Superconducting quantum circuits are the core technology behind the most sensitive magnetometers. An analogous device has now been implemented using a gas of ultracold atoms, with possible applications for rotation sensing.
When a magnetic field needs to be measured with the utmost precision, a superconducting quantum interference device (SQUID) is the instrument of choice1. Its exquisite sensitivity derives directly from a macroscopic manifestation of quantum mechanics, making it an archetype of quantum engineering. Reporting in Physical Review Letters, Ryu and colleagues2 demonstrate an analogue of a SQUID using an ultracold gas of neutral atoms known as a Bose–Einstein condensate. Here, the analogue to the magnetic field is a physical rotation, so the atomic device could prove useful for rotation sensing and vehicle navigation. More broadly, it strengthens the correspondence between atomic and solid-state systems. Because atomic systems are better understood and more easily controlled than their solid-state counterparts, atoms might eventually serve as a design platform for complex solid-state quantum devices.
A conventional SQUID is a small ring of superconducting material cut in half by two non-superconducting barriers. Wire leads connected to each side of the device allow a current to pass through it (Fig. 1a). Within each of the superconducting regions, electrons act like a coherent quantum wave. Because the current passing through the SQUID can take either path around the ring, the two corresponding waves can interfere: they can add constructively with the peaks of the waves lined up, or cancel destructively with the peaks of one wave aligned to the troughs of the other. The total current through the ring depends sensitively on the type of interference. For charged particles such as electrons, the way that the waves align is set largely by the magnetic field threading the ring, which makes the SQUID a good magnetometer.
In the atomic analogue demonstrated by Ryu and colleagues, the superconducting electrons are replaced by a Bose–Einstein condensate consisting of a few thousand rubidium atoms at nanokelvin temperature, isolated in an ultra-high-vacuum chamber3. Like the electrons, the atoms in a Bose–Einstein condensate act as a wave, allowing similar physics to be probed. Here, the atoms are held in a ring-shaped trap that has two small potential-energy barriers through which the atoms can tunnel (Fig. 1b). The authors created the ring trap using a technique known as a painted potential. For this, a laser beam is rapidly scanned across the trapping region and selectively turned on so as to illuminate only a ring-shaped area. The atoms are attracted to the laser light and confined within the ring. The tunnelling areas are produced by reducing the light intensity at two spots on the ring.
Because the atoms are neutral, the condensate version is not especially sensitive to magnetic fields. However, if the whole apparatus rotates, then the atoms will experience the Coriolis force, which twists the path of any object moving on a rotating platform. For example, on the Earth, the Coriolis force causes the circulating air flow of hurricanes and cyclones. It affects the atomic waves much like a magnetic field affects electrons. By measuring how the atoms move through the ring, even a tiny Coriolis force can be detected, making the system useful for sensing rotation4.
Ryu and colleagues' work builds on previous demonstrations (see, for example, ref. 4) of atomic systems similar to a SQUID, but for the first time uses the complete geometry of a ring with two barriers. The observed behaviour of the authors' atoms is in good accord with the phenomenological model used to describe superconducting devices. Indeed, for the atomic case, the expected behaviour can be derived nearly from first principles, so the system is on firm theoretical ground.
Rotation sensors are useful for vehicle navigation and other geophysical applications, and the atomic SQUID shows promise for advancing these technologies. A greater impact, however, may derive from the demonstration of how atomic systems can replicate solid-state devices. Although solid-state circuits have the practical benefit of not requiring lasers and vacuum chambers, developing a new device involves painstaking fabrication and characterization work. By contrast, the size and shape of an atom trap can be modified simply by reprogramming the behaviour of the laser beam. The atom system could thus serve as a design tool for complicated circuits, in which the geometry could be developed and optimized before being applied to superconductors. Although ordinary computer simulation can serve a similar purpose, a physical device can easily become too complex for simulation to be practical. Such complexity is common in quantum systems in which particle interactions are important and non-trivial, including high-temperature superconductors and, perhaps one day, quantum computers.
The idea that ultracold atoms could be used to simulate and explain solid-state systems has been a driving force in the atomic-physics community since the first observations of Bose–Einstein condensation5. Ryu and colleagues' demonstration that a useful device such as a SQUID can be implemented with atoms is a milestone in this effort. Nonetheless, substantial challenges remain. An immediate issue is the difficulty of using atomic systems to model macroscopic currents: the number of atoms in a condensate is relatively small, so there is no simple way to create a large current. The authors sidestep this problem by measuring a small current flowing through a barrier, rather than a large current passing through the ring as a whole (compare Fig. 1a and b). Although this set-up can be used for rotation measurements, it does not reflect the actual operation of a superconducting SQUID. A larger question is how well the correspondence between atoms and solid-state systems will hold up as the system's complexity grows. Until the systems become too complicated for computer simulation, the utility of the atomic experiments as a design platform for solid-state systems will probably be limited. Meeting the challenges involved will not be easy, but the steady progress in this field exemplified by Ryu and colleagues' achievement is encouraging.
Clarke, J. & Braginski, A. I. (eds) The SQUID Handbook Vol. 1 (Wiley-VCH, 2004).
Ryu, C., Blackburn, P. W., Blinova, A. A. & Boshier, M. G. Phys. Rev. Lett. 111, 205301 (2013).
Dalfovo, F., Giorgini, S., Pitaevskii, L. P. & Stringari, S. Rev. Mod. Phys. 71, 463–512 (1999).
Wright, K. C., Blakestad, R. B., Lobb, C. J., Phillips, W. D. & Campbell, G. K. Phys. Rev. Lett. 110, 025302 (2013).
Bloch, I., Dalibard, J. & Zwerger, W. Rev. Mod. Phys. 80, 885–964 (2008).
About this article
Physical Review A (2020)
Physical Review A (2020)
Physics Letters A (2016)