No abstract

I N connection with our measurements of the heat capacity of niobium in the normal and superconducting state, it was desirable to find a secondary thermometer for use in the range 2° to 20°K with the following requirements: reproducibility after cycling to room temperature high sensitivity, negligible change of calibration in the presence of a magnetic field. Such a thermometer has been the subject of a wide search by investigators in low temperature physics for many years. The most commonly used secondary thermometer, leaded phosphor-bronze, is reproducible but has sensitivity only below 7.2°K owing to the superconductivity of the lead. However, in addition to being insensitive above 7.2°K, it is also insensitive below 7.2°K in the presence of a magnetic field.The fact that carbon possesses desirable resistance-temperature properties in the liquid helium temperature range is well known, and considerable use has been made of carbon thermometers in various forms, such as india ink, Aqua Dag, etc. None of these carbon thermometers showed reproducibility from day to day, and in some cases the sensitivity was not sufficient. At the Low Temperature Symposium held at the National Bureau of Standards in March, 1951, J. R. Clement of the Naval Research Laboratory pointed out that a very convenient and sensitive carbon thermometer was available commercially in the form of carbon radio resistors, and that those rated at one watt manufactured by the Allen-Bradley Company showed high sensitivity in the helium range.Because a one-watt resistor was larger than desirable for mounting in our specimen of niobium, we chose a half-watt resistor at random from the stock on hand at Columbia University. The outer plastic covering was ground away so as to expose the carbon. The resistor was then covered with a 0.001-inch layer of clear glyptal lacquer and baked. It was then cemented into a cylindrical hole bored in the niobium. The total weight of the resistor and leads was then less than 0.2 gram. During the period April through August, 1951, the specimen and thermometer were cooled from room temperature to about 2°K seven times. Using a constant measuring current of 10 microamperes and a Wenner potentiometer, a careful resistance-temperature calibration was made each time using the vapor pressure of liquid helium. In this temperature range about 50 different calibration points were obtained. These points were found to lie on a smooth curve with over 90 percent of the points differing from the curve by less than 0.002 degrees.

No abstract

No abstract

The heat capacities of three samples of lanthanum have been measured in the temperature range 1.6 to to 6.5°K. A four-constant formula was found which represented to high precision the resistance-temperature relation of the carbon composition resistance thermometer from 1.6 to 7.2°K. Two superconducting transitions were found in each sample: one at 4.8°K and the other at 5.9°K. These are associated respectively with the hexagonal close-packed and face-centered cubic modifications of the metal. Below 2.5°K, a magnetic field of 10 000 gauss was found insufficient to quench completely the superconducting phase. The values of the normal heat capacity constants for the purest sample, averaged over the two crystal structures present, were determined by a thermodynamic analysis of the data to be y = (24.1 ±0.6) X10~4 cal/mole (°K) 2 , 0=142±3°K. The data are further analyzed for evidence of a law of corresponding states among superconductors.

EDITOR 635 sen ted in Fig. 1. The abscissa represents distance along the normal to the original shock with the origin at the corner, the scale being so arranged that the original shock has advanced a unit distance. The ordinate is the reduced pressure defect as determined by Lighthill, P= (p2-p)/e(p2-pi), where p is the pressure along the wall, pi and pi are respectively the pressures ahead of and behind the original shock and e is the angle of the bend in the wall. The sign of the quantity P is observed to be positive for both concave and convex corners. The theoretical treatment shows a logarithmic infinity in the pressure at the corner. This, of course, would not be obtained experimentally, but a relative maximum in the pressure does exist there.

I N a recent note to The Physical Review, Ogg 1 reported a large decrease in electrical resistance accompanying the rapid freezing of an approximately one-molar solution of sodium in liquid ammonia, and also the production of a persistent current, as in the classical Kammerlingh Onnes experiment, in a rapidly frozen ring of solution. The purpose of the present note is to report an unsuccessful attempt to repeat these experiments. It is not our intention to maintain that the effect does not exist, but rather to point out that the conditions under which the effect may be obtained are obscure.In all the experiments to be described, the solute was extremely pure distilled sodium obtained in thin-walled glass ampoules from the General Electric Vapor Lamp Company. These contained about 0.6 g of sodium which was dissolved in anhydrous liquid ammonia (99.95 percent) to make an approximately one-molar solution. Care was exercised to avoid contamination with moisture, but no attempt was made to measure the concentration of the solution with great accuracy because of the change in concentration always taking place due to evaporation of ammonia. For all the experiments the concentration, however, was within the range of 0.5-2 molar in which two liquid phases separate out on slow cooling. The magnetic field for the production of persistent currents was produced by an electromagnet capable of providing a field up to 1200 oersteds at a pole separation of 9 cm, the latter being sufficient to allow a Dewar flask to be inserted. Liquid nitrogen was used as a refrigerant for rapid freezing of the solutions.In the first experiments a closed system was used. The glass ampoule containing the sodium was crushed by a magnetically operated hammer and the resulting solution was forced into a glass tube one of whose vertical arms (A" bore, ^f" wall thickness, 6" length) was equipped with four sealed-in leads for resistance measurements by the potentiometer method. At the bottom of the tube was a glass ring of &" bore, A" wall thickness, and a If" mean diameter. It was found impossible with this apparatus to prevent separation of the liquid column and cracking upon freezing, so that no resistance measurements were obtained. After freezing the solution in the ring while in a field of 1200 oersteds, and then removing the magnet and bringing a sensitive compass needle near the Dewar flask, no persistent current was detected, although the compass needle gave an unmistakable response to a field as low as 0.05 oersted. Four trials were made, in two of which the Dewar flask was removed momentarily so that the compass needle could be brought even closer to the ring.i R. A. Ogg, Jr., Phys. Rev. 69, 243 (1946). See also April 1 and 15, 1946 issue, for erratum. R ECENTLY, in an article with the above title, 1 E. L.Hill considered the problem of determining the velocity as a function of position for a fluid rotating with constant angular velocity, according to relativity theory. However, his method of calculation appears questionable

No abstract