|Proceedings of the Second International Symposium on Non-Conventional Energy Technology (PACE, 1984)|
By RON BOURGOIN
What I should like to do is to review with you the field of superconductivity. As you know, superconductivity at this time is a phenomenon that is found to occur at the extremes of low temperature. For those of you who are comfortable with the Fahrenheit scale, we are talking about electrical behavior which occurs at around 400° below zero on the Fahrenheit scale. We're talking about 0°, or something close to 0° Kelvin, or about minus 273° Celsius.
Kammerlingh Onnes is attributed with the discovery of superconductivity in 1911. Now there were a few things that occurred prior to this discovery. In 1908, Kammerlingh Onnes was the first, as far as I have been able to determine, to liquefy helium. In 1908! And he was able thereby to obtain temperatures near 1° Kelvin, which had never been attained before.
Of course, there was considerable theory at the time of how metals might behave as the ambient thermal environment was reduced, and so his first work, as far as I am able to determine, was with platinum. He took platinum, cooled it down to his new found temperature, and found, essentially, a linear decrease in resistivity with reduction in temperature. When made tests specifically with pure platinum and impure platinum, he found that the resistivity reduction was a function of purity of material. Now, the point that I want to emphasize is this: the reduction in resistivity with temperature was no surprise. That had been expected. The near-zero resistance which was measured at these extremes of low temperature was low because he was already close to zero resistance anyway. So there was not a significant finding in that particular case.
Now, when he used mercury, he wanted to use a material with extremely high purity, and mercury, at the time, was the metal known with the highest purity. He cooled it down and he found the resistivity reduction essentially as he had found in the platinum, except that when he got down to about 4° Kelvin, he got a little surprise. The material dropped abruptly in resistance to measurably zero value. (I emphasize the measurably zero value because what it actually is we really don't know, but measurably it has a zero value.) He conducted some tests with impure mercury and found exactly the same behavior. That's quite an important point. In the case of platinum, he found that resistivity reduction was a function of temperature; in the case of mercury he found that it really didn't matter whether the material was pure or impure - it dropped in resistivity exactly the same way. So it didn't appear to matter to the mercury whether it was pure or impure.
You're already close to zero anyway when you get down to these extremes of low temperature, so I have often questioned the gain [of the measuring instrumentes - ed.] measuring the drop in resistance to a measurably zero value, especially when you consider the expense of cooling the materials to that degree of temperature.
Let's have a general review of the types of superconductors that are presently in existence. One is the Type One superconductor, which is essentially a metallic superconductor. Those materials have been found to enter the superconducting state at temperatures typically below 10° Kelvin. We find that the superconductivity is a surface effect. The current wants to travel on the surface. The electron drift velocities have been determined to be something of the order of 105 cm/sec. The thickness of the conducting sheath has been found to be something of the order of 10-5 cm. At one time it was thought that if you took a highly pure crystalline solid and, as you attained the absolute zero of temperature, the thermal vibrations would reduce in such a way that the mean free pass of the conduction electrons would be practically infinite, and so it was thought, you see, that the metal would provide almost perfect conductivity through the material - through the crystalline solid. However, this is not what occurs. If you take a pure metal of the highest purity you can obtain, ninety nine point nine out to the tenth decimal place somewhere, you find that the superconduction, that is this almost perfect conductivity, does not occur through the material. And yet the thermal vibrations at something of the order of a few degrees Kelvin is supposed to be significantly reduced, but when you crank the values of temperature through the equation, you find that the thermal agitation of electrons in the conduction band is still something of the order of 10 6 cm/sec. Now, you see, you've got a surface velocity of 105. So, apparently, you have a higher agitation in the structure, and that might be one of the reasons why the electrons are forced to the surface. As far as the theory was concerned at the time, what I want to emphasize is that a pure crystalline structure was expected to provide near infinite or near perfect conductivity at near absolute zero temperatures, which we have not found.
Another type of superconductor is called Type Two superconductor. Type Two superconductors are essentially alloys and metallic compounds. Can you think of any conducting structure less ordered than an alloy or metallic compound?
These materials have been found to enter the superconducting state at as high as 23° Kelvin. In these highly disordered structures we find that the superconductivity is through the material, and when we examine what's going on we find that there are naturally forming thin current filaments in the structure. When we do further work we find that the cross sectional diameter of these natural thin current filaments is something of the order of 10-6 cm. This is all in the literature. It also has been determined that the superconduction speed is of the order of 106 cm/sec.
