Calibrating the world's largest proving ring, capable of measuring forces up to 1,200,000 lb. With present facilities, the Bureau can calibrate such devices only to 110,000 lb. with deadweights; beyond that, indirect methods are used, with a resulting loss of accuracy. Three deadweight machines, the largest having a capacity of 1,000,000 lb., are being designed for the new NBS laboratories in order to increase the accuracy of such calibrations (page 31).

Catalytic Effects of Thermocouple Materials. In analysis of many industrial processes involving combustion of fuels such as hydrocarbons in air, it is often necessary to determine the temperature of exhaust products containing appreciable amounts of combustible gases and oxygen. This is particularly true in performance evaluations of gas-turbine power plants.

In this connection, it has long been known that platinum is a fairly good catalyst for many oxidation reactions; and early work at NBS showed a platinum shielded Chromel-Alumel thermocouple to indicate as much as 25 °F higher than the temperature of exhaust gases containing a small amount of unburned hydrocarbon. As a result of these findings, the Aeronautical Systems Division of the Air Force sponsored a program at NBS to determine the catalytic effects of all of the commonly used thermocouple materials.

The experimental technique utilized the resistance of electrically heated test wires of thermocouple elements to determine their temperatures. Power requirements at given wire temperatures were compared for low-velocity streams of dry air and of combustible mixtures to determine the magnitude of the thermal contribution from catalytic combustion. Experiments were conducted in mixtures containing up to 3 percent by volume of hydrogen, carbon monoxide, methane, and propane in air flowing at gas velocities from 0.16 to 4.38 cm/s over the resistance element.

No catalysis was exhibited by resistance elements made of gold, silver, Chromel, Alumel, or constantan. However, all resistance elements containing either platinum or palladium catalyzed the reaction of all combustible mixtures tested except those of methane. The magnitudes of the catalytic effects and their temperature limits, up to 2,000 °F, were determined; and some anomalous behaviors were explained.

Hypervelocity Missile in a Combustible Gas.

Stabilization of hyper

sonic combustion appears to be a prelude to its application to propulsion at hypersonic speed. With a view to this application, research on stabilization and properties of this kind of combustion continued by observation of a hypervelocity missile in a stationary combustible gas. In this program, sponsored by the Air Force, experiments showed that, as in nearly all combustion processes, oscillations driven by combustion may be expected at extremely high frequency. The likelihood of generation of thrust by combustion on external missile surfaces is suggested by a large reduction of the drag coefficient of the missile under certain conditions.

The experimental technique developed for these studies also permitted observations on the structure of detonation waves. Detonation appears as a shock wave followed by a combustion wave. The observed spatial separation was converted to ignition delay times, which in these experiments ranged from about one to ten microseconds. Chemical kinetics of chain-branching and chain-breaking reactions in the mixture of hydrogen and air were used to correlate and explain the observed delays.

Fuel Flowrate Studies. Under sponsorship of the Bureau of Naval Weapons, progress continued on the fuel flowrate standardization program for the aircraft industry, and on the evaluation of flowmeters. Many transfer reference meters were calibrated with liquid hydrocarbons for the Armed Services and for industry, in order to evaluate the accuracy of calibrators installed at other locations. The results of investigations over the past several years on turbine flowmeters were summarized for publication. It was shown that metering precision better than 0.2 percent was obtained for selected ranges of flowrate and viscosity when entrance conditions and meter orientation were suitably controlled. The readout instrumentation and transient response were also discussed.

Through such work more accurate flowrate calibration facilities are being maintained in the aircraft industry, and more suitable fuel flowmeters are being developed and manufactured.

The Bureau's work in electricity is primarily the development, improvement, and dissemination of the standards of measurement for electrical quantities; and the study of the electrical and magnetic properties of materials. Electrical standards must be established that are constant over long periods of time, uniform throughout the Nation, and compatible with other standards used throughout the world. Measurements of electrical quantities directly in terms of length, mass, and time ("absolute measurements”) are extremely difficult and are made only in the realization and confirmation of electrical standards of resistance, capacitance, inductance, and voltage. Calibration work is done by comparision with these electrical standards.

Absolute Measurement of Resistance. An evaluation, based on the prototype standards of length and time, of the unit of resistance maintained at the Bureau was completed. The evaluation was based on a nominally 1-picofarad symmetrical cross capacitor having a value computable to high accuracy from its mechanical dimensions. The computable capacitor was used to establish the value of a 0.01-microfarad capacitor which then, through the medium of a frequency-dependent bridge, established the value of a 10,000-ohm resistor. A comparision of that resistor with the group of 1-ohm standard resistors currently used to maintain the NBS unit of resistance established the value of the unit with an estimated accuracy of the order of 2 ppm.

It is expected that an even higher accuracy will be attained in the repetition. of these measurements. This method will greatly improve the Bureau's ability to check maintenance of the unit of electrical resistance through the use of a group of standard resistors. When combined with repetitions of determinations of the gyromagnetic ratio of the proton, the method can be used as a check of the stability of all types of electrical standards to a much greater accuracy than previously possible.

