Absolute Zero and the Conquest of Cold

by Tom Shachtman

  From the mid-1930s onward, scientists seemed to have most of the pieces of the deep-freeze puzzle spread out on a table in rough order, but they could not make the final assemblage. Along with the pieces having to do with magnetism, specific heat, and viscosity, there was another, provided also by Fritz London. He postulated that the difference between a metal in the normal state and in the superconducting state had to do with an "energy gap" involving something called the "Fermi surface," named after Italian physicist Enrico Fermi.* London believed that if one could figure out the microscopic mechanism of the energy gap that differentiated the two states, that would explain superconductivity and diamagnetism; he was also certain that the explanation would involve electron interaction at the Fermi surface.

  Once London had identified the energy gap, Mendelssohn contends, the path to assembling the pieces was clear, but mapping that path was a "formidable task requiring a superb knowledge of electronic phenomena in metals, great mastery of mathematical technique, and, above all, brilliant but controlled imagination."

  One man might not possess all of these talents, but a trio of men did: John Bardeen, Leon Cooper, and Robert Schrieffer. Their summons to the puzzle board began in 1950, when a telephone call reawakened Bardeen's interest in superconductivity. In the fifteen years since London had directed attention to the energy gap, nuclear technology had been developed for the purpose of separating the isotopes of uranium and was now being turned to more benevolent uses. Looking into newly separated isotopes of mercury, two sets of researchers independently deduced an important mathematical relationship regarding superconductivity: the temperature at which a metal became superconducting varied inversely with the square root of its molecular weight. In 1950 E. Maxwell at the National Bureau of Standards and Bernard Serin at Yale University both arrived at the inverse-square-root formulation and prepared articles about it; Serin also telephoned his friend John Bardeen.

  Bardeen was one of the prodigies of American science. The son of a medical-school dean and an artist, he graduated from high school at the age of fifteen and became one of the youngest students at Princeton's Institute for Advanced Studies in the mid-1930S. At Princeton and Harvard he studied the behavior of electrons in metals—and learned of the Londons' work in that area as it related to superconductivity. During the war Bardeen worked on magnetic fields given off by ships, and in the postwar era he teamed with Walter Brattain and William Shockley at Bell Labs to invent the transistor, for which the trio would win the Nobel Prize in 1956. Serin called Bardeen in 1950 because he knew Bardeen had come to believe that electron interaction at the Fermi surface of a metal could explain the onset of superconductivity,* and Serin's experiments and mathematical formulations provided additional evidence for this possibility.

  Bardeen tried to shape from the various clues a comprehensive theory that solved all the superconductivity puzzles, but he could not do so for a few years. Then he was able to invite to his home base, the Illinois Institute for Advanced Studies, Leon Cooper, who had recently received his doctorate from Columbia University in quantum physics. Because not enough office space existed at the institute, grad students and postdoctoral fellows were crowded into offices on Floor 3½ of a neighboring building, an enclave they labeled the Institute for Retarded Studies. In 1956 Bardeen asked Cooper to make room in his office for Robert Schrieffer, a doctoral candidate from the Massachusetts Institute of Technology (MIT). Bardeen had decreed that Schrieffer should spend a year in a laboratory as preparation for his doctorate, but when Schrieffer caused an explosion in the lab while welding metal in a hydrogen atmosphere, he was asked to concentrate on theory.

  By then Bardeen had extended his guesswork beyond the energy gap described by Fritz London, adding the idea that superconductivity as a "phase transition" must be produced by a change involving the spin of the electrons. Phase transitions are the transformations that occur when a gas becomes a liquid or a liquid becomes a solid. In this case, the transition was from normal conductance to the superconducting state. While Cooper was traveling on a subway, the revelation came to him. Just above the temperature for the onset of superconductivity, electrons acted normally—they repelled one another, and one result was resistance to an electrical current. But when the temperature reached down to the transition there would be an interaction among electrons made possible by phonons., Two previously isolated electrons with oppositely directed spins would bond into a "Cooper pair." This bonding would be accompanied by some extra residual attraction that influenced all the other electrons, until none of them repelled one another. This was the superconducting state. Raising the temperature above the transition point would again break the Cooper pairs, making them repel one another, which produced electrical resistance.

