John
PfE's ability to merge both WDS and EDS is fantastic! We are doing some tests to see how much we might run some elements by EDS and some by WDS and PfE is the bomb!! We are acquiring almost all elements on EDS and WDS simultaneously to be able to do exact head to head comparisons.
There is one point where there may be a problem with the matrix correction in a situation where oxygen is by WDS and F is by EDS (when both are by EDS, no problem), but with the mixture, there is apparently a flag set to remove oxygen equivalent to the measured F....but as I noted before, here oxyen is NOT by stoichiometry but explcitly measured. So in the current scenario, there are major problems with all WDS elements.
Here is biotite by EDS [in PFE with standards]...pretty good...
Element Na (ka) Mg (ka) Si (ka) Al (ka) P (ka) Cl (ka) K (ka) Ca (ka) Ti (ka) O (ka) F (ka) Fe (ka) Mn (ka) H () Li () Rb () Total
Average: 0.046 8.517 18.364 5.582 0.037 0.047 8.353 0.041 1.551 37.469 6.292 12.902 0.227 0.24 0.093 0.091 99.851
Here is WDS by itself (ignoring F by EDS)
Stat Na (ka) Mg (ka) Si (ka) Al (ka) K (ka) Ca (ka) Ti (ka) O (ka) Na (ka) Fe (ka) Mn (ka) Total
Average: 0.282 8.845 19.065 5.848 8.498 0.003 1.373 38.22 0 13.361 0.168 96.088
Published: 0.12 8.52 18.12 5.86 8.12 0.006 1.38 40 n.a. 13.05 0.18 95.78
But see now all elements by WDS + F by EDS, with much lower values for all elements plus total now is 49 wt%
Stat Na (ka) Mg (ka) Si (ka) Al (ka) K (ka) Ca (ka) Ti (ka) O (ka) Fe (ka) Mn (ka) F (ka) Total
Average: 0.146 4.184 9.546 5.751 5.187 0.002 1.088 11.656 7.464 0.086 3.517 49.051
Edit by John D.: Please don't paste from the little Copy button in the Analyze! window as it includes tabs for pasting into Excel. The tab characters aren't handled well in the forum. I replaced the tabs with spaces and it much more readable now.
Instead, simply select and copy text from the log window. Then after pasting it in, just use the Tt button in the post window to select a fixed width font and all the columns will line up nicely.
Hi John,
Interesting. Better send me the MDB file so I can troubleshoot this.
John,
Until I get the MDB file from you I can't diagnose your problem. It may be be just a calculation setting...
I took one of my EDS-WDS runs and analyzed some elements using WDS and some using EDS (at 50 nA!) and all looks good:
St 160 Set 2 NBS K-412 mineral glass, Results in Elemental Weight Percents
ELEM: Si Al Fe Mg Ca Si Al Fe Mg Ca Mn O
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC SPEC
BGDS: MAN MAN MAN MAN MAN EDS EDS EDS EDS EDS
TIME: 60.00 60.00 --- --- --- --- --- 40.00 40.00 40.00 --- ---
BEAM: 49.81 49.81 --- --- --- --- --- 49.81 49.81 49.81 --- ---
ELEM: Si Al Fe-D Mg-D Ca-D Si-D Al-D Fe Mg Ca Mn O SUM
XRAY: (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) () ()
91 21.184 4.757 --- --- --- --- --- 7.746 11.246 10.496 .077 43.597 99.103
