Theoretical Physics Conference, 1939
From GW Encyclopedia
THE FIFTH WASHINGTON CONFERENCE ON THEORETICAL PHYSICS
The theory of low-temperature physics was the focal point of discussion held in Washington January 26 to 28 under the joint auspices of the George Washington University and the Carnegie Institution of Washington, acting through its Department of Terrestrial Magnetism. The properties of liquid helium and of liquid hydrogen and deuterium, the interpretation of data on the adiabatic demagnetization of paramagnetic salts at temperatures below 1° K, and the phenomenon of superconductivity were the subjects discussed. In addition, the theory of certain very recent developments in nuclear disintegration and the theory of nuclear binding forces were discussed. Professors Niels Bohr, Harold C. Urey, Enrico Fermi, F. London, G. E. Uhlenbeck, J. H. Van Vleck, H. A. Bethe, G. Breit, E. U. Condon, I. I. Rabi, A. E. Ruark, F. Bitter, H. Grayson-Smith, F. Seitz, O. Stern, L. Rosenfeld, and many other physicists active in research were present.
Certainly the most exciting and Important discussion was that concerning the disintegration of uranium of mass 239 into two particles each of whose mass is approximately half of the mother atom, with the release of 200,000,000 electron-volts of energy per disintegration. The production of barium by the neutron bombardment of uranium was discovered by Hahn and Strassmann at the Kaiser-Wilhelm Institute in Berlin about two months ago. The interpretation of these chemical experiments as meaning an actual breaking up of the uranium nucleus into two lighter nuclei of approximately half the mass of uranium was suggested by Frisch of Copenhagen together with Miss Meitner, Professor Hahn's long-time partner who is now in Stockholm. They also suggested a search for the expected 100,000,000-volt recoiling particles which would result from such a process. Professors Bohr and Bosenfeld had arrived from Copenhagen the week previous with this news, and observation of the expected high-energy particles was independently announced by Copenhagen, Columbia, Johns Hopkins, and the Carnegie Institution shortly after the close of the Conference. Professors Bohr and Fermi discussed the excitation energy and probability of transition from a normal state of the uranium nucleus to the split state. The two opposing forces, that is, a Coulomb-like force tending to split the nucleus and a surface tension-like force tending to hold the "liquid-drop" nucleus together, are nearly equal, and a small excitation of the proper type causes the disintegration.
An interesting connection between nuclear physics and low temperatures, as pointed out by Professor Teller, is the fact that the balance between zero-point energy and potential energy is very similar in the liquid-droplet model of the atomic nucleus and in liquid helium-II. In fact, a similarity-transformation of linear dimensions by 1/105 and energies by 1010 , in accordance with the experiment, will transform the model of liquid helium-II into something very similar to the liquid-droplet model of the nucleus.
Professor Uhlenbeck introduced the discussion of the differences in the physical properties of the isotopic modification of hydrogen, H2, HD, and D2 at low temperatures with a consideration of how these differences arise. Classical physics would lead one to expect that the equilibrium properties, that is, properties which do not involve the time explicitly, as for example, vapor-pressures and molecular volumes, should be independent of the molecular weight if the interatomic forces of atoms of H and D are alike. These forces for isotopic atoms are so nearly the same that it has been concluded that the differences in these equilibrium-properties of the isotopes must be the result of quantum-effects. Two essentially different quantum-effects arise. The first arises because the de Broglie wave-length of a gas molecule depends upon its molecular weight and at low temperatures becomes so large compared with the size of the molecule that the diffraction-effects of the de Broglie waves upon collision of gas molecules become important. This has an important bearing upon the equation of state of the gas and the differences between the viral coefficients of H2 and D2 may be accounted for upon this basis. The second class of quantum-affects arise because of the large differences between the zero-point lattice-energies. These are responsible for the large differences in the vapor-pressures, molecular volume, triple points, and heat capacities of H2, HD, and D2. Calculations based upon simple harmonic oscillators and the simple considerations of Debye's theory of the solid state have been able to account quantitatively for the observed differences in the properties. Professor London gave an account of the theoretical considerations of Mr. Hobbs and himself based upon a similarity of the intermolecular-force fields and high zero-point energies in condensed hydrogen and helium. Thus the hydrogen molecules are in affect contained in small volumes having a large confining force at the boundaries but only small forces in the interior. Calculation of the heat of vaporization of the solids and molecular volume at 0° K are in agreement with the experimentally derived values. The anomalous character of the heat-capacities of the solid and liquid isotopes were discussed.
The differences in the properties of the ortho and para varieties of H2 and D2 were discussed. Dr. Karl Cohen, of Columbia University, outlined a theory developed by Professor B. C. Urey and himself to account for the larger heats of vaporization and smaller molecular volumes of the rotating (j = 1) variation of H2 and D2. These differences in the properties of the ortho and para varieties arise because of the rotation of the ortho molecules. One important affect is that the centrifugal force resulting from the rotation stretches this inter-atomic distance in the molecule and in effect makes the rotating molecule larger.
