The Dream of Anything Anywhere

The Engine
Dec 3, 2019 · 25 min read

For decades, 3D printing has seemed to hold a magical potential to build objects out of almost thin air. Is it finally starting to live up to the hype?

By Michael Blanding for The Engine
Illustrations by
Rodrigo Larraín & Julie Carles

It’s a busy Friday afternoon at Desktop Metal’s warehouse-sized office and factory space in Burlington, Mass. Men and women in business suits and khakis bustle around the entryway, picking at catering plates with sandwiches and cookies, while dozens of coders sit crammed into workstations in every inch of floorspace. “I’m super double-booked here, I’ve got customers, VARs, resellers …” says CEO Ric Fulop, apologetically, as he plops into a chair in a conference room.

He can be forgiven for seeming so stressed. In the three years since its founding, Desktop Metal has raised nearly $450 million in venture capital for its metal 3D printing technology. In January, it reached a valuation of $1.5 billion, achieving “unicorn” status, and giving hope to an industry that has at times seemed more hype than reality. “There is no hype,” Fulop insists. “Thirty years ago, it was a zero billion dollar industry. When we got into this three and a half years ago, it was a $5 billion industry. Now it’s a $9 billion industry. We think it’s going to be 10 times bigger over the next decade.”

At the heart of Desktop Metal’s breakthrough technology is a new printing process that builds up metal objects layer by layer, and then fires them in an oven to harden them. “By separating the shaping of the part from the thermodynamics, that allows us to make printers that are cost-effective and fast enough for mass production,” Fulop says. Its Studio System, suitable for small businesses and machine shops, goes for around $160,000, while its Production System machines go for upward of $750,000, designed for high-throughput industrial manufacturing.

Desktop Metal Production System Array

“I have a small part I could show to you in the back,” Fulop says. “One machine does a quarter-of-a-million parts per day — it’s a hundred times faster than the previous generation technology.” In addition, the process uses much less metal powder than previous techniques, so instead of costing $1,000 per kilogram of parts, it costs $50. “So it’s a 20th the cost for a finished part.” With efficiencies like that, he says, 3D printing could compete with mainstream manufacturing processes. “That’s where we are going with our Production System — we can enter the market where a lot of the capital gets spent.”

An Industry on the Cusp

Neal Stephenson’s 1995 science-fiction book, The Diamond Age, imagined a world in which families of the future could create anything they wanted on a specialized machine called a matter complier. Clothes no longer fit? You could throw them into a recycling bin, where they’d be torn apart molecule by molecule and reconstructed through nanotechnology into a new custom-designed outfit. On an industrial level, massive machines could print buildings out of diamond, and even whole artificial islands offshore.

[L-R] Ric Fulop, Founder & CEO, Desktop Metal; Jennifer Lewis, Professor at Harvard University & Co-Founder, Voxel8; Dayna Grayson, Partner, NEA

At the time Stephenson was writing, 3D printing technology was in its infancy, but seemed to hold unlimited potential to create anything in the home. Also known as additive manufacturing — since it builds shapes by adding material layer by layer, rather than removing it through machining or creating it by injection molding — 3D printing theoretically has multiple advantages over traditional manufacturing. It can allow for the creation of complex geometries and an endless iteration of designs, leading to prototypes in mere hours rather than the days or weeks.

On the other hand, the building process can be excoriatingly slow, and therefore expensive, and limited materials and low resolution can result in substandard quality. In the more-than-three decades since it first appeared, additive manufacturing has remained a niche process, regulated to making cheap prototypes or jigs and fixtures to aid manufacturing rather than products themselves. As recently as 2016, Inc. magazine said the technology was “dying.”

The last five years, however, has seen an unlikely surge in new additive manufacturing technologies — many developed at MIT — and a crop of innovative companies, many like Desktop Metal based in the Boston area. “Small advances in the platforms that were developed 30 years ago are leading to absolutely huge changes in their viability as a manufacturing platform,” says Jennifer Lewis, Hansjörg Wyss Professor of Biologically Inspired Engineering at Harvard University, and a co-founder of the additive manufacturing company Voxel8. “They are leading to higher surface finish, higher throughput, and faster build speeds.”

