Search

The most sensitive desktop-sized ionizer utilising electron bombardment is currently the solenoidal ion source. We present an alternate design for such an ion source whereby the solenoidal windings of the electromagnet are replaced by a shaped cylindrical permanent magnet in order to reduce the complexity and running costs of the instrument. Through finite element modelling of the magnetic field in COMSOL and experimental measurements on a small-scale prototype magnet stack, we demonstrate the required shape of the permanent magnet in order to generate the needed field, and the necessity of soft iron collars to smooth fluctuations along the central axis.

We present an analytic model applied to quasi-elastic scattering from an adsorbed surface species undergoing jump diffusion between adsorption sites described by a Bravais lattice combined with a basis of multiple points. The model allows for hops between adsorption sites which are both symmetrically and energetically inequivalent. We give results for 1-D hopping, which are applicable to a species jumping between the top and bridge sites along the [110] direction on an fcc-(110) surface or for jumps along a step edge. In 2-D, results for hopping between fcc and hcp hollow sites and between the bridge sites of an fcc-(111) surface are presented. These examples give characteristic signatures which will allow these forms of motion to be recognized in experimental data and will enable the underlying physical parameters to be extracted by comparison with the analytical forms derived here.

A classical molecular dynamics study of the diffusive motion of benzene molecules on graphite basal planes has been performed in the sub-monolayer coverage regime. Atomistic calculations were performed using the second generation forcefield COMPASS as well as the general purpose forcefields Dreiding and Universal. The COMPASS based calculations show evidence for a Brownian nature of the diffusion with a very small diffusion activation barrier of 11 ± 2 meV in agreement with recent helium and neutron spin-echo spectroscopy data [Fouquet P, Hedgeland H, Jardine AP, Alexandrowicz G, Allison W, Ellis J. Measurements of molecule diffusion on surfaces using neutron and helium spin echo. Physica B 2006;385–386:269-71]. Reasonable agreement is also found for the general purpose forcefields if screened charges are used in the description of the Coulombic non-bond interaction. The less computationally intensive Dreiding forcefield was shown to give a good qualitative description of the diffusion validating its applicability for future large scale calculation, i.e., for long times or enlarged systems. A potential energy surface (PES) has been established for the translational and rotational diffusive motion. The PES shows a peculiar dependence of the lateral diffusion barriers on the rotation angle of the benzene molecule which leads to a preferential selection of certain rotational states in the MD trajectories.

AFM, STM and diffraction of He and H2 have been used to assess Si(1 1 1)–H(1 × 1) surfaces for their potential as mirrors for matter-waves. The H-passivated samples are produced by wet-chemical methods and delivered to a different laboratory for diffraction measurements. We show that the surface is flat and homogenous over lateral scales of microns and that absolute He and H2 reflectivities of the order of ∼3% are obtained, even after 20 h storage under Ar and several days’ storage in UHV. These characteristics allow the use of Si(1 1 1)–H(1 × 1) as a highly reflective mirror for atoms and molecules, with application in a future He microscope or focused hydrogen nano-lithography system.

Medium energy ion scattering has been used to probe the early stages of the development of a surface alloy of platinum deposited on Cu(1 1 1). Our experiments reveal the presence of a novel two-layer “precursor” Pt–Cu alloy at low temperatures prior to formation of the expected Cu3Pt surface alloy at near 550 K. Comparison of the alloying behaviour of one monolayer of platinum on a nominally flat copper surface with that on a vicinal copper surface, Cu(12 12 11), demonstrates that the formation of this precursor alloy is mediated by steps on the surface. A model is proposed for the mechanism of formation of this precursor alloy, which occurs at much lower temperatures than previously observed for surface alloying of the platinum–copper system.

Isothermal measurements of the helium specular reflectivity from H–Cu(1 1 1) are used to demonstrate that adsorbed hydrogen is removed from the surface in an activated process according to first-order kinetics. The measurements were taken at low coverages with surface temperatures in the range of 210–245 K, where we derive an activation-energy of 64.3 ± 1.3 kJ mol−1. The observation allows us to rule out second-order, molecular desorption processes and supports the conclusion that the H atoms are transported from surface to sub-surface sites by an activated process. The results confirm an earlier suggestion that, in the low coverage limit, the sub-surface state for this system is more energetically favourable than other states. In addition, the measurements suggest abstraction processes take place with a probability of (4.3 ± 1.0)s0, where s0 is the initial sticking probability.

