“If the 20th century was the century of semiconductors, the 21st can become the century of the superconductor.”
That is how professor Mazhar Ali, head of a Delft University of Technology research team, puts their scientific breakthrough. Published in Nature, the discovery concerns superconductors, which are a class of materials that offers loss-free electrical conductivity. The research demonstrates that superconductors too can be made to only carry electrical current in a single direction — something that was previously thought impossible. The discovery may ultimately allow for electronics to become not only more efficient but also more performant by factors of hundreds, perhaps even unlocking the terahertz era, no less.
Superconductors are materials that carry current without any resistance. This means (all things being equal) that signals don't lose their integrity and there is no energy loss in the form of heat - irrespective of the actual physical distance separating the signal's origin and destination points - even if you were to measure it in Astronomical Units (AUs).
The problem with superconducting materials is that due to their characteristics, it becomes impossible to control the flow of electrons. Since there's no resistance throughout the length of the material, there's also no easy way to add it, and the current flows forwards as well as backward. The flow of electrons can usually be controlled through deployment of electromagnetic fields. But these are extremely hard to design, control and deploy at the nanometer scale of current manufacturing processes, which would throw a wrench into costs and scalability.
The scientists solved this by replacing classical components of Josephson Junctions, which are used to break material symmetry. To do so, they deployed a novel quantum material - which are essentially classical materials that have been manipulated (or shaved) towards their minimal possible size (usually only as thick as the molecules themselves). The scientists call their new design “Quantum Material Josephson Junctions” and have based it on the quantum material Nb3Br8.
Like graphene, Nb3Br8 is a 2D material that the research team theorized could be host to a net electric dipole - a particularly useful structure that allows for the superconducting symmetry to be broken. Due to this material, the researchers could now control the flow of current on their superconducting material, ordering the electron chaos towards having a "forward" path, where electrons can happily skim through the superconductor without any resistance, and a "backward" direction, the one scientists didn't want their free-moving electrons to go towards.
Imagine having a piece of velvet in your hands. If you trace a finger in the direction of the fibers, you're met with little to no perceivable resistance. But if you go against the fibers, the soft feel is largely gone - there's a difference between directions.
The Dutch Research Council (NWO) estimates that the superconducting efficiency advantage alone would allow for a 10% reduction in global western energy consumption compared to traditional semiconductors. At the same time, chipmakers would jump at the opportunity to reduce heat and wasted electrical power on their designs, as they always do when economically feasible.
But another element to this research is the type of performance increase superconductors can unlock compared to their traditional semiconductor counterparts (which, you guessed it, only deliver a part of the original current towards the destination due to the material's natural electrical resistance). According to Mahzar Ali, this new scientific breakthrough could pave the way for a transformative evolution in the chip manufacturing industry. Technology that has only been achieved with semiconductors can now potentially be made with superconductors - delivering up to 300 to 400 times the operating frequencies of classical materials. Ultimately, he said, the possibility is real for all sorts of societal and technological applications.
The researchers don't expect consumers, or even the most diehard PC enthusiasts, to have access to superconducting-based computing any time soon. While this may change with time and further materials research, that doesn't mean the breakthrough's impact is lessened. The researchers expect data centers and supercomputers to be the first to adopt superconductor designs. But with the increasing push towards cloud services, like Microsoft's Windows 365 Cloud OS and cloud gaming, there's a real impact on both environmental sustainability and the end-user experience.
As in all research — and especially the revolutionary kind — there is still a lot of work to be done before this translates into actual products. For now, the scientists are focused on reducing the operational temperature of their design. The aim is to achieve so-called “High Tc Superconductors” which would allow the Josephson diodes to operate at temperatures of 77 Kelvin (-192 ºC), high enough for cooling to be taken care of by liquid nitrogen, already used to enable the world's most impressive overclocking records on some of the Best CPUs and Best GPUs.
Another element to tackle is, naturally, production scaling. While the researchers' work proves that superconductors can actually be leveraged in nanodevices, there's a world of difference between manufacturing for academic purposes and the high-stakes yield game of foundries. If the researchers can devise a way for production to scale up towards millions of Josephson diodes in a single nanoscale chip, all bets are off for the terahertz race. That's enough to make even my old Netburst Pentium 4 huddle against a corner.