1
Electron Spectroscopy: Some Basic Concepts

1.1 Analysis of Surfaces

All solid materials interact with their surroundings through their surfaces. The physical and chemical composition of these surfaces determine the nature of the interactions. Their surface chemistry will influence such factors as corrosion rates, catalytic activity, adhesive properties, wettability, contact potential, failure mechanisms, etc. Surfaces, therefore, influence many crucially important properties of the solid.

Despite the undoubted importance of surfaces, only a very small proportion of the atoms of most solids are found at the surface. Consider, for example, a 1 cm cube of a typical transition metal (e.g. nickel). The cube contains about 9 × 1022 atoms of which about 6 × 1015 are at the surface. The proportion of surface atoms is therefore approximately 1 in 108 or 10 ppb. If we want to detect impurities at the nickel surface at a concentration of 0.1% then we need to detect materials at a concentration level of 0.01 ppb within the cube. The exact proportion of atoms at the surface will depend upon the shape and surface roughness of the material as well as its composition. The above figures simply illustrate that a successful technique for analysing surfaces must have at least two characteristics:

1. It must be extremely sensitive.
2. It must be efficient at filtering out signal from the vast majority of the atoms present in the sample.

This book is largely concerned with X-ray photoelectron spectroscopy (XPS) and Auger electron spectroscopy (AES). As will be shown, each of these techniques has the required characteristics but, in addition, they can answer other important questions:

1. Which elements are present in the near-surface region of a solid?
2. Which chemical states of these elements are present?
3. How much of each chemical state of each element is present?
4. What is the spatial distribution of the materials in the near surface region in three-dimensions and how does that vary with time?
5. If material is present as a thin film at the surface:
   1. How thick is the film?
   2. How uniform is the thickness?
   3. How uniform is the chemical composition of the film?

In electron spectroscopy, we are concerned with the emission and energy analysis of low energy electrons, usually in the range 20-2000 eV1 (the use of X-ray anodes that generate X-rays having a photon energy much higher than 2000 eV is becoming more popular). These electrons are liberated from the sample being examined as a result of the photoemission process (in XPS) or the radiationless de-excitation of an ionised atom by the Auger emission process in AES and scanning Auger microscopy (SAM). The distinction between AES and SAM is worthy of consideration. AES is a broad term that implies excitation of Auger electrons using a beam of electrons but makes no claim to be a technique that features high spatial resolution. SAM, on the other hand, always makes use of a finely focussed electron beam, typically in the range 10-100 nm, and provides results in the form of spatially resolved images derived from Auger electron data.

In the simplest terms, an electron spectrometer consists of the sample under investigation, a source of primary radiation, and an electron energy analyser all contained within a vacuum chamber, preferably operating in the ultra-high vacuum (UHV) regime. In practice, there will often be a secondary UHV chamber fitted with various sample preparation facilities and perhaps ancillary analytical facilities. A data system will be used for data acquisition and subsequent processing. The source of the primary radiation for the two methods is different; XPS making use of soft X-rays, most commonly monochromated Al Ka X-rays, although a twin anode arrangement is still often used (the most popular being Al Ka combined with Mg Ka), whereas AES and SAM rely on the use of an electron gun. The specification for electron guns used in Auger analysis varies tremendously, particularly as far as the spatial resolution is concerned, which, for finely focused guns, may be between 5 µm and < 10 nm. In principle, the same energy analyser may be used for both XPS and AES; consequently, the two techniques are often to be found in the same analytical instrument.

Before considering the uses and applications of the two methods, it is helpful to review the physical principles of the two processes along with their strengths and weaknesses.

1.2 Notation

XPS and AES measure the energy of electrons emitted from a material. It is necessary therefore to have some formal way to describe which electrons are involved with each of the observed transitions. The notation used in XPS is different from that used in AES. XPS uses the so-called spectroscopists' or chemists' notation while Auger electrons are identified by the X-ray notation.

1.2.1 Spectroscopists' Notation

In this notation, the photoelectrons observed are described by means of their quantum numbers. Transitions are usually labelled according to the scheme nlj.

The first part of this notation is the principal quantum number, n. This takes integer values of 1, 2, 3, etc. The second part of the nomenclature, l, is the quantum number which describes the orbital angular momentum of the electron. This takes integer values 0, 1, 2, 3, etc. However, this quantum number is usually given a letter rather than a number as shown in Table 1.1.

Table 1.1 Notation given to the quantum numbers which describe orbital angular momentum.

The peaks in XPS spectra derived from orbitals whose angular momentum quantum number is greater than 0 are usually split into two. This is a result of the interaction of the electron angular momentum due to its spin with its orbital angular momentum. Each electron has a quantum number associated with its spin angular momentum2, s. The value of s can be either +½ or -½. The two angular momenta are added vectorially to produce the quantity j in the expression nlj, i.e. j = |l + s|. Thus, an electron from a p orbital can have a j value of ½ (l-s) or 3/2 (l + s); similarly, electrons from a d orbital can have j values of either 3/2 or 5/2. The relative intensity of the components of the doublets formed by the spin orbit coupling is dependent upon their relative populations (degeneracies) which are given by the expression (2j + 1) so, for an electron from a d orbital, the relative intensities of the 3/2 and 5/2 peaks are 2 : 3. The spacing between the components of the doublets depends upon the strength of the spin orbit coupling. For a given value of both n and l, the separation increases with the atomic number of the atom. For a given atom, it decreases both with increasing n and with increasing l.

1.2.2 X-ray Notation

In X-ray notation, the principal quantum numbers are given letters K, L, M, etc. while subscript numbers refer to the j values described above. The relationship between the notations is given in Table 1.2.

Table 1.2 The relationship between quantum numbers, spectroscopists' notation and X-ray notation.