EDX

The displacement of target material electrons by primary electrons triggers the release of an x-ray photon, which is characteristic of the atom (element) from which it was released. Atoms hold electrons in orbit around their nuclei as a result of electrical charge differences between them, the nucleus carrying a positive charge and the electrons carrying a negative charge. These charges exactly balance each other. The arrangement of electrons around the nucleus of an atom is conventionally considered as a sequence of 'electron shells', with the innermost shell electrons possessing the lowest energy but with higher bonding energies, and the outermost electrons possessing the highest energies with lower binding energies. The shells are, also conventionally, labeled K, L, M, N, O, P, and Q from innermost to outermost. Because of the differing binding energy levels of the electrons around the nucleus, primary electrons of a higher potential energy (electron volts - eV) are required to displace K-shell electrons than are required to displace L-shell electrons, which is higher than that required for M-shell electrons etc. A simple diagram of the first three shells is illustrated below-

Primary electrons bombarding the atom are able to 'knock' shell electrons out of their orbits, which can be replaced by electrons from shells further out, that are able to 'jump' into the inner shell vacancies. There are a number of reactions that occur as a result of this, but for the purposes of explaining the generation of x-rays, only the following points are considered. The knocking out of an inner shell electron causes the atom to take on an excited, high energy state until the missing electron is replaced and the atom relaxes. There is therefore a difference in energy states. This difference can be large, and if it is, the excess energy can be released in the form of an x-ray (it can also be released in the form of an Auger electron, but is not considered here), which carries this energy difference, and has a wavelength that is characteristic of the atomic species from which it came. The following diagram is an illustration of this-

It should be noted that replacement of L and M shell (etc.) electrons can also be achieved by electrons jumping in from shells that are further out again, and can lead to a large number of generated x-rays of differing wavelengths. There are therefore a number of possible lines of x-rays available for analysis, but again this is not considered here, and references should be made to a relevant textbook for further details.

The most significant issue to note from this is that the x-rays generated from any particular element are characteristic of that element, and as such, can be used to identify which elements are actually present under the electron probe. This is achieved by constructing an index of x-rays collected from a particular spot on the specimen surface, which is known as a spectrum.

An example spectrum for phosphorous with fluorine. The element peaks for analysis have been marked in red. Where two peaks are shown for the same element, the smaller peak corresponds to one of the other lines of electron replacement, such as an 'M' shell electron jumping into the 'K' shell. The y-axis is an arbitrary pixel count.

(example spectrum taken from the Oxford Instruments INCA software.)

The x-axis of the spectrum is the x-ray energy scale, along which x-rays collected from various elements are registered, and which in turn form a series of peaks along the x-axis where each peak corresponds to a particular element. With modern software, it is possible to collect a series of spectrums for each point (pixel) analysed by the electron beam probe, as it is scanned across the surface of the specimen. If one spot only is analysed, quantification of the relative proportions of elements under that spot is possible, which is known as point analysis. If however a line of pixels is analysed, a line-traverse analysis can be performed which can highlight changing proportions of elements with distance along the line. This is known as line scanning. Finally, each pixel in an image can also be analysed, which illustrates the distribution of a chosen element across the image. This is known as dot-mapping of the elements, and can take up some considerable time as the electron probe is required to dwell on each separate point for a pre-determined period of time in order to collect enough data for analysis. In many applications of the technique to concrete petrography this is normally done in conjunction with a backscattered electron image, where each pixel of the image has an exact and corresponding pixel on each of the element maps, which are collected at the same time and at the same resolution. Another reason for choosing this image type is that efficient x-ray analysis requires a flat polished surface where there are few obstructions to x-rays leaving the surface of the specimen that may cause contamination of the signal. The software requires calibration to known standards before normal use.

In this technique, an image of the area for investigation is first captured, and the exact point for analysis is determined by the operator. The software is used to identify this point, and data capture can begin. With modern software, it is possible to identify multiple points on the specimen surface for analysis, either randomly selected or according to a pre-determined pattern (i.e. at intersections of a grid overlay or along a line). The software can then be set and left running to capture data automatically leaving the operator free to perform other functions. This is a standard feature and can be used in each of the three analytical variations.

Line scanning is useful for investigating the change in elements along a line of traverse on the specimen surface. Each pixel along the line is analysed and contains its own spectrum of data for selected elements. It is therefore possible to plot a series of individual line plots, showing the intensity of elements along the length of the line. The lines themselves can be placed at any position on the image, between any two points, and for any length up to the total available image dimension. Line point analysis data collection is very quick and large amounts of data can be collected over a relatively short period of time.

As mentioned above this technique is used to analyse the distribution of elements in an image. Having selected the elements for analysis, an image resolution setting is selected, and then the data collection begins. Selection of image resolution is important as the time required to collect data in this technique is significantly longer than for point or line analysis. Higher resolution images will require more time than lower resolution images. For example, an image whose dimensions are double those of another will capture four times as much data and consequently take four times as long to capture the same amount of data. Large amounts of data are captured anyway, and stored as separate element intensity images, which are not greyscale images. However, a standard greyscale image of the analysis area is normally captured in tandem with the element intensity images, or 'dot-maps' (see below). This has many advantages in later image analysis operations, as each of the images can be used in any arithmetic combination, with any of the other images during analytical image algorithms. It is therefore possible to 'reconstruct' many of the phases and features present in the greyscale image based on compositional element data.