92 21.232 4.749 --- --- --- --- --- 7.835 11.231 10.413 .077 43.597 99.135
93 21.160 4.742 --- --- --- --- --- 7.848 11.243 10.505 .077 43.597 99.172
94 21.130 4.759 --- --- --- --- --- 7.803 11.241 10.511 .077 43.597 99.118
95 21.214 4.765 --- --- --- --- --- 7.897 11.170 10.505 .077 43.597 99.225
AVER: 21.184 4.754 --- --- --- --- --- 7.826 11.226 10.486 .077 43.597 99.151
SDEV: .041 .009 --- --- --- --- --- .056 .032 .041 .000 .000 .049
SERR: .018 .004 --- --- --- --- --- .025 .014 .018 .000 .000
%RSD: .19 .19 --- --- --- --- --- .72 .29 .39 .00 .00
PUBL: 21.199 4.906 n.a. n.a. n.a. n.a. n.a. 7.742 11.657 10.899 .077 43.597 100.077
%VAR: -.07 -3.09 --- --- --- --- --- 1.08 -3.70 -3.79 .00 .00
DIFF: -.015 -.152 --- --- --- --- --- .084 -.431 -.413 .000 .000
STDS: 14 305 --- --- --- --- --- 395 12 358 --- ---
STKF: .4101 .1277 --- --- --- --- --- .6779 .4736 .1693 --- ---
STCT: 13489.6 6184.7 --- --- --- --- --- 12137.9 55715.5 8280.2 --- ---
UNKF: .1623 .0325 --- --- --- --- --- .0661 .0746 .0969 --- ---
UNCT: 5338.2 1573.4 --- --- --- --- --- 1183.3 8778.4 4740.9 --- ---
UNBG: 12.2 23.4 --- --- --- --- --- .0 .0 .0 --- ---
ZCOR: 1.3054 1.4633 --- --- --- --- --- 1.1841 1.5045 1.0816 --- ---
KRAW: .3957 .2544 --- --- --- --- --- .0975 .1576 .5726 --- ---
PKBG: 438.14 68.36 --- --- --- --- --- .00 .00 .00 --- ---
But there could be some odd interaction with the oxygen-halogen, but as I said, I can't tell until you send me the MDB file.
john
John
After playing around with a variety of settings, we were able to get PfE to yield a correct composition and get the problematic F-O issue to be a non-issue. As soon as we can, we will send you the mdb file so you can see how the initial problem occurred.
John
Hi John,
I found the problem. Actually two problems.
The clue was that the low totals you were seeing was because you had some elements disabled for quant in the unknown, but not in the standards. So when the program aggregated the intensities for duplicate elements, it found more counts for some of the standards than it did for the unknowns! That was the primary reason for the low totals. See here for a correctly aggregated example of the analysis of your biotite:
St 271 Set 2 Biotite - Wards, Bancroft, ONT, Results in Elemental Weight Percents
ELEM: Na Mg Si Al K Ca Ti O Na Fe Mn Na Mg Si Al P Cl K Ca Ti O F Fe Mn H Li Rb
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC SPEC SPEC
BGDS: LIN AVG AVG AVG AVG LIN LIN LIN EXP AVG AVG EDS EDS EDS EDS EDS EDS EDS EDS EDS EDS EDS EDS EDS
TIME: 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 .00 10.00 10.00 .00 .00 .00 .00 10.00 10.00 .00 .00 .00 .00 10.00 .00 .00 --- --- ---
BEAM: 10.10 10.10 10.10 10.10 10.10 10.10 10.10 10.10 .00 10.10 10.10 .00 .00 .00 .00 10.10 10.10 .00 .00 .00 .00 10.10 .00 .00 --- --- ---
AGGR: 3 2 2 2 2 2 2 2 2 2 --- --- ---
