Probe Software Users Forum
Software => Probe for EPMA => Topic started by: Mike Spilde on July 01, 2013, 12:55:59 PM
-
Hi PFE users,
John asked me to post my questions to the new users forum. I want to use the MAN background correction while measuring REE-bearing pegmatite minerals. I've used the off-peak backgrounds in the past on the JEOL, but for the entire suite of REEs and associated elements (32 total elements in the file), it is extremely difficult to get all the background positions correct. The MAN background method seems like it would be just the ticket. I'm not worried about trace levels, just the general level of REE (HREE vs LREE for example). So, I have some questions that I hope you could help with.
1) Has anything been published on this method?
2) The reference manual says the method is not good for high atomic number samples. Why not? Has anyone tested it on REE analyses? My samples are in the range of Z=30-60.
3) Paul showed me how to assign MAN fits using my calibration analyses, but as I am setting up the MAN assignments, what do I look for in terms of the fits? I have a number of standards calibrated and there is a good deal of scatter on some elements, so I take some standards out to get less scatter. What is considered a good intercept? I assume that approaching zero is good but a negative is not? How about the Rel% Deviation? I assume low is better but how low? 3 or less? 10 or less?
4) Does one always want to force a straight-line fit?
5) Are the values reassigned with each new file and calibration? If I copy a previous file but use a new calibration, are the values reassigned or are the previous values used?
Any advice would be appreciated.
Thanks, Mike
-
Hi Mike,
Thanks. I think others could benefit from my answers to your excellent questions. Therefore...
Hi PFE users,
John asked me to post my questions to the new users forum. I want to use the MAN background correction while measuring REE-bearing pegmatite minerals. I've used the off-peak backgrounds in the past on the JEOL, but for the entire suite of REEs and associated elements (32 total elements in the file), it is extremely difficult to get all the background positions correct. The MAN background method seems like it would be just the ticket. I'm not worried about trace levels, just the general level of REE (HREE vs LREE for example). So, I have some questions that I hope you could help with.
1) Has anything been published on this method?
The original paper is here:
http://probesoftware.com/Improved%20MAN%20%28Jour.%20Micros.%20Microa.,%201996%29.pdf
But you should contact Karstem Goemann who presented exactly what you want to do (MAN of REE minerals) at M&M in Nashville.
2) The reference manual says the method is not good for high atomic number samples. Why not? Has anyone tested it on REE analyses? My samples are in the range of Z=30-60.
Yes, I didn't try to "push the envelope" and limited myself to oxides and silicates but Karsten has used this for REE bearing minerals and Dave Wark also with excellent results on high Z super alloys (also presented at the same M&M). You might want to get their abstracts at least.
3) Paul showed me how to assign MAN fits using my calibration analyses, but as I am setting up the MAN assignments, what do I look for in terms of the fits? I have a number of standards calibrated and there is a good deal of scatter on some elements, so I take some standards out to get less scatter. What is considered a good intercept? I assume that approaching zero is good but a negative is not? How about the Rel% Deviation? I assume low is better but how low? 3 or less? 10 or less?
Rule of thumb : the lowest measured intensity (at a given average atomic number) is by definition the background. Since the intensities are corrected for continuum absorption the only thing that should be problematic are interferences and/or contamination and both of these issues will produce anomalously high intensities. So deselect the stds that give higher intensities and click the Update Fit button.
4) Does one always want to force a straight-line fit?
I only use that option if I have three points with two very close together so it gives a weird 2nd order fit. Or for two point fits obviously.
5) Are the values reassigned with each new file and calibration? If I copy a previous file but use a new calibration, are the values reassigned or are the previous values used?
It depends if you also loaded the standard intensities or not when loading a File Setup. If you don't load the standard (and MAN std) intensities, the assignements will be there, but no data. The standard intensities must be reacquired.
But if you *do* load the std intensities, you can simply reacquire the most important stds, usually for the major elements. The backgrounds usually do not change significantly for obvious reasons.
-
Although I have not used MAN for REE analysis I have used it for high Z on low Z with no trouble. This was not trace but around 1% Pt and 1% Pd on Alumina.
-
Although I have not used MAN for REE analysis I have used it for high Z on low Z with no trouble. This was not trace but around 1% Pt and 1% Pd on Alumina.
Hi Gareth,
Yes, this is still a low Z analysis for the unknown, but did you use the MAN also for the Pt and Pd standards?
A feature I don't use often enough is the fact that you can specify different bgds for the std and unk. For example, you could use the MAN bgd for your Al2O3 samples since the Z range is very limited and use off-peaks for the Pt and Pd stds which would be a pain to acquire a wide range of Z MAN stds for.
Of course the bgd on a trace element std (i.e., pure element) isn't very important from a statistical consideration. So in that sense it don't matter especially, but off-peaks on the Pt and Pt metal stds would be quick and easy. I like quick and easy. :)
john
-
Hi John,
does this mean --- and do you think it would make sense --- if I would use the same approach? Like, using offpeak backgrounds for PuO2, UO2, whatever stadards, but then measure on the sample with plenty of REEs and actinides using the MAN backgrounds? That might be very interesting to try.
Using MAN for standards and unknown did not work out.
Cheers
Ph
-
Hi Philipp,
Yes, you have a special situation with actinides and all.
And I don't think this feature of using off-peak backgrounds on stds and MAN bgds for unkns would help you much because your unknowns are pretty much like your stds. That is both your stds and unks are very, very high Z!
For Gareth's situation, his samples are almost pure Al2O3 so very low Z compared to his Pt and Pd stds. These samples are ideal for this bgd feature.
-
All, here is the link to the abstract of my M&M2011 presentation that John was referring to:
http://dx.doi.org/10.1017/S143192761100376X (http://dx.doi.org/10.1017/S143192761100376X)
Let me know if you can't access the abstract and I'll send you a copy.
There's not much about MAN in the abstract and it was also only part of the presentation, the larger part was on using multiple backgrounds to analyse Y, REE and others in uranium minerals like uraninite and coffinite.
The MAN data I presented there was for analysing florencite (a beam sensitive & water bearing REE-Al-phosphate) which only has a MAN of around 22. It worked very well for lower Z elements, still reasonably well for LREE, but not so well for HREE, see attached file. The graph shows an average for around 20 florencite measurements which were acquired using off-peak backgrounds, and comparing those results with the results using MAN backgrounds for the same dataset. The red curve is the "original" and the purple one the final "best" curve after including water by difference and using linear MAN fits.
This plot will show problems with both off-peak and MAN backgrounds. The Al deviation is due to limitations with off-peak positioning, potentially something similar for Ce, and La, but these are all major elements in florencite, so the effect is marginal and I did not investigate further.
The HREE are below detection limit, and the absolute values using MAN were around -0.1wt% for Dy and Er, and -0.05wt% for Yb, which is a bit of a concern. Back then I didn't have the time to sort this out for those particular samples. I tried to use conditions providing stable intensities, but the HREE were measured at the end of the program so there still might have been a issue with damage, sub-surface charging... If that's the case it's strange though didn't affect the later elements on the other spectrometers. One obvious thing to try would be to change the measurement order. The grains were too small to defocus further than 10µm.
Another issue is standards, not only secondary standards, but I also only have the Smithsonian orthophosphates for REE calibration at the moment. It is well known that their LREE standards (La to Gd) contain varying amounts of Pb (see http://dx.doi.org/10.6028/jres.107.056 (http://dx.doi.org/10.6028/jres.107.056)), so Pb needs to be individually determined and corrected in the reference composition, as far as that is possible as concentrations can vary within chips. Some of them contain minor amounts of other REE as well. John already said that both interferences and small amounts of the element in question will push the datapoint up in the MAN plots, so it is better to use the lower values, but if the reference composition of a standard is not correct it can go in any direction. I learned a lot about our standards looking at MAN plots (in terms of reference values, purity, contamination) and many of them are not as well characterised or "clean" as I'd like them to be.
I have since then used MAN to analyse monazite with good results, again majors & minors only, no traces! I'll see if I can put a few more details together and post them. Again, one problem is method validation, as I only have monazite "age" standards, none with a well constrained composition.
Cheers, Karsten
-
There is a simple method to compare the same intensity data using off-peak, MAN and Nth point background corrections.
First acquire a set of data for your elements of interest on some standards using normal off-peak backgrounds. Because you will also be fitting a MAN calibration curve, one also needs to acquire intensity data on standards that *do not* contain the elements of interest. For silicate elements one might select, in addition to the primary stds, standards such as MgO, SiO2, Al2O3, TiO2 and NiO, etc. Again, using full off-peak backgrounds.
Once the intensity data is acquired one can then perform quant analyses on the standards using the acquired off-peak background corrections.
Next, go to the Analytical menu and check the Use Off-Peak Elements For MAN Fit menu as seen here:
(https://probesoftware.com/smf/oldpics/i41.tinypic.com/6dvbc7.jpg)
Now fit your MAN backgrounds as usual using the Analytical | Assign MAN Fits menu, but note that the software will automatically utilize off-peak acquired element intensities for the MAN fit by ignoring the off-peak data for the standards and fit only the on-peak intensities!
Next go to the Analytical menu and check the Use MAN Correction For Off-Peak Elements menu. This tells the software to force off-peak measured elements to utilize the MAN calibration curve for the background correction as seen here:
(https://probesoftware.com/smf/oldpics/i44.tinypic.com/zn8s5l.jpg)
Now you can analyze your standards that were acquired using off-peak backgrounds, but now using the MAN calibration curve for comparison.
Finally go to the Analytical | Analytical Options dialog and select the Use Nth Point Calculation For Off-Peak Intensities as shown here:
(https://probesoftware.com/smf/oldpics/i39.tinypic.com/voy7ww.jpg)
Now uncheck the Analytical menu | Use MAN Correction For Off-Peak Elements menu and now you can analyze your off-peak acquired standards using *only* the off-peak intensities from the first point of every sample.
This way one can acquire and compare the same data set using off-peak, MAN and Nth Point background corrections. A white paper with more details is attached below. Remember that one needs to be logged in to see attachments!
-
Maybe this would be a good time to examine the method of Mean Atomic Number (MAN) background correction in more detail.
The example I am utilizing is from Paul Carpenter at Washington University (used with permission) and Paul acquired this MAN calibration data set for use in quantitative x-ray mapping with Probe Image and CalcImage as can be seen in the beautiful quant x-ray maps attached below. Note the stoichiometric *and* total quant maps which can be automatically generated and output with CalcImage.
Of course the MAN background correction is good to a few hundred PPM in silicates and oxides so it really is "overkill" for quant x-ray mapping, but it sure does save a lot of time because one doesn't need to acquire any off-peak x-ray maps!
So, what is the basis for the the MAN correction? The details are available in this publication:
http://probesoftware.com/Improved%20MAN%20%28Jour.%20Micros.%20Microa.,%201996%29.pdf
but it suffices to say that it is based on Kramer's Law, where in the absence of the element under investigation, the on-peak x-ray production of the material can be used to calibrate the continuum background. The MAN correction also contains an adjustment for absorption of the continuum of the emission line being investigated and for the matrix in question (both for the MAN calibration standards and the unknowns). Since the standard compositions are already known, the calculation of mean atomic number (MAN) is easily performed. In the case of unknowns, the mean atomic number of the sample is calculated during the ZAF or phi-rho-z matrix iteration.
So now let us examine the first element measured in these x-ray maps, Na ka as seen here in the Analytical | Assign MAN Fits menu available in both Probe for EPMA and CalcImage:
(https://probesoftware.com/smf/oldpics/i44.tinypic.com/2eek67n.jpg)
Of course, we can immediately see one problem, and that is that the NBS CuZn alloy is very much an outlier in our Na MAN fit. Why is that? Well it's because of our old friend, the Zn La on Na interference. Now, this interference could be corrected for in PFE- but let's not get carried away! There are plenty of other MAN standards to fit for Na, and besides, we aren't measuring materials with that high an average Z anyway!
So let's unselect that standard from our MAN fit by using a <ctrl> click of the mouse and then we click the Update Fit button to re-plot the fit as seen here.
(https://probesoftware.com/smf/oldpics/i43.tinypic.com/al0upx.jpg)
Note the now much more reasonable fit to the on-peak x-ray intensities from our standards, which is loaded by default based on whether the standard contains the element under investigation. Since the standard composition database indicated no Na in the NBS CuZn alloy, the software loaded that standard intensity. But for other standards that actually contain Na, for example, Kakanui Hornblende, the x-ray intensities for that standard material are not loaded for the default MAN fit.
To demonstrate the necessity of the continuum absorption correction, we can unselect the correction for continuum absorption by unchecking the checkbox and the significantly worse fit is quite evident. Why is this? Because every standard composition absorbs continuum radiation (i.e., at the characteristic energy of our elements of interest) in somewhat different degrees, depending on the physics details- that is why we have absorption edges in our EDS continuum spectra!
(https://probesoftware.com/smf/oldpics/i43.tinypic.com/2cgkmtz.jpg)
So clearly the correction for continuum absorption is necessary, as seen also in this plot of Mg ka in standards that do *not* contain Mg:
(https://probesoftware.com/smf/oldpics/i39.tinypic.com/91bfiq.jpg)
By including the correction for continuum absorption for Mg ka we are back in business as seen here:
(https://probesoftware.com/smf/oldpics/i42.tinypic.com/2vt41e9.jpg)
Don't worry, the program defaults all these options automatically! 8)
So, let's move on to a different situation as seen here for Mn ka:
(https://probesoftware.com/smf/oldpics/i44.tinypic.com/2w5m3h4.jpg)
Here, again, we see some outliers in our MAN fit, but these are *not* from spectral interferences from other elements, as was the case for Zn La on Na. These are standards which contain small amounts of Mn that have *not* been specified in the standard composition database. Paul puts it himself this way:
"1039 is our wollastonite standard that contains a low concentration of Mn but has not been characterized in the standard analysis yet. 1713 is Elba hematite that is also being adjusted, and the synthetic fayalite 465 also contains very low Mn. On a wavelength scan this level of Mn is not apparent."
By simply unselecting these standards which contain trace amounts of Mn, we can get a more accurate model of the background for our Mn measurements. Remember: background is by definition the lowest intensity one can measure! Interferences and "contamination" will all cause elevated outliers to the fit. Of course there are some subtle continuum artifacts as documented here:
http://probesoftware.com/Ti%20in%20Quartz,%20Am.%20Min.%20Donovan,%202011.pdf
but those are in the sub 100 PPM level! So not exactly significant for quant-xray mapping even when measuring off-peak images!
So after removing those standards from our fit which contain "undocumented" Mn, we obtain the following fit:
(https://probesoftware.com/smf/oldpics/i39.tinypic.com/261nntf.jpg)
For an element such as Si or Al we obtain beautiful fits using the default MAN assignments made automatically by the program as seen here:
(https://probesoftware.com/smf/oldpics/i44.tinypic.com/2k545v.jpg)
Now check out the gorgeous quant x-ray maps Paul acquired which are automatically corrected for background, matrix and spectral interferences in the attachments below!
It should also be mentioned that at zero concentration levels, using MAN background acquisition improves measurement precision by roughly the Sqr(2) and at the same time cuts the necessary acquisition time by 50%.
See the attachments posted here for details:
http://probesoftware.com/smf/index.php?topic=29.msg237#msg237
-
There is a post on using the MAN correction in conjunction with the blank correction in Probe for EPMA on trace element characterization (for simple matrices such as SiO2, TiO2, ZrSiO4), where a suitable zero or non-zero blank material is available here:
http://probesoftware.com/smf/index.php?topic=29.msg237#msg237
The idea is is that one can obtain excellent precision with the MAN background correction because it is the average of multiple standard points, for example, 5 measurements at 30 seconds each for each standard. If one is utilizing 3 or 4 or more standards per MAN fit (element), then the precision is excellent, much better in fact than counting off-peak for the same amount of time.
To illustrate this see the graph below which compares MAN to off-peak to Nth point off-peaks (where the off-peak is utilized only on the first point and re-used for the subsequent points) using the same data set for all calculations as described above:
(https://probesoftware.com/smf/oldpics/i43.tinypic.com/2eb4e4l.jpg)
The Cr off-peak data shows more variance due to the fact that we are subtracting two measurements from each other, while the MAN correction is essentially a constant at a given average atomic number which should not change much for a say SiO2 matrix. Of course the beauty of the MAN correction (unlike the Nth point method) is that it *can* account for changes in the matrix (and therefore background intensity) automatically without measuring the off-peaks! Here's another example:
(https://probesoftware.com/smf/oldpics/i43.tinypic.com/sz7g3b.jpg)
The full report can be found attached to this post:
http://probesoftware.com/smf/index.php?topic=4.msg189#msg189
-
I'm running some quantitative x-ray maps with a student on some melt inclusions to see if we can detect any variation in the glass chemistry and in setting up the standards I decided to acquire all ten elements using MAN backgrounds. Here is a sample of a standard and the amazing accuracy of the MAN correction (no off-peak measurements) even for trace elements:
St 305 Set 1 Labradorite (Lake Co.)
TakeOff = 40.0 KiloVolt = 15.0 Beam Current = 30.0 Beam Size = 5
(Magnification (analytical) = 20000), Beam Mode = Analog Spot
(Magnification (default) = 2524, Magnification (imaging) = 736)
Image Shift (X,Y): 50.00, .00
Specimen from HR Wenk, UC Berkeley, Geology
See Wenk and Kroll, Bull. Mineral. (1984), 107, 467-487
Number of Data Lines: 5 Number of 'Good' Data Lines: 5
First/Last Date-Time: 12/17/2013 04:08:12 PM to 12/17/2013 04:17:53 PM
WARNING- Using Time Dependent Intensity (TDI) Element Correction
Average Total Oxygen: .000 Average Total Weight%: 100.109
Average Calculated Oxygen: .000 Average Atomic Number: 11.560
Average Excess Oxygen: .000 Average Atomic Weight: 21.022
Oxygen Equiv. from Halogen: .010 Halogen Corrected Oxygen: -.010
Average ZAF Iteration: 4.00 Average Quant Iterate: 4.00
St 305 Set 1 Labradorite (Lake Co.), Results in Elemental Weight Percents
ELEM: Na Si K Al Mg Fe Ca Mn Ti F O H
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC SPEC
BGDS: MAN MAN MAN MAN MAN MAN MAN MAN MAN MAN
TIME: 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00 30.00
BEAM: 29.93 29.93 29.93 29.93 29.93 29.93 29.93 29.93 29.93 29.93
ELEM: Na Si K Al Mg Fe Ca Mn Ti F O H SUM
21 2.907 25.278 .101 16.170 .089 .309 9.575 .002 .009 -.014 46.823 .000 101.248
22 2.852 23.590 .109 16.328 .088 .291 9.500 -.001 .024 .047 46.823 .000 99.650
23 2.831 23.548 .105 16.421 .087 .359 9.596 .002 .005 .040 46.823 .000 99.817
24 2.807 23.594 .108 16.406 .087 .359 9.588 .001 .001 .014 46.823 .000 99.789
25 2.809 23.884 .097 16.474 .085 .284 9.531 .000 .022 .034 46.823 .000 100.043
AVER: 2.841 23.979 .104 16.360 .087 .320 9.558 .001 .012 .024 46.823 .000 100.109
SDEV: .041 .738 .005 .118 .001 .036 .041 .001 .010 .025 .000 .000 .652
SERR: .018 .330 .002 .053 .001 .016 .018 .001 .005 .011 .000 .000
%RSD: 1.44 3.08 4.95 .72 1.40 11.39 .43 181.78 83.39 101.80 .00 .00
PUBL: 2.841 23.957 .100 16.359 .084 .319 9.577 .000 n.a. n.a. 46.823 n.a. 100.060
%VAR: (.01) .09 4.03 (.00) 3.72 .42 -.20 .00 --- --- .00 ---
DIFF: (.00) .022 .004 (.00) .003 .001 -.019 .000 --- --- .000 ---
STDS: 305 358 374 305 12 395 358 25 22 835 0 0
STKF: .0155 .2093 .1132 .1275 .4736 .6779 .1693 .7341 .5547 .1715 .0000 .0000
STCT: 45.98 65.79 225.18 261.40 2197.75 139.51 107.68 1966.75 56.01 27.06 .00 .00
UNKF: .0155 .1841 .0009 .1275 .0006 .0027 .0871 .0000 .0001 .0001 .0000 .0000
UNCT: 45.98 57.90 1.86 261.40 2.79 .55 55.42 .01 .01 .01 .00 .00
UNBG: .94 .07 .90 .85 1.67 .22 .46 5.03 .06 .06 .00 .00
ZCOR: 1.8317 1.3023 1.1141 1.2828 1.4496 1.1948 1.0967 1.2151 1.2005 4.1769 .0000 .0000
KRAW: 1.0000 .8799 .0083 1.0000 .0013 .0040 .5147 .0000 .0002 .0003 .0000 .0000
PKBG: 50.09 786.71 3.06 307.94 2.67 3.49 121.95 1.00 1.18 1.16 .00 .00
INT%: ---- ---- ---- ---- ---- .00 ---- ---- ---- .00 ---- ----
I'd call the accuracy (%VAR line) of the K, Mg and Fe results "spurious accuracy" if I didn't see this all the time! ;)
-
My question pertains to overall uncertainty estimation on the MAN method. There is the uncertainty that comes from the variance in the on peak measurement. But there should also be a component of uncertainty in the estimation of background intensity from the Assigned MAN fits.
I've attached an example from an assigned MAN fit of Ca Ka on PET for some feldspars I'm running this week. I've applied a 2nd order polynomial fit to the data (I'm using R for this example, but the results are the same as PFE) and the dashed lines are the 1-sigma confidence intervals of that fit. The estimated slope is 1.044 +/- 0.334 and the intercept is 3.508 +/- 2.802 (standard errors for uncertainty).
(https://probesoftware.com/smf/oldpics/i44.tinypic.com/2l9l4i0.jpg)
When you use this to estimate the Ca Ka background intensity for Lake County Plagioclase with Zbar of 11.5885 you get 17.37 +/- 0.42 cps/30nA, which is a 2.5% 1-sigma uncertainty. Is this propagated through to the final estimate of uncertainty in PFE? How does that compare to magnitude of the uncertainty in the background intensity from off peak measurements?
-
Hi Sean,
This is a good question, but a little difficult to answer.
Yes, there is an uncertainty about the MAN fit, but that plot is a function of the average atomic number of the MAN fit, not the concentration of the element.
For example, in the following MAN plot, we can see that the one sigma % deviation for K ka is around 2 %, but in our unknown material, the calculation of the average atomic number is essentially a constant- if the material is even roughly homogeneous.
