Absolute Zero and the Conquest of Cold (7 page)

Among the candidates for fixed points proposed during the second half of the seventeenth century were, on the lower end, the temperature of mixtures of salt and ice, of ice and water, and of near-freezing water. In the middle range the suggested fixed points were the temperature of the deepest cellar of the Paris Observatory (believed not to vary between summer and winter); the congealing points of aniseed oil, linseed oil, and olive oil; the temperature at which butter melts; and that at which wax melts. On the higher end they were the heat of the healthy human body, measured under the arm or in the anus; the internal temperature of certain animals; the maximum summer temperature in Italy, Syria, and Senegal; the boiling points of pure alcohol, spirit of wine, and water; the heat of a kitchen fire hot enough to roast foods; and the presumed temperature of the sun's rays. Halley, another Fellow of the Royal Society, recommended the boiling point of spirit, yet he also revealed why it was so difficult to establish any fixed point: "The spirit of wine used to this purpose [must] be highly Rectified or Dephlegmed for otherwise the differing goodness of the spirit will occasion it to boil sooner or later, and thereby pervert the designed exactness."

The questions of fixed reference points and measuring liquids even lured Isaac Newton. An eighteenth-century historian of thermometers commented that in the way Newton "carried everything he meddled with beyond what anybody had done before him, and generally with a greater than ordinary exactness and precision, so he laid down a method of adjusting thermometers in a more definite way than had been done hitherto."

Newton forthrightly set his zero at the freezing point of water, but his indication of the exact location of that zero reveals the difficulty of being precise as to where a flag should be planted: he described the point as "[t]he heat of the air in winter, when the water begins to freeze; and it is discovered exactly by placing the thermometer in compressed snow, when it begins to thaw." Newton was wrong in many of his thermometric ideas. His scale erred in being self-referential—the meaning of its points could be understood only by reference to one another, and not by reference to any widely recognized standards—and in having a bias toward measuring the upper temperatures more finely than the lower. As Newton went up the scale, his descriptions became longer and more exact. Number 17 was "Greatest degree of heat of a bath, which a man can bear for some time without stirring his hand in it," but above that, his choices increasingly involved materials that ordinary people would seldom encounter, such as the boiling of mixtures of water and metals, and they also used ratios that smacked of the magical—for instance, he insisted that the heat of boiling water was three times that of the human body, six times that of melting tin, and eight times that of melting lead, and that the heat of a kitchen fire of heaped coals was sixteen or seventeen times that of the human body.

Newton contributed in a more novel and useful way in the matter of choosing the range of information the instrument could display, by proposing linseed oil as the fluid inside the thermometer bulb. This extract of flax was a brilliant choice, because it addressed the basic requirement that thermometers be able to measure accurately on the high and low ends of the scale. Being extremely viscous, linseed oil could remain liquid at temperatures above that at which water boiled and below that at which water froze. But linseed oil's viscosity made it slow to register changes—when the temperature was raised just a few moderate degrees, linseed oil would drain in a sluggish way down the sides of the tube on which the scale was marked. This long adjustment time made the linseed-oil thermometer difficult to use and may have been why a 1701 article on it by Newton in the Royal Society's
Transactions
was unsigned and only later attributed to him.

The next innovator in thermometry was French physicist Guillaume Amontons, regarded as a relatively obscure figure today, but one whose work provided a critical foundation for others in the exploration of the cold, even well into the twentieth century. The son of a provincial lawyer, Amontons was deaf since birth, and he had been largely self-taught in the sciences when in 1687 he first sought attention for his work from the Académie Royale. The twenty-four-year-old demonstrated a new hygrometer that used two liquids for measurement, one of them mercury, and followed this up a year later with ideas for multiliquid barometers and thermometers. More than the innovations of a craftsman, these were the constructions of a man who understood basic principles and how to apply them. The instrument maker for the Paris Observatory thought Amontons's ideas worthy enough to be discussed with members of the Royal Society on a visit to London. Put into a 1695 book, these ideas earned Amontons admission to the Académie.

