A substitute for Josephson Junctions

[pico61]

Interesting article

https://www.allaboutcircuits.com/news/from-unearthed-time-capsule-mit-discovers-replacement-josephson-junctions/

[octillion]

The cryotron is far from a "replacement for Josephson junctions".  Maybe in a few specific applications, but quantum computing is not one of them.

In superconducting quantum bits (qubits) the Josephson junction does not operate as a switch.  Rather, it is used as a non-linear inductor (in most cases; sometimes they are also used as a linear inductor, typically as an array).  This non-linearity is important to allow the different energy levels to be distinguished.  In a harmonic (linear) oscillator, like an ordinary LC, each photon of energy added to the system is done so at the same frequency.  In an oscillator made with a Josephson junction and capacitor, the transition from 0 to 1 photons takes more energy higher energy (higher frequency), than from 1 to 2 photons, and so on.  This allows specific energy levels to be controlled and read out.

Other uses for Josephson junctions in quantum computing are the formation of SQUID loops (two or more junctions in a loop), which allows for inductances that may be tuned with magnetic flux.  SQUIDs can be used to make tunable qubits, couplers, and filters.  SQUIDs are also used as extremely sensitive detectors for magnetic fields.

Josephson junctions are also used as non-linear inductors to form a number of (near) quantum limited amplifiers.  These are amplifiers which add the minimum amount of noise to an amplified signal, as dictated by physics.

As a linear inductor, arrays of Josephson junctions can form large inductances with corresponding self resonant frequencies much higher than would be possible in magnetic inductors.  Most of the inductance is "kinetic" rather than magnetic.

Josephson junctions are used as a sort of switch in single flux quantum (SFQ) circuits, but the cryotron is not a drop-in replacement here.

The cryotron is also unusable for things like voltage standards.

The Josephson junction isn't going anywhere.

[martinr33]

The device could have been useful to build the sub-1K control logic needed to build quantum computers at useful scale (quantum computing with superconducting electronics is more of an exercise in eror correction than anything else, with estimates ranging from 1,000 to 10,000 error correcting qubits per logical qubit - once we get good enough qubits).

However, Microsoft just announced CMOS logic that can work at 100mK temperatures, so the device is irrelevant. Also, as it depends on a heating effect, it is potentially problematic for a QC.

https://www.nature.com/articles/s41928-020-00528-y.epdf?sharing_token=iDICFbqAtEIiRgzzt1RzT9RgN0jAjWel9jnR3ZoTv0MdWLUhfzcY6l47T8lROo13YpOMcyNoDEftKCkPqYg9LpxD66Yj_Ho_TOF52Jt0my7JUBxy_Pjdko40Mh8LqA5iN5aO74xUb2uFA6COW38nQweN6oX5xHJ3fAGT1AfgqCKqe7aJAp4FuXGz3uQQgiXwL-al1qFnoyY_DDsr0oNVwqmaq_iXC9L5Re78jWlBXqE%3D&tracking_referrer=www.microsoft.com