So, the highly disordered structures that were supposed to provide all of this resistance, these imperfect structures which were supposed to contain all sorts of defects and impurities were never expected to provide the superconductivity that they do, yet we find that they superconduct at higher temperatures than the metals, and that the conducting speed of the superconduction current is increased by a factor of at least 10 times.
So you see that, when I saw all of that coming out in the literature years ago, I began to question, What is resistance? Wouldn't you? You're told for years and years and years that resistance is a function of thermal agitation of metal ions in the lattice. Resistance is a function of collisions of conduction electrons between each other. Resistance is a function of crystal defects. I could go on and on, but you know this as well as 1. And yet we find that when we cool the imperfect materials down to as high as 23° Kelvin, resistance goes to zero, or measurably so. So what happened to all of this resistance that's supposed to be provided by the defects in the structure, and the impurities, and the thermal vibration? Because, as I told you a while ago, the thermal agitation velocities occuring in the conduction band in the metals near 1° Kelvin or below is found to be something of the order of 10 6 cm/sec. What I didn't tell you is that, at room temperature those vibrations are of something of the order of 107. So we find a reduction in thermal agitation velocity of only a factor of 10. And yet we get all of this high conductivity in the alloys. Apparently, you see, when the material enters the superconducting stage the defects don't evaporate, the impurities don't evaporate, everything is still right there. They have found no changes whatsoever in the structure of the material. So what caused the resistance in the first place? You say, "Well, the thermal agitation in the structure has reduced significantly". Really? You call reduction by a factor of 10 significant? You're going from 10,000,000 cm/sec thermal agitation in the conduction band down to 1,000,000 - that's a significant reduction?
So what's happened? Well, it appears the key to a high temperature superconductor might be in that thin filament existing in the Type Two superconductor. We'll explore that a little bit later.
Let's talk about some of the general properties of superconductors. This is very important. The first general property is extremely high current, something of the order of 105 amps per cm2. 106 amps per cm2 is no surprise. As a matter of fact, they get up much higher than that, but you're talking about currant densities that are truly anomalous.
The superconductors have extremely long electron mean free path, which means they can go for a long, long way without colliding into anything, whatever that means. I never have understood how two like charged particles like electrons could collide anyway, but it's in the literature, and those of you who have read the textbooks on electrical theory know it's there. You've seen the word "collision" all over the place, and yet you know that can't collisions can't possibly happen.
The electrons have high electron drift. Obviously we're talking about something to the order of 105 cm/sec in a Type One and 106 cm/sec in a Type Two. When you calculate the electron drift velocity in a copper wire, say 20 gauge, at room temperature, you find something of the order of l/100th of a cm/sec. Even that's pretty liberal; I think it's quite less than that, depending on how much current we're passing through, of course. So you can see a significant increase in order of magnitude.
Next point is abrupt drop in resistance. The point that I want to emphasize here is that the in the field of superconductivity, the name superconductivity was derived from this phenomenon, which is the abrupt drop in resistance; it wasn't everything else that has come later.
The next point is the generation of extremely high magnetic fields - in the Type One material, something of the order of 600 gauss; in the Type Two material, something of the order of 200,000 gauss. That's in the literature. Niobium-Tin, for instance, has a field of 200,000 gauss.
The next point is extremely enhanced diamagnetism, and we've heard diamagnetism mentioned frequently here. What is it? You approach a diamagnetic material with a magnet, it has a tendency to run away from you. So, extremely high diamagnetism. It repels any incoming magnetic field.
The next point is extremely poor thermal conductivity.
The next point is that if you make thin films ["chips"] with this material, you find that when they enter the superconducting state, they do so at a particular temperature and they measure certain current magnitude and they also measure certain magnitude of magnetic field. Then, as you probably know, if you increase the current to a value that is called the "critical current", then you lose the superconductivity, or, if you apply a magnetic field in excess of its particular value, you find that you lose the superconductivity. But they have found - the sides [tapered edges - ed.] of the chip continue superconducting. The fields required to lose the superconductivity in the edges are significantly higher than in the rest of the chip. The current drift along the edges is much higher than it is in the remainder of the chip. But you know what they do: they feel that the sides are causing than all kinds of difficulty, so what they do is they clip them off. That is a fact. Because, you see, those darn sides are causing them to get erroneous measurements. The blasted things just won't go out of superconductivity. It just destroys the magnetic effects on the chip so when they want to force reentry of the chip into the normal conducting state, so they cut the things off. What they ought to do is to keep the edges and throw the rest away.
