Giauque, William Francis
GIAUQUE, WILLIAM FRANCIS
(b. Niagara Falls, Ontario, Canada, 12 May 1895; d. Berkeley, California, 28 March 1982)
physical chemistry, chemical physics.
Giauque received the 1949 Nobel Prize in chemistry for his research on chemical thermodynamics, particularly his pioneering and exhaustive investigations on entropy and low-temperature chemistry. He contributed greatly to establishing the third law of thermodynamics as a fundamental scientific law, invented the adiabatic demagnetization cooling pro cess, and demonstrated the natural occurrence of oxygen”s O17 and O18 isotopes and molecular hy drogen”s ortho and para forms. Giauque’s experimental researches were meticulous, most of them definitive, with improvements on his results coming only from refinements in technique. In his long and productive career, all of which he spent at the University of California (Berkeley), Giauque published 183 papers and trained 51 graduate students.
Giauque was the eldest of two sons and a daughter. His parents, William Tecumseh Sherman Giauque and Isabella Jane Duncan, were American citizens, which, according to the citizenship laws in effect, automatically made their children American citizens. Giauque received his elementary school education mainly in Michigan, where his father was a weight master and station agent for the Michigan Central Railroad. Upon his father’s death in 1908, the family returned to Niagara Falls Collegiate and Vocational Institute, intending to acquire the training he needed to help support them. By this time, Giauque’s mother had become a part-time seamstress and tailor for J.W. Beckman, a chemist with American Cyanamid Company. Her employment with Beckman proved a very fortunate development because convincing Giauque to switch to the five-year general (college preparatory) course the next year took the combined efforts of his mother and the Beckman family. Giauque selected electrical engineering, but lacking both finances and engineering experience, he planned to work for a short time in one of the power gen erating plants at Niagara Falls. Unable to find any engineering openings, he accepted a position with the Hooker Electro-Chemical Company across the river in Niagara Falls, New York. Hooker’s well organised laboratory impressed Giauque greatly, and the two years he spent there convinced him to study chemical engineering.
By 1916 Beckman had been transferred to Berke ley; hearing of Giauque’s new interest in chemical engineering, he suggested that Giauque attend the University of California at Berkeley, Gilbert Newton Lewis had arrived there in 1912 to serve as the chemistry department’s chairman and dean of the College of Chemistry, which included chemical engineering. He also attracted a first-rate faculty, among them Joel Hildebrand, George E. Gibson, William Bray, and Gerald E.K. Branch. Bechman spoke highly of Lewis’s research on the electron valence theory, thermodynamics and free energy and praised the research program Lewis had established. His recommendation of Berkeley’s program, combined with its ten-dollar total semester fee, easily persuaded Giauque to enroll in the College of Chemistry rather than attend the more expensive Massachusetts institute of Technology or Rensselaer Polytechnic Institute.
Giauque graduated with highest honors in 1920, receiving a B.S. degree in chemistry for a program of study that contained 25 percent engineering courses, Hildebrand, who taught Giauque, described him as outstanding. Two years later Giauque, described the Ph.D. in chemistry with a minor in physics. Gibson directed his dissertation althought Giauque during his graduate studies. Because of Giauque’s obvious promise. Lewis immediately offered him a faculty position. Giauque still had hopes of an engineering career, but the excellent research envi ronment that Lewis had created finally led him, after several months of ambivalence, to pursue a career in chemistry. Giauque remained at Berkeley for the rest of his life, moving from assistant (1922–1927) to professor (1934) in twelve years. On 19 July 1932, Giauque married Muriel Frances Ashley, They had two sons.
Giauque’s earliest investigations were on low temperature entropy and the third law of thermo dynamics. The German chemist Walther Nernst first stated the third law, then called the Nernst heat theorem, in 1906. According to Nernst, in any re action involving only solids and liquids (including solutions), the change in entropy approached zero as the temperature approached absolute zero. Five years later Max Planck, in the third edition of his Thermodynamik, argued that Nernst’s theorem did not hold for solutions and required modification. He suggested assigning zero entropy to each element at absolute zero and interpreted the third law to mean that all pure solids and liquids had zero entropy at absolute zero. But Lewis and Gibson pointed out that entropy measured randomness of a macroscopic state, and even in a pure solid or liquid some ran domness existed in its structure. Only a perfect crystal of a pure solid or liquid lost its entropy at definition of the third law accepted today: The entropy of a perfect crystal is zero at absolute zero.
