Copyright 1987, 1989, 1999 Jerry Emanuelson

Superconductivity was discovered in 1911 by Heike Kamerlingh Onnes, the Dutch physicist known for his research into phenomena at extremely low temperature.  In 1908, Onnes had become the first person to liquify helium.  He was investigating the electrical properties of various substances at liquid helium temperature (4.2 degrees Kelvin) when he noticed that the resistivity of mercury dropped abruptly at 4.2 K to a value below the resolution of his instruments.

In 1933, W. Meissner and R. Oschenfeld discovered that a metal cooled into the superconducting state in a weak magnetic field expels the magnetic field from its interior.  In 1945, the Russian physicist V. Arkadiev first performed the now-classic experiment of using this expulsion of a magnetic field to levitate a small bar magnet above the surface of a superconductor.

Advances in superconductivity continued to proceed slowly.  During the first 75 years of superconductivity research, the critical temperature (the temperature below which superconductivity is present) was raised by less than 20 degrees.  In 1973, a niobium alloy was produced with a critical temperature of 23.2 K.  This is still the highest temperature for a metallic superconductor.

Finally, in January of 1987, one of those rare authentic technological breakthroughs occurred when researchers at the University of Alabama at Huntsville and at the University of Houston produced ceramic superconductors with a critical temperature above the temperature of liquid nitrogen.

The search for a ceramic superconductor actually began several months earlier with research conducted by several laboratories around the world, with the most productive research coming from K. Alex Mueller and J. Georg Bednorz of IBM Zurich.  The IBM group published (in 1986) results of research showing indications of superconductivity at about 30 K ¹.  Building on the IBM work, several other laboratories soon demonstrated bulk superconductivity at 39 K in an oxide of lanthanum, strontium, and copper. Bednorz and Mueller received the 1988 Nobel Prize in Physics for their groundbreaking work.  At the time Bednorz and Mueller began their work, the idea of a high-temperature ceramic superconductor was considered to be so crazy that they did their research quietly -- not even telling their colleagues what they were doing.  They tried more than two hundred combinations of ceramic oxides before achieving success.

The group of physicists at the University of Houston, led by C. W. Chu and his associate M. K. Wu of the University of Alabama at Huntsville, discovered that they could raise the critical temperature to 52.5 K by applying a considerable amount of mechanical pressure (more than 10,000 atmospheres) to the sample.

Finally the University of Houston group came up with the clever idea of pressurizing the crystal from the inside by substituting similar atoms of a different size rather than using an external mechanical press.  The resulting yttrium-barium-copper-oxide compound is the material that has sparked the current technological revolution.  Chu produced the first superconductors at liquid nitrogen temperatures² by using a yttrium:barium:copper ratio of 1.2:0.8:1.0.  Analysis of the superconducting material by X-ray diffraction and convergent-beam electron scattering³ later showed that the crystal phase that actually forms the superconductor has a Y:Ba:Cu ratio of 1:2:3.

[Subsequent information reveals that originally too much credit was given to the University of Houston and not enough to the University of Alabama, though.  See Appendix A, below, for a revised version of the history of the discovery.]

Recent research indicates that the yttrium-barium-copper oxide superconductor is different in some fundamental ways from earlier metallic superconductors and possibly different from the earlier ceramic oxide developed at IBM.  The yttrium compound seems to work in an entirely different way from all earlier superconductors.

The discovery of the yttrium-barium-copper-oxide superconductor caused scientists all around the world to stop other projects and join the the search for even better superconducting materials, and new developments began occurring on a weekly basis.  In 1987, a number of researchers believed they had evidence of room-temperature superconductivity.  The evidence for this was usually the observation of an instantaneous drop in resistivity as the material is cooled from above room temperature.  It is generally believed that the only logical explanation of such a sudden drop in resistivity is that a small filament in the material is superconducting at higher temperature and suddenly "shorts out" part of the non-superconducting material as the filament reaches its critical temperature.  Some scientists, however, have expressed alternative explanations for the abrupt drop in resistance⁴.

As of this writing, no one has produced unmistakable bulk superconductivity at, or near, room temperature, although some scientists claim to have isolated some of the microscopic fragments from the tiny filaments of room temperature superconductor.  At least one finding in late 1987 indicated possible superconducting phases well above room temperature in a 0.03 percent fraction of a typical sample, but this finding was never replicated elsewhere.  The search for superconductors with ever-higher critical temperatures continues.

