Tracking instrument drift on long runs

[Jarrod Crum]

We make long runs analyzing glass samples on our JEOL 8530 microprobe.  The glasses contain 10-30 elements of interests which makes the list of standards quite long and time consuming to recollect during runs.  So my question is the following:  Can we select a glass standard and remeasure that periodically during a run to track instrument drift with time and correct for environmental effects (e.g., lab temperature swings) of the lab space?  I can't find a way to do this in the software, but I certainly haven't learned everything that the software is capable of.

Thanks for any feedback on this!

[John Donovan]

This is a very good question.

Ideally the EPMA lab should have temperature variations that are limited to +/- 0.5C but in the real world that is not always possible of course:

https://probesoftware.com/smf/index.php?topic=330.0

If temperature control of the lab is not sufficient Probe for EPMA has a built in standard intensity drift correction which is designed to apply a linear interpolation between standardizations during an extended duration probe run:

https://probesoftware.com/smf/index.php?topic=168.0

Therefore the standard intensity drift correction depends on the magnitude of the temperature drift in the lab, and is of course particularly important for PET crystals which have the largest thermal coefficient of expansion.

One could (in principle) include a standard analyzed as an unknown sample at repeated intervals in the automation list to track instrument drift, but there is no way to apply that drift to the standard (or unknown) intensities at analysis time. But I do know of users who have done this as a "accuracy check" for runs performed over many weeks or months.

So if the instrument drift in your lab cannot be adequately calibrated (linear interpolation) using the Run Standard Samples and Run Standard Samples (again) automation actions, what I would do is to utilize the Re-Standard Interval (hrs) feature shown here:



With this feature one can simply specify the re-standardization interval (in hours) and the software will automatically re-run the selected standards every X hours.  Note that the software will only run the selected standards, so you don't have to select *all* the standards, perhaps just the standards assigned to the major elements. This is because trace elements aren't going to drift much (if at all), since background intensities tend to be very stable over time and in the case of off-peak backgrounds, they are measured for every analysis (unless using Nth point backgrounds in which case they are measured every N points).

[Jarrod Crum]

Thanks for the information.  We have used the re-standardization interval but unfortunately the time needed to collect all the standards is about 12-24hr of run time.  The sample(s) collection time is usually less less time intensive so the run time can quickly become unpractical.  Also the temperature fluctuation cycles more often then the re-standardization interval. 

We will try adding in single standard as an unknown at a more frequent interval to see if that can be helpful.
 

[John Donovan]

Thanks for the information.  We have used the re-standardization interval but unfortunately the time needed to collect all the standards is about 12-24hr of run time.  The sample(s) collection time is usually less less time intensive so the run time can quickly become unpractical.  Also the temperature fluctuation cycles more often then the re-standardization interval. 

We will try adding in single standard as an unknown at a more frequent interval to see if that can be helpful.

Yes, that is why I suggested just using a shorter re-standardization interval with only the major element standards that are the most subject to drift.

Running a single standard as "drift monitor" is fine, but PFE won't be able to utilize that information. You need to run the standard as a standard for the standard intensity drift correction to be applied.

[John Donovan]

You might also check out the Drift utility that comes with Probe for EPMA. It's in your Probe Software Utility menu accessible from the Start button.