Almost a century after Heike Kamerlingh Onnes first discovered superconductivity, the factors that determine whether a system will be superconducting and at what temperature remain hard to pin down. However, advances in nanotechnology have given some good pointers where to look, as well as providing promising systems for exploiting superconductivity in real-world applications.
The fundamental requirement for superconductivity is the coupling of fermionic electrons into Cooper pairs. Theory paints a neat picture of how the resulting bosonic behaviour allows occupation of the same energy levels and leads to a host of exotic behaviour – zero electrical resistance and the expulsion of magnetic flux lines so that superconducting objects levitate on magnets, to name a few. Where the picture grows fuzzy is extrapolating from there what specific aspects a material system needs to become superconducting at a given temperature. While design principles to fabricate a room-temperature superconductor remain elusive, a lot has been learnt in the chase, bringing applications of superconductors in a range of sectors from imaging, testing and quantum cryptography ever closer.
Among the material systems where unusual electronic behaviour akin to Cooper pairing might be likely is the interface between perovskite oxides – in particular, LaAlO3 and SrTiO3 – where there is a discontinuity in the polarity of the crystalline lattice. Following the initial discovery of a highly mobile “2D electron gas” at the interface in 2004, Jochen Mannhart and colleagues then identified superconducting properties at the interface in a layer limited to just 20 nm in 2007. The transition temperature was a chilly 200 millikelvin, and the exact origins of the effect were unclear, but oxide interfaces remain a hotbed for exploring electronic and spintronic behaviour.
Since then several 2D structures have revealed superconducting behaviour where it does not exist in the bulk, an example being “grey” tin. The form of tin usually considered most useful is “white” tin, which has a conventional metal crystallographic structure, and was among the first superconducting materials to attract study. However, at low temperatures white tin will gradually transform into grey tin, which has a diamond cubic structure and is sometimes described as “tin pest”. To their surprise, Qi-Kun Xue, Ding Zhang and colleagues at Tsinghua University in China found that when they reduced the dimensions of tin to 2D stanene of just 2-20 layers, they could observe superconducting properties in grey tin too. Going even thinner to monolayers resulted in insulating properties.
“What we found is that the grey tin can be scientifically quite interesting,” Zhang told nanotechweb.org. As well as the fundamental science the discovery opens up, it also poses the opportunity to produce circuits from all one material, with superconducting wires of few layer stanene separated by insulating monolayers.
Triplet vs singlet superconductors
Graphene – the mother of today’s explosion of 2D material research – has also demonstrated superconducting properties. In most superconductors electrons pair up with opposite spins to give a singlet spin state with isotropic s-wave symmetry following the Bardeen, Cooper and Schrieffer (BCS) theory of superconductivity. However other types of superconductivity are possible where the spins are aligned in parallel in a triplet state with anisotropic p-wave symmetry or chiral d-wave symmetry.
According to theory p-wave triplet superconductors are very sensitive to defects so there is no chance of observing it unless the crystals have very high purity. In addition, the superconducting transition temperature hovers at a frosty 1.5 Kelvin, below even the reach of liquid helium refrigerants. However, when it comes to graphene theory gives a more positive outlook for observing p-wave superconductivity. As well as the possibility of doping graphene to achieve superconducting effects, placing single-layer graphene on a superconductor should enhance intrinsic electron pairing with p– or d-wave symmetry to the point that a superconducting state is triggered in the graphene at temperatures above 4.2 K, i.e. the temperature of liquid helium. This kind of intrinsic superconductivity by proximity was observed with s-wave symmetry on the s-wave superconductor rhenium in 2013, and at the beginning of 2017 researchers in the UK, Israel and Norway reported superconductivity by proximity with p-wave symmetry on the electron-doped cuprate superconductor Pr2−xCexCuO4 (PCCO) at 4 K.