What I want to discuss, basically, is a field called superfluidity. I'm not going to go over all the details of superfluidity, except to say this. When you take helium and you cool it and you get down to about 4° Kelvin, about 4.2°, to be exact, it enters the liquid state, which is no surprise. But then, if you keep cooling and cooling and cooling, you get to a point at where you have a completely inviscid fluid. Zero viscosity. That means that it has no internal friction and, once you get it going, it won't stop. They have passed liquid helium through extremely fine apertures, something of the order of a few Angstroms diameter. A few Angstroms - an atomic diameter is an Angstrom or a couple of Angstroms, or three, or something like that, and we're talking about a capillary of five to ten Angstroms through which you are passing liquid helium! And you measure no reduction in pressure from the point of input to the output. No loss in pressure. What does that mean? That means that there was no friction between the capillary walls and the liquid. There was also no viscosity - no internal friction in the liquid. There was nothing to slow it down. No dissipation of energy whatsoever.
They have determined that the superfluid state is due to the pairing of helium atoms in an orientation of antiparallel spin. What that means, basically, is that one rotates clockwise, adjacent to another that is rotating counterclockwise. Now, if you look at Bernoulli's theorem, what does he say ought to happen when you have parallel flow lines? Think of two ships out at sea that get to close to each other. You get parallel flow lines. What happens to the water in between? Parallel flow lines, in essence in an antiparallel configuration. What happens? You get attraction. Those of you who have ever sailed know exactly what I'm talking about.
Now, these gentlemen here, Tilly and Tilly, have written a wonderful textbook called Superfluidity and Superconductivity, and they have found some interesting parallels between electrical superconductivity and the field of superfluidity. What they're talking about is essentially this: Isn't it interesting that in the field of superconductivity three men, Bardeen, Shriefer and Cooper, in 1957 generated an acceptable theory for a phenomenon that was discovered in 1911, and, I believe, were awarded the Nobel Prize in 1972. What did they find? They found that the basic superconducting unit is antiparallel spinning electron pairs. So these gentlemen here, you see, said, "Look! You've got antiparallel spins leading to attraction of conduction electrons. You have essentially the same kind of thing happening at extremes of low temperature in liquid helium". And so what they have done is to applied equations from superconductivity to superfluids, and some the predictions they were able to make are wonderful. That is about a 250 page to 300 page book, and it's absolutely delightful for those of you who might be interested. I'm just giving you a background now, some of the things that I was seeing over the years, which eventually led to the research that I performed.
In the fields of whiskers and filaments, Howard L. Cobb, in 1946, reported finding some whiskers growing off of capacitors that were able to short out radio units. Sydney Arnold, at Bell Laboratories, in 1956, was talking about "tin whiskers" that were able to short out large capacitors. He conducted resistance measurements and found drops in resistance, very anomalous drops in resistance from what is expected from the standard law of resistivity. And he made this statement: "The current magnitudes exceeded those expected by a factor of 100". Those of you in the electrical and electronics fields know that we try not to pass any more than 100 amps/cm2 through a wire. You just don't pass more than that; otherwise it's going to evaporate. Arnold found 200,000 amps/cm2. And that was found about 1951 and the man, I'm sure, had to think a long time before he published his results. But he published them in 1956. There was a book by Bogenschutz in 1974 that talked about all of the problems that were being provided by metallic whiskers and thin filaments. These things were just shorting everything out. Well, why didn't somebody look at those fibers at the time and say, "Might it be that we have something that's trying to superconduct at normal ambient temperatures and higher temperatures?" - because these capacitors are at quite high temperatures, as you know.
