Showing papers on "Cathodoluminescence published in 1979"

Observation of dislocations in cadmium telluride by cathodoluminescence microscopy

Abstract: A one‐to‐one correspondence between deformation‐produced dark spots in cathodoluminescence (CL) micrographs and dislocations has been demonstrated in n‐type CdTe by comparing the CL pattern with the etch‐pit pattern developed with a new etchant which has been ascertained to reveal dislocations by a successive polishing and etching method.

Reactive molecular beam epitaxy of aluminium nitride

Abstract: Single‐crystal films of AlN have been fabricated on sapphire and silicon substrates by reactive molecular beam epitaxy (MBE) at 1000°–1200°C. The crystal quality of the films was examined by the reflection high‐energy electron diffraction technique and the measurement of cathodoluminescence. The results suggest that the epitaxial AlN films prepared by reactive MBE have good quality as compared with bulk single crystal. Nonlinear optical coefficients and the electromechanical coupling constant of surface acoustic waves were also measured. It was shown that the AlN films have unique characteristics and are promising for nonlinear optical and surface acoustic wave devices.

The Temperature Dependence of Rare‐Earth Activated Garnet Phosphors I . Intensity and Lifetime Measurements on Undoped and Ce‐Doped

Abstract: The temperature dependence of cathodoluminescence intensity and decay time in the range 10°–480°K is reported for both nominally pure and Ce3+‐doped yttrium aluminum garnet (YAG) crystals. Luminescence from characteristic lattice defect levels and from the 2D excited state of the Ce3+ impurity have both been investigated. Three lattice defect levels have been detected. Recombination is dominated by the shallowest centers at low temperatures but, at higher temperatures, thermalization effects reduce the net capture rates at the competitive defect centers. As a result of thermalization in the doped crystal the efficiency of energy transfer to Ce3+ is increased, but only at the expense of introducing a long afterglow into the Ce3+ decay near room temperature. The temperature‐dependent interactions between different centers in the lattice are adequately described by a simple kinetic model. Each center is characterized by three rate processes and by a thermalization activation energy which is derived from the decay time data. The experimental results are in good agreement with preliminary studies already reported by the authors.

Properties of VPE‐grown GaN doped with Al and some iron‐group metals

Abstract: The influence of doping with Al and some iron‐group metals on optical and electrical properties of GaN epitaxially grown from the vapor phase is reported. These impurities are the main inadvertent contaminants in GaN growth, and details on the transport of these elements to the deposition area during growth are discussed. Detailed studies are performed by the SIMS technique on impurity concentration in layers grown under different conditions of contamination. Variations in doping over the area of the grown wafers, as well as with depth into the layer, are investigated, and shown to be significant for Al and N2. These investigations are complemented by SEM cathodoluminescence topographs. Photoluminescence data strongly indicate the existence of a shallow bound exciton (binding energy 12 meV) to isoelectronic Al on Ga sites in GaN. No radiative states are observed from the iron‐group contaminants. These (notably Fe and Cr) cause deep states efficient in electrical compensation of the material, which is easily made highly resistive with Fe or Cr.

The Relationship Between Concentration and Efficiency in Rare Earth Activated Phosphors

Abstract: The concentration‐dependence of cathodoluminescence efficiency for the rare earth (RE) activators Ce3+, Eu3+, Gd3+, or Tb3+ in YAG is compared at 300° and 25°K and extends earlier work by the authors. Competitive recombination processes in the garnet phosphors include recombination at bulk lattice defect centers detected by near‐u.v. luminescence and recombination through some shunt pathway dominant at 300°K. The experimental data are interpreted using a kinetic model, and except for the activator Gd3+, can be fitted using reasonable parameters. The model includes a specific mechanism for activator excitation (a bound exciton state) and the effects of impurity scattering on the rate of recombination at lattice defect centers. It is shown that one of the most important factors governing the general shape of the efficiency vs. concentration curve is the activator thermalization factor, . This in turn depends on the depth of the impurity potential and the rate of energy transfer in the impurity‐exciton complex. The general theory as developed is useful in rationalizing earlier measurements of efficiency vs. concentration for RE activators in a number of oxide and oxysulfide lattices.

The characterization of highly‐zinc‐doped InP crystals

Abstract: Highly‐zinc‐doped InP crystals, grown along the 〈111〉 direction by the liquid‐encapsulated Czochralski technique, have been characterized by x‐ray topography, transmission cathodoluminescence, and transmission electron microscopy. It is observed that high zinc doping improves the perfection of crystals insofar as dislocations are concerned. However, not all of the dopant atoms are in solid solution, and it appears that some have clustered to form a high density (∼4×108 cm−2) of fine precipitates with an average size of ∼675 A.

