Continuous liquid interface production helps 3D printing live up to its namesake.
Carbon3D Continuous liquid interface production helps 3D printing live up to its namesake.

Anyone who has ever watched 3D printing in action can attest that it’s actually more like 2D printing but with a thicker medium. Joseph DeSimone, CEO and co-founder of Carbon3D, in Redwood City, Calif., pointed out this misnomer when he took the TED2015 stage on March 16—the same day that the cover story about his research in the March 20 issue of the journal Science went live online.

DeSimone went to Vancouver to publicly debut continuous liquid interface production (CLIP), a technology that Carbon3D has been developing for two years. Rather than 3D print an object by mechanically extruding, stacking, and curing layers of material—usually thermoplastic—CLIP takes a page from sterolithography (SLA). It prints an object by chemically accumulating micron-thick layers through photopolymerization—the hardening of a photopolymer from a liquid state into a solid using light.  

The process recalls scenes from Terminator 2 in which the T-1000 robot assassin emerges from a pool of liquid, and for good reason. In his TED Talk, DeSimone said that the movie inspired Carbon3D’s co-founders—who have backgrounds in chemistry, material science, engineering, and physics—to rethink how 3D printing can overcome its primary limitations: speed, object strength, and material choice. “There are mushrooms that grow faster than 3D-printed parts,” DeSimone said. “If we could [print] really fast, then we could start using materials that are self-curing and have amazing properties.”


How CLIP Works

Light triggers photopolymerization while oxygen inhibits it. “If we can control spatially the light and oxygen, then we can control this process,” DeSimone said. Instead of a printer bed and printer extruder, CLIP consists of an upward facing digital projection system that directs ultraviolet (UV) light through a composite window into a reservoir of UV-curable resin, and a build platform, or stage.

The key to the technology lies in the window, which is made from Teflon AF (amorphous fluoroplastic) 2400, permeable to oxygen, and transparent to light. It has the “characteristics [of] a contact lens,” DeSimone said. By controlling the flow of oxygen through this window, CLIP creates a dead zone of uncured resin with a thickness in the tens of microns—about the diameter of two to three red blood cells—in the reservoir between the growing object and the window. This liquid interface allows production to proceed continuously. Conventional SLA technology, for comparison, prints a 2D pattern on the window, delaminate it, refresh the resin, re-position the build platform, and then repeat the process in discrete steps.

Carbon3D

CLIP’s digital projector shines what is essentially a movie made from UV cross-sections of the object through this window and dead zone. The photopolymer resin solidifies where UV light hits it. Meanwhile, the build platform lifts the curing and growing object from the reservoir.

To analogize on the building scale, conventional 3D printing is like masonry construction, where walls rise course by course. CLIP is like cast-in-place concrete, but without the hassle of building formwork.

Carbon3D and the University of North Carolina


CLIP Advantages

Though CLIP will have to address issues of scale, material safety, and cost, it has the potential to overcome the three shortcomings of conventional 3D printing that DeSimone noted in his TED Talk. The first is speed. CLIP can print objects 25 to 100 times faster than conventional 3D printers. This translates to a rate of between 300 millimeters (11.8 inches) per hour to more than 1,000 millimeters (39.4 inches) per hour, depending on print resolution, according to Carbon3D’s Science article, “Continuous liquid interface production of 3D objects.”

This 2-inch-diameter Buckyball-like structure took CLIP 6.5 minutes to print, while it took a Polyjet printer three hours, and a SLA printer 11.5 hours, according to third-party tests commissioned by Carbon3D.
Carbon3D This 2-inch-diameter Buckyball-like structure took CLIP 6.5 minutes to print, while it took a Polyjet printer three hours, and a SLA printer 11.5 hours, according to third-party tests commissioned by Carbon3D.

In his TED Talk, DeSimone predicted that CLIP’s printing rate could eventually be 1,000 times faster than conventional printers. Two factors that currently limit the technology are: resin cure time, which affects how fast the build platform can pull the growing object from the reservoir; and the resin flow into the build areas, or how fast fresh resin can refill the volume in which the preceding UV image was created.

The second improvement is in the mechanical strength of printed objects. Because CLIP’s printing process resembles injection molding, its objects look and act monolithic. They’re smooth on the exterior, solid on the interior, and exhibit isotropic behavior; that is, they have the same mechanical performance along the x, y, and z axes. Conventional 3D printed materials often exhibit different strength and mechanical properties depending on the direction in which they were printed.

Finally, CLIP can print with a range of media, including ceramics, biological materials, and soft elastomers, which can act as dampeners, have high strength-to-weight ratios, and resist temperature swings. What makes CLIP most promising, DeSimone said, is its ability to “make a [commercial-quality] part in real time that has the properties to be a final part” and to connect “the digital thread” from design to prototyping to manufacturing.

For consumers and companies that have been skeptical of investing in 3D printing and manufacturing, CLIP may be the technological breakthrough that will change their minds. But they certainly won’t be the only ones convinced: Carbon3D has already raised $41 million in venture capital to commercialize the technology.