A detailed comparative study of Cu growth on the vicinal surface Pt(12 12 11) has been performed with helium atom scattering (HAS) and medium energy ion scattering (MEIS). These techniques are ideally complementary in that HAS is extremely sensitive to the electronic structure of the surface layer, while MEIS accurately probes the atom core structure of the surface. By combining these two techniques it has been possible to measure the initial structure of Cu on the stepped Pt surface and to study how the Cu redistributes across the surface and into the bulk on annealing. The HAS and MEIS data identify two significant diffusion transitions on the surface occurring at temperatures of 450 and 600 K.

We demonstrate sub-surface absorption of hydrogen on Cu(1 1 1) to be energetically favoured, even in the limit of zero coverage and without surface reconstruction. Helium atom scattering measurements indicate desorption of hydrogen from the surface at a lower temperature, than that required to remove all absorbed hydrogen. The stablest bonding site, which is invisible to the helium beam, is therefore not a surface site. Our experimental observations of the behaviour of Cu(1 1 1) differ from previous calculations and contrast with other systems, where population of subsurface sites generally occurs only after population of surface states, or after significant surface reconstruction.

We present a quantitative analysis on the electron energy loss spectra (EELS) of potassium (K) doped C 60 thin films collected as a function of time to explore the diffusion of K atoms at and near the surface at 100 K. The spectral features of K 3 C 60 , K 4 C 60 , K 6 C 60 and pure C 60 phases were observed to coexist before reaching saturation doping. The ratio of these phases can be controlled by the K deposition rate. At low deposition rate, we found no trace of K 3 C 60 at the surface. A quantitative analysis, based on fits using Lorentzian functions characteristic of each structural phase, is used to monitor the time and dose variation of the multiphase-film. The analysis gives the dissipation rates of K 4 C 60 and K 6 C 60 phases at the surface together with the peak broadening of the LUMO–LUMO+1 transitions of K 4 C 60 and K 6 C 60 structures as a function of grain size. The results give insights into the relative kinetic stability of K x C 60 grains, as x is varied.

We present atomic force microscopy (AFM) measurements from a passivated silicon crystal miscut by 0.1° and show the etching regime to be significantly different from surfaces with a larger miscut angle. A simple kinetic model is developed to explain the results and is used to derive the optimal etching conditions for nominally flat Si(1 1 1)–(1×1)H. We show that small changes in miscut angle can alter the kinetic steady state and promote the formation of deep etch pits, even on the least stable, 〈 1 1 2〉 miscut surface. Collisions of steps with these pits result in arrays of stable, self-aligned `etch hillocks' over micron dimensions. Following preparation, we use AFM to observe the initial growth of native oxide on the Si(1 1 1)–(1×1)H surface, and demonstrate that AFM is a sensitive probe to surface oxidation in the sub-monolayer regime.

We present a quantitative analysis of electron energy loss (EEL) spectra to investigate the initial growth of C 60 thin films on a graphite substrate at room temperature. The approach is to fit the experimental data to a parameterized simulation, which includes both a growth model and the treatment of electron scattering. A rate equation approach is used to characterize the growth by a single parameter, the probability of a molecule descending to the preceding layer when approaching an island edge. The scattering dynamics is described by a layer-dependent cross section and a constant mean free path. The results indicate that the initial growth of C 60 thin films on graphite is predominantly two-dimensional with a high degree of interlayer transport. Specifically, the simulation gives a probability of 0.88±0.10 for a C 60 molecule to cross an overlayer descending step.

A systematic procedure is described, which provides quantitative, model independent information from specular lattice-rod scans in helium–surface scattering. Interference effects between adjacent low-index terraces are used to give a precise measure of interlayer spacings. The method is illustrated with data from experiments on the heteroepitaxial growth of Co on Cu(111). The precision and the statistical significance of the procedure are discussed. In favourable circumstances, interlayer spacings can be obtained to a precision of better than 0.02 A.

We investigate the reflectivity of Si(111)-(1×1) H for helium atoms with particular reference to atom optical applications, which require a high specular reflectivity. Quantum close-coupling calculations using an empirical potential are used to identify the optimum scattering conditions. We show that a diffraction probability >0.5 in the specular channel can be obtained in a kinematic window near 20 meV with near-normal incidence (polar angle

The kinetic processes taking place during the adsorption of propyltrimethoxysilane (PTMS) onto polycrystalline iron and aluminium oxide surfaces have been investigated using X-ray photoelectron spectroscopy. Although the adsorption of PTMS on iron reaches equilibrium within a few seconds, the adsorption of PTMS on aluminium exhibits a dramatic time dependence. The presence of a single oscillation in the adsorbate coverage as a function of time indicates the existence of two distinct surface species, each of whose coverage varies dynamically, confirming the predictions of an earlier model.

The growth mechanism of a C 60 thin film on graphite substrate has been investigated by a quantitative analysis of electron energy-loss spectra, together with low-energy electron diffraction and thermal annealing measurements. It is found that the early stages of growth at room temperature are best described through two-dimensional island ripening in a fixed structure and predominantly by a layer-by-layer fashion.