Visualization of the grain‐boundary potential barriers of PTC‐type BaTiO3 ceramics by cathodoluminescence in an electron‐probe microanalyzer

Abstract: In PTC‐type BaTiO3 ceramics the grain‐boundary potential barriers can easily be made visible by means of cathodoluminescence in an electron‐probe microanalyzer without having a voltage applied. The visualization can be explained in the context of our finding that the intensity of the cathodoluminescence radiation depends sensitively on the number of charge carriers present in the conduction band. Thus, in PTC‐type BaTiO3 ceramics, but not in undoped or reduced materials, the deficiency of conduction electrons in the depletion layers at the grain boundaries causes a blackening of the grain‐boundary regions. The blackening of the grain boundaries of our PTC ceramics also appears below the Curie point, thus demonstrating the existence of grain‐boundary barriers in this temperature range.

Transmission cathodoluminescence: A new SEM technique to study defects in bulk semiconductor samples

Abstract: We report a new method of defect electron microscopy which is an extension of the well‐known method of cathodoluminescence. In this new technique, which we call transmission cathodoluminescence, the luminescence generated at the top surface of a thick semiconductor sample by the electron beam of a scanning electron microscope (SEM) is used to probe the sample volume beneath the luminescing region. The intensity of the transmitted luminescence collected by a small solid‐state detector mounted beneath the sample is displayed on the cathode ray tube of the SEM. We demonstrate the usefulness of this technique by imaging dislocations in GaAs substrates and GaAlAs epitaxial layers.

Lattice constants and band‐gap variations of the pentenary semiconductor system Cu1−yAgyInS2(1−x)Se2x

Abstract: Much interest has been expressed in the ternary IB‐III‐VI2 group semiconductors for use in electro‐optical devices such as solar cells. Subsets of these, AgInS2, AgInSe2, CuInS2, and CuInSe2, have been combined to form the pentenary alloy system Cu1−yAgyInS2(1−x)Se2x. With such an alloy the band gap may be varied while keeping the lattice constant fixed. Samples were prepared by reacting stoichiometric powder mixtures at about 900 °C. X‐ray diffractometry tests suggest the alloys maintained complete solid solubility throughout the system in a chalcopyrite‐type crystal structure. From cathodoluminescence studies on pressed bars of these powders the band‐gap energies were estimated at 300 °K. These tests suggest that the alloys are all direct‐band‐gap semiconductors.

Efficiency enhancement in manganese‐doped zinc silicate phosphor with AlPO4 substitution

Abstract: Large increases in the efficiency of the conventional P1 phosphor (Zn2SiO4 : Mn) have been achieved by substituting various combinations of III and V oxides for SiO2, e.g., Zn2−yMnySi1−2xAlxPxO4. The photoluminescence efficiency of these phosphors exhibit strong dependence on the AlPO4 substitution concentration x having highest efficiency for x=0.5–1.5%. Cathodoluminescence and decay measurements show that such materials can have CL efficiency greater than the commercial zinc silicates and still have comparable persistence characteristics.

Screen Weight Optimization

Abstract: A common visual subjective technique used in screening is shown to be related to the void transmission of the phosphor screen. A theoretical model of stacked phosphor particles is developed, which can be used to empirically determine the void transmission of a phosphor screen. It is shown that in the region where void transmission is no longer the dominant factor for light transmission, the optimized screen weight is obtained. By considering the electron density of the screen, the screen weight at which maximum cathodoluminescence occurs can be predicted and proves to be in good agreement with experimental results.

Evaluation of defects in InP and InGaAsP by transmission cathodoluminescence

Abstract: We present the results of a defect study of liquid‐encapsulated Czochralski grown InP substrates, and InGaAsP grown by liquid‐phase epitaxy using the technique of transmission cathodoluminescence (TCL). The TCL image of the dislocations permits an accurate measurement of the dislocation density and an estimate of the minority‐carrier diffusion length. In addition to the dislocation density, TCL examines material quality by imaging growth‐related defects, e.g., growth striations in InP and stacking faults in InGaAsP.

