by Academician Yu. OSIPYAN, Director of the USSR Academy's Institute of Solid State Physics

High-temperature superconductivity research everywhere is following three main directions. And just as you've said, the race for the highest possible temperature is still on. The past six months saw a breakthrough into a temperature range above 120K. And that means that during the whole of the previous 75 years scientists advanced only 15K.

The boom was triggered off by reports on the synthesis of lanthanum-based systems with a transition temperature of only 30- 40K. The next major step were yttrium-barium systems with a transition temperature close to 93K. And now even better compounds, based on bismuth and thallium, with a transition temperature of up to 120K have been developed.

As for the empirical search for new materials, scientists engaged in it are guided by certain qualitative considerations, such as the replacement of some ions with others. But they cannot become quantitative because we still do not understand the basic physical mechanism producing a state when a material loses its electric resistance. Understanding this mechanism is the goal of the second, and probably most important, direction of this research. Scientists are engaged in systematic studies of the available superconductors with the idea of seeing into the mechanism of their electric conductivity in the normal state, just before they become superconductors. This is essential for selecting the right basic model.

And then there is also a third area of research: technological uses of the new materials.

What are the main results obtained by Soviet scientists in this field to date to understand the mechanism of high-temperature superconductivity? What we do know is that the cause of this phenomenon, as follows from the old model of low-temperature superconductivity, is the pairing of current carriers and the formation of what are known as Cooper pairs - a complex of two electrons or two holes. The question is what causes this phenomenon, and it remains open.

The existing theory of Bardeen-Cooper-Schrieffer (BCS), which merited a Nobel Prize, suggested a phonon mechanism of formation of Cooper pairs. It appears, however, that this mechanism alone cannot account for high-temperature superconductivity. According to the latest findings, it is either not involved at all, or is involved in combination with some other mechanisms (with exciton, biexciton or polaronic, bipolaronic and so on).

We, at our Institute, have obtained what appears to be most important experimental results. To begin with, I'd like to mention our studies into the electrical properties of superconducting single crystals. These are determined by a complex atomic-crystal structure. Now try and imagine a prism 11-13 A high with a base in the form of a square measuring 3.5 x 3.5 A. It consists of three sub-cells in the center of each of which there is either a rare-earth or a barium atom. This structure, typical of high-temperature superconductors, is known as tetragonal. But if you cool the crystal, its structure will radically be altered at some 500- 600 ⁰ C, which is high above the temperature of transition to superconductivity It becomes orthorhombic, with the square at the base of the prism being replaced with a rhomb, something that alters all the crystallographic parameters accordingly. And it is these structures that we are investigating now.

The first measurements on yttrium-barium crystals indicate that in the normal state conductivity is anisotropic, that is it depends on direction. Then we investigated the "dimensional pattern" of conductivity. It was believed before that metallic conductivity occurs only along the AB planes, parallel to the base of the prism. In the C plane perpendicular to the base it was of a semiconducting nature.

These concepts provided the basis of the theory evolved by the Nobel Prize winner, Prof. Philip Anderson, of Princeton University A team of our researchers, led by Corresponding Member of the USSR Academy of

page 65

Sciences, I. Shchegolev ^* , have obtained results which cannot be described in terms of the Anderson model. It turns out that in a perfect single crystal conductivity in all directions (the C plane included) is of a metallic nature. This, to may mind, is one of the most important experimental results to date in the science of superconductivity We can now consider high-temperature superconductors as substances which, before transition into a superconducting state, are normal but anisotropic tridimensional metals (as different from dielectrics and semiconductors, the electric conductivity of metals increases as temperature drops). In other words, metallic properties manifest themselves throughout the volume of the crystal.

Later on our findings were confirmed by American researchers and by scientists in Japan. They were recognized as correct at last year's conference on high-temperature superconductivity in Interlaken, Switzerland.

The second important result concerns current distribution along the sample cross section. And without belittling the achievements of our counterparts in other countries, let me tell you again about our own achievements. As was believed before, superconducting current is restricted to some regions of the materials only, concentrating, for example, on twin boundaries. (In their low- temperature orthorhombic state all crystals are twin systems, their structure resembling a parquet floor, with the rhombs at the bases of prisms being at an angle to one another). But if it is true that the flow of current is restricted to some areas only, there is little hope of passing a strong homogenous current through such a sample.

We have been able to show how the current is distributed in a superconductor.

We knew that Meissner effect applied to superconductors. According to it, a magnetic field cannot penetrate into the sample, but is expelled out of it like a cork out of a bottle of champagne. But, as has been demonstrated by Acad. A. Abrikosov, in some types of superconductors the magnetic field does enter the sample, although not as an uninterrupted flux, but in the form of separate lines - magnetic field quanta, known as the Abrikosov vortices.

Using the electron microscope, we observed nonuniformity of the magnetic field distribution in a sample, depositing upon it a very thin layer of ferromagnetic powder. With the cooling of the material the powder particles regrouped and concentrated in circles in areas where the magnetic field was the strongest. These circles formed a correct lattice, which was in full agreement with the Abrikosov theory The circles were very small, less than a micron in size, and we had to make a "replica" upon which the pattern of vortices was imprinted and then studied under the electron microscope. The uniformity of the lattice attested to the uniform distribution of the current along the thickness of the crystal. The spaces between vortices were 0.3 mcm, and that was the maximum depth of the magnetic field penetration into the sample.

