Showing papers in "Philosophical Magazine in 1976"

Electronic properties of substitutionally doped amorphous Si and Ge

Abstract: It is shown that substitutional doping of an amorphous semiconductor is possible and can provide control of the electronic properties over a wide range. a-Si and Ge specimens have been prepared by the decomposition of silane (or germane) in a radio-frequency (r.f.) glow discharge. Doping is achieved by adding carefully measured amounts of phosphine or diborane, between 5 × 10−6 and 10−2 parts per volume, to obtain n- or p-type specimens. The room temperature conductivity of doped a-Si specimens can be controlled reproducibly over about 10 orders of magnitude, which corresponds to a movement of the Fermi level of 1·2 eV. Ion probe analysis on phosphorus doped specimens indicates that about half the phosphine molecules in the gaseous mixture introduce a phosphorus atom into the Si random network; it is estimated that 30–40% of these will act as substitutional donors. The results also show that the number of incorporated phosphorus atoms saturates at about 3 × 1019 cm−3, roughly equal to the number ...

The work-hardening of copper-silica v. equilibrium plastic relaxation by secondary dislocations

Abstract: Continuum plasticity is used to derive a new model for the plastic zone of secondary dislocations (B structure) observed around undeformable particles in a plastically deformed metal matrix. The mo...

Vacancy dipoles in fatigued copper

Abstract: The weak beam technique was used to perform an electron-microscopical study of dislocation dipoles in single crystals of copper fatigued beyond saturation at a resolved plastic shear strain amplitude of 0·3%. It has been shown previously that both primary edge dipoles and narrower faulted primary dipoles occur in these specimens (Antonopoulos and Winter 1976). An experiment designed to detect dipole flipping among the unfaulted dipoles gave a negative result and a sign determination revealed a two to one preponderance of the vacancy type. The faulted dipoles were found to be of vacancy type exclusively. It therefore seems likely that, after saturation, fatigue produces dislocation dipoles of vacancy type only. As an explanation of this it is suggested that non-linear elastic effects cause interstitial dipoles to have higher energy than vacancy dipoles. The practical consequence is that the persistent slip bands into which fatigue deformation concentrates (Winter 1974, Finney and Laird 1975) are g...

Theory of bimolecular reaction rates limited by anisotropic diffusion

Abstract: The theory of bimolecular diffusion-limited reactions including drift effects is extended to anisotropic diffusion and/or non-spherical reaction volumes. One-dimensional as well as isotropic two-or three-dimensional diffusion are limiting cases of the general theory. It is shown that, in the case of preferentially one-dimensional diffusion, diffusivities in the transverse directions about 107 times smaller than in the direction of fast diffusion may have an essential influence on the reaction kinetics. Such a situation appears to be realized in the diffusion-controlled annihilation of radiation-induced interstitials and vacancies in cadmium. Another interesting conclusion is that the rate of reactions between defects migrating in one dimension but along non-parallel lines (e.g. crowdion-crowdion reactions in f.c.c. metals) or reactions between H centres in alkali halides are controlled by two-dimensional diffusion and not by the much slower one-dimensional diffusion kinetics as has been assumed i...

Oxygen precipitation and the generation of dislocations in silicon

Abstract: Oxygen-rich precipitates in silicon, and the generation of dislocations at the interfaces between the precipitates and the matrix were observed. The precipitates are square-shaped plates with 〈110〉 sides and {100} habit planes. Prismatic dislocation loops are generated in silicon by the mechanism of prismatic punching. The loops are identified as interstitial loops with ½ 〈110〉 Burgers vectors and 〈110〉 axis. An ideal loop is rhombus-shaped with line senses in 〈112〉 directions. Nucleation of loops follows the mechanism of Ashby and Johnson (1969): a shear loop is nucleated on an initial slip plane and completes a rhombus-shaped prismatic loop by repeated cross-slips. To our knowledge, the loop nucleation process is identified for the first time for a system completely under internal stress.