That has finally put additive manufacturing on the cusp of being able to compete with more traditional technologies as a method for production, not just prototyping. Already, it’s made inroads in industries such as aerospace, defense, jewelry, and medical and dental devices, all of which require specialized tools and equipment. From $9.8 billion in revenues last year, it is predicted by industry analyst Wohlers Associates to grow still more to $15.8 billion by 2020 and $35.6 billion by 2024. (While the company doesn’t separate revenue for polymer and metal manufacturing, metal 3D printers represents over a third of all printer sales last year, $948 million out of $2.6 billion.)

The question, however, is whether additive manufacturing can grow out of specialized industries to gain wider adoption across manufacturing as a whole. “It is still less than 1 percent of most manufacturing markets, so getting to a tipping point is still far away,” says Dayna Grayson, an engineer and investor with New Enterprise Associates (NEA). “The markets are so large, however, that by the time you get to 5 percent, you can build some very significantly sized companies.”

Richard D’Aveni, a professor at Dartmouth’s Tuck School of Business and author of the 2019 book The Pan-Industrial Revolution, predicts that in five to 10 years, additive manufacturing may well fundamentally change the way the world creates products. Rather than producing items in China and other countries overseas, companies like GE will be able to create its products in small, mostly automated factories all over the U.S., cutting down significantly on waste and shipping costs. Getting “over the chasm,” however, will take effort and innovation. “The overall industry is at an inflection point, but it’s stalled a bit,” he says. “Moving into the next phase of mass manufacturing becomes a significant problem for almost every technology.”

Polymers for Prototyping

3D printing was born in 1983, when engineer Chuck Hull dreamed of a quicker way to make prototyped parts. “At the time, you designed the parts on paper, a tool designer would design a tool, and then an injection molder would inject the plastic,” he says. “It would take weeks and months, and if it didn’t work, you had to do everything all over.” At the time, he worked for Dupont on a process that used UV light to “cure” plastic photopolymers to put veneers on tabletops. He wondered if he could use the same process to make a three-dimensional object. “I thought to myself, these are really just thin sheets of paper — is there a way to combine all of these layers to make a prototype?” he says.

[L-R] Chuck Hull, Co-Founder & Chief Technology Officer, 3D Systems; Max Lobovsky. Co-Founder & CEO, Formlabs; Richard D’Aveni, Professor of Strategy, Tuck School of Business at Dartmouth College

The machine he created consisted of a build platform submerged in a tank of resin, through which he shone a single-point UV laser to draw the shape for each layer. Once cured, the platform lifted slowly to allow a new layer of liquid to be cured. Eventually succeeding in creating a small eyewash cup, he called his invention a stereolithography apparatus (SLA), after the Greek words meaning “solid stone writing.” In 1986, Hull formed the company 3D Systems, still one of the leaders in additive manufacturing today. Over the years, the company has refined its original process, which still mostly focuses on prototyping, but has ventured into some manufacturing as well.

3D Systems Figure 4 Production System

In 1998, for example, the company Align began using 3D Systems’ printers to create their progressive set of dental aligners for adults as an alternative to braces. Made from biocompatible polyurethane resin, the products take advantage of the rapid customization to design the device for an individual patient’s mouth. More recently, surgeons have used the company’s printers to make surgery guides to help them make incisions in the right place. “When I step back and look at all of the progress that has gone on, it’s become a pretty amazing invention,” says Hull about his legacy.

The advantage of SLA is its high degree of resolution. Its downsides include the slow speed of the process, and the narrow range of materials usable through photopolymerization. In addition, because the layers are built up one after another, it can lead to some “shale-like” weaknesses in the strength of the final product. The company aimed to address those limitations in its latest printer, released last year. Called Figure 4, after a figure in Hull’s original patent, it is designed for speed, with a newly designed range of polymers and release membranes so the layers peel off more quickly, as well as a separate UV curing station to finish the process and increase strength.

SLA isn’t the only type of 3D printing technology, however. Shortly after its invention, the University of Texas-Austin grad student Carl Deckard used a different approach, taking a bed of powdered resin and using a laser to heat and fuse the granules together. Known as Selective Laser Sintering (SLS), the technology was commercialized in the 1990s, and eventually sold to 3D Systems, which uses it alongside SLA. Another company, Arcam, developed a similar process using an electron beam; it was acquired by GE in 2016. SLS has the advantage of being able to produce stronger products out of plastic, nylon, and metal, but is even slower and more expensive than SLA, and similarly limited in materials.

