WDS element mapping with Thermo NS7

[kthompson75]

Questions have been asked about Phase Mapping in the Noran System 7 using the WDS inputs. The answer? Absolutely. I do this all of the time.

There are 2 phase mapping software packages in the Noran System 7. (1) COMPASS + X-Phase and (2) X-phase. COMPASS utilizes the PCA algorithms developed by Paul Kotula at Sandia National Labs. The results of using this algorithm are tremendous and detailed elsewhere. X-phase uses the EDS element maps to create phases. X-Phase is the effective equivalant of saying "Element-based Phase mapping." While COMPASS is the equivalent of saying "Spectrum-based phase mapping." Because WDS involves counts and not a spectrum, X-phase is very effective at creating WDS-based phase maps. The advantage with WDS element maps is that the data is orders of magnitude cleaner (better trace detection, better resolution) than with EDS element maps.

The main question is: How to get the WDS element maps into the Noran System 7 software. On my system, We have a single WDS spectrometer on an FESEM. I simply take several WDS maps in serial acquisition mode, move the map files into a common folder, update the header file and then run X-phase. It works like a charm. For those with a probe, 4 or 5 WDS spectrometers running simultaneously makes this more efficient. The trick here is getting the WDS counts data out of the probe and into NS7. If there are outputs on the spectrometer (typically a BNC), one can directly hook the spectrometer out directly into the NS7 WDS BNC input. We also sell a solution that pulls all 5 spectrometer output at once into the NS7. The trick, of course, is getting access to the WDS spectrometer outputs. If these aren't available, the alternative is a software solution.

It is great fun to perform WDS element phase mapping and COMPASS EDS spectral mapping on the same data set and see which provides a better result. So far, I have not found a situation COMPASS failed to find all of the phases. The WDS mapping falls short if I am not aware of an element present in the sample, so I typically run COMPASS live-time and then populate my elements for WDS anlaysis based on those results. Often-times, I run the COMPASS EDS only and then do a "long" (60 seconds per element) point-and shoot WDS analysis in the middle of each phase to get the best possible Quant of the phases.

[John Donovan]

There are 2 phase mapping software packages in the Noran System 7. (1) COMPASS + X-Phase and (2) X-phase. COMPASS utilizes the PCA algorithms developed by Paul Kotula at Sandia National Labs. The results of using this algorithm are tremendous and detailed elsewhere. X-phase uses the EDS element maps to create phases. X-Phase is the effective equivalant of saying "Element-based Phase mapping." While COMPASS is the equivalent of saying "Spectrum-based phase mapping." Because WDS involves counts and not a spectrum, X-phase is very effective at creating WDS-based phase maps. The advantage with WDS element maps is that the data is orders of magnitude cleaner (better trace detection, better resolution) than with EDS element maps.

Hi Keith,
We've talked about getting the WDS map data into the first few (empty) spectrum channels in Compass for phase extraction where a (WDS) trace (or minor light element) element might be helpful for discriminating similar phases.  Not a priority, but just curious where are we on that front?