by Michael W. Dorsey
Photos by Patrick O'Connor
What started as a small experiment 15 years
ago has grown into an internationally known university-industry alliance. Today, WPI's Metal Processing
Institute is helping advance both the
frontiers of metal
processing and a worldwide,
$50 billion industry.
The largest industry-university alliance in North America began with an exchange of small talk over dinner one evening in 1984. Diran Apelian, then head of the materials engineering department at Drexel University, was attending a black-tie event to accept an award on behalf of his research team for the best paper published that year in Metallurgical Transactions. Beside him at the head table was the chairman of a major aluminum casting company.
"He said, 'You know, the casting industry, particularly the aluminum casting industry, has no educational home anywhere in America'," Apelian remembers. "'If I want to learn about new developments in aluminum casting, I don't know where to go.'"
The next morning, on a plane back to Philadelphia, Apelian drafted a letter inviting 10 aluminum casting firms to join a new consortium where, for $10,000 a year, they could support research and development and reap the benefits. "To this day, I don't know why I did that," Apelian says. "My laboratory was busy. We were flush with contracts from the Department of Defense, NASA, the NSF and other agencies. We didn't need the money.
"But I guess I'm an opportunist. I knew the industry was going to grow, and that these guys would need a link to academia. I decided to send that letter out and find out what would happen."
Three days later, an envelope was dropped onto Apelian's desk. Inside was a check for $10,000, signed by Ray H. Witt, chairman of CMI International Inc. (since acquired by Hayes-Lemmerz International Inc.). "I knew then we were on to something," Apelian says.
In all, 10 checks arrived, and the Aluminum Casting Research Laboratory was born. In the years since, Apelian's prediction has come true. The aluminum industries have, indeed, grown tremendously, as manufacturers have replaced more and more of the steel in cars, airplanes and other vehicles with aluminum to reduce weight and improve fuel economy, and as many new applications have been found for the lightweight metal.
The rising demand for aluminum products has led to growing competition among metal processing companies and increasing pressure to both improve the quality of finished parts and push the envelope in manufacturing technology. That, in turn, has placed a premium on fundamental and applied research on metal alloys and the various techniques used to fashion them into useful shapes.
Often unable to carry out this research on their own, companies that supply raw metal, form it into finished products, and use those products have turned to universities for help. Most have come to WPI, where the Aluminum Casting Research Laboratory moved in 1990 when Apelian became the University's provost. From its inaugural 10 members, the ACRL has grown to 35 companies. In more recent years, the ACRL paradigm has been used to establish industry-university alliances in other facets of the metal processing industry: powder metallurgy, semisolid metal processing and metal heat treating. Together, they comprise the Metal Processing Institute, which, with almost 140 members, dwarfs any other industry-university alliance in North America.
In The Graduate, the young college grad played by Dustin Hoffman is pulled aside at a cocktail party and given some terse career advice. "Ben--I want to say one word to you--just one word--plastics."
MPI director Diran Apelian.
Diran Apelian has a similar scene in his life story, but for him the single life-changing word was "metallurgy." Born in Cairo, Egypt, of Armenian parents, Apelian came to the United States with his family at the age of 15. Speaking little English, he was enrolled in the 10th grade at Haverford High School in Haverford, Pa. Where others might have struggled, he thrived, becoming president of the student council in his senior year.
"My math teacher was fond of me," Apelian remembers. "I was interested in science and engineering, and one day I asked him what field I should go into. The Space Race had begun and he knew that America would be building a lot of rockets and would need new and better alloys. 'Metallurgy is where it's at,' he said. I didn't really know what that was, so I looked it up. It sounded fascinating."
Apelian enrolled at Drexel University with a major in metallurgical engineering. Taking full advantage of the university's co-op program and Pennsylvania's steel industry, he completed internships at U.S. Steel's Fairless Works, Techalloy and Ford Motor Co. One summer he worked as a National Science Foundation research fellow at the Franklin Institute, where he completed studies on pure beryllium.
"The 'Lehr und Kunst' model that we talk about a lot at WPI was something I really put into practice as an undergraduate," he says. "I was able to integrate the theory I learned at Drexel with practice. I think that had a very formative influence on me."
Apelian published the results of his work on beryllium and additional research in composite materials in Drexel's undergraduate research journal. He also edited Alloy, the newsletter of the university's Materials Engineering Department. With his bachelor's degree in hand, he moved on to MIT, where he earned a Sc.D. in materials science and engineering, working with Professor Merton Flemings, who would become his mentor and close friend. (Today, Flemings heads the Singapore-MIT Alliance.)
Rather than pursue an academic appointment, Apelian went to work in industry after completing his doctorate. He joined Bethlehem Steel's Homer Research Laboratories, where he helped develop the Ultra-Form Series, new high-strength, low-alloy steels that were featured (along with Apelian's photo) in Bethlehem's advertisements. Stronger than other steel alloys, they enabled auto makers to make thinner, lighter bumpers and other metal parts, reducing the weight of cars.