ELEM: Na Mg Si Al K Ca Ti O Na Fe Mn Na Mg Si Al P Cl K Ca Ti O F Fe Mn H Li Rb SUM
XRAY: (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) (ka) () () ()
140 .240 8.680 18.704 5.909 8.450 .002 1.404 37.487 .000 12.967 .254 .000 .000 .000 .000 .102 .000 .000 .000 .000 .000 5.456 .000 .000 .240 .093 .091 100.077
141 .302 8.924 18.519 5.920 8.418 .020 1.420 37.861 .000 13.430 .252 .000 .000 .000 .000 .010 .049 .000 .000 .000 .000 5.148 .000 .000 .240 .093 .091 100.697
142 .289 8.831 18.664 5.792 8.363 .010 1.395 37.017 .000 13.088 .194 .000 .000 .000 .000 .014 .000 .000 .000 .000 .000 4.523 .000 .000 .240 .093 .091 98.605
143 .246 8.622 18.538 5.800 8.476 .017 1.448 37.467 .000 13.239 .169 .000 .000 .000 .000 .029 .037 .000 .000 .000 .000 5.534 .000 .000 .240 .093 .091 100.047
144 .261 8.800 18.612 5.838 8.516 .012 1.438 37.611 .000 13.211 .210 .000 .000 .000 .000 .008 .004 .000 .000 .000 .000 5.224 .000 .000 .240 .093 .091 100.170
145 .280 8.635 18.707 5.735 8.456 .017 1.468 37.443 .000 13.097 .212 .000 .000 .000 .000 .024 .103 .000 .000 .000 .000 5.063 .000 .000 .240 .093 .091 99.665
146 .230 8.658 18.737 5.818 8.393 .004 1.448 37.564 .000 13.301 .131 .000 .000 .000 .000 .031 .091 .000 .000 .000 .000 4.804 .000 .000 .240 .093 .091 99.636
147 .242 8.639 18.883 5.736 8.631 .009 1.360 37.738 .000 13.030 .127 .000 .000 .000 .000 .094 .036 .000 .000 .000 .000 4.973 .000 .000 .240 .093 .091 99.922
148 .172 8.652 18.830 5.813 8.375 .028 1.345 37.528 .000 13.076 .188 .000 .000 .000 .000 .045 .097 .000 .000 .000 .000 5.368 .000 .000 .240 .093 .091 99.942
149 .146 8.824 18.965 5.742 8.463 .030 1.352 37.447 .000 13.100 .224 .000 .000 .000 .000 .013 .052 .000 .000 .000 .000 5.319 .000 .000 .240 .093 .091 100.101
AVER: .241 8.727 18.716 5.810 8.454 .015 1.408 37.516 .000 13.154 .196 .000 .000 .000 .000 .037 .047 .000 .000 .000 .000 5.141 .000 .000 .240 .093 .091 99.886
SDEV: .049 .108 .144 .066 .078 .009 .044 .222 .000 .140 .044 .000 .000 .000 .000 .034 .040 .000 .000 .000 .000 .311 .000 .000 .000 .000 .000 .538
SERR: .016 .034 .046 .021 .025 .003 .014 .070 .000 .044 .014 .000 .000 .000 .000 .011 .013 .000 .000 .000 .000 .098 .000 .000 .000 .000 .000
%RSD: 20.50 1.23 .77 1.13 .93 63.03 3.12 .59 .0000 1.06 22.56 .0000 .0000 .0000 .0000 91.90 84.35 .0000 .0000 .0000 .0000 6.06 .0000 .0000 .00 .00 .00
PUBL: .120 8.520 18.120 5.860 8.120 .006 1.380 40.000 n.a. 13.050 .180 n.a. n.a. n.a. n.a. n.a. .057 n.a. n.a. n.a. n.a. 4.120 n.a. n.a. .240 .093 .091 99.957
%VAR: 100.80 2.42 3.29 -.85 4.11 148.81 2.02 -6.21 .00 .80 8.89 .00 .00 .00 .00 --- -17.60 .00 .00 .00 .00 24.79 .00 .00 .00 .00 .00
DIFF: .121 .207 .596 -.050 .334 .009 .028 -2.484 --- .104 .016 --- --- --- --- --- -.010 --- --- --- --- 1.021 --- --- .000 .000 .000
STDS: 246 900 900 246 217 1108 22 900 0 900 227 0 0 0 0 1108 1108 0 0 0 0 1100 0 0 --- --- ---