Low-Voltage Red Cathodoluminescence of ZnSe Single Crystals

Abstract: Red cathodoluminescence from low-resistivity ZnSe:Cu, Al crystals has been observed under the excitation of low-energy electron beams below 50 V. Dependence of current and brightness on applied voltage as well as the emission spectrum are presented. The red emission of ZnSe tends to be subjected to degradation, while the blue and the green of ZnS previously observed do not.

Imaging of dislocations in InP using transmission cathodoluminescence

Abstract: We present an exact one‐to‐one correspondence between grown‐in dislocations revealed as etch pits and dark spots observed on transmission cathodoluminescence images of InP. This correspondence has been demonstrated for heavily tellurium‐ and sulfur‐doped InP. This is the first time that dislocations have been identified and imaged in InP using a luminescence technique.

Electrical, SEM, and TEM studies on ZnTe. II. Impurity segregation during long annealing

Abstract: The contrast obtained after long thermal annealing in high purity ZnTe, by SEM in the cathodoluminescence mode and by TEM is investigated. The contrast variations are due to segregation or extraction of residual impurities (LiZn, CuZn, AgZn) which control the electrical and optical properties of the ZnTe. These studies allow to approach the self-compensation problem in terms of limited solubilities of impurities, function of stoichiometry and temperature, the acceptor reserve being in the defects (inclusions, grain, and subgrain boundaries). A l'aide de la MEB en mode cathodoluminescence et de la MET, nous avons etudie les variations de contraste apres recuit thermique de longue duree dans le ZnTe. Les variations de contraste observees sont interpretees en termes de segregation et extraction des impuretes residuelles (LiZn, CuZn, AgZn) qui sont responsables des proprietes electriques et optiques du materiau. L'etude des recuits thermiques dans le ZnTe montre que les equilibres stoechiometriques sont regis pendant les recuits de courte duree par des equilibres locaux, c'est-a-dire dependant du voisinage de l'impurete (inclusions de Te). Pour les recuits de longue duree l'atmosphere de recuit devient preponderante et celle-ci contrǒle la solubilite ou l'extraction des impuretes.

Luminescence studies of the 1.36‐eV band and the edge emission in melt‐grown GaAs : Cu

Abstract: Using cathodoluminescence at 80–300 K, we have measured the injection‐level dependence, frequency response, and temperature dependence of the 1.36‐eV band and the 1.50‐eV edge emission in melt‐grown GaAs : Cu. The overall luminescence is relatively weak and the 1.36‐eV band is accompanied by a broad emission in the range 1–1.4 eV. We find no inherent injection‐level shift in the shape or peak energy of the 1.36‐eV band. With increasing temperature, the 1.36‐eV band quenches out with an activation energy of 0.1 eV or greater. The frequency response of the 1.36‐eV band is remarkably flat for deep‐center luminescence. However, the frequency response of the edge emission is unusually slow in GaAs : Cu, particularly in samples with higher Cu concentration. Variation in injection level causes a strong variation in branching between the 1.36‐eV band and the edge emission which is not explicable in terms of occupancy effects related to the relative kinetics of the two bands. It is suggested that the 1.36‐eV band results from a conduction‐band‐to‐acceptor transition whose kinetics are dominated by another process such as equilibration of a carrier trapping level in resonance with the conduction‐band continuum. The same process also controls the kinetics of the edge emission transition in GaAs : Cu with high Cu concentration.

Anomalous electrical and optical characteristics of GaAs/AlxGa1−xAs heterostructure materials

Abstract: GaAs/AlxGa1−xAs double‐heterostructure layers grown by conventional LPE have been examined by three separate electrical and optical techniques. Some material has exhibited characteristics not previously detected in epilayers grown singly. Our central observation is that Ge‐doped passive and active layers can have n‐type conductivity within the heterostructure.

Temperature, injection level, and frequency dependences of some extrinsic luminescence bands in ZnTe