One day, Apelian received a call from the head of the Materials Engineering Department at Drexel asking if he could teach a course on thermodynamics for a professor who was taking a sabbatical leave. "I said, 'My God, that's the course I struggled with the most!'" Apelian remembers. "But I taught it, and I loved it. I then taught an undergraduate course, and that was even more fun."
Eventually, he took a six-month leave from Bethlehem to teach at Drexel. At the end of the term, he was called to the Dean's office. "I have a dilemma," Apelian remembers him saying. "We have an annual faculty teaching award. I pay attention to this award because the students select the winner. You've won this thing hands down, and you're not even a full-time faculty member. How would you like to join us?"
Molten aluminum is poured into molds in one of the facilities of the Aluminum Casting Research Laboratory, the first of the centers that later became the MPI.
"Of course, that went right to my head," Apelian says. "I accepted an offer for about half of my salary at Bethlehem. It was the best thing I ever did."
Apelian excelled as a teacher and a researcher. In his lab, he conducted pioneering work in various areas of solidification processing, including molten metal processing and filtration of metals, aluminum foundry engineering, plasma deposition, and spray casting. He did research in many of the areas that would later come under the umbrella of the Metal Processing Institute, including aluminum casting, powder metallurgy and heat treating. The work received significant support from the government and industry. Over time, it has also resulted in more than 300 published papers and a long list of awards and honors.
In 1983 he won an appointment as head of the Materials Engineering Department. It was while serving in that post that he launched the Aluminum Casting Research Laboratory. He would later go on to serve as associate dean of the College of Engineering and associate vice president for academic affairs and graduate studies. In 1990 he was recruited to become provost at WPI.
Having made a commitment to scale back his own research to concentrate on administering WPI's academic and research programs, Apelian recommended to the ACRL's membership that the center stay at Drexel. But after negotiations with the university failed to produce the level of commitment it felt was needed, the ACRL steering committee voted unanimously to move the center to WPI. Satya Shivkumar, who had worked as a postdoctoral student under Apelian at Drexel, was recruited to head the laboratory.
Apelian served as provost for six years, presiding over a period that saw WPI grapple with a growing financial aid burden and other financial challenges. During his tenure, the University maintained strong enrollments despite a declining college-bound population and declining interest in engineering; expanded its global projects program; strengthened its diversity programs; nearly tripled its research funding; and won recognition for the first time as a national university.
In 1996 Apelian decided it was time for a change. He considered a number of senior academic positions at other universities, but ultimately decided to stay at WPI and devote his time to metal processing research and education. By then, the ACRL was the flagship laboratory of the Metal Processing Institute, which also encompassed the Powder Metallurgy Research Center (PMRC) and the Semisolid Metal Processing Center (SSMC).
The process of extracting metals from their ores and forming them into useful implements is one of mankind's oldest tech- nologies. Casting objects from bronze and copper was a well-established process in the Middle East as early as 3500 B.C. The use of iron to make tools and weapons followed about a thousand years later. Against this historical backdrop, the technology of aluminum processing seems like a Johnny-come-lately. The most abundant metal in the Earth's crust, aluminum was first identified as an element by Sir Humphrey Davy in 1808. It would take another half century to begin to develop solutions to the complex problem of extracting pure metal from the many materials, including bauxite, in which it is found in nature.
In 1885, world production of aluminum was just 15 tons. That figure rose to 8,000 tons by the turn of the century, increasing steadily to reach 680,000 tons just after World War II. The introduction of the aluminum beverage can in the 1960s accelerated the growth of the industry in the postwar years. In more recent times, rising fuel prices and the drive to build more fuel-efficient vehicles has placed a premium on the weight of cars and trucks. Since most of the weight of a typical vehicle is due to its metal frame, shell, axles, engine and drive train, finding ways to reduce the amount of metal in vehicles and replacing the steel traditionally used in cars and trucks with lighter metals has become a high priority for auto makers.
Aluminum, which provides up to a 55 percent weight savings over steel, but is also strong, rigid and easily recyclable, has been the alternative of choice. In fact, the use of aluminum in vehicles has doubled since 1991. The average car and truck produced today contains about 250 pounds of the metal, a figure expected to grow as much as 10 percent per year. Much of the aluminum in today's cars can be found in engine, transmission and drive train components, though the metal is expected to one day replace steel in body panels and structural components, as well. The use of aluminum in vehicles has fueled a dramatic growth in aluminum production worldwide over the past few decades, and catalyzed a proportionate decline in steel production.
The suite of research labs operated
by the ACRL is overseen by Research
Scientist Libo Wang.
