New Features/Improvements in Standard/CalcZAF

[John Donovan]

A point worth considering is that almost none of the macs we use were actually measured, they are fitted values. So the "range" of applicability does not exist. Equally, low kV analysis is a regime for which the correction algorithms were not developed but that does not prohibit their use, and indeed the phi-rho-z algorithm works well with most attention on the accuracy of mac values at low kV.

Paul

Hi Paul,
Yes, with the welcome exception of the empirical MAC measurements for low energy x-rays by Pouchou and Bastin, etc., whose use in CalcZAF is discussed here:

http://probesoftware.com/smf/index.php?topic=890.msg5699#msg5699

Fortunately, as you mentioned, the rest of the periodic table (except for some emission energies near some absorption edges), typically has significantly smaller MACs for most other emission lines.

So have you ever come across any x-ray measurements at beam energies above 50 keV?
john

[Paul Carpenter]

Just following up on the Quantitative Microanalysis using WDS and EDS course at Lehigh Microscopy School this year 2017. We use CalcZAF for discussion and demonstration of the correction parameters for both bulk and particle analysis. The entry formats for formula, weight percent, and standard database are excellent ways to conveniently enter compositions to then show correction parameters for various compounds.

We also use the CalcZAF input file to show how one can walk through the measured k-ratios for the Cu-Au as well as Si-Ir measurements discussed in the course.

The comparison of mac values from the different data sets is very useful as well.

Finally, the log window output includes the mac value for the sample, f-of-chi (i.e., f(x)), and the components of the ZAF correction. We also compare C with calculated k as an interpretation of the ZAF factor. The "use all correction algorithms" checkbox is used to effectively summarize the two measurements for the Si-Ir alloy that clearly shows the need to have a standard similar to the sample for accurate quantitative analysis. It is probably the most important demonstration in the course, and is important for both general microanalysis and those who use standardless EDS.

When I talk about software tools for the microanalyst, I usually discuss Casino, then CalcZAF, then DTSA-II, GMRfilm, and finally Penelope. These are all excellent tools that can be used to quickly highlight problems and solutions in EPMA.

Cheers,

Paul Carpenter

[Probeman]

When I talk about software tools for the microanalyst, I usually discuss Casino, then CalcZAF, then DTSA-II, GMRfilm, and finally Penelope. These are all excellent tools that can be used to quickly highlight problems and solutions in EPMA.

Hi Paul,
Glad to hear Lehigh went well this year with two concurrent probe sessions (one local and one remote)!   

Also it's good to know what free, downloadable applications are most useful for the community. If you have any additional comments you want to add about available EPMA/SEM applications please feel free to add your thoughts here also:

http://probesoftware.com/smf/index.php?topic=927.0

I think you had mentioned some ImageJ or Matlab? image processing scripts or macros at EMAS that you have found useful?
john

[John Donovan]

We just realized today that when the web site switched over to https for secure connections, the Help | Update CalcZAF code in CalcZAF will no longer work for automatic updating from the application.  So, just as was the case for Probe for EPMA, one has to manually download the latest CalcZAF installer from this link:

https://probesoftware.com/download/CalcZAF.msi

Then run the installer once to get the new https: compatible secure download version.  After this, one can again simply use the Help | Update CalcZAF menu to update the application as usual.

In addition, we also realized that the "Interactive Help" buttons as described here by Karsten Goemann:

https://probesoftware.com/smf/index.php?topic=67.0

were also broken by the new https:// secure connection, so updating CalcZAF to the latest version (12.4.5) will fix all these button links. 

Again, if you have Probe for EPMA you can update both CalcZAF and Standard simply using the Help | Update Probe for EPMA menu in PFE (if you've already updated to the "secure download" version of PFE).

Please let us know if you have any questions at all about this process.

[John Donovan]

Recently Ben Hanson at Corning Glass asked why the MACJTA.DAT mass absorption coefficients table did not contain MACs for emitters less than 1 keV. Armstrong's tabulated values which are found in the MACMATK.DAT, MACMATL.DAT and MACMATM.DAT text files show that he tabulated emitters down to boron.

The MACJTA.DAT binary file that I generated in 2008 does not contain emitters of energy less than 1 keV which was because there was a line of code from the McMaster MAC fit code that I did not modify when I added Armstrong's FRAME equations.

So I changed the code to only skip emission lines with energies less than 0.1 keV when calculating MACs using Armstrong's FRAME equations, which subsequently allows Ka emitters down to boron.

I really don't think any one should be using MACs generated using the FRAME fit equations. My impression is that the default MAC table (LINEMU.DAT) is more than adequate, unless one is quantifying very low energy emission lines, in which case one should really be utilizing empirically measured MACs as listed in the Empirical MACs menu in CalcZAF and Probe for EPMA.