Professor F. London of the Institut Henri Poincare of the University of Paris, who has been Visiting Professor at Duke University for some months, recently developed a theory which accounts for many of the phenomena observed in liquid helium at and below the “λ” transition point. As is well known, the phenomenon is such that the liquid helium does not become a solid upon lowering its temperature but, after passing through the λ point (sudden change of specific heat at 2.2oK), it becomes a superfluid having extremely low viscosity, a high heat-conductivity in the region of the λ point, and other very strange effects such as the "fountain-effect" and an astounding ability to "creep" up the walls of containers and tubes are observed. London has proposed that this behavior is a condensation of the Bose-Einstein gas into the lowest energy-states. However, the liquid helium is not an ideal Bose-Einstein gas since the effort of Yan-der-Waals forces is to create some spatial order. This will give a decrease in the density of the levels for the lowest states (proportional to say k4dk instead of k2dk as for free particle a where k is the wave-amber). It can be shown that such a non-ideal Bose-Einstein gas will have a jump in the specific-heat curve at the point where particles begin entering the lowest energy states. Professor London emphasized a possible analogy of the behavior of helium-II to the superconductivity-electrons in metals, and to the diamagnetism observed in many types of solids. He also proposed a theory for the viscosity and heat-conduction of helium-II (bearing an analogy with electrons in a metal} which makes use of free particles obeying a Boss statistics. The work of L. Tisza on these phenomena was discussed; Tisza states that not all of the atoms enter the lowest energy state and therefore a few of them have a finite momentum and exert a pressure. Only these few give rise to the viscosity which one observes; the heat-conduction arises from the change in pressure of these excited atoms. Professor Fermi and Professor London proposed several experiments which would throw further light on these problems.
Recent consideration bearing on the method of Giauque and Debye for obtaining temperatures below 1° K by the adiabatic demagnetization of a paramagnetic salt and the property of matter at these temperatures were discussed by Professor J. H. Van Vleck of Harvard University. The well-known method of liquifying a gas such as helium consists in isothermally compressing (thus lowering the entropy of the system) and then expanding adiabatioally (thus lowering the temperature while the entropy stays constant); magnetic cooling is quite similar— the magnet is the "entropy squeezer" and the field is released after establishing thermal insulation (pumping out the heat -transfer gas around the paramagnetic salt).
The experiments of Simon, Kurti, and coworkers on NO4Fe(SO4 )2 • 12H2O at temperatures below 1° K was discussed by Dr. M. H. Hebb and Dr. C. F. Squire. The absolute temperature scale was established in these experiments not by use of Curie's law relating the magnetic susceptibility with the temperature, but through the thermodynamic relationship T abs = ∆Q/∆S. The Curie law which is valid at temperatures of 1° K is no longer valid at the temperatures obtained by magnetic cooling because just the interaction forces which produce the cooling effect are responsible for the variation from Curie's law. Professor Van Vleck discussed the theory of these interaction forces — the splitting of electron-energy states by the crystalline electric field and the spin-spin interaction of the paramagnetic ions causing further splitting. Partition-functions and specific heats were calculated for several salts and agreement with experiment indicates that a representation of the local field acting on the spins of tibia type proposed by Onsager is better than the classical one of Lorentz. Spin-spin interaction could only be partially solved and agreement with experiments remains only qualitative. The theoretical interpretation of the ferromagnetism (hysteresis effects) found in iron-alum by Simon, and Kurti at 0.034 K remains quite unclear.
The theory of paramagnetic relaxation-time and the experiments of Professor Gorter and other Dutch physicists were discussed in great detail. Just as one has absorption and dispersion of the electric rector light, so one can have magnetic absorption and dispersion at about radio frequencies. The oscillators are the electron-spins, which are damped by the spin-spin coupling in about 10-9 /sec; the damping-time can be enormously increased by applying an external magnetic field, tinder these conditions the spin-spin coupling is too weak to "finance a turn over" of the dipoles but the spin-lattice interaction can turn then over.
The time required for spin-lattice interaction to establish thermal equilibrium has great significance since it might be the limiting factor in reaching still lower temperatures than have been attained up to now. According to Van Vleck and Kronig, non-adiabatic coupling between lattice vibrations and electronic motion determines this time. Quantitative calculations are still in a preliminary stage. Relaxation-time between nuclear-spin moments and electron-spins or with the lattice are very long and the cooling to extremely low temperatures by this interaction would require at least a day before equilibrium would be established. Professor Teller discussed the calculation for the time-effect. The matrix-element (perturbation-energy) which gives the transition-probability between the nuclear spin and electron-spin is quite small for paramagnetic salts. Perhaps the interaction with electrons in metals would be sufficient to cut down the relaxation on-time considerably.
The theory of superconductivity was briefly discussed by Professor F. London. It must be emphasized that the magnetic behavior is as important as the superconduction. The microscopic picture is not yet clear. From the behavior of liquid helium and that of diamagnetic in solids it is probable that superconductivity is a cooperative phenomenon causing very low level-densities for low energies.
February 1, 1939
G. F. Squire, University of Pennsylvania
F. G. Brickwedde, National Bureau of Standards
E. Teller, George Washington University
M. A. Tuve, Carnegie Institution of Washington
Photographic Credit: n/a
Author or Source: President's Papers/RG0002; Office of Public Relations
Document Location: University Archives
Date Added to Encyclopedia: January 12, 2007
Prepared by: Lyle Slovick
For more information about GW history
Special Collections Research Center 
The Melvin Gelman Library 
The George Washington University 
2130 H Street, NW Suite 704
Washington, DC 20052
Please send us your questions and comments about the encyclopedia.
This site is maintained by the Special Collections Research Center and the Web Development Group.