Personal Manufacturing

It would be hard to imagine life today without personal computing devices such as iPads and smartphones. Conceptualizing these handheld computers, though, would have required an even larger leap of the imagination in the 1960s, when computers took up entire rooms and needed punch cards to store data.

“Even the most ambitious, forward-looking people back then couldn’t have predicted all the things computers would be used for in 2011,” says Hod Lipson, director of the Creative Machines Lab at Cornell University. But thanks to visionaries such as the late Steve Jobs, a personal computing revolution began in the late 1970s that brought desktop computers into people’s homes and, eventually, into their pockets.

Lipson thinks that three-dimensional printers—those robotic machines that build solid objects layer by layer from powders, liquids, and pastes—are sitting on the verge of a parallel personal manufacturing revolution. “In 20 years, many people will have a 3-D printer in their kitchen for printing designer foods and other products,” says Lipson, who works with the technology. A surgeon could have one in the operating room for printing bone grafts or replacement blood vessels, and a chef might have one in the restaurant for printing gourmet meals with varying textures and tastes. “In 40 years, we’ll have a hard time explaining to our grandchildren how we lived without one.”

Although Lipson doesn’t see 3-D printers ever being able to compete with mass production facilities in terms of cost-effectiveness and manufacturing speed, he says that right now “there are a lot of things we don’t make because they’re not viable in small quantities.” Manufacturing objects with 3-D printing offers people the opportunity to design custom-made pieces—lampshades, jewelry, artistic knobs, and furniture—with complex geometries.

Fabrication of some everyday objects that are now being mass produced could even shift into the realm of 3-D printing. Eye-glass frames, for instance, are pumped out of factories in large quantities. Based on a head scan, however, consumers could use modeling software to design and print custom-fit frames at home, Lipson says. “Once you do that, why would you want ill-fitting mass-produced frames anymore—ever?”

Before this personal manufacturing revolution can take place, though, researchers will need to develop a broader array of robust printing materials and, of course, low-cost printers with user-friendly software. Some machines printing at low resolution with simple materials are now available for less than $4,000 on websites such as Makerbot.com. But as it was in the personal computing revolution, a lot of the initial, exciting materials development and testing will be carried out on high-end (>$50,000) machines in industry and academia.

Even though 3-D printing is a rapidly growing market, about 70% of its more than $1 billion in sales currently comes from printed prototypes or model parts made of substances such as plastics (C&EN, April 25, page 22). Businesses use these models to help settle on designs for products such as car parts, sneakers, and toothbrushes. To make the jump from prototyping parts to manufacturing functional objects for aerospace, medical, and home applications, scientists are beginning to work with more advanced materials such as conductive polymers, metals, hydrogels, and biopolymers.

When engineers developed 3-D printing—or additive manufacturing, as they prefer to call it—in the mid-1980s, the machines principally worked with polymer liquids and powders. Although thousands of printers are in use today, a lot of them still use only polymers, says Brent Stucker, an industrial engineer at the University of Louisville. Nylon is a particular favorite, he says, because it runs well in the machines and has good strength and durability.

Most often, manufacturers use laser sintering to convert nylon into parts with intricate shapes. In that technique, a roller spreads a thin layer of polymer powder across a platform. Then a laser traces the first layer of a computer-rendered model into the powder, heating and fusing it into a solid. After that, the platform lowers, the roller spreads more powder, and the laser does its job again, solidifying a second layer. The process repeats, layer by 100-µm layer, until the object is complete, usually within a few hours. Like an archaeologist at a dig site, the manufacturer then unearths the 3-D-printed part from the powder block, dusting it off carefully with a brush.

But not all polymer powders work with laser sintering. Acrylonitrile-butadiene-styrene (ABS), for instance, works well in a type of 3-D-printing process called fused deposition modeling. The polymer comes as a filament coiled onto a spool, and it feeds through a heated extrusion nozzle on a robotic arm. As the ABS melts, the extruder squeezes it onto a platform in successive thin layers to build up an object. But, Stucker says, ABS “doesn’t work worth beans” with laser sintering because the polymer’s high viscosity at its melting point makes its shape difficult to control.

Both nylon and ABS are compatible with injection molding, a standard plastics manufacturing process, because the machine holds the polymers at a high temperature before pushing them into a mold to cool. Three-dimensional printing, in which compounds are typically heated fast and then cooled without a molded support, works differently, Stucker points out. “Very few people have thought about the differences between how materials get processed in additive manufacturing and traditional manufacturing,” he adds. Traditional materials are not necessarily the best ones for the job. “There’s a huge opportunity for new material formulations in this industry, especially for polymer chemists.”