STKF: .0623 .0776 .1621 .1014 .1213 .3472 .5481 .1738 .0000 .0654 .4895 .0000 .0000 .0000 .0000 .1596 .0523 .0000 .0000 .0000 .0000 .0674 .0000 .0000 --- --- ---
STCT: 3523.9 3800.3 7176.8 4648.7 3806.2 11905.2 20191.0 3722.5 .0 604.9 4942.5 .0 .0 .0 .0 3136.0 876.4 .0 .0 .0 .0 1239.2 .0 .0 --- --- ---
UNKF: .0011 .0554 .1423 .0395 .0761 .0001 .0122 .1623 .0000 .1126 .0016 .0000 .0000 .0000 .0000 .0003 .0004 .0000 .0000 .0000 .0000 .0147 .0000 .0000 --- --- ---
UNCT: 64.2 2712.7 6301.9 1809.5 2389.1 4.7 449.5 3476.8 .0 1041.2 16.6 .0 .0 .0 .0 5.3 6.6 .0 .0 .0 .0 269.9 .0 .0 --- --- ---
UNBG: 117.0 20.5 13.6 8.8 17.7 21.4 38.8 9.5 .0 9.8 6.9 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0 --- --- ---
ZCOR: 2.1232 1.5746 1.3148 1.4715 1.1102 1.0894 1.1537 2.3117 .0000 1.1687 1.1889 .0000 .0000 .0000 .0000 1.3811 1.1946 .0000 .0000 .0000 .0000 3.5000 .0000 .0000 --- --- ---
KRAW: .0182 .7138 .8781 .3892 .6277 .0004 .0223 .9340 .0000 1.7214 .0034 .0000 .0000 .0000 .0000 .0017 .0075 .0000 .0000 .0000 .0000 .2178 .0000 .0000 --- --- ---
PKBG: 1.55 133.83 467.16 208.08 136.61 1.22 12.63 368.32 .00 108.26 3.44 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 --- --- ---
The second reason was that I was skipping aggregation of EDS elements that are duplicated with WDS elements, but that should be possible to do, so I will fix that code tonight.
Finally, I thought that Julien's recent run was complex but this I think yours is even more complex. Congratulations, John! You get the gold medal grand prize!
john
Continuing with this issue of aggregating duplicate elements when some are disabled for quant and others not...
Here is a simple test run where I acquired Si on two spectrometers for the two NIST mineral glasses. Std 162 (K-411 is the primary standard).
St 160 Set 1 NBS K-412 mineral glass, Results in Elemental Weight Percents
ELEM: Si Si Fe Mg Ca Al Mn O
TYPE: ANAL ANAL SPEC SPEC SPEC SPEC SPEC SPEC
BGDS: LIN LIN
TIME: 10.00 .00 --- --- --- --- --- ---
BEAM: 30.03 .00 --- --- --- --- --- ---
AGGR: 2 --- --- --- --- --- ---
ELEM: Si Si Fe Mg Ca Al Mn O SUM
XRAY: (ka) (ka) () () () () () ()
1 21.108 .000 7.742 11.657 10.899 4.906 .077 43.597 99.986
AVER: 21.108 .000 7.742 11.657 10.899 4.906 .077 43.597 99.986
SDEV: .000 .000 .000 .000 .000 .000 .000 .000 .000
SERR: .000 .000 .000 .000 .000 .000 .000 .000
%RSD: .00 .0000 .00 .00 .00 .00 .00 .00
PUBL: 21.199 n.a. 7.742 11.657 10.899 4.906 .077 43.597 100.077
%VAR: -.43 .00 .00 .00 .00 .00 .00 .00
DIFF: -.091 --- .000 .000 .000 .000 .000 .000
STDS: 162 0 --- --- --- --- --- ---
STKF: .2018 .0000 --- --- --- --- --- ---
STCT: 134.76 .00 --- --- --- --- --- ---
UNKF: .1614 .0000 --- --- --- --- --- ---
UNCT: 107.75 .00 --- --- --- --- --- ---
UNBG: 1.69 .00 --- --- --- --- --- ---
ZCOR: 1.3080 .0000 --- --- --- --- --- ---
KRAW: .7996 .0000 --- --- --- --- --- ---
PKBG: 64.69 .00 --- --- --- --- --- ---
The analysis is correct.
