Measurements of electron momentum density by positron annihilation are influenced by thepresence of defects in superconducting oxides. Both lifetime and Doppler broadening measurements are sensitive to the superconducting transition. Possible candidates for the defect species are discussed along with their role in modifying the positron response.

Microchemical and microstructural characterization, using established SEM and STEM techniques, has shown that most (>90% of the area) of the Bi-Sr-Ca-Cu-O thin films prepared by reactive magnetron sputtering on both NaCl and MgO substrates in the (100) orientation is the superconducting phase Bi2 Sr2 CaCu2 O8 (2212); with largely defect free single crystal grains oriented with the a-b conduction planes parallel to the substrate. The geometrical interconnectivity is high and the grain boundaries are clean and well ordered without any intermediate layers to limit the superconductivity. There were also small patches (>10%) in which the Bi- and Cu-rich phases, Bi2 SrCuO5 and CuO, had separated. The films had an onset Tc at 115K with zero resistance at 85K and a Jc of 105A/cm2 at liquid He temperature.

Electron diffraction, high resolution electron microscopy and STEM microanalysis have been carried out on bulk superconducting oxides of Bi-Sr-Ca-Cu-O, and on the new (TI,Pb)- Sr-Ca-Cu-O and Pb-Sr-R (rare earth)-Ca-Cu-O systems to elucidate microstructure and microchemistry. Real space imaging and microchemical analysis are virtually essential to provide information about the atomic structure of interfaces and other local defect structures as well as local chemical composition in the materials since microstructure controlscritical current parameter Jc. The HREM images have been interpreted where necessary using the multislice image simulations with 256 x 256 Fourier transform arrays. Superstructure modulations of the type present in Bi-based and TI-Ba-Ca-Cu based compounds are not observed with the Pb substituted thallium compounds examined and it seems likely that Pb substitution stabilizes the compositions.

Two pyrochlore solid solutions, Gd2(ZrxTi1-x)2O7 and Y2(ZrxTil-x)2O7, were investigated to clarify the relationships between composition, structural disorder and ionic and electronic transport. A defect chemical model was developed based on the results of electrical and x-ray diffraction measurements. Frenkel, reduction and bandgap energies were derived.

A detailed optical absorption study has been performed on a series of chemically reducedlithium niobate crystals. A threshold absorption above 3.8 eV has been identified as a bandgap excitation, while the entire subbandgap absorption has been demonstrated to be due to a single defect via optical bleaching experiments. A model based on a bipolaron is proposed to explain the observed optical properties.

W report here the nature of the hole traps in singly doged (Ce3+, Tb3+ or Sm3+) and doubly doped (Ce3+ or Eu2+ and Sm3+) MgS. Ce3+ switches its role from electron trap to hole trap when Sm3+ is added to the host. Stimulated luminescence involving these hole trapsare presented. On codoping with monovalent Li+ or Na+ ions V type hole centers can be significantly reduced in MgS doped with the trivalent ions.

The results confirming a significant role of dislocation—impurity interaction in changing local properties of plastically deformed semiconductor crystals have been presented.

A laser method for non-destructively mapping the distribution of aluminum-hole centers, Al-h, in quartz has been developed utilizing an aluminum-hole A3 absorption band nearly coincident with the He-Ne 632.8 nm wavelength. Al-h distribution for quartz sample with varying aluminum concentration compared with coloration and the dose dependence of competing aluminum compensation mechanisms Al-h and Al-OH is determined. From a comparison with electron spin resonance(ESR) results an approximate absorption coefficientcalibration of 0.1 cm-1/ppm Al-h is found.

The structural properties of the TiO2 - NbO2 solid solution have been investigated experimentally by the time—oi—flight neutron diffraction method. It is found that 1) the position parameter for oxygen ions decreases linearly with increasing Nb concentration, 2) the four short metal—oxygen distances increaselinearly with increasing Nb concentration, while the two long distances remain constant,3) the metal—oxygen—metal bond angle,α, exhibits a noticeable deviation from that found in rutile at low Nb concentration, 4) Nb4+ occupy both rgular cation and octahedral interstitial sites at low Nb concentration, but Ti44 ions are increasingly squeezed out from the regular cation sites to octahedral and tetrahedral interstitial sites at Nb concentrations higher than 30 mole%.

We present the results of Rietveld analysis on the neutron diffraction intensities of Er2Fe17, which crystallizes in the hexagonal Th2Ni17 phase. Of the reflection sets, h-k ≠ 3n, l = 2n+1 and l = 2n, the former fit better than the latter and are in general more intense. This suggests disordered transition-metal-substitution on the planes parallel to the c-axis resulting in stacking faults; ABABs... is the usual sequence of ordered transition-metalsubstitutions in the parent CaCu5 phase. Influence of such deformations on the interatomic distances inthe substitution zone and the consequent magnetic bonding and anisotropy is discussed.

We have investigated the “real time” diffusion and trapping of hydrogen insilicon with rapid 1 MHz capacitance and current-voltage measurements on boron and phosphorus doped Schottky-barrier and MIS structures. Hydrogen is introduced by implanting hydrogen ions into the front electrodes of these capacitors. As interpreted by numerical modeling, our data from p-type silicon indicate that the 300 K hydrogen diffusivity is =10-10cm2/s, that =10% of the mobile hydrogen is positively charged, and that these ions are trapped at boron acceptors with a capture radius of 70 Å. The behavior of the n-type capacitors was quite different; hydrogenation resulted in the appearance of a dense layer of positive charge quite near the silicon surface, and in some cases negatively charged species which compensate phosphorus donors.