Giauque demonstrated the correctness of Lewis and Gibson’s third-law interpretation in his doctoral dissertation and in his first publication (1923). He showed experimentally, from heat-capacity and heat of-fusion measurements, that glycerol glass (super cooled glycerol) at 70K had considerably more entropy than crystalline glycerol (about 5.6 cal mol-1 K-1) and concluded that this difference remained even at absolute zero. His third-law demonstrations continued in the 1920’s and early 1930’s with a series of investigations on diatomic gases in which he calculated their entropies theoritically from spectroscopic data and compared them with experimental entropies determined calorimetrically.
Giauque and several graduate students, among them R. Wiebe, H.L. Johnston, and J.O. Clayton measured low-temperature heat capacities and changes of state to obtain the calorimetic entropies for molecules such as hydrogen iodide, oxygen, nitrogen, nitric oxide, and carbon monoxide. For the spectroscopic entropies they used quantum-statisticcal equations developed within the last twenty years from band spectra studies on gaseous molecules. These included Otto Sackur’s and Hans Tetrode’s equations for gaseous that Richard Tolman and Harold Urey had independently derived for rotational energies (1923), and Hervey Hicks and Allan Mitchell’s equations for vibrational energies (1926).
All of these experiments, which Giauque began publishing in 1928, showed the thoroughness and high degree of accuracy that characterized his Research. Not only did the close agreement of his spectroscopic and calorimetric entropy values clearly support the third law of thermodynamics, but gi auque provided further confirmation when he showed that his experimental entropies and those calculated from equations for the entropy of formation were in excellent agreement. At the same time his experiments verified the use of quantum statistics and the partition function in calculating entropy.
As a result of calorimetric studies in 1852 by Julius Thomsen at Copenhagen and later by Marcelin Berthelot at Paris, most chemists at that time believed the quantity of heat evolved in an exothermic reaction measured the reactants’ affinity for one another. While this correlation often seemed to hold, some reactions that were clearly endothermic occurred spontaneously. Confusion also surrounded the meaning of entropy after introduction of the concept in 1854 by Rudolf Clausius at Bonn, and persisted until J. Willard Gibbs at Yale University, in a brilliant series of publications (1875–1878), showed that both the heat (enthalpy) and the entropy changes were necessary to establish a reaction’s spontaneity at constant temperature and pressure. To measure spontaneity. Gibbs in 1875 defined a new function called the “potential” and expressed its dependence on enthalpy, entropy, and temperature with the equation dζ = dχ–t dη, where ζ is the potential, χ the enthalpy, η the entropy, and t the temperature. In 1882 Hermann von Helmholtz at Berlin derived a similar relation in which he introduced the term “free energy” instead of Gibbs’s “potential,” For this reason chemists and physicists today speak of the Gibbs-Helmholtz equation, writing it as ΔG=ΔH — TΔS, where ΔG is Gibbs’s free-energy change; ΔH is the enthalpy change; and TΔS is the product of the entropy change and absolute temperature.
Gibbs (1876), Helmholtz (1882), and Jacobus van’t Hoff lapr1880) showed that equilibrium constant (K) and electromotive force (EMF) measurements pro vided additional ways of calculating free energy. At Berkeley, Lewis, Gibson, and Merle Randall, carried out numerous free-energy determinations from chemical equilibrium studies and EMF mea surments, while Giauque and his students calculated them from enthalpies and entropies measured ther modynamically, and from spectroscopic data.
Giauque’s research on the entropy of gases from 1928 to 1932 showed that the hydrogen molecule, H2, and molecules with similar ends, such as CO, NO, and N2O, crystallized with a definite amount of residual entropy as the temperature approached absolute zero. The entropy of solid hydrogen results from a disordered nuclear spin alignment of the molecule’s two protons. This alignment is parallel or in the same direction in the ortho form, and antiparallel or in opposite directions in the para form. Disordered crystalline arrangements cause the entropy found in the other molecules. Werner Hei senberg, at Niels Bohr’s Institute for The oretical Physics in Copenhagen, first suggested in 1927 that hydrogen and other elementary diatomic molecules existed in symmetrical (para) and antisymmetrical (ortho) forms. That same year Friedrich Hund, also in Copenhagen, pointed out that nuclear spin ac counted for an element’s hyperfine spectrum and on that spectral analysis therefore provided information on the two spin states.