One industry where 3-D printing can shine at fabricating polymeric and other parts is aerospace. Short production runs, like those required for aircraft parts, and complicated geometries are 3-D printing’s forte, explains David L. Bourell, director of the Laboratory for Freeform Fabrication at the University of Texas, Austin. The technology also saves on raw material costs compared with subtractive manufacturing techniques such as milling, in which machinists remove excess material during production.

Right now, aerospace engineers are exploring how to print aircraft parts such as wire holders and mounting plates out of high-performance electrically conducting polymers. These materials drain electricity so they don’t hold a static charge, which can lead to sparks and fires, Bourell says. “What’s all the rage at the moment,” he adds, “is to mix conductive multiwalled carbon nanotubes into the polymer powder.”

Both Bourell and Stucker collaborate with industry and government clients to test materials for 3-D printing. Bourell recently worked with the microelectronics industry to produce silicon-wafer holders that can stand up to the high temperatures of wafer manufacturing. His team mixed silicon carbide powder with phenolic polymer resin and used laser sintering to form the required part. To fill in any porous nooks and crannies left behind by the sintering process, the researchers “infiltrated” the part with molten silicon—a technique akin to soaking up liquid with a sponge.

Stucker, who uses laser sintering and a similar technique called electron-beam melting to fabricate solid objects, says his lab is currently examining powder and printing parameters for the Navy. After printing parts from a variety of materials—nylons, stainless steels, and titanium alloys—his team carries out fatigue tests and studies the microstructure of the pieces to understand how 3-D printing affects material performance.

Samuel M. Allen is also no stranger to developing new materials for 3-D printing. The professor of metallurgy at Massachusetts Institute of Technology has worked with colleagues Michael Cima and Emanuel M. Sachs to print metal parts without high-temperature tools like lasers. Cima and Sachs were the first to use the descriptor “3-D printer” in the early 1990s when they invented one at MIT for making ceramic objects.

This type of machine, which has been popularized since by firms such as Burlington, Mass.-based Z Corp., is smaller and less expensive than industrial printers that fuse materials with lasers and electron beams. The printer works by rolling out thin layers of powder and then tracing out a part’s structure layer by layer with liquid binders sprayed from ink-jet printheads. The powders can be metal oxides or plasters, and the binders can be polymers, acids, and even water.

To print metal parts without heat, Allen and his team used steel powder with low carbon content and an acrylic polymer binder to hold it together. As Bourell’s group did, Allen’s team strengthened the pieces by infiltrating them after the printing process with a molten high-carbon iron alloy.

Allen acknowledges that this method for printing metal parts isn’t used a lot today—laser sintering, a higher resolution technique, is more popular—but he says that the powder-bed ink-jet 3-D printers that originated at MIT can and are being used to make metal pieces in other ways. For instance, engineering students at MIT use 3-D printers to create intricate metal jewelry via a method called lost-wax casting. In this technique, a geometrically complex pendant or ring gets printed from wax powders. Then, a jewelry maker coats the wax model with plaster and later melts the wax out, leaving behind a ceramic cavity into which molten metal can be poured. Online printing services such as i.materialise.com fabricate jewelry based on customer designs this way as well.

The medical industry has also gotten into the 3-D printing business, using lost-wax and other techniques. Rock Hill, S.C.-based 3D Systems, for instance, offers 3-D printers and printing services for fabricating the metal framework of dental crowns and bridges. The firm uses laser sintering to directly make metal crowns from scans of plaster tooth molds. And it produces metal crowns with powder-bed ink-jet machines via lost-wax casting.

Others are developing 3-D printing for more advanced dental applications. Materials science researchers Jake E. Barralet of McGill University, in Montreal, and Uwe Gbureck of Germany’s University of Würzburg are focused on bone grafts for dental implant surgery. When a dentist implants an artificial tooth, Barralet says, there’s often not enough bone left in the patient’s jaw to firmly anchor the part. As a result, the dentist must graft either bone harvested from the patient or processed from a cadaver. To avoid this time-consuming bone-harvesting procedure, he and Gbureck are trying to make custom-fit 3-D-printed artificial bone-graft substitutes.

Because natural bone is porous and composed of calcium phosphate and collagen matrix, the researchers have been working with similar materials. So far, they have successfully printed blends of tricalcium phosphate and hydroxy­propyl­methyl­cellulose on a powder-bed ink-jet printer with phosphoric acid binder. The acid crystallizes the calcium phosphate, fusing the successive printed powder layers of the parts together.

Barralet and Gbureck also loaded a second printhead with recombinant bone morphogenic protein 2, a bioactive macromolecule that stimulates bone production, and printed it in various spots within the artificial bone parts (Adv. Funct. Mater., DOI: 10.1002/adfm.200901759). The researchers demonstrated that the protein retained most of its activity after the print process. According to Barralet, being able to print a bioactive just in the area where you want bone growth has a number of advantages: lower material costs, lower doses, and increased patient safety.