It was this strong upward trend in aluminum production, and the implications it held for the future of the aluminum industry, that set the stage for the Aluminum Casting Research Laboratory.
Like all of the centers within the Metal Processing Institute, the ACRL puts most of its resources into precompetitive research--work that explores fundamental issues in aluminum casting of interest to all of the laboratory's members. Ideas for research projects come from the 35 member companies and are reviewed twice a year by a steering committee made up of representatives of member firms, who serve on a rotating basis.
"People in industry bring us their problems. They tell us about the things they cannot do today because of technological barriers," Apelian says. "They may not understand the physics of a particular process. They may not know what to measure, or they may be measuring things that are not important because the sensors that will measure more important things haven't been developed yet. However, in all of these problems, there is a fundamental scientific question that we try to define and subsequently solve. It's a different model for university researchers.
"In the old model, a researcher gets an idea and writes a proposal. Who needs his results? He doesn't know. How will they be used? He doesn't know. How will his research impact society? He doesn't know. Our mission is to educate and to create knowledge, but also to make useful contributions. To do that, we need to know what the problems and issues are."
Funds for the work of the ACRL (and the other MPI centers2) come from the $15,000 annual fee each member pays, as well as in-kind donations of time, manufacturing facilities, and aluminum ingots and castings. The ACRL is full of aluminum ingots and cast parts made of almost every aluminum alloy produced in the world. "I never have to buy aluminum," says
Makhlouf Makhlouf, associate professor of mechanical engineering, who was named director of the ACRL in 1993.
The laboratory works on four projects at a time, each of which becomes the topic of a thesis or dissertation for a master's or Ph.D. candidate. The work is conducted in a suite of research facilities that are supervised by Research Scientist Libo Wang, who has been with the ACRL since its founding. The facilities include a casting lab and a separate analytical laboratory that has equipment for testing the mechanical properties of castings at a range of temperatures, and for testing fatigue strength and impact strength. The MPI also makes use of scanning and transmission electron microscopes, an X-ray diffraction laboratory and other material characterization facilities of WPI's Materials Science and Engineering Program.
One of the projects currently under way in the lab is examining the eutectic properties of aluminum alloys. When molten aluminum solidifies, jagged structures called dendrites, which are nearly pure aluminum, form first. The remaining material--a mixture of aluminum, silicon and other elements--forms in the spaces between the dendrites. "The material in the interdendritic spaces is known as the eutectic structure," Makhlouf says. "We want to know how and where it begins to form and how to control its formation. This will give us another degree of freedom as we design alloys for specific purposes."
A second project is looking at the feeding characteristic of aluminum, which is a way of measuring how quickly and smoothly the molten metal will fill the intricate details of a mold. "Some alloys flow nicely, and others do not," Makhlouf says. "We want to correlate these characteristics with the chemistry of the alloys. For example, will adding iron or magnesium produce better or worse feeding characteristics?
"Very little is known about this correlation, and yet this is fundamental information that companies can put to use immediately. For example, this knowledge may make it possible to make castings with thinner walls. If you have to make thick-walled castings just because you can't find an alloy with the right feeding characteristics, then you are unnecessarily increasing the weight of the part and the cost of your castings."
A third project is attempting to better understand--and identify ways to prevent--die soldering. After many cycles of heating and cooling, aluminum may begin reacting with the steel used to make aluminum casting dies.
To make castings, aluminum is heated in open crucibles where it is subject to contamination. The ACRL's clean metals program is addressing this problem.
If molten metal sticks to the die, parts of the casting or pieces of the die itself can be lost. When this happens, casters must replace or repair the expensive dies. "When soldering begins to occur, you have to continually change pins and liners inside the die, which causes downtime for the casting machines," says Makhlouf, who notes that results of the research to date indicate that changes in alloy chemistry may avert the problem.
The fourth of the current projects is identifying ways to enhance the fatigue properties of aluminum cast components through control of microstructure. Aluminum components undergo cyclic loading (tension and compression) that can lead to fatigue. Graduate students in the ACRL have made significant contributions that will help develop processing technologies that improve the fatigue properties of cast components.
In addition to the projects suggested by the ACRL membership, the laboratory has, over the past eight years, received more than $3.8 million from the U.S. Department of Energy to conduct basic work in aluminum casting. "The DOE is very much interested in reducing the weight of automobiles," Makhlouf says. "It also recognizes that aluminum casting is more energy-efficient and cleaner than steel production."
While ACRL companies do not choose the topics for the DOE projects, Makhlouf says each addresses a problem of general concern to the industry. In addition, through donations of supplies, expertise, and the use of their manufacturing floors, the consortium members collectively help WPI meet its cost share requirement to receive DOE funds. "I estimate that the raw materials we've supplied to WPI for research, whether castings or other materials, are probably worth about $100,000," says Paul S. Kennedy '67, president and CEO of Kennedy Die Castings Inc. in Worcester. "But I'm sure we've gained back at least as much in learning."