The most common type of 3D printing first appeared in 1989, designed by mechanical engineer Scott Crump of Minnesota. Looking to create a toy frog for his daughter, he took a hot glue gun and filled it with a mixture of polyurethane and candle wax, extruding a thin stream of heated material that stiffened as it cooled. Continuing to experiment with the technology, which he called Fused Deposition Modeling (FDM), he and his wife Lisa formed the company Stratasys, which has since grown to become the world’s largest 3D printing company, with more than $650 million in annual revenues.

Its printers use heated extrusion nozzles to squeeze out softened plastic filaments like a tube of toothpaste, laid down in layers to build up an object. The process is easy to use and faster than SLA or SLS, and allows for a much wider range of materials; however, does not have as high resolution, and is still not often cost-effective for mass production. Nevertheless, Stratasys has used it successfully to produce tools and prototypes for some 18,000 customers, including Airbus, Boeing, Lockheed Martin, NASA, Ford, and Volvo. Some have used it for specialized production parts, particularly in aerospace. Boeing, for example, says it saves $3 million for each 787 Dreamliner by 3D printing some 50,000 parts per plane.

Rise and Fall and Rise

Led by the popularity of FDM systems, the future looked bright for additive manufacturing by the mid-2000s, with the technology seemingly poised to go mainstream. Enter MakerBot, a compact 3D printer designed by a former art teacher–turned-entrepreneur Bre Pettis, who envisioned a 3D printer in every home, like a sci-fi matter compiler come to life. Pettis appeared on the cover of WIRED magazine in October 2012 confidently holding MakerBot’s Replicator 2 printer, with the bright orange coverline, “This machine will change the world.”

By then, the company had already sold more than 5,000 early versions of his machine to an enthusiastic crowd of hackers and DIY artists. Other companies such as PrintrBot and Solidoodle raced to join in the frenzy; 3D Systems created its own consumer system called The Cube; and in 2014, Stratasys acquired MakerBot itself for more than $400 million. The printers soon disappointed consumers, however, with constant hardware problems and clogged extruder nozzles. More crucially, it turned out most people simply weren’t interested in paying a premium price of $1,000 or more for a machine to print cheap plastic items at home.

Stocks that had risen 10-or 20- fold between 2009 and 2014 suddenly crashed back down to earth. (3D Systems went from $5 to $96 before dropping down to $9; Stratasys surged from $12 to $125 and back down to $16.) Publications and analysts that had been touting 3D printing as the next revolution were now declaring it dead. “There was a simplistic view that people were going to make things at home,” says Max Lobovsky, who watched MakerBot’s rise and fall as a grad student at MIT. “But of all the things we have at home, only a fraction can be made with anyone 3D printer.”

Even so, Lobovsky had personally benefitted from using 3D printers in digital fabrication workspaces at the MIT Media Lab, and wasn’t ready to declare the technology “over.” As he finished his master’s degree in 2011, he got together with two other MIT students to explore a new idea for a desktop printer. Instead of using FDM, it would use SLA, a technology that had been more or less passed over for personal use. “When we started, it only existed in these very large machines,” Lobovsky says.

Lobovsky and his colleagues created a new technique that inverted the usual SLA process, so only a small tank of resin was needed. They used a laser from Blu-Ray players, creating special software to calibrate it. Launching their company as Formlabs in Somerville, Mass., in 2017, they advertised their new machines on Kickstarter, with a price tag of just over $3,000. Instead of targeting home consumers, Formlabs aimed for small companies and independent craftspeople — so-called “prosumers,” who might not be able to afford a $50,000 machine from 3D Systems or Stratasys.

Formlabs Form 3 printer and components

The Kickstarter campaign made $3 million in preorders, and has been in the black ever since. Its success has attracted over $100 million in investment from the likes of Foundry Group, Autodesk, and former GE CEO Jeff Immelt, now a venture partner at NEA, earning Formlabs a $1 billion valuation by last August. The company now has 600 employees worldwide, and is already looking to hire more. With a build size of about 5 ½ x 5 ½ x 7 inches, Formlab’s sweet spot is in prototyping and molds. Its technology, for example, can print patterns in heat-resistant plastic used by jewelers as molds for metal, as well as molds for custom-designed dental liners and hearing aids. Last October, Formlabs announced a partnership with Gillette to create custom-designed razor blade handles, with 48 designs in 7 colors, under the tagline, “a man’s grooming tools should be as unique as he is.”

