Here is my wish list of PENEPMA binary alpha factors that need to be calculated. Presumably this would greatly improve accuracy for low kV work, since the ZAF corrections do not work very well in my experience. Is there a way that I can calculate these easily myself? Do I just need to make .par files? Thanks for any help/advice.
Ni-Ta, Ni-Hf, Ni-Re, Ni-W, Ni-Zr
Al-Hf, Al-Re, Al-W
Ta-Hf, Ta-Re, Ta-W, Ta-Co, Ta-Cr, Ta-Mo, Ta-Zr
Hf-Re, Hf-W, Hf-Co, Hf-Cr, Hf-Mo, Hf-Zr
Re-W, Re-Co, Re-Cr, Re-Mo, Re-Zr
W-Co, W-Cr, W-Mo, W-Zr
Zr-Co, Zr-Cr, Zr-Mo
-Dan
Hi Dan,
You could certainly run these yourself in Standard.exe, but since I'm already all set up to do it, I'll just fire them off myself and let you know as soon as I have them integrated into the matrix.mdb file so you can use them in your low overvoltage analyses in PFE.
john
Thanks John! That would be a tremendous help. Does the matrix.mdb file get updated during a general PFE update?
-Dan
I'm updating the matrix.mdb file about every two months now, but as soon as these new binaries are ready I will update the matrix.mdb file and let you know.
Download this file to see which binaries have been calculated. If a binary that you are interested in has not been calculated please let me know.
The number of calculated Penepma binaries for alpha factor matrix corrections has been proceeding well. Almost half the periodic table has been calculated at normal precision. Work is beginning on double precision calculations which will greatly improve accuracy at low overvoltage situations.
Here is an example of the fast Penepma derived matrix corrections on a measurement of NIST Hastalloy 2402, which while admittedly is somewhat heterogeneous, yields impressive accuracy in a difficult system at 15 keV:
St 651 Set 2 NIST SRM C2402 (Hastelloy C), Results in Elemental Weight Percents
SPEC: Co Si Mn V Cu
TYPE: SPEC SPEC SPEC SPEC SPEC
AVER: 1.500 .850 .640 .220 .190
SDEV: .000 .000 .000 .000 .000
ELEM: Fe Ni Cr Mo W
BGDS: LIN LIN EXP LIN LIN
TIME: 100.00 100.00 100.00 100.00 100.00
BEAM: 29.84 29.84 29.84 29.84 29.84
ELEM: Fe Ni Cr Mo W SUM
207 7.920 52.742 16.357 14.968 4.271 99.658
208 7.665 51.589 16.440 16.132 4.096 99.321
209 7.805 52.359 16.415 15.261 4.208 99.448
210 6.692 46.295 16.082 22.657 4.677 99.803
211 7.604 51.635 16.470 16.253 4.075 99.436
AVER: 7.537 50.924 16.353 17.054 4.265 99.533
SDEV: .488 2.633 .157 3.180 .244 .194
SERR: .218 1.178 .070 1.422 .109
%RSD: 6.48 5.17 .96 18.65 5.72
PUBL: 7.300 51.500 16.150 17.100 4.290 99.740
%VAR: 3.25 -1.12 1.26 -.27 -.57
DIFF: .237 -.576 .203 -.046 -.025
STDS: 526 528 524 542 574
STBE: 1.0000 1.0000 1.0002 1.0016 1.0007
STCT: 6346.7 20768.3 15241.3 8726.4 2235.2
UNBE: .9459 .9846 .9500 1.2690 1.5858
UNCT: 505.7 10739.8 2627.0 1185.5 60.4
UNBG: 19.5 118.8 71.3 14.9 4.7
KRAW: .0797 .5171 .1724 .1359 .0270
PKBG: 26.98 91.45 37.85 80.39 13.83
Ok, so that could be luck! But here's another measurement at 20 keV:
St 651 Set 3 NIST SRM C2402 (Hastelloy C), Results in Elemental Weight Percents
SPEC: Co Si Mn V Cu
TYPE: SPEC SPEC SPEC SPEC SPEC
AVER: 1.500 .850 .640 .220 .190
SDEV: .000 .000 .000 .000 .000
ELEM: Fe Ni Cr Mo W
BGDS: LIN LIN EXP LIN LIN
TIME: 100.00 100.00 100.00 100.00 100.00