(https://probesoftware.com/smf/oldpics/i44.tinypic.com/2lxjdx1.jpg)
This is why there is such an improvement in precision when using the MAN correction for trace elements as described here:
http://probesoftware.com/smf/index.php?topic=29.msg237#msg237
Now maybe we can just assume that the precision is constant for a given average atomic number (a not unreasonable simplification), so using the previous labradorite example we obtained 0.104 wt% K:
ELEM: Na Si K Al Mg Fe Ca Mn Ti F O H SUM
21 2.907 25.278 .101 16.170 .089 .309 9.575 .002 .009 -.014 46.823 .000 101.248
22 2.852 23.590 .109 16.328 .088 .291 9.500 -.001 .024 .047 46.823 .000 99.650
23 2.831 23.548 .105 16.421 .087 .359 9.596 .002 .005 .040 46.823 .000 99.817
24 2.807 23.594 .108 16.406 .087 .359 9.588 .001 .001 .014 46.823 .000 99.789
25 2.809 23.884 .097 16.474 .085 .284 9.531 .000 .022 .034 46.823 .000 100.043
AVER: 2.841 23.979 .104 16.360 .087 .320 9.558 .001 .012 .024 46.823 .000 100.109
SDEV: .041 .738 .005 .118 .001 .036 .041 .001 .010 .025 .000 .000 .652
SERR: .018 .330 .002 .053 .001 .016 .018 .001 .005 .011 .000 .000
%RSD: 1.44 3.08 4.95 .72 1.40 11.39 .43 181.78 83.39 101.80 .00 .00
So maybe we can just assume that the 2% average variance on the MAN fit applies to the "concentration" of the background as well? So since the background count rate from MAN is roughly 0.9 cps/nA and .1 wt% corresponds to roughly 1.8 cps/nA then the 2% variance of the MAN applies to background of the .1 wt.% or 0.001 wt.% or 10 PPM.
So if we add that to the 0.005 wt.% or 50 PPM (4.95% variance) we already obtained from the measurement, we obtain a total variance of roughly 0.006 wt.% or 60 PPM.
It makes sense because if we improve the precision of the MAN fit, the additional variance approaches zero. How does that sound?
ELEM: Na Si K Al Mg Fe Ca Mn Ti F O H SUM
STKF: .0155 .2093 .1132 .1275 .4736 .6779 .1693 .7341 .5547 .1715 .0000 .0000
STCT: 45.98 65.79 225.18 261.40 2197.75 139.51 107.68 1966.75 56.01 27.06 .00 .00
UNKF: .0155 .1841 .0009 .1275 .0006 .0027 .0871 .0000 .0001 .0001 .0000 .0000
UNCT: 45.98 57.90 1.86 261.40 2.79 .55 55.42 .01 .01 .01 .00 .00
UNBG: .94 .07 .90 .85 1.67 .22 .46 5.03 .06 .06 .00 .00
Edit: Actually, in thinking about this some more, the precision of the fit mostly depends on the accuracy of the continuum absorption correction! Note the change in fit variance with and without the continuum absorption correction checkbox checked in the Assign MAN Fits dialog above.
-
Hi John
I've just been listening to Karsten and Paul telling us about the wonders of MAN background usage and came back to Perth fired up and reach for the manual (Users guide and reference) which doesn't give any instruction on how to set it up rather a simple flavour of what might be done. So I'm sitting here scratching my head with the spectre of long mapping sessions with off peak background acquisition as I cannot get easily to the crux of what needs to be done. Having waded through a morass of questions and answers on the discussion forum, one question remains unasked and unanswered. That is quite simply, how do you do it. This may need a flow chart.
Specifically, do you need to model all backgrounds for all elements of interest on all specs and all crystals or is it sufficient to chose a position to get counts as a function of z per crystal per spec at a specific I and KV?
Please can I have an stepwise idiot's guide.
Cheers,
Malc.
-
No problem, you already have it. Click the Help | Getting Started Manual menu and scroll down to the Assign MAN Background Calibrations chapter around page 153. It will take you through the process mouse click by mouse click. Once you see how easily it is done, you will be both surprised and pleased.
john
-
I cannot over emphasize the importance of properly fitting one's standards for the MAN background fits.
The basic "rule of thumb" (and it is merely a "rule of thumb" so there are exceptions!), is that since we are trying to characterize the continuum as a function of average atomic number, and background is, by definition, generally the lowest intensity we can measure, then any standard which plots "above" the fit as seen here, is a good candidate to <ctrl> click to unselect and update the MAN fit:
(https://probesoftware.com/smf/oldpics/i58.tinypic.com/34dsn50.jpg)
In the above case, two standards that both do not contain fluorine, are seen to give an elevated background, why? Because P and Cr both interfere quite well with F ka...
In other situations one might actually find that they have a trace element contamination in their standard that they were not aware of (Probe for EPMA automatically deselects all standards that contain any of the element of interest in the MAN fitting) and therefore that standard will also plot "above" the general trend. Therefore "contaminated" standards should also be deselected and the fit updated similarly as we did with the above apatite and Cr2O3 standards which interfere with F ka.
In practice you do not have to know exactly why a particular standard intensity plots above the general trend of continuum intensities, just unselect and re-fit. Though as is often the case, the physics details might be interesting!
(https://probesoftware.com/smf/oldpics/i59.tinypic.com/2vsgajd.jpg)
For example, why does the copper standard intensity cause the intensity to "curve over" and decrease? Because in higher average atomic number materials, the self absorption in the material, for low energy x-rays such as fluorine, is increasing faster than the increase in continuum production due to the slowing down of the primary electrons.
-
John - is it possible to use a combination of MAN and off-peak backgrounds for samples where all elements were originally acquired with off-peak backgrounds? To elaborate, following your and Karsten's example, I acquired all the samples and standards with off-peak backgrounds, assigned the MAN fits after checking "Use Off Peak Elements for MAN Fit", and then applied them to my unknowns by checking "Use MAN correction for Off-Peak Elements". I was running a routine for REE-phosphates similar to Karsten's, and was looking to do a similar comparison (which turned out to be quite informative).
Sounds like you did everything correct. This is exactly why I implemented the ability to change all off-peak acquired elements to be treated as MAN background acquired. This makes sense since the on-peak data is already there, one simply needs to ignore the off-peak data! That way one can compare the same data using both off-peak and MAN (one can even treat a multi-point background as a normal off-peak sample!). 8)
This was the reason for this post:
http://probesoftware.com/smf/index.php?topic=4.msg189#msg189
I noticed that checking the "Use MAN Correction for Off-Peak Elements" option seems to apply the MAN background to all elements. Is it possible to post-process the data using a combination of MAN backgrounds for a few elements, and off-peak backgrounds for the rest? For example, could I use the off-peak backgrounds for the middle-heavy REE's, and use MAN backgrounds for La, Ce and P? Or would I need to have specified MAN or off-peak for each element prior to acquisition? Apologies if I'm missing something obvious here.
Yes, you can switch all off-peak elements to MAN to compare, but not selected elements. The only thing obvious you are missing is that if this is what you want to do, then do it! ;)
That is acquire "off-peak backgrounds for the middle-heavy REE's, and use MAN backgrounds for La, Ce and P". Remember, you can acquire a different background for every element *and* also acquire different backgrounds for the standards compared to the unknowns- if that became necessary...
You might also take a look at this post by Karsten Goemann on looking at REE elements:
http://probesoftware.com/smf/index.php?topic=4.msg17#msg17
-
There is a simple method to compare the same intensity data using off-peak, MAN and Nth point background corrections...
The above post:
http://probesoftware.com/smf/index.php?topic=4.msg189#msg189
mentions calculating the same data using both off-peak and MAN methods to compare the accuracy and precision.
Please note that one might consider making a copy of the original probe data MDB file to another folder, so that one copy can be treated using off-peak background corrections and the other copy can be treated using MAN background correction methods.
This allows the user to keep the results for each method without having to re-assign and re-calculate.
-
I am going to post something about MAN, but first I'll make a suggestion. John, you might consider making a separate MAN discussion page. Cause when you search for MAN or "MAN" you also get hits for many, mania, manuscript, etc.
As I hope MAN use takes off, it might be useful to make it easier for people for find MAN posts.
Just thinking....
-
John, can you clarify exactly under what conditions one should click the "Clear all MAN assignments" button?
Case in point. I have set up a 10 element glass MAN setup. I acquire all the MAN backgrounds, and life is good.
A month later, an occasion comes to use the MAN backgrounds but I need to add another element, P. Interesting, I think, as I want to add it on TAP, where it will be at a low angle and thus the background is curved, and since P in a minor element in the basaltic glasses, I have to worry about getting the proper background curvature. But wait.... if I use MAN, I don't have to worry about the modeling of the curvature, since the MAN takes it all into account. Another reason to use MAN, even for minor amounts (I'm expected 0.5-0.9 wt%).
But my question. Since I've added an element, do I need to "start all over" in the MAN assignments (click the clear button), or, preferably, just select "Assign MAN fits" and go thru the procedure.
Note also that in this procedure, I am re-acquired (after 2-3 months) the backgrounds on many of the MAN stds I acquired earlier this summer, so I can check how the bkg many have changed, as I assume it probably hasn't but would be nice to be sure.
So the question is, do you recommend clearing the prior MAN background assignment and "starting" fresh?
-
I am going to post something about MAN, but first I'll make a suggestion. John, you might consider making a separate MAN discussion page. Cause when you search for MAN or "MAN" you also get hits for many, mania, manuscript, etc. As I hope MAN use takes off, it might be useful to make it easier for people for find MAN posts.
I appreciate the thought. Right now I've tried to contain the MAN discussions into a main area. The main topic is for questions about the current MAN features and how to use them. The correct topic for these types of questions is the current topic:
http://probesoftware.com/smf/index.php?topic=4.0
which was started by Mike Spilde last summer. Your questions below in the next post are perfectly appropriate for this topic.
In addition there are some "specialty" topics on using MAN bgd corrections, such as...
1. Quantifying carbon using MAN bgd calibration curves:
http://probesoftware.com/smf/index.php?topic=48.0
Ironically, the carbon in steel quantification method used by many European EPMA users is a very simple form of our MAN method, but they only utilize a single standard (pure Fe) to obtain a background for carbon and do not perform a continuum absorption correction.
The main downside with using a single standard for the MAN regression is that if the average Z in the sample changes (e.g., unknown alloy steels and/or the carbon standard which is normally Fe3C), the background will be incorrectly calculated because the composition is different and therefore the average Z is different from our pure Fe standard used for the background. Our multi-standard MAN calibration curve allows the user to accurately apply the regression over a range of compositions and hence over a range of average Z.
2. Another MAN topic is using the MAN correction along with the blank correction to improve precision and reduce integration time for trace elements:
http://probesoftware.com/smf/index.php?topic=29.0
See especially reply#5 in the above Ti in Quartz topic.
3. There is also a topic here discussing the relationship between accuracy and precision in the MAN background correction:
http://probesoftware.com/smf/index.php?topic=307.0
I look forward to more discussion here as it is not a simple thing to think through.
-
John, can you clarify exactly under what conditions one should click the "Clear all MAN assignments" button? Case in point. I have set up a 10 element glass MAN setup. I acquire all the MAN backgrounds, and life is good.
A month later, an occasion comes to use the MAN backgrounds but I need to add another element, P. Interesting, I think, as I want to add it on TAP, where it will be at a low angle and thus the background is curved, and since P in a minor element in the basaltic glasses, I have to worry about getting the proper background curvature. But wait.... if I use MAN, I don't have to worry about the modeling of the curvature, since the MAN takes it all into account. Another reason to use MAN, even for minor amounts (I'm expected 0.5-0.9 wt%).
But my question. Since I've added an element, do I need to "start all over" in the MAN assignments (click the clear button), or, preferably, just select "Assign MAN fits" and go thru the procedure.
It's really a six or one half dozen of the other sort of thing. Either way is fine, it's just that if one made many changes to the setup such as adding several elements and changing the background types, etc., etc., it is entirely possible to correct everything in the Assign MAN Fits menu dialog, but it just might be easier to clear them and let the application re-assign everything automatically.
Then simply review all the MAN fits again to make sure they are good.
john
-
Hi All
I have a slightly more parochial question on MAN usage - or rather NOT usage.
I have built a pretty large setup for a wide variety of elements, many of which were copied in from old projects or elements.mdb
I decided to set everything up using off-peak backgrounds, so set both my unknowns AND my standards to use off-peak backgrounds in both the acquire>elements/cations AND acquire>acquisition options windows (see attached). I have then run a full set of standards and now when I try and analyse my standard data in the analyse window I get a message telling me I have no MAN data to fit for Ni (also attached). I don't just see this for Ni but all standards in this batch. Why is PfE looking for MAN data when I have set it to use off-peak BGs across the board?
Whats going on, what am I missing here?
Confused
Leeds
-
I decided to set everything up using off-peak backgrounds, so set both my unknowns AND my standards to use off-peak backgrounds in both the acquire>elements/cations AND acquire>acquisition options windows (see attached). I have then run a full set of standards and now when I try and analyse my standard data in the analyse window I get a message telling me I have no MAN data to fit for Ni (also attached). I don't just see this for Ni but all standards in this batch. Why is PfE looking for MAN data when I have set it to use off-peak BGs across the board?
Whats going on, what am I missing here?
Hi Leeds,
One possibility is that because of the standard intensity drift correction feature, the program will scan for all undeleted standards and try to load them for the drift interpolation.
So double check that all MAN standards are deleted and try again. You can also clear the MAN assignments but that shouldn't be necessary.
I also assume the Analytical | Use Off-Peaks For MAN Fits menu is unchecked?
-
Thanks for the reply.
Will try that tomorrow.
Leeds
:-\
-
Hi Again
I've done as you suggested and deleted the old standards, some of which were imported from projects using MAN BG.
My problem has gone away. All is now good.
Thanks Probeman!
:D 8)
-
Hi John,
I have a question regarding the collection of backgrounds for quant maps of a single phase e.g. Mg Calcite or Plagioclase- which we do often. In a single mineral backgrounds are constant & for majors peak to background ratios are high so background represents small percentage - making it tempting to ignore backgrounds. What we should do is MAN backgrounds the only reason we dont is to save time by not collecting standard data for standards which do not contain the elements. But I suppose the machine's pretty stable and we could just calibrate for MAN backgrounds every few months as larger errors for backgrounds would not matter? Have you thought about having an option were you could specify a single background in counts per second for the whole map - for this case were your mapping elements within a single phase e.g. here Mg-calcite for minor and trace elements - and the background could be specified for the map from a single quant point.
Thanks
Ben
-
I have a question regarding the collection of backgrounds for quant maps of a single phase e.g. Mg Calcite or Plagioclase- which we do often. In a single mineral backgrounds are constant & for majors peak to background ratios are high so background represents small percentage - making it tempting to ignore backgrounds. What we should do is MAN backgrounds the only reason we dont is to save time by not collecting standard data for standards which do not contain the elements. But I suppose the machine's pretty stable and we could just calibrate for MAN backgrounds every few months as larger errors for backgrounds would not matter? Have you thought about having an option were you could specify a single background in counts per second for the whole map - for this case were your mapping elements within a single phase e.g. here Mg-calcite for minor and trace elements - and the background could be specified for the map from a single quant point.
Hi Ben,
Good question.
Basically you want to do an MAN background calibration curve using a single point... which assumes of course that you have a constant composition (or at least a constant average Z which is essentially the same). However, as you point out, for major elements the background (if it is smaller than the measurement precision) can be ignored, or at least roughly approximated.
So here's an example of what to do if you wanted to say x-ray map (or point analyze) Ca and Mg on say dolomites that are (roughly) constant in composition. By the way, to include CO3 in the matrix calculation by various methods, see this topic:
http://probesoftware.com/smf/index.php?topic=92.0
Let's start with adding Ca and Mg to a spectrometer and then go to the Acquisition Options dialog as seen here and specify that the standard acquisition use off-peak backgrounds but the unknowns (your dolomite point analyses or maps) use the MAN background. Here is the Acquisition Options dialog:
(https://probesoftware.com/smf/oldpics/i60.tinypic.com/2u63e45.jpg)
This gets more important as one starts to add additional elements to analyze for, but then one will have multiple standards that *do not* contain the element of interest and can therefore be utilized in the MAN calibration for variation in average Z!
Next we select our standard positions and run them:
(https://probesoftware.com/smf/oldpics/i60.tinypic.com/nqz0j8.jpg)
Once the standards are acquired we can analyze them as unknowns because we utilized off-peak backgrounds as seen here:
St 135 Set 1 Calcite (analyzed), Results in Elemental Weight Percents
ELEM: Mg Ca C O
TYPE: ANAL ANAL SPEC SPEC
BGDS: LIN LIN
TIME: 20.00 20.00
BEAM: 30.01 30.01
ELEM: Mg Ca C O SUM
1 -.017 40.141 12.000 47.952 100.076
2 .028 39.956 12.000 47.952 99.936
3 -.010 39.994 12.000 47.952 99.935
AVER: .000 40.030 12.000 47.952 99.982
SDEV: .024 .098 .000 .000 .081
SERR: .014 .057 .000 .000
%RSD: ---- .24 .00 .00
PUBL: n.a. 40.031 12.000 47.952 99.983
%VAR: --- (.00) .00 .00
DIFF: --- (.00) .000 .000
STDS: 222 135 0 0
Now we go to assign the MAN background for our unknowns using the Analytical menu as seen here:
(https://probesoftware.com/smf/oldpics/i58.tinypic.com/2lbkwvm.jpg)
But whoa, we now get this error message:
(https://probesoftware.com/smf/oldpics/i59.tinypic.com/29diizd.jpg)
Why? Because our standards were acquired using off-peak backgrounds and so the program ignores the off-peak standards by default for the MAN calibration. So we can change that default by clicking this menu in the same Analytical menu as seen here:
(https://probesoftware.com/smf/oldpics/i59.tinypic.com/2zjgydh.jpg)
Now we click the Analytical | Assign MAN Fits menu again and now we see this for the first element Mg:
(https://probesoftware.com/smf/oldpics/i57.tinypic.com/1z50sqc.jpg)
and this for Ca:
(https://probesoftware.com/smf/oldpics/i62.tinypic.com/6gfy51.jpg)
Yes, we're only using a single point for each MAN calibration curve so it is a constant background intensity for that element, but depending on what standard you pick you'll get a reasonable background estimate for major and maybe even minor elements. Or you could just acquire one or two more darn standards and get a proper slope for your MAN calibration curves!
Helpful hint: if you aren't sure which of your standards might have similar average Z values compared to your unknowns, you can use this handy menu the the Standard.exe app as seen here:
(https://probesoftware.com/smf/oldpics/i61.tinypic.com/2prbmhd.jpg)
-
In practice you do not have to know exactly why a particular standard intensity plots above the general trend of continuum intensities, just unselect and re-fit. Though as is often the case, the physics details might be interesting!
(https://probesoftware.com/smf/oldpics/i59.tinypic.com/2vsgajd.jpg)
For example, why does the copper standard intensity [cuprite #114] cause the intensity to "curve over" and decrease? Because in higher average atomic number materials, the self absorption in the material, for low energy x-rays such as fluorine, is increasing faster than the increase in continuum production due to the slowing down of the primary electrons.
After seeing the plot above some might ask, why stop there? Maybe the std 140 (Rutile) is also either contaminated with fluorine (unlikely I'm guessing) or perhaps the Ti La line slightly interferes with the F Ka line? After consideration of the fact that Ti L emission overlaps *strongly* with N ka, we might guess that perhaps the tail of a Ti L emission line could cause a slight interference for F Ka?
The easiest way to try and answer this question is by examining an x-ray database such as the NIST 2005 database in PFE available from the Xray | Xray Database menu as seen here:
(https://probesoftware.com/smf/oldpics/i57.tinypic.com/2gvpgro.jpg)
Note that the wavelength range was specified by entering F Ka in the Specify Range controls as seen above. Now to see if any Ti emission lines could in principle, interfere with the F Ka line, we can select an element to highlight, in the case Ti, by using the Highlight Element control as seen here:
(https://probesoftware.com/smf/oldpics/i62.tinypic.com/hvujx2.jpg)
Or one can select multiple emitters by clicking the Periodic Table button and using the periodic table dialog as seen here:
(https://probesoftware.com/smf/oldpics/i60.tinypic.com/1z2oz1c.jpg)
In either case no Ti lines are found near the analytical emission line for fluorine. Another method which can provide a nominal estimate of the possible overlaps is by using the Options | Interferences menu in the Standard.exe app. Begin by selecting a the potentially problematic Ti standard or another similar standard such as synthetic TiO2 as seen here:
(https://probesoftware.com/smf/oldpics/i58.tinypic.com/1zgxqu1.jpg)
This will cause the following window to open with the TiO2 composition loaded automatically as seen here:
(https://probesoftware.com/smf/oldpics/i60.tinypic.com/33axsom.jpg)
Now, using the Unknown button, add a small amount of fluorine to the material because the software only check for elements in the material as seen here:
(https://probesoftware.com/smf/oldpics/i62.tinypic.com/dfhe9g.jpg)
Finally, click the Calculate button to checking for overlaps where as we can see here, nothing is reported for F Ka:
(https://probesoftware.com/smf/oldpics/i60.tinypic.com/fmq2w0.jpg)
-
Hi John,
Thanks. I get you where the average atomic number is constant (as in a single phase) so we don't have to correct for changing atomic number within the map we can use a single MAN standard. With best MAN data given where atomic number of MAN standard is same as unknown.
By using even a single MAN standard with similar atomic number (as is the case for most silicates/carbonate standards) we improve the quantification compared to no background. The error due to difference in atomic number of MAN standard and unknown is much smaller than the error with no background.
A good compromise for major elements where p/b ratio is large and background is small % (<1%) of total counts?
-
Thanks. I get you where the average atomic number is constant (as in a single phase) so we don't have to correct for changing atomic number within the map we can use a single MAN standard. With best MAN data given where atomic number of MAN standard is same as unknown.
By using even a single MAN standard with similar atomic number (as is the case for most silicates/carbonate standards) we improve the quantification compared to no background. The error due to difference in atomic number of MAN standard and unknown is much smaller than the error with no background.
A good compromise for major elements where p/b ratio is large and background is small % (<1%) of total counts?
Hi Ben,
Yes that is correct if you are just measuring trace elements and none of the matrix elements. Of course you can't use a silicate to measure the background for Si ka!
But, if you are measuring just traces in silicates or carbonates, you've already measured a bunch of "on-peak" or MAN backgrounds on the standards you are using for your trace elements, e.g., Fe in TiO2.
So, even if you used off-peak backgrounds for your standard calibration, and you plan on using MAN backgrounds for your unknown quant x-ray mapping runs, you can still use those off-peak measured standards as backgrounds for the MAN calibration. This involves using the Analytical | Use Off-Peak Elements For MAN Fit menu as described here:
http://probesoftware.com/smf/index.php?topic=4.msg189#msg189
-
In case anyone is interested, I've attached a pdf of the trace MAN presentation I gave at the EPMA 2016 topical conference this week in Madison (remember, you have to be logged in to see attachments).
The full paper will be out in the Aug issue of the Amer. Min. this year.
-
Here's a couple of MAN plots from my trace zircon run posted here:
http://probesoftware.com/smf/index.php?topic=73.msg5130#msg5130
In these MAN plots of the P ka line, here is the plot with the defaults assigned by the program:
(https://probesoftware.com/smf/gallery/395_11_10_16_4_07_28.png)
Note that P plots well above the trend. Quantifying this amount shows an apparent 220 PPM of P in zircon, but is it real?
That is there is either a trace amount of phosphorus in this synthetic zircon or there is a spectral interference from Si or Zr. Since this is a synthetic zircon from Oak Ridge and we know it is very pure, we can check for nominal interferences using the Standard Assignments dialog as seen here:
(https://probesoftware.com/smf/gallery/395_11_10_16_4_14_38.png)
And in fact there is a very small interference from the Zr La line, but it's a big emission line so it doesn't take much of a tail to create an apparent 220 PPM of P.
Once we remove the zircon standard from the MAN fit we obtain this plot:
(https://probesoftware.com/smf/gallery/395_11_10_16_4_07_47.png)
And we are good. Remember, an artifact like this can be corrected for using the normal interference correction in PFE, or we can assign a blank correction (assuming we have a spectroscopically pure zircon standard- or at least one that is well characterized for trace phosphorus using a technique other than EPMA) to deal with it. Remember, you should not assign both a spectral interference and a blank correction as you will get a "double correction". Fortunately, PFE will remind us that this is a no-no!
john
-
1) Is it possible to easily extract the raw data behind MAN (i.e. the standard intensities and Z-bar values)? The new graphics are very nice, but being able to extract the raw data to plot up off-line would be very useful for figures/papers/posters/etc. I realize I can export the count rates for the standards from the "Output" menu, and then look up the Z-bar values in the standard database, but... well, I'm lazy.
Hi Owen,
I'm lazy too. That is why the Output | Save Custom Output Format #8 (MAN) already outputs the standard intensities and zbars.
2) This is most likely operator error on my part, but is there a reason the last element in the MAN list reverts to a curved fit, even when you've told it to force the straight-line fit?
To explain further, I open the "Assign MAN Fits" window, click the last element, and click "Force Straight Line Fit", which does its job nicely. Then I hit OK, and the dialog closes.
Well you do realize that one needs to check this "Force Straight Line Fit" checkbox and click the Update Fit button for each element that you want to force a straight line fit for, if you want it to stick...
john
-
I realize this may seem like a nitpick, but I caught this while I was reducing some data, and couldn't figure out why the numbers for one element would sometimes change without me (seemingly) doing anything. It's easy to do unwittingly.
Yes, very helpful.
It's clearly a small bug. I will squash it tomorrow!
john
-
Hi Owen,
OK. The fix for that last MAN fit assignment glitch is fixed and ready to download the latest PFE.
It was subtle... had to do with the order things are loaded! But it *only* affected the last element for this reason!
john
-
Hi Owen,
OK. The fix for that last MAN fit assignment glitch is fixed and ready to download the latest PFE.
It was subtle... had to do with the order things are loaded! But it *only* affected the last element for this reason!
john
Awesome! Thanks much!
No worries, thanks for the excellent beta testing report! It's a lot easier to reproduce issues when I get a detailed description! :-*
-
I really enjoy using the MAN, really! However, there is a series of error message that really starts to annoy me. The problem occurs when a lot of different standards are loaded in the run and the MAN has not been acquired in all of them. Whereas I do acknowledge and appreciate that the error appears (it can be legit in some case), it get really really really annoying when the same error repeat itself over 50 times for 12 elements (and ~20 standards). Please, please, please... can we find a better way to list this, e.g., buffer in a memory all the problematic standards / element that have missing data, and then maybe simply report it in the log window?
Second, I run into a big problem that actually resulted in a totally screwed up MDB file where I cannot use the MAN anymore (so as all my unknown are acquired with MAN I basically cannot reprocess my data anymore...). The issue started happening when I decided to acquire a new set of MAN acquisition just for the element mapping I was performing for the same user. The difference with my previous MAN acquisition was that I moved the Si Ka from a TAP (used for quant) to a PET (to enable mapping along with Al, Mg [TAP], K [PET] and Fe [LiF]). Ever since I do have this new calibration, calling the assign MAN routine returns me absolutely... nothing (i.e., PfE cannot find any MAN acquisition for any standard, and I know this is not true). To further complicate that story, my latest standardization and MAN acquisition on standard included two column conditions (15 and 10 keV). Clearing the MAN assignment and re-doing everything does not solve the problem. Disabling a set of standard and keeping only the one I need does not help neither. Exporting the results into a new file (and importing the std associate to the mapping setting only) does not solve the problem neither. My ultimate solution: use my backup before the file was screwed up and re-acquire again my standards in, this time, a NEW file. I guess, this is my fault, as I should have prepared a new sample right away for this new setting, but I was hoping to continue using the other MAN assignment for Al, Mg, K and Fe, which had not changed.
Is there a solution to fix this problem? I cannot open the MDB file and manually delete the "faulty" standard acquisitions as I only have access to the newest Microsoft Access software (which cannot read anymore the MDB file format). The file is too big to be attached on this forum, but I can send you (JD) a copy by email.
Best,
Julien
-
I really enjoy using the MAN, really! However, there is a series of error message that really starts to annoy me. The problem occurs when a lot of different standards are loaded in the run and the MAN has not been acquired in all of them. Whereas I do acknowledge and appreciate that the error appears (it can be legit in some case), it get really really really annoying when the same error repeat itself over 50 times for 12 elements (and ~20 standards). Please, please, please... can we find a better way to list this, e.g., buffer in a memory all the problematic standards / element that have missing data, and then maybe simply report it in the log window?
Sorry you are so frustrated. Adding another button to cancel out should be feasible. Note that if you load different standards into a new run you should probably first clear the MAN assignments and then you shouldn't get all these errors.
Is there a solution to fix this problem? I cannot open the MDB file and manually delete the "faulty" standard acquisitions as I only have access to the newest Microsoft Access software (which cannot read anymore the MDB file format). The file is too big to be attached on this forum, but I can send you (JD) a copy by email.
It's difficult to understand what you might have done here. Better send me the MDB file to look at.
john
-
Hi John,
Thanks for the quick response. This is not so urgent (and this is now the week-end and we probably all need a lot of rest after that week). I learnt from my mistakes. However, I was thinking the MAN code could handle the following situation (sorry for the vagueness of my previous post):
(1) The researcher wanted chlorite analysis. Easy. The day before their venue, I acquired all required elements that we agreed on the week before (Si, Al, Fe, Mn, Mg, Ca, Na, K, F and Cl) in several standards, all with MAN (no element with 2-pts bkg). The automated run I had to get the MAN on all my standards (maybe a little too many, but I’m still testing…) took 2-3 hours.
(2) Once they arrived, they also requested additional elements: Ba, V, Co, Ni and Zn. I didn't want to worry about setting right my background for these element that are expected to be minor (or maybe trace), and I also wanted to ease potential interference correction, two great reasons to opt for the MAN background correction. To avoid re-running all my standards with this now 2 to 3 minute long setup, I opt to just run it in 5 (new!) standards, for Ba (barite), Ni (pentlandite), Zn (sphalerite) and, V and Co (metals - I know, not ideal, but I’m doing with what I have right now...).
(3) So after that, I ended up with many standards that had no data for Ba, V, Co… and 5 other standards that had no data for Si, Al, Fe…
(4) At this point, I managed to get into the MAN assignment tool and set the assignment (after 50+ clicks on these error message). I guess an easy fix for this would be to generate ONE error for ALL elements / standards missing and have that reported at the end in the main log window of PfE (e.g., a red message).
(5) After this, the MAN was properly assigned and the treatment of the unknown analysis worked fine, we got all the data we needed for the day.
(6) Things got complicated during the evening. We opt to perform some quantitative mapping on these chlorite (for Si, Al, Mg, Fe and K). Based on my spectrometer setting, I was in need of putting Si on a PET crystal (the only two TAP reserved for Al and Mg). To enable the MAN correction, I had thus to reacquire MAN data that would include also Si on PET; I did this in a sub-set of my Si, Al, Fe, Mg and K standards. Stupidly enough I did that in the same MDB file...
(7) This is were the files started getting “bogus”. After this acquisition, I started having a bunch of errors than the MAN was not available for Al, or Si, or another element… To solve this error, I opt to clean all MAN assignment, and start from scratch. However, when I returned to the MAN assignment window (after many clicks), the MAN assignment constantly returned me an error that no MAN data were available, and this for every single element in the MAN assignment window. Also, Si is only listed once, whereas it should be listed twice (once for Si on TAP and one for Si on PET - actually there should even be a third entry, as I had also acquired MAN data in my standards for Si on PET at 10 keV (all other data were acquired at 15 keV).
I have sent you the good (early backup) and the bad file (after the latest std acquisition).
Again, no rush on this, I am just wondering why this has screwed up the file (so I won’t reproduce this), and if it is possible to fix this easily.
Best,
Julien
-
Hi Julien,
I will look at your file and see if I can handle this type of situation better.
I will only mention at this time that if I have a run with lot of MAN elements and I want to add some additional trace elements, I generally just add them as off-peak elements- though note that you need to make sure they are specified as off-peak both for both standards and unknowns, as seen in the Acquisition Options dialog.
I think something like the above will avoid the problems you are seeing. That said the software should be able to handle the situation you describe, so I will look into that this weekend. With the current software I think it's just a question of unselecting standards which do not have the MAN data acquired for those elements.
john
-
Hi Julien,
Well I wasn't sure I was going to be able to figure out what the issues were with this probe run file, because I have to hand it to you: you get the EPMA Grand Prize for most complicated probe run ever!
(https://probesoftware.com/smf/gallery/1_13_11_16_4_11_36.jpeg)
I mean, you have duplicate elements, at multiple keVs, using MAN backgrounds, aggregate intensities, with different element setups, etc.... I hardly knew where to start! Anyway, I managed to isolate the culprits and here is the summary of the bug fixes (from the version.txt file):
11/13/16 Fix several bugs in MAN fit code when duplicate elements at different operating voltages are
present in the same run (Allaz).
Here is a screenshot of the new (and improved!) MAN Fit display dialog (note both 15 and 10 keV elements):
(https://probesoftware.com/smf/gallery/1_13_11_16_12_59_42.png)
Below are two of your standards. First one at 15 keV:
St 340 Set 2 Pyrope, Results in Elemental Weight Percents
ELEM: Al Mg Si K Fe Ti Cr Mn Ca O
TYPE: ANAL ANAL ANAL ANAL ANAL SPEC SPEC SPEC SPEC CALC
BGDS: MAN MAN MAN MAN MAN
MAN1: 1.28970 -.12378 .018498 -.38521 -.03082
MAN2: .035652 .261638 .012660 .142803 .080721
MAN3: .003343 -.00301 -.00016 -.00163 -.00074
ABS%: -28.69 -30.48 -25.48 -5.67 -.14
TIME: 30.00 30.00 30.00 30.00 30.00 --- --- --- --- ---
BEAM: 19.86 19.86 19.86 19.86 19.86 --- --- --- --- ---
AGGR: --- --- --- --- ---
ELEM: Al Mg Si K Fe Ti Cr Mn Ca O SUM
251 11.5262 12.7897 19.4222 .00188 6.33455 .29975 1.40944 .24008 3.10892 44.7774 99.9101
252 11.5370 12.7413 19.7032 -.00008 6.37568 .29975 1.40944 .24008 3.10892 45.0867 100.502
253 11.5198 12.7502 19.4433 -.00427 6.40858 .29975 1.40944 .24008 3.10892 44.7897 99.9655
254 11.5865 12.7641 19.6021 .00188 6.30480 .29975 1.40944 .24008 3.10892 45.0107 100.328
255 11.5434 12.7980 19.3988 .00321 6.33253 .29975 1.40944 .24008 3.10892 44.7711 99.9051
256 11.4746 12.7441 19.4013 -.00203 6.40277 .29975 1.40944 .24008 3.10892 44.6964 99.7753
AVER: 11.5312 12.7646 19.4952 .00010 6.35982 .300 1.409 .240 3.109 44.855 100.064
SDEV: .03634 .02414 .12715 .00282 .04215 .000 .000 .000 .000 .155 .28413
SERR: .01484 .00986 .05191 .00115 .01721 .00000 .00000 .00000 .00000 .06335
%RSD: .31516 .18915 .65223 2853.97 .66281 .00000 .00000 .00000 .00001 .34593
PUBL: 11.4211 12.2716 19.3656 n.a. 6.80919 .29975 1.40944 .24008 3.10892 44.4140 99.3397
%VAR: .96 4.02 .67 --- -6.60 .00 .00 .00 .00 .99
DIFF: .11012 .49296 .12953 --- -.44937 .00000 .00000 .00000 .00000 .44131
STDS: 325 336 301 334 326 --- --- --- --- ---
STKF: .4390 .4784 .2543 .1130 .6544 --- --- --- --- ---
STCT: 1010.49 2735.00 119.94 243.57 431.67 --- --- --- --- ---
UNKF: .0800 .0890 .1435 .0000 .0537 --- --- --- --- ---
UNCT: 184.05 508.70 67.71 .00 35.42 --- --- --- --- ---
UNBG: 1.57 1.81 .11 1.04 .84 --- --- --- --- ---
ZCOR: 1.4421 1.4344 1.3582 1.1253 1.1845 --- --- --- --- ---
KRAW: .18214 .18600 .56454 .00001 .08204 --- --- --- --- ---
PKBG: 117.885 282.635 616.119 1.00184 43.3686 --- --- --- --- ---
And another at 10 keV:
St 336 Set 3 Periclase, Results in Elemental Weight Percents
ELEM: Al Mg Si K Fe O
TYPE: ANAL ANAL ANAL ANAL ANAL CALC
BGDS: MAN MAN MAN MAN MAN
MAN1: 1.10474 .003876 .012899 -.10651 -.11700
MAN2: .024212 .167921 .009920 .067070 .045766
MAN3: .001162 -.00215 -.00012 -.00060 -.00051
ABS%: -30.62 -10.37 -20.36 -2.37 .12
TIME: 30.00 30.00 30.00 30.00 30.00 ---
BEAM: 20.23 20.23 20.23 20.23 20.23 ---
AGGR: ---
ELEM: Al Mg Si K Fe O SUM
329 -.00400 60.2003 .00175 .00322 .04239 39.6411 99.8848
330 -.01423 60.2350 .00515 .00404 .06666 39.6659 99.9625
331 -.01595 60.3602 -.00681 .00431 .00401 39.7152 100.061
332 -.01114 60.3954 -.00088 .00203 .14669 39.7898 100.322
333 -.00462 60.3592 -.00599 .00016 .10331 39.7532 100.205
334 -.01185 60.3800 -.01198 .00276 .07546 39.7462 100.181
AVER: -.01030 60.3217 -.00313 .00275 .07309 39.719 100.103
SDEV: .00495 .08242 .00628 .00152 .04918 .056 .16341
SERR: .00202 .03365 .00257 .00062 .02008 .02297
%RSD: -48.062 .13663 -200.83 55.1365 67.2864 .14164
PUBL: n.a. 60.3027 n.a. n.a. n.a. 39.6973 100.000
%VAR: --- (.03) --- --- --- .05
DIFF: --- (.02) --- --- --- .02127
STDS: 325 336 301 334 326 ---
STKF: .4702 .5285 .2806 .1146 .6410 ---
STCT: 602.80 1765.08 69.94 100.27 75.88 ---
UNKF: -.0001 .5285 .0000 .0000 .0006 ---
UNCT: -.09 1765.08 -.01 .02 .07 ---
UNBG: 1.03 1.36 .08 .51 .30 ---
ZCOR: 1.5250 1.1414 1.2958 1.1365 1.2571 ---
KRAW: -.00014 1.00000 -.00009 .00021 .00091 ---
PKBG: .91587 1296.33 .92694 1.04123 1.22577 ---
A couple of notes. First the code now displays and re-checks the fit of each MAN element when you click the OK button. Also, if you are in aggregate intensity mode, the MAN window only displays the first instance of the duplicate element. If you want to see the MAN fit for any subsequent duplicate elements, just unselect the Use Aggregate Intensity mode under the Analytical | Analysis Options menu dialog.
Whew! In any event, the bottom line is that now Probe for EPMA shouold be able to handle any combination of element setups, with duplicate elements, at multiple keVs, for MAN background processing now.
john
PS For those that are interested, I've attached Julien's MDB file below (with his permission).
PPS This new software will allow one to handle multiple keVs in their MAN background fits for probe runs going back to v. 11.4.0, which is when I added the additional keV arrays for the MAN fit.
-
In going through the MAN fits in Julien's probe run, I noticed that some standards might have certain elements that are not reported in the standard.mdb compositional database. Remember, the MAN background fit is a fast and easy way to check that your standards are as clean and complete as one would hope. The outliers will tend to fall above the general trend line of the MAN fit, making them easy to spot.
Now some of these are definitely on-peak interferences, but some may be "contamination" in the sense that they are present- but not accounted for! For example in the MAN fit for Cl Ka we see this plot:
(https://probesoftware.com/smf/gallery/1_13_11_16_1_00_00.png)
And in fact when we analyze the standard (from the same MDB file as above):
St 328 Set 1 Kaersutite, Results in Elemental Weight Percents
ELEM: Si Al Na Mg F Ca Ti Cr K Cl Fe Mn H O
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC CALC
BGDS: MAN MAN MAN MAN MAN MAN MAN MAN MAN MAN MAN MAN
MAN1: .433228 -.69047 -.22037 -3.8524 -.53121 .458636 -.33698 -.45214 -.33213 -.05578 -.04689 -.11476
MAN2: .260354 .402656 .195662 .737859 .126105 .092872 .287473 .475453 .137779 .046313 .083034 .060886
MAN3: .000071 -.00552 -.00238 -.01295 -.00247 .000815 -.00163 -.00079 -.00152 -.00042 -.00079 -.00041
ABS%: -22.83 -28.84 -49.17 -35.61 -74.17 -3.34 -2.50 -1.34 -4.96 -11.57 -.46 -.83
TIME: 30.00 30.00 20.00 20.00 20.00 30.00 30.00 20.00 20.00 20.00 30.00 30.00 --- ---
BEAM: 20.01 20.01 20.01 20.01 20.01 20.01 20.01 20.01 20.01 20.01 20.01 20.01 --- ---
AGGR: --- ---
ELEM: Si Al Na Mg F Ca Ti Cr K Cl Fe Mn H O SUM
51 18.7167 6.47933 1.83938 7.97634 .33305 7.98541 3.10663 -.00252 .84801 .02933 8.83063 .12070 .23000 42.5990 99.0920
52 18.7057 6.47333 1.86776 7.85883 .24068 7.91836 3.12160 .01332 .87714 .02290 9.10796 .11198 .23000 42.6275 99.1770
53 18.4669 6.41391 1.85091 7.78733 .20441 7.86146 3.10923 .00973 .84554 .03115 9.73464 .12887 .23000 42.4083 99.0823
54 18.7365 6.57265 1.90582 7.86262 .30850 7.89659 3.07578 .01559 .87882 .02731 9.06527 .11481 .23000 42.6877 99.3779
55 18.8403 6.57865 1.85397 7.99179 .31942 7.90964 3.10957 .01488 .83560 .02806 8.94850 .12863 .23000 42.8628 99.6518
AVER: 18.6932 6.50357 1.86357 7.89538 .28121 7.91429 3.10456 .01020 .85702 .02775 9.13740 .12100 .230 42.637 99.2762
SDEV: .13735 .07062 .02569 .08651 .05575 .04528 .01710 .00746 .01970 .00308 .35091 .00774 .000 .164 .24125
SERR: .06143 .03158 .01149 .03869 .02493 .02025 .00765 .00334 .00881 .00138 .15693 .00346 .00000 .07328
%RSD: .73478 1.08590 1.37868 1.09564 19.8258 .57212 .55081 73.1317 2.29834 11.0899 3.84039 6.40067 .00001 .38431
PUBL: 19.4000 6.30000 1.95000 7.64000 .14000 8.15000 3.01000 n.a. .84000 n.a. 8.77000 .13000 .23000 43.1680 99.7280
%VAR: -3.64 3.23 -4.43 3.34 100.86 -2.89 3.14 --- 2.03 --- 4.19 -6.92 .00 -1.23
DIFF: -.70679 .20357 -.08643 .25538 .14121 -.23571 .09456 --- .01702 --- .36740 -.00900 .00000 -.53094
STDS: 301 325 301 336 349 322 344 319 334 350 326 342 --- ---
STKF: .2543 .4390 .0509 .4784 .0611 .1623 .5545 .2862 .1130 .0620 .6544 .2861 --- ---
STCT: 588.46 1014.03 194.06 2765.22 18.32 402.37 1438.72 777.01 242.84 84.89 431.68 164.84 --- ---
UNKF: .1433 .0452 .0093 .0511 .0007 .0738 .0265 .0001 .0078 .0002 .0774 .0010 --- ---
UNCT: 331.56 104.46 35.42 295.25 .22 182.89 68.80 .24 16.84 .32 51.04 .58 --- ---
UNBG: 2.95 2.57 .98 2.28 .18 1.74 3.04 5.52 1.14 .42 .89 .60 --- ---
ZCOR: 1.3048 1.4381 2.0061 1.5456 3.9095 1.0729 1.1709 1.1484 1.0941 1.1992 1.1808 1.1990 --- ---
KRAW: .56344 .10302 .18252 .10677 .01177 .45452 .04782 .00031 .06934 .00373 .11825 .00353 --- ---
PKBG: 113.266 41.6888 37.2975 130.257 2.21103 105.973 23.5975 1.04375 15.7533 1.75341 58.0413 1.96671 --- ---
we obtain about 270 PPM of Cl.
john
-
Many thanks, John! For the price - of course - that I gladly accept. However, this is not yet my most complicated run, you got to wait my new idea for fast, precise and accurate monazite analysis (and other REE minerals ;) 30+ elements, expecting to use MAN, TDI, n-th point, multipoint bkg... Potentially also multiple analytical conditions (trace vs major) and use of EDS... This got to be causing other troubles, so get ready :D :D :D
And thanks A LOT for these fixes! Effectively it works great now!
About the Kaersutite standard, the MAN is here used on this secondary standards (and many many more) to indeed check for the presence of an unlisted element. My periclase for instance contains some Fe (not so much a surprise), but also Al (and maybe Si), around 130-180 ppm Al, and is clearly visible on the MAN plot (data from a "regular" TAP crystal)!
(https://probesoftware.com/smf/gallery/1_25_11_19_8_17_48.png)
I might also remind other readers that peak interference will obviously also be a problem. I am here thinking of the analysis of REE minerals, for which multiple interferences exists and the MAN will pick it! And here is again another Wonderful MAN power: once you identified the peak interference, you can just use the data acquired on your standards for the MAN. No data are lost!
Let's recap:
- Identification of impurities or unrecognized element in standard at the ~100 ppm level maybe lower (??? to check ???)
- Cut your analysis time in half
- Allow identification of peak interference, and unusable MAN bkg data due to this interference can then be use for the interference correction itself!
- No worry about potential interferences on background.
I still wonder what (a) the long term stability is, (b) what is the variability, and (c) what is the error on the calibration curve (it should be possible to calculate an error envelope on the MAN fitting line / curve, using the repeated background measurement on the standards). I am getting some nice data on the long term reproducibility as I type, I might one day share this... Just need time to compile a huge set of data covering 3 month of testing and many standards. There are still some behavior on the MAN fits I cannot explain with impurity in std or peak interference, notably on the REE business... Got some idea, though (problem of absorption edges in the standard?)... More to come soon - hopefully!
Cheers to the PfE Master!
Julien
-
And thanks A LOT for these fixes! Effectively it works great now!
You are most welcome. It is a pleasure to work with so many experts such as yourself. As someone once said: it takes a village! :)
I still wonder what (a) the long term stability is, (b) what is the variability, and (c) what is the error on the calibration curve (it should be possible to calculate an error envelope on the MAN fitting line / curve, using the repeated background measurement on the standards). I am getting some nice data on the long term reproducibility as I type, I might one day share this... Just need time to compile a huge set of data covering 3 month of testing and many standards. There are still some behavior on the MAN fits I cannot explain with impurity in std or peak interference, notably on the REE business... Got some idea, though (problem of absorption edges in the standard?)... More to come soon - hopefully!
The error on the MAN calibration curve regression is discussed in detail in my recent Amer. Min. paper here:
http://epmalab.uoregon.edu/publ/A%20new%20EPMA%20method%20for%20fast%20trace%20element%20analysis%20in%20simple%20matrices.pdf
However, I can summarize:
1. Because we generally only measure the standards for the MAN calibration curve once per analytical session or mapping run, the regression error on the MAN fit *must* be treated as an accuracy error, as opposed to a precision error. Yes, if we remeasured the MAN regression curve for each point analysis or each mapping pixel, then we would need to include the MAN regression error as a precision issue. But no one would ever re-run the MAN calibration for every analysis or pixel! We need to think carefully about this as it is not an intuitive conclusion, but it is (surprisingly) correct. If you don't agree, please read the above paper (see tables 3a, 3b and 3c in particular), and maybe also read this post here:
http://probesoftware.com/smf/index.php?topic=307.msg3190#msg3190
2. The stability of the measured continuum intensities over time are exceptionally good so far as I have seen, mostly because we are measuring the intensities at the on-peak positions when the element is *not* present. Therefore there is no peak, only a smooth x-ray continuum.
3. In fact, if one re-measures the MAN curve within a single probe run, any changes in the continuum intensities over time are automatically applied to the unknown samples using the built-in drift correction in PfE, just as the primary and interference standards are drift corrected with multiple standard sets! But since the MAN intensities are not actually on a peak (though they are *at* the peak position!), the stability is much better than the primary standards or the interference standards. Especially better than the interference standard intensities, where the interfering intensities are often measured on the *side* of a peak (causing the the spectral overlap!) :o
4. That said, a difference in detector bias voltage or PHA window, could change the absolute intensity of the continuum for the MAN curves, so I generally re-measure both the primary standards and the MAN standards for each new run or re-run. It only takes a few minutes as the primary standards and the MAN standards are often the same standard (though not for the same element!).
5. However, if you are only utilizing MAN standards for major and even minor elements, you can probably skip re-measuring the MAN standards, but as I said above, it won't hurt to re-measure them and it only takes a few minutes for typical run of 5 or 10 elements.
6. Finally, absorption edges in the MAN standards are automatically taken care of by performing an absorption correction on the continuum intensities. This is because each MAN standard absorbs a particular continuum energy depending on the composition of that MAN standard. This was John Armstrong's contrbution to the paper. If you want to see the effect of absorption edges in the MAN curve, just turn off the continuum absorption correction in the Assign MAN Fits dialog. There are also examples in the above Amer. Min. paper for Na Ka and K Ka with and without the continuum absorption correction. The absorption correction is generally only noticable at low energies.
john
-
Hi John,
I'm using the newest version of PfE (11.6.9) and I'm getting some odd behaviour when I'm trying to assign man fits, I'm getting some errors about missing standards at the start, but then when I get to the window and verify the fits and hit OK the display cycles through all the elements generates a warning ("Warning no man standards have been assigned yet to calculate the man background fit on spectrometer 2 on crystal TAP at 20 keV" which is odd because Na on spectro 2 is not part of this particular sample) and stops. The only way to get out of the MAN assignment and fit window is to hit cancel, which I assume cancels the man fit. Is this a bug or is my run not set up properly?
Cheers
Glenn
-
I'm using the newest version of PfE (11.6.9) and I'm getting some odd behaviour when I'm trying to assign man fits, I'm getting some errors about missing standards at the start, but then when I get to the window and verify the fits and hit OK the display cycles through all the elements generates a warning ("Warning no man standards have been assigned yet to calculate the man background fit on spectrometer 2 on crystal TAP at 20 keV" which is odd because Na on spectro 2 is not part of this particular sample) and stops. The only way to get out of the MAN assignment and fit window is to hit cancel, which I assume cancels the man fit. Is this a bug or is my run not set up properly?
Hi Glenn,
This is because of Julien's issue above where he had multiple samples with different elements, but he still wants all of the MAN fits to be assigned to *all* MAN elements in *all* samples in the run.
So, yes. The new behavior of the software is to go through all samples in the entire run (if they are not disabled), and check for missing MAN assignments. So if you are getting warnings for a sample you don't care about, just disable it and re-load the MAN fits. Or acquire at least one MAN standard for that element with missing MAN standards.
Sorry for the change in behavior, but because of the potential of multiple sample setups with different or duplicate MAN elements, I had to check all samples in the probe run (unless they have been disabled).
john
-
Hi Glenn,
I was thinking that instead of disabling the offending samples so they don't get included in the MAN fit assignments, one could also just use the disable quant flag for the specific element causing the problem in the Elements/Cations dialog. But when I look closer, although I am checking for the disable acquisition flag, I am not checking for the disable quant flag.
But I wonder if you can utilize the disable acquisition flag for the offending element, in the offending samples, to avoid the missing MAN standard warnings? I note that one can disable acquisition on a specific element on a specific sample, so long as it doesn't have any data...
john
-
Hi there,
I'm using MAN backgrounds for analysis of hydrous basaltic glasses as it reduces the analysis time by half reducing the chance of beam damage (also running TDI in case there still is beam damage). Just wanted to check that a specified by difference element (H in this case) is included in the iterative calculation of the mean atomic number?
Thanks!
Ery
-
Hi there,
I'm using MAN backgrounds for analysis of hydrous basaltic glasses as it reduces the analysis time by half reducing the chance of beam damage (also running TDI in case there still is beam damage). Just wanted to check that a specified by difference element (H in this case) is included in the iterative calculation of the mean atomic number?
Thanks!
Ery
Hi Ery,
Good question. Here is how it works:
The average Z for the standards used in the MAN fit is calculated (as expected) based on the composition of the standards as defined in the standard.mdb database. So as long as the standard composition never changes, the average Z will always be the same.
However for unknowns, we have to utilize the average Z from the calculated composition. So, the average Z calculation for MAN includes the measured elements and also the specified elements whether they be by stoichiometry or by difference. So yes.
More details are in my recent trace MAN paper:
http://epmalab.uoregon.edu/publ/A%20new%20EPMA%20method%20for%20fast%20trace%20element%20analysis%20in%20simple%20matrices.pdf
Of equal importance is the fact that when calculating water (or any element) by difference (or by stoichiometry), one must include these elements by difference in the matrix correction, as is the case with Probe for EPMA. See the paper attached to this post, comparing water by difference included and not included in the matrix correction:
http://probesoftware.com/smf/index.php?topic=61.msg4303#msg4303
john
-
Hi Ery,
I just realized that you can see the calculated Z-bar or average Z in two places in Probe for EPMA. First here in the log window output:
Un 27 MAM IW2 C4-ext-1
TakeOff = 40.0 KiloVolt = 15.0 Beam Current = 20.0 Beam Size = 10
(Magnification (analytical) = 8000), Beam Mode = Analog Spot
(Magnification (default) = 1267, Magnification (imaging) = 1267)
Image Shift (X,Y): .00, .00
Number of Data Lines: 6 Number of 'Good' Data Lines: 6
First/Last Date-Time: 03/13/2007 06:29:57 PM to 03/13/2007 06:49:54 PM
WARNING- Using Exponential Off-Peak correction for p ka
WARNING- Using Time Dependent Intensity (TDI) Element Correction
Average Total Oxygen: 43.852 Average Total Weight%: 100.331
Average Calculated Oxygen: 43.852 Average Atomic Number: 13.324
Average Excess Oxygen: .000 Average Atomic Weight: 22.489
Average ZAF Iteration: 3.00 Average Quant Iterate: 3.00
Oxygen Calculated by Cation Stoichiometry and Included in the Matrix Correction
And also here in the Analyze! window:
(https://probesoftware.com/smf/gallery/1_20_02_17_10_24_23.png)
The average Z value will change visibly as the specified concentrations or elements by difference are modified.
-
Hi,
I've just noticed something. When you have elements on multiple spectrometers - and your viewing the MAN curves individually (not something you would normally do - only for testing, comparing spectrometers.. etc). The MAN is wrong for those standards which contain a high abundance of the element on multiple spectrometers.
e.g. If I view Al individually for Al2O3 and have it on 3 spectrometers.
Total is 189.46 (3*47.5 plus oxygen 47.075) and z-bar is 11.761
Whereas if I aggregate the spectrometers - Total is 100% and Z-BAR is correct 10.666
This is not a problem for analysis - for you'd never use something with 190% total, BUT something to be aware of when doing such tests
Ben
-
Hi John,
I think I missing something obvious - but using the custom output #8 (MAN) - what's the difference in units for intensity - I can't get them to match.
(https://probesoftware.com/smf/gallery/453_10_04_17_2_37_19.png)
(https://probesoftware.com/smf/gallery/453_10_04_17_2_37_32.png)
The intensity scale is difference. But if I divide rawcnt by seconds and nA it does not help (e.g. 8/10sec/20nA).
Also 520 Al-metal looks different
Thanks
Ben
-
Hi Ben,
The difference is that the raw MAN intensities output to a file are *not* corrected for continuum absorption, but the intensities in the MAN plot are corrected for continuum absorption.
Try comparing the raw intensities to the raw plot, by unchecking the Correct for Continuum Absorption checkbox.
That doesn't mean that I have gotten something mixed up somewhere! :D
john
-
I've just noticed something. When you have elements on multiple spectrometers - and your viewing the MAN curves individually (not something you would normally do - only for testing, comparing spectrometers.. etc). The MAN is wrong for those standards which contain a high abundance of the element on multiple spectrometers.
e.g. If I view Al individually for Al2O3 and have it on 3 spectrometers.
Total is 189.46 (3*47.5 plus oxygen 47.075) and z-bar is 11.761
Whereas if I aggregate the spectrometers - Total is 100% and Z-BAR is correct 10.666
This is not a problem for analysis - for you'd never use something with 190% total, BUT something to be aware of when doing such tests
Hi Ben,
Yes, when one has a 190% total, how should one calculate the matrix correction (or average Z) for such a non-physical situation? Answer: any way you feel like! ;D
The reason the average Z calculation is off when the spectrometers are not aggregated in your example is (I think) because you have several times the Al (from three spectrometers) relative to the specified oxygen (from the standard database), so the average Z is biased towards Al (Z = 13), rather than oxygen (Z = 8 ).
Try analyzing it in Analyze! using the stoichiometric oxygen option (in Calculation Options dialog).
john
-
Hi John,
The difference is that the raw MAN intensities output to a file are *not* corrected for continuum absorption, but the intensities in the MAN plot are corrected for continuum absorption.
Thanks - that makes sense - and was it!
Try analyzing it in Analyze! using the stoichiometric oxygen option (in Calculation Options dialog).
john
Thanks - that worked - keeping Al and O in proportion - great suggestion
Ben
-
Hi,
I just found what I was looking for - on the MAN output to excel you have both
rawcnt: data uncorrected for absorption.
corcnt: data corrected for absorption
Thanks
Ben
-
I just found what I was looking for - on the MAN output to excel you have both
rawcnt: data uncorrected for absorption.
corcnt: data corrected for absorption
Yes, that's right. In fact it also outputs all the matrix corrections for those MAN standards so you could also work backwards from the corrected counts...
Note that years ago I originally tried a continuum specific correction from NIST (because the continuum correction should be different from the bulk correction due to the anisotropy of the continuum emission), but it gave worse results than the bulk correction, so the bulk matrix correction term is what we currently use for the MAN continuum absorption correction.
john
-
Is there an option to output the intercept, slope, curvature, Rel % Deviation for the MAN fits for all the elements in a run, and maybe even which standards are being used? Playing around with different standards, off peak positions, forcing straight line versus not, etc. Be nice to have these values easily tabulated.
-
Is there an option to output the intercept, slope, curvature, Rel % Deviation for the MAN fits for all the elements in a run, and maybe even which standards are being used? Playing around with different standards, off peak positions, forcing straight line versus not, etc. Be nice to have these values easily tabulated.
Currently these parameters are output from the Output | Save Custom Analysis Output | Save Custom Analysis Format #8 (MAN):
LINE Data file line number
SNUM Sample number (standard number)
SNAM Sample name (standard name)
ZBAR Sample average atomic number
RAWCNT Raw count intensity
CORCNT Corrected count intensity (absorption corrected)
ZEDCOR Atomic number correction factor
ABSCOR Absorption correction factor
FLUCOR Fluorescence correction factor
ZAFCOR Total ZAF correction factor
ONTIM On-peak (MAN) counting time
BEAM Beam current in nA
BEAMCUR Beam current (combined condition samples)
And if you put the software in DebugMode from the Output menu, you will see this output in the log window:
Elements:
si fe mg ca p al na ba k ti
MAN Assignments:
835 835 336 374 336 12 835 336 358 336
12 12 374 835 374 14 12 358 835 358
22 14 835 14 835 27 14 12 12 374
27 22 14 22 12 28 22 14 14 835
28 28 22 28 14 285 27 27 22 12
285 285 27 0 22 0 28 28 27 14
0 0 28 0 27 0 285 160 28 27
0 0 0 0 28 0 0 285 160 28
0 0 0 0 0 0 0 0 285 160
0 0 0 0 0 0 0 0 0 285
BackgroundTypes:
MAN MAN MAN MAN MAN MAN MAN MAN MAN MAN
MAN Fit Orders:
2 2 2 2 2 2 2 2 2 2
MAN Absorption Correction Flags:
-1 -1 -1 -1 -1 -1 -1 -1 -1 -1
MAN Counts:
74.1 21.7 13.9 29.4 3.5 16.7 22.1 44.1 30.3 1.3
26.1 6.2 16.2 92.0 3.6 27.2 9.4 47.4 81.6 1.6
51.7 6.1 22.4 26.3 9.3 32.1 6.8 42.3 25.5 1.9
55.3 9.5 13.9 43.0 3.3 30.3 8.1 44.1 24.0 8.8
58.4 15.3 15.7 59.7 3.0 29.4 8.3 93.4 39.2 1.2
50.0 8.4 17.6 .0 5.5 .0 7.8 94.5 53.7 1.2
.0 .0 14.5 .0 6.9 .0 9.2 49.2 54.8 3.4
.0 .0 .0 .0 6.3 .0 .0 55.5 32.0 3.5
.0 .0 .0 .0 .0 .0 .0 .0 34.4 1.4
.0 .0 .0 .0 .0 .0 .0 .0 .0 2.0
MAN Fit Coefficients:
-14.766 -3.2199 -6.5750 -8.5227 -.40542 30.2930 .112198 -.54570 -5.1947 .052925
5.84314 .946438 2.68385 3.62923 .482494 -.80475 1.00449 3.86722 3.24887 .100007
-.05119 -.00766 -.02100 -.02692 -.00311 .097303 .019536 -.00100 -.02401 .002169
-
And this output as well:
MAN fit data and coefficients for si
Order 2, npts 6
StdAss: 835 12 22 27 28 285
Z-bars: 45.8151 10.4267 16.3920 22.9433 23.7170 14.4794
Counts: 74.1008 26.1434 51.6858 55.3129 58.4085 50.0244
AbsCor: 1.96292 1.57869 1.28689 1.61488 1.68766 1.17433
Counts: 145.454 41.2723 66.5140 89.3238 98.5737 58.7449
Coeffs: -14.745 5.84054 -.05111
MAN fit data and coefficients for fe
Order 2, npts 6
StdAss: 835 12 14 22 28 285
Z-bars: 45.8151 10.4267 10.8047 16.3920 23.7170 14.4794
Counts: 21.6734 6.21936 6.07033 9.48054 15.2525 8.44972
AbsCor: 1.10816 .996548 .998913 1.03724 1.00089 1.02129
Counts: 24.0175 6.19789 6.06373 9.83355 15.2662 8.62964
Coeffs: -3.2217 .946670 -.00767
MAN fit data and coefficients for mg
Order 2, npts 7
StdAss: 336 374 835 14 22 27 28
Z-bars: 11.0372 11.9907 45.8151 10.8047 16.3920 22.9433 23.7170
Counts: 13.8886 16.2342 22.3551 13.8539 15.7346 17.5643 14.4872
AbsCor: 1.54794 1.44002 3.23244 1.38991 1.87119 2.72537 2.94422
Counts: 21.4987 23.3775 72.2615 19.2557 29.4423 47.8691 42.6536
Coeffs: -6.5667 2.68299 -.02098
MAN fit data and coefficients for ca
Order 2, npts 5
StdAss: 374 835 14 22 28
Z-bars: 11.9907 45.8151 10.8047 16.3920 23.7170
Counts: 29.4093 91.9869 26.2621 42.9830 59.6713
AbsCor: 1.07150 1.10091 1.04570 1.00488 1.05021
Counts: 31.5120 101.269 27.4624 43.1929 62.6674
Coeffs: -8.5125 3.62767 -.02687
MAN fit data and coefficients for p
Order 2, npts 8
StdAss: 336 374 835 12 14 22 27 28
Z-bars: 11.0372 11.9907 45.8151 10.4267 10.8047 16.3920 22.9433 23.7170
Counts: 3.49284 3.59269 9.28106 3.25934 3.04344 5.48866 6.90243 6.28710
AbsCor: 1.36107 1.34961 1.63430 1.35544 1.40684 1.16702 1.38028 1.42656
Counts: 4.75401 4.84874 15.1680 4.41783 4.28163 6.40539 9.52732 8.96890
Coeffs: -.40611 .482553 -.00311
MAN fit data and coefficients for al
Order 2, npts 5
StdAss: 12 14 27 28 285
Z-bars: 10.4267 10.8047 22.9433 23.7170 14.4794
Counts: 16.7464 27.2437 32.0534 30.2572 29.4431
AbsCor: 1.95626 1.20535 2.01195 2.13817 1.31384
Counts: 32.7604 32.8382 64.4899 64.6951 38.6836
Coeffs: 30.2930 -.80475 .097303
MAN fit data and coefficients for na
Order 2, npts 7
StdAss: 835 12 14 22 27 28 285
Z-bars: 45.8151 10.4267 10.8047 16.3920 22.9433 23.7170 14.4794
Counts: 22.0557 9.42896 6.80209 8.13324 8.26629 7.81730 9.16492
AbsCor: 3.95005 1.49808 1.75421 2.60603 4.07311 4.48371 2.06884
Counts: 87.1213 14.1254 11.9323 21.1955 33.6695 35.0505 18.9608
Coeffs: .115846 1.00402 .019550
MAN fit data and coefficients for ba
Order 2, npts 8
StdAss: 336 358 12 14 27 28 160 285
Z-bars: 11.0372 12.2316 10.4267 10.8047 22.9433 23.7170 12.6937 14.4794
Counts: 44.0832 47.4109 42.3110 44.0621 93.3551 94.4881 49.2239 55.5154
AbsCor: .951038 .972255 .942275 .948707 .946385 .952120 .960851 1.00869
Counts: 41.9249 46.0955 39.8686 41.8020 88.3499 89.9640 47.2968 55.9981
Coeffs: -.54606 3.86727 -.00100
MAN fit data and coefficients for k
Order 2, npts 9
StdAss: 358 835 12 14 22 27 28 160 285
Z-bars: 12.2316 45.8151 10.4267 10.8047 16.3920 22.9433 23.7170 12.6937 14.4794
Counts: 30.2908 81.6375 25.5278 23.9502 39.1737 53.7147 54.8469 31.9706 34.4008
AbsCor: 1.04931 1.14161 1.05343 1.06993 1.01340 1.06181 1.07488 1.05296 1.05273
Counts: 31.7845 93.1984 26.8917 25.6251 39.6985 57.0348 58.9537 33.6638 36.2146
Coeffs: -5.1908 3.24845 -.02400
MAN fit data and coefficients for ti
Order 2, npts 10
StdAss: 336 358 374 835 12 14 27 28 160 285
Z-bars: 11.0372 12.2316 11.9907 45.8151 10.4267 10.8047 22.9433 23.7170 12.6937 14.4794
Counts: 1.28070 1.59672 1.94603 8.76543 1.19729 1.19741 3.37633 3.52606 1.44701 1.97932
AbsCor: 1.01933 1.04191 1.03182 1.04960 1.01001 1.01684 1.01439 1.02050 1.02978 1.08067
Counts: 1.30545 1.66364 2.00795 9.20017 1.20928 1.21758 3.42492 3.59834 1.49010 2.13899
Coeffs: .053961 .099936 .002170
-
The forum...it does work ;)
-
Hi,
I was just doing MAN for fluorine, clearly a tricky element, it seems to work better without correction for continuum absorption
With absorption correction TiO2, SrTiO3, Cr2O3 plot much higher compared to Barite and HfO2
(https://probesoftware.com/smf/gallery/453_05_09_17_6_33_12.png)
(https://probesoftware.com/smf/gallery/453_05_09_17_6_33_25.png)
In the second image, only those standards containing Fe lie significantly above the curve - due to Fe La interference (see wavescan below). giving following.
(https://probesoftware.com/smf/gallery/453_05_09_17_6_41_38.png)
Removed Fe standards - TiO2 disappointingly low
(https://probesoftware.com/smf/gallery/453_05_09_17_6_37_03.png)
-
Hi Ben,
Yes, I've seen this too.
Because there are so many elements (e.g., L lines) that interfere with fluorine, it is difficult to find a decent set of MAN standards that will work. I generally use the off-peak method for minor/trace fluorine.
An unrelated issue for oxygen is that almost any standard one could utilize for the oxygen MAN curve is oxidized. I have tried MAN calibrations for oxygen using Ge and Au, but haven't found an MAN standard useful at low Z.
john
-
Hi Ben,
Also, on the issue of the low intensity for F Ka in TiO2, note that the MAC for F ka in Ti is quite large, around 14500 (actually that is for F Ka in pure Ti. In TiO2 the MAC is more like 13700).
So by not utilizing the continuum correction in the MAN curve you are significantly undercorrecting the F Ka intensity in TiO2.
Here is your F Ka MAN calibration curve at 20 keV (F Ka at 20 keV?), with the continuum absorption correction and with the Fe bearing standards removed. As you point out there are some on-peak interferences from Fe L lines on the F Ka peak position. So removing those gives us this:
(https://probesoftware.com/smf/gallery/395_05_09_17_10_29_25.png)
The high outlier is Cr2O3 and I would suspect a small interference from Cr L also (as I tell my students, if you have a high outlier in the MAN curve it is usually an interference or contamination- so just remove it from the fit- after all background is by definition the lowest thing we can measure). So now we get this curve:
(https://probesoftware.com/smf/gallery/395_05_09_17_10_29_41.png)
It's not a terrible fit, but it is worth noting (as you did) that without the absorption correction the fit is better, but the TiO2 is way low as one would expect from the MAC. It might be worth looking at a wavescan of F ka on your barite standard and see if there is anything unusual in the background shape.
This is worth further study I think. It reminds me of something that Paul Carpenter once said: using the MAN background method is a good way to check for undeclared elements in one's standards. See here for that discussion:
http://probesoftware.com/smf/index.php?topic=4.msg499#msg499
john
-
Hi all
I saw this post with interest. F is something that our Geo's always want to measure, and I have resigned myself to never using the MAN fit for it due to the many L line interferences on it when constructing MAN fits. More recently however we have started to quant map fluorine in micas (which needs correction for Fe and a small Mg II order correction for high Mg micas), so I went on a quest to find a set of standards that would work OK for creating a MAN fit for F. I have attached a couple of the latest with/without some higher Z stds. Its seems to work OK, but bear in mind this is not quanting super low level F. It seems to have reasonable accuracy when compared to spot analysis of the same mineral down to ~0.5 wt%.
Haven't had a chance to refine it any further yet though, although I am sure there are some better standards I can use on my other standard blocks to fit it a bit better now that I know what to expect.
If you are interested this is the list of standards populating that MAN fit:
Astimex albite
Astimex Barite
Astimex Sanidine
In-house kyanite standard
Astimex BN
P&H Orthoclase
P&H Corundum
P&H Periclase
Astimex Gold
Astimex Hafnium metal
Astimex Zinc metal
Astimex Germanium metal
Astimex Cuprite
Cheers
P.S. forgot to mention, is anyone else getting that weird display bug as seen in the first image in the MAN fits when they display errors..?
-
Hi Ben,
This is all good stuff.
I haven't looked into F by MAN myself, so the fact that you and Ben are investigating this is quite interesting to me.
I see the display bug, and it looks like it's related to the display of error bars. I've seen this kind of thing before in other plots, so I will try to reproduce it myself so I can fix it, but if you don't mind could you send me your MDB file to my email? It's not very reproducible because I just looked at 15 or so MAN plots and none of them show that issue so far...
Thanks!
john
-
Hi John
Yes its definitely related to displaying the error bars. You are right its a little random when it will happen or not. Sometimes after excluding a standard and updating the fit it will appear, sometimes not. Sometimes after zooming in and out on the MAN fit it will appear, sometimes not...
I will email you the dropbox link for the displayed MDB file
Cheers
-
Thank you John and Ben, that's very helpful; Ben I think I'll try some of the metal standards you suggest; and John good point about TiO2 MAC.
-
I wonder if its the very high MACs for F in Ti and Cr that are the problem when using absorption correction. I've tried it for TAP crystal where interferences no longer problem - such as Fe. And Ti and Cr still elevated. - Picture to be uploaded
I guess the point is for F when generating MAN curve, avoid interferences and high MACs where possible
Ben
-
About time I said something. Too quiet for too long. I've taken to always doing F by MAN along with everything else. Ben W... your ideas on stds for this are good and in line with those I tend to apply too. Interferences stick out like a dog's balls and can be removed easily, but make sure you've plenty of standards to give choice. I've been doing this routinely for plenty of apatites - along with the STDI corrections to sort out issues for F analysis on apatite (and restricting the counting time to 40 s) - and phyllosilicates. I make abundant use of references to provide confidence in what I am up to.
-
Hi John
Yes its definitely related to displaying the error bars. You are right its a little random when it will happen or not. Sometimes after excluding a standard and updating the fit it will appear, sometimes not. Sometimes after zooming in and out on the MAN fit it will appear, sometimes not...
I will email you the dropbox link for the displayed MDB file
Cheers
Hi Ben,
I found and fixed the MAN display issue you noticed with regard to errors bars and the "Defaults" button. You should be able to update later tonight and all should be good.
john
-
About time I said something. Too quiet for too long. I've taken to always doing F by MAN along with everything else. Ben W... your ideas on stds for this are good and in line with those I tend to apply too. Interferences stick out like a dog's balls and can be removed easily, but make sure you've plenty of standards to give choice. I've been doing this routinely for plenty of apatites - along with the STDI corrections to sort out issues for F analysis on apatite (and restricting the counting time to 40 s) - and phyllosilicates. I make abundant use of references to provide confidence in what I am up to.
Hi Malc,
If you get a chance it would be interesting to see a few of your MAN plots for fluorine.
john
-
Hi all
Following on from Ben's data, I have collected a bit more data for F on PC0 and TAP xtals, and attached images of the MAN fits on PC0 and TAP with/without continuum correction. I think at least on my setup, I arguably get a better fit with absorption correction (esp. in the specific higher Z stds I am using).
Interestingly for both PC0 and LTAP, std 559 (TiO2) sits above the plot with continuum absorption correction ON, much like your data. As per the database there is a 6th/7th order overlap of Ti on F for PC0...buts its so miniscule and high order that it has to be inconsequential? There is no reported overlap for Ti on F for TAP?
On PC0, when the absorption correction is removed, the point sits far below the fitted line. Conversely on LTAP when absorption correction is removed the point sits in line with other data.
Any thoughts..?
Cheers
-
It has to be an interference from Ti L or oxygen K. Anyone have a high precision scan of F Ka in TiO2 handy?
john
-
I have one and it's not super high resolution or super high count times (steps of 4 units, 4 sec/point), but I still can't see anything obvious in this scan on one of my TiO2?
Cheers
-
Here you go John... as requested. One of the more recent ones.......
-
Hi Malc,
It would be really interesting to see how your TiO2 standard plots on this curve. Do you have a F ka MAN plot with TiO2?
John
-
It would also be interesting to see how Ti metal plots up on these F Ka MAN curves, but it would be important that the Ti metal is freshly polished at least. The F Ka emission volume is very shallow, so the oxidation layer should be as thin as possible so the continuum absorption calculation is accurate.
-
With regard to Ben Wade's MAN plots for F Ka on TAP and the TiO2 outlier, here are some more of my wandering thoughts.
1. The MACs for F Ka in TiO2 are quite large. For F Ka in Ti the tables MACs (from CalcZAF) are:
MAC value for F Ka in Ti = 14588.35 (LINEMU Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for F Ka in Ti = 14540.00 (CITZMU Heinrich (1966) and Henke and Ebisu (1974))
MAC value for F Ka in Ti = .00 (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for F Ka in Ti = 14986.59 (MAC30 Heinrich (Fit to Goldstein tables, 1987))
MAC value for F Ka in Ti = .00 (MACJTA Armstrong (FRAME equations, 1992))
MAC value for F Ka in Ti = 14277.71 (FFAST Chantler (NIST v 2.1, 2005))
MAC value for F Ka in Ti = 14500.00 (USERMAC User Defined MAC Table)
and for F ka in O they are:
MAC value for F Ka in O = 12439.63 (LINEMU Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for F Ka in O = 12390.00 (CITZMU Heinrich (1966) and Henke and Ebisu (1974))
MAC value for F Ka in O = .00 (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for F Ka in O = 11927.75 (MAC30 Heinrich (Fit to Goldstein tables, 1987))
MAC value for F Ka in O = .00 (MACJTA Armstrong (FRAME equations, 1992))
MAC value for F Ka in O = 11863.62 (FFAST Chantler (NIST v 2.1, 2005))
MAC value for F Ka in O = 12400.00 (USERMAC User Defined MAC Table)
I don't have Ben's MDB file handy at the university, but yesterday from home I tried selecting the lowest MAC table values (FFAST), and the MAN fit for TiO2 (#559) was still about the same amount over the trend line. So it doesn't seem possible that the MAC could be off that much.
Interestingly if we look at the calculated MAN absorption corrections as seen here (from running PFE in DebugMode):
MAN fit data for Mg ka, SP4 TAP, 15 keV
Std J MANSet Npts Z-bar Cps AbsCor
504 1 1 1 14.3913 .715813 1.58141
501 2 1 2 10.7106 .523931 1.49128
541 3 1 3 12.0184 .548867 1.43226
559 4 1 4 16.3937 .584618 1.83249 <- TiO2
578 5 1 5 20.5879 .668368 2.41281
701 6 1 6 10.7190 .618357 1.33443
503 7 1 7 13.4430 .615176 1.54753
549 8 1 8 6.12880 .349537 1.25008
571 9 1 9 11.7470 .557961 1.40367
575 10 1 10 10.6460 .638117 1.31164
589 11 1 11 14.0000 .831370 1.09576
594 12 1 12 4.00000 .280255 .964030
602 13 1 13 27.0000 .713668 3.01841
619 14 1 14 21.0000 .834364 1.82118
514 15 1 15 26.6522 .592982 3.27371
we can see that the largest corrections are for F Ka in some of the other standards. So it could be the MAN continuum absorption correction, but why is TiO2 more of an outlier than the other standards with larger corrections?
Note: many years ago I attempted to apply a true continuum absorption correction using Myklebust's continuum expression, but I got worse regression results than just using the normal phi-rho-z bulk absorption correction...
-
I split off the rest of the discussion regarding this F Ka ROI mystery peak artifact on TAP crystals (and possibly related standards containing Ti) to a new topic:
http://probesoftware.com/smf/index.php?topic=992.0
This is because we suspect this mystery peak to be a diffraction artifact of some kind. Another topic of interest could also be this one:
http://probesoftware.com/smf/index.php?topic=611.0
on JEOL H-type crystal artifacts, or this one on peak shape changes over time:
http://probesoftware.com/smf/index.php?topic=854.0
john
-
There are times (like right now) where I have been using a MAN fit I created days ago, and there is no need to go thru revising it. Someone is asking me to show/prove that it as good as I say it is. However, I see no way where I can simply show the display. Nor is there any where where I can find a listing of which standards I ended up using for each element MAN fit, nor of the fit parameter.
Is there a feasible way that these plots could be stored somewhere? (There are many many Display options under the "Run" tab.) Likewise there are many List options under the same Run tab; might the MAN standard options be listed? Or is all of that data volatile?
-
Hi John,
I don't see a problem with using the existing MAN plot dialog. You don't have to "revise" anything, just click on each element to see the MAN plot for it, and optionally click the Copy to Clipboard button to capture the plot graphics to paste into a document or whatever.
If you accidentally change some assignments in the MAN dialog and don't want to have them saved, just click Cancel button instead of the OK button to exit.
Am I missing something?
john
-
Roger.
-
I am doing some quality control checking of MAN results at minor and trace el levels: Mg, Sr, Mn and Fe in carbonates (using defocused 20 micron beam). I have wavescan data to compare the MAN-predicted bkg counts with. Sr, Mn and Fe MAN numbers match nicely with the wavescan numbers. The particular sample which I am testing has a mean Z of 12.6.
From both wavescans and discrete background measurements on the sample (which has a very small distinct peak; backgrounds are clear), the background should be 12.5.
However, the MAN fitted background, using the continuum absorption correction, gives a background of 17, which doesn't seem correct. If I turn off the cont. abs. correction, there is a wonky curve but the value at a mean Z of 12.6 is closer to the actual measured background level.
(Note: conincidence that the carbonate Z is 12.6, and the actual back ground is 12.5 c/s/20nA)
Again all other 3 elements, Sr, Mn and Fe work great with the con. abs. corrected MAN values being what the wavescans show.
Could there be something weird going on for Mg?
MAN stds were acquired in normal acquisition mode (not TDI). Carbonates later were run in TDI.
Attached are 2 plots
-
Hi John,
Both plots are correct. But it is a little unintuitive, so good question!
The plotted (absorption corrected) MAN intensities in the Assign MAN Fits dialog are the "ideal" background intensities. They cannot be compared to actual measurements. But by turning off the absorption correction in the MAN dialog, you are seeing the raw measured values, however which cannot be fitted together, because each MAN standard has different matrix physics. But if the unknown matrix is the same as an individual MAN standard, then you can compare this "raw" intensity to off-peak measurements, as you observed. Please re-read the Donovan, et al. 2016 paper in Amer. Min for a complete explanation:
http://epmalab.uoregon.edu/publ/A%20new%20EPMA%20method%20for%20fast%20trace%20element%20analysis%20in%20simple%20matrices.pdf
In short, this is because each MAN standard is a different matrix and therefore needs to be corrected for the specific emission line individually before fitting. Then once we have the equation of fit for these "ideal" intensities for the MAN standards, we can apply that fit to an actual sample after it is "de-corrected" for the actual absorption in the sample.
If you want to compare the calculated MAN background intensity with off-peak measurements, see the procedure here:
http://probesoftware.com/smf/index.php?topic=4.msg189#msg189
Then compare the line in the output labeled "UNBG:". Which are the MAN *or* off-peak intensities for the sample.
john
-
Hi John,
It is my understanding that if we do have a MAN correction curve for an element, and if we do have analyzed an element with 2-point background, it should be possible to switch between the 2-point background correction and the MAN. And even if it is not the case, we could use the 2-point background for determining a MAN background correction curve.
My problem is that - for testing purpose - I want to switch between the 2-points background and the MAN background, yet I can NOT change the background type in elements/cations window. The background field are all grayed out, and I cannot change a thing...
(https://i.ibb.co/stYrhL0/Screenshot-2019-01-28-at-13-21-55.png) (https://ibb.co/jHXFd7c)
I do have checked the option "Use MAN correction for Off-peak measurement..." in menu Analysis, and I do have a correct set of MAN correction (the program indeed asked me to reset the MAN correction, which I did).
(https://i.ibb.co/fXG0f8N/Screenshot-2019-01-28-at-13-20-55.png) (https://ibb.co/zmHfp82)
(https://i.ibb.co/0y7PH1x/Screenshot-2019-01-28-at-13-24-27.png) (https://ibb.co/1GgHFBk)
Maybe I'm doing something wrong? Note that at first, I was using the “shared background” in this run. To enable the MAN background, I did the following:
- REMOVE the shared background
- Activate the “Use MAN correction for off peak elements”
- Hit the “Clear All MAN assignments”
- Re-assigned the MAN fits.
Nonetheless, in each case (with shared bkg or with linear 2-pts bkg after resetting the MAN fit), the MAN option in elements/cations remain unavailable… Am I doing something wrong? Any help would be greatly appreciated!
Julien
-
Hi Julien,
I think you are basically doing it correctly. You don't need to change anything in the Elements/Cations dialog. Those options are for the background *acquisition* type (we'll modify the description to make it more obvious!).
I don't quite understand how you are seeing the MAN curve fits if you acquired off-peak standards and the Use Off-Peak Elements for MAN Fits menu is unchecked. Maybe you had it checked earlier at some point?
Anyway, see here for an explanation of using standards acquired using off-peak bgds, but for the MAN fit:
https://probesoftware.com/smf/index.php?topic=4.msg189#msg189
The steps to use MAN backgrounds based on your off-peak acquired standards, is to first check the Use Off-Peak Elements for MAN Fits menu and then assign the MAN fits. This allows the program to use standards acquired with off-peak backgrounds, for the MAN assignments, by "throwing away" the off-peak measurements on those standards.
Next if your samples are acquired using MAN backgrounds you are done. But if your samples are off-peak, then you need to check the next menu down which is the Use MAN Correction for Off-Peak Elements menu. This will force the program to utilize MAN background corrections even if the sample (std or unk) data was acquired using off-peak backgrounds.
To get real fancy you can also check the Use Interpolated Off-Peaks for MAN Fit checkbox in the Analytical | Analysis Options dialog as described here:
https://probesoftware.com/smf/index.php?topic=987.0
With this method, first suggested by Ben Hanson at Corning, the program obtains the MAN background intensities by interpolating the off-peak measurements. This option can be very useful when you want to use MAN backgrounds for faster acquisition (especially quant x-ray maps), but your standards are contaminated with traces of the element of interest, for example using MAN fits for O ka!
I hope this makes sense, but let me know if you are still having trouble. Feel free to also send me your MDB file.
john
-
Hi John and all
My wish would be to be able to (post acquisition) mix and match elements with MAN and off-peak post acquisition rather than all one or the other... I am sure it probably wouldn't be simple to implement though...
Cheers
-
Hi John and all
My wish would be to be able to (post acquisition) mix and match elements with MAN and off-peak post acquisition rather than all one or the other... I am sure it probably wouldn't be simple to implement though...
Cheers
Hi Ben,
Yeah it would be more than a little bit of work, I'll have to think how it might be possible. But you can always do what I do, which is decide which elements should be MAN and which elements should be off-peak *before* you acquire data! ;)
john
-
Thanks John,
I agree with Ben and it might be possible if you would enable the background fields two-point, MAN, MPB... in the Elements/Cations window. Then you could ask your software to check which option is being selected in the Elements/Cations window, and if MAN is selected (and if a MAN correction is available), then it uses the MAN for the background and if the two-point option is selected, then it would use the two-points. This solution could be further extended (with certainly more work...) for the MPB and more specifically shared background. Right now, you have an option to enable or remove the shared background, but if you do have a shared background, and if the MPB option in Elements/Cations is set to MPB, then it uses the shared background. Otherwise it would use the classical two-point background or the MAN background. Of course, if you do so, the software should be able also to determine if (a) MAN background is available, and (b) if a set of MPB or shared background is available.
This solution might also add some more clarity: the user can simply check in the Elements/Cations to see what background correction is being used for that specific element, whereas the current solution (your answer to my question) is done by clicking an option in a menu.
Julien
-
Hi,
I have been trying to use some advanced measurement (or rather processing) techniques, i.e. multi-point background fitting, interference corrections and (now) mean atomic number background calculation WITHOUT Probe for EPMA (Punk’s not dead!). Although it is quite demanding, I have been quite successful in some example tasks (in all these approaches) having only measured intensities, CalcZAF and MATLAB.
Measurement of Pb in my complex material (REE-, actinide-, Zr-, Ti-, Nb-rich minerals) seems to be very complicated. Recently, I have been trying MAN method. So, here I share my experience with fitting the Z vs background curve without PfE, but I also have a question.
I carefully selected some standards for background measurement at PbMb position (c. 58010 on LPET of our Cameca SX100). The intensities were measured at 15kV, 180 nA for 240-300 s.
The approach of how to calculate the continuum absorption is mentioned in Ware and Reed (1973), which is referred in Donovan et al. (2016). However, for better understanding, I recommend a book “Scanning Electron Microscopy and X-ray Microanalysis” by Goldstein and co-workers and its electronic supplement, where the method of Philibert, Duncumb and Heinrich is explained in detail. At first, I tried to calculate their example on NiKa absorption in Ni-Fe alloy in Excel, then I got the same results in CalcZAF. Continuum absorption is calculated similar to the absorption of a characteristic line with the exception of zero concentration of the emitter in the former case. The values of f(chi), which is the fraction of the generated x-rays, which were not absorbed, can be simply and quickly calculated in CalcZAF. First of all I selected Philibert/Duncumb-Reed matrix correction and MAC table (Ware and Reed refer to Heinrich, 1966; however MAC’s for PbMb are not published in this set – see below). Then, I loaded a standard from our dataset, e.g. GdPO4. Subsequently, I added PbMb line to the calculation with Pb concentration set to zero. After the calculation, the value of f(chi) for PbMb can be found in results. This approach seems to work well.
(https://probesoftware.com/smf/gallery/1783_07_06_19_1_29_52.jpeg)
Here is my average Z vs corrected background intensity diagram:
(https://probesoftware.com/smf/gallery/1783_07_06_19_1_32_31.jpeg)
Somewhat enigmatic for me are the deviations of the two Zr-bearing standards, ZrO2 and pure Zr. I have studied a WDS scan of a wide region around PbMb performed on a pure Zr, but there should be absolutely no noticeable zirconium (or some contaminant) peak at that position – according to the scan as well as according to the x-ray spectral database. Any idea for this behaviour?
(https://probesoftware.com/smf/gallery/1783_07_06_19_1_34_05.jpeg)
According to calculated absorption corrections, Zr is a somewhat stronger absorber (f(chi)=0.55) than most other standards, at least those with similar Z. Surprisingly, the raw background intensities before the correction seem to be much closer to the general trend of the most other standards (except for TiO, which shows much higher f(chi)=0.85 = much weaker absorber that the others, which usually have c. 0.6-0.7 or lower).
(https://probesoftware.com/smf/gallery/1783_07_06_19_1_35_40.jpeg)
One observation I find interesting or surprising: I tried to explain the deviation of the Zr-bearing standards in my diagram. So, I decided to compare the values of all the MAC’s of PbMb in Zr (originally, I selected Heinrich, 1966) among various datasets. Then, I found out, that there are no values for absorption of PbMb in the dataset of Heinrich, 1966 and CalcZAF automatically took MAC’s from a different dataset (McMaster, I think). The MAC's for PbMb in Zr from different datasets, however, are relatively similar, so the backgrounds of Zr-bearing still deviate.
One more question, on PfE – is the continuum absorption correction done using the approach by Philibert (1963) as referred in Ware and Reed (1973) in all cases or can it be switched to a different approach?
Best regards, Jakub Haifler
-
Hi Jakub,
Wow. Very impressive work. I must say, I think you would really have a blast with PFE on your instrument, maybe someday?
As to why the Zr containing materials are behaving like this, there are a lot of parameters involved here so it will be interesting to figure this out. As you point out, the mass absorption coefficient values being the most dominant for this correction. I will have to try some MAN measurements myself, using the Pb Mb emission line, and see what I can see, but in the meantime I will try to answer a few of your questions as much as possible.
But first I ran a *simulation* of these MAN curves in Probe for EPMA (in demo mode) and got this MAN curve for your standard calibration materials:
(https://probesoftware.com/smf/gallery/395_07_06_19_2_38_24.png)
Turning off the continuum absorption correction I got this:
(https://probesoftware.com/smf/gallery/395_07_06_19_2_38_39.png)
Note that stds 40 (Zr oxide) and 540 (Zr metal) both drop below the curve without the absorption correction, so that is consistent with your measurements. The default MAC table in Probe for EPMA (and CalcZAF) is Henke (for emission lines less than 10 keV). Looking in CalcZAF we can see the MACs for this emission line in Zr:
MAC value for Pb Mb in Zr = 2580.40 (LINEMU Henke (LBL, 1985) < 10KeV / CITZMU > 10KeV)
MAC value for Pb Mb in Zr = .00 (CITZMU Heinrich (1966) and Henke and Ebisu (1974))
MAC value for Pb Mb in Zr = 2750.53 (MCMASTER McMaster (LLL, 1969) (modified by Rivers))
MAC value for Pb Mb in Zr = 2620.36 (MAC30 Heinrich (Fit to Goldstein tables, 1987))
MAC value for Pb Mb in Zr = .00 (MACJTA Armstrong (FRAME equations, 1992))
MAC value for Pb Mb in Zr = 2528.02 (FFAST Chantler (NIST v 2.1, 2005))
MAC value for Pb Mb in Zr = 2620.00 (USERMAC User Defined MAC Table)
The Henke MAC is quite a bit smaller than the MAC30 table value which you used. So this might cause the over correction you are seeing for the Zr containing materials.
As to what continuum absorption correction is being utilized in Probe for EPMA for the MAN curve fit, it is whatever absorption correction is currently selected in the ZAF/Phi-Rho-Z method. Originally we tried several absorption corrections specifically intended for continuum emissions, but we got worse results than just using the absorption corrections from the currently selected matrix correction. So now the MAN method in PFE just uses the absorption correction from whatever matrix correction has been selected.
By the way, you might want to contact Julien Allaz, Mike Jercinovic and Karsten Goemann as both have worked on using the MAN method with complex high Z materials. All three will be at the QMA 2019 conference in Minnesota later this month and all will be giving presentations on various background correction methods including MAN, off-peak and MPB (multi-point backgrounds) and Karsten will present the MPB method with "shared" backgrounds (that is utilizing normal off-peak measurements from other emitters measured on the same spectrometer).
Also see these topics:
https://probesoftware.com/smf/index.php?topic=4.msg17#msg17
https://probesoftware.com/smf/index.php?topic=9.msg7764#msg7764
Or maybe your materials are slightly contaminated in Pb? I can assure you that some of the Smithsonian REE standard are contaminated with Pb! Note how my MAN curves excluded these materials automatically because it knows from the standard composition database that these materials contain Pb!
john
-
Hi John,
thank you very much for your answer. As you wrote, PfE would be a great help, particularly for such complicated measurements I am doing at the moment. On the other hand, these are extraordinary and are only for my research, I do not perform such calculations for routine measurements. So, the increased effort is acceptable in this case. Unfortunately, I am not powerful enough to affect what software and hardware equipment we have in our laboratory. But as you wrote – maybe sometime. I hope, my little DIY is not going to break your business with PfE :), but may allow some people, particularly students, to try how the method works.