In any case, I've attached the newly generated MACJTA.DAT binary file below (remember to login to see attachments), and if you want, this new file should be copied to your C:\ProgramData\Probe Software\Probe for EPMA folder.  Note that the ProgramData folder is hidden by default in Windows, so you'll need to "unhide" it using the Folder Options in Windows Explorer if it is not visible on your computer.

But just for general interest here is a comparison of some emitter - absorber pairs as generated from CalcZAF (from the X-ray menu) using the new MACJTA.DAT file attached below:

MAC value for O Ka in Si =    8155.72  (LINEMU   Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for O Ka in Si =    8790.00  (CITZMU   Heinrich (1966) and Henke and Ebisu (1974))
MAC value for O Ka in Si =        .00  (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for O Ka in Si =    8063.47  (MAC30    Heinrich (Fit to Goldstein tables, 1987))
MAC value for O Ka in Si =    8669.79  (MACJTA   Armstrong (FRAME equations, 1992))
MAC value for O Ka in Si =    7544.10  (FFAST    Chantler (NIST v 2.1, 2005))
MAC value for O Ka in Si =    8790.00  (USERMAC  User Defined MAC Table)

Note that the McMaster value is zero because those equations are only designed for photons greater than 1 keV.

MAC value for Mg ka in Fe =    5239.40  (LINEMU   Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for Mg ka in Fe =    6120.70  (CITZMU   Heinrich (1966) and Henke and Ebisu (1974))
MAC value for Mg ka in Fe =    5395.10  (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for Mg ka in Fe =    5518.65  (MAC30    Heinrich (Fit to Goldstein tables, 1987))
MAC value for Mg ka in Fe =    6089.59  (MACJTA   Armstrong (FRAME equations, 1992))
MAC value for Mg ka in Fe =    5089.56  (FFAST    Chantler (NIST v 2.1, 2005))
MAC value for Mg ka in Fe =    5522.00  (USERMAC  User Defined MAC Table)

By the way, I do not know why the MAC values generated from Armstrong's FRAME equations are slightly different from his tabulated values (in CITZMU.DAT).

MAC value for Na Ka in Mg =     814.41  (LINEMU   Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for Na Ka in Mg =     770.10  (CITZMU   Heinrich (1966) and Henke and Ebisu (1974))
MAC value for Na Ka in Mg =     884.35  (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for Na Ka in Mg =     811.13  (MAC30    Heinrich (Fit to Goldstein tables, 1987))
MAC value for Na Ka in Mg =     783.81  (MACJTA   Armstrong (FRAME equations, 1992))
MAC value for Na Ka in Mg =     746.84  (FFAST    Chantler (NIST v 2.1, 2005))
MAC value for Na Ka in Mg =     810.00  (USERMAC  User Defined MAC Table)

But here is where the "wheels come off" for the FRAME equations (apparently the FRAME equations from Armstrong were subsequently improved by Armstrong in his MACMAT*.DAT files tabulation):

MAC value for C ka in Ag =    8178.73  (LINEMU   Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for C ka in Ag =    5507.00  (CITZMU   Heinrich (1966) and Henke and Ebisu (1974))
MAC value for C ka in Ag =        .00  (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for C ka in Ag =    5545.52  (MAC30    Heinrich (Fit to Goldstein tables, 1987))
MAC value for C ka in Ag =   43402.74  (MACJTA   Armstrong (FRAME equations, 1992))
MAC value for C ka in Ag =    5809.43  (FFAST    Chantler (NIST v 2.1, 2005))
MAC value for C ka in Ag =    8190.00  (USERMAC  User Defined MAC Table)

Again I suggest staying with the default (LINEMU.DAT) values from Henke for most work.

[John Donovan]

Philipp Poeml recently pointed out to us that an error is given in CalcZAF when attempting to edit the "additional" x-ray line energies in the default x-ray line database.  This has now been fixed and is ready for updating.

Update CalcZAF by simply using the Help | Update CalcZAF menu. If you have Probe for EPMA, simply use the Help | Update Probe for EPMA menu, and all will be automatically updated.

This error occurred only when editing the default x-ray line and/or the default x-ray fluorescent yield databases using these menus in CalcZAF:



*and* when the x-ray line being edited was one of the "additional" x-ray lines, e.g., Ln, Lg, Lv, Ll, Mg, or Mz emission lines.

The quantification of these "additional" x-ray lines was working fine, just the edit code was not quite right.  Anyway, all fixed now!   

[John Donovan]

This is a tiny change, but we recently modified the CalcZAF Export Format (from the Probe for EPMA Analyze! window right click menu) to export the full sample type, name and line number (if not outputting the intensity averages), so when the data is loaded into CalcZAF using the File | Open menu, the full sample name/line number shows as seen here:



This can be useful in some circumstances, e.g., for some reason one is proccesing multiple data lines in CalcZAF exported from PFE.

