I've definitely been using the coating correction. For the originally Au-coated glasses, it made a difference, but not enough seemingly. That probably stems from the fact that they were coated in another lab with a questionable amount of Au (~5nm estimate apparently not correct).

Getting all standards and unknowns coated in C, by me, has made a huuuge difference. On my unknowns, the stoichiometric vs calculated oxygen is off by less than 2 wt%, so I think I'm doing something right.

And thanks for that link; it clears up a lot of questions I had about the coating correction. Cheers!

The Forensics Chemistry journal isn't necessarily on your radar but there is a paper in there https://www.sciencedirect.com/science/article/pii/S2468170920300400?via%3Dihub that I published recently with some GSR colleagues that is relevant to this topic.  It basically discusses validation procedures to ensure that an SEM/EDS system is performing and QC procedures to check and document the performance.

NIST paid the tariff so it is available for free under the Creative Commons license.

I've been working all week on this, and I think that the largest source of error for the USGS glasses was that they were Au-coated.    That's it.

Obviously I've done a lot of other things to tighten up my probe work, but taking off the Au and coating with C puts Si-Al-Mg compositions within 3% or less variance. I wish it would have been higher on my "Things To Check" list!

I'm also finding that for Na K, mineral standards seem to work better when analyzing glass SRMs as unknowns.

Now I just have the original challenge of how to analyze a glass. Thanks for all the help and time, John!

Julien Allaz and I have been discussing the oxygen-halogen correction in PFE and wanted to share our recent efforts to improve the way PFE handles oxygen-halogen containing samples.  It's a little complicated.

First of all it should be noted that the program properly handles the oxygen-halogen correction on unknown samples for all situations, particularly when oxygen is calculated by stoichiometry. It also handles the situation where an unknown is calculated by simply specifying the amount of oxygen. But of course in that case, the oxygen-halogen correction has already been applied.

PFE also handles the situation where a standard is analyzed and the oxygen in the standard database was *not* already corrected for the oxygen-halogen equivalence (#282). As seen here:

Oxygen Calculated by Cation Stoichiometry and Included in the Matrix Correction
Oxygen Equivalent from Halogens (F/Cl/Br/I), Subtracted in the Matrix Correction

St  282 Set   2 Fluor-phlogopite (synthetic), Results in Elemental Weight Percents
 
ELEM:        F      Si      Mg      Al       K       O
TYPE:     ANAL    SPEC    SPEC    SPEC    SPEC    CALC
BGDS:      LIN
TIME:    20.00     ---     ---     ---     ---     ---
BEAM:    29.99     ---     ---     ---     ---     ---

ELEM:        F      Si      Mg      Al       K       O   SUM 
     6   8.453  20.002  17.307   6.404   9.281  38.218  99.665
     7   8.766  20.002  17.307   6.404   9.281  38.086  99.846
     8   8.536  20.002  17.307   6.404   9.281  38.183  99.713

AVER:    8.585  20.002  17.307   6.404   9.281  38.163  99.741
SDEV:     .162    .000    .000    .000    .000    .068    .094
SERR:     .094    .000    .000    .000    .000    .039
%RSD:     1.89     .00     .00     .00     .00     .18

PUBL:    9.020  20.002  17.307   6.404   9.281  41.775 103.789
%VAR:    -4.83     .00     .00     .00     .00   -8.65
DIFF:    -.435    .000    .000    .000    .000  -3.612
STDS:      835     ---     ---     ---     ---     ---

Note that the "published" (PUBL) amount of oxygen from the standard database is based on the assumed cation stoichiometry, and is not correct, but the halogen correction does make the proper correction.

Previously if the standard composition was already corrected for the oxygen-halogen correction as in the following standard (#284), the program performed a double correction of the oxygen-halogen equivalence (see above posts). However, Julien and I added some code that checks for a negative excess oxygen from the standard database, and if it is negative, it zeros out the deficit oxygen before entering the matrix correction routine. The new code is seen here:

If sample(1).Type% = 1 Then
analysis.WtPercents!(chan%) = ConvertTotalToExcessOxygen!(Int(1), sample(), stdsample())
If UseOxygenFromHalogensCorrectionFlag And sample(1).OxideOrElemental% = 1 Then
If analysis.WtPercents!(chan%) < 0# Then analysis.WtPercents!(chan%) = 0#  ' zero out oxygen equivalence of halogen from standard database
End If

' For unknowns, use specified oxygen weight percent
Else
analysis.WtPercents!(chan%) = stdsample(1).ElmPercents!(ip%)
End If

Here is an example of a standard, already corrected for the oxygen equivalence of halogens, analyzed with the oxygen-halogen correction:

Oxygen Calculated by Cation Stoichiometry and Included in the Matrix Correction
Oxygen Equivalent from Halogens (F/Cl/Br/I), Subtracted in the Matrix Correction

St  284 Set   2 Fluor-phlogopite (halogen corrected), Results in Elemental Weight Percents
 
ELEM:        F      Si      Mg      Al       K       O
TYPE:     ANAL    SPEC    SPEC    SPEC    SPEC    CALC
BGDS:      LIN
TIME:    20.00     ---     ---     ---     ---     ---
BEAM:    30.01     ---     ---     ---     ---     ---

ELEM:        F      Si      Mg      Al       K       O   SUM 
     9   8.806  20.002  17.307   6.404   9.281  38.069  99.870
    10   8.832  20.002  17.307   6.404   9.281  38.059  99.885
    11   8.824  20.002  17.307   6.404   9.281  38.062  99.880

AVER:    8.821  20.002  17.307   6.404   9.281  38.063  99.878
SDEV:     .013    .000    .000    .000    .000    .006    .008
SERR:     .008    .000    .000    .000    .000    .003
%RSD:      .15     .00     .00     .00     .00     .01

PUBL:    9.020  20.002  17.307   6.404   9.281  37.980  99.994
%VAR:    -2.21     .00     .00     .00     .00     .22
DIFF:    -.199    .000    .000    .000    .000    .083
STDS:      835     ---     ---     ---     ---     ---

STKF:    .1715     ---     ---     ---     ---     ---
STCT:    59.95     ---     ---     ---     ---     ---

Note that both the published concentration of oxygen is corrected for in the standard database, and the stoichiometric oxygen in the standard is properly corrected as expected.

Part of the reason for this issue is that many, many years ago we decided to allow the user to handle the deficit (or excess) oxygen in the standard compositional database, but then include that deficit (or excess) oxygen in the matrix calculation. For example, when analyzing magnetite, the default is to include the excess oxygen by simply loading the speciifed oxygen concentration from the standard database.

But (and here it gets complicated), if the user decides to simply calculate oxygen by stoichiometry for this standard (instead of loading the specified oxygen from the standard database as is the default), the program automatically includes the deficit (or excess) oxygen from the standard database.

But if the standard composition in the standard database is already corrected for the deficit (or excess) oxygen, this results in an double correction during the matrix correction. The new code above should prevent this double correction.

This is slightly on the wrong board... but at customer request, we added (in Pathfinder x.x) an "incandescent" option to replace the rainbow scheme for COMPASS component maps.
It can be set in the NSSMachineSettings.ini file.
In the [Image] group.
Set UseIncandescent=1.