Aside from molds and mass customization, however, SLA has also gotten a second look as a mass-manufacturing technology. One of the most promising new additive manufacturing companies, the Redwood City–based Carbon, started with a similar concept to stereolithography, but rather than building an object painstakingly layer by layer, it envisioned a rapid, continuous process. “We wondered if we could grow parts out of a puddle, like the T-1000 in the Terminator movie,” says Carbon founder Joe DeSimone, a University of North Carolina chemistry professor and winner of the Lemelson-MIT prize in 2008, referring to actor Robert Patrick’s character famously emerging out of a pool of mercury-like liquid in 1991 film Terminator 2: Judgment Day. “In other words, could the mass of an object be derived from a source of liquid resin below it?”

Carbon production platform

The technique he developed, called continuous liquid interface production (CLIP) uses a special window that controls the flow of both light and oxygen to allow photopolymer resin to solidify on an inverted build plate that moves slowly upward from a pool of liquid. “This enables the generation of a continual liquid interface,” says DeSimone, “and thus the ability to rapidly grow layerless parts.” The process allows the UV light to flash a pattern all at once, rather than drawing it with a laser, allowing parts to be built 25 times faster than previous SLA processes with less waste and a smoother finish. It also makes parts stronger, since they don’t have the same shale-layer effects of traditional SLA. “With our materials, we are able to achieve parts with properties that compare to injection molded parts,” DeSimone says. The company has designed resins for silicones, polyurethanes, elastomeric polyurethanes, and rigid high-temperature materials such as epoxies.

Founded in 2013, the company raised nearly $700 million by late last year, both from VCs such as Baille Gifford and ARCHina Capital; and companies including GE, Johnson & Johnson, and Adidas. It crossed the $1 billion valuation threshold last year (making it the third additive manufacturing “unicorn” along with Desktop Metal and Formlabs), and as of June, reached $2.4 billion. Starting in 2017, Adidas began using Carbon’s technology to create plastic cushioned midsoles for its Futurecraft running shoes for the consumer market. More recently, Carbon has revealed a larger printer called the L1, with a build volume 10 times that of its previous printer. Sports equipment company Riddell has used the technology to make custom-designed football helmet liners for NFL players, using some 140,000 individual struts, printed in two materials at once. “Certain areas of the liner function differently than other areas in order to optimize energy management in the event of an impact,” DeSimone says. Ridell plans on releasing the helmets to the consumer market late this year or early next year.

[L-R] Joe DeSimone, Co-Founder & CEO, Carbon; Josh Martin, Co-Founder & CEO, Fortify; Duncan McCallum, CEO, Digital Alloys

Another company, Fortify, also uses a projector rather than a laser to polymerize plastics, but rather than build the structure continuously, it creates it layer by layer, a technique called Digital Light Processing (DLP). Created by Josh Martin and Randall Erb at Northeastern University, Fortify also integrates electromagnets into the printing process it calls FluxPrint, which can control the orientation of fibers embedded in photopolymeric resins to give the structure added strength, stiffness, or thermal conductivity. “It’s very programmable and wireless in nature,” says Martin, who co-founded the company Fortify, based in South Boston, in 2016. “We can pretty precisely control these additives without needing to use a large energy potential.”

By changing the magnetic fields during printing, the company can selectively polymerize different areas, building up materials with unique properties, voxel by voxel. For example, Fortify has used its magnetic 3D printers to produce electrical connectors for electric vehicles — complicated structures that need to be embedded with RF (radio frequency) properties, at the same time with-standing high temperatures beneath the hood. “To make those with tooling processes is super-expensive,” he says. In addition, the company has been able to make high-performance plastic tools for injection molding. “The 3D printing space has been trying to crack this for decades, and they have not been able to perform under high levels of pressure,” Martin says. “We can take a process that would usually require three months, and turn it around in a week or less.”

Expanding to Metal

While additive manufacturing originally used plastic polymers, one of the major innovations has been the adaptation to metals. German firm EOS first commercialized a technology called Direct Metal Laser Sintering (DMLS) in 1995, a variation on SLS that shoots a high-wattage laser into a bed of powdered metal. The process allows for precision parts with complex geometries, even though it can be slow and expensive, with a single part costing anywhere from $500 to $2,000 to make. For that reason, the process has made the most inroads in the aerospace industry; GE has used it to print parts for jet engines, and SpaceX and Virgin Galactic have used it to create parts for their rockets.