As for my data: I did some additional measurements. Something is clearer now, something not. I did two 5-point background measurements on Zr (btw, we have two pure Zr standards – intensities in both somewhat differ, but neither is compatible with my MAN curve). I observe that intensities at PbMb position are c. 7% higher than what I would expect from the exponential fit through the other four points. So, some contamination seems to be highly probable (may be Pb, Mo, Bi, Nb... at this position). On the contrary, the theoretical background at PbMb calculated from the fit remains incompatible with the MAN curve.
(https://probesoftware.com/smf/gallery/1783_12_06_19_3_29_13.jpeg)
So, after additional measurements, following 12 standards are compatible within a single fit: YAG, Fe2O3, YPO4, LaPO4, GdPO4, GdAlO3, MnTaO4, CrTa2O6, CaWO4, Hf, Ta, W. I excluded TiO and Ti, as TiKb(II) shows overlap with PbMb. Similar to Zr, Sn also have PbMb-intensity higher than a fit through 4 other points nearby, so the contamination may also occur. Pure elemental standards Cr, Fe are not compatible with the MAN curve (they give higher PbMb-intensities), but I do not have additional data for resolving the reason.
(https://probesoftware.com/smf/gallery/1783_12_06_19_3_30_41.jpeg)
Very mysterious for me is the behaviour of cheralite, CaTh(PO4)2, intensity of which shows very deep “negative” anomaly. It has avg. Z of 51.158, but PbMb-intensity of only 1.24 cps/nA (based on 3 measurements). CaWO4 and MnTaO4 with very similar avg. Z give c. 1.53-1.54, which is the theoretical, MAN-curve-compatible, background value at that Z.
I read some older comments on the MAN method (from about 2013). According to them, the method works well for low-avg Z matrices, while not so well for high-avg Z, which is not a very good message for me. However, I have not found much information on that. I think, my MAN curve looks relatively nice (of course, I have not explained incompatibility of Fe, Cr and particularly cheralite yet). Moreover, the used standards are relatively simple compounds (pure elements, binary or ternary compounds). So, I guess, the complex natural material is going to be much more tricky.
I am somewhat confused of a more general question of whether there are some serious limits and irregularities in this method. Or what should be the main pitfall for the high-Z matrices?
One example: from the many posts in this thread we can see that the MAN curve is usually fitted by a linear/exponential/2nd order polynomial, fluently increasing, function. This seems to be a rule: the higher Z, the higher the (corrected) background intensities. But for example in this post, John's data show a concave curve (which John explains there).
https://probesoftware.com/smf/index.php?topic=4.msg1070#msg1070
Best regards, Jakub
-
Hi Jakub,
Please understand that there is nothing extraordinary about using the MAN background correction. My guess is that, at least for geologists who utilize the software, the MAN correction is used almost all the time, at least for the major and minor elements (note that one can utilize a mix of MAN and off-peak elements for any given sample setup). The reason for it being utilized so often, is that the MAN correction saves so much time time (and improves precision) on major, and even minor elements, where very small accuracy errors in the MAN calibration curve, are generally smaller than the precision of the peak intensity measurement.
For typical geological samples with K emission lines and low to moderate average Z materials, this is usually the case. In my experience with typical silicates and oxides, the MAN background correction is accurate down to 200-300 PPM, even without a blank correction. The use of the MAN correction with higher Z matrices and high Z emission lines is still an area of active investigation. But maybe a little "MAN" history would be appropriate here...
The very first rationale for the MAN considerably preceded my own initiation to EPMA, and was invented to deal with EPMA instruments which utilized fixed monochromaters for commonly measured elements. For example, on my first EPMA instrument, which I inherited at UC Berkeley in the 1980s, it had 4 fixed monochromaters (for Si, Fe, Ca and Al), and 4 tunable spectrometers for another 4 elements. The reason of course being that fixed monochromaters *cannot* be de-tuned for the off-peak measurement! So people came up with the idea of utilizing a calibration curve of standards which did not contain the elements of interest, for the background correction for these "fixed" spectrometers. Which meant of course that using the MAN correction, we could measure 8 elements in 10 seconds! This was very useful as you might imagine for beam sensitive materials.
Since then the MAN background correction has been significantly improved and as my Amer. Min. 2016 paper showed, the MAN background correction can even be utilized for improving sensitivity for trace elements. Basically this results in a 40% or more improvement in detection limits, in 1/2 the acquisition time, because only the on-peak intensity is measured. And of course, of interest to you, one automatically avoids off-peak interferences, because there are no off-peak measurements when using the MAN correction! In addition, interpolation issues with non-linear backgrounds and/or absorption edges simply disappear with the MAN method because, again, only the on-peak position is utilized.
But even though the MAN method can result in better accuracy than off-peak methods when such off-peak issues are present, the main caveat for the MAN method remains accuracy (at least at trace levels), because we are not performing a direct measurement of the continuum on the sample. Therefore for ultimate accuracy, the MAN method is commonly combined with the so-called "blank" correction. The blank correction works well for relatively simple matrices where suitable (usually synthetic) blank samples can be obtained, e.g., quartz, zircon, pyrite, TiO2, Fe2O3, Fe3O4, etc. Of course for complex materials, zero blanks may be difficult to obtain. But it is also of interest to note that the blank correction as implemented in Probe for EPMA, can be applied to non-zero blank samples, so there is that option.
Now having said all that, the remaining practical difficulties of the MAN method is obtaining pure standards that do not contain the element of interest. As you pointed out in the linked post, many standards have not been properly characterized for trace elements. For example, I have two synthetic zircons, one appears to have a couple hundred PPM more phosphorus than the other. I have heard from many colleagues that utill they started utilizing the MAN method, they had not realized how contaminated some of their standard materials were. On the other hand, the MAN method can teach us quite a lot that we didn't already know about our standards!
Just as a small aside, Ben Hanson at Dow Corning came up with a really crazy idea for the MAN method, which is to utilize the *off-peak* interpolated intensities for the MAN calibration curve, or "offset MAN" as he called it, as described here:
https://probesoftware.com/smf/index.php?topic=987.msg6447#msg6447
"Crazy as a fox" as they say! The reason he requested this because he found that many of his standards weren't as pure as he thought they were. I thought it was a good idea for another reason: sometimes I've wanted to utilize the MAN method for oxygen analysis, and it's difficult to find standards that don't contain even trace oxygen (since we live in an ocean of oxygen)! Whatever.
It boils down to this: almost without exception, the lowest intensity one can measure (by definition!) is the background. Yes, there are possible sample and/or detector absorption edges and "holes" in the background continuum due to secondary Bragg diffraction (e.g, the hole in the continuum seen in FeS2 near the Au Ma position and in some PET crystals at the Ti Ka position, though there is still some debate about this), but these are fairly rare. So in general, any standards that fall above the general trend of the MAN calibration curve are either from contamination (from the element being present), or an interference from another analytical peak.
The general strategy at the user level is to simply remove those (higher intensity) offending standards from fit of the MAN curve. Now because you are conducting a bit of a research project here with your high Z materials, you are interested in exactly what is causing these outliers in the MAN calibration curve. And that is great and you should pursue this line of inquiry because it is interesting and helpful for greater understanding of the details.
As to particular difficulties with high Z matrices, the main difficulty is due to the fact that the P/B ratios are worse for high Z matrices since from Kramers Law we know that we produce more continuum in higher Z materials. Second, when analyzing high Z emission lines, we are generally restricted to L or even M emissions and these are also typically lower P/B than K emissions. So combined this means that the calibration of the MAN curve is even more critical for high accuracy work in high Z materials, especially at low concentrations.
That said, there are advantages to using the MAN method even in complex high Z materials, as has been pointed out (no off-peak interferences and no interpolation). But I think other approaches such as the MPB (multi-point background) and even "shared" bgds can be useful for such matrices. I believe Karsten Goeman and Julien Allaz have published these methods and also have examples of these alternative off-peak methods posted elsewhere on the forum.
-
Just wanted to point out that the MAN fit calculations in the Assign MAN Fits dialog in Probe for EPMA can be modified to utilize the more physically accurate electron fraction or Z fraction averaging method. Of course, this only pertains to average Z calculations for compounds.
That is to say, when compared to the traditional mass fraction based methods for calculating average Z. The mass fraction averaging method includes the mass of the neutrons, which has little to no effect on continuum (or backscatter) production, and therefore adds arbitrary errors to average Z calculations. See attached abstract from 2019 below (login to see attachments).
Here is an example for the Ca Ka emitter which is a fairly moderate energy x-ray using the traditional mass fraction average Z method:
(https://probesoftware.com/smf/gallery/395_10_02_21_12_09_19.png)
And here is the same data, utilizing the electron or Z fraction method, specifically Z^0.7 electron fraction averaging:
(https://probesoftware.com/smf/gallery/395_10_02_21_12_09_50.png)
Why the 0.7 exponent? It's because that is necessary to account for the screening of nuclear charge by the inner orbital electrons in higher Z atoms.
In fact this is why we see the backscatter vs. Z trend start to "curve over" around atomic number 30:
(https://probesoftware.com/smf/gallery/395_10_02_21_12_24_38.png)
-
Hi all,
I’ve found myself in a predicament with MAN background and interference corrections, and I’m hoping someone can point out the glaring error of my ways.
I was measuring Co in Fe-Ni meteorites on our Cameca by both off peak quant points and MAN-corrected quant maps and have come across a disparity with the results for the Co to the tune of around 2000ppm.
As the Cameca is currently busy doing other things, I decided to test this out using our JEOL to see if I could replicate and simplify the problem using just the pure Fe and Co standards.
I measured Co Ka and Fe Ka on Fe metal and Co metal using off peak backgrounds. The Fe region is pretty clear and I can use a linear fit between the two background measurements, but the Fe Kb interferes with the Co Ka and the -ve side background ends up moving further up the Fe Kb shoulder. I could have used a -ve background on the low side of the Fe Kb, but that sits on the other side of the Fe K absorption edge:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_48_48.jpeg)
Zooming in on the background positions for the Co Ka, the exponential fit still over-estimates the background intensity underneath the Co Ka (as seen with a wavelength scan over the Co Ka in the Fe metal standard):
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_04.jpeg)
This results in reporting ~250ppm negative values of Co in Fe metal:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_14.jpeg)
As I’m trying to figure out a solution for data that I have previously measured on the Cameca, I thought to switch everything over to MAN background subtraction.
MAN curves look ok:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_22.jpeg)
But now the same data as above, that was reading ~-250ppm, now reads +900ppm.
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_30.jpeg)
And adding Fe in to the MAN fit highlights that I’ve either (i) got Co in my Fe standard, or (ii) there is an Fe interference on the Co Ka.
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_38.jpeg)
I don’t think there is 900ppm Co in my Fe metal standard, and looking at the user’s data, I have acquired MAN calibrations on fayalite, pyrite and hematite alongside Fe metal and all show elevated Co content proportional to the Fe content of the phase.
Looking back at the wavelength scan of Co Ka in the Fe metal standard, I can see that there are two components to be subtracted from the Co Ka intensity: (i) the Bremsstrahlung background (roughly drawn on in maroon, and across the Fe K absorption edge) and (ii) the Fe Kb shoulder:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_48.jpeg)
At this point, you think “that’s easy, just add the Fe as an interference correction on the Co measurement!”.
Annoyingly, this isn’t working:
The Fe metal results with Fe interference correction on Co:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_55.jpeg)
(https://probesoftware.com/smf/gallery/796_07_05_21_8_50_04.jpeg)
And without Fe interference correction on Co:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_50_12.jpeg)
“Use Assigned Interference Corrections on Standards and Unknowns” is enabled globally:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_50_20.jpeg)
And I should add that I have tried both with and without "Use Interpolated Off Peaks for Man Fit".
Fe and Co were both collected on all standards on both the Cameca and JEOL runs, and neither are working.
The JEOL reports 1000ppm Co in the Fe standard when looking at the standards (as opposed to when measuring the Fe standard as an unknown) and the Fe standard is in the standard database as being pure Fe.
To summarise:
• Off-peak measurements are under reporting Co by ~250ppm
• MAN background subtracted measurements are over reporting Co by ~900ppm
• I can’t get the interference correction working for MAN-background subtraction of elements acquired with off peak backgrounds for data acquired on both our Cameca and JEOL instruments, even using the most up to date Probe for EPMA.
I really want to figure out where it (or more likely, me) is going wrong with the MAN-subtracted interference corrections!
Anyone have any thoughts?
-
Hi Jon,
I just quickly read through your post and think I understand what is going on. Specifically, the interference corrections are *not* applied to the MAN fit curves (it is possible to do in principle of course, but we have not implemented it).
What you want to do is remove all standards containing Fe from your Co ka MAN curves (use <ctrl> click to deselect them and click the Update MAN Fit button) because, yes, Fe Kb does interfere with Co Ka.
The other thing to keep in mind is that the *accuracy* MAN bgd correction is limited to around 100 to 200 PPM in the silicate-oxide z-bar range (10 to 20 or so), and is slightly worse for high average atomic number materials. So depending on your accuracy needs, this may be where the blank correction needs to be applied to improve accuracy further. See:
https://probesoftware.com/smf/index.php?topic=307.0
and
https://probesoftware.com/smf/index.php?topic=29.0
for more details. Or did I miss something?
-
Hi John,
I think the issue is that the interference correction isn't being applied to the MAN background subtracted measurements (that were acquired with off peak background measurements).
The MAN curves look good to me, and I've removed all the Fe-bearing phases from the Co Ka MAN fit.
I wish I did have a decent blank calibration (or even a decent FeNi primary or reference) standard! My worry with the real world samples though is that I'd need a dozen FeNi blank calibration samples to account for the "blank" X-ray intensity to change notably as the Fe content varies (as its on the Fe Kb shoulder for Co Ka).
-
Hi Jon,
You can see the magnitude of the interference correction for each element, in the log window output. Look for the line labeled INT%.
Note that depending on the situation you don't necessarily require a perfect matrix match for the blank correction. However, you should not attempt to correct an interference correction using the blank correction if you are already applying an interference correction, or you will get a double correction (the software should warn you about this I think).
Feel free to send me your MDB file and I can take a look.
john
-
Here's an example I found of a Fe Kb on Co interference correction on pure Fe:
Un 3 Iron metal, Results in Elemental Weight Percents
ELEM: Fe Ni Co Mo
BGDS: LIN LIN LIN LIN
TIME: 160.00 160.00 200.00 210.00
BEAM: 49.87 49.87 49.87 49.87
ELEM: Fe Ni Co Mo SUM
126 100.109 -.017 .000 .000 100.093
127 99.927 -.021 .000 -.001 99.905
128 100.363 -.030 -.003 -.001 100.330
129 100.134 -.009 .001 -.004 100.122
130 100.130 -.023 .001 .004 100.113
AVER: 100.133 -.020 .000 .000 100.113
SDEV: .155 .008 .001 .003 .151
SERR: .069 .003 .001 .001
%RSD: .15 -38.89-2445.43-1900.01
STDS: 526 528 527 651
STKF: 1.0000 1.0000 .9978 .1305
STCT: 454.61 475.74 1669.26 51.00
UNKF: 1.0013 -.0002 .0000 .0000
UNCT: 455.21 -.09 .00 .00
UNBG: 1.24 1.42 3.82 .50
ZCOR: 1.0000 1.1015 1.0174 1.2544
KRAW: 1.0013 -.0002 .0000 .0000
PKBG: 367.64 .94 1.00 1.00
INT%: ---- ---- -100.05 ----
And here is a NIST NiFe alloy that actually does contain 220 PPM of Co:
Un 7 NiFe NBS alloy, Results in Elemental Weight Percents
SPEC: Mn Si Cr Cu
TYPE: SPEC SPEC SPEC SPEC
AVER: .300 .320 .060 .038
SDEV: .000 .000 .000 .000
ELEM: Fe Ni Co Mo
BGDS: LIN LIN LIN LIN
TIME: 160.00 160.00 200.00 210.00
BEAM: 49.89 49.89 49.89 49.89
ELEM: Fe Ni Co Mo SUM
146 51.728 47.016 .020 .012 99.493
147 51.605 47.176 .020 .014 99.533
148 51.701 47.263 .019 .006 99.707
149 51.536 47.223 .021 .006 99.503
150 51.721 47.090 .015 .007 99.551
AVER: 51.658 47.154 .019 .009 99.557
SDEV: .084 .100 .002 .004 .087
SERR: .038 .045 .001 .002
%RSD: .16 .21 12.19 39.05
STDS: 526 528 527 651
STKF: 1.0000 1.0000 .9978 .1305
STCT: 455.29 476.37 1668.43 57.20
UNKF: .5553 .4482 .0002 .0001
UNCT: 252.81 213.48 .31 .03
UNBG: 1.12 1.54 4.12 .50
ZCOR: .9303 1.0522 1.0295 1.3263
KRAW: .5553 .4482 .0002 .0005
PKBG: 227.59 139.90 1.07 1.06
INT%: ---- ---- -80.46 ----
-
I finally found an example of using interferences with MAN backgrounds (works with both off-peak and MAN) and found this of the NIST mineral glass:
Un 3 NBS K-412 mineral glass, Results in Elemental Weight Percents
SPEC: Na Al Ca Ti Fe O B Sr Ba
TYPE: SPEC SPEC SPEC SPEC SPEC CALC SPEC SPEC SPEC
AVER: .043 4.906 10.899 .000 7.742 43.008 .000 .000 .000
SDEV: .000 .000 .000 .000 .000 .048 .000 .000 .000
ELEM: Mg F F Si P K Mn
BGDS: MAN MAN MAN MAN MAN MAN MAN
TIME: 240.00 240.00 240.00 40.00 90.00 60.00 240.00
BEAM: 49.77 49.77 49.77 49.77 49.77 49.77 49.77
ELEM: Mg F F Si P K Mn SUM
XRAY: (ka) (ka) (ka) (ka) (ka) (ka) (ka)
91 11.610 .001 .014 20.723 .003 .005 .059 99.031
92 11.628 -.007 .020 20.727 .004 .003 .056 99.062
93 11.648 -.017 .009 20.681 .006 .002 .062 98.985
94 11.622 -.010 .008 20.652 .004 .000 .059 98.876
95 11.619 -.018 .007 20.759 .003 .003 .058 99.092
96 11.623 .010 .001 20.652 .006 .004 .059 98.898
AVER: 11.625 -.007 .010 20.699 .004 .003 .059 98.991
SDEV: .013 .011 .007 .044 .001 .002 .002 .088
SERR: .005 .004 .003 .018 .000 .001 .001
%RSD: .11 -154.03 68.50 .21 26.55 68.39 3.08
STDS: 12 831 831 14 285 374 25
STKF: .4736 .1544 .1544 .4101 .1600 .1132 .7341
STCT: 724.20 50.52 380.32 268.26 122.73 222.19 119.00
UNKF: .0774 .0000 .0000 .1581 .0000 .0000 .0005
UNCT: 118.28 -.01 .06 103.40 .02 .05 .08
UNBG: .56 .15 4.13 .19 .12 .99 .17
ZCOR: 1.5029 3.9668 3.9668 1.3094 1.3980 1.0939 1.2044
KRAW: .1633 -.0001 .0002 .3855 .0002 .0002 .0007
PKBG: 212.57 .96 1.01 547.83 1.19 1.05 1.46
INT%: ---- -2.47 -17.60 ---- ---- ---- ----
TDI%: .099 -.627 .521 .000 .000 .020 -.513
DEV%: .0 .0 .7 .0 .0 84.1 .0
TDIF: LOG-LIN LOG-LIN LOG-LIN ---- ---- LOG-LIN LOG-LIN
TDIT: 257.00 257.50 258.67 .00 .00 78.67 259.33
TDII: 119. .143 4.20 ---- ---- 1.04 .253
TDIL: 4.78 -1.95 1.43 ---- ---- .0423 -1.37
The first fluorine channel is using LTAP, and the second fluorine channel is using a PC1 crystal. Note that the interference correction is larger with the PC1 LDE type Bragg crystal as expected.
-
Hi John,
Thanks for looking at this for me. I've been scratching my head about this for the last two days and I can't figure out where I've messed up!
I've not had a problem with either MAN or interference corrections before, so I'm lost as to where I've gone wrong. And then doubly confused that I've managed to do it again on a completely new setup on a different instrument!
Having checked that "Use Assigned Interference Corrections on Standards and Unknowns" is enabled in the Analysis options menu, I recalculate the Fe metal "unknown" and get the following:
Un 2 std6012_Fe_metal_test_01
TakeOff = 40.0 KiloVolt = 15.0 Beam Current = 10.0 Beam Size = 0
(Magnification (analytical) = 40000), Beam Mode = Analog Spot
(Magnification (default) = 100000, Magnification (imaging) = 100000)
Aperture Number: 2
Image Shift (X,Y): .00, .00
Number of Data Lines: 12 Number of 'Good' Data Lines: 12
First/Last Date-Time: 05/06/2021 07:27:49 PM to 05/06/2021 07:49:53 PM
WARNING- Using MAN Background Instead of Off-Peak Background Correction
Average Total Oxygen: .000 Average Total Weight%: 99.928
Average Calculated Oxygen: .000 Average Atomic Number: 26.001
Average Excess Oxygen: .000 Average Atomic Weight: 55.850
Average ZAF Iteration: 2.00 Average Quant Iterate: 3.00
Un 2 std6012_Fe_metal_test_01, Results in Elemental Weight Percents
ELEM: Fe Co
BGDS: MAN MAN
TIME: 20.00 20.00
BEAM: 10.00 10.00
ELEM: Fe Co SUM
51 99.817 .099 99.916
52 99.816 .100 99.916
53 99.645 .084 99.729
54 100.247 .097 100.344
55 99.404 .103 99.507
56 100.016 .097 100.113
57 99.772 .087 99.859
58 99.843 .105 99.947
59 99.476 .104 99.580
60 99.609 .099 99.708
61 100.267 .115 100.382
62 100.056 .082 100.137
AVER: 99.831 .098 99.928
SDEV: .277 .009 .278
SERR: .080 .003
%RSD: .28 9.62
STDS: 6012 6013
STKF: 1.0000 1.0000
STCT: 817.76 878.02
UNKF: .9983 .0010
UNCT: 816.43 .84
UNBG: 1.83 2.43
ZCOR: 1.0000 1.0209
KRAW: .9984 .0010
PKBG: 446.56 1.35
INT%: ---- ----
I see the INT% is recorded as ---- despite the interference correction being selected?
I'll email over the mdb file to see what you think.
-
Hi Jon,
Be sure to use the "Tt" button for text output so the columns line up!
This is a silly question but did you actually assign the Fe on Co interference in the Standard Assignments dialog?
-
This is a silly question but did you actually assign the Fe on Co interference in the Standard Assignments dialog?
Just looked at your MDB file and yeah, you didn't actually assign any spectral interferences! :o
I mean the software is good, but it can't read your mind! :P
-
I meant to say, I un-did all the corrections on the mdb file for you so that you had a blank database to go at!
-
Annoyingly, this isn’t working:
The Fe metal results with Fe interference correction on Co:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_49_55.jpeg)
Assuming this is right, of course!
-
Hi Jon,
OMG! You found an honest to God bug in the code!
