By Dr Matthew Dyson, Technology Analyst, IDTechEx
Mention an electronic circuit and you are likely to picture a printed circuit board (PCB) – a rigid rectangle in a characteristic green colour with copper lines and a bewildering array of components soldered onto it. But does adding electronic functionality means using a PCB and thus requires shoehorning a rigid rectangle into the product?
The emerging approach of 3D electronics suggests not. Instead of making them separately on a rigid board, 3D electronics instead involves integrating electronic functionality within or onto the surface of objects. Whilst antennas and simple conductive interconnects have long been added to the surface of injection-moulded plastic objects, 3D electronics is undergoing extensive innovation with new materials, metallisation methods and manufacturing methodologies.
Whilst aerosol and material jetting enable conductive interconnects to be applied to surfaces, in-mould electronics and 3D printed electronics enable complete circuits to be integrated within an object. Between them the various approaches offer multiple benefits that include simplified manufacturing, reduced weight and novel forms. With 3D electronics, adding electronic functionality no longer requires incorporating a rigid, planar PCB into an object then wiring up the relevant switches, sensors, power sources and other external components.
Figure 1: The status of different 3D electronics technologies are detailed in the IDTechEx report “3D Electronics 2020-2030: Technologies, Forecasts, Players”
Electronics on a surface
The best-established approach to adding electrical functionality onto the surface of 3D objects is laser direct structuring (LDS), in which an additive in the injection-moulded plastic is selectively activated by a laser. This forms a pattern that is subsequently metallised using electroless plating. LDS saw tremendous growth around a decade ago, and is used to manufacture 100s of millions of devices each year, about 75% of which are antennas.
However, despite its high patterning speed and widespread adoption, LDS has some weaknesses that leave space for alternative approaches to surface metallisation. First, it is a two-step process that can require sending parts elsewhere for plating, thus risking IP exposure. It has a minimum resolution in mass production of around 75um, thus limiting the line density, and can only be employed on moulded plastic. Most importantly, LDS only enables a single layer of metallisation, thus precluding crossovers and hence substantially restricting circuit complexity.
Given these limitations, other approaches to applying conductive traces to the surfaces of 3D objects are gaining ground. Extruding conductive paste, a viscous suspension comprising multiple conductive flakes, is already used for a small proportion of antennas, and is the approach of choice for systems that deposit entire circuits onto 3D surfaces.
Aerosol jetting is another metallisation approach, in which a relatively low viscosity, usually conductive ink, is atomised. This spray is then combined with an inert carrier gas and ejected from a nozzle. Aerosol jet has two notable advantages: it is capable of resolutions as fine as 10um, and the nozzle can be placed a few millimetres away from the surface thus facilitating patterning of 3D surfaces with complex surface geometries. The downsides are the cost of the complex atomisation and delivery process, and the requirement to re-optimise the process for different inks.
An advantage of digital deposition methods of the incumbent LDS technology is that dielectric materials can also be deposited within the same printing system, thereby enabling crossovers and hence much more complex circuits. Insulating and conductive adhesives can also be deposited, enabling SMD components to be mounted onto the surface.
In-mould electronics (IME) offers a commercially compelling proposition of integrating electronics into injection-moulded parts, reducing manufacturing complexity, lowering weight and enabling new form factors since rigid PCBs are no longer required. Furthermore, it relies on existing manufacturing techniques such as in-mould decoration and thermoforming, reducing the barriers to adoption. The basic principle is that a circuit is printed onto a thermoformable substrate, and SMD components mounted using conductive adhesives. The substrate is then thermoformed to the desired shape and infilled with injection-moulded plastic. IME is especially well suited to human-machine interfaces (HMIs) in both automotive interiors and the control panels of white goods, since decorative films can be used on the outer surface above capacitive touch sensors.
Whilst IME is likely to dominate HMI interfaces in the future due to the ease of manufacture and compatibility with established manufacturing techniques, it does bring technical challenges. Chief among these is developing conductive and dielectric materials that can withstand the temperature of the thermoforming process along with the heat and pressure of injection moulding. As such, materials suppliers are developing portfolios of materials aimed at IME, with conductive inks that can be deformed without cracking. Additional challenges include the development of electronic design software that can account for bending on circuits and developing SMD component attachment methods that are reliable under the moulding process.
Fully 3D-printed electronics
Least-developed technology is fully 3D-printed electronics, in which dielectric materials (usually thermoplastics) and conductive materials are sequentially deposited. Combined with placed SMD components, this results in a circuit, potentially with a complex multilayer structure embedded in a 3D plastic object. The core value proposition is that each object and embedded circuit can be manufactured to a different design without the expense of manufacturing masks and moulds each time.
Fully 3D-printed electronics are thus well suited to applications where a wide range of components need to be manufactured at short notice. Indeed, the US Army is currently trialling a ruggedised 3D printer to make replacement components in forward operating bases. The technology is also promising for applications where a customised shape and even functionality is important, for example, medical devices such as hearing aids and prosthetics. The ability of 3D-printed electronics to manufacture different components using the same equipment, and the associated decoupling of unit cost and volume, could also enable a transition to on-demand manufacturing.
The challenges for fully 3D-printed electronics are that manufacturing is fundamentally a much slower process than making parts via injection-moulding since each layer needs to be deposited sequentially. Whilst the printing process can be accelerated using multiple nozzles, it is best targeted at applications where the customisability offers a tangible advantage. Ensuring reliability is also a challenge since with embedded electronics post-hoc repairs are impossible; one strategy is using image analysis to check each layer and perform any repairs before the next layer is deposited.