Nuclear spins, like electron spins, are difficult to reverse; and when a molecule interacts with electromagnetic radiation, the resulting electronic band spectrum contains two different sets of lines, one for each spin alignment. Molecular hydrogen’s spectrum showed that the more intense set belonged to ortho-hydrogen and corresponded to odd rotational levels (odd rotational quantum numbers), while the fainter set represented para-hydrogen and even rotational levels. The two forms produced a regular pattern of lines that alternated in intensity, and from the intensities it appeared that molecular hydrogen contained a three: one ortho: para mixture at ordinary temperatures. Edward Condon, at that time a National Research fellow in Germany, had carried out theoretical calculations on the hydrogen molecule in 1927 showing that the ortho-para equilibrium was temperature dependent though established only slowly. In a letter to the Berkeley laboratory he suggested that evidence of the two forms might result from keeping hydrogen at liquid air temper atures for two to three months. Condon expected to see a marked difference in hydrogen’s heat capacity if a transition occurred between the two forms.
Giauque began the suggested experiments late in 1927. He and H. L. Johnston obtained twenty grams of pure hydrogen by electrolyzing water, and after keeping it in a steel container at the temperature of liquid air (85 K) for 197 days (19 October 1927–3 May 1928), they observed a decrease of 0.04 cm (0.4 Torr.) in its vapor pressure at the triple point. The change occurred because as the temperature decreased, fewer and fewer molecules had sufficient rotational energy to remain in the higher ortho (odd rotational) energy levels. More and more molecules reversed their nuclear spins to assume the lower-energy para form and occupied the lower (even) rotational levels. Giauque and John ston’s experiment indicated an entropy difference of 4.39 cal mol-1 K-1 for the two forms, which clearly supported the third law interpretation that Lewis and Gibson gave in their 1920 paper.
Karl Bonhoeffer and Paul Harteck, at the Kaiser Wilhelm Gesellschaft in Berlin, first separated ortho and para-hydrogen in 1929, using ten grams of char coal to adsorb a small amount of hydrogen gas. The charcoal acted as a catalyst in establishing the equi librium between ortho-and para-hydrogen. They succeeded in keeping the charcoal at the temperature of liquid hydrogen (20 K) for about twenty minutes, and after pumping off the gas they showed, from its higher thermal conductivity, that it consisted of 99.7 percent para-hydrogen. That same year Arnold Eucken and Kurt Hiller, at the Technische Hoch schule in Breslau, measured a change in the heat capacity of hydrogen they had cooled to 90 K for periods of four to fourteen days. By the 1930’s low temperature research established beyond doubt that molecular hydrogen existed in two forms and that a catalytic separation produced almost pure para hydrogen.
To account for the entropy values of CO, NO, and N2O, Giauque believed that in the crystal state some of the molecules had heat-to-head arrangements (CO. CO, NO. NO, N2O. N2O) and others had head-to-tail arrangements (CO. OC, NO. ON, N2O. O2N). Every molecule’s orientation in the crystal was not the same as required for zero entropy. Such disorder could account for the small entropy at 0 K. For CO the entropy was 1.1 cal mol-1 K-1. In a completely random crystalline ar rangement of CO, NO, or N2O, the entropy increased to a maximum of S = R In 2, or 1.38 cal mol-1 K-1. While the behavior of these compounds seemed to fall outside the third law, Giauque’s results proved the law’s validity only for perfect crystalline order in the lowest energy state.
Giauque’s third-law investigations resulted in his most significant accomplishment, the invention in late 1924 of cooling by adiabatic demagnetization. His new cooling method enabled scientists to understand better the principles and mechanisms of electrical and thermal conductivity, to determine heat capacities, and to investigate the behavior of superconductors at extremely low temperatures. Michael Faraday in London had conducted the first systematic low-temperature research beginning in 1823, when he used compression and cooling with ice-salt mixtures to liquefy such gases as chlorine, sulfur dioxide, ammonia, and carbon dioxide. In 1877 Carl von Linde at Munich developed a com mercially practical refrigeration process based on the expansion of ammonia gas, and in 1877 and 1878 Louis Cailletet at Chaacutetillo—sur-Seine reached temperatures lower than 80 K and liquefied the “permanent gases,” oxygen, nitrogen, nitrogen dioxide, carbon monoxide, and acetylene. Linde and Cailletet used the Joule-Thomson effect (1852), in which a compressed and cooled gas, after expansion through a small opening, cools further be cause the expanding gas expends some of its kinetic energy in overcoming intermolecular attractions.