Precipitation and trapping of hydrogen in copper

Abstract: Copper samples containing dissolved hydrogen up to a few hundred atomic p.p.m. have been prepared by quenching from equilibrium at higher temperatures. Measurements of the annealing of the residual resistivity up to 150°C show that the hydrogen precipitates around room temperature and the precipitation is controlled by diffusion of the hydrogen. Microscope investigations confirm this result and show that the hydrogen precipitates as bubbles which form preferentially at grain boundaries and dislocations. The number of bubbles formed was typically 109 cm−3. Consideration of bubble growth leads to the conclusion that the increase in bubble volume during growth is due to plastic deformation of the lattice near the bubble and that the precipitation is a diffusion limited process, conclusions which are confirmed by the experimental evidence. A model for the trapping of the dissolved hydrogen at impurities is also discussed. It is shown that by quenching quickly to low enough temperatures the hydrogen i...

The increase in the conductivity of chalcogenide glasses by the addition of certain impurities

Abstract: It is suggested that in chalcogenide glasses, elements such as Cu or Mn can have coordination number 4 and be negatively charged, and that the compensating positive charge is provided by D+ centres. The law of mass action ensures that the concentration of D-centres is then small. Under these conditions the Fermi energy is no longer pinned and the activation energy for conduction is shown to be decreased by 1/3. Similar considerations are applied to the effect of O, Cl and K on the conductivity of amorphous Se.

Finger-like crack growth in solids and liquids

Abstract: When voids or cracks grow by diffusion on grain boundaries in stressed solids, it is sometimes observed that the void periphery becomes finger-like, the fingers advancing ahead of the main body of the void. In this, grain-boundary voids in solids resemble voids in liquids, or in very viscous adhesives, all of which show a transition from a stable smooth periphery to an unstable, finger-like one as the void (or crack) growth-rate increases. We have analysed these phenomena, adapting and extending results due to Taylor and Zener. This allows us to derive conditions for the instability to occur, and expressions for the finger spacing and for the overall growth-rate of the finger-like crack

Weak-beam study of dislocation structures in fatigued copper

Abstract: Single crystals of pure copper oriented for single slip were fatigued to saturation at low constant plastic strain amplitude. A weak-beam study of the dislocation microstructures revealed a very high density of dipoles in both the ‘wall’ and ‘matrix’ structures. Secondary dislocations were not observed

Defects in electron-irradiated germanium

Abstract: The growth and structure of defects in electron-irradiated Ge has been followed at room temperature and at higher temperatures between 200° and 300°C. At room temperature, small interstitial loops of approximately edge character on {311} planes are observed and eventually grow into dendritic structures. At the higher temperatures, rods elongated along the 〈110〉 directions form first but then tend to widen into plates of interstitial type on {311} planes with a displacement vector R = 0.04〈611〉. It is possible that all of these defects are of the same type and correspond to aggregates of Ge interstitials, but the possibility of impurity effects cannot be ruled out. The defects tend to transform into perfect dislocation loops on annealing.

Magnetic viscosity in ferromagnets

Abstract: The thermal activation theory of magnetic viscosity is reviewed Equations are derived relating the maximum slope of the magnetization decrement in (time) curve to absolute temperature, irreversible susceptibility, and the rate of change of activation energy with field at constant temperature, (∂∊/∂H)T The activation energy ∊ is, in general, a function of the applied field H and an intrinsic function of the temperature T If, however, (∂∊/∂T)H=0 then (∂∊/∂H)T≃28·5k/(dH/dT) where H is the field at which the viscosity curve slope is a maximum Deviations from this relationship indicate an intrinsic temperature variation of the activation energy ∊ and thus enable thermal activation and intrinsic temperature effects to be separated Experimental data on magnetic viscosity are re-examined from this point of view

The ion pair distribution functions in molten rubidium chloride

Abstract: The partial pair distribution functions for molten rubidium, chloride have been determined by neutron scattering from isotopically doped samples. The Cl isotope was varied from the natural abundance for 35Cl and 37Cl to enriched 35Cl and enriched 37Cl. By using a specially prepared vanadium container, a much simpler container scattering correction was achieved compared with that for the silica container used in other work on halides. The results are compared with the molecular dynamics simulation of Dixon and Sangster: the agreement is good.

