Volume 28, Issue 2, 01 February 1957
Index of content:
Study of Boiling in Short Narrow Channels and Its Application to Design of Magnets Cooled by Liquid H2 and N228(1957); http://dx.doi.org/10.1063/1.1722696View Description Hide Description
A study has been made of the critical heat input, Q̇c, to liquid H2 and N2 boiling in an annular gap, of cross‐section area, A, between a vertical cylindrical heater of height, Z, and various concentric plugs. The data are in approximate agreement with calculations based on an assumption of homogeneous and frictionless two phase flow. The pressuredrop across the channel is spent primarily to accelerate the vapor being formed and the liquid carried with it. The derived relation does not contain any empirical coefficients or exponents. For ρv,2≪ρL the general equation reduces to ,where L is the latent heat of vaporization, ρv,2 the exit vapor density, ρL the liquid density, g the gravitational acceleration, Ze the hydrostatic head equivalent to the pressuredrop across the heater, and f 2 the mass fraction of vapor in the fluid leaving the heater.
Forced flow experiments give f 2. At Q̇c equal to that found in natural convection boiling, f 2=̇0.5 for both liquids. For Z=3 or 6 in. and a channel thickness of 0.022 in. the ratio of observed to calculated Q̇ is 0.95 for N2 and 0.70 for H2. The calculation is equally good for critical natural convection experiments, Ze =Z, and for subcritical as well as critical forced flow measurements, which extend up to Ze =70 Z.
The optimum operating vapor pressure is discussed for magnets cooled by ideal natural convection boiling. A simple roll‐type of coil is discussed and is shown to be superior in some respects to more conventional windings.
28(1957); http://dx.doi.org/10.1063/1.1722698View Description Hide Description
A model is proposed for the mechanism of material removal in ultrasonicmachining. From this model and fundamental physical principles, a machining rate equation is derived for the special case where solid, circular tools are used in conjunction with a puddled slurry. A comparison of the predictions of the theory with the experimental data shows that the theory gives the correct relationship between the machining rate and the variables involved in the machining process. Suggestions for the improvement of the ultrasonicmachining process are presented, and the method to be used in generalizing the theory is outlined. The poor reproducibility of the machining rate measurements is explained on the basis of the machining rate equation.
28(1957); http://dx.doi.org/10.1063/1.1722699View Description Hide Description
Performance characteristics of the Lcathode, including emission, evaporation rate, and evaporant composition, are presented, followed by an investigation of the constitution of the emitting surface. Activation of the external tungstensurface is effected by the migration thereover of strongly adsorbed BaO issuing from the pores. The active surface, which is continuously renewed during life, comprises a nearly complete oxygen monolayer on tungsten covered in turn by a complete bariummonolayer. The oxygen in the dipole layer enhances emission and drastically increases the sticking time of barium. The cathode emission is close to the maximum that can be expected from a barium activated tungsten dispenser cathode.
28(1957); http://dx.doi.org/10.1063/1.1722700View Description Hide Description
An experimental study of the L cathode is presented, including the chemistry of carbonate decomposition and barium generation, the origin of BaO in the evaporant, the barium transport mechanism through the porous plug, and the factors determining cathode life. Carbonate decomposition is accompanied in low porosity cathodes by the deleterious reaction
28(1957); http://dx.doi.org/10.1063/1.1722701View Description Hide Description
The work functions of germaniumsingle crystals have been obtained by measuring the contact potential differences between the germanium crystals and a gold reference whose work function was measured by the Fowler method. Work function values for the bombardment‐cleaned surfaces of three samples having comparable resistivities were the same within experimental error (4.78±0.015 ev). Variation in the doping concentration resulted in work‐function changes smaller than those predicted by a simple Fermi level shift. Adsorption of oxygen at pressures of about 1×10−7 mm Hg resulted in work‐function increases of about 0.20 ev. The clean surface work‐function values could be restored after O2adsorption by heating at 500°C for 15 min. For those cases in which changes were noted, adsorption of hydrogen and nitrogen at higher pressures (10−3 mm Hg) resulted in decreases of work function, while CO adsorption increased the work function. Measurements of the effects of strong electric fields and intense illumination on the work function suggested that these effects were associated with rather thick surface layers which could be removed by vacuum heating or ion‐bombardment cleaning. The photoelectric‐threshold level was near or slightly above the Fermi level for the ion‐bombardment cleanedsurfaces. After oxygen adsorption the threshold level was below the Fermi level.
