In high Tc superconductors, the low critical current densities in polycrystalline materials have been attributed to a combination of critical current anisotropy and poor superconducting coupling across grain boundaries. Theoretical calculations indicate that although the flux pinning should vary roughly inversely as grain size, the polycrystalline critical current behavior could possibly be understood in terms of stresses due to the grain boundary. Experiments have been conducted to increase the coupling between adjacent grains by modifying grain boundary chemistry. These include adding either a conducting layer or a superconducting layer at the interfaces. The effect of additions such as Ag, B, Bi, Ga, and In to produce a conducting layer and the alloying of RBa2Cu307 with another superconductor to produce a superconducting layer were analyzed by measuring Tc, Jc, and observing changes to the microstructure. Early results indicate some Jc enhancement with silver addition. However, the addition of a different superconductor appears more promising.
The fabrication of Ti/Pt/Au ohmic contacts on diamond using two transmission line model masks during the photolithography step was modified as a result of the adverse effects on the resistance from the rectifying lip created by the overlap of the two masks, and the possible inhibition of carbide formation due to the presence of oxygen on the diamond surface before metallization. The first modification consisted of decreasing the rectifying lip by diffusing a small amount of Ti from beneath the contact defined by the first mask, and decreasing the overlap of the two masks from 5 to 2 μm, which is close to the minimum allowable by our photolithography techniques. The second modification consisted of the desorption of oxygen from the diamond surface using a heat treatment in vacuum and cool down in purified hydrogen. As a result of these changes, the contact resistance was decreased by more than two orders of magnitude from 8.1 × 10-2 Ω-cm2 to 1.2 × 10-4 Ω-cm2.
Ohmic contacts on diamond were fabricated using a Ti/Pt/Au metallization scheme and two transmission line model masks. The double mask system is used to isolate the Ti by surrounding it with Pt. Pt is expected to prevent Ti diffusion either to the contact surface or along the diamond surface during annealing. In this study we found that this system of masks was successful in preventing the Ti diffusion as revealed by Auger analysis on the diamond and contact surfaces. An Auger depth profile revealed that the Ti was confined to the diamond/contact interfacial region, and was most likely combined with Pt.
Ohmic contacts have been fabricated on a naturally occurring type IIb diamond crystal using an annealed Ti/Pt/Au trilayer metallization where the Pt served successfully as a barrier to Ti diffusion into the Au capping layer. However, a specific contact resistance could not be reliably determined using transmission line model measurements. Auger microanalysis revealed the presence of Ti on the diamond surface near the ohmic contact pads. The most likely origin of the Ti on the diamond surface was determined to be lateral diffusion from beneath the contact pads. This would have produced a nonuniform concentration of Ti across the diamond surface which, in turn, would have affected the diamond sheet resistance in a complicated way.
Transmission electron microscopy was used to analyze the microstructure of YBa2Cu3O7−x/Y2O3/YBa2Cu3O7−x trilayer structures deposited by off-axis sputtering on MgO substrates with varying degrees of roughness. Substrate surface hillocks with a peak-to-valley height of about 4.5 nm were found to contribute to strain that extended through the film and disrupted the smoothness of the Y2O3 layer. In some cases, these hillocks served as nucleation sites for yttria precipitates. Such defects may seriously jeopardize the realization of weak-link Josephson junctions.
The growth of (100) diamond/iron/copper multilayer structures has been examined by reflection high energy electron diffraction, extended X-ray absorption fine structure, and scanning electron microscopy in an effort to determine the thickness limit for metastable face-centered-cubic Fe on (100) diamond. Both copper films deposited on iron layers with thicknesses below 1.4 nm and the iron layers themselves were found to be face-centered cubic single crystal, while films grown on iron that was 2.0 nm and thicker and the iron itself were found to be polycrystalline. This critical thickness range of 1.4–2.0 nm compares well with the theoretically calculated value of 1.8 nm. This value was determined using the mechanical equilibrium theories (Matthews-Blakeslee and van der Merwe) with a lattice parameter for face-centered cubic iron that was derived by estimating the functional form of the linear thermal expansion coefficient and extrapolating the Poisson's ratio for austenitic stainless steel to the temperature of interest. The shear modulus, and intrinsic stacking fault energy for fcc Fe from ∼ 1350°C down to below room temperature have also been estimated. A more likely room temperature lattice parameter for fcc Fe than is usually assumed was estimated to be 0.3579 nm. The measured in-plane lattice parameter of strained fcc Fe on diamond was 3.54 ± 0.1 Å.
