Diamond anvil cells have been used to generate a wide range of pressures, from 0.1 GPa to over 400 GPa (for reference, the center of Earth is about 360 GPa). Samples are squeezed between two diamond anvils and studied using infrared, visible, and x-ray probes. In the past year a new synchrotron beam line has become available at CHESS for the general user for diamond anvil cell work using x rays. This has opened up new areas of research as the experimenters need only to bring a sample in a diamond anvil cell and can leave with the x-ray data mostly analyzed. Current x-ray diffraction work at CHESS on materials subjected to static pressures up to 400 GPa are reviewed. Both energy dispersive and Laue diffraction techniques have been applied to phase transition, equation of state, and state of stress problems. Since most samples at very high pressures are powders, energy dispersive diffraction is most often used. An example of this is xenon which turns metallic at 150 GPa. Since the plasma frequency of xenon is in the infrared and because of the presence of an absorption band at 2 eV, xenon is a transparent blue metal at this pressure. The energy dispersive diffraction data provided the structural and equation of state information needed to understand the physics of the problem. An example of Laue diffraction using diamond anvil cell is the study of the state of stress of diamond anvils themselves. In an ongoing experiment at CHESS, the tips of highly stressed diamonds are analyzed by studying the energy distribution of various Laue spots using a solid state detector.
The brightness of third-generation synchrotron radiation sources is so high that the x-ray optics of diffractometers can be aberration-free and their instrument functions can be perfectly symmetric. At the same time there is sufficient intensity that no further compromises are required and experiments are brief. Adequate stability and resolution can be obtained only if care is taken to eliminate the effects of primary beam heating on the x-ray optics. At present state of the art, the quality of data need depend only on the quality of the sample be it a powder, a polycrystal, or a thin film. Achievements include lattice parameter determination in the parts-per-million range, high intensity, and high resolution for structural studies, anomalous dispersion for structure refinement and phase identification, and high resolution angle and energy scanning in the same instrument. Most recently we have demonstrated aberration-free powder patterns at SSRL and perfect monochromators for one-third kW x-ray beams at NSLS.
The development of ultrahigh-brightness x-ray sources makes time-resolved x-ray studies more and more feasible. Improvements in x-ray optics components are also critical for obtaining the appropriate beam for a particular type of experiment. Moreover, fast parallel detectors will be essential in order to exploit the combination of high intensity x-ray sources and novel optics for time-resolved experiments. A CCD detector with a time resolution of microseconds has been developed at the Advanced Photon Source (APS). This detector is fully programmable using CAMAC electronics and a MicroVax computer. The techniques of time- resolved x-ray studies, which include scattering, microradiography, microtomography, stroboscopy, etc., can be applied to a range of phenomena (including rapid thermal annealing, surface ordering, crystallization, and the kinetics of phase transition) in order to understand these time-dependent microscopic processes. Some of these applications are illustrated by recent results performed at synchrotrons. New powerful x-ray sources now under construction offer the opportunity to apply innovative approaches in time-resolved work.
Interference among Bragg beams and effects of polarization state mixing in a multibeam diffraction process leads to a new method of determining noncentrosymmetric phases when an elliptically polarized x-ray beam is used. This idea is demonstrated by an experiment on a GaAs single crystal. With a structurally known noncentrosymmetric crystal the same method can be used to measure the degree of a circular polarization.
A new flaw reconstruction scheme is proposed for locating and sizing flaws embedded in materials as well as estimating their absorption coefficients in an unified manner. Based on x- ray projections and flawless prototype tomogram, the proposed method locates flaws by detecting their convex hulls, fits elliptic models into the convex hulls with identical moments of inertia, and estimates flaws' absorption coefficients using least-squares principle. Due to the capability of annihilating false-alarm flaws, the proposed method improves the detection performance of the convex hull method. For convex type of flaws, simulation results show that the proposed method can produce very accurate estimates of flaw intensity as well as flaw size and orientation for full-view and limited-view noiseless cases.
A series of experimental measurements was conducted for the characterization of a transmission circular zone plate. The zone plate (ZP), with a primary focal length of 40 cm for 8 keV photons, was illuminated by monochromatized synchrotron x rays. Focusing efficiency of the ZP was measured as a function of x-ray energy between 5 - 1 keV, from which the ZP thickness was determined. Focal spot sizes at the first, second, and third order focus were measured, and they agreed very well with the calculated values. Images of a 1000 mesh/inch gold grid were also obtained at the three focal planes. The grid scans indicated that the spatial resolution is about 2 (mu) in the image obtained at the third order focus.
