WDS-EDS Quant: Ensure that EDS time is longer than sum of all WDS elements!

[wonachlas]

Hello Microprobers,

We encountered some strange behavior when conducting combined WDS-EDS measurements with PfE using Pathfinder EDS software.

I was originally setting the EDS acquisition time to be equal to the sum of all WDS elements (in Acquisition Options of PfE). I had 5 WDS spectrometers, each with 2 elements, each measured with MAN backgrounds for 10 s. Thus I set EDS acquisition time to be 20 seconds, so that it would be acquiring at the same time as WDS without adding any extra time. This created a very subtle problem that was tricky to diagnose: I noticed that one of my second-in-order WDS elements had higher than normal RSD on standards (~5-10%, should be <1%). When acquiring data on a fluorescent sample, we could see that the beam was turning on/off the sample a few times at the start and at the end of each measurement. It is normal for Pathfinder to turn the beam on/off at the start of the measurement (wish it didn’t, but always does). However, it was not normal for the beam to be turning on/off at the end of the measurement. As it turns out, Pathfinder was cutting the beam at the end of the 20 s EDS acquisition time. However, the WDS elements were still finishing up (because of the added spec drive time), such that the last few seconds of each WDS element were interrupted by Pathfinder cutting the beam, then PfE immediately commanding to turn it back on. The beam was cut from the sample for only a few milliseconds, and apparently this process does not take exactly the same amount of time for each measurement, thus creating variations in total counts for WDS elements measured second. All of the second order elements had slightly higher RSD on the standards, but it was most apparent for Mg (measured after Na on Sp1 LTAP), which ultimately helped us narrow down the problem. If the second order elements were of lower abundance, if the WDS time was only slightly longer than the EDS time, or if the sample was not fluorescent, this problem might be happening unnoticed in the background with only minimal (but potentially significant) effect.

Takeaway: When performing combined WDS-EDS, always make sure to set your EDS acquisition time to be slightly longer than the actual sum duration of all WDS elements.

[John Donovan]

What I do is simply disable the faraday cup control in the Thermo AutomationClient software.

Otherwise Thermo Pathfinder will insert the faraday at the *beginning* of the EDS-WDS acquisition, also affecting the WDS counting.

Some say it's a feature, others say it's a bug!   

[Probeman]

It should also be pointed out that there is really no need to measure the beam current with Thermo Pathfinder when acquiring WDS and EDS spectra in Probe for EPMA. Because the quant/display routines in PFE automatically correct the EDS net intensities for beam drift using the faraday cup measurements made by PFE.