Bioprinted parts don’t necessarily have to be structural implants made of hard, inelastic materials like artificial bone, though. Some engineers are attempting to print more elastic tissue implants such as replacement blood vessels and heart valves.

At last month’s Biotechnica Fair in Hannover, Germany, an interdisciplinary research team from that country’s Fraunhofer Society exhibited an artificial blood vessel that it fabricated with 3-D printing. Because capillary vessels can have diameters of 20 µm or less, most 3-D printing techniques—with resolutions of 50 µm and higher—are not suitable for creating such fine vasculature. So the team, led by chemist Günter Tovar, designed its own printing system, which combines ink-jet and two-photon polymerization technology. The two-photon laser system precisely cross-links a blend of copolymers to create the vessels’ fine structure.

To test the elasticity of the printed vessels, the researchers plan to flow liquid through them in a bioreactor designed to mimic a pulsed bloodstream. In addition, the team will modify the inside of the capillaries with anchoring peptides and endothelial cells to line them like real blood vessels.

Jonathan T. Butcher also wants to print tissue structures for implantation in the body. But the Cornell biomedical engineer doesn’t want to add cells to his parts after fabrication. He wants to print them directly into the artificial tissue. “That way you don’t have to worry about the cells trying to crawl into your scaffolding” and getting where they need to go, he says.

To print live cells, however, 3-D printing techniques such as laser sintering and fused deposition modeling—methods that use extreme heat—won’t work. So Butcher teamed up with Cornell’s Creative Machines Lab. The engineers in that lab, led by Lipson, have developed a 3-D printer that works based on a room-temperature robocasting technique. This printer extrudes different materials from multiple syringes on a robotic arm. The technology relies on substances that harden on their own or that firm up when exposed to ultraviolet light.

Butcher now uses one of these machines to print artificial heart valves. Being able to print a custom heart valve, with intricate anatomical geometry, for a child with a congenital defect would be especially useful, Butcher says. Children, who are growing rapidly, require multiple open-heart surgeries over time so that doctors can implant progressively larger prosthetic valves. By printing a valve made of a patient’s stem cells and a temporary scaffold, surgeons can instead implant the piece and then sit back while the cells degrade the scaffold and remodel the tissue into a native structure all their own. When children grow, this structure grows with them.

So far, Butcher and his team have printed hydrogel scaffolds made of poly­ethylene­glycol­di­acrylate and polyester amide, cross-linking them with a UV lamp added to the 3-D printer. They also add alginate, which can increase the material’s viscosity for easier printer extrusion, and they add gelatin to which the cells and polymers can adhere.

When asked whether robocasting has a high enough resolution (about 300 µm) for printing lifelike valves, Butcher says he thinks the method will work just fine. “There are a lot of expensive ways of making the same thing,” he says. “Other 3-D printers cost hundreds of thousands of dollars” and take up a lot of real estate. To get a surgeon or hospital to buy into 3-D printing technology, the activation barrier needs to be lowered, he adds. “Just like the personal computer,” Butcher says, “we want something available for the masses that doesn’t occupy a whole room.”

The Creative Machines Lab at Cornell developed the robocaster Butcher uses several years ago and shared the machine’s blueprints online in 2007 as part of the open-source Fab@home project. The website lists the parts needed to build a 3-D printer and gives directions on how to put it together. To truly “do it yourself” would cost about $1,200, Lipson says, depending on the type of parts a person chooses.

Since the release of the blueprints, Lipson says, his team has seen a lot of interesting YouTube videos and comments on the Fab@home forum in which people have shared their uses of the homemade printers. Food printing is something that has been quite popular.

“People above the Arctic Circle in Norway were putting Nutella in the machine,” says second-year Cornell grad student Jeffrey I. Lipton. “Others put Cheez Whiz and frosting in.” Lipton and other members of the Fab@home team recently connected with chef David Arnold of the French Culinary Institute, in New York City, to get in on the food-printing craze. Together, the group printed tasty scallops, celery, and even turkey sausage in interesting shapes.

For the scallops, Lipton says, the team blended the seafood and added transglutaminase—often called meat glue—so that the mixture would resolidify once extruded. For the celery, they added agar to create a celery gel.

The food-printing craze is catching on in other research settings as well. Last year, scientists at the University of Exeter, in England, built their own 3-D printer with a heated extruder for printing gourmet chocolate (Virt. Phys. Prototyping, DOI: 10.1080/17452751003753212). Chocolate has different material phases, says lead researcher Liang Hao, so precise temperatures are needed to control its flow and solidification during printing.

“Food might be the killer app” to propel the personal manufacturing revolution forward, Cornell’s Lipson says. In the personal computing revolution, the first application that drove the market was gaming. Food printing might just be the thing to catch people’s attention so that they demand faster, cheaper, higher resolution 3-D printers at home, he says. “The race is on.”