Volume 31, Issue 4, 01 April 1960
Index of content:
31(1960); http://dx.doi.org/10.1063/1.1735653View Description Hide Description
A discussion is given of the processes by which atoms are displaced from their lattice sites by incident fast electrons. In addition to direct displacement by the incident electron, expressions are derived for the fraction of atoms displaced by fast moving atoms and by fast secondary electrons, both of which are generated initially by the incident electron. A discussion of thermal spikes is given. A thermal spike process is discussed that might lead to the generation of defects with x irradiation in materials that exhibit a threshold energy for a displacement by a charged particle.
31(1960); http://dx.doi.org/10.1063/1.1735654View Description Hide Description
X‐ray examinations of brittlefractures in polycrystallinechromium, produced at temperatures between 193°K and 623°K, have been made. Estimates of the amount of deformation present at the fracturesurfaces showed that very large plastic strains were associated with brittle cleavage fracture at the higher temperatures, such that the length of a self‐sustaining crack calculated according to the modified Griffith crack formula of Hall, was more than an order greater than the mean grain diameter of the material. Plastic strains of this order would be expected to result in ductile behavior. It is suggested that brittle failure was caused by the presence of a fine dispersion of precipitate particles, which delayed the action of cross slip in sealing off the early stages of cracking. The spacing of the precipitate particles was estimated to be between 6000 and 7000 A.
31(1960); http://dx.doi.org/10.1063/1.1735655View Description Hide Description
Experimental observations are presented of dislocation multiplication, of the defectstructure left behind by a moving dislocation, and of cross‐glide of individual dislocations in LiF crystals. New dislocation loops form at many different sites in the wake of a moving dislocation. These loops have the same Burgers vector as the parent dislocation but do not, in general, lie on the same atomic plane. The rate of formation of new loops depends upon the magnitude of the applied stress. Such creation of new loops leads eventually to the formation of a wide glide band.
A moving screw dislocation trails many line defects behind it that lie parallel to its direction of motion. The existence and nature of these trails and the observed dislocation multiplication can be explained in terms of a mechanism which involves the formation, by cross‐glide, of jogs on a screw dislocation. This cross‐glide multiplication mechanism was originally proposed by Orowan and by Koehler. It is demonstrated that cross glide occurs easily in LiF, so that this mechanism is plausible. Some interesting complications arise when jogs are formed that are longer than several atomic spacings but less than several hundred. The defect trails exert a dragging of the screw dislocations that is not negligible compared to the yield stress of a crystal.
31(1960); http://dx.doi.org/10.1063/1.1735656View Description Hide Description
The mobilities of positive ions in H2 and D2 have been measured, using a pulsed Townsend technique. For hydrogen, the mobility μ0 corrected to 0°C was 11.8 (cm2/v×sec) at E/p 0=26(v/sec×mm Hg), increasing to a maximum of 15.3 at E/p 0=48, then decreasing to 11.6 at E/p 0=150. For deuterium, the mobility was about 0.75 the value for hydrogen throughout the range of E/p 0. The experiment was performed with uranium‐purified gas in an ultra‐high vacuum system. Ions were not identified as atomic, diatomic, or triatomic; there was no conclusive evidence of more than one ion at any value of E/p 0. The H2mobility data are significantly lower than those attributed to Mitchell in the range 20<E/p 0<40, and lower than the value ≈12.5 for zero‐field mobility generally reported.
31(1960); http://dx.doi.org/10.1063/1.1735657View Description Hide Description
The five independent elastic constants for yttriumsingle crystals have been determined by the pulse‐echo technique over the temperature range, 4.2–400°K. The experimental values extrapolated to 0°K are: C 11=8.34±0.02, C 33=8.01±0.02, C 44=2.690±0.006, C 12=2.91±0.03, and C 13=1.9±0.4 in units of 1011 d/cm2. The degree of elasticanisotropy is low throughout the entire temperature range. The curves illustrating the temperature dependence of the elastic constants show several inflections, and the curve for C 11 crosses that for C 33 near 390°K. The behavior is somewhat unusual, and is probably explicable on the basis that the degree of band overlap in yttrium is quite sensitive to interatomic spacing. Such a sensitivity has previously been postulated to explain the temperature dependence of the Hall coefficient.
31(1960); http://dx.doi.org/10.1063/1.1735658View Description Hide Description
31(1960); http://dx.doi.org/10.1063/1.1735659View Description Hide Description
A correlation is reported between certain optical extinction bands in sapphire,quartz, the silver and alkali halides, and the calculated plasma resonance of electrons in colloidal particles which are either known or expected to be present. This correlation suggests not only that pile neutrons produce lithium particles in lithium fluoride, but aluminum particles in sapphire, and silicon particles in quartz.