At least ten elements from the lanthanide series have been successfully substituted for yttrium in making the new oxide superconductors.  Some of the alternate compounds have properties that are superior to the yttrium compound.  For example, neodymium-barium-copper-oxide is superior in its capability of excluding a magnetic field⁵, and thulium-barium-copper-oxide remains superconducting under much stronger magnetic fields than the yttrium compound⁶.

Although not usually mentioned in discussions of superconductivity in either the scientific or popular literature, it is likely that the capability of shaping a magnetic field and shielding magnetic fields will become one of the more important industrial applications of these superconductors.

Type I vs. Type II superconductors.

An important characteristic of all superconductors is that the superconductivity is "quenched" when the material is exposed to a sufficiently high magnetic field.  This magnetic field, H_c , is called the critical field. In the early superconductors, including all of the elemental superconductors except niobium, the superconductivity is quenched in relatively low magnetic fields.

In contrast, type II superconductors have two critical fields.  The first is a low-intensity field H_c1, which partially suppresses the superconductivity. The second is a much higher critical field, H_c2, which totally quenches the superconductivity.  The upper critical field of type II superconductors tends to be two orders of magnitude or more above the critical fields of a type I superconductor.  Therefore, it is the advent of the type II superconductor that has made possible the manufacture of superconducting magnets of incredible strength.

The main factor limiting the field strength of the conventional (copper wire) electromagnet is the I²R (power) losses in the windings when sufficiently high currents are applied.  (Power = current squared times resistance.)  In a superconductor, where R = 0, the I²R losses are obviously not a problem.

The new ceramic oxide superconductors are type II superconductors and early research indicated upper critical fields that appear to be at least as high as anything yet discovered.  One paper⁶ stated that the upper critical field of yttrium-barium-copper-oxide is 14 Tesla at liquid nitrogen temperature (77 degrees Kelvin) and at least 60 Tesla at liquid helium temperature.  The similar rare earth ceramic oxide, thulium-barium-copper-oxide, was reported to have a critical field of 36 Tesla at liquid nitrogen temperature and 100 Tesla or greater at liquid helium temperature.  (By comparison, the strongest household magnets are rarely stronger than .05 Tesla.)  It is apparent that if the problems of working with these brittle ceramic materials can be overcome, magnets of incredible strength can be made for a multitude of potential applications.

The first tantalizing indications of very high critical fields in these new superconductors has since given way to pessimism because of the phenomenon of "flux creep," a process whereby lines of magnetic flux penetrate the material and quench the superconductivity in a very unpredictable way.

If flux creep and other unusual problems can be overcome, incredibly powerful magnets could be made for use in fusion reactors for energy production, particle accelerators, advanced transportation systems (such as high-speed maglev trains), and perhaps most important of all, advanced medical imaging devices.

Magnetic resonance imaging devices are currently the most important market for the "old-style" superconductors.  Nuclear magnetic resonance now enables physicians to obtain detailed images of the interior of the human body without surgery or exposure to ionizing radiation.  Magnetic resonance imaging devices are now available only at major hospitals.  They are very bulky machines because of the huge amount of thermal insulation required to keep the liquid helium from evaporating.  The liquid helium to operate a magnetic resonance imaging device costs about $30,000 per year ⁷.  It has been estimated that the use of liquid nitrogen superconducting magnets could save $100,000 per year in overall operating costs for each magnetic resonance imaging device⁸.  In addition, the initial cost of the machines would be far lower and the physical size of the machines would be much smaller.  It is entirely conceivable that the new oxide superconductors could make it economically feasible for magnetic resonance imaging devices to be located in many clinics and doctor's offices.

Some Consequences of Zero Resistance.

When a current is induced in a doughnut-shaped superconductor, the current will continue to circulate in the ring until an external influence causes it to stop.  In the 1950s, "persistent currents" in superconducting rings immersed in liquid helium were maintained for more than five years without the addition of any further electrical input.  The kinds of external influences that might cause a persistent current to stop include being exposed to a magnetic field that tends to oppose the existing current flow, being exposed to a magnetic field above the critical field of the material, and the superconductor being warmed above its critical temperature.

A superconductor cannot be shorted out.  If the effects of moving a conductor through a magnetic field are ignored, then connecting another conductor, such a piece of copper, across a superconductor will have no effect at all.  In fact, by comparison to the superconductor, copper is a perfect insulator.