[John Donovan]

Based on a suggestion by Brian Joy in the linked post here:

https://probesoftware.com/smf/index.php?topic=508.msg9243#msg9243

we first added some additional parameter output in CalcZAF (and PFE in DebugMode). First in the normal CalcZAF intensity to concentration output we added (an additional column of, as you will soon see why) the assigned standard numbers just before the unk/std matrix factor ratios:

Chromite from Raw K-ratios-elemental

STANDARD PARAMETERS (TOA= 40):

 ELEMENT  STDNUM STDCONC STDKFAC   Z-BAR  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR                 
   Fe Ka     895  72.080   .6781 20.9954   .9969  1.0000  1.0662  1.0629
   Cr Ka      24  68.337   .6400 18.9506   .9962   .9999  1.0719  1.0678
   Ti Ka      22  59.939   .5547 16.3920   .9950  1.0000  1.0861  1.0806
   Al Ka      13  52.926   .4353 10.6462  1.1676  1.0000  1.0412  1.2157
   Mn Ka      25  77.446   .7341 21.1658   .9973  1.0000  1.0578  1.0549

 ELEMENT STP-POW BKS-COR   F(x)e   F(x)s      Eo      Ec   Eo/Ec
   Fe Ka  1.0939   .9747   .9846   .9877   15.00  7.1120  2.1091
   Cr Ka  1.1038   .9712   .9799   .9837   15.00  5.9900  2.5042
   Ti Ka  1.1251   .9653   .9720   .9770   15.00  4.9670  3.0199
   Al Ka  1.0604   .9819   .8874   .7600   15.00  1.5600  9.6154
   Mn Ka  1.0808   .9787   .9827   .9854   15.00  6.5390  2.2939


SAMPLE: 1, TOA: 40, ITERATIONS: 3, Z-BAR: 17.53337

 ELEMENT  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR STP-POW BKS-COR   F(x)u      Ec   Eo/Ec    MACs  STDNUM uZAF/sZAF
   Fe ka  1.0235  1.0000  1.1012  1.1271  1.1454   .9614   .9620  7.1120  2.1091 175.948     895    1.0604
   Cr ka   .9993   .9530  1.0946  1.0425  1.1351   .9643   .9806  5.9900  2.5042 82.3870      24    .97631
   Ti ka  1.0058   .8822  1.0859   .9634  1.1191   .9703   .9665  4.9670  3.0199 135.747      22    .89154
   Al ka  1.6720   .9995   .9812  1.6398   .9537  1.0288   .5307  1.5600  9.6154 2436.99      13    1.3488
   Mn ka   .9973   .9958  1.1182  1.1104  1.1618   .9625   .9853  6.5390  2.2939 64.2586      25    1.0526

 ELEMENT   K-RAW K-VALUE ELEMWT% OXIDWT% ATOMIC% FORMULA KILOVOL                                       
   Fe ka  .27180  .18431  20.774   -----  10.316    .241   15.00                                       
   Cr ka  .47690  .30522  31.818   -----  16.971    .397   15.00                                       
   Ti ka  .00550  .00305    .294   -----    .170    .004   15.00                                       
   Al ka  .10830  .04715   7.731   -----   7.947    .186   15.00                                       
   Mn ka  .00200  .00147    .163   -----    .082    .002   15.00                                       
   Mg                      6.290   -----   7.177    .168
   V                        .122   -----    .066    .002
   O                      33.040   -----  57.270   1.340
   TOTAL:                100.232   ----- 100.000   2.340

And then we added the separate unk/std matrix factors for the absorption, fluorescence and and atomic number factors as seen here:

SAMPLE: 1, TOA: 40, ITERATIONS: 3, Z-BAR: 17.53337

 ELEMENT  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR STP-POW BKS-COR   F(x)u      Ec   Eo/Ec    MACs  STDNUM uZAF/sZAF   uA/sA   uF/sF   uZ/sZ
   Fe ka  1.0235  1.0000  1.1012  1.1271  1.1454   .9614   .9620  7.1120  2.1091 175.948     895    1.0604  1.0267  1.0000  1.0328
   Cr ka   .9993   .9530  1.0946  1.0425  1.1351   .9643   .9806  5.9900  2.5042 82.3870      24    .97631  1.0031   .9531  1.0212
   Ti ka  1.0058   .8822  1.0859   .9634  1.1191   .9703   .9665  4.9670  3.0199 135.747      22    .89154  1.0108   .8822   .9998
   Al ka  1.6720   .9995   .9812  1.6398   .9537  1.0288   .5307  1.5600  9.6154 2436.99      13    1.3488  1.4320   .9995   .9424
   Mn ka   .9973   .9958  1.1182  1.1104  1.1618   .9625   .9853  6.5390  2.2939 64.2586      25    1.0526  1.0000   .9958  1.0571

