Fuel cells are typically made from three materials that have to withstand heat ranging from room temperature to 800° Celsius. But because the materials expand at different rates when heated, degradation and cracking can occur where they meet. To help solve this problem, Denis Cormier, a professor of machining and manufacturing at the Rochester Institute of Technology, is developing fuel cells that are produced by 3-D printers, microlayer by microlayer. Instead of having potentially weak seams binding them together, the materials blend into one another. “You gradually transition the materials,” Cormier explains. It’s an intricate weaving process that can’t be done by conventional manufacturing technologies.
Such delicate fabrication is among the breakthroughs that enthusiasts hail as the first stirring of an American industrial revival. So-called advanced manufacturing brings new and emerging technologies — 3-D printing, or additive manufacturing, as well as robotics, telematics, and nanotechnology – to making what we now use or may invent, and in ways that can adapt to change. It could “offer the potential to produce higher quality and a wider variety of products — even customizing products for just a few or even a single buyer — and do so at low cost,” former U.S. Commerce Secretary Gary Locke told Congress last March. Heralding “a renaissance in American manufacturing,” President Obama has enlisted research-and-development talents from six leading universities and a half-dozen major companies in a $500 million partnership. The results could mean bright prospects for engineers in a variety of disciplines, even if they don’t generate huge numbers of jobs.
If, as Locke contended, manufacturing remains vitally important to U.S. national and economic security, it is in dire need of a reboot. The world’s largest manufacturer, the United States produces 19 percent of global output. But the sector, which employs 64 percent of American engineers, represents just 13 percent of GDP, compared with 77 percent for services. In 1979, it employed 19.5 million Americans; last year, that figure had shrunk to 11.5 million, just 6 percent of the total workforce. A generation after jobs and production began drifting offshore, much of what’s left of U.S. manufacturing is low-tech, “making it much more vulnerable to low-wage competition” from overseas, notes Rob Atkinson, president of the Information Technology and Innovation Foundation (ITIF). In high-tech goods, the United States has gone from a trade surplus a decade ago to a deficit of $81 billion last year.
Clearly, jobs are on the line, but so is the edge in innovation that underpins U.S. economic strength. That’s because America’s share of global research and development is shrinking along with its traditional manufacturing base. As governments elsewhere, particularly in Asia, expand science and technology training, multinational companies increasingly find hospitable locales for R&D outside the United States, taking advantage of local talent, proximity to markets, and sometimes lower costs. Harvard Business School professors Gary Pisano and Willy Shih, writing in 2009, contended that without a strong manufacturing base, American innovations will be few and far between. There are, the pair wrote, “relatively few high-tech industries where the manufacturing process is not a factor in developing new – especially, radically new – products . . . An economy that lacks an infrastructure for advanced process engineering and manufacturing will lose its ability to innovate.” The President’s Council of Advisors on Science and Technology (PCAST) noted that “proximity still encourages people to exchange the knowledge most critical to innovation.” Many high-tech products invented in the United States — including “fabless” chips, LCD monitors, and PCs — no longer can be made here because of a loss of knowledge and skills. When products such as smartphones serve as battlefield gear, loss of this capacity has a national security dimension. “We may be denied these things in the future,” says Gary Fedder, the professor of electrical and computer engineering who heads the Robotics Institute at Carnegie Mellon University (CMU).
What’s needed are new, improved ways to make things we all know and cutting-edge methods to create products that don’t yet exist – techniques that let the United States produce more of what is invented and designed here. That’s the goal of advanced manufacturing, or what some call high-value-added production.
Of the new methods, additive manufacturing is perhaps the most radical. It is the complete opposite of the subtractive methods of manufacturing – the milling, chipping, cutting away at a chunk of material to create a product or part – that underpin most mass production. Specialized 3-D printers come in different varieties and use a range of techniques. But they all digitally scan a 3-D CAD design, and virtually divide it into ultra-thin slices. Then they essentially print out and layer each microslice atop another using powder or molten liquids as “inks” -- which can encompass metals, resins, composites, and plastics -- and fuse them together, perhaps with heat or lasers. “It is closer to how nature does it,” says Radovan Kovacevic, head of the Research Center for Advanced Manufacturing at Southern Methodist University in Texas.