The diamagnetic effect that causes a magnet to levitate above a superconductor is a complex effect.  Part of it is a consequence of zero resistance and of the fact that a superconductor cannot be shorted out.  The act of moving a magnet toward a superconductor induces circulating persistent currents in domains in the material.  These circulating currents could not be sustained in a material of any finite electrical resistance.

These circulating persistent currents form an array of electromagnets that are always aligned in such as way as to oppose the external magnetic field.  In effect, a mirror image of the magnet is formed in the superconductor -- with a north pole below a north pole and a south pole below a south pole.  If the magnet is moved or rotated, the "mirror image" of the magnet rotates with it.  A disk magnet levitating over a superconductor may be spun rapidly about its longitudinal axis without affecting its levitation.  Diamagnetism that is strong enough to levitate a magnet can only occur in a superconductor.  For this reason, the "levitating magnet" test is one of the most accurate methods of confirming superconductivity.

Preparation of the Ceramic Oxide Superconductors

The most common method of making the yttrium-barium-copper oxide superconductors is to mix dry powders of yttrium oxide, barium carbonate, and cupric oxide in the proper molecular ratios.  (Barium carbonate is used instead of barium oxide because commercially available barium oxide is very impure.  Also, much of the barium oxide tends to be converted into barium hydroxide and barium carbonate upon exposure to air.)

After the chemicals are thoroughly mixed, they are heated in a furnace at about 950 C for about 18 hours.  This forms the basic crystal structure and eliminates the carbon dioxide from the barium carbonate.  The chemical reaction that occurs in this step is:

This first firing results in a porous dark gray or black clump.  If the powders are not mixed well enough, a green phase will appear as well.  This mass is ground into a fine powder and placed again in the furnace.  This time a oxygen flow is maintained as the material is heated to about 950 degrees Celsius for about 18 hours.  After this firing, the sample will show some signs of superconductivity.  But an additional grinding and firing under flowing oxygen is usually necessary to obtain a good quality superconductor.  In the process of being heated under oxygen flow, the oxygen content of each crystal unit is increased from 6.5 atoms of oxygen to approximately 7.  There was considerable uncertainty during much of 1987 about the "ideal" oxygen content.  One early report showed that a good yttrium-based superconductor had an oxygen content of 6.72 ⁵.  It is now generally believed that the compound should have an oxygen content that is as close to 7 as possible. The chemical formula of the superconducting compound is:

The rate of cooling in this final firing must not exceed 100 degrees per hour until the temperature of the material is below 500 degrees Celsius.  This slow rate of cooling is an absolute necessity for producing superconductors with a critical temperature above the temperature of liquid nitrogen.

For the individual preparing these superconductors for the first time, the most difficult part of the preparation is obtaining the correct oxygen flow.  Adequate oxygen flow must be maintained until the temperature of the sample has dropped below 400 degrees Celsius.  If the oxygen flow is removed too soon, much of the necessary oxygen will be lost.  One unique characteristic of this material is its ability to transport oxygen ions.  It was once thought that this compound may have many uses as a transporter of oxygen in applications having nothing to do with superconductivity⁹ but little has been done with this idea.

In the earliest samples of superconductor made by Colorado Futurescience, we relied on sintering to produce superconductors of the desired size and shape rather than pressing the material into pellets.  In sintering, the particles in the powder are fused together by heating the material for an extended period of time to a temperature that is below the melting point of the material.

The only common materials that are suitable for use as a container for firing this compound are platinum and alumina.  We chose to use alumina for obvious economic reasons.  Conventional laboratory-grade porcelain and this compound tend to mutually contaminate each other when porcelain labware is used for the firings.  Occasionally, even alumina becomes stained by the compound, but the contamination is easily removed by soaking the alumina labware in hydrochloric acid.

There are several problems with the results of the ordinary methods of preparation.  One problem is that the material is very brittle and breaks easily under repeated thermal stress.  This problem can be relieved somewhat by cycling the temperature several hundred degrees up and down during the final heating cycle.  This causes the material to develop thermal stress cracks at unstable locations during the heating process.  These small thermal stress cracks are then re-sintered before the material is finally brought back down to room temperature.

This brittleness of this material is one of the most important problems to be overcome before many of its potential commercial uses are realized.  Much progress has been made recently in this area.

