The Migration of Printed Electronics to 3D
Emerging 3D-print technology allows manufacturers to produce not only physical components, but also the electronics that enable them to function, adding a new wrinkle to additive manufacturing.
The 3D printer is much more than a toy for hobbyists. This technology has enabled the development of creative products with real value and has the potential to revolutionize manufacturing. The basic benefits of additive manufacturing, where 3D parts are deposited layer by layer in a precise pattern, lie in the ability to quickly change the physical design of an object by editing a software file, while avoiding the material waste inherent in subtractive processes like machining and metal electroplating. Longer term, the marriage of electronics and 3D printing might allow us “to create structures you can’t make any other way,” suggests Lawrence Gasman, president of SmarTech Markets Publishing.
The leading commercial applications for 3D printing include building manufacturing prototypes for various industries, allowing engineers to inexpensively test out new designs and produce tooling that can aid in the manufacturing process. Today, prototyping represents the majority of the activity in 3D printing, but there has also been considerable progress in medical applications, including everything from 3D models of anatomy for training surgeons to 3D-printed medical implants.
There is, however, great potential beyond the hype that suggests that anything can be 3D printed. It appears that the integration of 3D printing and printed electronics is finally resulting in viable products, such as miniature antennas, improved test fixtures, and modular components that can be integrated into electronic products. SmarTech, which has produced market research reports on the 3D-printing industry for several years, is recognizing this trend with its first report on 3D printing in the electronics industry, published in August.
Additive manufacturing relies on one of several methods, the choice of which depends on the material to be printed, performance requirements, and the customer’s budget for equipment. Here are the technologies that are most applicable to 3D printing for electronics applications.
The machines that most people associate with 3D printing use fused deposition modeling (FDM), in which a filament of material is melted and deposited onto a surface in layers. The printers are relatively inexpensive, ranging from tabletop machines for hobbyists to more substantial equipment for industrial use. FDM printers primarily print various plastics, although the filaments can be infused with metals or other materials.
Selective laser sintering (SLS) uses powders as the raw material. As the name implies, a laser sinters (melts) the powders to fuse them in precise locations, avoiding the need to heat the entire part to produce the desired mechanical strength. SLS is versatile and can work with polymers, metals, ceramics, and even exotic materials such as food-based powders, but the machines are expensive and best suited to high-end applications.
Many high-end 3D printers use stereolithography (SLA), in which a UV laser is used to selectively solidify liquid resins made from photosensitive polymers. The process can print high-quality objects with high resolution, and, historically, came with a cost to match. The newest generation of printers is much less expensive, with models that are even accessible to hobbyists.
Inkjet printing is a common technique for high-resolution 2D printing, but it can be used in a 3D context by printing multiple layers on top of one another. Unlike FDM and SLS, however, inkjet cannot be considered a true 3D approach, as it is best suited to applications where the thickness (or height, z) is much less than the length and width (x and y) dimensions.
Aerosol Jet printing, a process patented by Optomec, is related to inkjet printing and has been touted as a 3D-printing technique because of its capability of printing features onto curved surfaces rather than only to planar substrates. Like inkjet, however, it does not truly print 3D structures. The process generates a mist from liquid raw materials and focuses the aerosol stream of material to deposit droplets onto the substrate. An integrated laser sinters the deposit, similar to the process in an SLS machine, without heating the underlying substrate. The Aerosol Jet process can create features as small as 10 microns, printed over steps and contours.
The Importance of Materials
3D printing has historically been focused on printing a variety of plastic materials. When it has been extended to printing metals, it usually involves bulk components for aerospace and similar applications. High-end printers can create very detailed structures with micron-scale resolution, but the materials are not tailored for producing reliable electronics.
For this reason, the focus of 3D-printed electronics is not to print complex circuits, but rather to take advantage of applications that encompass both mechanical and electrical function and require conductive traces to make the electrical connections. This is similar to the evolution of 2D-printed electronics, where early commercial products relied on fairly low-resolution conductive traces in applications that were not extremely demanding.