Copper films were grown on a single crystal diamond substrate using an iron seed layer. The effect of the crystalline structure of the iron seed on the Cu films was studied with extended x-ray absorption fine structure (EXAFS) and scanning electron microscopy (SEM). The EXAFS study shows that the 10 Å Fe seed layer is in an fee structure, and has collapsed into a bec structure by the time 20 Å of Fe has been deposited. In the SEM pictures it is observed that subsequent layers of Cu grow as continuous films for thin fcc-Fe seeds, and grow in an island mode for the thick, bcc-Fe seeds.
Using transmission electron microscopy and energy dispersive x-ray spectroscopy analysis, the microstructure and composition of a YBa2Cu3O7−x/Y2O3/YBa2Cu3O7−x trilayer film deposited on MgO by off-axis sputtering at 670 °C and 100 mTorr was investigated. The YBa2Cu3O7−x layer in contact with the substrate was found to be mainly single phase. However, the top YBa2Cu3O7−x layer was embedded with 5–10 nm crystalline Y2O3 (yttria) particles, which disturbed the local microstructure. The top YBa2Cu3O7−x layer also had barium copper oxide particles covering its surface. The partial decomposition of the top YBa2Cu3O7−x layer may have been due to resputtering of the yttria layer which locally altered the deposition conditions. In particular, the oxygen pressure in the vicinity of the growing film may have increased due to sputtering of the yttria layer by plasma gas atoms and the sputtered target atoms. As a result, deposition occurred under conditions away from the thermodynamic stability line.
Grain growth during sintering of the polycrystalline high onset temperature superconductor YBa2Cu3O7−x has been examined by measuring grain sizes at various temperatures for different times above 900°C. The grain size distribution is found to be predominately lognormal after some sintering time. The mean grain size increases with both temperature and time. The standard deviation of the grain size distribution increases with temperature, but appears to be independent of time. The grain growth law may have an exponent of five. The grain growth rate may be affected by the grain size of the calcined powder.
A model incorporating the thermodynamic equilibrium oxygen content, oxygen in-diffusion, and oxide phases for the oxygenation of the YBa2Cu3O7−x is described. For sintered polycrystals, grain growth and the resulting grain size distribution are included. The model is used to calculate the volume percent of each oxide phase for several processing conditions of sintered specimens and is compared with available results on the quantification of phases present. Such a comparison indicates that as the concentration of impurity phases increases so does the concentration of the less oxygenated 123 phases, suggesting that impurities coating 123 grains act as oxygen indiffusion barriers. The model is also used to investigate the uniformity of the oxygen content in large twinned and detwinned single crystals such as have been used for measuring superconductivity parameters.
A thin film, produced by carbon ion implantation and outdiffusion at the temperature of irradiation, has been confirmed to be diamond using micro-Raman spectroscopy and electron diffraction. The Raman spectra contained a definitive diamond peak shifted slightly from that of natural diamond. This shift may be due to residual strain and is consistent with the broad linewidth. Fragments of the film over a micron in size were examined with transmission electron microscopy and found to be untwinned, single crystals. The film had been produced on a polycrystalline copper substrate. The carbon ions were at an ion energy of 120 keV and the irradiation was carried out to a high dose.
We report phase segregation based on cerium content in thick (>70μm) single crystals and a polycrystalline sample of Nd2−xCexCuO4−y with nominal composition x=0.15. The crystals showed segregation into sharply defined sheets of materia l with different cerium concentrations. These sheets in the a-b-plane of the crystals were apparent when polished cross-sections were viewed with crossed polarizers in visible light. The different color of each layer was correlated with cerium compositions by electron microprobe analysis. AC susceptibility showed a strong, well defined transition at 24.5 K and a second, weaker transition at 20 K. The appearance of the 20 K transition was shown to be associated with a layer of material with reduced cerium content. A polycrystalline sample also displayed small intra-grain regions with reduced cerium content.
The morphology of the nucleation and growth surfaces of oxygen-acetylene combustion deposited diamond thin films has been studied using scanning electron microscopy. The diameters of both nucleation cells and growth surface grains have been measured and analyzed statistically. The general shape of the distributions is found to change from nucleation to growth surface for each film. The frequency distribution of cells is generally normal, whereas, the distribution for the growth surface is found to be always lognormal. The change in shape of the distributions from normal for the nucleation surface to lognormal for the growth surface can be explained from a stochastic theory of growth.