  The Cooper-pairs idea set the stage for a complete explanation of superconductivity. The trio now applied themselves to extending the Cooper-pairs idea to explain how such pairing could affect all the electrons at the Fermi surface. Working in part by analogy, they likened electron pairs of a potential superconductor to separated couples on a crowded dance floor. When a dance couple at the edge of the floor begins to move in unison, their motion affects the movements of other couples—or, in terms of electron pairs, it could be said that the first Cooper pair generates residual attraction that makes other pairs bond. On the dance floor, the added motion spreads from couple to couple in a wavelike pattern until all the couples on the floor are moving in unison. Another analogy: The Cooper pairs are like a heavy ball rolled on a mattress; the springs do not bounce back immediately as the ball passes, but stay down for a moment, during which a route is smoothed for other balls to precisely follow the first one's path—as though attracted by the first ball.

  What was needed, as Schrieffer later put it, was a quantum-mechanical depiction of a wave function that choreographed the dance of 1023 couples—the number of electrons involved. This equation was so difficult to figure out that Schrieffer almost gave up on the task and considered changing his dissertation to one having nothing to do with superconductivity. But Bardeen convinced him to keep working at it for another month, while the advisor went to Stockholm to accept the 1956 Nobel Prize. During that time, Schrieffer figured out how to state the equation for the wave function of the dance, and when Bardeen returned, the trio tidied up the theory and set to work trying to see if it would explain all previous experimental results.

  As they did, their excitement grew, because what came to be called—after their initials—the B-C-S theory could explain the abrupt onset of superconductivity, the way in which magnetism was expelled from the interior of a solid, the rise in the specific heat at the transition temperature, and even superfluidity. All of these phenomena had to do with the action of Cooper pairs. Above the critical temperature, the pairs were apart or broken, and resistance and other normal attributes, such as magnetic susceptibility, existed; at the critical temperature, the Cooper pairs came together, a bonding that drastically changed the structure, permitting electric currents to course without resistance, eliminating interior magnetism, raising the heat capacity, and enabling liquid helium to flow end lessly. The connection between superconductivity and superfluidity now became clear: both had to do with the ability to sustain currents at a constant velocity over a long time, without any apparent driving force—in superconductivity, a current of electricity, and in superfluidity, the flow of liquid helium atoms.

  Bardeen recalled that in early 1957, when he, Cooper, and Schrieffer compared their theory with other scientists' experimental data, "we were continually amazed at the excellent agreement obtained." When they came across apparent discrepancies, they discovered by redoing the math of the other experiments that these calculations contained errors, not contradictions of their theory. They submitted a short communication to the Physical Review and followed it up several months later with a more detailed paper. Superconductivity, the greatest puzzle in the vicinity of Ultima Thule, had at last been solved.

  13. Mastery of the Cold


  FOR TEN THOUSAND YEARS, civilization had progressed by means of ever greater control of heat; in the twentieth century, progress came about through control of the cold.

  In 1912 Clarence Birdseye went ice fishing in Labrador, in temperatures 20 to 40 degrees below freezing. Imitating the natives, after hauling up his catch through ice several feet thick, Birdseye would throw it on a line over his shoulder. In minutes, because of the frigid air temperature, the fish would freeze solid. He noted that even when these fish were cooked weeks later, they tasted fresh, unlike fish frozen by slower methods. A chemist, Birdseye knew the general rule that the longer it takes for crystallization to occur, the larger the individual crystals that are formed. The reason for the fresher taste of quick-frozen fish, he determined through research, was that while in regular freezing ice crystals formed and grew in the frozen flesh, rupturing cell walls and compromising cell integrity, in quick-freezing solidification took place so rapidly that only very small ice crystals formed, and the cell walls did not break. Maximum crystal formation occurred between—1° and—5°C; by passing rapidly through that zone on the way to lower temperatures, quick-freezing avoided excess crystallization.