Abstract: Using cathodoluminescence measurements between 80 and 300 K, we have investigated the temperature, injection level, and frequency dependences of three extrinsic luminescence bands in nominally undoped ZnTe. At 80 K, our material shows strong edge emission and broad extrinsic bands near 1.59 eV and 2.08 eV. The peak position and band shape are independent of injection level for both these extrinsic bands. 80 K measurements of frequency response from 50 Hz to 50 MHz show that the 2.08 eV band has an exponential time decay with a time constant of 0.045 μsec. The 1.59 eV band shows a more complex frequency dependence which indicates that one component of the response has a time constant of 1.85 μsec. At temperatures above 80 K the 2.08 eV band quenches with an activation energy of 0.22 eV which indicates that the transition originates near a band edge. In contrast, the 1.59 eV band does not quench with a well-defined activation energy. The peak of the 2.08 eV band follows the temperature dependence of the edge emission energy, whereas the peak of the 1.59 eV band shows the opposite behavior. From the 80 K measurements, we conclude that the 2.08 eV band results from a conduction-band-to-acceptor transition, and that the 1.59 eV band results from an intracenter transition between localized levels of a compact complex. Above 160 K, a third extrinsic band begins to appear near 1.80 eV. This band becanes increasingly prominent with increasing temperature. The frequency dependence of the 1.80 eV band at 166 K indicates an exponential time decay with a time constant of 0.37 μsec. However, the frequency dependence changes with temperature and a faster component appears in the response at 300 K. We conclude that 1.80 eV band near 160 K results primarily from an intracenter transition in a compact complex, and that there is a contribution from free-to-bound transitions at 300 K.

Evidence for the interaction between the free exciton and the F - centres in KI

Abstract: 2014 The resonant emission observed at 2 125 Å in KI irradiated with electrons at very low temperature increases when the crystal is illuminated by infrared light. This phenomenon is due to the ionization of the Fcentres, which are efficient quenching centres for the free excitons. LE JOURNAL DE PHYSIQUE LETTRES Tome 40, 1 er juillet 1979, Classification Physics Abstracts 61.80F 71.35 78. 60H Recently [1], a new mechanism concerning primary defect formation was introduced, suggesting that the precursor states responsible for their creation are states of the free exciton (F.E.). On the other hand, experiments on KCI have shown it was possible to reach these precursor states by optical pumping in the exciton relaxed state absorption band [2]. In the iodides, and particularly in KI, the free exciton emits a narrow luminescence band labelled resonant emission, the properties of which are well known (see for instance the references given in [3]). This band can be observed under electron irradiation of the crystal, in spite of the fact that most of the created excitons with this kind of excitation are self-trapped excitons (STE) due to capture of electrons by self trapped holes (Vk centres). We have studied the influence on the F.E. luminescence of an additional irradiation of the sample during the electron irradiation. This second irradiation was made with a light of wavelength in the STE main absorption band region [4] which, following ref. [1] could induce an increase in the F.E. emission due to an STE -+ F.E. transition. For the resonant emission observation, we use the experimental procedure described in ref. [3]. The instantaneous electron energy deposition rate is 1022 eV/cm-3 S-1. The pumping source is initially a 1 W YAG : Nd (*) Equipe de recherche associée au C.N.R.S. (**) Equipe de recherche n° 5 du C.N.R.S. Laser, the 1.06 g emission of which is well adapted for the exciton absorption band in KI [4]. Then, we observe that, when the crystal is under I.R. stimulation during the electronic excitation, the resonant luminescence signal IR increases by AIR. This quantity can be, following the conditions, of the order of magnitude of the initial signal. AIR persists unchanged during the optical stimulation. The increase and decrease of AIR at the beginning and end of the stimulation are faster than the time constant of the apparatus (100 ms), which agrees with an electronic effect. However the following results show unambiguously that the observed effect is not due to the STE -+ F.E. transition, but to the influence of the stimulation on the centres created by the irradiation, in spite of the fact that the defect creation rate in KI near liquid helium temperarature is very low : (a) the resonant luminescence changes with irradiation time : it decreases with dose, whereas AIR increases (Fig. 1) : AIR is about zero with small doses, increases with the irradiation time to a certain limit. This effect is independent of the origin of the studied crystals (Crystal Growth Laboratory Salt Lake City; Laboratoire de Physique Cristalline, Orsay), and is also, roughly, independent of impurity content; (b) AIR for determined conditions, depends on the stimulation intensity but a saturation can be reached for a given intensity of the stimulation light; (c) By using a light source of 150 W followed by a Bausch and Lomb monochroArticle published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyslet:019790040013031300 L-314 JOURNAL DE PHYSIQUE LETTRES Fig. 1. Free exciton luminescence intensity, excited by a pulsed electron source, with (8, ! ) and without (0, A) additional light stimulation (see text). The results are displayed for 2 samples : a ultra pure KI (0, ~) and a Li-doped (10 ppm) KI ( 1, A). Electron voltage : 30 kV ; Average current ~ 0.1 1 gA x cm-2 ; Repetition state : 600 Hz; Light source : 90 W xenon lamp + « MTO J 721