In one two-year project, the laboratory conducted an exhaustive study that correlated the chemistry of 24 aluminum alloys, covering the entire range of compositions used in industry, with their microstructure and physical and mechanical properties. The project involved studies of 20,000 cast specimens prepared for the lab by Kennedy Die Castings. The result was the publication of a textbook by the North American Die Casting Association; the book has become an invaluable reference for aluminum casting companies. A similar project, which is relating the casting characteristics of die-casting alloys with their mechanical properties, is nearing completion. That database will be released on a searchable CD-ROM.
To make an aluminum casting, solid metal ingots are heated to about 660 degrees centigrade and melted. Then the molten metal is poured from open crucibles into molds. At various points in this simple process, the metal can become contaminated with foreign gas or solid matter that can significantly detract from its performance and desirability. The problem of melt cleanliness is one of the most important barriers standing in the way of more widespread use of aluminum in automobiles, since porosity caused by gas bubbles and weak points caused by solid inclusions can make aluminum parts more prone to cracking and fatigue.
The ACRL is currently in the fourth year of a five-year project funded by the DOE aimed at better understanding how aluminum becomes contaminated, how contaminants can be detected and measured, how contamination can be avoided, and how dirty metal can be cleaned. The program has already yielded some significant breakthroughs.
For example, work by the ACRL has led to the standardization of a test, known as the reduced pressure test, commonly used in industry to detect hydrogen gas in molten aluminum. "The same company can do this test and come up with a different result each time," Makhlouf says. "Inconsistencies in results makes it impossible for one company to relate its findings to those of another company. We have developed a standard way of conducting the test that will produce consistent results. The American Foundryman's Society is recommending this approach to its members. We hope it will eventually become standard practice."
The problem of measuring solid inclusions has been solved through research in MPI's Nondestructive Evaluation Laboratory, headed by Reinhold Ludwig, professor of electrical and computer engineering. Sergey Makarov, visiting research professor in MPI, Ludwig and Apelian have developed a sensor that uses a combination of a strong electric current in the presence of a magnetic field to create what are known as Lorentz forces. These forces can act on particles as small as 10 microns in diameter and direct them toward optical sensors for recording and analysis.
WPI has licensed the patented process to Heraeus Electro-Nite Co. in Philadelphia, which is developing a commercial version of the sensor it hopes to bring to market in the near future. The sensor, which will be more sensitive and less expensive than other instruments currently used by industry, can be incorporated into a feedback loop that will enable casters to make real-time changes in their process (for example, adjusting the temperature of the melt or introducing chemical additives) to keep inclusions under control.
Ludwig says it may be possible to use the same technology not only to detect inclusions, but to separate them from the molten metal. "If we can apply strong enough
Lorentz forces on these particles, it may be possible to direct them to a specific location and remove them from the melt. There is a lot of industrial interest in this. We are also exploring how we can use the same forces to remotely steer or mix a melt."
As part of the DOE-funded Clean Metals Project, the ACRL is looking at other technologies for removing inclusions from aluminum melts, including activated filters and methods for floating inclusions to the top of the melt. The latter can already be accomplished using a device called a rotary degasser, which injects an inert gas into the melt from a graphite rotor. Rising gas bubbles capture inclusions and float them to the top of the melt. Then, either the inclusions are skimmed off, or clean metal is drawn from the bottom of the melt.
"People have been using this device for a long time, but nobody knows the optimum way of operating it," Makhlouf says. "It's all trial and error. We are attempting to get a better handle on what is really going on."
To address this deficiency, Ph.D. candidate M.D. Maniruzzaman has been developing a computer model of a rotary degasser, a complex program that must simulate the fluid dynamics of a two-phase (liquid and gas) turbulent flow, as well as model the collisions between the gas bubbles and inclusions. "The model is nearly done," Makhlouf says. "Next, we will begin 'what if' games. What if we change the speed of the rotors, the flow rate of the gas, or the design of rotor? How will that affect the success of the process, and can we come up with guidelines for its optimum operation?"
Rotary degassers, filters and sensors would not be needed at all if were possible to prevent inclusions and gas contamination in the first place. Researchers in the ACRL haven't discovered a method for doing that yet, but they have determined that two methods currently used in industry to reduce melt contamination don't work very well. "One important source of inclusions is oxygen in the air, which reacts with aluminum to produce aluminum oxides," Makhlouf says. "Continuous ladling of the aluminum mixes the inclusions into the molten metal. Companies have covered the melt with inert gases or with salt fluxes to keep oxygen out, but we have found that cover gases don't work, and we have not yet found a flux that doesn't have an adverse effect on the metal."