The theory and practice of x-ray absorption spectroscopy are discussed in brief, together with the kind and quality of the atomic-scale structural information that it can yield. Its application to special materials systems is emphasized.
For anisotropic structures such as those found for the high Tc superconductors, the utility of x-ray absorption measurements can be enhanced by taking advantage of the inherent polarization of synchrotron radiation. Polarization dependent measurements have been made on all four classes of high Tc materials (La, Y, Bi, and Tl based Cu oxides) after orienting them magnetically. All show similar polarization dependence for their near edge structure which is characteristic of planar Cu compounds. The paper concentrates on results for YBa2Cu3O7-x materials. The two types of Cu sites in this material have different polarization dependence, and measurements with the x-ray polarization parallel and perpendicular to the c-axis can be used to distinguish their contributions to the near edge structure. We can clearly resolve the 4p(pi) contributions from the two sites. The extended fine structure also shows distinct polarization dependence which allows the separation of contributions from in-plane and out-of-plane bonds. Measurements on oxygen deficient materials show contributions from Cu1+ which is determined to be on the Cu(1) (linear chain) sites from the polarization dependence, and to vary linearly with the O concentration.
Zn, Ni, Co, and Fe substitute for Cu in YBaCu3O7 and strongly suppress the transition temperature Tc. Surprisingly, Zn at low concentrations suppresses Tc the most and Tc goes to 0 at 8 - 10% Zn. There is still controversy as to the Zn site -- plane sites Cu(2) or chain sites Cu(1). We present x-ray absorption data (XAFS) for several concentrations of Zn and for samples from four different sources. The XAFS data indicate that many samples have significant ZnO contamination. Analysis of samples for which the ZnO contamination is negligible indicates that the Zn is mostly on the planes but displaced along the c-axis.
A new approach to the analysis of x-ray absorption fine structure (XAFS) data is presented. It is based on the use of radial distribution functions directly calculated from a single-particle ion hamiltonian containing model potentials. The starting point of this approach is the statistical average of the XAFS for an atomic pair. This average can be computed using a radial distribution function (RDF), which can be expressed in terms of the eigenvalues and wavefunctions associated with the model potential. The pair potential describing the ionic motion is then expressed in terms of parameters that are determined by fitting this statistical average to the experimental XAFS spectrum. This approach allows the use of XAFS as a tool for mapping near-neighbor interatomic potentials, and allows the treatment of systems which exhibit strongly anharmonic potentials which can not be treated by perturbative methods. Using this method we have analyzed the high temperature behavior of the oxygen contributions to the Fe K-edge XAFS in the ferrosilicate minerals andradite (Ca3Fe2Si3O12) and magnesiowustite (Mg0.9Fe0.1O). Using a temperature dependent anharmonic correction derived from these model compounds, we have found evidence for a local structural change in the Fe-O coordination environment upon melting of the geologically important mineral fayalite (Fe2SiO4). We have also employed this method to the study of the axial oxygen contributions to the polarized Cu K-edge XAFS on orientated samples of YBa2Cu3O7 and related compounds. From this study we find evidence for an axial oxygen-centered lattice distortion accompanying the superconducting phase transition and a correlation between this distortion and Tc. The relation of the observed lattice distortion to mechanisms of superconductivity is discussed.
The complementary nature of x-ray absorption fine structure (XAFS) spectroscopy and x-ray diffraction (XRD) is described. In particular XAFS records the local structure while XRD detects the long-range crystallinity enabling the heterogeneity of materials like single-phase catalysts to be explored. Both measurements can be combined to facilitate novel in situ experiments. We have used a horizontal energy dispersed x-ray beam and a photodiode array to detect transmission XAFS and a curved position sensitive detector positioned in the vertical plane to record XRD. This arrangement has been used to follow the formation of mixed-oxide catalysts.
We apply high resolution x-ray diffraction to study the development of the c(2 X 2) phase of Pb on Ni(001). We observe lineshapes in agreement with the scaling theory of Ohta, Jasnow, and Kawasaki, but with an anomalously slow time dependence of average domain size R on time t, R (alpha) t0.12. The presence of 1% - 3% carbon on the surface apparently acts as a weak random field. We are able to understand the observed power law dependence and the observed effect of C density in an adaptation of the analysis of the random field Ising model given by Grinstein and Fernandez.