The sweet spot for 3D printing is micron-scale resolution, with features between 10 and 100 microns in size. This is the realm where screen printing struggles to achieve sufficient performance, while photolithography is overkill.
In order to meet this demand, manufacturers are developing silver inks designed specifically for 3D printing. For many product designs, the inks need to be dispensed onto curved, not flat, surfaces, so they need to be viscous enough to hold their shape while not being so viscous that they cannot be dispensed. They also need to cure at low temperatures, ideally at room temperature, because they are usually being co-printed with plastic materials that cannot withstand the elevated temperatures typical of standard electronics processing.
Silver-based inks are preferred for their high conductivity and ease of dispensing. In the high-value, low-volume applications associated with 3D printing, cost of materials is not a critical factor. In the long term, if 3D printing extends to higher volume, cost-sensitive applications, it may be desirable to switch to more economical copper inks. But, at present, the challenges associated with oxidation of copper make it not worth considering.
3D-printed electronics can use standard plastic materials, but other choices may be preferable for optimum performance. Voxel8, a spin-out company from Professor Jennifer Lewis’ research group at Harvard, is developing a process to use epoxy because its thermal and mechanical properties are closer to materials that are used in printed circuit boards (PCBs). “We are actively working on new materials all the time,” says Karl Willis, VP of product development. “The 3D-printing industry has found a home with high-value applications,” which often demand customized materials.
Real-World Applications of 3D-Printed Electronics
Several companies are making commercial products that either combine 3D plastic parts with conductive printed traces or merely print conductive lines onto curved surfaces. The first approach, which is the most exciting and the most likely to change the way electronics are manufactured, merges the capabilities of FDM or SLA with inkjet printers. These 3D printers can co-print filament-based plastics and conductive silver inks to create a single, fully 3D part with either exposed or embedded conductive traces.
Voxel8 is rapidly developing this technology. It began shipping its 3D printer in May as a developer’s kit consisting of the printer, required software, and printing materials. The plastic filament is PLA, a thermoplastic resin chosen because it’s a familiar material that is easy to use. The silver ink is a proprietary formulation that is highly conductive and dries quickly at room temperature.
Voxel8 showcased sample products at the Consumer Electronics Show in January, including a printed watch with embedded microcontroller and LEDs and a small quadcopter drone with an embedded PCB. The ability to pause printing to insert traditional electronic components and then print additional layers of metal and plastic over them opens up further potential applications of 3D printing.
Creating More Versatile Test Fixtures
Companies that design hardware need to build custom test fixtures to determine whether it is working as designed. Since test fixtures are a highly customized, low-volume product, they are a perfect application for 3D printers. For example, RightHand Robotics makes robotic arms with fingers that can grasp objects. The tactile sensors on the fingers need to be tested, and Voxel8 designed and built a fixture with an embedded microcontroller and an LED that lights up when the sensors are functioning properly. (Left) Similar fixtures could be designed to test a wide variety of electromechanical parts.
Printing antennas is nothing new, but as electronics shrink, shrinking antennas along with the rest of the components causes problems with performance, such as reduced bandwidth and efficiency, and consequently reduced frequency range. One possible way around this problem is to deviate from the traditional planar antenna with a design consisting of lines that meander to fill a hemispherical shape. (Right) 3D printers enable printing such an antenna configuration directly. In contrast, traditional methods would involve printing the patterns onto a Kapton sheet, which would then be glued to the structure, resulting in a much weaker construction.
Voxel8 has partnered with The MITRE Corporation to produce small, high-performance conformal antennas for a US government project. These antennas would be extremely difficult, if not impossible, to fabricate using conventional methods. As of press time, the companies were preparing to release a prototype.
PCBs: A Disruptive Use of 3D Printing
While Voxel8 is successfully embedding traditional PCBs into 3D-printed objects, startup Nano Dimension is printing the PCBs themselves. Simon Fried, co-founder and chief business officer, explains the motivation: “We were looking to establish a disruptive application, something unique, that would be at the forefront of new product design and technical innovation.” They felt that the PCB industry was the right place to start.