It's because you first acquired your standards (and unknowns) using off-peaks. But because your backgrounds were not properly selected as seen here:
(https://probesoftware.com/smf/gallery/1_07_05_21_12_44_50.png)
whereas I would choose backgrounds as seen here (green lines):
(https://probesoftware.com/smf/gallery/1_07_05_21_12_45_05.png)
The next thing we both overlooked is that the program was warning us with this message (seen in red in the log window):
Warning- Negative interference counts (-0.2100179) for Co ka on interference standard 6012
Unfortunately the program only prints this warning message the first time it occurs because some people were complaining about too many warning messages!
Equally unfortunate is that because you wanted to try MAN backgrounds, even through you acquired your samples using off-peak bgds, you (correctly) selected to use the two Analytical menu check menus (Use Off-Peak Elements for MAN Fits and Use MAN Correction for Off-Peak Elements). Which normally works just fine.
But, the code that loads the interference correction counts (that is, the pseudo Co counts from the Fe Kb interference), did *not* take into account the fact that the Use Off-Peak Elements for MAN Fits flag was checked! It does respect this flag in the load MAN intensity code, but somehow I never implemented it into the load standard intensity (and standard interference intensity) code! :-[
I guess the reason no one (until now) has noticed this is that normally one either acquires things using off-peak acquisitions or using MAN acquisitions. In other words, if you had properly selected the Co off-peak backgrounds, or created a new run using MAN acquisitions, the buggy code would have worked and we wouldn't have found this issue.
So I salute sir!
So here is the old code for loading the standard intensities:
' Correct the data for dead time, beam drift and off-peak background (results returned in sample(1).CorData!()) ****commented out 05/07/2021 because it was not handling the Use OffPeakElementsForMANFlag properly****
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(0), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(2), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(2), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
and here is the corrected code that handles this flag (as it already does everywhere else in the code):
' Correct the data for dead time, beam drift (and off-peak background if NOT using off-peak sample elements) (results returned in sample(1).CorData!())
If UseOffPeakElementsForMANFlag = False Then
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(0), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(2), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(2), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
Else
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(1), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(3), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(3), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
End If
I do vaguely remember looking at this code when I added the Use Off-Peak Elements for MAN Fits menu feature, but at the time I was just thinking that this only needed to be added to the load MAN intensities code. But if you're loading standard intensities using this MAN menu flag, then you really need to do it for the standard and interference standard intensities also. Doh! >:(
Basically the program was still calculating the interference intensities for the Co Ka in Fe using the off-peak method, which had the poorly chosen off-peak positions. And yes, it was also loading the standard intensities using off-peak intensities, but of course those are usually very high intensities and the bgd correction would be very small, and so I suspect that is why no one ever noticed this when using the Use Off-Peak Elements for MAN Fits menu flag for their off-peak acquired standards!
Anyway here are the results with the new code:
Un 2 std6012_Fe_metal_test_01
TakeOff = 40.0 KiloVolt = 15.0 Beam Current = 10.0 Beam Size = 0
(Magnification (analytical) = 40000), Beam Mode = Analog Spot
(Magnification (default) = 100000, Magnification (imaging) = 100000)
Image Shift (X,Y): .00, .00
Number of Data Lines: 12 Number of 'Good' Data Lines: 12
First/Last Date-Time: 05/06/2021 07:27:49 PM to 05/06/2021 07:49:53 PM
WARNING- Using MAN Background Instead of Off-Peak Background Correction
Average Total Oxygen: .000 Average Total Weight%: 99.708
Average Calculated Oxygen: .000 Average Atomic Number: 26.000
Average Excess Oxygen: .000 Average Atomic Weight: 55.847
Average ZAF Iteration: 1.08 Average Quant Iterate: 4.00
Un 2 std6012_Fe_metal_test_01, Results in Elemental Weight Percents
ELEM: Fe Co
BGDS: MAN MAN
TIME: 20.00 20.00
BEAM: 10.00 10.00
ELEM: Fe Co SUM
51 99.701 -.006 99.696
52 99.700 -.004 99.696
53 99.529 -.020 99.509
54 100.130 -.007 100.122
55 99.288 -.001 99.288
56 99.899 -.008 99.892
57 99.655 -.017 99.639
58 99.726 .001 99.727
59 99.359 .001 99.360
60 99.492 -.005 99.488
61 100.149 .011 100.160
62 99.938 -.022 99.916
AVER: 99.714 -.006 99.708
SDEV: .277 .009 .277
SERR: .080 .003
%RSD: .28 -148.00
STDS: 6012 6013
STKF: 1.0000 1.0000
STCT: 818.74 879.45
UNKF: .9971 -.0001
UNCT: 816.43 -.05
UNBG: 1.83 2.43
ZCOR: 1.0000 1.0209
KRAW: .9972 -.0001
PKBG: 446.59 .98
INT%: ---- -107.45
I am re-compiling now and will upload this code fix in a few hours...
-
OMG! You found an honest to God bug in the code!
Good news for me: I thought I was going mad! Thanks for looking in to this!
Although, as you said, it only came to light because I messed up the background position…
I did contemplate putting the low side background on the other side of the Fe Kb, but given the samples are >90wt% Fe, I was concerned about the influence of the Fe K absorption edge on the background intensity:
(https://probesoftware.com/smf/gallery/796_07_05_21_8_48_48.jpeg)
The above is a wavescan over the Co Ka region on Fe metal. Co Ka position in red, Fe K edge in orange, background positions in green. (the above plot is from the multipoint background window, is there a way to put absorption edges on the Plot! window wavescans?)
I decided to keep the low side background on the same side of the Fe K edge (and Fe Kb) as the Co Ka, and let the exponential background take care of the curvature. That was the plan, anyway...
Thanks again for sorting this so quickly – this would’ve been playing on my mind all weekend!
-
Yeah, trying to deal with interferences using a fancy background fit is something that I've tried (and recently discussed with our new consultant Will Nachlas), but it just doesn't appear to work most of the time.
Better to get good off-peak background positions and do the interference properly, or even better, just use the MAN background method! Then one never has to worry about off-peak interferences!
Yes, you can display absorption edges in the Plot! window. Just click the Load X-ray Database button, and check the Show Absorption Edges checkbox, then click Close and Zoom Full to re-draw the graph (see orange marker):
(https://probesoftware.com/smf/gallery/1_07_05_21_2_01_27.png)
-
OK, we just uploaded v. 12.9.6 of Probe for EPMA which now utilizes the MAN background correction not only for the standards used for the MAN background curve fit, but also for the background correction of the standard intensities and interference standard intensities. Use the Help menu as usual to update Probe for EPMA.
The point being that when the Use Off-Peak Elements for MAN Fits menu is checked it was using off-peak elements for the MAN curve (by only utilizing the on-peak intensities from the off-peak acquisitions), but it was still using off-peak background correction for loading the standard and interference standard intensities.
So when Jon had the off-peak positions improperly located on the edge of the Fe Kb peak, there were no (positive) interference intensities found and so the program did not apply the interference correction to the sample (what would be the point?).
Now one *could* argue that there should be a separate flag for the background correction of the standard and interference standard intensities, but I think it makes sense that if you're using off-peak elements for the MAN corrections they should be applied to the primary standard (and interference) intensities also. One still has the option to acquire the unknowns only using MAN acquisitions.
I also adding a warning if one has turned on the interference corrections in the Analysis Options dialog, but no specific interference corrections were assigned in the Standard Assignments dialog.
-
We made one small change to the code today that I think makes explaining what we are doing with these MAN menu flags much easier, and that is what this post will be about. If you have been following the discussion between Jon Fellowes and I above, read on. If not, you might want to review the above posts. Here I will be discussing these two MAN menu flags found in the Analytical menu as seen here:
(https://probesoftware.com/smf/gallery/1_08_05_21_3_09_35.png)
Remember, both of these two MAN flags in the Analytical menu only affect the *analysis* of standard or unknown samples, *not* the acquisition of standard and unknown samples! So please keep in mind that when I use the words "acquire" or "acquisition" I mean the actual process of acquiring sample data (from the Acquire! or Automate! windows), but when I use the words "analyze" or "analysis", then that refers to the process of quantifying already acquired sample data (from the Analyze! window or Output menu).
In short when the "Use Off-Peak Elements for MAN Fits" menu is checked, it instructs Probe for EPMA to utilize off-peak elements acquired in standard samples for constructing the MAN background curve fit. Checking the "Use MAN Correction for Off-Peak Elements" menu instructs the software to perform an MAN background correction, when loading primary standard (and interference standard) intensities, and also when performing the background correction during quantification of standard or unknown samples.
So here are the four situations explaining how these (quite useful) MAN menu analysis flags can be utilized:
1. Both "Use Off-Peak Elements for MAN Fits" and "Use MAN Correction for Off-Peak Elements" menu items are unchecked.
This is the default mode of Probe for EPMA and means simply that one can acquire off-peak acquisitions on both one's standard samples and also one's unknown samples. During the actual analysis (as opposed to the acquisition!), the off-peak background correction method will be applied to both standards and unknown samples in the Analyze! window. This is essentially the traditional background acquisition method for EPMA instruments. But remember, one must have no spectral interferences on any of the off-peak background positions!
Alternatively this is also the situation when using MAN background acquisitions for both one's standard samples and one's unknown sample acquisitions. The easiest way to utilize this option is to click the Change All to MAN button in the Acquisition Options dialog from the Acquire! window, so all elements for both standards and unknowns will be changed from the default off-peak acquisitions to MAN background acquisitions.
2. The "Use Off-Peak Elements for MAN Fits" menu is checked, but the "Use MAN Correction for Off-Peak Elements" menu item is unchecked.
In this case, the user plans to acquire off-peak backgrounds on their standards, but will acquire MAN backgrounds on their unknowns (to reduce acquisition time and/or improve sensitivity). By checking the "Use Off-Peak Elements for MAN Fits" menu, the user can acquire off-peak backgrounds on their standards, but still utilize the on-peak standard intensities for the MAN curve fit to correct the background on their unknown samples. This option also allows one to utilize the "Use Interpolated Off-Peaks for MAN Fit" option in the Analysis Options dialog.
3. The "Use Off-Peak Elements for MAN Fits" menu is unchecked, but the "Use MAN Correction for Off-Peak Elements" menu item is checked.
This option is utilized when the standards are acquired using MAN backgrounds, but the unknown samples are acquired using off-peak backgrounds. Then the user decides that they want to perform the MAN background correction method for the unknown samples. This is a good option when you want to compare the results from both off-peak and MAN background methods, on the same data points.
4. Both "Use Off-Peak Elements for MAN Fits" and "Use MAN Correction for Off-Peak Elements" menu items are checked.
Here the user acquired both their standards *and* unknown samples using off-peak background acquisitions, but wants to analyze their standards and their unknown samples using the MAN background correction method. This is the situation that Jon Fellowes utilized in his above examples.
So after all that explanation(!), here is an analysis of Jon's Fe metal standard which should not contain any Co (run as an unknown), after both these Analytical menus are checked, and the interference for Fe Kb on Co was specified in the Standard Assignments dialog from the Analyze! window:
Un 2 std6012_Fe_metal_test_01
TakeOff = 40.0 KiloVolt = 15.0 Beam Current = 10.0 Beam Size = 0
(Magnification (analytical) = 40000), Beam Mode = Analog Spot
(Magnification (default) = 100000, Magnification (imaging) = 100000)
Image Shift (X,Y): .00, .00
Number of Data Lines: 12 Number of 'Good' Data Lines: 12
First/Last Date-Time: 05/06/2021 07:27:49 PM to 05/06/2021 07:49:53 PM
WARNING- Using MAN Background Instead of Off-Peak Background Correction
Average Total Oxygen: .000 Average Total Weight%: 99.708
Average Calculated Oxygen: .000 Average Atomic Number: 26.000
Average Excess Oxygen: .000 Average Atomic Weight: 55.847
Average ZAF Iteration: 1.08 Average Quant Iterate: 4.00
Un 2 std6012_Fe_metal_test_01, Results in Elemental Weight Percents
ELEM: Fe Co
BGDS: MAN MAN
TIME: 20.00 20.00
BEAM: 10.00 10.00
ELEM: Fe Co SUM
51 99.701 -.006 99.696
52 99.700 -.004 99.696
53 99.529 -.020 99.509
54 100.130 -.007 100.122
55 99.288 -.001 99.288
56 99.899 -.008 99.892
57 99.655 -.017 99.639
58 99.726 .001 99.727
59 99.359 .001 99.360
60 99.492 -.005 99.488
61 100.149 .011 100.160
62 99.938 -.022 99.916
AVER: 99.714 -.006 99.708
SDEV: .277 .009 .277
SERR: .080 .003
%RSD: .28 -148.00
STDS: 6012 6013
STKF: 1.0000 1.0000
STCT: 818.74 879.45
UNKF: .9971 -.0001
UNCT: 816.43 -.05
UNBG: 1.83 2.43
ZCOR: 1.0000 1.0209
KRAW: .9972 -.0001
PKBG: 446.59 .98
INT%: ---- -107.45
Now we can see that the Co concentration is within 60 PPM of zero with a variance of 90 PPM. Not bad considering!
For those interested in the code, basically we utilize the UseOffPeakElementsForMANFlag in the UpdateGetMANStandards procedure as seen here (this has not changed):
' Correct the data for dead time, beam drift (and off-peak background if NOT using off-peak sample elements for MAN fits)
If UseOffPeakElementsForMANFlag = False Then
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(0), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(2), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(2), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
' Correct the data for dead time, beam drift (and MAN background if using off-peak sample elements for MAN fits)
Else
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(1), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(3), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(3), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
End If
And in the UpdateGetStandards code we now utilize the UseMANForOffPeakElementsFlag for performing an MAN correction when loading the standard (and interference standard) net intensities as seen here:
' Correct the data for dead time, beam drift (and off-peak background if NOT using MAN for off-peak elements)
If UseMANForOffPeakElementsFlag = False Then
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(0), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(2), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(2), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
' Correct the data for dead time, beam drift (and MAN background if using MAN background corrections for off-peak elements)
Else
If Not UseAggregateIntensitiesFlag Then
Call DataCorrectData(Int(1), UpdateTmpSample())
If ierror Then Exit Sub
Else
Call DataCorrectData(Int(3), UpdateTmpSample()) ' skip aggregate intensity load
If ierror Then Exit Sub
Call DataCorrectDataAggregate(Int(3), sample(), UpdateTmpSample()) ' perform aggregate intensity based on unknown
If ierror Then Exit Sub
End If
End If
Note the first line of each code section which corresponds to the Analytical menus!
-
Thanks again for your work sorting out this problem. I've updated our processing computer's version of PfE and reprocessed the user's data and the element ratios are falling exactly where they should.
I often run the first standardisations in a session as off peak measurements and run a wide range of standards that can be later converted to a MAN curve, as more often than not, the user will at some point want to collect quantitative maps. The extra time required to do this isn't usually a problem, as it'll be running automated overnight. This has also saved my bacon on this occasion!
One other flag for the creation of a MAN curve - that I initially mis-read as a global "use MAN curve" checkbox - is the "Use Interpolated Off Peaks for MAN Fit" checkbox in the Analysis Options menu:
(https://probesoftware.com/smf/gallery/796_11_05_21_12_36_55.jpeg)
I think I'm right in saying that this option switches between using the uncorrected (background, interference, matrix...) on peak measured values for elements not present in a standard to populate the MAN curve plot versus using the interpolated off peak positions? The latter has the benefit of removing any error from minor/traces of an element in impure standards, but at the expense of increased error (assuming the backgrounds were measured at 1/2 the time as the on peak measurement), issues with the interpolation (e.g. shape, as above!) and spectral interferences on both the high and low side backgrounds.
I was wondering whether this checkbox would be better suited to the Assign MAN Fits screen, though?
If the rationale for switching to the interpolated off peak measurements for standards in the MAN curve is because of trace contamination in standards, is it worth being able to select on a per element or per standard basis whether we want to use the on peaks or interpolated off peaks? I appreciate that would be a headache-inducing amount of work though, and this is more of an open question as to whether this would be useful for anyone else. Admittedly, I can't think of many times that I would've needed it!
-
One other flag for the creation of a MAN curve - that I initially mis-read as a global "use MAN curve" checkbox - is the "Use Interpolated Off Peaks for MAN Fit" checkbox in the Analysis Options menu:
(https://probesoftware.com/smf/gallery/796_11_05_21_12_36_55.jpeg)
I think I'm right in saying that this option switches between using the uncorrected (background, interference, matrix...) on peak measured values for elements not present in a standard to populate the MAN curve plot versus using the interpolated off peak positions? The latter has the benefit of removing any error from minor/traces of an element in impure standards, but at the expense of increased error (assuming the backgrounds were measured at 1/2 the time as the on peak measurement), issues with the interpolation (e.g. shape, as above!) and spectral interferences on both the high and low side backgrounds.
I was wondering whether this checkbox would be better suited to the Assign MAN Fits screen, though?
If the rationale for switching to the interpolated off peak measurements for standards in the MAN curve is because of trace contamination in standards, is it worth being able to select on a per element or per standard basis whether we want to use the on peaks or interpolated off peaks? I appreciate that would be a headache-inducing amount of work though, and this is more of an open question as to whether this would be useful for anyone else. Admittedly, I can't think of many times that I would've needed it!
Hi Jon,
Very reasonable questions (and thank-you again for noticing this subtle issue).
So yes, the purpose of the "interpolated MAN" option is for exactly what you describe: it is possible that many of one's standards are more contaminated with trace elements that one would have expected. For me that says "get better standards"! But, this option can also be useful in constructing an MAN curve for say oxygen (dues to surface oxidation), or carbon (due to carbon contamination) on one's standards, etc. We all could take advantage of this!
This idea for this originally came from Ben Hanson at Corning Glass. The option to utilize off-peak interpolated standards for the MAN curve even though the element in question is declared in the standard database comes from Phil Orlandini at UT Austin. This is described in more detail in this topic here:
https://probesoftware.com/smf/index.php?topic=987.0
The three of us also wrote this up in an M&M abstract here:
https://www.cambridge.org/core/journals/microscopy-and-microanalysis/article/modified-mean-atomic-number-background-correction-using-offpeak-interpolated-measurements/6EBB9664B8EDFA5703C05F85CD706CAE
Though I do think this is an interesting option, I do have minor concerns when the element in question is present in large concentrations due to absorption edge effects on the continuum, and also off-peak interferences as you mentioned. But it usually works well for major and minor element background corrections, if (sadly) one has crappy (sorry, impure) standards!.
As to why this option is hidden away in the Analysis Options dialog, well maybe that is because I would consider this an advanced option that should only be utilized by "experts". Also, this flag requires reloading of all MAN intensities, so honestly it really must be set before attempting to review the fits in the MAN Assignments dialog (I could probably come up with a few other excuses, but let's leave it there for now). :)
As for assigning this sort of thing on an element by element basis, well, I'll just say that I'll leave it to you and others to present a case for why that would be worth implementing! ;D
-
As some of you may know, the Z (or electron) fraction averaging method for calculation of average atomic number (Z-bar) in Probe for EPMA, used in the MAN background fit method, allows one to specify an exponent for the best fit.
Recent work by Moy et al. confirms that there is a small dependency of the continuum intensity on the emission line energy (as a function of the best fit Z fraction Z-bar).
To facilitate testing of this expression and its effect on the MAN curve fit, we have implemented the equation from Moy et al., into the Probe for EPMA Assign MAN Fits menu dialog as seen here:
(https://probesoftware.com/smf/gallery/1_30_06_21_9_09_49.png)
For most purposes the default exponent of 0.7 seems to work well, but this new button should allow further testing. It will be interesting to try and understand the physics behind all this!
-
Hi John,
Did you find some solutions for possibility of choosing and changing MAN/off peak correction method for different elements after acquisition was done?
Next question is: how I can upload data for MAN curves plotting from earlier (old) sessions which were in the same conditions of course?
Thank you a lot.
-
Did you find some solutions for possibility of choosing and changing MAN/off peak correction method for different elements after acquisition was done?
The best explanation is in this post above:
https://probesoftware.com/smf/index.php?topic=4.msg9953#msg9953
In addition to the options described above, using the standard/unknown background options in the Acquire! | Acquisition Options dialog, one can also on a per element basis:
1. Acquire off peaks on standards and unknowns
2. Acquire off peaks on standards and MAN of unknowns
3. Acquire MAN on standards and off-peaks on unknowns
4. Acquire MAN on standards and unknowns.
And then also add to those options, multipoint background (MPB) acquisitions... Again, on an element by element basis.
As for changing the background types after one has already acquired data, there are several options, assuming that one acquired the standards and/or unknowns using off-peak or MPB originally.
1. Using the MAN options described in the link above one can analyze off-peak data as though it were acquired using MAN methods.
For instance, one acquires off-peaks on their standards and unknowns, then one can utilize the on-peak measurements from the standards to create MAN cruves, which can then be applied to the off-peak unknowns for MAN analysis of off-peak data.
Or one can acquire MAN data on standards, and off-peak data on unknowns, and again, utilize the MAN method for the unknowns.
Again, see the post linked above for details...
One can also acquire MPB data on standards and/or unknowns and then from the Elements/Cations dialog, switch the MPB options to using one of the off-peak methods. In this case the multiple off-peaks are combined and then utilize one of the off-peak methods, e.g., linear, hi-only, slope-lo, average, etc.
And then, of course, there is the "shared" multi-point background method described here:
https://probesoftware.com/smf/index.php?topic=9.msg1579#msg1579
where one acquires normal off-peak backgrounds on multiple (at least two) elements per spectrometer (at least one), and it creates a pseudo- MPB curves for extrapolating backgrounds using off-peak backgrounds from several elements on that spectrometer/crystal pair...
Whenever I am trying out different background methods on various elements, I generally make a number of sub folders (MAN, Off-peak, etc), and then copy the original MDB file into those sub folders and then set the flags differently for each MDB depending on what I am trying to accomplish, e.g., comparing off-peak to MAN processing for trace elements as described in our 2016 paper here:
https://epmalab.uoregon.edu/publ/A%20new%20EPMA%20method%20for%20fast%20trace%20element%20analysis%20in%20simple%20matrices.pdf
Next question is: how I can upload data for MAN curves plotting from earlier (old) sessions which were in the same conditions of course?
The best method is to utilize the Load File Setup button from the Acquire! New Sample/Setup button and browse to the old run, and select the sample setup you want to load. Then it will ask if you want to load the standard intensities, and you click Yes.
It will then load all your primary, interference and MAN standards along with the MAN assignments from the old run, into the new run.
Yes. It's like magic! 8)
-
Thank you John, let me clarify a bit.
1. Before I am not sure in my "MAN" skills, I want to measure my samples by classic way with off peak background correction. Yes, we can separately calculated the same measuring results with "MAN" and "off peak" methods. Then we can compare calculation results.
But on this way a little change of calculation concentration of element "A" with changing a calculation method from "off peak" to "MAN" will not affect to concentration another elements which concentration calculated only by "off peak".
This is why I asked you about this possibility.
2. Yes, this way we can upload any standards from "old" acquisitions. But I haven't find the way I can delete some of "old" standards which I uploaded to new (current) MDB. Sometimes the number of standards I want ignore is too much and it will be suitable to delete some of them. Is it possible?
3. Small addition question.
When we plot a MAN curve with standards which include searching element "A", the system uses "off peak" background data. What equation the system apply to calculate "on peak" background for MAN curve - the equation we have chosen for "off peak" background correction or something else?
-
1. Before I am not sure in my "MAN" skills, I want to measure my samples by classic way with off peak background correction. Yes, we can separately calculated the same measuring results with "MAN" and "off peak" methods. Then we can compare calculation results.