BEAM: 29.90 29.90 29.90 29.90 29.90
ELEM: Fe Ni Cr Mo W SUM
212 7.761 52.059 16.630 16.097 4.175 100.121
213 7.574 51.048 16.596 17.276 4.102 99.996
214 7.342 49.414 16.456 18.757 4.209 99.579
215 6.950 47.710 16.403 20.405 4.268 99.137
216 7.256 49.131 16.603 18.563 4.134 99.086
AVER: 7.377 49.873 16.538 18.220 4.177 99.584
SDEV: .310 1.703 .101 1.626 .065 .476
SERR: .139 .761 .045 .727 .029
%RSD: 4.20 3.41 .61 8.93 1.55
PUBL: 7.300 51.500 16.150 17.100 4.290 99.740
%VAR: 1.05 -3.16 2.40 6.55 -2.62
DIFF: .077 -1.627 .388 1.120 -.113
STDS: 526 528 524 542 574
STBE: 1.0000 1.0000 1.0002 1.0016 1.0007
STCT: 13994.5 51175.2 30140.3 12198.0 2796.8
UNBE: .9700 1.0078 .9747 1.3558 1.7182
UNCT: 1064.3 25322.8 5120.5 1655.8 68.2
UNBG: 31.8 204.9 109.1 19.0 5.3
KRAW: .0761 .4948 .1699 .1357 .0244
PKBG: 34.53 124.61 47.92 88.24 14.03
And yet another at 25 keV which is pretty tough for matrix corrections:
St 651 Set 4 NIST SRM C2402 (Hastelloy C), Results in Elemental Weight Percents
SPEC: Co Si Mn V Cu
TYPE: SPEC SPEC SPEC SPEC SPEC
AVER: 1.500 .850 .640 .220 .190
SDEV: .000 .000 .000 .000 .000
ELEM: Fe Ni Cr Mo W
BGDS: LIN LIN EXP LIN LIN
TIME: 100.00 100.00 100.00 100.00 100.00
BEAM: 30.03 30.03 30.03 30.03 30.03
ELEM: Fe Ni Cr Mo W SUM
217 7.848 52.058 16.556 16.219 4.141 100.221
218 7.565 50.339 16.474 17.953 4.270 100.002
219 7.207 48.493 16.428 19.914 4.433 99.875
220 7.243 48.395 16.514 19.730 4.225 99.506
221 7.581 50.382 16.618 17.531 4.144 99.656
AVER: 7.489 49.933 16.518 18.269 4.243 99.852
SDEV: .266 1.526 .073 1.556 .120 .282
SERR: .119 .683 .033 .696 .053
%RSD: 3.55 3.06 .44 8.52 2.82
PUBL: 7.300 51.500 16.150 17.100 4.290 99.740
%VAR: 2.58 -3.04 2.28 6.84 -1.10
DIFF: .189 -1.567 .368 1.169 -.047
STDS: 526 528 524 542 574
STBE: 1.0000 1.0000 1.0002 1.0016 1.0007
STCT: 23100.2 88504.6 46670.5 14785.3 3113.6
UNBE: 1.0002 1.0332 1.0016 1.4466 1.8458
UNCT: 1729.7 42771.5 7706.4 1886.1 71.8
UNBG: 43.9 288.2 147.2 21.0 5.8
KRAW: .0749 .4833 .1651 .1276 .0231
PKBG: 40.46 149.39 53.35 91.01 13.49
In most cases, the precision is worse than the accuracy, and the accuracy is averaging 2-3% or better in a heavily fluorescent system.
I wanted to mention that the latest Matrix.mdb database of k-ratios from Penfluor/Fanal Monte Carlo (fluorescence effects from Fanal) calculations is provided in the recent CalcZAF.msi update (v. 12.8.3) here:
https://probesoftware.com/download/CalcZAF.msi
Be sure to download and install this CalcZAF.msi installer, *before* you run the ProbeforEPMA.msi installer, or you might not get your Matrix.mdb database updated. Once can also directly download this Matrix.mdb database file from here (note, the filename is case sensitive!):
https://probesoftware.com/download/Matrix.mdb
This Matrix.mdb file should be copied to the C:\ProgramData\Probe Software\Probe for EPMA folder, and overwrite your existing file.
There are over 460K Monte Carlo derived k-ratios covering much of the periodic table in here. Seems to me this might make an interesting data visualization project for some bright young person!
The details are found in the paper attached to this post:
https://smf.probesoftware.com/index.php?topic=152.msg8276#msg8276