Bad Totals for Beam Sensitive Samples

[Richard Walshaw]

Hi All
For a few weeks now I have been getting low totals (96-98%) on feldspars and glasses. This suddenly occurred while using the same standards and conditions (10nA; 2-10um defocus) I have always used. I have checked through backgrounds, matrix corrections and everything else I can think of. This only occurs with beam sensitive samples – it does not occur with olivine, pyroxene or garnet, which all give good analyses.
My gut feeling is that since a certain date I have been damaging my feldspars and glasses, its as though PFE is not setting the beam to defocus properly.  Confusingly it does appear to defocus when I check using a fluorescent mineral but this is only visible with larger spots and is hard to assess down at the 2um level.
Has anyone any thoughts on this? Sound familiar? Am I missing something?
Please also check the attached screenshot . The beam conditions displayed via the Analyze! window on the left show a 2um defocus. The conditions displayed in the Automate! Window on the right show the same conditions without a defocus.
What are the two fields actually displaying here and which set of conditions actually apply to the analysed sample?

Richard

[John Donovan]

Please also check the attached screenshot . The beam conditions displayed via the Analyze! window on the left show a 2um defocus. The conditions displayed in the Automate! Window on the right show the same conditions without a defocus.
What are the two fields actually displaying here and which set of conditions actually apply to the analysed sample?

Hi Richard,
I have no idea why you are suddenly seeing low totals on your beam sensitive materials.  I guess I would first have to ask: is the TDI correction turned on properly to correct for this?  Maybe you could post an analysis or even better a TDI fit plot from PFE for Na and Si (the main culprits), though K and Al should also be treated if present. This is accessed from the Standard Assignments button.

Also please remember that, although not always necessary, one can also "turn on" TDI acquisitions for standards also as seen here:



Therefore if your standards are also beam sensitive you can correct for this- especially important for materials such as apatite where the crystal orientation in the standard mount affects the standard intensities.

[Richard Walshaw]

Thanks John
I have looked into TDI but will double check everything.
Can you tell me why the two sets of beam conditions for the same sample in the attached image do not agree?

Richard

[John Donovan]

I have looked into TDI but will double check everything. Can you tell me why the two sets of beam conditions for the same sample in the attached image do not agree?

Please feel free to post a couple of TDI plots. 

I don't see any attached image.

[Philipp Poeml]

I guess Richard is talking about the kyanite conditions2.jpg attached to post #1.

[John Donovan]

I guess Richard is talking about the kyanite conditions2.jpg attached to post #1.

Ah, thanks, I see now.

Yes, the conditions from the Acquire! window are the last column condition sent to the instrument, so they are usually "current".

The conditions you have circled in the Automate! window are the "digitized conditions" parameters which are then assigned to the various positions samples for automatic acquisition with different conditions for different samples.  The display are actually "disabled" (more gray) to indicate that you haven't selected the "Use Digitized Conditions" option in the Automate! window.

Here is a discussion using "digitized multiple sample setups":

http://probesoftware.com/smf/index.php?topic=111.msg405#msg405

To see the enable/disable on these different automation options try selecting these different "automation basis" options and see the enable/disable selection change:

[Richard Walshaw]

Well, I appear to have fixed my problem now. I have rebuilt my setups again and am now getting correct values from feldspar and glass standards. I still have no idea what the issue was, but its gone!

Edit by John: good to hear, but high totals in the unknowns usually can be attributed (besides possible spectral interferences and/or heterogeneous interaction volumes) to a bad data point(s) on your standards.

That is to say, if some *low* intensities are averaged into the standard it will usually appear as *high* totals in the unknowns. The Report button in the Analyze! window is your friend in this situations.

That or try the Run | List Anomalous Intensity Data for Standard or Unknowns menu as described here:

http://probesoftware.com/smf/index.php?topic=168.msg725#msg725

For general suggestions on bad analytical totals see this page in the Reference manual:

http://probesoftware.com/download/PROBEWIN.pdf#page=233