"It is this interdisciplinary approach to problem solving that makes MPI stand out in the field of research," Apelian says. "These examples point out the strength that emerges when different disciplines, in this case electrical, mechanical and materials engineering, are brought together to address fundamental technical and scientific issues."
Though artisans may have been using metal powders to make jewelry and other artifacts even before the advent of metal casting, the modern powder metallurgy industry began in the 1920s, when the technology was first used to produce products like self-lubricating bearings and carbide cutting tools. Rather than molten metal, its raw ingredients are metal powders--mostly iron, steel and copper--which are produced by "atomizing" molten metal with high-pressure air or water. Mixed with lubricants, the powders are poured into molds and compressed under a force of about 50 tons per square inch. The resulting "green" compacts are heated in a furnace at a temperature below the melting point of the metal--a process called sintering--which fuses the powder particles.
The finished parts require virtually no machining and typically cost less to produce than comparable cast parts. The economies of powder metallurgy made the process attractive to the rapidly growing automotive industry during the years following World War II, and auto makers now dominate the customer base for the industry, with about 70 percent of its output going into cars and trucks.
The Powder Metallurgy Research Center, the second component of MPI, was launched in 1991, with Ulf Gummeson as its first director. Gummeson, now executive director of the PMRC, studied mining and metallurgy in his native Sweden and worked in powder metallurgy at Hoeganaes AB and Hoeganaes Corporation, where he rose to the rank of president by the time the firm was sold to American interests in 1968. He served as general manager of Nuclear Metals and president of New England High Carbon Wire Co. before rejoining Hoeganaes in 1978, He was president of the Metal Powder Industries Federation from 1987 to 1991, and received the organization's prestigious Powder Metallurgy Pioneer Award in 1996.
At its inception, the PMRC was given a two-pronged research agenda. One prong would have a technical focus--work aimed at better understanding and improving powder metallurgy itself. The other would have a management focus--work aimed at gaining insight into the workings of the P/M industry.
"This is an industry in transition," says Chickery Kasouf, associate professor of management. "We are seeing more and more consolidation and the emergence of large companies, but it remains a largely fragmented industry with many small firms, few if any of which have a large enough share of the market to affect industry outcomes. They are also stuck between large powder suppliers and large customers."
For the past eight years, Kasouf has been a principal investigator in a series of studies of powder metal parts producers. The series began with a four-round study, directed by Kasouf and former mechanical engineering professor David Zenger, with considerable assistance from Gummeson. "That study identified key management and engineering research issues in the industry," Kasouf says. "The results became the genesis of a proposal to the Alfred P. Sloan Foundation's Industry Studies Program."
With an initial round of support from the foundation, Kasouf, Zenger and Apelian conducted studies that looked at the relationships between parts producers, at the methods they use to estimate costs, at the increasing globalization of the industry, and at how P/M companies obtain and use information about their customers. Jacqueline Isaacs, assistant professor of mechanical, industrial and manufacturing engineering at Northeastern University and an affiliate assistant professor in the MPI, and Kevin G. Celuch, professor of marketing at Illinois State University, joined the research team partway through the first round. The Sloan Foundation has since funded a second round of studies (with Kasouf and Apelian as principal investigators and Isaacs and Celuch as co-investigators) that expand on these topics and also look at how electronic commerce is affecting the way parts producers interact with customers.
Marge Wood, manager of laboratories for the Metal Processing Institute,
provides support for student and faculty researchers.
The research funded by the first round of the Sloan award found that powder metal parts producers face several serious challenges. For instance, their customers are looking to them to provide more value-added engineering services, at the same time demanding even lower prices. In addition, parts producers often use inappropriate models for estimating their costs, which undervalue their own research and development work (for example, the expertise required to convert parts from cast iron or steel to powder metal). The problem becomes magnified over time, as customers expect regular price reductions after a contract is awarded. The globalization of the auto industry--with U.S. companies establishing overseas subsidiaries and foreign firms setting up plants in North America--have caused changes in the marketplace that may require new competitive strategies.
Overlaying these challenges, and hindering their solution, is the reluctance of parts makers to share information with each other or even with their customers, for fear of losing their competitive advantage. "It's a very wary industry," Kasouf says.
But the sharing of information can be a critical element in the development of positive buyer-seller relationships that increase competitive advantage and long-term success, he adds. For this reason, the factors that affect these relationships form a focus area in the second round of the Sloan study. "We are conducting a longitudinal study to find out how firms manage their relationships over time," he says. "We want to know, for example, how they choose accounts, how they maintain a profitable business, and how they manage information flow."
In one study already completed, Kasouf and Celuch looked at the factors that determine how effectively companies gather and use information about their markets. Market intelligence, he notes, is an essential strategy in market-oriented companies, and market orientation is directly related to business performance and employee satisfaction. "We interviewed customers of parts producers and asked them how confident they are in their firm's ability to handle information," he says. "We found that a lot of information flow is driven by individual employees' perceptions of efficacy--in other words, how individual employees view their own ability to use information and how much they value that information. The good thing is that efficacy can be easily increased through good employee management."