Another problem that needs to be overcome is the development of economical techniques for growing single crystals of this material.  The methods described above, as well as many similar methods, produce a random array of crystals of the material.  The random crystals are bonded together by the sintering process to form a continuous superconductor, but the current-carrying capability is substantially reduced as compared to a single crystal.

Many of the applications requiring a large current-carrying capacity will require growing single crystals of this compound.  As of this writing, various methods have been tried such as hydrothermal synthesis ⁹; i.e. growing the crystals in water at extremely high temperatures and under extremely high pressure (to keep the water from vaporizing at the high temperatures).

One of the most promising and active areas of research is in thin films of these materials.  This is of great importance in incorporating these material into semiconductor devices.  In fact, a new type of transistor using the critical magnetic field to switch currents on and off is now under development.

Summary.

High-temperature superconductivity in ceramic oxides is a new technology in which advances are occurring at a rapid pace which is unprecedented. This technology ended a long period in superconductivity research in which little progress was made.  Even though the mechanical properties of these materials present a serious problem in many practical applications, many of the other properties of these ceramic oxides show tremendous potential for revolutionary applications.

Bibliography:

1. J. G. Bednorz and K. A. Muller, Z. Phys. B, p.189, vol 64, 1986.

2. M. K. Wu, C. W. Chu, et al., Physical Review Letters, p. 908, vol. 58, March 2, 1987.

3. P. M. Grant, et al., "Superconductivity above 90 K in the Compound YBa₂Cu₃O_x : Structural, transport, and magnetic properties," Physical Review B, pp. 7242-7244, vol. 35, no. 13, May 1, 1987.

4. Geballe, T.H. and J.K. Hulm., "Superconductivity - The State That Came in from the Cold," Science, pp. 367-375, vol. 239, no. 4838, January 22, 1988

5. J. M. Tarascon, et al., "Oxygen and rare-earth doping of the 90-K superconducting perovskite YBa₂Cu₃O_x," Physical Review B, pp. 226-234, vol. 36. no. 1, July 1, 1987.

6. J. J. Neumeier, et al., "Thulium barium copper oxide: A 90-K superconductor with a potential 1-MG upper critical field," Applied Physics Letters, pp.371-373, vol. 51, no. 5, August 3, 1987.

7. M. D. Lemonick, "Superconductors!" Time, pp. 64-75. vol. 129, no. 19, May 11, 1987.

8. J. Wilson and O. Port, "The New World of Superconductivity," Business Week, pp. 98-100, April 6, 1987.

9. Judith Goldhaber, "Riding the Heat Wave," LBL Research Review, pp. 2-11, vol. 12, no. 2, Summer, 1987.

Appendix A:

In the text above, the story of the discovery of superconductivity above the temperature of liquid nitrogen is told in the way that it was reported at the time in the scientific and popular press.  Many subsequent reports, beginning with an article in the August 5, 1988 issue of Science, gave a very different accounting of the events.  It now appears that most of the work leading directly to the discovery was made at the University of Alabama at Huntsville, not at the University of Houston.  The actual discovery was made by graduate students Jim Ashburn and Chuan-Jue Torng at Alabama -- based on an idea by Ashburn.  As often happens with grad students, they got none of the credit at the time.

The relevance of the pressurization work at the University of Houston to the discovery still remains controversial.  Ashburn and Wu maintain that the pressurization experiments had nothing to do with the discovery of the yttrium-barium-copper oxide material, but the University of Houston group has always maintained that there was a connection.

Anyone wanting to investigate further the actual events leading to the discovery of the first material to superconduct above the temperature of liquid nitrogen should consult the following sources:

Pool, Robert, "Superconductor Credits Bypass Alabama," Science 5 August, 1988. pp. 655-657.

Chu, C.W. "High-Temperature Superconducting Materials: A Decade of Impressive Advancement of Tc," IEEE Transactions on Applied Superconductivity, Vol. 7, No. 2, June 1997, p. 80.

Chu C.W., "High Temperature Superconductivity," History of Original Ideas and Basic Discoveries in Particle Physics, ed. H. B. Newman and T. Ypsilantis, Plenum Press, New York, 1996, p. 793.

Chu, C.W., "Superconductivity Above 90 K and Beyond," HTS Workshop on Physics, Materials and Applications, ed. B. Batlogg et al., Singapore, World Scientific, 1996, p. 17.