Note that the standard intensity is 134.76 cps/nA. Even though I disabled quant for the 2nd Si channel in the primary standard. This is what the software is supposed to do! The disabled quant channel in the primary standard is temporarily enabled, to allow the proper loading of the primary standard intensity for a sample which has both channels enabled...
Now here is std 162 (K-411) where again, the 2nd Si channel is disabled for quant:
St 162 Set 1 NBS K-411 mineral glass, Results in Elemental Weight Percents
ELEM: Si Si Fe Mg Ca Al Mn O
TYPE: ANAL ANAL SPEC SPEC SPEC SPEC SPEC SPEC
BGDS: LIN LIN
TIME: 10.00 --- --- --- --- --- --- ---
BEAM: 30.00 --- --- --- --- --- --- ---
AGGR: --- --- --- --- --- --- ---
ELEM: Si Si-D Fe Mg Ca Al Mn O SUM
XRAY: (ka) (ka) () () () () () ()
2 25.383 --- 11.209 8.847 11.057 .053 .077 43.558 100.184
AVER: 25.383 --- 11.209 8.847 11.057 .053 .077 43.558 100.184
SDEV: .000 --- .000 .000 .000 .000 .000 .000 .000
SERR: .000 --- .000 .000 .000 .000 .000 .000
%RSD: .00 --- .00 .00 .00 .00 .00 .00
PUBL: 25.382 n.a. 11.209 8.847 11.057 .053 .077 43.558 100.183
%VAR: (.00) --- .00 .00 .00 .00 .00 .00
DIFF: (.00) --- .000 .000 .000 .000 .000 .000
STDS: 162 --- --- --- --- --- --- ---
STKF: .2018 --- --- --- --- --- --- ---
STCT: 67.41 --- --- --- --- --- --- ---
UNKF: .2018 --- --- --- --- --- --- ---
UNCT: 67.41 --- --- --- --- --- --- ---
UNBG: .66 --- --- --- --- --- --- ---
ZCOR: 1.2577 --- --- --- --- --- --- ---
KRAW: 1.0000 --- --- --- --- --- --- ---
PKBG: 102.66 --- --- --- --- --- --- ---
Again the analysis is correct, but note now that the standard counts are 67.41. Why? Because the standard has the 2nd Si channel disabled for quant, so it *doesn't* aggregate the two Si channels. This is what it should do.
john
Now let's add a third(!) Si channel to another spectrometer...
So we standardize using three spectrometers, but when we reanalyze our previous secondary standard (std 160), we see a problem:
St 160 Set 2 NBS K-412 mineral glass, Results in Elemental Weight Percents
ELEM: Si Si Si Fe Mg Ca Al Mn O
TYPE: ANAL ANAL ANAL SPEC SPEC SPEC SPEC SPEC SPEC
BGDS: LIN LIN LIN
TIME: 10.00 .00 .00 --- --- --- --- --- ---
BEAM: 30.02 .00 .00 --- --- --- --- --- ---
AGGR: 3 --- --- --- --- --- ---
ELEM: Si Si Si Fe Mg Ca Al Mn O SUM
XRAY: (ka) (ka) (ka) () () () () () ()
3 30.839 .000 .000 7.742 11.657 10.899 4.906 .077 43.597 109.717
AVER: 30.839 .000 .000 7.742 11.657 10.899 4.906 .077 43.597 109.717
SDEV: .000 .000 .000 .000 .000 .000 .000 .000 .000 .000
SERR: .000 .000 .000 .000 .000 .000 .000 .000 .000
%RSD: .00 .0000 .0000 .00 .00 .00 .00 .00 .00
PUBL: 21.199 n.a. n.a. 7.742 11.657 10.899 4.906 .077 43.597 100.077
%VAR: 45.47 .00 .00 .00 .00 .00 .00 .00 .00
DIFF: 9.640 --- --- .000 .000 .000 .000 .000 .000
STDS: 162 0 0 --- --- --- --- --- ---
