a lot of capital, small companies have a hard time handling the financing, and venture capitalists often do not like those kinds of ventures.

Eagar: The reduction in industrial investment in materials has had an important effect on universities. Since there has been no pull for Ph.D.’s in corporate research labs, the ratio of master’s degrees to doctorates has been increasing. Students who might previously have pursued a Ph.D. now get an M.S. degree in engineering and then an M.B.A. Then they go to work in the investment arena, where they make more money and have more responsibility.

Q: Displacing old materials with new ones has been a challenge. Is there a conundrum to nanotechnology, namely that the large sunk investment in silicon plants will make companies resist change? Or will a new set of companies arise?

Eagar: Companies often have a difficult time changing. I suggest you read The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail, by Clayton M.Christensen.

Weber: Some companies do not change, and those companies go out of business. But we are always looking to the future. Competitive pressures are often the impetus for transforming old technologies and extending their life. An example of this is the continuing success of silicon versus gallium arsenide, despite predictions of silicon’s demise.

Q: At Lehigh, we teach students about the fiber-composite fan blades that almost caused the demise of Rolls Royce and about the airframe companies’ aluminum-lithium alloy development program. How do you approach these episodes in your models?

Williams: One has to understand the failure modes of materials and have a high enough level of confidence to act based on that understanding. Regarding the specific examples you mention, the GE90 engine has successfully used a fiber composite for blades. For aluminum-lithium alloy development, there was probably overoptimism about the how easily the technical challenges could be overcome.

Q: The time-to-market situation varies widely between fields. How does NSF’s investment strategy reflect this variation? What is the government plan for infrastructure areas and for slower-moving industry segments? Weber: This challenge is one of many that NSF must address. Our intent is certainly to keep programs going in metals, ceramics, and so on. When a particular material is challenged, threatened with displacement, it tends to

improve, so we must address a wide range of processing issues, availability, and so on, as well as new materials, such as polymer composites. NSF tries to keep a balanced portfolio based on the challenges it sees. Today nanotechnology is one such challenge.

Q: What does the panel think of the chemistry content in the education of materials scientists and engineers?

Bement: Purdue’s departments are big enough to have their own materials groups. They have woven together a consortium across the campus to provide chemistry information in several areas, such as synthesis research models. This seems to be working effectively.

Q: Jim Williams mentioned a threefold improvement in the thrust-to-weight ratio of jet engines, but the slope of the improvement graph was dropping to near zero. Why?

Williams: Some of the slowing may be because we are still using the same classes of materials. Efforts to improve these materials further have not been very successful, so perhaps we need new materials. Also, the emphasis has shifted from improved performance to lower cost with equal performance.

Bement: We often overlook the business implications of risk and uncertainty. We can offer a real technology advantage, but when all the costs are summed up, it may still be a nonstarter.

Q: We are confronted by a workforce shortage in electronic materials. How can NSF help with this as we try to wire up the country?

Weber: A shortage of people is a problem, but we will also find smarter ways to solve the problem and, thereby, probably reduce the need for people. In the longer term, NSF is concerned about the intellectual base of the country. In the past, we have relied on immigration, but that is changing, and we have to figure out how to meet our needs ourselves.

Q: We see more integration of teams in photonics, for example, than in structural materials. Is this a result of our education system?

Bement: Design processes are different for electronics than for structural systems.

Q: Ceramics are a mature technology. Metals are a mature technology. Silicon has been around for nearly 50 years. What if silicon stops being the driving force?

Eagar: Moore’s law is going to fail, and the law of diminishing returns will eventually apply. Biomaterials is an exciting area. Photonics is an exciting area. Nanotechnology is a surrogate for these areas. But we can’t ignore the older materials and will need a balanced view.

Mitchell: Materials are not exciting by themselves. It’s what you do with them that’s exciting. Silicon became a unique material because of what we can do with it. So I have no specific answer regarding what the new materials will be, but new societal needs will create needs for new materials.

Bement: There are limitations in nature, but solutions may be different. For example, we may get around apparent limitations by using networks rather than single devices. We have to be able to understand the limitations early enough to pioneer the correct path.