Transmission electron microscopy studies of displacement cascades in Cu3Au

Abstract: Displacement cascades produced in ordered Cu3Au by the impact of Cu+ions of energy 5–100 keV are studied using the method described in Part I. It is found that each incident ion of energy ≳ 10 keV creates, at the core of the resulting displacement cascade, a zone of reduced long-range order which can be imaged in the electron microscope. The sizes of displacement cascades in Cu3Au are well described by the analytical theory of Sigmund and co-workers. The best evidence for this comes from a comparison of experimental and theoretical image contrast profiles at an ion energy of 10 keV, when reasonably direct physical arguments can be used to link experimental and theoretical parameters. At ion energies ≳ 30 keV many of the disordered zones develop extrusions, and the formation of some discrete sub-cascades is observed. The various mechanisms which might give rise to the disordering are discussed.

Asymmetry of slip and shape changes during cyclic deformation of α-iron single crystals

Abstract: During cyclic deformation at prescribed plastic strain amplitude (net zero strain) initially round α-iron single crystals became elliptical. These shape changes could be related satisfactorily to the asymmetry of slip (of screw dislocations) in tension and compression by a simple model which, furthermore, permitted an estimate of the relative contributions of screw and non-screw dislocations to dislocation glide. Systematic investigations showed that the shape changes were most pronounced at high strain amplitudes and cyclic strain rates. The shape changes were found to depend on the strain amplitude in much the same way as the mechanical saturation stresses. An interpretation was proposed in terms of cyclic micro- and macro-strain ranges corresponding to the motion of predominantly non-screw dislocations and screw and non-screw dislocations, respectively. The possible significance of the discussed effects with regard to fatigue failure of b.c.c. metals was pointed out and supported by experiment...

Dislocations in polyethylene crystals: line energies and deformation modes

Abstract: Explicit expressions are obtained for the line energy of straight dislocations in orthorhombic polyethylene crystals. It is found that the 〈0, 0, c〉 screw dislocation is the perfect dislocation with the lowest line energy whilst the 〈0, 0, c〉 edge has negative line tension and is therefore unstable. The possible dissociation of perfect dislocations into partials is discussed, and it is shown that both (110) and (310) twin dislocations can arise from such dissociations. Numerical calculations also show that homogeneous nucleation of 〈0, 0, c〉 dislocations, and twin dislocations, has a very high probability of occurrence under a stress of the order of the yield stress of polyethylene crystals. These results help to explain the readiness of polyethylene crystals to deform by c axis slip, twinning, and monoclinic transformation modes. The supression of c axis slip by irradiation and cross-linking is also explained by the very high line tension of the 〈0, 0, c〉 screw dislocations.

The solid electrolyte transition and melting in salts

Abstract: Thermodynamic, electrical and structural data for simple solid electrolytes are collated, and the case for considering the solid electrolyte state as a separate ion-disordered phase is presented quantitatively. Attention is directed to some remarkable regularities and similarities in the transitions from poorly-conducting to well-conducting states.

Atomic simulation of the dislocation core structure and Peierls stress in alkali halide

Abstract: Dislocation core behaviour was studied by determining the equilibrium arrangement of ions surrounding an a/2〈110〉 {110} edge dislocation in potassium chloride. A computer simulation method was used to solve the equations of motion of the ions interacting through a Coulombic potential, with a Born-Mayer repulsive contribution, between nearest neighbours. Two flexible boundary schemes were employed and their application and importance to the results are described. The observed difference between the Volterra description of the ion positions and the calculated equilibrium position constitutes the core field. This core field was examined in detail but its most important feature is a lattice expansion of about 0-5 b2 per unit dislocation length. Shear stresses were also applied to the models using flexible boundary conditions with results indicating a Peierls stress in the range of 12-7 to 32-4 MPa which is in reasonable agreement with experiment. The case of moving the dislocation was quite sensitive...

The effect of electron interaction on variable-range hopping

Abstract: The effect of interaction between electrons on variable-range hopping is investigated. It is shown that interaction between electrons on localized states produces a band gap for single-particle excitations, but not for those of the many-electron system. Variable-range hopping with σ proportional to exp (– B/T1/4) still occurs in the limit of low T, but interaction may lead to deviations at higher T, and decreases the pre-exponential factor.