Los Angeles-based company Relativity Space is working to create the first entirely 3D printed rocket, raising $145 million this October from Bond and Tribe Capital, bringing its total investment up to $185 million. The company uses a massive FDM printer with 18-foot robotic arms to deposit melted metal wire on a spinning turntable to make round parts such as fuel tanks. The company, which was founded in 2015, plans on launching its first rocket in 2021. With a payload of up to 1,250 kilograms, the rocket is designed to use 100 times fewer parts than traditional rockets (1,000 compared to 100,000), and be constructed in just 60 days rather than several years.

Recently, other companies such as Desktop Metal have taken metal printing’s success in aerospace and introduced new processes to bring down cost for other industries as well. Rather than use heat to fuse metal wire or powder, Desktop Metal uses a process invented by MIT professor and co-founder Ely Sachs to fix powder with an adhesive binder instead. Past a series of protective doors in Desktop Metal’s factory, row upon row of mini fridge–sized machines whir productively. Each has a pair of mechanical arms that pushes sticks of wax and metal powder through a nozzle, similar to FDM, building up objects layer by layer on a build plate. When they are done, a technician will place the objects in a debinder where a liquid solution will remove the wax, creating open channels inside the object. Then it is put into a hot furnace where the metal shape will be sintered, hardening as it shrinks by up to 20 percent. At the same time, the sintering bonds the metal from all directions, fusing the layers together to increase overall strength.

The process, which the company calls its Studio System, began shipping in June to companies including Ford, Stanley Black & Decker, Goodyear, and Owens Corning, which use it to create prototypes, molds, jigs, and fixtures to help their manufacturing process, says VP of Product Larry Lyons. In addition, he says, vehicle companies including BMW and Caterpillar are using it to print spare parts on demand. “Caterpillar has a 40-year guarantee on spare parts,” he says. “So they just have these massive warehouses all over the world filled with parts they might ship out once a month.” By printing parts on demand, they can dramatically lower costs for consumers, who could have a part made right at the dealership.

Through another set of doors is Desktop Metal’s much larger Production System, which uses a different process called Binder Jetting. Large machines drop a thin layer of powder on a build plate, then a print head runs over it, dropping binder layer by layer, allowing for exquisite resolution of detail at 100 times the speed of SLS. When printing is done, workers in white hazmat “rabbit suits” remove excess powder, which can be recycled. The system can be used to create dozens of parts at once, all fired together in an industrial-sized furnace, resulting in the company’s 20-times cost savings. The company plans to start shipping its first systems by the end of the year.

Desktop Metal isn’t the only company in the Boston area to pioneer metal manufacturing, however. In nearby Watertown, Markforged was founded by engineer Greg Mark as a one-stop-shop for 3D printing, employing a range of technologies and materials. Among them is its new Metal X printer, which uses a system similar to Desktop Metal’s Studio System to lay down metal powder encased in plastic binder into 3D shapes that are then hardened through sintering. The company has raised $137 million to date, from the likes of Matrix Partners, Summit Partners, Microsoft, and Porsche.

While binder jetting can increase speed for metal production, it does have its drawbacks. As sintering shrinks products, it can create a very slight variation between them. While software can compensate for that variation to an extraordinary degree, there may still be a 3 percent range of variability. While that isn’t a problem for small objects, issues obviously increase with size.

Down the street from Desktop Metal in Burlington, startup Digital Alloys has been using a different method for metal printing that allows for larger sizes. Similar to FDM, it starts by extruding a metal filament through a nozzle. However, rather than using heat to melt the metal, it sends an electric current through the wire, liquefying it just at the point of contact. “It’s the same physics that heats a coil in a toaster,” says company CEO Duncan McCallum, a mechanical engineering graduate from MIT and longtime venture capitalist, who co-founded the company in 2017. Called “joule printing,” the process creates less waste than machining, and can build layers quickly, without the need for sintering. The company says it sees “full material fusion” between layers, creating a density of 99.5 percent and a tensile strength stronger than cast metal, comparable to wrought metal. “We’re low-cost, the quality is exceptional, and we get a very dense metal right off the printer,” McCallum says.

A Digital Alloys print in progress

It terms of cost, the company’s sweet spot, says McCallum, are objects “larger than a tennis ball, but smaller than a beach ball.” For those parts, he says, “we can do it faster and cheaper than any other solution we’re aware of.” (The process doesn’t currently have high enough resolution for smaller objects, but could scale to larger objects with a larger printer.) A titanium fuselage bracket for aerospace, for example, uses 90 percent less material and can be made in 70 percent of the time, cutting costs by 60 percent — from $980 to $385. So far, the company has been producing parts in-house for clients including Boeing and Ford. It plans to sell its printers to companies starting in 2021. “First we’re building the cookbook,” McCallum says, “then we plan to give the cookbook to others.”

