1. SEM and EPMA
Some Advanced Topics
October 7, 2009

2. Beyond the Basics Stereo SEM EDS of particles: beware! TTL Detectors
X-ray mapping
Feature Sizing, Chemical Typing
FIB

3. Stereo SEM
Why? Graphical demonstration of 3D shapes of rough or complex objects
How? Very generally …..
Acquire two SEM images of the same object at the same magnification but by tilting the stage slightly.
Color each image red or green in Photoshop.
Then in Photoshop overlay the two images, while wearing red-green stereo glasses, shift them slightly apart until you have the correct 3D effect.
Show downstairs in 2nd floor visualization lab for maximum effect…
Exact detailed instructions on wall in room 308

4. Example: porous sandstone

5. Stereo SEM Details For rough objects, use 3-7° tilt
For smooth objects, use more, 7-15°
Use less tilt if higher magnification
If you need to refocus, do NOT change objective lense setting, rather move sample stage Z up or down.
Cited reference in literature: Heuser, 1989, Protocol for 3-D visualization of molecules on mica via the quick-freeze, deep-etch technique, Journal of Electron Microscopy Technique, vol 13, p

6. EDS of particles

7. Particles - 1
Mass effect/error: electrons escape from sides of small particles if E0 is large enough, so quantitative analysis will be in error because different elements (with different binding energies) will be affected differently
Goldstein et al, 1992, p. 479, 481

8. Particles - 2
Absorption effect of non-flat upper surface: different path length from normal flat geometry and the “normal” way we do quantitative analysis is to use FLAT polished standards for calibration -- so we could have “too much” x-ray intensity for particles
Goldstein et al, 1992, p. 479, 481

9. Particles - 3
Variable effect of geometry of trajectory between beam impact area on non-uniform surface and the location of the detector -- so we get different results from the same material, depending upon where we place the electron beam.
Goldstein et al, 1992, p. 479, 481

10. Results from 2006 777 student project on EDS of rough samples with VP-SEM
The first column shows the actual chemical composition, followed by the average composition of 20 points on different grains, followed by the variation (=standard deviation) of those individual measurements

11. Silicon Drift Detectors
In the past ~5 years, the SDD has created much enthusiasm. It uses a small Peltier cooler (no LN) and the ringed design and lower capacitence permits hundreds of thousands of counts per second, with Mn Ka resolution of <130 eV.
Image from Bruker web page

12. Silicon Drift Detectors - 2
“The major distinguishing feature of an SDD is the transversal field generated by a series of ring electrodes that causes charge carriers to 'drift' to a small collection electrode. The 'drift' concept of the SDD (which was imported from particle physics) allows significantly higher count rates.” - Wikipedia
For a simulation, go to

13. EDS X-ray Mapping-1
BSE imaging provides rapid distinction of some/many/most chemically distinct phases in a multiphase sample.
However, in many cases, having explicit 2D chemical information is value. This is where “x-ray mapping” comes into play.
X-ray mapping has changed tremendously since it was first introduced in 1956 as a combination of WDS (crystal diffraction spectrometry) and SEM: in those days it was called “dot mapping” as literally dots would be painted onto the CRT. To capture it, a photograph would have to be snapped of the screen. It generally was pretty grainy. And only 1 or 2 or 3 elements could be acquired simultaneously (depending on how many spectrometers were on the electron probe).

14. EDS X-ray Mapping-2
EDS was developed in the late 1960s and soon began to be used for X-ray mapping.
For the next 30 years or so, users would select “regions of interest” (ROIs), essentially the peak areas of particular elements -- sometimes limited to a finite value like 8 or which would then be used to “color in” a 2D area over which the beam would scan, either one very long and slow scan, or many faster scans that would be averaged.
Note: To this point, all the above maps included both the characteristic X-ray being chosen AND the background/ continuum under the characteristic peak.

15. EDS X-ray Mapping-3 Things have changed a lot in the past decade.
Digital pulse processing have taken over from the older analog processing, making shorter time constants and thus higher count rates possible (e.g. 30,000 cps vs 3,000 cps before) which create many more opportunities -- larger areas, shorter times.
Improved electronics and software to remove the “ROI” limits, and instead ACQUIRE THE WHOLE SPECTRUM AT EACH PIXEL! This is known as “spectral imaging”
And the development of SDD (Si-Drift Detectors) boosts the count rates to the hundreds of thousands of cps.

16. X-ray Mapping and the clock
Reed, 1996, Fig 6.1, p. 102
Due to the low count rate of detected X-rays, dwell times in the past generally needed to be hundreds of milli-seconds. A 512x512 X-ray map at 100 msecs took ~8 hours to acquire. The improvements to EDS systems with improved digital processing throughput allows 1-10 msec dwell times, dropping the x-ray mapping times down to the hour or less range, in many cases

17. Spectral (“hyperspectral”) Imaging
Here any and all x-rays detected are mapped to each pixel over which the beam is scanning. This is both very powerful (see elements not known to be present before starting), but also loses something relative to other ‘slower old-fashioned’ maps--lower counts in peak channels.

18. X-ray Mapping
This is an old-fashioned map using older EDS software + WDS channels with the SX51 electron probe, slowly moving across the sample, then applying colors to elements. This instrument is built to be more stable
3 X-ray maps combined; each element set to a color, and then all merged together in Photoshop. The maps took ~8 hours to collect.
than the SEM so there is no smearing as in the previous image.

19. Here is an example of false compositional contrast, an artifact of the background being a function of Z (MAN). Specimen is Al-Cu eutectic; X-ray maps are (a) Al, (b) Cu, (c) Sc. The contrast in (c) suggests Sc is present in the Cu-rich phase. However, there is no Sc, only the background in the Cu-rich phase is elevated relative to the background in the Al-rich phase. Thus one needs to be aware whether the background is or is not subtracted from X-ray maps, esp. when looking at minor elements where the continuum is a major component.
Continuum Artifact
Goldstein et al, 1992, Fig 10.6, p. 535

20. Feature Sizing, Chemical Typing
This is a valuable feature of the SEM for locating and identifying a particular mineral in a mixed population (e.g. K-rich minerals mixed with K-poor minerals.
Acquire BSE image with good contrast so the K-rich mineral is distinct
Set a desired brightness level as the test for the mineral in question (if brighter than, then acquire short EDS spectrum of center of grain)
Set elements to be evaluated; set short (~2 second) EDS acquisition time
Set the stage coordinates of the corners of the area; Run the program
Return the next day and look at the list of grains it acquired EDS spectra; order from high to low K; then drive to each grain to verify it is the grain you want.
The next 3 slides go through the process…

21. 1. Set the BSE intensity and then set the range to select one phase

22. 2. Select the elements needed to find the phase

23. 3. After SEM done, check the grains the software suggests are the ones you want

24. There are other features to the EDS software…
There are other features to the EDS software…. Finding the “needle in the haystack” using the “Maximum Spectrum” where an artifical spectrum is generated which has in each channel the highest intensity found ANYWHERE in the sample. Thus if only one pixel has a significant amount of one element, it will stand out.