Volume 34, Issue 4, 01 April 1963
- general session
- nuclear and antiferromagnetic resonance
- thin films‐1
- change interactions and magnetic transitions
- magnetic anisotropy; spin waves
- devices and instrumentation
- magnetic moments and ordering
- thin films—2
- magnetoelastic and magneto‐optic effects
- spin‐wave instabilities and magnetoelastic coupling
- magnetization processes
- rare earths
- ferromagnetism; superconductivity
Index of content:
34(1963); http://dx.doi.org/10.1063/1.1729522View Description Hide Description
Theoretical consideration is given to the behavior of the anisotropy field of iron—nickel films ranging in composition from 65% Ni—35% Fe to 90% Ni—10% Fe as a function of applied external stresses and magnetic fields in various directions in the plane of the film. It is postulated that internal stresses in the substrate film system may account for observed discrepancies. An expression is derived which relates the magnetoelastic parameter (η) to the magnetostriction (λ s ) of the film. This parameter is measured as a function of composition and the predicted value of λ s agrees within the limits of precision to that observed for bulk material. Assuming the films to be isotropic in the plane of the substrate, a relation between applied stress and change in the anisotropy field is developed and within limits experimentally verified. It is postulated that observed discrepancies are due to nonlinear anisotropic internal stresses and dispersion in the direction of the easy axis of the film.
34(1963); http://dx.doi.org/10.1063/1.1729523View Description Hide Description
The simplified cylindrical distribution function has been applied to the determination of the degree of orientation in a moderately oriented linear polyethylene fiber. The function has been calculated using the first two and ten lines of the scattering pattern and, though poorly resolved, yields useful orientation information. It is concluded that the greater the knowledge of the molecular structure of the material under investigation the less the amount of intensity data required to produce a useable function. Although the technique is applicable to both crystalline and noncrystalline materials, the standard methods, applicable only to crystalline materials, are simpler. Therefore, the method can be considered to apply mainly to moderately oriented noncrystalline fibers.
34(1963); http://dx.doi.org/10.1063/1.1729524View Description Hide Description
Piezoresistive characteristics of a diffused layer on the surface of a semiconductor having a cubic crystallographic structure are analyzed. The surface is assumed free of external forces, and the current component normal to the surface within the layer is assumed zero. The equations derived are useful in the design of diffused piezoresistive electromechanical transducers. The electric fields in the layer are related to the sheet current densities and the stresses by diffused piezoresistive coefficients, which are defined in integral form. These coefficients are functions of the fundamental piezoresistive coefficients, the crystal orientation, and the unstressed conductivity profile function in the layer, but are independent of the layer thickness. The coefficient of major interest is numerically evaluated as a function of surface impurity concentration for p‐type silicon and n‐type germanium‐diffused layers with Gaussian and complementary error function impurity distributions. Measurements on diffused samples are in agreement with the analysis. The magnitude of a nonlinearity in layer piezoresistance arising from the variations of the bulk coefficients with doping is estimated.
34(1963); http://dx.doi.org/10.1063/1.1729525View Description Hide Description
High‐field electron emission currents were drawn from a point cathode made from a single crystal of n‐type gallium arsenide. Stable currents were observed for several cathodes over a wide range of voltage with an upper limit on the currents which could be drawn from the material.
34(1963); http://dx.doi.org/10.1063/1.1729526View Description Hide Description
When a constant electric field is applied to a dielectric, the current density per unit field decreases with time from an ``initial conductivity'' to a ``final (or steady‐state) conductivity.'' A study is made of this time‐dependent polarizationeffect for ``pure'' NaCl crystals in the range from 50° to 200°C. The validity of Ohm's law and the superposition principle is demonstrated for these crystals; this establishes the linearity of the formal equation which relates polarization,electric field, and their time derivatives. Also studied are the effect of prolonged current flow, the effect of deliberate introduction of an air gap between the crystal and one of its electrodes, and the effects of impurities, deformation, x‐ray irradiation, and annealing. It is concluded that the results are not consistent with the well‐known space‐charge polarization theory of Joffé, according to which the buildup of space charge occurs because of blocking of the current carriers at one or both electrodes. Rather, it is necessary to regard the time‐dependent polarization as a dielectric relaxation phenomenon. According to this viewpoint, the final conductivity, and not the initial value, represents the true ionic conductivity. Possible relaxation mechanisms are discussed in terms of defect clusters and charged jogs on dislocations.
