The various surface phases formed on a K covered Au(110) surface at different coverages, deposition and annealing temperatures were characterized by LEED, AES and TDS. The structural characteristics and thermal stability of these phases were compared with those of other alkali covered fcc(110) transition metal surfaces, where the presence of the K(ad) induces a "missing row" reconstruction similar to that present already on the clean Au(110) surface.At K coverages < 0.3 monolayers (ML) the "missing row" reconstruction of the substrate is stabilized by the presence of K(ad). The K adlayer is adsorbed on a (1 x 2) substrate at coverages between 0.1 and 0.3 ML, while at theta-K less-than-or-equal-to 0.1 a (1 x 3) substrate is more stable. The rather mobile adlayer orders on these substrates at low temperatures only, where additional streaks in the LEED pattern indicate a continuous sequence of ordered structures with increasing coverage. At coverages > 0.3 ML the (1 x 2) coexists with islands of a (mixed layer) c(2 x 2) phase, which completely covers the surface at theta-K = 0.5.The formation of the thermodynamically stable phases is partly connected with kinetic barriers; all structural transformations that require a rearrangement of the substrate are activated while those that merely involve a reordering of the adlayer occur instantaneously even at low temperatures. K deposition at low temperatures (about 100 K) leads to the formation of metastable phases where the K adlayer is adsorbed on a (1 X 2) substrate, independent of coverage. Structural transitions are induced also by loss of K(ad) during annealing to higher temperatures, where stable higher coverage structures transform into lower coverage phases.The close agreement of the structural, thermodynamic and kinetic properties of the alkali induced "missing row" reconstructions on other fcc(110) transition metal surfaces with those of the ordered phases on the K-covered Au(110) surface, where the "missing row" reconstruction is more stable already for the clean surface, provides convincing experimental evidence for a common physical origin of these reconstructions.

A video-based analysis system for reflection high-energy electron diffraction (RHEED) is described which simultaneously measures the intensities and profiles of multiple diffraction beams. This system is used to record real-time RHEED intensity oscillations for layer-by-layer epitaxial growth. Fast Fourier transform analysis of the oscillation data is used to directly determine the growth rate and to accurately obtain phase information about the oscillations. This system is demonstrated and compared to other methods of recording RHEED oscillation data.

The intensities of several reflection high-energy electron diffraction (RHEED) beams have been recorded during molecular-beam epitaxial growth of GaAs(100) using a novel video intensity measurement system that records multiple RHEED beam intensities simultaneously. The RHEED beam intensities were recorded at varying angles of incidence and crystal substrate azimuth angles. Strong oscillations in the intensities in specular and nonspecular beams with the same period but varying phases have been measured. As noted by other investigators, the phase relationship of the oscillations of the various beams has been found to vary with incident and azimuthal angle. The results are examined with regard to recent studies of the role of Kikuchi processes on the phase of the specular beam. In contrast to other reports, it is found that although the diffracted intensities in the vicinity of the elastically diffracted beams are influenced by inelastically scattered electrons from the Kikuchi lines, these effects can not account for all the phase behavior observed.

The chemisorption of SiCl4, Si2Cl6, and chlorine on Si(111)7 × 7 has been characterized using soft X-ray photoemission with synchrotron radiation, thermal desorption spectroscopy, and Auger electron spectroscopy. SiCl4 dissociatively chemisorbs on room temperature Si(111)7 × 7 with an extremely low sticking coefficient, with only SiCl remaining on the surface. In contrast, Si2Cl6 chemisorbs with ∼ 500 times greater probability and then partly dissociates into SiClx (x = 1, 2, 3) fragments. A monolayer of Cl deposited directly also contains SiCl, SiCl2, and SiCl3 surface species, but they are created via reaction with substrate Si atoms and have lower Si2p core level binding energies. Upon heating the surface all the adsorbed Cl is removed via desorption of silicon chlorides, primarily SiCl2, indicating that SiCl4, Si2Cl6, and chlorine will etch Si(111)7 × 7 if an additional reactant is not avail to remove the surface Cl. Interestingly, the different reactivities of SiCl4 and Si2Cl6 upon adsorption can be explained by the dynamics of different adsorption mechanisms.

Considerable progress was made in the last years in utilizing reflection high energy electron diffraction (RHEED) intensity oscillations for determination of the growth rate and the composition in molecular beam epitaxial (MBE) growth of GaAs, AlGaAs and InGaAs. In this paper we report data from RHEED, X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) to show that while such practice is now routine in MBE, it may not be applicable in chemical beam epitaxy (CBE), at least not in the case of CBE growth of InxGa1−xAs on GaAs.

Indium antimonide has been epitaxially grown directly on sapphire. Reflection high-energy electron diffraction, transmission electron microscopy, and scanning electron microscopy data are presented to show that the indium antimonide layer is epitaxial, has an abrupt interface with the sapphire, and grows in the 〈111〉 direction. Mobility data show room-temperature mobilities as high as 1×104 cm2 /V s from some regions on the wafer.

