In 1911 a Dutch physicist, Heike Kammerlingh Onnes, was experimenting with mercury, electricity and liquid helium. He was very surprised to learn that as the temperature of the mercury approached absolute zero, all electrical resistance disappeared. Electricity flowed through the (now solid) mercury without losing any energy whatsoever. Also, the mercury did not heat up as a result of the current flowing through it.
Since then this behaviour, known as superconductivity, has been found in 25 chemical elements and in numerous other ceramics, alloys, organic crystals and even polymers. The prospect of materials that do not lose energy when conducting electricity promises a world of much greater energy efficiency. It also allows novel technologies, formerly off-limits, to flourish. Incredibly strong electromagnets carrying huge, infinitely circulating currents could make levitating trains a reality as opposed to a sci-fi writer's pipe dream. Rather more prosaic technological problems have been solved by ingenious applications of the unique properties of these materials.
Resistance is the process by which electrons flowing in a solid lose energy through collisions with the crystal lattice. As nearly all electrical conductors are solids, and virtually all of these conduct current by mobilising a sea of electrons carrying negative electrical charge, resistance is an unavoidable phenomenon. As far as electronic and electrical engineers are concerned, resistance is far from 'useless'; it allows electrical energy to be converted into heat, and helps moderate the flow of electricity in complex circuitry. At other times, it's a bane; power cables need to carry incredibly high voltages in order to minimise resistive losses. Electromagnets need to be constantly 'topped up' with electrical power just to maintain a magnetic field, and so on. All of this 'leaking' electrical energy gets wasted as low-grade heat - as anyone who works in an office full of computers can testify.
How Electrons Move in Ordinary Conductors
Subatomic particles, of which electrons are a typical example, come in two basic flavours: fermions and bosons. This distinction is drawn on the basis of their spin. A spinning charge is equivalent to a circulating current and circulating currents set up magnetic fields. Spin is, therefore, equivalent to the particle's in-built magnet field. The amount of spin a particle has is not arbitrary - fermions have spin equal to an odd multiple of one half, and bosons have either zero or integral spin (1,2,3... etc).
A consequence of being either a fermion or a boson is the way it affects how the particle interacts with its peers. Electrons are fermions and this has a profound effect upon the way they move in solids. In an electrical conductor electrons are free to move virtually anywhere, and so much interacting goes on. Fermions interact in such a way that any two in close proximity force each other to take up different energy levels - one loses and the other gains energy in the process1. Add more fermions, and these too take up their own unique energy levels. There are billions and billions of electrons jostling for elbow room and each takes up its own distinct place in the energetic pecking order2.
At any one time the electrons in a conductor are normally moving in all directions. If one then applies a voltage, more electrons want to move in one direction than in another. Cramming even more of these fermions into a volume of space means that they have to adopt higher energy levels and, in so doing, leave behind gaps in other levels3. The stage is now set for these high-energy electrons to collide with obstacles, lose energy (as heat), drop back to the vacant energy levels, pick up energy and move back up to the high energy levels, and so on. Electrical energy gets dissipated as heat, and the material exhibits electrical resistance.
How Electrons Move in Superconductors
Clearly something is occuring in superconductors to make the electrons behave differently at low temperatures. It used to be thought that there were two kinds of electron 'fluid' in a superconductor: the superconducting kind predominating at low temperatures and the ordinary, resistive kind predominating at normal temperatures.
In fact what happens to the electrons is that at low temperatures they stop being fermions and start being bosons. Bosons are capable of all happily occupying the same energy level within the same volume. So when one applies an electrical field, they don't need to increase energy levels in order to start moving. If they don't increase their energy, they can not lose it again and the current flows without resistance.
When superconductors are cooled to the point where they lose resistance, otherwise known as the 'critical temperature', all the electrons sink down in energy to a sort of 'sump' level where they become bosons and share exactly the same energy level. This strange behaviour occurs because at low temperatures the electrons forget about the fact that they have like charges (and should therefore repel each other) and start pairing up4. As one electron moves through the crystal lattice it deforms it slightly and the deformation attracts another electron, rather like a racing car slipstreaming another. The spins on these two electron 'quasi-particles' now add up to one or zero (depending upon whether the magnetic fields are aligned or in opposition), and they effectively become bosons. Instead of the lattice disrupting the progress of electrons, it now aids it and the pairs move through it without resistance.
As the temperature of the substance increases, the lattice vibrations swamp this subtle attractive effect between electrons, and the pairs break up. How high a temperature this occurs at, is dependent upon the atoms making up the crystal lattice. To extend the racing car analogy, the vibrations act like a ferocious side wind, swamping any advantage that slipstreaming confers. Placing the material in a too-strong magnetic field also disrupts superconductivity. The pairs of electrons have spins that point in opposite directions and are therefore separately repelled and attracted by the field. Once the field gets too strong, it pulls the pairs apart and the material stops superconducting.
Types of Superconductors
Generally, the bigger the atoms in the crystal lattice, the 'squishier' and more deformable they become. Thus the attractive effect becomes stronger. As heavy elements make better superconductors than light ones, they have a higher critical temperature. Alloys of certain metals also make better superconductors than elements. Niobium makes excellent superconducting alloys. For example, many superconducting electromagnets are wound from niobium alloy wire.
All superconductors suffer from the same drawback - they need to be cooled in order to show superconductivity. Up until 1986 coaxing materials to behave in this exotic fashion came at a substantial price. The only coolant that would achieve the necessary temperatures was liquid helium, a fractious and expensive substance boiling at 4K, tricky to handle and even trickier to make. Other cheaper (and less troublesome) coolants such as liquid nitrogen, which boiled at 77K, could not achieve the temperature necessary.