 ELEMENT   K-RAW K-VALUE ELEMWT% OXIDWT% ATOMIC% FORMULA KILOVOL                                       
   Fe ka  .27180  .18431  20.774   -----  10.316    .241   15.00                                       
   Cr ka  .47690  .30522  31.818   -----  16.971    .397   15.00                                       
   Ti ka  .00550  .00305    .294   -----    .170    .004   15.00                                       
   Al ka  .10830  .04715   7.731   -----   7.947    .186   15.00                                       
   Mn ka  .00200  .00147    .163   -----    .082    .002   15.00                                       
   Mg                      6.290   -----   7.177    .168
   V                        .122   -----    .066    .002
   O                      33.040   -----  57.270   1.340
   TOTAL:                100.232   ----- 100.000   2.340

The reason for the extra std number column is that these separate A, F and Z factors are only output when the user checks the "Verbose Mode" menu (in the Output menu).  And when that menu is checked there is a lot more output so the std numbers in the standard section scroll off the screen.

In Probe for EPMA, if you want to see these matrix factors you will need to first check the DebugMode menu (again from the Output menu) for the unk/std matrix factor ratios, and also the VerboseMode menu for the separate matrix factor ratios.

Hope that is good for everyone. 

[John Donovan]

We recently made a small improvement in the output for Penepma calculations.

Now after you have performed 1 or more Penepma simulation(s) you can output an array of continuum intensities at 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 keV energies for all Penepma sub folders (from running Penepma in "batch" mode).

These right now are an average of 5 channels (2 * 2 + 1) at the 1 keV intervals, and the program will ask if you want to output generated or emitted continuum intensities using this new button accessible from the Batch Mode button in the Penepma GUI:



Then you will get an output file with the average intensities and average variances that can be imported into Excel or whatever.



And of course average Z bars calculated using mass and Z fractions.

[John Donovan]

Continuing our discussion of new features in Probe for EPMA added since early summer, we now have a new effective takeoff angle estimator dialog in the CalcZAF application as shown here:



This dialog in CalcZAF can be used to derive the "effective" takeoff angle on one's WDS spectrometers when there is an apparent disagreement between the various spectrometers as shown here:



And discussed previously in several topics:

https://probesoftware.com/smf/index.php?topic=1466.msg11339;topicseen#msg11339

https://probesoftware.com/smf/index.php?topic=1466.msg11535;topicseen#msg11535

https://probesoftware.com/smf/index.php?topic=1535.msg11937#msg11937

The dialog itself is quite simple:



An example of these effective takeoff calculations are shown here:



This new dialog is an attempt to more easily determine one's effective takeoff angle for various spectrometers instead of simply trying different takeoff angles in CalcZAF as shown here:



Of course the main issue here is accuracy because one is comparing measured k-ratios to a theoretical k-ratio. Therefore, such issues as differences in carbon coating, oxidation layers, hydrocarbon contamination, can become quite important.  Especially since to obtain the best sensitivity one would attempt to measure k-ratios that are highly affected by absorption...

Finally, once one has determined the actual "effective" takeoff angles for each spectrometer and crystal combination, one can then edit the SCALERS.DAT file as described here:

https://probesoftware.com/smf/index.php?topic=40.msg12018#msg12018

in order to take advantage of these new takeoff angle calibrations.  The use of these effective takeoff angles in the SCALERS.DAT file in the Probe for EPMA absorption correction, won't affect your k-ratios , but it should allow you to obtain consistent concentrations across multiple spectrometers and crystals.

Pretty cool feature, I'd say...   😎

[Probeman]

I note that there seems to be a new capability in CalcZAF to calculate intensities (k-ratios) from concentrations using all matrix corrections:



That is to say, this option used to be disabled unless one was calculating concentrations from intensities.

This could be useful in teasing out what is the "correct" k-ratio when evaluating effective take off angles for our various spectrometers:

https://probesoftware.com/smf/index.php?topic=1569.0

[John Donovan]

As many of you know, we distribute the UofO EPMA standard database as the default standard composition database with CalcZAF (and Standard). 

Of course you can make your own (default) standard composition database by importing the standards from your JEOL or Cameca instrument using the Standard application to first create a new standard.mdb file (see attached pre-installation pdf, appendix A for Cameca and appendix B for JEOL) and importing your standard compositions, or you can enter new standard compositions "from scratch".

Of course if you create a new standard.mdb file you will lose access to the UofO standard compositions, so we have now added a copy of that standard database file to the CalcZAF distribution starting with v. 13.6.6. The copy of the UofO standard database is called (appropriately enough!): standard_UofO.mdb.