Almost simultaneously, in 1877 Raoul Pictet at Geneva developed a cascade process in which each gas in a group of several gases with decreasing critical temperatures and triple points, such as sulfur dioxide, carbon dioxide, and oxygen, liquefied the group’s next member. The process liquefied a gas by compression at the critical temperature, the highest temperature at which it existed as a liquid, and cooled it to its triple point, the lowest temperature at which it existed as a liquid, by boiling under reduced pressure. Because no liquids have critical temperatures and triple points between nitrogen’s boiling point (77 K), hydrogen’s critical point (33.3 K) and triple point (14 K), and helium’s critical point (5.2 K), the cascade process failed to liquefy hydrogen and helium. A solution to the problem of reaching these temperatures finally appeared near the turn of the century. In 1895 Linde considerably improved Joule-Thomson cooling with the invention of his regenerator or heat-interchanger cyclic cooling. In 1898 James Dewar at the Royal Institution in London combined the Joule-Thomson, Linde, and cascade processes to liquefy hydrogen (20.4 K), and in 1908 Heike Kamerlingh Onnes at the Cryogenic Laboratory in Leiden used the combined process to liquefy helium (4.2 K). The low temperature study of matter (cryogenics) made available laboratory temperatures of 5.0–0.8 K and led in 1911 to Kamerlingh Onnes’ discovery of superconductivity in metals such as mercury, tin, and lead.
Kamerlingh Onnes also studied the magnetic susceptibility of the paramagnetic compound gadolinium sulfate octahydrate, Gdsub2 (SOsub4)3. 8H2O, at liquid he lium temperatures. These measurements became Giauque’s starting point in 1924 when he calculated the effect of a magnetic field on the octahydrate’s entropy and showed theoretically that application and subsequent adiabatic removal of the field at liquid helium temperatures produced additional cooling. Adiabatic demagnetization suggested a new method of reaching temperatures near absolute zero. Two years later, in a theoretical paper published at Zurich on 11 December 1926, Peter Debye used the same gadolinium compound to describe in detail the principle of adiabatic demagnetization cooling. Debye’s paper appeared eight months after Wendell Latimer at Berkeley publicly discussed Giauque’s work for the first time, at the California Section meeting of the American Chemical Society (9 April 1926).
In 1924 it was well known from Nernst’s heat theorem and the Gibbs-Helmholtz equation that the heat capacities of substances become very small and approach zero at temperatures below 10—15 K. At these temperatures a substance loses practically all its thermal entropy, and magnetization/soldemagnetization should produce no further significant cooling. But paramagnetic compounds, such as gadolinium, cerium, and dysprosium salts, have thermal and magnetic entropy. In the absence of a magnetic field at low temperatures, they no longer have appreciable thermal entropy but still possess magnetic entropy because their atomic magnets have an irregular arrangement. Application of a powerful magnetic field forces the atomic magnets to line up with the field, reducing the magnetic entropy. A cooling bath re moves the heat generated by the entropy decrease. Giauque recognized that if he insulated the compound thermally and removed the field under adiabatic conditions, the total entropy must remain constant. By removing the field, the atomic magnets return to their random arrangement and increase the magnetic entropy. Temperature measures thermal motion; therefore, the accompanying decrease in thermal entropy that corresponds to a decrease in motion results in the temperature’s lowering. Because magnetic entropy is a factor in cooling only at low temperatures, Giauque pointed out that cooling by adiabatic demagnetization is most effective at tem peratures produced by the evaporation of liquid helium (1 K). He compared the sequence of steps in magnetic cooling with the three steps in the re frigeration process, using an idealized expansion engine.