What they should have done is stopped all of this work that was going on at the extremes of low temperatures and the helium refrigeration devices and so on and so forth, and gone to work on these thin whiskers and filaments. Stanford R. Ovshinsky, in 1966, received a United States patent for a symmetrical current controlling device, and what the man found was this: he was able to make a chip which offered resistance in the several millions of Ohms. When he got to a certain threshold voltage, he found that the resistance went down to something far less than one Ohm, and the only way that he could explain what was going on was to hypothesize the creation of very thin current filaments inside his chip. The drop in resistance from several million Ohms down to less than one occured instantaneously and THERE WAS NO EXCESSIVE GENERATION OF HEAT IN THE DEVICES. So much for i2r heating. Article by Marcus and Rottersmann in 1967. They found that in making chips they were making thin stripe conductors going all over the chip - and they were being shorted out. They examined the situation and found extremely thin filaments that had grown inside the thin film conductors themselves. Now, those thin film conductors were something of the order of 200 to 1,000 Angstroms in diameter. If there were thin filaments forming inside that, how big were they? They had to be significantly less than 1,000 Angstroms in diameter. That's in the literature. A.K. Jonscher, in 1969, found thin filaments in glassy insulator substate material. It seems that nature is going to find a path of least resistance whether we provide it or not.
D.E. Thomsen in 1980, in Science News, reported work with Saran wrap. Those of you who have ever seen Saran wrap under an electron microscope know that there are flbrils going all over the place. If you put an Ohmmeter across the thing, you'll measure several millions of Ohms. But take it and stretch it: align those fibrils inside the material and you find that the resistance drops down by several orders of magnitude, like 105 or 106; resistance drops by a factor of about 1,000,000. What's the size of them? Roughly 500 Angstroms in diameter.
IBM at the present time is doing some work on nanobridges, extremely fine line conductors on chips. These things are about 20 to 200 Angstroms in cross-sectional diameter. That is one particular method at this time of making very thin film stripes. Bell labs is presently working with quantum well wires. It's a rectangular thin film conductor that has a thickness of 200 Angstroms and width of about 200 Angstroms. The length is as long as you want to make it. M.l.T. is making what it calls "narrow lines". All of these people are making narrow lines of essentially the same dimensions. My question is why are they making them but they are not reporting resistance.
My method of making thin filaments is to prepare a chemical colloidal suspension. I take extremely fine metallic particulates and I suspend them in high dielectric mediums such as an epoxy resin. I heat the mixture to the melting point of the metal, which in the case of bismuth is something of the order of 271 °K. The only reason for doing that is to bring the particle into the molten or semimolten state. Then you apply electrodes to a container such as teflon or a glass beaker with pins, opposite polarity facing each other. You apply the voltage, which in some cases can be several thousand volts, wait awhile, and you'll get conductivity in just a matter of a minute or so. Then you retract one set of electrodes as far as you want to go. If you'll take cross sections of your mixture, you'll find little thin fibers going all over the place.
It's a very easy way to make filaments. As far as controlling the diameters and the lengths, I'm still working on that. I'm trying to get into the Ph.D. program in Materials Science at North Carolina State University, and hope that if I'm invited back next year that I will have some good reports for you. But this is presently what I'm doing. I do have a United States patent for the process. What it's doing at the present time is nothing, but there is a publication next month, the Bismuth Institute, which is headquartered in La Paz, Bolivia, has gotten very interested in my findings, and has decided to publish a very long abstract of the patent. The publications center for the Bismuth Institute is in Bussells, Belgium.
There are several theories of high temperature superconductivity. The Soviet theories say this: That if you make an extremely thin filament process, or film, and you encapsulate it by a polarizable medium, then the high current flow, assuming there will be one through the thin filament process, will induce positive charges all over the surface of the thin filament. And that extremely high positive field will serve to reduce the repulsive effect between electrons, and let those natural spin forces take over. And they must be extremely short range. They are saying basically that you create a positive, attractive field for the electrons. So you can get pairing that way.
Another theory, as expressed by my old professor, Dr. Robert Carroll, says that if you take a thin film of a material that normally contributes very few electrons to the conduction process, and yet presently contains high drift velocities, you ought to be able to confine the electrons into an extremely thin channel and force them to pair thereby. Just upon entry to an extremely thin channel.