Research on the processing of powder metals, bottom, under the direction of Mark Richman, is one focus of the Powder Metallurgy Research Center.
Electronic commerce has opened new avenues for gathering information about customers' needs and business plans, but the powder metallurgy industry has approached this new medium with suspicion, Kasouf says. "The industry is concerned that using the Internet to exchange information with customers will create an auction situation, where customers will expect parts producers to bid on each new part, enhancing the perception of their products as commodities. They would prefer a relationship where customers appreciate the value they add to products.
"The reality is, electronic commerce is inevitable. It's simply going to become a question of how the industry deals with it. In our future studies, we will look at how the industry can use the possibilities--the positives--of electronic commerce to its advantage."
Just as the research Kasouf and his team are conducting is helping the powder metallurgy industry adapt to changing times, it is also preparing students to thrive in the competitive world of industry, where an understanding of globalization, buyer-seller relationships and electronic commerce are vital tools. "It's really important to remember that it's not just PMRC members who benefit from this work," Kasouf says. "Students become immersed in critical industry issues."
Samyukta D. Warty, a graduate student who worked with Kasouf on the portion of the Sloan study focused on buyer-seller relationships, agrees. "My work helped me understand the supply chain management industry. As a result, I now have a job in a supply chain management company that works with buyers and sellers developing client-specific software for the e-business world. A thorough knowledge of clients' expectations, obtained from my research, gave me a big advantage over other candidates for the job. It not only helped me get the job, but excel in it."
Like the other centers within the MPI, the Powder Metallurgy Research Center provides an opportunity for all sectors of the industry (from powder suppliers, to parts producers, to equipment manufacturers, to parts consumers) to support fundamental research that can benefit everyone--work that is often beyond the scope of any single corporate R&D department. Directed by Apelian and Mark Richman, associate professor of mechanical engineering, whose research interests include theories that govern rapid granular flows and models for the compaction of powder metals, the laboratory has conducted a variety of research projects on such topics as the densification of powder metal compacts and the delubrification of compacts during sintering.
Another line of research, conducted by Ludwig and the Nondestructive Evaluation Lab, has attacked a problem that PMRC members early on identified as one of the industry's most pressing needs--the ability to detect cracks before compacts are sintered. "Cracks typically occur early in the compaction process, due to press misalignment, blunt tools, and so on," Ludwig says. "You may create thousands of parts and send them into the furnace for a typically long and costly sintering process, only to find out all are faulty due to one misaligned press. Therefore, the intent of our work has been to detect cracks in the green compact state, before it becomes too expensive to correct them."
The cracks in question are very small--as thin as 20 microns--and can occur on or under the surface of a part. Before Ludwig and his students accepted the challenge, no researchers had attempted to develop a system that would enable companies to detect cracks, trace them back to their source, and correct manufacturing errors--all in real time.
The team experimented with the use of eddy currents, ultrasound and X-rays without success, and then turned to electrostatics. Their solution, the product of four years of research and development, is a unique sensor that consists of hundreds of tiny needles. The needles are pressed against the compact and current flow is initiated. The result is a voltage distribution that uniquely characterizes a good part. Deviations in the voltage pattern indicate the presence of a crack, which is instantly pinpointed by software developed by Ludwig's team.
"We have developed a test system where you take a compact coming out of the compaction press, place it in a mold and push a button," he says. "An arm driven by a stepper motor places the sensor against the part and the current is injected. The resulting voltage is analyzed by a PC, and either a red or a green light flashes, telling the press operator whether or not the part is flawed. The process fits in with the natural rhythm of the assembly line, so it is truly online testing."
With the encouragement of the PMRC membership, the technology has been licensed to a company that will soon deliver a beta version of a shop-floor system for several parts makers to test. When this system becomes available to the P/M industry, it will make an important contribution to its competitiveness, according to Ludwig. "Powder metallurgy is an inexpensive way to produce parts, so it has been attractive to end users," he says. "But at the same time, those end users have been applying pressure on parts producers to turn out flawless parts. Having the means to detect and eliminate cracks early on should help realize that goal."
Somewhere between the solid and molten states of metal alloys lies a class of materials with unusual properties. Called semisolid metals, they flow like liquids but can be handled like solids. Unlike pure metals, which have a single melting point, alloys melt over a range of temperatures. If they are heated to a point within that range and then stirred as they cool to break the long, tree-like dendrites that characterize solid alloys into small, round crystals, the result is a material resembling warm butter.