  It took Birdseye another ten years to work out a commercial fast-freezing technique, using a calcium chloride brine solution that kept the temperature at—45°F and applying the intense cold to the undersides of metal plates along which the fish traveled in a conveyor-belt mechanism. Achieving the low temperature was the least difficult problem; Birdseye's operation would have been possible fifty years earlier had anyone realized the commercial potential of quick-freezing food.

  Birdseye raised some capital and built his first plant in New York in 1923; its processing machine weighed 20 tons, making it all but immovable. With additional capital and investors he built a second plant in Gloucester, Massachusetts, in 1924, and he was soon doing a brisk business and examining the possibility of freezing foods other than fish. By 1928, 1 million pounds of quick-frozen foods were being sold annually in the United States, the preponderance of them by Clarence Birdseye's company; just before the 1929 stock market crash, the Birds Eye company was bought by Postum cereals, the precursor of General Foods.

  Fast-freezing machinery had grown progressively lighter in weight, to the point where it would shortly be movable from site to site to take advantage of peak harvests. Food experts waxed rhapsodic that this would allow utilization of ever greater percentages of harvested fruits and vegetables, leaving still less to rot in the fields or in storage. The main obstacle to wide acceptance of frozen foods was the consumer, who looked askance at such items. Still, the promise for the future was palpable: in the words of a contemporary industry journalist, "Strawberries could be eaten in December, and you didn't have to shell peas or wash spinach any more." A test of the retail-market appeal of fresh-frozen fish, vegetables, and fruits was held in 1930, but the arrival of the Depression soon put Postum's plans for an enormous expansion of the industry on hold.

  The first third of the century also saw the emergence of two other industries based on mastery of the cold: artificial refrigeration and air conditioning. At the turn of the century, though iceboxes were in regular use in about half of American homes, only 1 percent of U.S. households featured bulky, expensive, artificial-refrigeration machines. These were variations of what Linde had produced for industrial firms, based on the laws of classical thermodynamics and advances in compression machinery. Further spread of mechanical refrigerators was hampered by the need to connect the machines to local sewer systems to ensure frequent disposal of their dangerous refrigerant chemicals—ammonia, ethyl chloride, or sulfur dioxide. A less dangerous refrigerant, carbon dioxide, was not as widely used because it required more pressure to function properly. By 1915, using knowledge about refrigerant chemicals and low-temperature handling techniques pioneered by basic researchers such as Dewar and Onnes, the General Electric Company developed better carbon-dioxide-based refrigerators that were independent of sewer connections, self-contained, and small enough for the average home; during the following fifteen years, a million such refrigerators were sold by GE and its main competitors, Kelvinator and Frigidaire.

  In 1902 the new building of the New York Stock Exchange opened, with an air-cooling system for its trading room, designed by Alfred Wolff. In 1905 Stuart Cramer coined the term air conditioning to describe what the machinery he was installing would do to the air in a southern textile factory—control its volume, humidity, temperature, and recirculation, with the emphasis on humidity control rather than on interior cooling. The third father of the field was Willis Carrier, a young engineer who fitted factories with temperature- and humidity-control units for the Wendt Brothers of Buffalo between 1908 and 1915, when he and a half-dozen compatriots were fired because the likelihood of American entry into the war convinced the Wendts that the future for air-control units was dim. Carrier and the engineers formed their own firm, which by 1918 had completed installations in factories serving more than seventy industries. The units they installed were huge, central air-conditioning machines, too large for use in private homes.