When Giauque began calculating the low tem peratures achievable with magnetic cooling, he had neither the expensive large-scale equipment to conduct experiments nor the thermometer to record the readings. He planned not merely to measure low temperatures but also to use magnetic cooling in his low-temperature thermodynamics research. Paramagnetic salts were ideal for use because of their high heat capacities at low temperatures, though Giauque later experienced difficulty in making good thermal contact with the cooled salt. The equipment he required included a magnet with a strong ho mogeneous field (8, 000—20, 000 gauss); a hydrogen and a helium compressor; a purification system for removing oil, air, and other gases from the helium; and vacuum pumps for hydrogen recovery and re duction of liquid helium’s temperature. Because this apparatus was not immediately available, Giauque and his graduate student D. P. MacDougall succeeded in carrying out the first adiabatic demagnetization cooling that produced temperatures below 1 K (0.53 K) only on 19 March 1933. Nine years had passed since Giauque first conceived of cooling by adiabatic de magnetization. The Leiden group had known all along of the research, but to Giauque’s astonishment they never attempted the cooling experiments before he did.
In the experiments, Giauque and MacDougall placed a sixty-one-gram sample of paramagnetic gad olinium sulfate octahydrate in a copper calorimeter tube. A vacuum jacket filled with helium gas to conduct heat from the compound surrounded the tube. The tube and jacket rested inside a copper lead Dewar flask to which Giauque and MacDougall added liquid helium through a vacuum-jacketed transfer tube to a height of one meter and then placed the flask within the copper coils of a solenoid magnet. Low-viscosity cooling oil (kerosene) pumped rapidly over bare copper conductors removed heat and promoted efficient heat transfer, which, Giauque found, was the principal problem in designing so lenoid magnets. An inductance bridge measured the gadolinium sulfate’s magnetic susceptibility.
The cooling progressed in three stages: (1) the electric current through the copper coils caused the atomic magnets to line up, releasing heat and de creasing the entropy of the paramagnetic compound; (2) when the cooling stopped, the compound was insulated against heat flow by evacuating helium gas from the surrounding jacket; (3) the electric current was turned off, quickly demagnetizing the compound, which did magnetic work by inducing an electric current in the copper coils. Because no heat entered, the atomic magnets absorbed energy and cooled the compound.
Measuring such low temperatures presented a problem. The commonly used constant-volume gas thermometer, even one containing helium, deviated from ideal behavior at these temperatures and was in error. An alternative was to measure the salt sample’s magnetic susceptibility, which, according to Curie’s Law (1905), varied inversely with the absolute temperature. Giauque made the first sus ceptibility measurements with a coil of several thousand turns of fine copper wire around the in sulating vacuum jacket. As the temperature de creased, the alternating current flowing through the coil also decreased while the salt’s magnetic sus ceptibility increased. From the relation between current and susceptibility, Giauque obtained magnetic susceptibility values and then calculated the absolute temperature from Curie’s law. For gado linium compounds the Curie constant C is 7.880, giving the Curie-law equation.
T = 7.880solmagnetic susceptibiligy.
The Curie equation provided good low-temperature values. But upon approaching absolute zero (1 K) it failed because it allowed entropy values to decrease asymptotically during magnetization and conflicted with the accepted view that entropy was finite. Gi auque obtained true thermodynamic temperatures by plotting the change in enthalpy between zero field and some constant field (H) against entropy, or . The graph showed the variation of T with magnetic field strength and Giauque’s calculation of the absolute temperature from the equation T = dHsoldS at constant magnetic field.
By 1938 Giauque had improved temperature mea surement by inventing an extremely sensitive amor phous carbon (lampblack) resistance thermometer for work below 1 K. It consisted of a single layer of glass-lens paper (which he had chosen for its very loose open structure) applied to the sides of a twelve-inch-long glass sample tube. Lampblack mixed with a large amount of ethyl alcohol was painted on the paper and then coated with a col lodio—ethyl ether-alcohol solution. Two platinum wires connected the carbon layer to tungsten ter minals sealed into the glass wall. The thermometer measured temperatures accurately and precisely and was suitable for low temperatures because its re sistance had a high temperature coefficient and changed little with magnetic field strength. By this time Giauque had decreased temperatures from 1 K, obtained by evaporating liquid helium, to 0.004 K with adiabatic demagnetization.