Moving down the dimensions
While electronics may seem the obvious sector for exploiting zero-resistance phenomena, as electronics becomes increasingly preoccupied with reducing feature sizes the next question is how low can you go before a wire loses its superconducting effects? In 2000 A. Bezryadin, C. N. Lau and M. Tinkham at Harvard in the US tackled the issue using measurements of ultrathin superconducting nanowires made from carbon nanotubes coated in superconducting Mo–Ge alloy. At high temperatures thermal excitations give rise to phase slips, which disrupts the processes that cause superconductivity. Bezryadin, Lau and Tinkham argued that quantum tunnelling of phase slips could also effectively localize Cooper pairs, suppressing superconductivity if the wire were thin enough for the normal-state resistance of the wire to be greater than the quantum resistance of Cooper pairs.
Nanowires have proven to be fruitful for exploring some of the exotic phenomena found alongside superconductivity too. Majorana fermions are particles that are their own antiparticle. Predicted by Ettore Majorana in 1937, they could be useful in quantum computing but remained unobserved for the next 75 years. In 2012 a team of researchers led by Leo Kouwenhoven at Delft University of Technology and Eindhoven University of Technology identified what they believed were Majorana fermions in a semiconductor wire wrapped in a superconductor. Although some had suggested that the signature of the Majorana fermions observed could be attributed to scattering, recent experiments by the same group with atomically smooth interfaces have ruled this out, providing strong indication that the systems do harbour Majorana fermions.
As well as carbon nanotubes, superconducting properties have been conferred on other nanostructures such as anodized alumina arrays by coating in superconducting materials. A popular choice of coating is YBa2Cu3Ox. Discovered by Paul Chu at the University of Houston in 1987, the superconducting transition temperature of YBa2Cu3Ox is a balmy 93 Kelvin. Other cuprates have demonstrated “high-temperature” superconducting properties since. While the transition temperature of YBa2Cu3Ox is still way below freezing, what made Chu’s discovery so significant is that above 77 K liquid nitrogen suffices as a coolant, and this is much easier to handle than liquid helium.
As well as coating nanotemplates, researchers have also produced YBa2Cu3Ox nanostructures through micromachinining and electrospinning, but the final product usually requires heat treatment for superconducting properties to appear. While the initial powder provides a degree of versatility for coating arbitrary structures, the heat treatment makes the material brittle. William Rieken, Atit Bhargava and colleagues at Nara Institute of Science and Technology showed that they could produce a powder of YBa2Cu3Ox nanorods through a solution processing approach, which is not only simple but does not require any further heat treatments. As a result they could paint the powder on structures without leaving them brittle. “Our materials don’t represent a new material, but a new way of thinking about superconductors and opens a route to designer superconductors, tailoring them for specific uses,” says William Rieken.
Josephson junctions and photon detection
Many of the applications of superconductivity – from quantum key distribution in cryptography, long range 3D infrared depth imaging, and integrated circuit testing, to fundamental tests of quantum mechanics – rely on exploiting the phenomenon for photon detection. Superconducting wires are very sensitive to incident photons as these can break up the Cooper pairs thus suppressing superconductivity. Devices based on this effect have successfully detected single photons in the visible and infrared regions of the spectrum, but had been less sensitive to lower energy photons. A graphene sheet contacted at both ends by a superconductor has demonstrated the ability to detect microwave photons as well, an important region of the electromagnetic spectrum for astronomers to detect cosmic background radiation and determine how galaxies form.
The basic structure for these devices – a non-superconductor or insulator with superconducting contacts at either end – is described as a Josephson junction. The material between the superconductors is a weak link, which provides a channel for a supercurrent – a current that could in theory run forever with no applied voltage. Observations of the effect had been dismissed as breaches in the insulator between the superconducting contacts until Brian David Josephson explained the phenomenon with predictions of the supercurrent based on the mathematical relationship between current and voltage in 1962.
