Index of content:
Volume 28, Issue 4, 01 April 1957
28(1957); http://dx.doi.org/10.1063/1.1722759View Description Hide Description
Germanium single crystals pile‐irradiated at 20°C and 1014 to 1017 n/cm2 etch differently from similar crystals which have not been irradiated. The irradiated crystals when etched with CP‐4 show a finer surface granulation. This difference in etching behavior is reduced on annealing. Kinematic studies indicate an activation energy for the annealing process between 1.6 and 1.8 ev. The etching behavior of silicon single crystals shows a similar change on pile irradiation. The present work, though preliminary in nature, suggests a new experimental approach to the study of radiation damage.
28(1957); http://dx.doi.org/10.1063/1.1722760View Description Hide Description
Water‐tank observations of the flow phenomena associated with boundary‐layer transition have revealed that a two‐dimensional discrete vortex line, which is considered to be the consequence of an amplified perturbation wave, has a strong tendency, in shear flows, to form three‐dimensional vortex loops with a marked transverse wavelength. The formation and development of the vortex loop is found to be the essential feature preceding the creation of a turbulent spot which takes place near the top of the vortex loop and near the outer edge of the boundary layer. This formation and development provides the guiding principle of laminar‐to‐turbulent transition in wakes and jets as well as in the boundary layer.
28(1957); http://dx.doi.org/10.1063/1.1722761View Description Hide Description
General equations governing the hydrodynamic behavior of an ideal compressible fluid in which chemical reactions and internal relaxations proceed are formulated. For one‐dimensional flow the equations are transformed to characteristic form, in which the ``frozen'' or high‐frequency sound velocity plays a role analogous to the unambiguous sound velocity in the nonreactive case.
28(1957); http://dx.doi.org/10.1063/1.1722762View Description Hide Description
Measurements of the intrinsic tensor permeability [μ] of unsaturated Niferrite at 9300 mc, using cavity perturbation techniques, have revealed an anomalous resonance for negative (anti‐Larmor) circularly polarized fields. The resonance has a line width of 70 oe and occurs very near the coercive force which is 13.2 oe. The usual resonance for positive (Larmor) circularly polarized fields occurs at 2750 oe. An interesting consequence of the low‐field resonance is that below saturation K″ is negative, where K=K′−iK″ is the off‐diagonal component of [μ]. Above saturation K″ is positive. The change in sign of K″ is shown not to violate energy considerations.
These effects are explained by a theory involving rf demagnetizing poles on Bloch walls based on a model with a physically probable domain structure. The theory extends the Polder‐Smit mechanism in unmagnetized media to include applied dc fields and circularly polarized rf fields. It is applicable to all ferrites, both mono‐ and polycrystalline, having domain walls free to move in dc fields of the order of the coercive force. Since the anomalous resonance depends upon the domain structure and appears to occur generally, it is a probe of considerable usefulness for the study of magnetization processes and high‐frequency phenomena in ferrites.
28(1957); http://dx.doi.org/10.1063/1.1722763View Description Hide Description
An equation relating the potential applied to and the current flowing in a liquid dielectric has been derived. The theoretical equation predicts that beyond a certain threshold potential the current is independent of the applied potential and is proportional to the number of ions produced per unit volume per unit time.
Measurements of the number of ions produced in cyclohexane as a function of temperature and water concentration were made. It was found that in the pure state about 200 ions/cm3‐sec were produced, while in the water saturated cyclohexane at 50°C about 1.6×105 ions per cm3 per sec were produced. It was concluded that the conductance in the pure state was due to ionization by cosmicradiation.
When the cyclohexane was saturated with water, a large increase in the conductivity was observed as the cyclohexane passed through the freezing point. This behavior corresponds to an increase in the dielectric constant as noted by Seyer and Barrow.
Deposit Spot Patterns from Low‐Index Planes of Metal Single Crystals in a New Theory of Cathode Sputtering28(1957); http://dx.doi.org/10.1063/1.1722764View Description Hide Description
The generation of the deposit spot patterns from low‐index planes at lower and higher ion energy is explained by mechanical double and triple collisions of the perpendicularly incident ions with lower and upper surface atoms of the target plane. The effective collision spheres are assumed to be determined by the largest closed electronic shells of the ion and the target atom with radii smaller than the radii of the largest electronic orbits of these shells. A second factor has to be taken into account for the explanation of the shape and other features of the single spots of the patterns. This is the attenuating influence of the different electron densities in the paths of the ion within the lattice, because the largest closed electronic shells exceed the size of the collision spheres and produce the highest electron density in close‐packed rows of the lattice.
These two concepts are sufficient to explain the number and arrangements of the spots in the patterns of low‐index planes, the specific shape of the single spots, the alterations of the spot patterns connected with higher ion energy or with accidental deviations of the prepared crystal plane from the ideal crystallographic plane, the deviation of the symmetry axis of the single spots in (110) fcc patterns from the diagonal direction of the unit area, and finally the different shape and size of the single spots in (111) spot patterns of silver and copper.