His main contributions to thermometry and to the study of cold came by way of a detour. Amontons tried to create a "fire-wheel" that used the heat of a fire to expand air and make it move a wheel; he failed, but the principles he elucidated were sound and would be built on in later research by others on heat engines, horsepower, and friction. He also used some of the fire-wheel research to create better thermometers. Heating three unequal masses of air and water in glass bulbs submerged in boiling water, he demonstrated that the masses "increase [d] equally the force of their spring by equal degrees of heat," causing the air pressure to rise by one-third of an atmosphere in each bulb—1 atmosphere being equal to the pressure of air at sea level, 14.7 pounds per square inch.

From this pressure work Amontons drew two important conclusions. First, even if the heated air had been afforded "the liberty of extending itself" instead of remaining confined within the glass bulbs, it still would not have increased its volume by more than one-third.
^*
Second, since the water in all three bulbs boiled at the same temperature, despite variations in the volume of the water and air in the bulbs, the boiling point of water was proved to be a constant, one that could be used with confidence as a fixed point on a thermometer. Based on these results, Amontons designed a new air-based thermometer employing sealed glass; the sealing prevented distortions in the readings that would otherwise come from changes in atmospheric pressure.

Amontons also advanced the cause of better thermometry by his tough criticism of that anonymous 1701 paper on the linseed-oil thermometer in the
Transactions
of the Royal Society—not knowing it had been written by Newton. He lambasted the author's contention that a body that on the author's self-referential scale had a temperature of 64 was twice as hot as one that registered at 32, charging that not enough was currently known about heat and cold to support such an assumption.

As his critique of Newton's thermometry suggests, Amontons was a purist who did not often engage in the speculation rampant among researchers of the era. But he was a man who did pay attention to the implications of his mathematical computations; so, having proved that air contracts by a fixed proportional amount when cooled, he could not avoid calculating what
might
happen to air if its temperature were radically reduced, to well below the freezing point of water. Would air become denser, and as the temperature dropped further, would air become a liquid? Would that liquid be water—current thinking favored that conclusion—or something else? In the total absence of heat, would there be any air pressure? In a 1703 paper, Amontons evolved a simple equation showing that a total absence of heat was theoretically possible.

In the equation, the product of pressure times volume equals the product of temperature times an unknown constant. From this Amontons drew the clear implication that if by some means the product of pressure times volume became zero, on the other side of the equation the temperature could fall to an "absolute zero." Amontons did not come right out and say there was an absolute zero, because he considered such a thing incompatible with what was currently known about nature and with what he believed.

For Amontons, absolute zero was a hypothetical construct to be imagined, not to be realistically pursued. Two years after his article was published, he died of an internal inflammation, at the age of forty-two. Central to his legacy was his understanding that in the grand scheme of things, human beings and most other life on earth lived not far, in temperature terms, from the freezing point of water, and that the country of the cold that began at the freezing point was far more vast than human beings had previously believed, promising temperatures below what any scale then in existence could measure, down to an almost mythical point, an absolute zero, the end of the end.

Around 1702, while Amontons was doing his best work in Paris, in Copenhagen the astronomer Ole Rømer, who had calculated the finite speed of light, broke his leg. Confined to his home for some time, he took the opportunity of forced idleness to produce a thermometer having two fixed points, marking his scale at 7.5 for the melting point of ice and at 60 for the boiling point of water. His zero was thus well below the freezing mark, and supposedly represented the temperature of a mixture of salt and ice, while blood heat on this scale happened to fall at 22.5, or three times the melting-ice temperature. Rømer wasn't concerned very much with the upper and lower limits of his scale, because his primary work was in meteorology, which dealt with temperatures in the middle range. Six years after his broken-leg episode, Gabriel Daniel Fahrenheit visited him in Copenhagen. The young, Polish-born man had become fascinated by the making of scientific instruments and wanted tips on the techniques involved. And he may also have had another important reason for visiting Rømer.