The technology has been around for decades and was used mainly to make “rapid prototypes” of parts. But now it’s increasingly used to make the parts themselves. It’s estimated that 20 percent of current 3-D printer output is for actual components. “We’re just seeing the tip of the iceberg, and it will be enormous,” says Terry Wohlers, a Colorado consultant and expert in the technology.
The advantages of 3-D printing are many. It can be inexpensive, since there are no dies, molds, and tooling to make. There is much less waste since it uses no more material than necessary to make the part — another cost-savings. And it can allow for more design freedom. “Geometrically, you can make stuff you couldn’t make before,” says Richard Hague, a professor of innovative manufacturing at Britain’s Loughborough University. For instance, researchers are developing printers to make the first nonflat circuit boards, which can be used to accommodate curvier product designs. They’ll soon make it easier to embed sensors and other electronics within materials, giving a big boost to efforts to create smart or self-healing materials. And additive manufacturing allows for customization; absent the cost of dies and molds, it will become economically feasible to make and profitably sell truly individualized products.
For now, it’s a technology best suited to low-volume, high-quality production. Think aerospace components, super-luxury automobiles, furniture, light fixtures, and medical devices and equipment. Boeing, for example, uses it to make an air vent that requires just two pieces; the old, subtractive way of making the vent required the machining of nearly 20.
Still, “it will not overtake mass production,” Kovacevic says. Everything from computers to cars to crayons will rely on traditional manufacturing for years to come. Though additive manufacturing is sometimes called rapid prototyping, “there’s nothing rapid about it,” Cormier says. It only speeds development time because there’s no need to make dies and molds. Mass production machines can spit out parts in seconds; 3-D printers need a minimum of many, many minutes, and usually several hours. Cormier estimates it will be at least a decade before we see 3-D printers fast enough to make parts by the millions. Part of Hague’s research at Loughborough is developing ways to speed up the process. For now, if you need a million plastic trash cans, stick with injection-molding. But if you need a thousand or fewer, “it’s ridiculous to make a mold,” says Cormier.
Additive manufacturing “is an exciting idea, but we have to be careful not to overpromise,”cautions Olivier de Weck, an associate professor of aeronautics and engineering systems at MIT. Still, over the next decade, as researchers find ways to accelerate the speed of 3-D printers, an increasing number of components needed for mass production will be printed. That, of course, eventually could mean some manufacturing comes back home from low-wage countries. If you can fabricate the parts you need quickly and cheaply where you need them, why ship stuff in from overseas?
It’s also important to make manufacturing more flexible to better keep pace with ever changing consumer demands. That’s the focus of the University of Michigan’s Engineering Research Center for Reconfigurable Manufacturing Systems, which is developing technologies that range from wireless networks to virtual simulations. The goal: to help plant managers quickly reduce — or ramp up — production to meet demand fluctuations, or speedily retool to change products. “Being able to deliver the right product at the right time that the market demands is difficult to do in many industries,” particularly the automotive, says A. Galip Ulsoy, a manufacturing professor and the center’s deputy director. It takes years to design and build a plant, and most are built to produce a specific product. “But it’s easy to get out of sync with the market.” Gasoline prices soar, and buyers suddenly want more fuel-efficient cars. The center’s research aims to help, say, an engine plant deftly and quickly segue from V-8s to V-6s.
At CMU’s Robotics Institute, research professor Ralph Hollis is designing hardware and software to speed the design and production of high-precision, electromechanical products using fully automated assembly systems. That’s a trend largely pushed by demand for more customized goods and the short life cycles of high-tech products. MIT’s de Weck says manufacturers also are gaining agility by moving away from large-scale, big-batch factories to small-scale, less capital-intensive facilities.
Fixtures in modern factories, robots are very good at doing the same chore over and over again, ad infinitum. But what’s largely absent from manufacturing are autonomous or semiautonomous robots that rely on sensors and can make decisions, operate in less-structured environments, and not necessarily do the same tasks endlessly. These kinds of robots already are used for nonmanufacturing duties, such as scoping out the innards of oil wells or disarming roadside bombs in war zones. But they are coming to manufacturing, too, Ulsoy says. One early use of thinking robots: autonomously guided vehicles for parts storage and retrieval, replacing inflexible conveyor belts. A new challenge, currently a focus of the multiagency National Robotics Initiative, is creating robots that work in symbiotic relationships with people. “That still requires a lot of research,” CMU’s Fedder says. “That’s the reason you don’t see robots and humans interacting -- it’s dangerous. One swing of a (robotic) arm could kill you.”