Solutions of the Equations of Space Charge Flow for Radial Flow between Concentric Spherical Electrodes31(1960); http://dx.doi.org/10.1063/1.1735660View Description Hide Description
The solutions of the equations of space charge flow of electron beams for radial flow between concentric spherical electrodes are formulated in such a manner as to permit the introduction of arbitrary initial conditions. This allows one to solve a variety of physical problems with the aid of a table of Airy functions. A particular case, which exhibits a potential minimum between electrodes, is presented as an illustrative example.
31(1960); http://dx.doi.org/10.1063/1.1735661View Description Hide Description
The attenuation of the plane shock wave produced in a solid by a flying plate of the same material is treated neglecting dissipative processes and effects of material rigidity. Explicit formulas for the position of the shock front and the shape of the pulse as functions of time are obtained by application of Friedrich's method. A numerical example for an aluminum target and projectile is presented to illustrate some of the features of the calculation, and an experiment is proposed to test the theory. The experiment should also allow a reasonably precise measurement of sound velocity immediately behind the shock front to be obtained. This possibility applies equally well to the case of target and projectile of different materials.
Motion of 180° Domain Walls in Metal Electroded Barium Titanate Crystals as a Function of Electric Field and Sample Thickness31(1960); http://dx.doi.org/10.1063/1.1735663View Description Hide Description
The domain dynamics of polarization reversal in metal electroded BaTiO3single crystals have not been adequately described in the literature. The present paper contains quantitative measurements of the velocity of the sidewise motion of 180° domain‐walls in BaTiO3 crystals, most of which were electroded with 200‐A thick films of Au. The wall velocity has been measured as a function of several variables including the applied electric field, the crystal thickness and the impurity content added to the crystal growth melts.
The wall velocity as a function of the field E is described by v ∞ exp (−δ/E) where δ and v ∞ are constants over several decades of velocity. The quantities δ and v ∞ are the activation field and extrapolated wall velocity for E=∞, respectively. This field dependence of the wall velocity is the same as that reported previously for liquid electroded specimens. The thickness dependence of the wall velocity, which is through δ, is very pronounced and similar to that observed in measurements of other variables under different experimental conditions. The present data are explained in terms of a surface layer estimated to be of the order of 100‐A thick. It is suggested that high fields within the surface layer may give rise to electron field emission and the electroluminescence which is observed during polarization reversal. Effects on the wall motion which persist for times of the order of minutes are ascribed to dielectric relaxation phenomena. The observed dependence of the wall motion on the impurity content of the cyrstal, and several other parameters, may result from changes in the characteristics of the surface layer.
31(1960); http://dx.doi.org/10.1063/1.1735664View Description Hide Description
A gold needle point in contact with a gold flat has been used to study the formation and initial growth of single‐asperity solid‐state bonds. In the absence of external pressure, the growth of the bonded area can be expressed analytically by a diffusion‐controlled rate equation. Values for the coefficient of self‐diffusion determined from these experiments are in reasonable agreement with those given in the recent literature.
31(1960); http://dx.doi.org/10.1063/1.1735665View Description Hide Description
A simple two‐dimensional model, in which uniform hard spheres are made to move steadily and apparently nearly at random, is described. At low sphere density the model exhibits ``gas‐like'' behavior. As the density is increased the behavior becomes more ``liquid‐like'' and then ``crystallization'' occurs. Many dynamic atomic phenomena, believed to occur in the gas, liquid, or solid state, are illustrated by the model.
31(1960); http://dx.doi.org/10.1063/1.1735666View Description Hide Description
The kinetic energy distribution of secondary positive ions liberated from a solid metallic target of beryllium under bombardment by positive ions was measured in a mass spectrometer provided with an energy analyzer. In conformity with earlier investigations, it was found that an appreciable fraction of the ions was liberated with energies less than 5−10 ev, although some secondary ions of more than 200 ev were found. The distributions appeared to be at least partially Maxwellian in character. Although errors in measurement were large, there appeared to be little dependence of the yield of secondary ions on the mass of the bombarding ion.
31(1960); http://dx.doi.org/10.1063/1.1735667View Description Hide Description
The present paper briefly reports about the results of examinations of surfaces of CP4‐etched Ge. The origin of the so called ``1st‐order structure'' observed on the micrographs is discussed. It is shown that while etching proceeds, the ending points of some dislocations are reached, or some of them become passivated. In both cases, the etch pits no longer grow and their traces give origin to the 1st‐order structure.