Born in 1686 in Danzig, Fahrenheit had been orphaned when he was fifteen by the sudden death of both parents from mushroom poisoning on a single day in 1701. He had then been sent by his guardians to Holland as an apprentice to a bookkeeping firm, but he had run away from that position so many times that his guardians had had a warrant put out for his arrest. Going from city to city, he developed an interest in scientific instruments, teaching himself by visiting laboratories, refusing to settle anywhere, perhaps out of fear of apprehension and of being returned to his apprenticeship. What Rømer may have done for Fahrenheit, in addition to exciting his interest in thermometers, was arrange for the withdrawal of the arrest warrant, a courtesy that Dutch authorities would have extended to Rømer as the mayor of Copenhagen.

Until 1716, as Fahrenheit traveled from place to place, he worked on his thermometers, solving technical problems and becoming a competent glass blower. For several years, he sought the patronage of German philosopher and mathematician Gottfried Wilhelm von Leibniz, and it was only after Leibniz's death that Fahrenheit decided to settle permanently in the Netherlands and begin cultivating the patronage of Hermann Boerhaave, the celebrated physician, chemist, and botanist. (Boerhaave's fame was so great that a letter sent from China reached him even though the address read only "Boerhaave, Europe.")

Fahrenheit sent to Boerhaave and to other leading scientists samples of his thermometers, and he sought commissions to make more. Boerhaave commissioned several and also requested that Fahrenheit do some experiments for him. In these, Fahrenheit established that the boiling point of water is always a function of the atmospheric pressure and that for each atmospheric pressure the boiling point of water is fixed. He drew from these facts the implication that height and depth could be measured by a thermometer, so long as it was able to accurately record the point at which water began to boil. Boerhaave reported Fahrenheit's experiments in his influential chemistry textbook
Elementa Chemiae.

To make his thermometers, Fahrenheit adopted but also importantly altered Rømer's scale. Finding the numerical markers Rømer had used for melting ice and blood heat, 7.5 and 22.5, "inconvenient and inelegant on account of the fractional numbers," he tried to set his zero lower by making his blood-heat mark higher, at 24; this put his melting-ice temperature at 8. He made some early thermometers with this scale, but the aesthetics of these numbers still did not satisfy him, and he also wanted a scale on which each degree reflected a fixed percentage change in a liquid—as did the devices made by Boyle, Newton, and Hooke. With inexact implements, he nonetheless managed to create a scale on which a 1-degree rise or fall produced a change of one five-hundredth the initial volume of spirit of wine at the zero point.

There his innovations might have ended had Fahrenheit not been unusually inquisitive, willing to master new languages to advance his knowledge—French, so he could read Amontons and other contributors to the
Mémoires de l'Académie Royale,
and English, so he could read Boyle, Newton, and Hooke in their native tongue. After reading Amontons, Fahrenheit switched to using mercury and recalibrated his scale. On his newer thermometers, each degree corresponded to one ten-thousandth the initial volume of the mercury, the same proportion as in Boyle's and Newton's thermometers. To achieve that ratio, he had to quadruple the values on his scale—which turned out well, because with the melting-ice point at 32° and the blood-heat point at 96°, he now had a scale on which the key numbers were still divisible by 4, but on which the range was greater, making it easy to work with. The ability to more accurately locate blood heat was quite important, because Fahrenheit knew of Boerhaave's interest in measuring body temperature and wanted his influential patron to adopt (and recommend) his devices.

Perhaps to preserve his ability to exclusively manufacture these thermometers, Fahrenheit did not publish the calculations that had led to his scale. Knowles Middleton, the modern authority on the history of thermometers, suggests that all instrument makers concealed such matters, or obfuscated them, to prevent others from replicating their instruments without paying for them. Thus while Fahrenheit promised to provide Boerhaave with "accurate descriptions of all the thermometers which I make, and of the way in which I have ... attempted to rid them of their defects, and by what means I have succeeded in doing so," he actually withheld the secret of the volumetric measurements that had helped him arrive at the important numbers of 0, 32, and 96. An unintended consequence of this concealment was that for hundreds of years afterward, scientific historians wrote that Fahrenheit's scale was arbitrary.