Electron Microprobe Analysis (EMPA) Laboratory

Uses for EMPA

EMPA is useful for determining the elemental composition of a solid sample. Because the instrument can also resolve and display small features that appear on the surface of a sample, the method is particularly adept at identifying the composition of small inclusions embedded in a matrix. The user can view the surface at high magnification (up to 40,000X), identify the features of interest, and then analyze the composition of those features. The technique will also identify trace elements or bulk contaminants. However, unlike XPS, EMPA samples a volume about 1-3 microns in radius extending into the sample, so it is not especially sensitive to the composition of the very near surface.

The instrument

The JEOL JXA-733A Superprobe has 5 wavelength-dispersive spectrometers and a Be-window SiLi energy dispersive detector. The instrument is fully automated by Advanced Microbeam hardware and PROBE-PRBSE microanalysis and digital imaging software. The lab has over 360 standard reference materials for quantitative bulk or thin film analysis. Quantitative analysis of bulk specimens (> 1000 nm thick) is performed using the CITZAF matrix correction algorithms incorporated into the PROBE software. Thin film (10-1000 nm) analysis is performed using the program STRATA.

Basic theory of operation

The electron microprobe is used for chemical analysis of very small solid samples. A finely focused (approx. 1 micron) beam of electrons is directed onto a flat polished specimen. The high energy electrons ionize the inner shells of the target elements in the sample. The decay of the ionized state results in the emission of characteristic x-rays.

Using Bragg's and Moseley's laws to determine the energy spectrum of x-rays permits the identification of the wavelengths unique to every element in the periodic table. Identifying which elements comprise a specific sample constitutes qualitative analysis. The method is simple and very fast.

Quantitative analysis is accomplished by measuring the intensity of the characteristic wavelengths for each element in the sample. The unknown sample intensities are then compared to intensities measured on standard reference materials of known compositions. Necessary corrections are applied by computer and the results can be displayed as weight % or atomic proportions.

The spatial relations of different materials or particles can be displayed either by moving the beam over the sample or moving the sample under the beam.

Detection limits and spatial resolution

The instrument is capable of analyzing the light elements C-N-O-F with a detection limit and sensitivity of about 500-1000 ppm and the heavier elements Na up to U with a detection limit and sensitivity of about 300 ppm (depending on the mean atomic number of the matrix and x-ray counting times).

The spatial resolution for quantitative analysis is of the order of 1-3 microns. The spatial resolution for secondary and backscattered electron imaging depends on the accelerating voltage and beam current and is approximately 100-200 nm.

Restrictions on samples

Samples must be solid and vacuum compatible. We bond them to the mounting block with epoxy. A sample can range in size from 0.5 mm x 0.5 mm x 0.5 mm to about 1 cm x 5 cm x 5 cm. The sample must be conducting. If it is not conducting, we can sputter coat it with about 30 nm of carbon to make it conducting, but you must inform us in advance if the sample is insulating or if you are uncertain.

Before analyzing thin films (< 1000 nm thick), we generally simulate the expected results based on the estimated composition using the STRATA software to determine if EMPA will have the desired sensitivity. If the same elements appear in both the film and the substrate, then determining the amount of those elements in the film will generally become increasingly difficult as the film becomes thinner. EMPA can determine the film thickness if the density is accurately known, or the density if the film thickness is accurately known.

Other sources of information on EMPA

A good place to start is http://darkwing.uoregon.edu/~mshaf/epmahome/otherlab.htm.

For further information about EMPA at Stanford, please contact Bob Jones at
bobjones@pangea.stanford.edu or (650) 725-1677.