31(1960); http://dx.doi.org/10.1063/1.1735668View Description Hide Description
The velocities of dislocations have been measured as a function of stress for both undeformed and deformed crystals. The motion is slower at a given stress in a strained crystal, and extra stress is needed to give the same velocity in a strained crystal as in one that is unstrained. It has been found that the extra stress is just equal to the difference in macroscopic flow stresses for the two states of a crystal.
Existing theories of strain‐hardening do not correctly predict the behavior of LiF. Barrier‐type theories are ruled out because the distance that a dislocation can move in a strain‐hardened crystal does not seem to be limited. Theories of Taylor hardening or of cutting of a dislocation forest do not give the observed dependence of strain‐hardening on dislocation density. It is observed that the strain‐hardening increment is proportional to the dislocation density (also to the plastic strain), and the proportionality constant is 3–5 dynes/disl. The data are consistent with the idea that the hardening is due to defects left in the wakes of moving dislocations. These defects would interfere with the movement of subsequent dislocations on the same or nearby glide planes.
31(1960); http://dx.doi.org/10.1063/1.1735669View Description Hide Description
A large positive piezoresistive effect has been observed in p type PbTe. The following room temperature coefficients were determined: πHydrostatic=+150±50; π11=80±30; π=+220±50, all in 10−12/d. The shear coefficient π44 is estimated to be +255×10−12 cm2/d. The calculated value of π12 was +35±40×10−12 cm2/d. These measurements were performed on single‐crystal, p type PbTe with a mobility of approximately 700 cm2/v sec and a carrier concentration of about 1018 per cm3. Consideration of the coefficients indicates the major axes of the energy ellipsoids are oriented in the  direction, which is in agreement with magnetoresistance measurements.
31(1960); http://dx.doi.org/10.1063/1.1735670View Description Hide Description
A molecular beam of sodium chloride can be partially sorted with respect to the rotational levels of the molecule when it is passed through an inhomogeneous electric field. The possibility of state selection of the beam in one of the lower rotational states of the molecule (J=2; |M| =0, 1, 2) has been investigated by studying the molecular trajectories for the rotational states J=1; |M| =1, 0 and J=2; |M| =2, 1, 0. A typical calculation for a state selector is given.
31(1960); http://dx.doi.org/10.1063/1.1735671View Description Hide Description
A transmission electron microscope study of ordered (001) films of Au‐50 at. % Cu made by evaporation in vacuo onto cleaved rocksalt at 400°C showed (a) the presence of a 6 sided roughly hexagonal single crystal area in the film about 2 μ across apparently surrounded by polycrystals, the 3 pairs of sides being nearly parallel to , respectively, (b) streaks within this crystal interpretable as twins on (111), (c) holes through the film tended within this crystal to be polyhedral, developing sides parallel to 〈100〉 and 〈110〉. The latter observation indicates that the effective supersaturation during evaporation is relatively low.
31(1960); http://dx.doi.org/10.1063/1.1735672View Description Hide Description
The irregular line profiles associated with crystals that have a random structure which is coarse relative to the irradiated area are regarded as the result of the composition of a characteristic distribution and a set of broadening and translating processes.
The inversion of this composition is expressed operationally. A differential operator associated with the transform of the intrinsic distribution is introduced. This operator reduces the line profile to the set of broadening and translating elements. The operations are such as may be performed by an analogue computer. Preliminary experiments show that it may be possible to obtain the desired resolution. The effects of ``noise'' and distortion are investigated.
31(1960); http://dx.doi.org/10.1063/1.1735673View Description Hide Description
Silicon crystals containing aluminum have been annealed at various temperatures within the range 700° to 1350°C, particular emphasis being given to the range 1200° to 1300°C. Precipitation, which occurs during the heat treatment in the bulk crystal and on various structuraldefects, has been detected by transmission optical microscopy. A lowering of the aluminum concentration in the neighborhood of the precipitate particles after heating at 1250°C has been revealed at the surface by a novel technique involving the etching of crystals which also contain phosphorus, where internal cylindrical and planar p‐n junctions are formed round individual dislocations and grain boundaries, respectively. At this temperature, no precipitation is observed on dislocations in plastically deformed crystals. The precipitation of aluminum is interpreted in terms of a chemical reaction involving the coprecipitation of oxygen and possibly carbon.