Substitutional doping in amorphous semiconductors the As-Si system

Abstract: The amorphous alloy system As-Si has been prepared in thin film form using the plasma decomposition of silane-arsine mixtures. Electrical and optical measurements show effects characteristic of sub...

Climb dissociation of dislocations in sapphire (α-Al2O3)

Abstract: Edge dislocation dipoles, formed in sapphire during basal glide at elevated temperatures, break up into strings of loops by self-climb. One stage in this process is the transformation of narrow dipoles into faulted dipoles. This is due to the climb dissociation of each dislocation in the dipole, according to the reaction 1/3[11 0]→1/3[11 0] + 1/3[01 0], with one pair of partials annihilating to form the faulted dipole. The stacking fault energy is estimated to be ∼400 erg/cm2, based on the maximum width of the faulted dipoles. A second stage, where the faulted dipoles rotate through 30° to the exact edge orientation, occurs to reduce the elastic energy.

Anisotropy in grain boundary segregation in copper-bismuth alloys

Abstract: The fracture surfaces of an alloy of copper containing 60 p.p.m. of bismuth have been analysed by Auger electron spectroscopy. Alloys annealed in the temperature range 400° to 700°C before fracture all show marked anisotropy in the segregation of bismuth to the intergranular fracture surface. This is in marked contrast to the results obtained by Seah and Hondros (1973) for sulphur and tin segregation in iron. Possible reasons for this contrasting behaviour are discussed briefly.

The α—β transition in the cristobalite phases of SiO2 and AIPO4 I. X-ray studies

Abstract: The α—β transition in the cristobalite phases of SiO2 and AlPO4 has been studied using X-ray powder diffraction. The transitions show marked hysteresis but also occur over a range of about 20°C in which the α and β phases coexist. The lattice parameters show a smooth and slow variation with temperature for each phase and a discrete difference between phases corresponding to volume differences of about 4%, showing that the transitions are of first order. The hysteresis has been studied and it is concluded that the phase coexistence in the bulk sample occurs because different crystallites undergo the transition at different temperatures. The structural aspects of the transformations are discussed and it is shown that the phase change involves a small movement of the atoms around four fold screw axes together with a disordering of oxygen atom positions

Vibrational spectra and topological structure of tetrahedrally bonded amorphous semiconductors

Abstract: Calculations of the vibrational spectra of continuous random network models of tetrahedrally bonded amorphous semiconductors are presented. Nearest neighbour central and non-central force constants are used. The results show that the shape of the TO-like peak of the spectrum depends on both the topology and bondangle distribution of the network. Comparison with experimental data suggests that a-Ge contains slightly fewer odd-membered rings than the Polk model and an average bond-angle distortion of 10° or more. On the other hand the shape of the TA-like peak depends more directly on the topology of the network, giving a clearer distinction between the results for structures containing both odd and evenmembered rings and those containing even rings only. Tentative comparison with experimental data for a-GaAs suggests a structure containing very few oddmembered rings, while the data for a-Ge indicate a significant number of oddmembered rings.

Charge density waves in layered metals observed by X-ray photoemission

Abstract: Temperature dependent X-ray photoemission spectroscopy (ESCA or XPS) studies on the Ta 4f lines in 1T-TaS2, 1T-TaSe2, 1T-Ta0.97Hf0.03S2 and 2H-TaSe2 are reported. Striking effects arising from charge density waves (CDWs) are observed at low temperature. While electron diffraction evaluates the periodicity and direction of the CDWs, these XPS observations allow the different configurations of the CDWs with respect to the crystal lattice to be deduced. The results indicate the existence of CDW fluctuations in commensurate phases, and that in some incommensurate and ‘quasi-commensurate’ phases the CDWs cannot maintain long-range coherence. In 1T-Ta0.97Hf0.03S2 the Hf impurities pin the incommensurate CDWs into domains.