28(1957); http://dx.doi.org/10.1063/1.1722765View Description Hide Description
Equations are derived for the impurity distribution which arises by evaporation of impurities from the surface of a homogeneously dopedsemiconductor. The rate of evaporation is assumed to be proportional to the surface concentration. When a rectifying metal contact is made to a semiconductor with such an impurity distribution, the dependence of capacitance on the applied voltage can be used to determine the proportionality constant between rate of evaporation and surface concentration.
28(1957); http://dx.doi.org/10.1063/1.1722766View Description Hide Description
Some experimental results are presented of the flow generated by the explosion of a sphere of high‐pressure gas. Glass spheres of 1, 2, and 5‐in. diam containing air, He, or SF6 at overpressures up to 21 atmos were used to produce a spherical analog of the shock tube. The results show that the shock sphere may be applied to the study of explosions, implosions, and spherical wave interaction under controlled and known initial conditions.
28(1957); http://dx.doi.org/10.1063/1.1722767View Description Hide Description
Diffraction of a wave by an aperture of any shape in a thin screen is treated by a new method—``the geometrical theory of diffraction.'' This is an extension of geometrical optics which accounts for diffraction by introducing new rays called diffracted rays. They are produced when incident rays hit the aperture edge and they satisfy the ``law of diffraction.'' A field is associated with each ray in a quantitative way, by means of the optical principles of phase variation and energy conservation. In addition ``diffraction coefficients'' are introduced to relate the field on a diffracted ray to that on the corresponding incident ray.
By this method a simple formula is obtained for the field diffracted by any aperture. It yields the field in the aperture, the diffraction pattern and the transmission cross section. Explicit formulas and numerical results are given for slits and circular apertures. The accuracy of the results increases as the wavelength decreases, but they are useful for wavelengths even as large as the aperture dimensions.
28(1957); http://dx.doi.org/10.1063/1.1722768View Description Hide Description
The magnetostriction of oriented polycrystalline MnBi was found to be higher than any published value for a metal. A value of −250×10−6 was measured parallel to a field of 22 000 oersteds oriented perpendicular to the c axis of the aligned hexagonal crystallites. This field magnetized the specimen to only slightly more than half of saturation. A tentative value of −800×10−6 for the magnetostriction at saturation was obtained by plotting the magnetostriction as a function of (B‐H)2 and extrapolating to saturation.Magnetic anisotropy constants of K 1=9.1×106 ergs per cm3 and K 2=2.6×106 ergs per cm3 were obtained by fitting a calculated curve to the experimental magnetization curve.
28(1957); http://dx.doi.org/10.1063/1.1722769View Description Hide Description
This paper extends earlier work by Johnson and Donal on the anomalous flicker effect, observed in tubes with a tungstencathode under space charge limited condition, and attributed to positive ions emitted by the cathode. This extension involves the observation of the individual current pulses causing the effect and the measurement of the flicker effectnoise spectrum over a wide frequency range. The results indicate that these pulses are due to sudden bursts of about 104 positive ions emitted by the cathode in less than 1 μsec and trapped in the space charge for many μsec. The trapping time of the ions is determined from the shape of the pulses and from the noise spectrum, and the results are found to agree. An experiment that verifies directly that the effect is caused by positive ions also shows the way to eliminate the effect. A discussion of the nature of these sudden bursts is offered and various mechanisms for the observed decay of the pulses are presented.
28(1957); http://dx.doi.org/10.1063/1.1722770View Description Hide Description
The Polder tensorsusceptibility is derived for a single crystalferrite possessing cubic symmetric anisotropy energy. The tensor is derived for the 110 plane of the cube and its elements are exhibited as a function of the angle that the applied field makes with a cube edge, as well as of saturation magnetization, applied field, anisotropy constant, and frequency. The analysis is valid for ferrites with anisotropy fields much less than externally applied magnetic fields. The tensor is no longer diagonal for circular polarizations, except along the 100 and 111 axes.
Exact Current‐Voltage Relation for the Metal‐Insulator‐Metal Junction with a Simple Model for Trapping of Charge Carriers28(1957); http://dx.doi.org/10.1063/1.1722771View Description Hide Description
The exact solution to the diffusion equation for one kind of charge carrier is obtained for the metal‐insulator‐metal junction where trapping of the charge carriers may occur in the body of the insulator. The effect of surface states is not considered. The electric potential distribution is given in closed form and the exact dc current‐voltage relation is given in parametric form for any degree of trapping. A graph of the current‐voltage relation is shown for the special case where the two boundary metals are the same.
A brief description of some experimental difficulties is given when test junctions were prepared entirely by the vacuum evaporation method.
