Associate Professor Mazhar Ali and his research group at TU Delft have discovered one-way superconductivity without a magnetic field, which has been believed impossible since its discovery in 1911 – to this day. The discovery, published yesterday in Nature, uses 2D quantum materials and paves the way for superconducting computers. Superconductors can run electronics a hundred times faster, all without losing power (one-way superconductivity without magnetic field). Ali: “If the 20th century is the age of semiconductors, the 21st century will be the age of superconductors.”
In the 20th century, many scientists, including Nobel laureates, were confused by the nature of the superconductivity discovered in 1911 by the Dutch physicist Kamerlingh Onnes. In superconductors, current flows through the wire without resistance, which means that controlling this current or even blocking it is almost impossible – much less current flow, but in one way and not the other. That Dr. Heng Wu and Dr. Yaojia Wang, the lead researcher in Ali’s group who conducted this study (one-way superconductivity without magnetic field), were able to make the superconductivity one-way necessary for computer science: it could be compared to the invention of a special type of ice that gives you zero friction when skating in one. in a way but invincible friction on the other hand.
Superconductor: super fast, super green
The advantages of using superconductors in electronics are twofold. Superconductors can speed up electronics a hundred times, and implementing superconductors into our daily lives can make IT greener: if you spin a superconducting wire from here to the moon, it will conduct energy without loss. Thus, using superconductors instead of conventional semiconductors could save up to 10% of all Western energy supplies, according to the NWO.
(Im) possibility of superconductivity application
In the 20th century and beyond, no one was able to solve the barrier of superconducting electron formation but to go in one direction, which is a basic feature required for computers and other modern electronics (see, for example, diodes that also go in one direction). In normal conduction, electrons fly around as separate particles; in superconductors, they work in pairs, without loss of electricity. In the 1970s, IBM researchers tried the idea of superconducting computers, but had to stop their efforts: in their documents, IBM said that without non-reciprocal superconductivity, it would be impossible for a computer to run on superconductors (one-way superconductivity without magnetic field).
Interview with Associate Professor Mazhar Ali on paper on the creation of proof of the concept “The Fieldless Josephson Diode in a van der Waals Heterostructure”:
Question: Why, when one-way works in normal semiconductor mode, is there one-way superconductivity that didn’t work before?
A: Mazhar Ali: “The electrical conductivity of semiconductors like Si can be unidirectional thanks to a fixed internal electric dipole, so a network is built on the potential they can. The textbook example is called” pn junction “; we don’t beat two semiconductors: one has more electrons (-) and the other has more holes (+). This can break the symmetry and can result in “one-sided” properties because, for example, forward vs. backward is no longer the same. There is a difference between going in the same direction as the dipole and against it; it’s like swimming with a river or swimming in a river. ”
“Superconductors are absolutely not analogous to this one-way idea without a magnetic field; because they are more related to metals (ie conductors, as the name suggests) than to semiconductors, which always lead in two directions and have no constructed potential. Similarly, Josephson Junctions (JJ), which are sandwiches of two non-superconductor superconductors, also have classic superconductor barrier materials and no special symmetry-breaking mechanism that leads to a difference between “forward” and “backward”.
Question: How did you do the first impossible?
A: “This is actually the result of one of my group’s basic research studies. In what we call “Quantum Material Josephson Junctions” (QMJJ), we replace the classical JJ barrier material with a quantum barrier material, where the intrinsic properties of the quantum material can modulate the bond between two superconductors in new ways. The Josephson diode is an example: we use the quantum material Nb3Br8, which is a 2D material such as graphene theorized for the placement of an electric dipole network, as our quantum material barrier for its selection and placement between two superconductors.
“We were able to remove several atomic layers on this Nb3Br8 and create a much thinner sandwich – but a few atomic layers thick – that was needed to make Josephson diodes and could not be made from normal 3D materials. Nb3Br8, part of a group of new quantum materials developed by our colleagues, Professor Tyrel McQueen and his team at Johns Hopkins University in the United States, and an important report on our first awareness of the Josephson diode.
Question: What does this discovery mean in terms of impact and applications?
A: “Many technologies are based on older versions of the JJ superconductor, such as MRI technology. Quantum calculations are now based on Josephson Junctions. Technology that was previously only possible with semiconductors can now be made with a superconductor with this building block. This includes faster computers, such as terahertz computers, which are 300 to 400 times faster than the computers we use today. It can affect all kinds of social and technological applications. If the 20th century is a century of semiconductors, then the 21st century will be a century of superconductors. ”
“The first direction we need to do for commercial use is to increase the operating temperature. Here we use a simple superconductor that limits the operating temperature. Now we want to work with so-called “high Tc superconductors” and see if we can operate Josephson diodes at temperatures above 77 K because that will allow cooling with liquid nitrogen. The second thing that needs to be done is scale production. Although we have proven well that it works with nanodevices, we have done very badly. The next step is to explore how you can scale the output of millions of Josephson diodes on a single chip.
Question: How confident are you in your case?
A: “There are many steps that all scientists must take to maintain the robustness of science. The first is to ensure that their results are repeatable. In this case, we made different devices from the left with different sets of materials, and we always found the same properties, even though different people measured on different machines in different countries. . This tells us that the Josephson diode is the result of our combination of materials and is not an unrealistic result of impurities, geometry, mechanical or error or user interpretation.
We also did experiments with a “smoking gun”, which further reduced the possibility of translation. In this case, to make sure we have a superconducting diode effect, we actually tried to “switch the diode; as we did. in the reverse direction and showed that we never measured resistance (superconductivity) in one direction and actual resistance (normal conductivity) in the other direction.
“We also measured this effect when applying magnetic fields of different sizes and showed that the effect was evident in the applied field 0 and disappeared in the applied field. A zero field is an important point for technological applications. This is because magnetic fields at the nanometer scale are very difficult to control and limit, so for practical applications, it is normally desirable to work without the need for local magnetic fields.
Q: Is it realistic for ordinary computers (or even KNMI and IBM supercomputers) to use superconductivity?
A: Yes! Not for people at home, but for server farms or supercomputers, it’s good to run it. Central computing is truly the way the world works today. Each intensive calculation is performed in centralized facilities, where localization brings a great advantage in the field of energy management, heat management, etc. It is real-time when the challenges mentioned in the next question will be overcome, that it will turn into centralized supercomputers!
More information
On May 18-19, Associate Professor Mazhar Ali and his colleagues’ prof. Valla Fatemi (Cornell University) and Dr. Heng Wu (TU Delft) hosted “Superconducting Diode Effects Wo
Reference: “The field-free Josephson diode in a van der Waals heterostructure” 27 April 2022, Nature. DOI: 10.1038/s41586-022-04504-8