But on this way a little change of calculation concentration of element "A" with changing a calculation method from "off peak" to "MAN" will not affect to concentration another elements which concentration calculated only by "off peak".
This is why I asked you about this possibility.
If the background method changes, the change in composition will usually be very small. But that small change in composition will still affect the matrix correction of the other elements. However the change from that will also be an even smaller change.
2. Yes, this way we can upload any standards from "old" acquisitions. But I haven't find the way I can delete some of "old" standards which I uploaded to new (current) MDB. Sometimes the number of standards I want ignore is too much and it will be suitable to delete some of them. Is it possible?
You cannot delete samples in Probe for EPMA. One should never delete scientific data. However, one can disable samples if they are not useful from the Analyze! window. And/or enable them.
3. Small addition question.
When we plot a MAN curve with standards which include searching element "A", the system uses "off peak" background data. What equation the system apply to calculate "on peak" background for MAN curve - the equation we have chosen for "off peak" background correction or something else?
If you select the option for using off-peak standards for generating the MAN curves, the program will only utilize the on-peak measurements and simply ignore the off-peak data.
Unless one utilizes the so called "interpolated off-peak MAN" method:
https://probesoftware.com/smf/index.php?topic=987.0
Then the off-peak (or MPB) fit method is utilized for the on-peak background.
-
1,2 - thank you.
3 - Yes, I asked you about "interpolated off-peak MAN" method. What equation the system apply to calculate "on peak" background for MAN curve - the equation we have chosen for "off peak" background correction?
-
3 - Yes, I asked you about "interpolated off-peak MAN" method. What equation the system apply to calculate "on peak" background for MAN curve - the equation we have chosen for "off peak" background correction?
The interpolated off-peak MAN method utilizes whatever off-peak (or MPB) fit method is assigned to that standard sample (from the Elements/Cations dialog).
-
Great!
John, thank you very much.
-
Greetings John and everyone!
Could you help to find a solution to compare background levels (for MAN curve) of the same standard/element which was measured at the same conditions but in different sessions (different MDB).
I tried to upload the standard measuring from different sessions with standard way (acquisition/new sample/... way) but it is impossible? because the system will be confused?
I wonder to use a set of identical data to obtain an average for routine measurements of main elements.
Also we can quick (just quality) compare a contamination of minor elements in the close samples which were measured in different sessions etc.
Thank you a lot.
-
Could you help to find a solution to compare background levels (for MAN curve) of the same standard/element which was measured at the same conditions but in different sessions (different MDB).
If you want to compare MAN background values from multiple runs you should be able to export those using the Output | Save User Specified Format Output menu from each run and compare them in an external program. Note also from the Output | Save Custom Analysis Output menu there is a Format #8 (MAN) output menu that might also be useful.
I tried to upload the standard measuring from different sessions with standard way (acquisition/new sample/... way) but it is impossible? because the system will be confused?
You can also import standards into Probe for EPMA from multiple MDB files using the Acquire! window, New Sample/Setup button dialog, Load File Setup button, but you'll need to (completely) close Probe for EPMA, before loading the same standards from other probe runs. Hey, it's a feature, not a bug! :)
Then once you have all the multiple (MAN) standards loaded from all your runs, you can open the Analytical Assign MAN Fits menu and then *uncheck* the Plot Last MAN Set Only checkbox as seen here:
(https://probesoftware.com/smf/gallery/395_23_08_22_9_47_26.png)
-
Or you could use "Drift.exe" and just search for a specific element or element/standard combination. It should tabulate all the intensities over all the mdbs with time stamps into a single xls spreadsheet. However, these intensities are not corrected for continuum absorption. Nevertheless, drift makes it easy to compare how the intensities vary from run to run over time.
-
S - Bi interference?!
I recently found a subtle(!) error in my methodology using MAN backgrounds:
the apparent interference of the S absorption edge on Bi M-alpha measured on PET.
I thought that this might be of interest to the community, and so am describing it here.
Using Bi2Te3 as the Bi-standard, and having established a precise MAN curve for Bi M-alpha, I thought all would be well.
However, the subsequent Bi contents of the various sulfide minerals were substantially above detection limits (unexpected, and almost certainly incorrect).
A clue: all the standard materials that contain sulfur fall well off the established MAN curve for Bi M-alpha.
However, no spectral interferences are listed by Probe-for-EPMA software for Bi M-alpha by sulfur.
What is going on?
I finally looked at the WDS scans of bismuth selenide, ZnS, and FeS2 tabulated by Sandrin Feig at Central Science Laboratory: EPMA - Method Development Tool.
The absorption edge for S is close to S K-beta, and this edge leads to a slightly elevated background under the Bi M-alpha position.
I think that, when large amounts of S are present, this elevation of the background leads to a misleading small value for analyzed Bi.
I suppose that I could move to the Bi M-beta line to avoid the problem.
But for the present, I simply put in an overlap correction, using synthetic ZnS (99.995% purity).
I would be interested in the experience and opinions of others on this apparent problem.
Thanks, Andrew
-
Interesting. My general "rule of thumb" for background (off-peak or MAN) is that it's generally the lowest intensity one can measure. Again, just a "rule of thumb", there are exceptions (holes in the continuum, e.g., Au La in Fe bearing minerals).
Can you share a screenshot of the plot for the Bi La MAN calibration curve with us?
-
Interesting. My general "rule of thumb" for background (off-peak or MAN) is that it's generally the lowest intensity one can measure. Again, just a "rule of thumb", there are exceptions (holes in the continuum, e.g., Au La in Fe bearing minerals).
Can you share a screenshot of the plot for the Bi La MAN calibration curve with us?
Attached is a PDF of the Bi M-alpha MAN assignment and fit.
The 8 reference materials used in this fit are: Cr2O3 (#2), TiO2 (#5), MgAl2O4 (#13), Cr (#27), Ag (#34), Ti (#45), Sb (#55), and SiC (#133).
The relative-percent-deviation is 1.79, the intercept is -0.0135, and the Z-fraction is the default 0.7.
Also attached is a PDF of wavelength scans (JEOL units, PETJ) from Sandrin Feig's excellent EPMA-Method Development Tool.
The S K-edge should be circa 160.7 mm, and the scans shown include: Bi2Se3, ZnS, and FeS2.
There is a tail from S K-beta1 on its high side, which extends under the Bi M-alpha peak (at about 164.0 mm).
I attribute this tail to the effect of the S K absorption edge.
This S-related tail presumably gives rise to fictive Bi M-alpha intensity, when S is abundant.
-
I attribute this tail to the effect of the S K absorption edge.
Upon closer inspection, the tail on the high side of S K-beta cannot be due to the S absorption edge.
If anything, the effect of the S absorption edge would be to lower(!) the background in this region.
Attached is a plot of the S K-edge in JEOL units for a PET crystal.
And a plot of the Ar and S K-edges in a scan of marcasite from Sandrin Feig's EPMA-Method Development Tool.
So, what gives rise to the elevated background between the S K-beta and S K-alpha peaks?
Particularly, close to the S K-beta peak (on the high JEOL mm side, which is the low energy side)?
I will have to check where are the radiant-Auger emission lines for S.
-
I will have to check where are the radiant-Auger emission lines for S.
Bingo! (possibly).
From the Auger emission catalog of Coghlan & Clausing (Atomic Data 5, 317-469, 1973):
For sulfur, the KM1M1 Auger emission line is at 2.438 keV, and the KM1M23 Auger emission line is at 2.4475 keV.
It is possible for Auger transitions to be present in X-ray spectra as radiant-Auger-emission lines.
In the attached scans of Bi2Se3 and FeS2 from Sandrin Feig's EPMA Method-Development Tool, there are small peaks at 2.437 and 2.445 keV.
Bi M-alpha is at 2.422 keV in this image.
So, now my interpretation is that tails from S Auger-emission lines near S K-beta are interfering on the Bi M-alpha line.
If this is correct, it might be useful to have the various Auger-emission lines added to programs such as Probe-for-EPMA,
and the EPMA Method-Development Tool.
-
Interesting. My general "rule of thumb" for background (off-peak or MAN) is that it's generally the lowest intensity one can measure. Again, just a "rule of thumb", there are exceptions (holes in the continuum, e.g., Au La in Fe bearing minerals).
Can you share a screenshot of the plot for the Bi La MAN calibration curve with us?
Attached is a PDF of the Bi M-alpha MAN assignment and fit.
The 8 reference materials used in this fit are: Cr2O3 (#2), TiO2 (#5), MgAl2O4 (#13), Cr (#27), Ag (#34), Ti (#45), Sb (#55), and SiC (#133).
The relative-percent-deviation is 1.79, the intercept is -0.0135, and the Z-fraction is the default 0.7.
Also attached is a PDF of wavelength scans (JEOL units, PETJ) from Sandrin Feig's excellent EPMA-Method Development Tool.
The S K-edge should be circa 160.7 mm, and the scans shown include: Bi2Se3, ZnS, and FeS2.
There is a tail from S K-beta1 on its high side, which extends under the Bi M-alpha peak (at about 164.0 mm).
I attribute this tail to the effect of the S K absorption edge.
This S-related tail presumably gives rise to fictive Bi M-alpha intensity, when S is abundant.
When you first described this issue I thought it was because you might have been using some sulfur containing materials for your MAN curve for Bi Ma, but now I see that you did not. So this is actually just a normal spectral interference issue and is properly corrected just as you did by utilizing a standard containing sulfur, but no Bi.
Or am I missing something?
-
So this is actually just a normal spectral interference issue and is properly corrected just as you did by utilizing a standard containing sulfur, but no Bi.
No, unfortunately, this is not just a normal spectral interference issue.
It is interference by Radiative-Auger-Emission lines (RAE lines) on the Bi M-alpha X-ray emission line.
The closest interferences given by the Probe-for-EPMA software in the vicinity of the Bi M-alpha line are shown in the attached JPG file.
The only interference listed as coming from sulfur is the S SKBN line, about -1.645 mm away from the Bi M-alpha line.
Whereas there are 2 RAE lines of sulfur, S KM1M23 and S KM1M1 that are closer to the Bi M-alpha position.
Please see the attached PDF, which shows the S RAE lines with respect to the main emission lines and satellite lines of sulfur.
As the RAE lines are not always insignificant, my suggestion is that they be included into the Probe-for-EPMA software in some fashion.
The Coghlan & Clausing 1973 reference is a good start for lines between 10 eV and 3000 eV.
Some further examples of RAE lines in X-ray emission spectra are given by Takahashi et al. (2001) Extended x-ray emission fine structure and high-energy
satellite lines state measured by electron probe microanalysis. Surface and Interface Analysis 31, 118-125.
-
So this is actually just a normal spectral interference issue and is properly corrected just as you did by utilizing a standard containing sulfur, but no Bi.
No, unfortunately, this is not just a normal spectral interference issue.
It is interference by Radiative-Auger-Emission lines (RAE lines) on the Bi M-alpha X-ray emission line.
OK, that is cool. But the Calculate WDS Interferences button in PFE is just a nominal calculation for trying to figure out these interferences in a general sense. This nominal calculation has nothing to do with the actual quantitative spectral interference correction in PFE. So if the nominal calculation doesn't list a specific interference, that doesn't mean it can't be corrected for.
In other words if a material with S but no Bi shows a positive net intensity it could be a contamination, or an interference from another element, or an interference from the element in question.
If these Auger lines are actually detectable, it would be neat to include them as a calculation/display option. But the quantitative interference correction itself should take of care of it either way is my assumption.
Did you try the quantitative interference correction using a standard with S but no Bi? How does it do?
-
So this is actually just a normal spectral interference issue and is properly corrected just as you did by utilizing a standard containing sulfur, but no Bi.
No, unfortunately, this is not just a normal spectral interference issue.
It is interference by Radiative-Auger-Emission lines (RAE lines) on the Bi M-alpha X-ray emission line.
OK, that is cool. But the Calculate WDS Interferences button in PFE is just a nominal calculation for trying to figure out these interferences in a general sense. This nominal calculation has nothing to do with the actual quantitative spectral interference correction in PFE. So if the nominal calculation doesn't list a specific interference, that doesn't mean it can't be corrected for.
Did you try the quantitative interference correction using a standard with S but no Bi? How does it do?
The Probe-for-EPMA software is so useful and reliable, that I am guilty of relying on it to figure out my interferences for me!
That is, I am (or have been) highly reliant on PFE's "nominal calculation" to indicate all probable interferences.
This is why I suggest adding in the RAE lines (perhaps just those from 0.6 to 3.0 keV, observable on TAP and PET).
And yes, as I mentioned in my first post on this topic - I simply put in an overlap correction, using synthetic ZnS (99.995% purity).
It works sufficiently well, thank you!
-
Did you try the quantitative interference correction using a standard with S but no Bi? How does it do?
The Probe-for-EPMA software is so useful and reliable, that I am guilty of relying on it to figure out my interferences for me!
That is, I am (or have been) highly reliant on PFE's "nominal calculation" to indicate all probable interferences.
This is why I suggest adding in the RAE lines (perhaps just those from 0.6 to 3.0 keV, observable on TAP and PET).
And yes, as I mentioned in my first post on this topic - I simply put in an overlap correction, using synthetic ZnS (99.995% purity).
It works sufficiently well, thank you!
Ok, great, just double checking.
If someone has a ASCII table with these Auger lines feel free to post it. We would like to have at least the text labels, energies and nominal intensities for each emission line.
-
[If someone has a ASCII table with these Auger lines feel free to post it. We would like to have at least the text labels, energies and nominal intensities for each emission line.
The Coghlan and Clausing 1973 reference is: Atomic Data and Nuclear Data Tables, volume 5, issue 4, pp. 317-469, and may be found at
https://doi.org/10.1016/S0092-640X(73)80005-1
More recent progress in simulation and calculation of electron spectra and Auger kinetic energies are given by:
Smekal, W., Werner, W.S. and Powell, C.J., 2005. Simulation of electron spectra for surface analysis (SESSA): a novel software tool for quantitative Auger‐electron spectroscopy and X‐ray photoelectron spectroscopy. Surface and Interface Analysis: An International Journal devoted to the development and application of techniques for the analysis of surfaces, interfaces and thin films, 37(11), pp.1059-1067.
Werner, W.S., Smekal, W. and Powell, C.J., 2011. NIST Database for the Simulation of Electron Spectra for Surface Analysis (SESSA). In Version 1.3, Standard Reference Program Database 100, US Department of Commerce. National Institute of Standards and Technology Gaithersburg, MD.
Dean, J.W., Chantler, C.T. and Ganly, B., 2022. Ab initio calculations of Auger electron kinetic energies: Breadth and depth. Radiation Physics and Chemistry, 200, p.110472.
Unfortunately, the online NIST X-ray Photoelectron Spectroscopy Database is not comprehensive:
https://srdata.nist.gov/xps/Default.aspx
-
I also calculated data for MAN BG curve. It is a bit different topic but could you explain how I can download the calculated MAN BG to PFE.
The mean atomic number (MAN) background correction curve is not a static fit. It is a dynamic fit based on the current selection of standards available and/or selected by the user and also corrected for standard intensity drift in real time and of course the current standard compositions. Also to a very small degree, the current dead time constants and equations.
In addition it is affected by the choice of the model for the calculation of average Z. Traditionally we have utilized a mass fraction model, but we now know (publication coming soon), that mass has nothing to do with continuum production and so one should generally select the Z Fraction model which is based on the number of atoms and their atomic numbers which scales roughly as Z^2/3. There is a topic here that explains this:
https://probesoftware.com/smf/index.php?topic=1221.msg10187#msg10187
To answer your question though, one can only import the standard intensities utilized in the MAN fit from another Probe for EPMA probe run. This can be done at any time whenever the Load File Setup button is selected from the Acquire! window New Sample/Setup button and the user clicks Yes to the question: Do you want to load the standard intensities? after the sample setup has been loaded.
-
Here is the new Z fraction average Z calculation using a fixed exponent as described in our recent paper this summer:
https://academic.oup.com/mam/article-abstract/29/4/1436/7224307
(https://probesoftware.com/smf/gallery/1_23_09_23_8_11_11.png)
Here the Z fraction exponent is fixed at 0.7 for all MAN elements. However, starting with v. 13.4.4 of Probe for EPMA you can now change this exponent to a zero and the program will automatically calculate the optimized exponent based on the emission line energy (as described in the paper linked above).
Here we've entered a zero for the Z fraction exponent and clicked the Update Fits button. We can can now see the calculated "optimized" Z bar exponent is 0.67 which produces a better Rel % Deviation fit to the MAN standards:
(https://probesoftware.com/smf/gallery/1_23_09_23_8_11_27.png)
-
Note that if you have these two options checked in the Analytical | Analysis Options menu dialog:
(https://probesoftware.com/smf/gallery/1_23_09_23_8_38_53.png)
you will see the MAN fit details for the optimized Z fraction exponent, "ZEXP:" parameter:
St 310 Set 2 SM Fayalite, Results in Elemental Weight Percents
ELEM: Si Al Ca K Ti Cr Fe Mn
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL
BGDS: MAN MAN MAN MAN MAN MAN MAN MAN
MAN1: .165963 2.59049 -.08976 -.10492 -.36600 -1.2218 -.18728 .109394
MAN2: .196322 -.22066 .068148 .119022 .255554 .312196 .102628 .039946
MAN3: .000000 .012323 .000000 -.00014 .000000 -.00291 .000000 .001477
ABS%: -27.70 -39.15 -3.56 -5.55 -1.30 -.28 .18 .00
ZEXP: .64 .63 .69 .68 .70 .72 .73 .73
TIME: 40.00 40.00 40.00 20.00 20.00 40.00 25.00 25.00
BEAM: 19.97 19.97 19.97 19.97 19.97 19.97 19.97 19.97
Assuming the MAN Z Fraction exponent is set to zero...
-
Here is the new Z fraction average Z calculation using a fixed exponent as described in our recent paper this summer:
https://academic.oup.com/mam/article-abstract/29/4/1436/7224307
(https://probesoftware.com/smf/gallery/1_23_09_23_8_11_11.png)
Here the Z fraction exponent is fixed at 0.7 for all MAN elements. However, starting with v. 13.4.4 of Probe for EPMA you can now change this exponent to a zero and the program will automatically calculate the optimized exponent based on the emission line energy (as described in the paper linked above).
Here we've entered a zero for the Z fraction exponent and clicked the Update Fits button. We can can now see the calculated "optimized" Z bar exponent is 0.67 which produces a better Rel % Deviation fit to the MAN standards:
(https://probesoftware.com/smf/gallery/1_23_09_23_8_11_27.png)
A related request - can you arrange for Probe-for-EPMA to export the graphic of the MAN assignment and fit for each X-ray line used?
At present, I use the Windows Snip tool and dump the graphic into Word (see attached).
In this way, I can the reference materials used (or most of them), the Rel% Deviation, the Curvature, the Slope and Intercept, as well as the Z-bar Exponent.
Further question - what are acceptable values or ranges in the MAN fit for: the Rel% Deviation, the Curvature, the Slope and Intercept?
At present, I like to have Rel% Deviation certainly less than 3, preferably less than 2.
I like to have the Intercept close to zero (not always possible, unclear as to why).
What do others find acceptable?
Thanks,
Andrew
-
A related request - can you arrange for Probe-for-EPMA to export the graphic of the MAN assignment and fit for each X-ray line used?
At present, I use the Windows Snip tool and dump the graphic into Word (see attached).
In this way, I can the reference materials used (or most of them), the Rel% Deviation, the Curvature, the Slope and Intercept, as well as the Z-bar Exponent.
If you click the Print Additional MAN Fit and Correction Parameters checkbox as described here:
https://probesoftware.com/smf/index.php?topic=4.msg12053#msg12053
much if not all of this info is output to the log window.
Further question - what are acceptable values or ranges in the MAN fit for: the Rel% Deviation, the Curvature, the Slope and Intercept?
At present, I like to have Rel% Deviation certainly less than 3, preferably less than 2.
I like to have the Intercept close to zero (not always possible, unclear as to why).
What do others find acceptable?
The MAN fits in your Word document in the post above look really good to me. I guess it just depends on the sensitivity/accuracy you require. The 2016 MAN paper goes into this in some detail:
https://epmalab.uoregon.edu/publ/A%20new%20EPMA%20method%20for%20fast%20trace%20element%20analysis%20in%20simple%20matrices.pdf
-
A related request - can you arrange for Probe-for-EPMA to export the graphic of the MAN assignment and fit for each X-ray line used?
At present, I use the Windows Snip tool and dump the graphic into Word (see attached).
Note that you can also use the <alt> <Print Screen> key combo to capture just the current window to the system clipboard.
We could also add a "Clipboard" button to the MAN dialog.
-
The latest version of Probe for EPMA v. 13.5.9 now includes a Clipboard button in the MAN window and also outputs the MAN coefficients and relative % deviations (and exponents if utilizing the Z Fraction MAN fit option) if using the Print MAN Fit and Correction Parameters to Log Window (and Use Detailed Output) option from the Analytical | Analysis Options dialog which can be see here for mass fraction MAN Z bar calculation:
St 358 Set 2 Diopside (Chesterman), Results in Elemental Weight Percents
ELEM: Na Si K Al Mg Fe Ca Mn O H
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC SPEC
BGDS: MAN MAN LIN MAN MAN MAN MAN LIN
MAN1: 8.21064 17.6972 15.1856 -23.560 3.60143 4.96464
MAN2: .298724 .235361 1.31971 5.10899 .135131 2.51710
MAN3: .039522 .000000 .016978 -.09491 .014340 .018263
REDV: 5.88 11.90 .77 3.37 .93 1.89
ABS%: -45.67 -17.88 -27.57 -30.56 -.79 -2.99
And here for Z fraction MAN Z bar calculation (constant exponent) with the constant exponent:
St 358 Set 2 Diopside (Chesterman), Results in Elemental Weight Percents
ELEM: Na Si K Al Mg Fe Ca Mn O H
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC SPEC
BGDS: MAN MAN LIN MAN MAN MAN MAN LIN
MAN1: 4.05116 17.6299 15.3465 -30.781 1.65561 -2.6997
MAN2: .850560 .251887 1.24988 6.25317 .366658 3.66492
MAN3: .028771 .000000 .025809 -.12994 .010452 -.00973
REDV: 7.10 11.91 1.88 1.81 .22 4.33
ABS%: -45.67 -17.88 -27.57 -30.56 -.79 -2.99
ZEXP: .70 .70 .70 .70 .70 .70
And here for Z fraction MAN Z bar calculation (variable exponent):
St 358 Set 2 Diopside (Chesterman), Results in Elemental Weight Percents
ELEM: Na Si K Al Mg Fe Ca Mn O H
TYPE: ANAL ANAL ANAL ANAL ANAL ANAL ANAL ANAL SPEC SPEC
BGDS: MAN MAN LIN MAN MAN MAN MAN LIN
MAN1: 2.00334 17.6443 14.9056 -32.456 1.84603 -3.2513
MAN2: 1.13292 .253212 1.30209 6.52835 .344140 3.74527
MAN3: .021317 .000000 .025652 -.13876 .010783 -.01194
REDV: 7.63 11.92 2.34 1.50 .16 4.45
ABS%: -45.67 -17.88 -27.57 -30.56 -.79 -2.99
ZEXP: .62 .64 .63 .63 .73 .69
-
We also modified the Report output table format to include MAN statistics as requested by Andrew Locock:
(https://probesoftware.com/smf/gallery/1_17_10_23_8_02_55.png)