  The United States emerged from the Great War in a stronger economic position than any other combatant nation; this economic strength began to express itself in many ways, among them growing use of the cold, for instance in skyscrapers and in motion-picture theaters. As commercial skyscrapers reached ever higher into the sky, their architects realized that at about twenty stories above the street, windows should not be opened, to prevent the havoc of wind gusts, more prevalent there than at ground level. Recognizing also that in summer the absence of outside airflow might make the buildings unacceptably hot, developers began to install central air conditioning in their taller buildings and to permanently seal their windows. In the 1920s, air cooling was installed in American motion-picture theaters, making them for the first time usable in summer and creating widespread appreciation among patrons for cool interior air. That appreciation, combined with the growing popularity of home refrigerators, led in 1928 to the Carrier firm's designing the first single-room home air conditioner. Because the Carrier machine controlled humidity as well as temperature—an exceedingly difficult task—the manufacturer was unable to price it lower than $3,000, which meant that these first air conditioners were beyond the reach of almost all Americans.

  The Tennessee Valley Authority and other rural-electrification projects of the 1930s spread the use of artificial refrigeration into new geographic areas. Shortly, despite the Depression, the majority of American homes featured refrigerators. Only the eclectic continued to use true iceboxes in their homes when they could afford the mechanical alternative: the poet e. e. cummings refused to buy a refrigerator and give up his icebox, because he was fond of the man who twice a week delivered ice to his Greenwich Village flat.

  Tracer bullets had easily exploded the World War I dirigibles that had been filled with hydrogen gas rather than helium, because helium had been too expensive and difficult to obtain. After the war, as a byproduct of examining routes to improving the efficiency of extracting and manufacturing gaseous helium for dirigibles, the U.S. Bureau of Mines in 1924 first produced liquefied natural gas (LNG), a combination of methane and other gases that liquefaction could reduce in volume 580 times. Here was a fourth industry whose existence depended largely on the ability furnished by application of the cold to condense its products for transport and storage. Widespread use of LNG as a heating fuel was hampered by the relative cheapness of fossil fuel oil found at the same locations as the gas and by a disastrous explosion near a natural-gas facility in 1944. These factors put the nascent industry on hold for several decades, until the 1970s, when the Organization of the Petroleum Exporting Countries (OPEC) hiked oil prices to the point where LNG became an attractive alternative and supertankers were built to make its transport equally economic.

  Large-scale production of liquefied gases for other uses started before that of LNG, because the numbers were so compelling
: a ton of gaseous oxygen occupied 26,540 cubic feet at normal temperature but only 31 cubic feet when liquefied; the ratios were similar for nitrogen, hydrogen, argon, neon, and helium. All sorts of manufacturers began to rely on the liquefied, deeply cooled products of air-separation plants that could be easily transported and then returned to gaseous form when needed. The technology of electric lighting depended on the availability of inert gases such as neon and argon to fill light bulbs. Steel making could no longer be accomplished properly without blast furnaces into which compressed oxygen was introduced. For welding—as predicted by Georges Claude, the developer of the French cryogenic industry—oxy-acetylene supplanted air-acetylene. Liquefied gases were also used in the production of many industrial chemicals, fertilizers, and photographic films.

  Leo Dana, the American chemist who had spent a year working with Kamerlingh Onnes, came home and found a job with the Union Carbide Company in 1923, shortly after it had bought the Linde air-separation operation in the United States. Among the first problems Dana tackled was the need for better insulation, so that liquefied gases could be more readily transported and stored. His "powder in vacuum" insulation boosted Union Carbide's ability to sell its products, and it enabled basic-research laboratories to maintain a steady supply of liquid gases.

  Too small to be noticed as a market just then was the use of liquefied gases as fuel for the experimental rockets being constructed by Robert H. Goddard in the United States and a small group of scientists in Hitler's Germany. The rocketeers needed the big thrust that controlled burning of hydrogen and oxygen could provide, and only by using liquefied gases could they pack enough fuel into their rockets to enable the devices to reach thousands and then tens of thousands of feet up into the sky.