28(1957); http://dx.doi.org/10.1063/1.1722702View Description Hide Description
A procedure is developed for averaging over the thermal translational motion of a gas scatterer in order to obtain observed angular averages from a knowledge of the differential cross section in the center‐of‐mass system. The results are applied to the simple case of spherically symmetrical elasticscattering with constant cross section.
28(1957); http://dx.doi.org/10.1063/1.1722703View Description Hide Description
The dependence of line width of the ferromagnetic resonance on the shape of the sample is examined on phenomenological grounds. If a damping term of the Bloch‐Bloembergen type is assumed in the equation of motion of the magnetization, the line width for an ellipsoidal sample is given by . The Landau‐Lifshitz form of damping leads to a line width which is independent of the shape of the specimen.
28(1957); http://dx.doi.org/10.1063/1.1722704View Description Hide Description
An unpinned dislocation damping model (based on the Mott‐Nabarro solution hardening theory) was previously used to explain internal friction at room temperature and is now extended into the high‐temperature range. With the model it is possible to explain the high‐temperature internal friction which has been observed in the kilocycle frequency range for measurements made at very low strain amplitudes. In addition, the damping due to dislocation climb is considered.
28(1957); http://dx.doi.org/10.1063/1.1722706View Description Hide Description
Copper wires were cold worked at room temperature to approximately a 15% reduction in area and were then irradiated at temperatures between 100° and 150°C with 1.25‐Mev electrons. The residual resistivity was observed to decrease as a function of exposure at temperatures above 100°C. The higher the temperature at which the irradiation was performed, the greater was the rate of resistivity decrease. From these data, it is concluded that one of the primary defects produced by electron irradiation becomes mobile in the temperature range, 100°–150°C. It is suggested that interstitials and vacancies produced by the irradiation initiate a process which results in the annihilation of dislocations. From an analysis of the temperature dependence of the rate of decrease, a value for the activation energy for vacancy migration in copper has been deduced: 1.28±0.10 ev.
28(1957); http://dx.doi.org/10.1063/1.1722707View Description Hide Description
In solid materials the properties of transient creep, internal friction and dispersion can be correlated on the basis of a linear theory, if the strains are small throughout. The theory, first given by Boltzmann, is developed in some detail. Dissipative properties of the material are shown to depend on a function of time called the ``creep function'' φ(t).
The general condition 1/Q≅constant is examined. This condition is compatible with a creep function of logarithmic type. Velocity dispersion of plane or spherical waves is very small in such a medium. Graphs of 1/Q, phase and group velocities vs frequency are given, and a quantitative relationship between creep and internal friction constants is derived. The results are substantially applicable to the problem of losses in dielectrics, by carrying out the electromechanical analogy.
28(1957); http://dx.doi.org/10.1063/1.1722708View Description Hide Description
The solubility of air bubbles is observed in a variety of experiments. Freely rising bubbles exhibit a solution rate more than twice that of bubbles which are stationary: i.e., trapped on the walls of the container. The theory of solution of stationary bubbles is given but the theory of a free bubble is difficult and has not been solved. Stationary bubble observations lead to a determination of the coefficient of diffusivity: 2.9×10−5 cm2/sec at 27°C; the temperature dependence is considerable, obeying Einstein's equation relating diffusion and viscosity. The effect of surface contamination on diffusion is analyzed; bubble solution in the presence of contaminants is not greatly altered. Experiments are described showing that air bubbles of dimension less than 1μ lodged on hydrophobic particles are not soluble and can exist indefinitely. It is shown that hydrophobic particle residues frequently remain after bubble solution, even in relatively pure water. These residues function as bubble nuclei for boiling or for cavitation.
28(1957); http://dx.doi.org/10.1063/1.1722709View Description Hide Description
Two maseroscillators have been constructed and operated. The behavior of one maser has been observed under various operating conditions by comparing it with the second maser which is used as a reference standard. The expeimental setup is shown and experimental results are given. The frequency‐pulling effects of the beam and the oscillation amplitude as a function of cavity‐tuning are compared with the theoretically expected behavior. It is shown that the simple behavior predicted by theory exists only for oscillation frequencies which are well removed from line center.
The theoretical curves are obtained from a new analysis which includes the velocity distribution in the beam. A comparison is made between the velocity‐distributed and uni‐velocity theories. It is shown that the introduction of the velocity distribution drops the maximum beam efficiency from 1 to 0.76 and causes a region of stability to appear in the beam frequency‐pulling function.