A Complete E, P, V, T, S Thermodynamic Description of Metals Based on the P, u Mirror‐Image Approximation34(1963); http://dx.doi.org/10.1063/1.1729527View Description Hide Description
The well‐known practice of using the mirror‐image, in the P, u plane, of the shock Hugoniot curve about a vertical line through the shocked state PH, uH is shown to give a complete thermodynamic description of metals when the U (shock velocity) vs uH relationship and α≡V ‐1(δV /δT) P≈0 are known. Use of the experimental relationship U = c 0+auH (a and c 0 constants) and α = constant, leads to a thermodynamic description which results in the metal appearing less compressible than if described by the Mie—Grüneisen equation of state. Furthermore, the existance of an anomalous behavior of cp in the low pressure neighborhood (<10 to <50 kbar depending on the metal) of the initial state rules out the simultaneous existence of a Hugoniot satisfying the linear U vs uH relationship, of isentropes satisfying the mirror‐image assumption, and of a constant value of α in this neighborhood. Thermodynamic functions for 16 metals are calculated up to 2 Mbars and compared with the results obtained from the Mie—Grüneisen equation of state.
34(1963); http://dx.doi.org/10.1063/1.1729528View Description Hide Description
This paper develops the equations needed to determine the surface diffusion coefficient (Ds ) when a point source is used and solute is drained off from the high diffusivity surface layer into the bulk. The assumptions used are essentially those used by Fisher for his analysis of grain boundary enhanced diffusion, and the results are equally simple to use. The conditions under which Ds can be most easily and accurately determined are discussed.
34(1963); http://dx.doi.org/10.1063/1.1729529View Description Hide Description
It is found that Ge can be plastically deformed by indentation at temperatures down to 78°K without introducing cracks. Photomicrographs are presented which show the characteristic etch‐pitpattern of dislocation half‐loops resulting from annealing and etching of indented (111) surfaces.
Magnetic Field Effects on an Abnormal Truncated Glow Discharge and Their Relation to Sputtered Thin‐Film Growth34(1963); http://dx.doi.org/10.1063/1.1729530View Description Hide Description
Superposition of a radially symmetric magnetic quadrupole field onto an abnormal glow discharge operating at E/P of 105 produces an additional ionization region in the negative glow which is not observed in the absence of a magnetic field and which increases the ion current density at the cathode by more than an order of magnitude. Preliminary Langmuir probemeasurements show the electric field gradient in this region to be small and an average electron energy of only 0.8 eV. Data are given to show how sputtering ratio measurements in such a system can be correlated to secondary electron emission at the cathode and charge transfer in the Crooks dark space. Ion current density profile measurements across the cathode as well as corresponding cathodeerosion and anode deposition profiles are given.
34(1963); http://dx.doi.org/10.1063/1.1729531View Description Hide Description
The five independent elastic constants of single crystals of the hexagonal allotrope of thallium have been determined by the ultrasonic pulse‐echo technique over the temperature range, 4.2°–300°K. The values at 300°K in units of 1011 dyn/cm2 are C 11=4.080, C 33=5.280, C 44=0.726, C 12=3.54, and C 13=2.9. The compressibility derived from these elastic constants is in good agreement with the compressibility obtained by Bridgman from hydrostatic measurements on polycrystalline samples.
34(1963); http://dx.doi.org/10.1063/1.1729532View Description Hide Description
Pyrolytic graphite was cleaved in a high vacuum to obtain clean surfaces, and the secondary electron emission properties of these surfaces were determined. Pyrolytic graphite has a maximum secondary emission yield of 1.0 at 300‐eV primary electron energy for electrons incident normal to the basal plane. The most probable energy of the emitted secondaries is 2 eV, with one‐half of the secondaries having energy in excess of 8 eV. A maximum secondary emission yield of 0.75 at 350‐eV primary energy was measured for normal incident electrons on a face that was perpendicular to the basal plane. The backscattered fraction for pyrolytic graphite was found to be 0.05. No differences were observed in the secondary electron emission properties of single‐crystalline pyrolytic graphite and pyrolytic graphite that was less well ordered.
34(1963); http://dx.doi.org/10.1063/1.1729533View Description Hide Description
The electron density of a pure InSb crystal has been synthesized using very accurate x‐ray diffraction intensities. After subtracting the electron density of the palladium core from both kinds of atoms, the distribution of the valence electrons was determined. By this means it was established that, on the average, 0.45±0.05 electrons are transferred from the Sb atoms to In atoms, as compared to a transfer of 0.34 or 0.42 electrons deduced from infrared absorption measurements. The effective ionic charge to be used in polar‐scattering calculations for InSb was calculated to be 0.17±0.02 e based on the electron density, as compared to 0.18 e or 0.16 e determined from electron mobility measurements.
34(1963); http://dx.doi.org/10.1063/1.1729534View Description Hide Description
Utilization of the Hall effect in the frequency conversion of microwaves with gain was investigated. A conversion gain which seemed to be feasible was found to be dependent on the control sensitivity of the conversion system. A high control sensitivity can be achieved through use of a high‐mobility Hall sample, like InSb, an efficient coupling of a modulating field to the sample, and a suitable high‐energy pump source.