"A semisolid metal has some important advantages," says Andreas Alexandrou, professor of mechanical engineering and director of the Semisolid Metals Processing Center (SSMC). "Because it behaves like a solid, it can be picked up by a robot and moved around. But because it behaves like a liquid, it can be injected into a die, just like molten metal. As a semisolid material, it won't entrain gases, so porosity does not develop. Plus, because it is cast at lower temperatures than molten metal, it cools faster, there is less wear on the dies, and there is less shrinkage, so cycle times and maintenance are reduced, and dimensional tolerances are tighter. And finally, semisolid castings are superior to liquid castings and can be made thinner."
Not surprisingly, interest in semisolid aluminum casting is high, especially in the automotive industry. But, Alexandrou says, there is little fundamental knowledge about the process, which was developed at MIT just 25 years ago and placed in the public domain only within the past five years. In particular, he notes, commercial use of the process is limited by a lack of understanding of the complex rheology, or flow properties, of semisolid materials. "They don't flow like simple liquids," he notes. "Instead, they exhibit a highly non-Newtonian behavior."
The behavior of semisolid materials is further complicated, as they fall into a class of substances known as structured or thixotropic materials, which have internal microstructures that continually change as the material is processed. "Also," Alexandrou says, "the solid elements of these two-phase materials tend to stick together, creating what is known as yield stress. You have to exceed this stress before the material will flow.
"We are at a critical juncture where the future of this process will be decided," he adds. "Currently, there are only a few companies using this process, and the raw material tends to be expensive. But if technical, physical and economic issues are resolved, it has the potential to become a multibillion-dollar industry."
The SSMC is currently the only university laboratory in the United States conducting research in semisolid metals, and its approach to the problem is garnering the attention of researchers and corporations worldwide. With funding from the Department of Energy, Alexandrou has been developing sophisticated mathematical models of semisolid metals and their behavior when subjected to the stresses during casting.
The models are being verified by comparing the results with in-depth and painstaking experiments conducted at three locations: at WPI, by Apelian and his students; at Oak Ridge National Laboratory, by Srinath Viswanathan; and at MIT, under the direction of Merton C. Flemings, the inventor of the rheocasting process. The SSMC also has research relationships with universities in China, England, France, Germany and Russia, the most extensive of which is with the University of Aachen in Germany, where, in addition to intellectual exchanges, WPI undergraduate students have completed Major Qualifying Projects.
Seeking to become more innovative, the heat treating industry came to WPI to establish a research center.
The ultimate goal is to develop computational tools that can be used to optimize the process and develop new alloys that are ideally suited for semisolid processing. Alexandrou says these computational tools could be used to analyze the way semisolid metal fills complex die shapes and to optimize conditions for high-quality parts. "We are at the forefront in the development of such models," he notes. "We helped create the model that is currently being used by such companies as Aluminum Pechiney, Volkswagen, Fiat, Yamaha and Hitachi."
The modeling of the unusual properties of semisolid metals was Alexandrou's entree into the metal processing field. A mechanical engineer with expertise in fluid dynamics, he was the director of WPI's aerospace research group six years ago when he attended a faculty seminar given by Apelian that addressed the unanswered questions surrounding the use of semisolid metals. The presentation suggested to Alexandrou that the fluid flow models he and his students were developing at the time might hold the answers to some of those questions.
Alexandrou says working with the SSMC's 18 members has been a process of mutual education. As the member companies have come to appreciate the complexities of the semisolid metal process and the importance of mathematical models, Alexandrou and his students have learned about the realities of industry.
"We do work that is practical and theoretical at the same time," he says, "because industry doesn't often view things from a theoretical viewpoint, and universities don't always see things from a practical viewpoint. By talking with each other, we bring the discussion to exactly the right level."
"If you look at the literature from the 1890s up to the 1920s, you see that a lot of the work in metals and materials during that time was on heat treating," Apelian says. "After that, with the development of better microscopes, crystallography, X-rays and so on, it was possible to see inside the metal crystals, and the focus of the field changed. Today, the heat treating industry is realizing that it has not been as innovative as it needs to be. Just as the casting industry did 15 years ago, it went looking for an academic home."
After an extensive search of university research laboratories across the country, the heat treating industry turned to WPI. Having developed a new vision for the industry that placed a heavy emphasis on research and development, the American Society for Metals (with over 50,000 members) and the Metal Treating Institute (whose president is L. Thomas Benoit Jr. '66, president of Flame Treating and Engineering Co. in West Hartford, Conn.), approached WPI about establishing a center dedicated to research and education in heat treating. The Center for Heat Treating Excellence was inaugurated with a ribbon-cutting ceremony and symposium on Sept. 1, 1999. "With more than 60 member companies, this is the largest center in the MPI," Apelian says.