Prototyping PCB designs is typically an expensive, time-intensive process, which is becoming a more serious concern as product life cycles get shorter. The facilities that fabricate prototype boards are almost exclusively located in Asia, but the designers are more likely to be in North America. The iterative process of prototyping is tedious: Designers send a file to a prototype house, wait two weeks or so for the boards to be fabricated and shipped, and then repeat until the design is finalized.
Bringing prototyping in-house would save weeks of time and protect valuable intellectual property, but the required equipment is expensive and bulky. Nano Dimension developed its DragonFly 2020 3D Printer (left) specifically to solve this problem, building prototypes within hours and vastly speeding up the process.
The DragonFly printer co-deposits two materials, a highly conductive silver nanoparticle ink and a functional dielectric ink, to build an entire PCB from the ground up. Nano Dimension designed its inks in-house to ensure they would work well together. A printed PCB prototype does not just need to look like the real thing, as might be the case for a mechanical part prototyped using 3D-printed plastic, but it needs to function like a real PCB. The company claims that its printed PCBs can.
Nano Dimension is currently working with beta customers and plans to start shipping printers near the end of 2016. One caveat: While these printed PCBs are multilayered structures, the “3D” aspect of Nano Dimension’s products is minimal, as the PCBs are flexible and each layer is only a few microns thick.
Building a Modular Cellphone
Another startup is producing truly 3D electronic components as it moves toward its vision of tomorrow’s consumer electronics being assembled from modular pieces. Nascent Objects is addressing e-waste by creating a platform where users can print a variety of modules – speakers, cameras, sensors, and more – and join them together to make a functioning device. Putting them together is supposedly as simple as inserting a SIM card into a cellphone, so it is easy to upgrade a device rather than discard it when improved components become available.
Nascent Objects does not build printers. Its IP lies in its printing process and template-based software, which can be used with many different commercially available 3D printers that use SLA or inkjet printing. The modules consist of a plastic housing, which can be made from a variety of materials, with embedded conductive traces and the ability to encapsulate components such as sensors and LEDs. Its “main board” includes an embedded microcontroller that runs the functions of the other modules.
Moving Toward Volume Production
3D printing is a great tool for prototyping, where the value of being able to change designs on the fly makes the process much more efficient despite the fact that printing a single product can take many hours. The long-term goal of suppliers, however, is to move toward higher volume production in order to expand into more lucrative markets. In order to do this, it will be necessary to vastly increase throughput. Printing thicker layers or moving the printheads faster, though, is not the answer, as the result will be low-quality parts that do not meet customer requirements.
The answer for higher volume 3D printing lies in parallelization, which can be achieved through various approaches. Lewis’ group at the University of Illinois, and later at Harvard, came up with a method to print from up to 64 printheads at once. The design needed to meet several challenging criteria, including the ability to withstand the high pressures required to print highly viscous fluids from a continuous filament; achieve uniform flow rates across the nozzle array; allow simultaneous patterning of more than one material; and be scalable for large area, rapid manufacturing. The design, which Voxel8 is in the process of commercializing, can print a 3D part that might currently take an entire day to print using a single nozzle in less than half an hour.
HP has created a 3D printer expected to begin shipping in late 2016 that can print an array of parts simultaneously. The company claims that its Multi Jet Fusion proprietary architecture (left) can dispense 30 million drops of fluid per second. The process appears to be similar to SLS, but it uses both a fusing agent and a detailing agent, and is much faster. Initial applications are not geared toward printed electronics, but HP’s vision includes the ability to tailor the properties of each voxel (unit cube, or 3D pixel) independently. If this were extended to control of conductivity and producing translucent voxels, it could be possible for the system to create printed lenses, sensors, or parts with embedded electronics.
The promise of faster production at lower cost could be what it takes to launch applications beyond prototyping and truly disrupt the manufacturing process. Parallelization will need to happen at both the nozzle level and the system level, where multiple printheads dispensing from multiple nozzles are part of a network of several machines that create an efficient assembly line. When 3D printers can reliably co-print functional polymers and conductive inks fast enough to produce hundreds of parts per day rather than a handful, the industry may be ready to revolutionize electronics manufacturing.