The absorption band spectra from which Giauque had calculated the entropies of diatomic gases led unexpectedly to the discovery of oxygen’s two iso topes. A band spectrum contains many strong and weak lines. By applying quantum statistics to their distribution pattern. Tolman at the California Institute of Technology in 1923. Birge at Berkeley in 1926, and others determined the molecular energy levels of these gases. Giauque had used the same quantum statistical distribution patterns of the molecular en ergy levels to calculate entropies and had found excellent agreement with values obtained from the third law and from entropies of formation. In 1928, while examining atmospheric oxygen band spectrum photographs provided by Harold D. Babcock of Mount Wilson Observatory. Giauque and Johnston noticed some unaccounted-for weak lines in the spectrum. In 1928 Robert Mulliken at New York University had shown that the spectrum’s strong doublets belonged to molecular oxygen (O16-O16). Giauque also knew that many of the weak lines that Babcock had discovered and measured belonged to the molecule’s higher energy states. Yet he and Johnston could not account for all the weak lines. Giauque never left anything of significance unexplained and considered his entropy calculations un satisfactory because they failed to account for the additional weak lines. The origin of these lines re mained unknown until early 1929, when Giauque recalled awaking one morning and suddenly realizing that the lines came from oxygen isotopes.
Francis W. Aston, J. J. Thomson’s assistant at Cambridge and the world’s authority on mass spec troscopy, had established the existence of neon, sulfur, chlorine, and sili conisotopes by 1929, but had not found any isotopes of oxygen. Detailed frequency calculations by Giauque and Johnston in January 1929 proved that the O16-O18 molecule gen erated one set of weak lines. By May their calculations identified another set of very weak lines that Babcock had reported but had not associated with the oxygen spectrum. They belonged to the O16-O17 molecule. Their presence confirmed the existence of the O17 isotope that P. M. S. Blackett and others in 1925 had reported to result from collisions between alpha particles and nitrogen nuclei. By assigning masses of 18 and 17 to the two isotopes, Giauque and Johnston succeeded in calculating accurately all of the oxygen spectrum’s weak lines from the positions of the strong lines. Articles announcing the discovery of O18 and O17 appeared in the 2 March and 1 June 1929 issues of Nature.
Giauque’s unexpected discovery that normal at mospheric oxygen contained small amounts of the isotopes O17 (0.037 percent) and O18 (0.204 percent), in addition to the abundant O16 (99.759 percent), caused a problem with the chemists’ and physicists atomic mass values. Chemists arbitrarily had assigned an exact relative atomic mass of 16 to oxygen’s unsuspected three-isotope mixture as their standard for atomic mass determinations and continued to use this value even after 1929. Physicists now based their atomic mass standard on the lightest and most abundant oxygen isotope, O16. The mean atomic mass of atmospheric oxygen was then 16.0044 on the mass spectrometric scale. To convert from the physicists’ atomic mass by 1.00027 (16.0044/16). The two scales remained in use until 1961, when the International Union of Pure and Applied Chem istry and the International Commission of Atomic Weights abandoned O16 and adopted the carbon12 isotope with C = 12.0000 as the new standard. On this scale the chemists’ atomic mass of atmospheric oxygen decreased to 15.9994 from 16.0000. The change in 1961 was the first since the 1860’s, when the precise atomic mass determinations made by Jean—Servais Stas at Brussels had made 0 = 16.0000 the accepted atomic mass standard. Prior to that time chemists had assigned various masses to ox ygen: Thomas Thomson’s 0 = 1, William H. Wol laston’s 0 = 10, and Joumlns Jacob Berzelius’ 0 = 100.
Of the two oxygen isotopes Giauque discovered in 1929, O18 provided scientists with an isotopic tracer that enabled them to study the photosynthesis and respiration mechanisms and led Harold Urey, George Murphy, and Ferdinand Breckwedde in 1931 to the spectroscopic identification of hydrogen’s isotopes. Giauque’s isotopic research also demonstrated Hei senberg’s earlier prediction that a molecule retains a half-quantum unit of vibrational energy (zero point or residual energy) even in its lowest energy quantum state. Thus vibration motion within an atom did not stop at 0 K.
In his nearly sixty-year career at Berkeley, Gi auque interrupted his research on low-temperature entropy, adiabatic demagnetization, and oxygen isotopes only once. This occurred between 1939 and 1944, when he directed a classified engineering program that designed and built a mobile liquid oxygen generating plant. The government required liquid oxygen for medical and survival purposes and for use in rocket fuel. The heat exchangers designed in the program were prototypes of the large units constructed later for the liquefaction of natural gas.
When G.N. Lewis became chairman of Berkeley’s chemistry department in 1912, his policy required all faculty members to teach introductory chemistry. From the time of his appointment as instructor in 1922, and continuing every semester for thirty-four consecutive years, Giauque taught a discussion laboratory section. Until retirement in 1962, his teaching duties also included advanced physical chemistry and chemical thermodynamics. For fifteen years (1945–1960) Giauque served as adviser for Letters and Science students majoring in chemistry. Despite a no-nonsense, strictly business image that students sometimes found forbidding. Giauque was at heart a storyteller and a humorous personality who often developed close relations with his graduate students.