Standard Cells Under Vibration. Standard cells are normally used under vibration-free conditions. Even so, the question frequently arises as to the effect of vibration on the electromotive force of standard cells. The Bureau exposed a number of unsaturated standard cells to vibrations at frequencies from 5 to 2,000 c/s and applied forces from 1 to 30 g. With a d-c galvanometer used as null detector, such vibrations appeared to have insignificant effects. However, when observations were made with a cathode-ray oscilloscope, an a-c voltage was observed. At frequencies above 150 c/s, the magnitude of the a-c voltage appeared to be dependent on the applied force and to range from 20 to 120 microvolts. At frequencies below 150 c/s, the magnitude of the a-c voltage was no longer proportional to the applied force because of the resonance of various components, especially free mercury above the cell septa. Vibrations of these magnitudes have no lasting effects on the electromotive force unless the cells are quite old or exhibit excessive hysteresis; in such cases several days may be required for the cells to recover their initial electromotive force.

Electrode Kinetics. Electrode reaction mechanisms may be studied in many ways. One method, presently under consideration at the Bureau, involves the measurement of the impedance of electrode-electrolyte systems as a function of the frequency of an applied alternating field. The kinetics were studied in terms of electrode relaxation processes using alternating currents of frequencies from 50 c/s to 50 kc/s. Electrodes studied included silver, cadmium, zinc, and manganese dioxide. The last two electrodes were also studied in combination in dry cell electrolytes. Equations based on electrostatics and electrodynamics were developed to explain observed phenomena. Two electrode processes occur at manganese dioxide electrodes; whereas a single process, namely, simple charge transfer, occurs at the other electrodes. The exchange currents of silver in silver nitrate and of cadmium in cadmium sulfate were found from impedance measurements at low and high frequencies to be 110 amp/m2 and 140 amp/m2, respectively.

Assembling the gage blocks used to form the NBS computable cross capacitor. The capacitor was used to re-evaluate fundamental unit of resistance in terms of the prototype standards of length and time (page 36).

Electrical Properties of Molecular Solvents at High Temperature. There are very few data on the behavior of dilute solutions of electrolytes in molecular solvents at high temperatures. Nonionic solvents at high temperatures are few in number and there is a need for extensive study of their properties. At present, the dielectric constant of boric oxide is under study at the Bureau. At frequencies from 1 to 2 kc/s, both dielectric constant and loss increase sharply above 500 °C; the dielectric constant rises to above 300 at 900 °C, indicating that very large units of the threedimensional boric oxide must be inferred if one is to interpret the electrolytic conductivity and the interionic attractions of salt solutions made with boric oxide solvent. Measurements of these values are now in progress.

Metal Oxide Solubilities in Molten Salts. Electrochemical corrosion of metals in molten salts at elevated temperatures is greatly influenced by the oxygen content of the environment. As a part of a broad study on this problem, the Bureau determined the solubility of the oxides of the metals of the first transition series (titanium, vanadium, chromium, manganese, iron, cobalt, nickel, and copper) in molten sodium chloride at 900 °C. Vanadium pentoxide reacts violently with molten sodium chloride to form chlorine, whereas the oxides of the other first transition metals are stable and only slightly soluble (less than 0.1 mole percent) in molten sodium chloride. Results for Cu2O, COO, and Mn30, indicated that the mechanism of dissolution is complex, since the metal-oxide ion ratio is several times greater than would be expected for a simple solution process. Furthermore, the solubilities of the oxides, although low, are much greater than would be indicated by thermodynamic calculations.

4

Metal-Molten Salt Interactions and Stoichiometry. Extensive mass transfer is commonly observed for metals immersed in molten salts at elevated temperatures. Such transfer limits the use of molten salts as heat exchangers and the range of use of reference electrodes in molten salt systems. Reactions between silver and molten sodium chloride were investigated at 820 °C and 940 °C in a pure oxygen atmosphere and in a nearly oxygen-free (10-5 atm) environment. In each case, silver and oxide ions are formed in the melt with the concentration of the silver ions exceeding that expected from the stoi chiometry of monovalent silver oxide. At low oxygen pressures, less than 0.2 atm, the rate of silver-oxygen interaction is zero, or independent of the oxygen pressure. At higher oxygen pressures, the rate is controlled by diffusion of reaction products away from the reaction site. If the solubility product of the oxide is exceeded locally by slowness of diffusion, metallic silver crystallizes out since the oxide is unstable. This process accounts for extensive mass transfer of metallic silver in molten sodium chloride.

Differential Thermocouple Voltmeter. A basic problem in electrical standardizing laboratories is the transition from d-c measurements (which are closely related to the fundamental standards) to a-c measurements at power and higher frequencies.

A recent contribution of the Bureau to the solution of this problem consists of the development of a "differential thermocouple voltmeter." This