Knowledge Base

ARXPS is especially useful as it provides non-destructive information about the thickness and composition of very thin films which are not suitable for sputter depth profiling and also materials (e.g. polymers) which may be irreversibly damaged by argon ions.

An excellent introduction to ARXPS is given by Royston Paynter [1] together with a Mathematica Workbook in the Supplementary Information.

ARXPS can give information therefore on :

• Ordering of layers in an ultra-thin film [3]
• Thickness of layers [1-3]
• Distribution of elements and chemical states within the film (depth profile reconstruction) [3]

Figure 1. Simple illustration of an ARXPS measurement of a silicon oxide layer on a silicon wafer. Note how at the more grazing angle, the top most oxide (and the sub-oxides) are more prominant as we are now more surface sensitive

ARXPS measurements are realtively simple to perform. XPS data is collected at a series of angles, typically from 0° to 75° photoemission angle. Five or six angles are usually sufficient and Cumpson [2] has recommended acquiring data at 0°, 40°, 55°, 63° and 70° for materials such as polymers which exhibit weak elastic scattering, and 0°, 33°, 45°, 54°, 60° for samples containing elements of larger atomic number (e.g. an iron alloy), where elastic scattering is stronger. Performing measurements at these angles concentrates the data acquisition at the higher photoemission angles, where the analysis is most sensitive to the sample surface.

The principal behind why we see a change in the surface sensitivity arises from the change in effective attenuation length when we revolve a sample around it’s axis.

If we think of our general equation for determining the attenuation length of an electron within a sample, we have a parameter which determines the amount of radiation which is reduced as a result of passing Δx thickness of a sample.

μ = lim (Δx → 0) (1 – (I / I₀) / Δx

Where μ = the attenuation coefficient

I = intensity of radiation following transmission

I₀ = intensity of incident radiation

Attenuation length is then determined from the reciprocal of this.

λ = 1/μ

And gives us a general equation for attenuation (of radiation OR electrons)

I/I₀ = exp(-Δx/λ)

When we revolve our sample, we are increasing the effective path length from the point of photoelectron emission – to the material surface, and hence our Δx value – so we will be reducing the intensity of our I for those electrons that originate deeper within the material. The end result of this is that our eventual spectrum is more representative of the atoms from closer to the surface.

ARXPS involves analysing data from a sample at different angles, and identifying the changes that arise as a result of this change in surface sensitivity – outlined in the previous section.

Controlling the angle of analysis at a number of points permits ‘snapshots’ of the chemical environments up to a particular depth. For thin films (< ~9 nm for Al Kα X-rays), it is possible to observe the changing chemistry as you travel through the Z-dimension of the film.

By monitoring the changing intensity of a substrate peak as a function of analysis angle, provided the inelastic mean free path / attenuation depth is known – it is possible to calculate the thickness of a film or overlayer on a flat surface using ARXPS.

Is = Is0.exp(-d/λs(o).cos(θ)

Where:

Is = Intensity of the substrate signal following attenuation through the overlayer

Is0 = Intensity of the substrate signal before attenuation through the overlayer

d = film thickness

λs(o) = IMFP of substrate electrons, through overlayer material

θ = Take off angle

Is0 is unknown, but with multiple analysis angles, you may determine this either through graphical means, or using simultaneous equations to cancel down.