STKF: .2018 .0000 .0000 --- --- --- --- --- ---
STCT: 134.76 .00 .00 --- --- --- --- --- ---
UNKF: .2412 .0000 .0000 --- --- --- --- --- ---
UNCT: 161.05 .00 .00 --- --- --- --- --- ---
UNBG: 2.46 .00 .00 --- --- --- --- --- ---
ZCOR: 1.2785 .0000 .0000 --- --- --- --- --- ---
KRAW: 1.1951 .0000 .0000 --- --- --- --- --- ---
PKBG: 66.40 .00 .00 --- --- --- --- --- ---
So what went wrong? Well our secondary standard 160, now has Si on three channels, but our primary std (162), still only has two channels for Si (note that the standard counts are still 134.76). So we have fewer counts on our primary standard when only two Si channels are aggregated, as opposed to three Si channels on our secondary standard (the sample being analyzed).
If we reacquire our primary standard with three Si channels, all is good again.
St 160 Set 2 NBS K-412 mineral glass, Results in Elemental Weight Percents
ELEM: Si Si Si Fe Mg Ca Al Mn O
TYPE: ANAL ANAL ANAL SPEC SPEC SPEC SPEC SPEC SPEC
BGDS: LIN LIN LIN
TIME: 10.00 .00 .00 --- --- --- --- --- ---
BEAM: 30.02 .00 .00 --- --- --- --- --- ---
AGGR: 3 --- --- --- --- --- ---
ELEM: Si Si Si Fe Mg Ca Al Mn O SUM
XRAY: (ka) (ka) (ka) () () () () () ()
3 21.410 .000 .000 7.742 11.657 10.899 4.906 .077 43.597 100.288
AVER: 21.410 .000 .000 7.742 11.657 10.899 4.906 .077 43.597 100.288
SDEV: .000 .000 .000 .000 .000 .000 .000 .000 .000 .000
SERR: .000 .000 .000 .000 .000 .000 .000 .000 .000
%RSD: .00 .0000 .0000 .00 .00 .00 .00 .00 .00
PUBL: 21.199 n.a. n.a. 7.742 11.657 10.899 4.906 .077 43.597 100.077
%VAR: .99 .00 .00 .00 .00 .00 .00 .00 .00
DIFF: .211 --- --- .000 .000 .000 .000 .000 .000
STDS: 162 0 0 --- --- --- --- --- ---
STKF: .2018 .0000 .0000 --- --- --- --- --- ---
STCT: 198.43 .00 .00 --- --- --- --- --- ---
UNKF: .1638 .0000 .0000 --- --- --- --- --- ---
UNCT: 161.05 .00 .00 --- --- --- --- --- ---
UNBG: 2.46 .00 .00 --- --- --- --- --- ---
ZCOR: 1.3070 .0000 .0000 --- --- --- --- --- ---
KRAW: .8116 .0000 .0000 --- --- --- --- --- ---
PKBG: 66.40 .00 .00 --- --- --- --- --- ---
Note that the standard counts are now 198.43 cps/nA.
This I believe is the reason for the problems that John Fournelle saw is his very complicated run. That is depending on whether it's the primary standard, or the secondary standard (or unknown), that has fewer or more duplicate elements acquired, the observed analytical total will either be too high or too low!
Bottom line, the software can handle elements in standards that are disabled (relative to the sample being analyzed), but it can't handle a situation in which the primary standard and another sample have different numbers of duplicate elements!
:o
Edit by John: Or (I should add), the same number of duplicate elements, but not the same exact spectrometers!!!
Man, this gets complicated fast...