Initially, according to Apelian, the research projects for the new center will be conducted not just at WPI, but at other university and industry sites worldwide. Four major projects were launched in early 2000. The first, which focuses on the effects of microstructure and service properties on solution heat treatment of aluminum alloys, is being undertaken at the University of Connecticut. Another project, conducted by Yiming (Kevin) Rong, associate professor of mechanical engineering at WPI, is developing an analytical tool for part-load design and temperature control in loaded furnaces. The third is a major study of quenching during thermal processing; it is directed by Richard Sisson Jr., professor of mechanical engineering and interim ME department head at WPI. The fourth project, to be completed at Illinois Institute of Technology, is investigating the prediction of distortion and residual distress in heat-treated components.
As with the other centers in the MPI, the marriage between WPI and the heat treating industry was built on a fundamental principle of supply and demand. "Universities have two products," Apelian says, "the students we educate, and the knowledge that we create and disseminate. Companies need two things: innovation, and people who can implement the innovation. When you put those together, it's a no-brainer."
Apelian says member companies receive the innovation they seek from the MPI's research projects, the results of which are communicated to the members of each center during two annual workshops on campus (which also provide a forum for member companies to discuss issues facing their respective industries); through technical seminars that address key problems each industry is encountering; and through technical education programs for interested members, including customized continuing education programs offered onsite.
Paul Kennedy says he has never had any doubts about the value of his company's membership in MPI. "Some of the best research in casting anywhere in the world is happening at WPI," he says. "Our membership gives us immediate access to that research, and enables us to drive it, as well. We've developed several new alloys as a result of research in the ACRL, and I know we would not have been able to do that had we not played a role in shaping that work. A lot of people think of casting as 'old technology,' but it isn't. It's really valuable for our people to see that and to get involved in the exciting things happening in the field. It helps them get out of the box."
As for the people side of the equation, Apelian says the activities of MPI produce a steady supply of men and women who are not only well-versed in the theory of metal processing, but who are familiar with the practical realities of putting that theory into practice in the commercial world--often through firsthand experience. "The real and true beneficiaries of everything we do are the students," he says. "Our faculty members, through their research, become much more aware of the issues facing the industrial sector. They integrate that understanding into their classroom instruction, so theory and practice comes forth in an integrated way."
Through Major Qualifying Projects and graduate research projects, students get a true taste of real-world engineering. "The undergraduates who complete MQPs through MPI typically complete a paid internship in industry the summer before they work on their project," he says. "They learn about industrial processes and they experience the corporate culture. When they come back to campus, they have a project topic and are ready to run with it. It's 'Lehr und Kunst' at its best."
At the graduate level, master's and Ph.D. candidates can get the same kind of firsthand experience by completing industrial internships. Students in this program spend half of their time in industry completing their thesis or dissertation work. "To use a business term, we 'outsource' the research," Apelian says. "There are many processes and types of equipment that we don't have at WPI, and in reality we don't want to have them here on campus because the technology changes every few years. Through this program, the facilities of industry become our laboratories."
Every year, four to six students take part in the program, each supported by about $40,000 per year from sponsoring companies. Another 20 students are supported through the fees member companies pay or through research grants from government and industry organizations. All benefit from the close contact they maintain with the members of the MPI's laboratories, who follow closely the progress of their research, and who prove attentive--and critical--audiences when they present their results.
"Each of our projects has a focus group of member companies," notes ACRL director Makhlouf. "They are in constant touch with the students. They get to know them well, and the students become quite familiar with the companies."
Three weeks before the biannual workshops, progress reports written by the students go out to member companies. "They read this stuff," Apelian says. "They bring the reports back to campus with all kinds of red marks. Then the students stand before this group of representatives from these leading companies (often including the vice presidents for research) and they make their presentations. I can't imagine a better experience. Knowing that what you are doing is important to somebody is an incredible incentive. We have a built-in motivation engine in the system."
Not surprisingly, students who've gone through this experience make especially attractive job candidates. "All of my students are hired almost the moment they finish," Makhlouf says. "And about 80 percent are hired by ACRL companies."
With the incorporation of the Center for Heat Treating Excellence and the growing maturity and international recognition of the other components of the MPI, Apelian has been taking time to think about how far things have come since he seized that opportunity more than 15 years ago. "We've built a strong organization," he says. "We've made some real advances in research and some fundamental contributions to near-net-shape manufacturing. We've prepared many talented men and women to make their own contributions to the field. And through all of this, we've gained incredible visibility for the MPI and for WPI."
But Apelian says the MPI's most rewarding product may well be the change it has brought about in the industries it serves. "When a center like this first starts, there is a certain amount of paranoia," he says. "Companies are naturally reluctant to share information and engage in cooperative research. But over time, as they have seen how everyone benefits from the work we do here--how, indeed, a rising tide floats all ships--the MPI has become a true learning organization, where companies learn from us and from each other. That has been incredibly rewarding."
Last modified: Jun 13, 2000, 13:31 EDT