28(1957); http://dx.doi.org/10.1063/1.1722772View Description Hide Description
The high current density existing at near‐breakdown reverse bias of an n‐p junction produces, by Ohm's law, an electric field. If the effect of this field is included in the solution for diode current, it is found that it can significantly affect the breakdowncharacteristic. It can, for example, produce an increase in injected photocurrent resembling multiplication, even though multiplication is not occurring in the junction itself. Alternatively, if multiplication is present, it can produce instability and breakdown before the avalanche in the space‐charge region itself becomes unstable (i.e., before the multiplication factor becomes infinite). These predicted phenomena have been observed in a series of germaniumn‐p junctions. Closely allied with these phenomena is the observation of a region of ``soft breakdown'' in many germanium junctions, which appears to be a bulk effect.
28(1957); http://dx.doi.org/10.1063/1.1722773View Description Hide Description
Various aspects of the magnetic anisotropy of electrodeposited elongated single‐domain Fe and FeCo alloy particles are examined with a view to better understanding their process of magnetization. The initial increase of coercivity with increase of the angle between alignment direction and measuring field is in qualitative accord with the prediction of the chain‐of‐spheres model with fanning, as previously proposed, and in contrast to coherent rotation models. Analyses of high‐field torque curves and of the fields at which rotational hysteresis vanishes suggest that an anisotropy is present which is a little greater than predicted by the chain‐of‐spheres model but less than that predicted by the Stoner‐Wohlfarth ellipsoid model, for the observed dimensional ratios. Calculations of the rotational hysteresis in single domain‐particles are extended to the chain‐of‐spheres model. A study of the rotational hysteresis enables a relatively sensitive choice between several models of the magnetization process. Comparison of the observed and predicted values for the rotational hysteresis integral, indicates strongly that a nonuniform rotation process, like fanning in a chain‐of‐spheres, is operative in electrodeposited elongated single‐domain particles.
28(1957); http://dx.doi.org/10.1063/1.1722774View Description Hide Description
Under certain conditions an electron beam can be subjected to a process of scaling in which the transverse components of position, velocity, and force are scaled by a common factor while longitudinal components remain unchanged. In analyzingbeams produced by electron guns under conditions where thermal velocity effects have to be taken into consideration, the application of transverse scaling simplifies the treatment by allowing a reduction in the number of independent parameters required. Design curves for Pierce‐type guns are reproduced in normalized and condensed form. These curves may be used in choosing design parameters to produce a prescribed beam.
28(1957); http://dx.doi.org/10.1063/1.1722775View Description Hide Description
The equilibrium distribution of copper between octahedral and tetrahedral sites in the ferrites varies appreciably with temperature. The rate at which these ions approach their equilibrium distribution at a fixed temperature, after quenching from a higher temperature, is much greater for the slightly reduced materials.
Previously unreported deviations from a smooth variation of saturation induction with annealing temperature have been observed in ferrites containing nickel and/or copper that have been prepared to achieve a high degree of homogeneity and hence to be relatively free from defects.
28(1957); http://dx.doi.org/10.1063/1.1722776View Description Hide Description
The dependence of saturation induction at room temperature upon composition for the substitution of nonmagnetic ions into BaFe12O19, into BaM2Fe16O27 (where M2=Ni2, Ni1.5Cu0.5, Zn2), and into the mixture BaNiFe14O23 is given. Evidence is presented for the existence of points of magnetic compensation in a number of series wherein aluminum is substituted for iron. The distribution that gallium takes in these structures appears to depend upon the sites occupied by divalent metal ions.
28(1957); http://dx.doi.org/10.1063/1.1722777View Description Hide Description
The dynamic behavior of 4–79 molybdenumPermalloy operating in a reversible region of its hysteresis loop is analyzed theoretically and experimentally. A useful expression is presented which relates B, H, time, and magnetic viscosity. This expression is applied to determine the behavior of the magnetic material when magnetized (1) by sinusoidal currents having frequencies in the kilocycle/second region and (2) by step‐function currents producing transient voltages with time constants in the millimicrosecond region. The theoretical results show that there is a correlation between the behaviors under these two extreme cases of current excitation. This correlation is verified by measurements made on thin tapes having nominal thicknesses of ⅛, ¼, ⅓, and ½ mil and the results of these measurements agree with theory. Various effects of magnetic viscosity on the dynamic behavior of the core material are shown and an expression is derived which makes it possible to solve theoretically certain kinds of magnetic engineering problems whose theoretical solutions have not been achieved previously. These problems pertain to devices such as shift registers, high‐frequency transformers, and high‐frequency magnetic amplifiers which contain cores made of thin ferromagnetic tape.
28(1957); http://dx.doi.org/10.1063/1.1722778View Description Hide Description
Single crystals of barium oxide were exposed at a temperature of 1423°K or 1473°K to barium vapor of different pressures ranging from 3.4×10−5 mm Hg to 7.6×10−2 mm Hg. The resulting blue coloration was measured in an absorption spectrometer, which provided estimates of the number of color centers per cm3, NF . The relation between NF and the number of barium atoms per cm3 in the vapor, Nv , was . This relation can be tentatively explained by assuming total dissociation of the color centers into oxygen vacancies and free electrons at the temperature of formation.