New man-made materials with wholly unique properties, especially those derived from nanotechnology, also loom large among advanced-manufacturing technologies. “Innovative manufacturing is going smaller and smaller,” says de Weck, who is also executive director of MIT’s new Production in the Innovation Economy project. Two promising new materials under development in MIT labs: a fiber composite material that’s conductive, and so can carry electronic signals that could, say, rid planes of wire bundles and also protect aircraft from lightning; and a plastic coating infused with microsensors that could be used on airplane wings, slightly altering its texture to accommodate changes in airflows, or on bridges. According to PCAST, “Materials such as graphene, buckyballs, and carbon nanotubes that have nano-scale crystalline structures could serve markets for data storage, energy, optoelectronics, avionics, defense, and packaging. Potential products include highly attuned chemical and biological sensors, fuel cells, touch screens, lightweight body armor, and airframes.” There is also potential for “entire new classes of pharmaceuticals based upon nanostructures for broad classes of disease such as cardiovascular disease and cancer.”
For engineering students of all stripes, the hoped-for boom in advanced manufacturing should provide great job prospects – if they can be attracted to the field. “There are opportunities for people with every engineering skill,” Fedder says. “Manufacturing crosscuts through all disciplines.” A number of schools already encourage students to work with industry, among them Olin College, MIT, Michigan, and CMU. Still, mention manufacturing to 18-year-old freshmen, de Weck says, “and they look at you like you’re crazy. People still think it is a greasy, oily, dirty job that is old-fashioned. People don’t think it’s sexy.”
Whether the revolution in manufacturing will turn into an engine of employment for the broader population of job-seekers is a question. MIT President Susan Hockfield told a 2010 conference on manufacturing technologies that 17 million to 20 million new jobs need to be created in the coming decade if America is to fully recover from the Great Recession. “And it is very hard to imagine where those jobs are going to come from unless we seriously get busy reinventing manufacturing.” But part of the genius of advanced manufacturing lies in the promise of productivity gains. “As we develop tools to become more productive, we produce an increasing number of goods, but we need fewer and fewer people to do it,” Michigan’s Ulsoy says. Along with its many other benefits, Wohlers says, manufacturing with 3-D printers “does reduce labor needs dramatically.”
The labor force required will need to be numerate and tech-savvy, able to operate and program computerized machines, troubleshoot complex equipment, use math to make logistical decisions, and read blueprints. Those with at least an associate’s degree from a community college will have a leg up.
Incorporating cutting-edge technologies into manufacturing will, in many cases, require massive amounts of capital. Cost is also holding back efforts to update legacy industries like textiles. The know-how needed to manufacture electronics-embedded “smart clothes” exists, but is still too expensive. “The problem is, how do you manufacture them in a cost-effective way?” Fedder says. Last June, PCAST argued that no single company was prepared at this stage to risk the sums needed to develop the potential of nanoscale carbon materials, flexible electronics, or nanotechnology-enabled medical diagnostic devices. The upshot is Obama’s Advanced Manufacturing Initiative, comprising CMU, MIT, Georgia Tech, Stanford, the University of California, Berkeley, the University of Michigan, and industrial partners Caterpillar, Corning, Dow Chemical, Ford, Intel, and Northrop Grumman. But when the White House could come up with only half the $1 billion PCAST recommended, ITIF’s Atkinson was disappointed: “It’s a step in the right direction, but it’s a day late and dollar short.” Another possible solution for the financing of risky research would be the creation of industry consortia, like the chip industry’s Sematech. Other countries, meanwhile, aren’t standing still. ITIF estimates that Japanese government spending on manufacturing R&D as a share of GDP is 35 times higher than the U.S. government’s, and Germany’s is 20 times more. The return of “Made in America” is not yet a sure thing.
Thomas K. Grose is Prism’s chief correspondent, based in London