The Basics of CAICISS

 

CAICISS is a novel ion scattering technique that enables true surface specificity. Due to the physics employed in low energy ion scattering, the chemical and structural information is limited to the first few of atomic layers and can be derived in a layer-by-layer fashion. CAICISS therefore has the potential to follow chemical and structural changes in real time.

For full details on the theory of CAICISS, see Chapter 3 of Marc Walker's PhD thesis.

 

The Equipment

 

For the co-axial impact collision ion scattering spectroscopy system built a Warwick, low energy ions are produced using a Nielson ion source, extracted and focused, using an Einzel lens, along a flight path to the main chamber. The flight path includes in-order away from the ion source, a collimating aperture, a set of vertical and horizontal electrostatic deflection plates for electrostatic steering and chopping of the beam, an off-axis port aligner and a chopping aperture. The off-axis port aligner is utilised to induce a bend of 2 degrees into the flight path to ensure that neutrals, produced by the ion source, fail to pass through the final aperture. Chopping of the beam is achieved using a fast rise time pulse applied to one set of the deflection plates in order to sweep the beam across the final aperture. The resulting short pulses (~ 50 ns) then pass through the centre of a micro-channel plate detector and down the remaining section of flight tube toward the sample. Ions and neutrals that are scattered back off the sample at angles close to 180° are detected in time-of-flight mode using the micro-channel plate detector.

 

 

The Physics of CAICISS

 

One of the attractions of CAICISS is the relative simplicity of the physics involved. The ideal scattering process can be described accurately by a binary collision model [1, 2], which conserves energy and momentum, as shown in figure 1, and leads to an equation describing the scattering process [1]:

A = M₂/M₁........ M₁ = mass of incident ion......... M₂ = mass of target atom........ E₀ = kinetic energy of incident ion........ E₁ = kinetic energy of scattered ion........ θ₁ = scattering angle with respect to incident trajectory.

A derivation of the kinematic factor equation is given here.

 

FIGURE 1 - The binary collision model.

 

From the equation above it can be seen that if the incident ions have a constant mass and energy and the kinetic energy of scattered ions can be measured, then the mass of the scattering atom can be deduced. Considering the equation further, the advantages of scattering through 180º during data analysis can be seen immediately.

If quantitative information is required (e.g. relative concentrations of species), species-dependent effects such as ion neutralisation and scattering cross-section must be considered [2]. Indeed, the scattering cross-section is dependent not only on the scattering species but also on the ion-atom interaction potential [1, 2]. This potential can be described in two ways – the Molière approximation to the Thomas-Fermi model [3, 4], or the Ziegler-Biersack-Littmark (ZBL) function [5]. From this the relationship of scattering angle θ₁ and impact parameter can be computed, giving a set of ion trajectories of the form shown in figure 2. Note the existence of a shadow cone behind the scattering atom, into which no incident ions can penetrate. If another atom lies within the shadow cone, it is sheltered from the incident flux and hence cannot contribute to the scattered signal. For low energy ions the width of the shadow cone is typically a large fraction of the interatomic spacing in the surface region. Figure 3 shows the effects of the shadow cone over the top three atomic layers. The shadowing caused by surface atoms blocks out the majority of the second layer and all of any deeper layers, and hence the scattered signal is almost exclusively of surface origin [1]. This feature makes CAICISS a good option for studying the structure and composition of solid surfaces. In more typical CAICISS experiments [6, 7, 8], 2-3 keV incident ions are used which results in narrower shadow cones and hence contributions to the scattered signal from the top two or three atomic layers. This is still an excellent degree of surface specificity compared to electron spectroscopy techniques which expose several atomic layers [1].

To investigate surface structure with CAICISS, the orientation of the crystal with respect to the incident ion beam is changed during the experiment. This change can either be to the angle of incidence (polar angle) or the orientation of the crystal (azimuthal angle). As the angle is varied atoms will move into and out of the shadow cones, giving a backscattered signal which has an angular dependence. An example of a polar angle scan is shown in figure 4.

The polar scans show several features of importance in determining the structural and chemical makeup of the crystal. The scan begins with the angle of incidence set to 0º, such that the beam direction is parallel to the surface and each surface atom is shadowed by the previous atom in the atomic row. As the angle of incidence increases, the surface atoms begin to emerge from the shadow cones until the edge of the cone from one atom is directly focussed on the next atom in the chain, producing a large increase in the backscattered ion yield due to trajectory focussing [9]. This is known as the surface peak, with the relevant angle of incidence known as the critical angle (geometry shown in figure 5). The critical angle can be related to the interatomic spacing of the atoms on the surface, which is vitally important for characterisation of the surface.

As the angle of incidence is increased further, atoms in the second layer begin to emerge from the shadow cones. Again, when the edge of a shadow cone from a surface atom is directly focussed on a second layer atom an increase in backscattered yield is seen. The angle at which this occurs can be used to determine the interlayer spacing between the surface and the second layer. This can be used to look for surface relaxations, in which the topmost interlayer spacing may change as a result of, for example, heating or adsorption. Similar effects are seen as the shadow cones from the surface and second layer atoms are focussed onto atoms in the third layer. Experimental data is then simulated using packages such as the FAN code developed by H. Niehus to determine the surface structure of the sample. Fits of angular peak positions and peak intensities are used to determine the likely structure and elemental composition of each layer in the surface region. As already discussed, very little contribution is seen from deeper layers. To summarize, the polar scans not only offer information on the chemical makeup of the surface region, but also interatomic and interlayer spacing information which assist in the characterisation of the surface region.

The azimuthal scans show blocking dips along low-index axes in the crystal, helping to characterise the structure of the surface and distinguish any reconstructions which may have occurred. Azimuthal scans are generally performed at grazing incidence to minimise contributions from layers below the surface.

In conclusion we see that CAICISS is an excellent tool for surface-specific studies of solids. It is almost unique in its surface specificity has been used in several different types of investigations [6, 7, 8].

 

FIGURE 2 - A schematic of a shadow cone resulting from the Coulombic repulsion between the incident ion (red lines) and target atoms.

 

FIGURE 3 - At certain angles and energies the shadow cones prevent incident ions reaching sub-surface atoms.

 

FIGURE 4 - A polar angle plot for 3 keV helium ions backscattered from a platinum-covered Cu(100) surface.

 

FIGURE 5 - An illustration of the critical angle and how it can be used to derive shadow cone radius or interatomic spacing.

 

References

1. D.P. Woodruff and T.A. Delchar, Modern Techniques of Surface Science –Second Edition, Cambridge University Press, Cambridge, UK, 1994
2. D. Briggs and M.P. Seah, Practical Surface Analysis, 2nd Edition, Vol. 2: Ion and Neutral Spectroscopy, Wiley, Cirecester , UK , 1992
3. G. Molière, Z. Naturf. 2a (1974), 133
4. S.A. Cruz et al, Nucl. Instru. Meth. 194 (1982), 659
5. J.F. Ziegler et al, The Stopping Range of Ions in Solids, Vol 1, Pergamon Press, New York (1985)
6. M Aono et al, Nucl. Instr. and Meth. B 64 (1992), 29
7. T. Choso et al, Appl. Surf. Sci. 121/122 (1997), 387
8. M Katayama et al, Nucl. Instr. and Meth. B 99 (1995), 598
9. H. Niehus et al, Surf. Sci. Rep. 17 (1993), 213

 

 

Contact Details

•