Industrial production processes, by and large, rely on robotic assembly lines that place, package, and connect a variety of disparate components. While the manufacturing world is dominated by robots, there are applications where the established processes of serial ‘pick and place’ and manipulation of single objects reach scaling limits in terms of throughput, alignment precision, and the minimal component dimension they can handle effectively.


By contrast, the emerging methods of engineered self-assembly are massively parallel and have the potential to overcome these scaling limitations.”So far researchers have focused on small area proof-of-concept demonstrations testing new self-assembly concepts – for example testing a different driving force, agitation, or component handling mechanism – inside a small container to define the environment where self-assembly takes place,” Heiko Jacobs, a professor at Technische Universität Ilmenau and Chair of the university’s Nanotechnology Group, tells Nanowerk. “Over the past 10 years, tremendous knowledge has been gained.

However for many observers it leaves out the proof that any of the processes can be scaled to enable high volume production of real devices and products as promised at the outset.””Personally, I am convinced that this is possible and we are now at a stage where we have gained sufficient experience that will allow us to put the knowledge together to begin building ‘self-assembly machines’ which provides a deep motivation to go forward,” he continues.In a paper in the June 27, 2014 online edition of Advanced Materials (“A First Implementation of an Automated Reel-to-Reel Fluidic Self-Assembly Machine”), Jacobs’s team together with researchers from the University of Minnesota, describe the first example of a device to assemble and distribute LEDs over large areas for solid state lighting applications.It provides the blueprints and operational parameters of a reel-to-reel (RTR) fluidic self-assembly platform to assemble and electrically connect semiconductor chips with a yield clearly exceeding a 99% benchmark set by robotic pick and place machines.

“I am convinced that processes of self-assembly will eventually be used in the manufacturing industry of the future,” says Jacobs. “Today it might appear as science-fiction to some critical observers. However, researchers worldwide including my group continue to work out the details.”