We present a new approach to gain detailed surface-molecule potential information by combining neutron and helium atom scattering. First results are presented for a prototype study of benzene molecule diffusion on graphite surfaces. The measurements show good agreement and confirm the transferability of the results between the two methods. Future instrumental improvements are described that will improve the sensitivity of NSE by a factor of 50 for studies of molecules adsorbed on surfaces and mesoporous substrates.

The growth of ultra thin Cu films (up to 25 ML) on Pt(111) at a substrate temperature of 340 K has been investigated using helium atom scattering (HAS) and scanning tunneling microscopy (STM). In the first monolayer the growth is 2D-pseudomorphic with nucleation at step sites. No reconstruction of the substrate is observed. Diffusion limitation along the step edges leads to the formation of branched island structures which coalesce leaving randomly distributed holes (on average 30 A in diameter) as revealed by STM measurements.

The growth of K on Si(100), at 150 K, has been studied in situ by helium atom scattering. The transition from a corrugated, semi-conducting surface charge density to a smooth, metallic surface is clearly visible in the data. Below one monolayer we identify a single ordered overlayer at a coverage corresponding to 0.5 ML with respect to the Si atom density. A close packed metal film develops near 1 ML and reaches maximum order at 1.25 ML. The density of the metal film is consistent with a c(4 × 2) structure with a lattice mismatch of 3.5% in the [011] direction (overlayer expansion) and 2.3% in the [011] direction (overlayer compression) or vice versa. At higher coverages, the film appears to develop by island growth.

Measurements of the helium atom scattering reflectivity during growth, adsorption-desorption equilibrium and isothermal desorption of K on Cu(100) are reported. An orientational order-disorder phase transition is observed between 123 and 223 K in the submonolayer condensed phase. Adsorption-desorption isotherms have been obtained for a range of temperatures between 263 and 448 K. The data indicate a Stranski-Krastanov growth mode for this system, with the formation of 3D clusters after one monolayer, and give a heat of adsorption of about 1± 0.1 eV per atom near the completed first monolayer. The form of the isotherms suggests repulsive adsorbatc interactions in this coverage and temperature regime.

The growth and structure of potassium on Ge(100) has been studied at 300 K using scanning tunnelling microscopy. In the early stages of growth (approximately 0.1 ML), constant current topographs show small growth nuclei in two distinct surface sites: between and atop the Ge(100) dimers, respectively. Growth of both sites proceeds anisotropically with a preference for the 〈011〉 direction. Topographs obtained at higher coverages suggest the growth is locally more isotropic. Images showing regions near to surface steps display a higher local coverage on the upper terrace than on the lower, suggesting a preferential nucleation of growth at steps on the upper terrace. Near saturation coverage, the K atoms form clear one-dimensional chains along the 〈011〉 direction, similar to those on the K/Si(100) surface. The tunnelling characteristics are observed to change dramatically close to saturation coverage, at which point the surface develops a metallic character.

The interaction of silane with Cu(100) has been studied with high resolution helium atom scattering. Two stable structures are found to exist depending on the surface temperature during adsorption: (a) in the temperature range 173 to 300 K a structure exhibiting a weak p(1 × 1) helium diffraction pattern with a small corrugation is observed, probably consisting of SiHχ fragments, and (b) between 300 and 550 K a complex, but well ordered incommensurate overlayer is observed. The latter structure has a significantly larger corrugation than the low temperature phase. At temperatures between 600 and 650 K the surface undergoes a significant rearrangement that returns the corrugation, as seen by the helium atoms, to that of the clean surface. This change suggests either a complete desorption of the surface silicon, or its dissolution into the copper lattice.

Specular helium atom scattering has been used to characterise ultra-thin iron films grown epitaxially on a copper (100) substrate. Results are reported for films up to ten monolayers thick for three growth temperatures: 220, 298 and 420 K. In addition we report the results of subsequent thermal treatment of the films. The results suggest that, at low temperatures, growth is diffusion limited and proceeds by the formation of simulataneous multilayers. Near room temperature, growth is more organised but in the early stages, up to approximately three monolayers, films are less well ordered than when slighly thicker. After approximately seven monolayers the quality of the films grown at 298 K deteriorate, possibly as a result of the formation of misfit dislocations. Growth at 420 K proceeds in a layer-by-layer manner but Auger spectroscopy suggests that there is significant intermixing with the substrate at this temperature; that is, the film appeas to develop as an epitaxial surface alloy. None of the as-grown films are thermally stable. We present indirect evidence that the alloy film undergoes a spinodal decomposition at lo0w temperatures, leading to the precipitation of an enriched surface layer.