28(1957); http://dx.doi.org/10.1063/1.1722710View Description Hide Description
A variational method is used to calculate the distribution of energy reflected from an echelette diffraction grating. Experiments performed using a klystron generating 3.2‐cm electromagnetic waves which impinge upon a large scale grating agree with the calculations to within a few percent, the error being that expected from the variational formulation.
28(1957); http://dx.doi.org/10.1063/1.1722711View Description Hide Description
The transversely magnetized slab operating in a parallel‐plane‐type transverse electric mode may be examined from an operational viewpoint. Characteristics of the wave‐guide mode emerge from considerations of symmetry and some results are derived that are not evident from a first‐order perturbation theory. Operators are developed associated with various transverse strata of the wave‐guide cross section corresponding to slabs of different media within the guide. The use of these operators is systematized in application to propagation problems in both rectangular and coaxial type transmission lines. Ferromagnetic resonance is explored in ferrites obeying a reactive Polder tensor‐type description and some features of resonance are discussed.
28(1957); http://dx.doi.org/10.1063/1.1722712View Description Hide Description
Merz's picture of spike‐shaped domains of reversed polarization emerging from an electrode is analyzed with particular attention to its electrostatic aspects, taking into account the inhomogeneous piezoelectric distortions which accompany domain formation. The depolarization energy of a domain is evaluated. The interaction energy of the applied field and the depolarization field is shown to be negligible. The charge flowing into the electrodes as a domain expands is shown to be related primarily to the reversal of spontaneous polarization, rather than to the changes of the depolarization field. Our final results are similar in form to the expression originally given by Merz. Although a detailed domain wall theory is not yet available, we have combined an existing domain wall estimate with our electrostatic expressions, and this results in an implausibly large critical nucleus. Some effects of nonlinearities in the dielectric behavior and of internal charge distributions are discussed.
28(1957); http://dx.doi.org/10.1063/1.1722713View Description Hide Description
Area contacts of twenty different metals were made on n and psilicon. The contacts were applied by the use of jet plating techniques, with the exception of alkali metal contacts which were pressure contact or mercury amalgam contact. A qualitative correlation is shown to exist between the work function of the metals and the rectification of these metals on n and psilicon. I–V characteristics taken on eleven of these metal‐silicon contacts lend further support to this picture. Consideration of a quantitative work function model is made difficult due to the many errors and interpretations involved in using metalwork function values.
Transistor structures were made and studied for several of the metal‐silicon contacts. From analysis of this transistor data it is found that an excess current three orders of magnitude greater than theory predicts, must be present in the diodes made from these metals. This excess current is not adequately explained by any presently known mechanism.
28(1957); http://dx.doi.org/10.1063/1.1722715View Description Hide Description
Heretofore the observation of grain boundaries with the field emission microscope has been limited by the accidental and infrequent appearance of suitable specimens. A process for preparing polycrystallinetungsten samples by bending a tungsten wire, etching a tip at the bend and heating is described. Both large and small angle grain boundaries result. As many as five tungsten grains may share a cross section of less than one micron in diameter. All the parameters required to describe the geometry of the boundaries may be obtained experimentally. Detailed observations on the dislocation structure of boundaries with angles as great as 6° are potentially possible. However, the field emission intensity from the region immediately adjoining and including the grain boundary must first be increased appreciably. Possible methods for achieving improved electron emission from the grain boundary region are discussed.
28(1957); http://dx.doi.org/10.1063/1.1722716View Description Hide Description
The buildup of electroluminescent brightness is compared to the buildup of photoluminescence under various conditions which affect the electron population in traps. A qualitative explanation is proposed as to the mechanism of buildup and conclusions are drawn which confirm the observation that electroluminescence is confined to small regions in the phosphor crystals.
28(1957); http://dx.doi.org/10.1063/1.1722717View Description Hide Description
The integrated effective resonance neutron absorption cross section of uranium‐238 cylinders has been experimentally investigated in the Hanford Test Pile. The results of reactivity measurements were interpreted to yield the effective resonance integral as a function of the surface‐to‐mass ratio of the uranium cylinders. The experiments were extended to determine the Doppler coefficients resulting from heating the uranium in a constant neutron spectrum in the reactor. The effective resonance integral, , can be approximated as a function of the surface‐to‐mass ratio by .The Doppler coefficient of resonance escape probability for the reactor is ,and the value assigned to the coefficient of the volume absorption of the resonance integral is .