As well as photon detectors, researchers at NIST in the US have used Josephson junctions to build artificial synapses. Connections in the brain respond to the history of signals that have passed through them giving learning functions. Neuromorphic electronics researchers are keen to emulate synaptic connections to produce electronics with added functionality. Schneider and colleagues built Josephson junctions consisting of two layers of superconducting materials separated by an insulating silicon matrix embedded with nanoscale clusters of manganese. When an applied electric current exceeds a critical level, voltage spikes are produced that mimic the action potentials – signal spikes – that neurons generate.
“These artificial synapses are in fact better than their biological counterparts,” Schneider told nanotechweb.org. “They can fire much faster – 1 billion times per second compared to a brain cell’s 50 times per second using just one ten-thousandth as much energy.”
Superconducting devices out and about
Despite the great potential, superconducting devices are often confined to the lab because of all the bulky cooling equipment required. Progress has been made here too. Robert Hadfield at the University of Glasgow and colleagues in STFC Rutherford Appleton Laboratory in the UK, Single Quantum B. V. in the Netherlands and KTH Royal Institute of Technology in Sweden miniaturized a platform for superconducting photon detectors that operate at 4 K. As the researchers point out in their report, “Although the need for liquid cryogens has been eliminated by the use of practical closed-cycle cryocoolers, such bulky, power hungry systems are not truly capable of mobile operation.”
The researchers go on to describe how they have developed a fully closed-cycle miniaturized cooling platform based on Stirling and Joule–Thomson (J–T) cycles that can reach a base temperature of 4.2 K and is the size of a desktop printer. The design has already been launched aboard the Ariane 5 rocket in 2009 as part of the Planck mission, where it operated “flawlessly for the entire mission duration of nearly 4.5 years (>39 k h)”. This cooler design has now been adapted specifically to house for example the superconducting nanowire single-photon detectors.
Take two Josephson junctions in parallel and you get a superconducting quantum interference device (SQUID) – a highly sensitive magnetic flux detector. In the SQUID, current splits through each of the Josephson junctions but the presence of a magnetic flux will generate a screening current to cancel the flux. This screening current loops round both Josephson junctions adding to the current in one arm and subtracting from it in the other. As the flux in the loop increases, the screening current will change direction to either increase or decrease the flux to an exact integer quanta of flux.
A key application of SQUIDs is in biosensing. Their high sensitivity can detect iron concentrations in body organs for magnetoencephalography, magnetocardiology, fetal magnetocardiology and biomagnetic liver susceptometry. They can also detect injected magnetic nanoparticles that are functionalized to target specific proteins, cells or antigens to track, image and diagnose disease. Keiji Enpuku, Yuya Tsujita, Kota Nakamura, Teruyoshi Sasayama and Takashi Yoshida at Kyushu University in Japan reviewed recent progress in magnetic biosensing techniques using SQUIDs in the Superconductor Science and Technology focus collection on SQUIDs in biomagnetism.
Unleashing the full potency of this biosensing tool means detecting magnetic-nanoparticles inside the human body with SQUIDs, and this raises additional challenges. “For in vivo human measurements the system must be non-invasive and conform to the anatomic restrictions requiring sensitive detectors and dedicated setups,” explain Oswaldo Baffa and colleagues at Universidade de São Paulo and Universidade Federal de Goiás in Brazil in a report in the same focus collection. They go on to describe a system based on an a.c. biosusceptometer to induce magnetism in manganese ferrite-based magnetic nanoparticles surface-coated with citric acid, which they detect using a second-order axial gradiometer coupled to a radio frequency SQUID. They achieved limits of detection of 8–11 × 109 at distances of 1.1–2.5 cm. While noting a number of possible improvements to the system they conclude, “The results found show that the bio-susceptometric technique has good sensitivity and temporal resolution since measurements take around 30 s, and might have interesting applications in the real-time in vivo detection of nanoparticles after systemic injection.”
Although studies of nanostructures have made huge contributions to advancing understanding of superconductivity, many aspects of the phenomenon remain a marvellous mystery and a great stimulant for further research. Equally as fascinating again is the creativity in applying superconductors in such a diverse array of fields, and here without a doubt the tiny world of nanostructures has had a huge impact.