Correlation between resistivity increment and volume fraction of G.P. zones in an Al-3·2 wt % Zn-2·2 wt % Mg alloy

Abstract: The correlation between resistivity increment caused by G.P. zones and their volume fraction has been investigated in an Al-3·2 wt % Zn-2·2 wt % Mg alloy. It is found that the resistivity increment is a linear function of the 2/3 power of the volume fraction. The results show that the resistivity increase is mainly determined by the total volume of zones and is not sensitive to their size distribution

Transmission electron microscopy studies of displacement cascades in Cu3Au: I. The diffraction contrast of disordered zones

Abstract: Heavy-ion or neutron bombardment of the ordered alloy Cu3Au may result in the creation of regions of reduced long-range order (‘disordered zones') within the volumes of individual displacement cascades. We present calculations for the electron-microscope contrast in superlattice dark-field reflections from such disordered zones, based on two models: a straightforward scattering matrix model, where the disordered zones are considered to be uniformly disordered, and all the contrast arises from the difference in structure factor of these zones and the ordered matrix; and a more sophisticated method, based on an alternative Bloch wave approach, which allows the effect of the strain contrast of any dislocation loops produced by the collapse of cascades to be included. The theoretical calculations are tested by the observations of disordered zones in Cu3Au produced by energetic Cu+ ion bombardment. Good agreement between the contrast calculations and the experiments is found.

The electrical conductivity of fluid selenium up to supercritical temperatures and pressures

Abstract: The electrical conductivity of fluid selenium has been measured as a function of temperature and pressure to 1750°C and 1200 bars, respectively. The conductivity isobars exhibit strong increases to nearly-metallic behaviour in selenium above 1300°C at supercritical pressures. As the fluid is heated above 1500-1600°C the conductivity isobars drop sharply toward more insulating behaviour in the pressure range investigated here.

Work-hardening and recovery of dispersion hardened alloys

Abstract: Experiments have been carried out in order to test theories of the work-hardening and low temperature recovery in dispersion hardened crystals. Single crystals of copper containing small (∼ 500 A diameter) particles of SiO2 have been deformed in tension at temperatures between 77 and 450 K, and the primary and secondary dislocation distributions have been determined quantitatively by transmission electron microscopy at 400 kV. It is found that in the temperature range in which the work hardening rate decreases with increasing temperature and at which static recovery occurs (∼300 K), the number of both primary and secondary dislocation loops associated with the particles drops significantly with increasing temperature of deformation. It is shown that this behaviour is consistent with the climb and shrinkage of Orowan loops by pipe diffusion (Gould, Hirsch and Humphreys 1974), but that it is not consistent with the theory of Atkinson, Brown and Stobbs (1974) in which recovery is attributed to proce...

Magnetic viscosity in ferromagnets: II. Co5Ce permanent magnet particles

Abstract: Magnetic viscosity measurements have been made on Co5Ce particles with diameters from 1–5 to 8 μm and over a temperature range from 95 to 290 K. The temperature variation of the properties could not be explained in terms of thermal activation effects alone. A wall barrier model has been used to analyse the data and to separate intrinsic temperature and thermal activation effects. The temperature variation of barrier interaction strength is shown to be the same for all particle sizes and to be the major factor in the rise in coercive force as the samples are cooled.

Quenching of high quality gold single crystals

Abstract: An improved quenching technique is described that allows compact and delicate samples to be quenched with high quenching rates otherwise only achieved with the Joule heating method. High quality gold single crystals 1 mm in diameter with dislocation densities lower than 2000/cm2 have been quenched from temperatures between 700° and 950°C using this technique. It was found that vacancy losses during the quench were negligible in these crystals. The formation energy of vacancies in gold was found to be (0-97±0-03) eV. Polycrystalline gold wires 1 mm in diameter have been quenched from temperatures between 730° and 1000°C in the same way as the single crystals. A comparison with the results on single crystals shows that the main type of sinks for vacancies in polycrystalline Au during a quench are dislocations. A theory by Seidman and Balluffi describes these losses quantitatively. The specific resistivity attributable to vacancies was determined to be (1·69±0·20)μΩ cm/at.-%.