The effects of different hydrogen or equivalent deuterium treatments on CdTe superficiallayers are studied. Surface chemical analysis is performed by x-ray photoelectron spectroscopy and hydrogen profiles are controlled by secondary ion mass spectroscopy using deuterium. Hydrogen plasma etching leads to a clean and nearly stoichiometric surface, and the surface abrasion rate is higher than the hydrogen in-diffusion rate. Deuterium implantation by a Kaufman source leads to surface depleted of tellurium and prevents them from further oxidation. Annealing under deuterium gas produces a similar surface, free fromoxide, but not stable with regard to further oxidation.

We present a kinetic model for hydrogen diffusion and deactivation of shallow dopants incrystalline silicon which accounts for some of the effects of dopant type and concentration on observed deuterium profiles. The predictions of this model are in accord with recent Secondary Ion Mass Spectrometry (SIMS) profiles of deuterium in both n— and p—type silicon for a range of resistivities after passivation in a deuterium plasma. Dopant type and concentration effects are explained by assuming the presence of a charged species and by allowing for the formation of immobile deuterium molecules. The deuterium profiles are controlled by chargestate—mediated reactions which form deuterium molecules and pairs with shallow donors or acceptors. Earlier research has found that the hydrogen interstitial in silicon has a deep donor state in the band gap. The observed behavior of the deuterium profiles implies that the Fermi level passes through this donor state for a resistivity between 1 and 10 Q—cm in boron—doped material at 125 ºC; i.e. the state is at Ev + 0.30 ± 0.10 eV.

Techniques of low-temperature photoluminescence, Auger electron spectroscopy, and X-ray photoelectron spectroscopy are used to study the residual defects remaining after the rapid thermal annealing of ion-implantation induced damage in As+ implanted (100) silicon. Rapid thermal annealing resulted in the electrical activation of a major portion of arsenic in the implanted zone, but a shallow region confined within a depth of =200 Å from the surface contained As complexed with Si and O atoms. Surface characterization results agree with the observed degradation in transport properties of the diodes constructed on the shallow N+/P As+ implanted B doped p-Si wafers.

Transition metals in silicon produce a number of photoluminescence bands.

In this paper we discuss and compare the electronic and vibronic properties of luminescence bands commonly related to iron, niquel, copper and gold impurities.

Defects in crystalline silicon caused by a BF+2 ion implantation dose of 3x1015cm-2 have been studied using plan-view and cross-section TEM and SIMS. Dislocations form at annealing temperatures above 1000ºC and are pinned below the Si surface by fluorine bubbles ∼10 nm diameter at a density of ∼1011cm-2. This microstructure is essentially stable at 1150ºC from 4 to 400 minutes.

The annealing behavior of the lattice damage induced by 1 MeV Si+ implantation, at a dose of 3 x 1015 cm-2, into (100) GaAs has been examined by high resolution TEM, RBS and SIMS. The samples were annealed at temperatures ranging from 500 to 900ºC for 10 sec by rapid thermal annealing (RTA). Two types of defects, perfect and partial dislocation loops, lying in the GaAs {111} planes, were found throughout the as-implanted region, with the highest defect density concentrated in a buried band centered at 1 μm below the surface. After annealing at 500ºC, the total defect density showed little change, but the defects had become partial dislocation loops. With further increase in the annealing temperature, the surface region became increasingly defect free as the remaining individual dislocations increased in size, accompanied by a decrease in their number density. RBS dechanneling analysis showed that after annealing at 900ºC, the crystal lattice of the near-surface region had recovered to the level of a pristine GaAs. SIMS measurement of the Si profile showed little movement of the implant species upon RTA processing, and the Si peak position correlated well with the maximum defect density in the defect band.

The interaction between impurity species and dislocations is an area of tremendous importance both from a scientific viewpoint and from the need to understand gettering mechanisms. Low temperature gettering will be a vital tool for future VLSI requirements. However, the direct and unambiguous measurement of point defects gettered onto dislocations isdifficult.

We have made a systematic DLTS study of a deep level which is clearly related to dislocations. The measurements were made in silicon deposited by molecular beam epitaxy (MBE). This near mid-gap state appears to result from the complexing of an impurity or defect with the dislocation. In silicon with relatively high dislocation densities we observe classical DLTS spectra. As the dislocation density becomes low we see clear evidence of “crowding” of the gettered impurities; even though the total number of defects is not high the local concentration near the dislocation causes spectral distortion. We are able to model these spectral anomalies by observing detailed capacitance transients.

In this paper we aim to present an overall description of the electronic properties of the gettered impurity or defect. The insight that this gives into gettering mechanisms isalso discussed.

Transmission electron microscopy at 300kV has been used in a thorough examination of CMO3 and bipolar processed wafers of Czochralski silicon. Diffraction contrast and tilting experiments have been performed on the defects which were observed and a number of new observations have been made on stacking faults and {001} platelets induced by oxygen precipitation.

We present high resolution transmission electron microscopy results of Pbl-xEuxTe alloys that show evidence for a compositional instability for x ≈ 0.5 when the alloys are grown on BaF2 substrates. The Pbl-xEuxTe solid solution becomes stable at room temperature if a buffer layer of PbTe is grown on the BaF2 substrate prior to thegrowth of the Pb1-xEuxTe layer. The stabilization ofthe Pb1-xEuxTe solid solution is the result of the additional energy term due to the strain between the Pbl-xEuxTe film and the PbTe buffer layer. The estimated critical temperatures for decomposition of the Pb1-xEuxTe alloys with and without the PbTe buffer layer are ≈0 K and.≈ 366 K, respectively in accord with our1 experimental observations. We also present models for the structure of the decomposed phaseand use them to obtain simulated images using computing methods. The simulated images are compared with those obtained experimentally.