In addition to winning the 1949 Nobel Prize in chemistry, Giauque earned many other honors in his lifetime. They included two honorary degrees-an Sc. D. from Columbia University (1936) and an LL.D. from the University of California Section’s J. Willard Gibbs Medal (1951) and its California Section’s G. N. Lewis Award (1955). the Charles Frederick Chandler Foundation Medal (1936), and the Franklin Institute’s Elliott Cresson Medal (1937). Giauque was elected to the National Academy of Sciences in 1936, and he held membership in the American Philosophical Society from 1940. He died of com pications from a fall.
1. Original Works. Giauque’s papers are in the Bancroft Library’s Archives, University of California, Berke ley. There is s sixty-one-page oral history (1974) on Gi auque in the archives, but because Giauque’s work is his Nobel Lecture, “Some Consequences of Low Temperature Research in Chemical Thermodynamics,” s in Nobel Lectures: Chemistry 192–1962 (New York, 1964), 227–250. There is a collection of his articles, The Scientific Papers of William F. Giauque: Low Temperature Chemical and Magneto Thermodynamics 1, 1923–1949 (New York, 1969).
A bibliography of Giauque’s articles articles and reviews (1923–1978) is included in his papers. Many of them prior to 1962 appeared in Journal of the American Chemical Soceity, After 1962 Giauque published increasingly in Journal of Physical Chemistry and Journal of Chemical Physics. Some of his important articles are “The Third Law of Thermodynamics. Evidence from the Specific Heats of Glycerol That the Entropy of a Glass Exceeds That of a Crystal at the Absolute Zero,” in Journal of the American Chemical Society, 45 (1923), 93–104, with G.E. Gibson; “Paramagnetism and the Third Law of Thermodynamics. Interpretation of the Low-Temperature Magnetic Sus ceptibility of Gadolinium Sulfate,” ibid., 49 lapr1927), 1870–1877; “The Entropy of Hydrogen Chloride. Heat Capacity from 16°K, to Boiling Point, Heat of Vaporization, Vapor Pressures of Solid and Liquid,” ibid., 50 (1928), 101–122; “Symmetrical and Antisymmetrical Hydrogen and the Third Law of Thermodynamics. Thermal Equilibrium and the Triple Point Pressure,” ibid., 3221–3228, with H.L. Johnston;“An Isotope of Oxygen, Mass 18, Inter pretation of the Atmospheric Absorption Bands,” ibid., of Oxygen, Mass 17, in the Earth’s Atmosphere,” ibid., 3528–3524, with H.L. Johnston; “The Entropy of Hy drogen and the Third Law of Thermodynamics. The Free Energy and Dissociation of Hydrogen,” ibid., 52 (1930), 4816–4831; “The Calculation of Free Energy from Spec troscopic Data,” s ibid., 54 (1932), 3135–3142, with C.W. Clark; “Expriments Establishing the Thermody namic Temperature Scale Below 1degK. The Magnetic and Thermodynamic Properties of Gadolinium Phosphomo lybdate as a Function of Field and Temperature,” ibid., 60 lpr1938), 376–388; with D.P. MacDougall; and “Amor phous Carbon Resistance Thermometer-Heaters for Magnetic and Calorimetric Investigations at Temperatures Below idegK,” ibid., 1053–1060, with C.W. Clark.
II. Secondary Literature. In addition to the oral history, other accounts of Giauque are D.N. Lyon and K.S. Pitzer, “William Francis Giauque,” in University of California. In Mermoriam (Berkeley, 1985), pp. 152–156; William Jolly, From Retorts to Lasers (Berkeley, 1987), which has a chapter on Giauque and also gives a valuable historical account of the Berkeley chemistry department; “Giauque Awarded Nobel Prize for Low Temperature Research,” in Chemical and Engineering News, 27 (28 November 1949), 3571; and “William Francis Giauque,” in McGraw-Hill Modern Scientists and Engineers (New York, 1980), 191–193. Giauque’s papers also contain four short unpublished biographical memoirs, written in 1948 and 1949, totaling ten pages (no authors given).
Anthony N. Stranges