This led up to an unambiguous and most important conclusion: superconducting single crystals can carry heavy currents. Our findings concerning the magnetic field structure were later confirmed by Bell Telephone experts in the US.

I would also like to mention the optical studies conducted by the team of Prof. V. Timofeyev ^** Superconducting materials contain copper and oxygen. One question causing a lot of arguments is whether the replacement of some isotopes of oxygen

* From 1994 - an academician. - Ed.

^** From 1990 - Corresponding Member of RAS .- Ed.

page 66

by others alters the temperature of transition superconductivity? The first results gave a negative answer to this question. But then Los Alamos physicists came up with what looked like an impressive confirmation of the opposite. It was not accepted, however, because the method of sample preparation was not quite correct. What Timofeyev did was to stage technically similar, but optical studies. This is because the main question can be viewed from a broader premise. According to the BCS theory, a replacement of isotopes should influence not only the superconducting transition temperature, but also certain other properties, such as optical spectra. Timofeyev's experiments (he replaced ¹⁶ O with the heavier ¹⁸ O isotope) confirmed such influence. I feel that we shall soon be able to answer the question of how the replacement of isotopes influences the temperature of superconducting transition.

Interesting work has been done by Prof. V Ginzburg, Acad. L. Gorkov, and a number of other scientists.

It is difficult to tell about their work in simple terms. But still I'll try to give you a general outline of the new theory of Acad. A. Andreyev. It touches upon some of the things I have mentioned before.

On the strength of a number of experiments it was believed that the boundaries of twins can either increase or weaken superconductivity. The two conclusions seemed to be mutually exclusive. The new theory of Andreyev leads to some very important new conclusions which lend themselves to experimental verification.

The first concerns the distribution of the magnetic flux in a sample. Remember the Abrikosov vortices penetrating the superconductor? A magnetic vortex line cannot break abruptly In accordance with Kirchhoff rules taught at school, current does not end at some point, but either branches off, or flows through a different "channel". Andreyev has demonstrated that this is exactly the way superconducting current behaves in the presence of twins. And if current can branch off and follow a different "channel", it should be able to pass around the obstacles in its path.

This makes it possible to continue work on a fuller theoretical model describing the behavior of a superconducting current in a material.

The second important conclusion from Andreyev's theory is that vortex lines are sometimes divided into sections capable of moving on their own. These conclusions are of great importance in as much as they help us to understand what prevents vortices joining together and what homogeneities in a superconductor can act as barriers. In the final analysis we are talking of a material being able to pass greater or smaller critical current.

And this ability, in turn, has a crucial role to play in various technological applications, say, when we want to produce materials with definite properties. In the immediate future we can speak of producing superconducting cables for electrical power engineering. But this is easier said than done. As I have said on many occasions, to make a cable from a superconducting material is the same as making it from concrete. The material is very brittle and it has to be rolled and drawn. This is not to mention the fact that in the process it loses and gains oxygen which alters its superconducting properties.

Or take the interaction between the superconducting current and the magnetic structure of the material. I have already described how a magnetic field penetrates a sample in the form of vortices. But that picture was static, and in reality it becomes dynamic as soon as electric current is passed through the material. Stability of the superconducting state depends on the intensity of vortex motion or on how the vortices are distributed. With magnetic penetration increasing, the distances between the vortices are reduced owing to their growing concentration until it reaches a value when the magnetic field destroys the superconducting state. Thus the heavier the current, the greater the probability of losing superconductivity. We can now produce superconducting cables of considerable length that can pass a current of 1,000A/cm ² . And we think we can improve upon this figure.

Another technological aspect is making film for computer and microchip applications. In this field superconductors will have to handle alternating and not direct current and will be exposed to strong electrical fields which will affect their behavior. All parameters of their transition to a superconducting state will be changed (e.g. transition will occur at lower temperatures). When using such materials for alternating current, especially at high frequencies, a temperature "safety margin" should be ensured. Today these problems are being investigated at many academic research centers and colleges.

But one can say with confidence even now that at very high frequencies and at liquid nitrogen temperatures these materials retain their superconducting properties. And they are not lost even in the pulse mode of operation, with the pulse length being of the order of several picoseconds (1 picosecond is 10 ^-12 s).

Today we can make superconducting films that are in no way inferior, and sometimes even superior, to those produced abroad. And our next task is to make matrices containing superconducting and semiconducting materials and launching their commercial production. We have developed such a hybrid matrix scheme, but the technology of its production is still very complicated.

But we do have some interesting results in the field of chemical synthesis. The method of self-sustaining high-temperature synthesis (SHS) developed under the direction of Prof. A. Merzhanov ^* is very effective and has been assessed highly in the United States and other countries. Licenses have been bought by many industrial countries. This method yields superconductors that are in no way inferior to materials produced in other and more labor-consuming ways.

Our progress in all areas of high-temperature superconductivity studies can be judged by the desire of countries where similar research is under way to cooperate with us.

Quite recently researchers at the Institute of Solid State Physics of the USSR Academy of Sciences and the Los Alamos Laboratory in the United States published the results of their joint studies. We exchange samples and verily each other's results. And there is also cooperation between Soviet and American theoretical physicists.

^* From 1997 - an academician.- Ed.

Permanent link to this publication: