There has been a slow but steady improvement in EM Probes over the last 50 years. Some of these basic changes include: 1) The diffusion pump has been replaced with a turbo pump, 2) High stability field emission electron sources that allow for sub-micron spatial resolution mapping at low voltages, 3) Computer upgrades, 4) Operating system changes to latest window versions, 5) Basic improvements to BSE and SE detectors, 6) Automated stages, 7) Synthetic multi-layer diffraction ‘crystals’ that improve detection of low energy x-rays and 8) Integration of energy dispersive spectrometers (EDS) to supplement the wavelength dispersive (WDS) spectrometers and provide more flexible analysis options.
Now one improvement that needs to be made, for a dramatic change, is the replacement of the proportional counter (PC) with a silicon drift detector (SDD) [ 1- 5]. The replacement of the PC with a SDD (called WDS-SD) allows high order diffracting to be completely discriminated against, along with a dozen other benefits. The replacement of the gas counter with an SDD sensor resolves many issues and brings additional benefits [ 1- 7]. These include:
1. It provides complete high-order x-ray line filtering (pulse height analysis, PHA) due to the higher energy resolution of the SDD compared to the gas counter ( figure 1a). This also provides improved low energy x-ray analyses. In other words, no high order diffraction included in collection.
2. An SDD sensor can directly replace a gas counter in a WDS and is referred to as a WDS-SD.
3. Only moderate cooling of the SDD sensor is required.
4. A WDS-SD has effectively no energy shift as a function of x-ray count rates as opposed to the significant shifts seen in gas counters at high count rates ( figure 1b).
5. Better energy resolution.
6. It allows for the identification and removal of counter artefacts (e.g. escape peaks, sum peaks) not separable in the gas counter energy spectrum ( figure 1a). In other words, almost no interferences from extraneous X-Rays.
7. It has essentially no sensitivity to room temperature or atmospheric pressure variations.
8. It can accommodate much higher count rates than a conventional gas counter.
9. It produces superior peak-to-background ratios (WDS-SD) over a conventional WDS-GC. However, the current count rates are lower due to the current non-optimal size/shape of the SDD detector/sensor, but larger area SDD’s or smaller Rowland circle can partly compensate for this.
10. Removal of the variable efficiency factors in the gas counter opens the possibility to directly calculate background intensities, allowing for analysis without measuring backgrounds.
11. Better low energy performance (improvement in the light and mid element range).
Fig. 1.
a)) WDS-SD spectrum measured at the Cu Kα peak position, counts (Y) in both linear and log scales versus energy (X) [ 1- 3]. First order bremsstrahlung is under the main peak. Second order bremsstrahlung is shown coinciding with pulse pile up while everything else is an artefact of the system, and b) WDS-GC versus WDS-SD comparison of peak shift/position versus count rate, peak position (Y) versus count rate (X) for Iron, reproduced from reference [ 2].