The high‐power test was performed by using a pulsed microwave pump. Conversion gains were measured as a function of the pumping power and showed some ``saturation'' effects. These effects were explored and explained, based on the behavior of carriers in the sample under high‐power microwave fields. In particular, a breakdown field beyond which the saturation seemed to disappear was found.
The present S band experimental model still suffered a conversion loss of 25 to 30 dB. Results have, however, revealed the possibility of a practical microwave Hall converter with gain.
34(1963); http://dx.doi.org/10.1063/1.1729535View Description Hide Description
A theoretical study has been made of the transmission of a monochromatic, well‐collimated beam of radiation normally incident on idealized material containing only two energy levels. The nonlinear partial differential equations governing the variation of population and photon density in space and time have been solved exactly for arbitrary initial conditions under the assumption that spontaneous emission and thermal relaxation from the excited state can be neglected. These conditions are satisfied in practice when the radiation is in the form of a pulse whose duration is short compared with the characteristic times of these relaxation processes. Two cases are considered in detail: (a) when all atoms are in the ground state and the material absorbs; and (b) when there is a population inversion and the material amplifies. In case (a), the radiation, with a characteristic velocity, ``bores'' its way through an optically dense substance emerging delayed in time and leaving the material in a perfectly transparent or saturated state. In case (b), an incoming pulse is amplified and sharpened—to a degree determined by the gain of the medium. In addition to these cases, the dependence of the apparent spontaneous emission lifetime on the size and population distribution of the material is mentioned.
Some Causes of Resonant Frequency Shifts in Atomic Beam Machines. I. Shifts Due to Other Frequencies of Excitation34(1963); http://dx.doi.org/10.1063/1.1729536View Description Hide Description
The quantum theory of an atomic beam machine is set up in matrix form. A new method is then used to derive the Bloch‐Siegert shift in the resonance. The results are extended to the case of Ramsey‐type excitation. Finally the Bloch‐Siegert shift is computed for the present atomic beamfrequency standards and found to be well below the accuracy of measurement.
Some Causes of Resonant Frequency Shifts in Atomic Beam Machines. II. The Effect of Slow Frequency Modulation on the Ramsey Line Shape34(1963); http://dx.doi.org/10.1063/1.1729537View Description Hide Description
The effect of slow frequency modulation of the exciting radiation on the Ramsey line shape observed in an atomic beam experiment is formulated theoretically. It is shown that the presence of second harmonic in the modulation can introduce measurable frequency shifts, whether observed directly or with a servo system.
34(1963); http://dx.doi.org/10.1063/1.1729538View Description Hide Description
Previously, five phonon energies were used to explain 16 peaks in the measured lattice absorptionspectrum of GaAs. It is shown here that the five phonon energies are not unique and that a different set of values is also compatible with the available infrared data both with respect to the absorption peak positions and the temperature dependence of the absorption. A similar difficulty is also noted for some previously published work on CdS. It is suggested that difference band absorption measurements might show which sets of phonon energies are correct.
34(1963); http://dx.doi.org/10.1063/1.1729539View Description Hide Description
Observations of Bitter patterns were carried out for two varieties of Ni–Fe alloy thin filmselectrodeposited without magnetic field onto mechanically polished copper substrates. These two sorts of films were composed of 80Ni‐20Fe ``Film 1'' and 82.5Ni‐17.5Fe ``Film 2'' and were Fe‐rich and Ni‐rich from the composition of 81.5Ni‐18.5Fe with minimum coercive force. The behavior of magnetic domains was examined during various magnetic processes and under externally applied stress. It was found that: (i), in ``Film 1,'' the easy axis is parallel to the scratches on the substrates and the domain walls fall into those scratches; and (ii), in ``Film 2,'' an averaged easy direction is induced perpendicular to the scratches in several films, whereas in the other films it was not well‐established. These results are well explained, being consistent with the hysteresis characteristics, by assuming that magnetostriction is positive in ``Film 1'' and negative in ``Film 2'' and that magnetoelastic and shape anisotropies both are induced, accompanying the scratches on the surface of the substrates.
34(1963); http://dx.doi.org/10.1063/1.1729540View Description Hide Description
Space‐charge, field, and potential distributions in inhomogeneously doped crystals are determined by a nonlinear differential equation which requires numerical computation in order to be solved. By means of numerical computation an attempt is made to obtain the potential, field, and space charge throughout the crystal. For exponential impurity distributions bounding curves can be obtained which allow one to obtain the complete distributions. A few examples are computed which demonstrate the difficulties encountered in obtaining solutions.