Specular helium atom scattering has been used to study the early stages of growth in epitaxial iron films on a copper (100) substrate. Results are reported for films, up to five monolayers thick, grown at 298 K. The data show that, in the case of helium scattering, the specular intensity can give a picture of the vertical growth morphology even in situations when it is not possible to separate the sharp, coherently scattered contribution to the specular peak from the broad, incoherently scattered contribution to the peak. We conclude that film growth begins with the formation of three-dimensional islands having a small mean size. As the film thickens, there is a trend towards layer-by-layer growth and an increased island size. The results enable us to role out earlier suggestions that the growth begins in a layer-by-layer fashion with a single or bi-layer structure.

The phonon spectra of both clean and cobalt covered Cu(100) have been studied with helium atom scattering, and the results compared with those previously obtained with EELS for nickel and iron overlayers on Cu(100) [1,2]. In the Γ-M direction on clean Cu(100), both Rayleigh and bulk band edge phonons were observed. In the Γ-X direction, surface Rayleigh modes with dispersion similar to that obtained in EELS [3] occurred together with a previously unobserved, higher energy mode. No change in phonon dispersion was found for cobalt covered Cu(100).

We show that existing theories which relate the helium-metal potential to the metallic surface charge density do not lead to a unique prediction for the surface charge density profile for a given experimental potential. A simple calculation of the corrugation of the attractive van der Waals potential for Cu(110) and Ag(110) shows that this is a contributary factor and that its inclusion improves the agreement between theories and also between experimental and calculated values for the charge density corrugation amplitude.

Helium atom diffraction intensities from ordered missing dimer structures on Si(001) surfaces indicate that, in stable (2 × n) n = 8, 11 configurations, the unit cell contains a pair of missing dimers rather than the single missing dimer predicted by theory.

Results are presented on the diffraction of 4 He atoms from the 1×2 reconstructed surface of Pt(110). The diffraction pattern observed indicates that the surface can be considered to have a one-dimensional corrugation. From the position of the rainbow maximum the corrugation amplitude is estimated to be ∼1.5 A. Structural models for the surface are used to derive corrugation funtions from the superposition of atomic charge densities. The results of hard wall scattering calculations from these large corrugations are compared with the experimental data. The results confirm that the large corrugation amplitude could arise from a displaced or missing row model. However, limitations in the present analytical theory prevent any definite distinction to be made concerning the detailed nature of the atomic displacements. These limitations are discussed in the light of the present results and previously published data.

High resolution helium diffraction has been used to investigate the adsorption of oxygen on a stepped W(001) surface. The initial adsorption proceeds via sites on the (001) terraces with adsorption on the surface steps preferred at higher coverages (θ>0.25). The transition from adsorption, above the surface plane, to incorporation, below the surface plane, occurs at a much lower temperature on the stepped surface than on unstepped W(001) surfaces. In the incorporated phase, the surface structure is characterized by a (2×1) LEED pattern and a (1×1) helium diffraction pattern.

An efficient method for the approximate calculation of atom diffraction intensities from a corrugated hard wall is presented. The calculations make use of Fourier transform techniques and are particularly advantageous for systems with large unit cells. Applications of the technique are given. These relate to helium atom scattering experiments from a vicinal tungsten surface with (001) terraces of approximately 30 A length separated by single atomic steps. Comparison of calculated results with the experimental peak profiles show that at a sample temperature of 200 K the variance of the terrace lengths is 56% of the mean length. The validity of the kinematic approxiamation for helium diffraction is considered.

We present electron energy loss analysis of the hydrogen induced commensurate-incommensurate displacive phase transition on the W(100) surface. Coverage dependent results indicate that the transition from the ( 2 × 2 ) to the (1×1) phase occurs via a tilted dimer model for the intermediate incommensurate phase. The combination of both lateral and vertical shafts of the substrate W atoms is indicative of mixing between two lattice vibrational modes: the Rayleigh branch with displacements normal to the surface plane, and a phonon branch with displacements parallel to the surface. The temperature dependent results clearly provide evidence for both states which play an important role in order-disorder (particularly) and first order (possibly) phase transitions.

Results of helium diffraction from Si (001)−(2 × n), n ≈ 10 are presented. The (2 × n) unit cells are shown to contain missing dimer defects alongside dimerized surface atoms. The experimental distribution of scattered intensity is compared with current models of the silicon dimer geometry and it is shown that all the main features in the observed spectra can be explained with a symmetric dimer configuration.

A general expression for the detector response to a given beam flux distribution is given. Illustrative examples are worked out for some simple idealized cases and it is shown that both the measured lineshape and the measured intensity depend on the details of incident beam and detector geometry.