A series of superlattices were grown by molecular beam epitaxy on (100) GaSb substrates that had been miscut by 2, 3, and 4 degrees toward the <011> direction. These superlattices were then studied by scanning all possible  or  (asymmetric) reflections with high resolution multiple-crystal x-ray diffractometry. In addition, the (400) (quasi-symmetric) reflection was scanned. From peak splittings we extracted mismatch and tilt parameters for the epitaxial unit cell. We compared our results for the nontetragonal component of the distortion to calculations based on the coherent strain model of Hornstra and Bartels. We find that this model, which was developed for epitaxial growth on a general (hkl) plane, also describes our results for growth on vicinal (100) planes. The resolution of our data is sufficient to establish that the distortion was not purely tetragonal. A monoclinic unit cell symmetry adequately describes our results.
It has been known for decades that lipid molecules can be spread on the surface of water to form monolayers, and that these monolayers undergo phase transitions as a function of pressure and temperature. However, it is only in the past few years, and only because of the availability of synchrotron radiation, that the structures of a number of Langmuir monolayer phases have been determined. In most cases these results contradict the assumptions commonly made earlier.
The structure of the solid-liquid interface is of fundamental importance in chemistry, but progress in understanding this interface has been slow, due to the lack of in-situ probes that provide information at atomic scales. Recently, in-situ surface x-ray scattering measurements have provided insight into the microscopic nature of solid-liquid interfaces and this paper discusses experiments on electrochemically deposited monolayers of Pb, Tl, and Bi on Ag and Au (111) electrodes. Tl and Pb form 2-D, incommensurate hexagonal solids that are compressed relative to bulk and rotated by 4 - 5 degree(s) with respect to the substrate. As the applied electrode potential decreases, the in-plane atomic spacing also decreases, since the chemical potential of the monolayer increases. From these data, the 2-D compressibility of the monolayer can be calculated. We find that the compressibility is only slightly dependent on substrate, being smaller on Ag(111) than on Au(111). For Tl/Ag(111), the intensity of the Ag surface diffraction changes when the monolayer is adsorbed. This results from a substrate- induced modulation of the atomic positions in the incommensurate monolayer and we have quantified this modulation. Bi/Ag(111) forms an unusual structure: a rectangular lattice that is uniaxially commensurate with the hexagonal surface. There are two Bi adatoms per rectangular unit cell and one adatom is displaced from the centered position by 0.35 angstrom. The commensurate Bi rows lie along the rows of threefold hollow sites on the Ag(111) surface. This unusual structure reflects the tendency toward covalent bonding found in Bi and a fortuitous match between the atomic spacings of the Ag substrate and the close packed planes of bulk Bi. In contrast to Tl and Pb where the compressibility is isotropic, Bi/Ag(111) compresses anisotropically and this maintains the uniaxially commensurate structure. Our results show that for these metal monolayer systems the adatom-adatom interactions determine the atomic structure of the monolayer and the adatom-substrate interactions only weakly affect this structure. Furthermore, the structure is not influenced by the presence of the large concentration of adsorbed water molecules or anions.
X-ray standing waves generated above a mirror surface during total external reflection have been used to locate a heavy atom layer which was embedded hundreds of angstroms above the mirror surface in a Langmuir-Blodgett multilayer. This same method has also been used to map out the ion distribution in the diffuse double layer that forms at the electrolyte/charged surface interface.
We present initial results of an in situ structural investigation of the underpotential
deposition of copper on an iodine covered platinum/carbon layered synthetic microstructure,
using x-ray standing waves generated by specular (total external) reflection and Bragg
diffraction. We also compare the result of surface coverage isotherms derived from both
electrochemical and x-ray measurements.
Application of fast-acting digital computers and technical means of signal treatment allowed autoradiographic analysis in material science by investigating component distribution of the base and impurities in optical monocrystals. Heterogeneity of the growing crystals caused by the unevenness of component distribution of the base and impurities dictates the necessity of improved methods of study. One of the methods widely used in crystal homogeneity investigations, activation radiography, is based on the secondary (beta) -irradiation registration. Analysis of autoradiographic features is connected with the elucidation of the optical density distribution of the exposed and treated photoemulsion, used in the quality of detector of the secondary (beta) -irradiation. In this paper the possibilities of the digital image processing (DIP) by the autoradiographic investigation of the monocrystals, on the basis of hufnium dioxide, stabilized by the different rare-earth elements are shown.