The intensities of several reflection high-energy electron-diffraction beams have been recorded simultaneously at varying angles of incidence and crystal substrate azimuth angles during molecular-beam-epitaxial growth of GaAs(100). Strong oscillations in the intensities of specular and nonspecular beams with the same period but varying phases have been measured. The phase of the oscillations of the various beams has been found to vary with incident and azimuthal angle. A kinematic calculation based upon a simple model for epitaxial growth is presented, and its prediction concerning phase is compared with the experimental results. The results are also examined with regard to recent studies of the role of Kikuchi processes on the phase of the specular beam.

For the reconstructions present during epitaxial growth, the diffraction experiment of choice is reflection high-energy electron diffraction (RHEED). In the RHEED geometry, the momentum transfer normal to the surface is small and can be varied by changing the angle of incidence at constant incident energy. The resulting rocking curve, or diffracted intensity versus angle of incidence, is comparable to a low-energy electron diffraction IV profile. We present rocking curve results for the 〈001〉 and 〈11̄0〉 principle azimuths for 10-keV diffraction experiments from GaAs(110). This known surface structure is the best tested structure to develop the analysis. The measurements are compared to a fully convergent dynamical N-beam RHEED calculation which includes as many propagating and evanescent beams as needed. An important issue is to determine what extent the diffraction is dominated by bulk features and single-layer resonances. This is addressed by selectively eliminating beams in the analysis and comparing beam emergence predictions to experimental observations.

An accurate and fast dynamical theory for calculating reflection high-energy electron diffraction (RHEED) rocking-curve intensities at 10–40 keV is presented. The application of RHEED to surface structural analysis is demonstrated for the first time on a reconstructed surface by comparing data to calculated spectra for GaAs(110). The technique is highly sensitive to structural changes along specific directions due to its forward-scattering geometry.

Isoelectronic doping using antimony has been shown to reduce traps and improve material properties during epitaxial growth of Si doped GaAs(100) and AlGaAs(100). In this study, the effect of the antimony dopant on the optimal growth temperature is examined with the aim of producing high-quality heterostructures at lower temperatures. High-quality films of GaAs and AlGaAs have been grown by molecular-beam epitaxy at the normal growth temperatures of 610 and 700 °C, respectively, and 50–100 °C below this temperature using varying small amounts of Sb as a dopant. Electrical properties of the films were then examined using Hall mobility measurements and deep-level transient spectroscopy. Isoelectronic Sb doping of GaAs and AlGaAs allows the growth temperature for these materials to be reduced by up to 80 °C with trap densities and Hall mobilities at least comparable or better than those for films grown at optimal temperatures. It should therefore be possible to grow GaAs and AlGaAs heterostructures at one temperature without sacrificing the electrical properties of the individual materials.

A detailed theoretical/experimental spin-polarized low energy electron diffraction (SPLEED) investigation of Cu(100) is under way to explore the utility of SPLEED measurements in structure determinations for surfaces of low atomic number. Preliminary SPLEED calculations show that, in certain energy ranges, the results are very sensitive to the assumed structural model. Good agreement between theory and experiment is obtained using the surface model, which incorporates multilayer relaxation, that was derived from analysis of earlier LEED intensity data. Work is continuing to refine this model by use of SPLEED data.

The results of an exploratory polarized LEED study of the order/disorder transition at a Cu 3 Au(100) surface are reported. The data provide no evidence of a sudden change in surface composition or local order at the bulk transition temperature, and are consistent with the results of earlier LEED and low energy ion scattering studies.

A simple UHV offset manipulator is described that not only allows a target crystal to be moved to any point on a circle centered on the manipulator axis but also provides indepedent θ and φ rotations at each position.

The use of laser annealing in the preparation of a clean, well-ordered Ni(001) surface has been investigated. The surface was cleaned by argon ion bombardment and subsequently irradiated with the fundamental output of a high power, Q-switched ruby laser. The laser treated surface was characterized by use of Auger electron spectroscopy and both conventional and spin-polarized LEED, and the data compared to that from a surface prepared by thermal annealing. Laser irradiation of a sputtered surface at energy densities of ∼0.8 J cm−2 leads to complete removal of residual argon without segregation of carbon or other contaminants at the surface. However, irradiation at an elevated temperature ∼200 °C is required to obtain a well-ordered surface. At elevated temperatures, laser annealing provides a surface with cleanness and surface order comparable to that obtained by conventional thermal annealing.

Combined measurements of both electron spin polarization and intensity are described for several LEED beams from Ni(001) and Ni(001)c(2 × 2)Te surfaces. Polarizations and intensities were determined in the energy ranges 20–110 eV and 25–200 eV, respectively. Measurements on the specular 00 beam were undertaken at several angles of incidence in the range 10.2–18° for azimuthal angles φ = 0° and φ = 45°. The 01 beam, and the 12, 12 beam from the Ni-Te surface, were studied at normal incidence. Sizable polarization features, ¦P¦ ≲ 25%, were observed, even for Ni(001), indicating that polarized LEED measurements may be used to study surfaces of low atomic number. Pronounced changes in the polarization features were evident following deposition of a c(2 × 2) tellurium adlayer. The experimental data are in general agreement with the results of initial calculations undertaken using a relativistic multiple-scattering theory.