Widening the Scope

One of the most difficult challenges in additive manufacturing is how to print with several colors or materials at once. One of the newest entrants into the 3D printing space, HP, has taken some steps to solve that problem with a technology called Multi-Jet Fusion (MJF), developed in 2015. A sort of cousin of SLS, MJF starts with a bed of polymer powder. However, instead of shooting it with a laser, the printer heats the entire bed almost to its melting point, and then passes a print head over the area with thousands of small nozzles that deposit an infrared-sensitive ink. When a high-power infrared energy source passes over the same area, it fuses it to the powder underneath. Some 10 times faster than SLS and half as expensive, the process also allows printing in several different colors at once, with the ability to control down to individual voxels (the 3D equivalent of pixels). The technology can be used to print multi-color prototypes, as well as objects such as custom-designed cell phone cases.

Another technology that uses a similar technique is called Material Jetting (MJ). Created by PolyJet in 1999, and since acquired by Stratasys, it forgoes the messy powder bed to print droplets of ink directly onto the print bed. The drops are either heated, setting as they cool, or subjected to a UV light to cure them, similar to SLA. The process is fast, relatively clean, and allows for control not only of colors, but also of the materials themselves, able to mix different types of polymers on the fly for different voxels. Stratasys’ J750 machine, for example, can print in six different materials with hundreds of thousands of color options, and has been used to print anatomical models.

“If you were an alien coming down from space and had never seen 3D printing before, you would say that this is objectively the technology everyone should bet on,” says Davide Marini, CEO of Inkbit, a company based in Medford, Mass., that also uses material jetting. There are two problems, however: Any material too viscous will clog the tiny nozzles (5 to 15 microns wide), limiting materials. Second, random variations in the ways the nozzles spray the ink lead to imperfections as layers build. To compensate, Stratasys’ J750 machine sweeps a scraper across the surface between layers; however, that slows the process and further limits the materials, since anything too sticky, such as epoxies, will cling to it.

Marini studied mechanical engineering in Milan and worked as an investment banker in London before coming to MIT to study biomaterials. There, he learned of new technology being developed by MIT scientist Wojciech Matusik at the Computer Science and Artificial Intelligence Laboratory (CSAIL) to address the limitations in MJ. He developed chemicals that would be liquid enough to shoot easily from the nozzles, but then change when exposed to UV light, to create more complex materials such as epoxies. In addition, he developed an ingenious system to cut down on errors using machine vision and artificial intelligence.

As ink is laid on the object, a lens scans it at high resolution to find any random errors. The print head then automatically compensates when laying down the next layer, placing more or less ink in spots to make the surface flat, obviating the need for a scraper. “We can scan at a resolution of 20 microns without any change to speed,” says Marini, who spun the company out of MIT in 2017 with Matusik and funding from Italian packaging company IMA.

Inkbit has since formed a partnership with Johnson & Johnson to create products such as medical devices combining multiple materials at high resolution. One device, for example, features tiny channels of less than a ½ millimeter in size. With the speeds possible, an intricate plastic device that would cost $350 using SLA could be produced at the cost of only a few dollars, Marini says. Next, the company plans on installing several beta machines at partner companies to continue to test the technology before releasing it more widely.

The possibilities for multiple materials go beyond combining polymers. At Harvard, Jennifer Lewis’ lab uses a technique similar to material jetting, but using a pneumatic system to extrude a paste-like ink through larger nozzles at room temperature, allowing for a much wider range of materials. At the Wyss Institute, Lewis has been a pioneer in printing 3D artificial organs with living cells. On the manufacturing side, however, she helped found the company Voxel8, which uses a method called ActiveMix to build complex products.

One of the company’s main projects is 3D printing shoes. Unlike SLA processes used to print midsoles, Voxel8 is focusing on the fabric upper sole of the shoe, using its technology to embed polyurethane inks into the fabric. “We can take a piece of textile and screen print all of these zonal features and patterns, both for aesthetics and also functional purposes,” says Lewis. For example, by the way material is printed, the printer could make parts of the shoe stiffer than others. Currently, she says, the company is working with two of the top five athletic footwear companies to develop mass-customized shoes.