Always uppermost in the minds of the scientists who made these materials, was the need to increase the critical temperature of superconductors. Progress was slug-like for 75 years after Onnes' first discovery. The highest temperature then recorded was 23.2K, in a niobium-germanium alloy, and the prospect of physicists producing anything that superconducted at liquid nitrogen temperatures was deemed fanciful.
In 1986, Georg Bednorz and Alex Muller produced a dense black ceramic, based upon copper, oxygen and lanthanum, that superconducted at 35K. This was, comparatively speaking, a huge advance which hinted at further dramatic increases in critical temperature. Within a year, the similar yttrium barium copper oxide (YBCO) had been discovered by Paul Chu and his colleagues. This superconducted at a then-unimaginable 93K, 16K above the boiling point of liquid nitrogen.
The record now stands at about 130K. This is about -140°C. However the phenomenon was instantly christened 'high temperature superconductivity', such had been the dampened expectations of the field's practitioners up to that point5. Nobody really knows how the electrons in high-temperature superconductors pair up but the mechanism appears to be radically stronger than that operating in conventional superconductors. These ceramics are very easy to make - school pottery departments (which deal with ceramics on a daily basis) often have the right equipment, and the raw materials can be bought from chemical suppliers.
Apart from the obvious drop in electrical resistance to zero, superconductors exhibit some other remarkable properties. Foremost of these is the Meissner effect. Placing a superconductor in a magnetic field causes an electrical current to circulate in the material. This generates its own field that exactly opposes that applied. The superconductor behaves like a perfect magnetic mirror. Superconductors can therefore be used as perfect magnetic shields. High temperature superconductors lend themselves to a dramatic demonstration of this phenomenon. If one places a tiny magnet on top of a slab of this material and pours liquid nitrogen around the substance, the magnet eventually levitates off the surface and hangs suspended in space.
Applications for Superconductors
The obvious application for a substance that acts like a magnetic mirror, is as a means of levitating objects. If one wants to levitate a train from a magnetic track, for example, one also needs to levitate all the refrigeration equipment needed to keep the superconductor cold. This application, therefore, is dependent upon someone making a room temperature superconductor.
On the other hand, if one is prepared to stay in one place, the ability to generate incredibly strong magnetic fields can be very useful. Nuclear Magnetic Resonance imaging devices and spectrometers rely upon magnetic fields and tend to work better the higher the field does. NMR scanners, now widely used to diagnose medical complaints, could not work effectively without superconducting magnets. The currents carried by the coils of these magnets would probably melt ordinary, non-superconducting copper wire.
The ability to create magnetic bottles, where the neck of the bottle is formed by a constricted magnetic field, is essential to anyone who wants to control the behaviour of very hot gases. Nuclear fusion and developments in plasma rocketry, for example, also use superconducting coils to do just that. The VASIMR rocket, currently under development, is a highly efficient and powerful device for interplanetary missions.
Electrons can behave as waves as well as particles but these effects are invisible on a macroscopic scale. However superconducting electrons, because they do not scatter off each other or lattice atoms, begin to co-operate over scales of centimetres and show wave-like behaviour that manifests itself in some unusual ways. If two superconductors are separated by a very thin layer of insulator, the electrons 'tunnel' through the layer and in so doing generate extremely high-frequency alternating currents, which are very sensitive to magnetic fields. This is known as the Josephson Effect (after its discoverer who won the Nobel Prize). By connecting two such junctions in parallel in a ring configuration, the resulting Superconducting Quantum Interference Device (SQUID) can be used to detect incredibly small magnetic fields such as brainwaves.
The Josephson Effect also has another, potentially incredibly important application; it could be used to make super-fast computers. The current through the junction is very sensitive to external magnetic fields. Therefore placing a wire carrying a current - and hence generating a field - next to it, can be used to turn the current on and off. The resulting Josephson logic gate works blindingly fast; switching speeds of nine trillionths of a second have been demonstrated. What is more, unlike conventional silicon circuits they do not generate any heat whatsoever.
The limiting factor constraining the temperature at which these devices function, is the material from which they are constructed. There is a further constraint in that it is not always that easy to make devices out of the materials; mercury is hardly an ideal substance for making wires! Niobium-germanium alloy is fairly ductile and so can be fashioned without too many problems. YBCO is a ceramic, and these do not lend themselves easily to forming into wires. Thus for the foreseeable future we are stuck with helium superconductors. There have been reports recently of a ceramic that superconducts at room temperature. But since there have been so many false alarms in the past, a future where we hover rather than drive to work still looks a long way off. Nevertheless, NMR scanners, sensitive magnetic sensors and novel interplanetary rockets could not exist now without these remarkable materials, and the equally remarkable discovery of Kammerlingh Onnes in 1911.
1 This process arises from a phenomenon known as the Pauli Exclusion Principle, which stipulates that no two interacting fermions can have the same quantum number.
2 Known as a band.
3 This is a simplified view of what really goes on, but it is more or less accurate enough.
4 Called Cooper pairs after the man who discovered them and won the Nobel Prize for Physics as a result.
5 In 2001, some Japanese scientists took magnesium and boron, ground them up together and heated them to make magnesium diboride. This material, which is hardly novel, superconducted at 39K. This is nowhere near the 130K of copper oxide-based ceramics, but it opens up a whole new line of enquiry for research.