And so, what I started doing was to look around for materials. All this seemed very exciting to me. It seemed there was a lot of theory saying that these things ought to superconduct at normal ambient temperatures. And also the findings with the edges on thin films and with the whiskers. If I could speak with Mr. Arnold of Bell Laboratories, I would ask him this: "How come that first whisker that made contact with the ground did not evaporate?" This man was talking about capacitors in the order of a hundred microfarad. What about the instantaneous current drain through that whisker? It must have been truly enormous. So you see, you look at all of those things and then a picture begins emerging, and that is that if you can make extremely thin conductive filaments they ought to superconduct, or you ought to see something akin to superconductivity at normal ambient temperatures, and possibly higher temperatures. This is what I did. I started looking at bismuth, which has a resistivity that is about 100,000 TIMES LESS than what it should be according to its conduction electron density. The resistance is high, but NOT AS HIGH AS IT OUGHT TO BE. Also, the electron mean free path, that is, the distance between collisions, to use the present vocabulary, is extremely long, so you have extremely long mean free path in the bismuth. This is in bulk, at normal ambient temperatures. Bismuth is the most diamagnetic of all of the metals. If you don't believe it, approach it with a magnet and see what happens. I did that on purpose. I made extremely thin filaments that I could see, put them on a table top, approached them with a magnet - and off the tabletop they went, instantaneously. If that's not diamagnetism, I don't know what is. Also, bismuth, in the bulk form, has the next to the lowest thermal conductivity of all the metals. And, also, another of the things that I liked about bismuth is that it has extremely low wear resistance, or a very low coefficient of friction. I put all of those things together and they spelled a partial superconductor at room temperature. Remember the first list 1 gave you and compare it with what I've just given you. So then I thought that if I could make bismuth filaments down to a couple of hundred Angstroms in diameter, I ought to be able to observe superconductivity behavior at ambient temperatures. All of the details are in the patent.
I worked and I worked and I worked and I was down to a micron diameter of fibers. I noticed that I needed a certain voltage, and then suddenly, I got very rapid oscillations all over the place, tremendous instability, and then I would increase the current just by a little bit, and the needles would just go out of sight. The voltage went down to practically zero, as far as I could measure; the current tended to rise to infinity. Now, I didn't like the idea of having a voltage drop - they tell you you're not supposed to have a voltage drop in a superconductor. So I worked to eliminate the voltage drop. So I learned to prepare a colloidal suspension having smaller particulates suspended, and worked at that and, my wife can tell you, I will never forget the day that this happened - I succeeded in making the suspension; I went ahead and formed the fibers as I had normally done; I took resistance measurements at the end of the trials, and they measured ZERO RESISTANCE. That was one of the happiest days. I immediately ran out of that laboratory, took my wife out and we had some wonderful wine and a wonderful meal, and just had a great time. I have made hundreds of these, and I can tell you that it works as I say it does. I have not measured resistance. I am not exactly certain what's going on, and that's why I'm going on into this program and hopefully doing some additional work.
As far as the magnetic field is concerned, I don't need to get into details. You can surmise what they would be as well as I can. Potential applications include the obvious computer applications. One application which might not be so obvious is the possibility of developing a synthetic spinal cord bridge. That excites me more than anything. If we are able to find a way to make something that will not generate resistance and i2r heating in the body, insert a material that would be accepted by the body (and, by the way, bismuth is very nice) and bridge gaps in the spinal cord where fibers have not regenerated, wouldn't that be wonderful?! We've got a lot of work to do. There's a possibly of very thin microelectrodes for single neuron studies, which they desperately need. You have a zero resistance electrode - think of the activity that you could measure in the individual neurons. The extended lengths of fibers - think of the possibility of low loss cables. Think of the possibilities of the generation of high magnetic fields, without auxilliary refrigeration mechanisms. Think about windings for motors and generators. And I could go on and on, but I've got to quit at some point. Thank you all very much.
Q. In one of your comments you were surprised that the voltage dropped to zero. Why did you not want this? And second question is, can you comment on "frozen-in" fields in superconductors?
A. Are you talking about the so-called "fluxoids"? [A. Yes.] Regarding the first question - I wanted a voltage drop to zero. If I said that, I certainly didn't mean it. When I found initially a voltage drop I didn't want it because I knew that in order for a material to be superconducting, it would have to provide no potential drop. Regarding the second question, in superconductors you can have a persistent field which stays there and "locks" the superconducting filament in place. In my case, we are talking about a single one of those filaments. 90 we are not talking about exactly the same thing.