Lewis and her colleagues have also used the system to embed electronic components, such as sensors and batteries, into fabrics to create wearable electronics. “You name it, we have printed almost every class of functional material you can imagine, and we have print heads that can switch and mix on demand.” While many of these methods are still under development, Lewis believes it is only a matter of time before more manufactured products embrace the capabilities of 3D printing — not only to replace their current manufacturing techniques, but to allow for the creation of new forms and materials not possible any other way.

“I’m not so much a believer in the idea that every home will have a 3D printer in 10 years, but the technology is penetrating evermore into companies’ production platforms,” Lewis says. “It’s providing an opportunity to design new materials, voxel by voxel, in a way we’ve never been able to do before.” We may never see the day in which some matter complier allows us to create everything we can possibly desire. But we may see a day — and soon — in which many different types of additive manufacturing technologies will combine to create significant parts of the objects we use, drive, and wear every day.” +


Material World

In addition to new technology and machines, additive manufacturing will need a new class of high-quality materials if it is going to truly compete with traditional manufacturing techniques. As a doctoral student at Caltech, Raymond Weitekemp never intended to go into 3D printing. “I came into additive manufacturing kicking and screaming,” he says. He’d seen the crash after MakerBot failed to live up to expectations, and was skeptical about the industry. Working to develop high-performance materials in the lab, however, he stumbled upon a new photosensitive polymer, when he used the wrong catalyst for a reaction. “I like to say I half-invented it,” he says, “it was completely unexpected.”

That polymer, which he called COR alpha (short for Cyclic Olefin Resin) turned out to be 10 times as rugged and durable as other photopolymers, and able to withstand high temperatures without losing strength. “In Izod impact tests, most people can do 10 to 30 joules per meter, but we can do 100 to 300,” he says. In 2016, Weitekemp has since spun the technology into the Berkeley-based company, Polyspectra, whose sole purpose it is to create new high-quality materials for additive manufacturing.

[L-R] Jay Rogers, Co-Founder & CEO, Local Motors; Raymond Weitekemp, Founder & CEO, polySpectra

Going Places

After specialty industries such as aerospace and medical and dental devices, the automobile industry may be the best hope for additive manufacturing breaking through to the mainstream. Already, companies including Ford, BMW, and Volkswagen have gone from using 3D printing for tools and prototypes to 3D printing their first parts for use in commercial vehicles.
“Automobiles are the big bellwether that everyone is watching,” says Tuck professor Richard D’Aveni. “So far, it’s only making slight inroads, but it’s marching towards wider use.”

Meanwhile, one company isn’t waiting for the rest of the industry to catch up. Arizona-based Local Motors has already started using additive manufacturing on a heroic scale to produce the world’s first 3D-printed cars. The company is the brainchild of Jay Rogers, an Iraq war veteran who saw two friends killed during their deployments, due to outdated military vehicles. When he returned from overseas in 2006, he enrolled in Harvard Business School with the intent of creating a company to get new technology into military vehicles more quickly.

At the time, 3D printing was still a nascent technology for manufacturing, and Rogers’ company was looking at rapid methods for laser cutting metal; but after he saw a demonstration of an SLS machine, he realized that cars could be printed more quickly and more durably using a polymer frame. Switching his focus to civilian use, Rogers founded Local Motors and used a giant FDM printer to lay multiple layers of carbon-fiber-reinforced plastic to create a convertible buggy called Strati — the first 3D printed car — revealed at the Detroit Auto Show in 2014. “Our first car took 44 hours to print,” he says. “That’s not industrial speed. But the next year, we made
a car double the size in half the time.”

Since then, the company has created a larger shuttle bus called Olli, printed in two large parts — top and bottom — that are fitted together. While the other 2,000 parts of the car, including tires and electronics, are conventional, the bus is 90 percent 3D printed by volume. After exhaustive crash testing in which the frame was tooled for maximum safety, the company has sold the car commercially in 10 cities, where it operates on college, government, and assisted living campuses at speeds of up to 30 miles an hour. The company is still working its way through a thicket of laws to certify its cars for highway use, but Rogers eventually predicts that 3D printing will create cars that are both safer and cheaper than traditional vehicles. “It has nothing to do with the capability, and everything to do with regulation,” Rogers says. “We believe the best way to show we can do it is to do it.”

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade