G.H. Wheatley

Energy spectra of 6–32 keV neutral and ionized Ar and He scattered from Au targets; ionized fractions as functions of energy

Abstract The neutralization of ions is an important aspect of low energy ion scattering for surface analysis. Electrostatic energy analyzers (ESA) have been used almost exclusively in such work, and information on charge neutralization efficiencies is needed for quantitative interpretation of ESA data. In the past, the occurrence in low energy ion spectra of surface peaks and low backgrounds due to scattering from inside the solid has been attributed to preferential neutralization of ions which penetrate beyond the surface. In the work to be described, a time-of-flight technique was used to measure energy spectra of both neutral and ionized Ar and He scattered at 90° from a polycrystalline gold target. Incident energies of 6–32 keV were used. The energy spectra of neutral Ar scattered from polycrystalline gold exhibit sharp surface peaks, and double scattering shoulders, over this entire energy range. For He there is a gradual downward slope toward lower energy rather than a sharp surface peak. The behavior in both cases is attributed to large scattering cross-sections which cause a loss of beam particles during penetration. A calculation using a 1 r 2 potential illustrates this effect as a function of energy for helium. In the present experiments we find that the ion fraction of scattered argon does indeed depend on depth of penetration. This is in contrast to the behavior of He and H at higher energies, e.g. 100 keV, in which cases the charge state depends on emergent velocity but not on depth of penetration. The characteristic shapes of ion scattering spectra in this energy range appear to result from both neutralization and beam attenuation inside the target.

Low-energy neon-ion scattering and neutralization on first and second layers of a Ni(001) surface

Abstract The scattering and neutralization of 2.4 and 5 keV Ne+ ions on the Ni(001) surface have been studied by time-of-flight (TOF) and electrostatic analyzer (ESA) techniques. The scattering yield of neutrals plus ions (by TOF) is strongly dependent on crystal orientation, in one direction being reduced by the shadowing of 2nd layer atoms by 1st layer atoms, or in another being increased by focussing of ions onto the 2nd layer by 1st layer atoms. Ion yields (by ESA) show little of this variation since the ions are largely neutralized on scattering from the second layer. The results thus demonstrate and explain the first layer selectivity of low-energy ion scattering by ESA for a case in which there is no shadowing of second layer atoms by the first layer. On the other hand, the ability to measure and distinguish first and second layer scattering of neutrals and ions by TOF suggests the possibility of composition analysis of individual layers of single crystal alloys and compound semiconductors.

Investigation of low-energy ion scattering as a surface analytical technique

Abstract The energy spectra of rare gas ions backscattered from various solids have been monitored for primary energies in the range 40 keV to 500 eV. At the higher primary energies peaks in the backscattered energy spectrum, which correspond to the masses of atoms on the sample surface, tend to broaden considerably towards lower energies. As the primary energy is reduced, these peaks become sharp, and below 5 keV primary energy the technique is suitable for elemental surface analysis of solids, including polycrystals and amorphous materials. The sensitivity of the apparatus to trace impurities has been estimated by examination of precalibrated test samples, and is found to be ∼ 5 × 10 −4 monolayers for a heavy element such as gold, and by extrapolation ∼ 10 −1 to 10 −2 monolayer for oxygen. Sputtering by low energy helium ion beams is found not to be a serious problem during analysis. Argon ion beams, on the other hand, may be used to study depth profiles of composition in alloys. This report is intended to bring to light some of the capabilities and present limitations of low energy ion scattering as a surface analytical technique.

Energy and mass spectra of neutral and charged particles scattered and desorbed from gold surfaces

Surface analysis by low energy (~0.1–10 keV) ion scattering has, in the past, been performed primarily by means of electrostatic energy analysis. This technique though very useful, ignores the large neutral fraction of the scattered particles. In this paper a low energy time-of-flight (TOF) spectrometer is described, capable of providing TOF spectra of both charged ions and neutrals. This system has been used to obtain energy spectra of 8 keV Ar+ ions scattered off polycrystalline gold targets. Unexpected features of these spectra were shown to be due to the sputtering of hydrogen, and other species, off the gold surface and into the TOF channel. The mass to charge ratios of these atomic species were determined by measuring the flight times of ions through an electrostatic analyzer. An alternative technique for TOF analysis of beam-desorbed neutrals, as well as ions, is also proposed.

Charge states of 25–150 keV H and 4He backscattered from solid surfaces

Abstract Measurements have been made of the ion-fractions of H and 4 He backscattered with energies of 25–160 keV from Cu, Au, and Si surfaces which were etched and washed but not atomically clean. The ion-fractions for H range from 0.37 at 25keV to 0.92 at 160 keV, and for 4 He from 0.10 at 30keV to 0.58 at 150 keV, depending to a small extent on the target material. Where comparisons can be made the data agree rather closely with results of others for particles traversing thin foils. The data are useful for calibration of an electrostatic analyzer in surface analysis. Plots of ion-fraction against particle velocity show a primary dependence on velocity, as expected, but there is a small difference in slope between the H and He curves. Charge states of particles scattered from surface impurities did not deviate significantly from those of particles scattered from the substrate at the same energy.

Comparison of a time-of-flight system with an electrostatic analyzer in low-energy ion scattering

Abstract Neon ions with primary energy of 5 keV were scattered from gold and silicon and energy spectra were obtained by both time-of-flight (TOF) and electrostatic analyzer (ESA) methods. At this energy the neutral plus ion spectra obtained by the TOF method are not so sharply peaked as the ion spectra of the ESA, owing to neutralization effects on natural linewidth and also to somewhat poorer energy resolution. However, the ion dose required for the TOF technique was substantially lower and a significant reduction in surface disorder on silicon was detected by Leed. Neon ions were neutralized more efficiently on a gold surface than on silicon. Evidence of sputtered Si + ions having relatively high energy, e.g. 800 eV, was found by a combination ESA-TOF measurement. The Si + ions are evidently knocked out of the first atom layer by Ne reflected from deeper layers.

Scattering of low energy Ne+ ions on Ni(001) and Ni(001)Au (segregated) surfaces

Abstract Low energy (2.4–9.5 keV) ion scattering of Ne from Ni(001)Au (segregated) surfaces on a Ni-1% Au single crystal has been studied using a time-of-flight (TOF) system which collects both neutrals and ions thereby revealing, at certain crystal orientations, the scattered yield from second layer as well as first layer atoms. Good agreement was found between experimental and computer simulated spectra for both the clean Ni and Au-segregated surfaces. In the latter case the overlayer containing ∼90% Au and 10% Ni was assumed to have the same FCC(001) structure as the clean Ni. LEED patterns from the segregated surface, however, show extra spots indexed approximately as (2 × 6). Azimuthal scans of the Ni scattering yield coming predominantly from the second layer in the segregated case show similar symmetry, i.e. the FCC(001) (1 × 1) symmetry, for both clean Ni and Au-segregated surfaces. However, the gold overlayer is evidently not strictly substitutional, in view of the LEED patterns. No evidence was found of Au in the second layer.

Studies of solid surfaces with 100 keV 4He+ and H+ ion beams

Abstract Backscattering of 4He+ and H+ ions incident at 100 keV was used to detect foreign atoms on silicon and graphite surfaces, and also surface disorder on silicon. The ion beams were supplied by an ion-implantation accelerator. Energy spectra of the backscattered ions were measured by an electrostatic analyzer (ESA) which had 3 % energy resolution. Foreign atom sensitivity of this system, expressed as the atom density required to yield 10 counts, is 1.3 × 1012/cm2 for gold, 9 × 1012/cm2 for iron, and 3 × 1014/cm2 for oxygen, at an ion dose of 0.8 microC per point. Use of an 8 microC dose lowers these limits an order of magnitude at the expense of greater surface damage and longer measurement time. The effective scattering yield as measured in the ESA departs somewhat from the Rutherford Z2 dependence because of ion neutralization which depends on ion energy which in turn depends on the mass of the target atom. Ion channeling was used to supress scattering from silicon atoms beneath the surface which tends to obscure the peaks for light elements such as oxygen and carbon. The channeling phenomenon was also used to study surface disorder due to abrasion damage and damage done by the ion beam itself. Sputtering of metal films by the 100 keV 4He+ beam was negligible. However, iodine was removed from silicon surfaces at a detectable rate. This technique at 100 keV, in common with backscattering at higher energies, does not require UHV conditions. It can therefore be used in a simple vacuum system to analyze surfaces of technological interest which might be altered by the heating or the heavy ion bombardment which are required to remove adsorbed background gas layers in several other surface analytical techniques.

Observations of electron reflection at tungsten (001) surface in the energy range 1–10 eV

Abstract Measurements of electron reflection at tungsten (001) surface are reported. The measurements were made with a display-type LEED apparatus operated in the retarding mode. Results are given for clean tungsten and for surfaces formed by adsorption of gases on tungsten. The variation of diffraction intensity (arbitrary units) as a function of electron energy in the range 1–10 eV is described for the 00 and 01 beams. Measurements on clean tungsten are reported for a sequence of values of the angle of incidence ψ0, for each of two azimuthal orientations of the primary beam (0 and 45° referred to the 01 direction in the crystal surface). The measurements on the 00 beam fill in the gap between those of Khan, Hobson and Armstrong and of Zollweg for normal incidence (ψ0 = 0) and those of Propst and Edwards (ψ0 = 53°) and are correlated with these earlier measurements. The chief observations are: A — a narrow peak near 4 eV which is insensitive to both variation of primary-beam orientation and gas adsorption; B — a very narrow dip-peak combination whose location can be correlated with the 01 beam threshold (grazing-emergence) energy and whose location and shape are highly sensitive to gas adsorption. The observations support suggestions that A is connected with an inelastic-scattering threshold and that B is a surface-state resonance.

ATOMIC POSITIONS OF AU ATOMS ON A NI(110) SURFACE

Abstract The arrangement of Au atoms segregated to the (110) surface of a Ni-0.8at% Au alloy was studied with low energy ion scattering (LEIS) in combination with low energy electron diffraction (LEED). The results of the LEED analysis have been published earlier [1]. From this analysis it is known that at the very low coverage of around 0.1 monolayer (ML) a 1 × 1 symmetry exists. At high coverages ( ∼ 1 ML) a 7 × 7 overlayer structure with c(2 × 4) subunits was observed. At intermediate coverages also other symmetries were observed such as (5 × 1) and c(2 × 2). By comparing angular LEIS scans with the results of computer simulations, we were able to determine part of the Au configurations giving rise to these observed symmetries. At low coverage the Au atoms were found to occupy near-substitutional sites in the Ni top layer. No Au atoms were observed on top of the Ni surface, nor in the second Ni layer. The 7 × 7 reconstructed surface is formed by a hexagonal, incommensurate, Au overlayer on top of the Ni(110) structure. A complete 7 × 7 structure would give rise to a coverage of 1.28 ML of Au atoms.