Performing Integrated WDS and EDS Acquisition in PFE

[John Donovan]

While Ryan McAleer and I were looking into how we might decrease the time it takes to start an EDS acquisition using the JEOL MEC EDS interface, when using the TDI acquisition feature in Probe for EPMA:

https://smf.probesoftware.com/index.php?topic=79.msg13791#msg13791

we found another "interesting" behavior of the JEOL EDS software.

Previously we found that the JEOL EDS software automatically inserts the Faraday cup (PCD) when the EDS acquisition is complete. Of course when one is still acquiring WDS elements, this is not a good thing.  So in addition to item #2 listed here:

Ryan McAleer makes some suggestions for getting the best results from the JEOL EDS interface on iSP100/iHP200F instruments:

1. You should run a single EDS spectrum acquisition in the JEOL software prior to any EDS acquisitions with Probe for EPMA. Doing seems to get the JEOL software interface into a correct state. Just one EDS acquisition for each time starting the JEOL software. 

2. To avoid issues with the Faraday cup inserting before the WDS acquisitions are finished, you must uncheck the "Insert PCD after reservation execution is completed" [PFE will prompt you about this once per run]

3. The user should run a single EDS acquisition after they change the EDS time constant in Probe for EPMA. This is to ensure that these EDS conditions are properly saved for subsequent EDS element analyses. Or one can just set the EDS time constant they want in both JEOL (Analysis-->EDS analysis basic conditions) and Probe for EPMA. Otherwise the documented time constant on that first point will be incorrectly saved [we are looking into why this occurs].

4. If there is any hangup during EDS comm, RESTART the JEOL COMPUTER.  If you just restart the software (either through "ResetEPMA" or through the task manager) the "reservation" of the EDS will not clear and you will run into a variety of problems.

we have to add a new item because the JEOL EDS software also seems to change the beam mode once the EDS acquisition is completed, even if WDS elements are still being acquired!

Here are Ryan's observations:

During EDS analysis: JEOL software indicates scan coils on and spot mode--equivalent of pfe digital spot mode I think.



After EDS ends (but WDS still going), changes to this configuration: pfe equivalent of scan mode. I think when EDS run ends JEOL reverts to a standard state, which is probe scan, note that the beam stays defocused if you have that set, so if the Mag is set to 270K for analysis then maybe there is little practical effect. Becomes an issue if you want to make a valiant attempt to measure something smaller than FOV at 270K (<500nm).

So, just FYI.

[Ben Buse]

Hi a question about pulling intensities...

The PFE software stores the EDS spectra at time of analysis, and then at a later time it asks the JEOL EDS to extract the net intensity from the spectra stored in PFE - is that correct?

Which leads to two further questions

a. Can the PFE ask the JEOL offline software PC-SEM to do this task?

b. If PFE had the spectra for each pixel in a map (if this was extracted from JEOL pts and given to PFE), could it then ask the JEOL EDS to give net intensities for each spectra?

 

[Ben Buse]

With rgds to question a. interestingly testEDS works, on JEOL offline computer PC-SEM

But PFE doesn't seem to

[John Donovan]

With rgds to question a. interestingly testEDS works, on JEOL offline computer PC-SEM,
But PFE doesn't seem to

That's probably because you are trying to run PFE at the same time TestEDS is connected and the JEOL LibEDSIS DLL won't allow that because it isn't multi-threaded.

The PFE software stores the EDS spectra at time of analysis, and then at a later time it asks the JEOL EDS to extract the net intensity from the spectra stored in PFE - is that correct?

Not quite.

PFE acquires the EDS point spectra at time of acquisition and stores it to the run database. At the same time, PFE extracts the net intensities and stores them also, unless during an automated acquisition with EDS the user answered the question "Do you want to store EDS net intensities during the automation?" with a No, as described here:

https://smf.probesoftware.com/index.php?topic=79.msg13471#msg13471

The reason for this question is because when one has many points per sample, every time PFE acquires a new data point, it lists the net intensities for both WDS and EDS elements for that sample.  For WDS elements there is essentially no "overhead" because the WDS data is stored in the local MDB file.  But for EDS elements PFE needs to query the EDS server and obtain the net intensities from the OEM spectrum deconvolution software, which takes a few seconds per point. 

If there are hundreds of points per sample, this starts to add up, so the user is asked whether they want to do this net intensity extraction for every point acquisition, or do it later after the automation has completed, when it will take much less time (because it only needs to obtain the net intensities once per sample/point).

So yes, normally PFE acquires the EDS spectrum, obtains the EDS net intensities for any EDS elements, and then stores those also. If EDS elements are added subsequently to the sample, when the quant is run again, or even just the raw data is displayed, PFE will again obtain the EDS net intensities and store them again.

Which leads to two further questions

a. Can the PFE ask the JEOL offline software PC-SEM to do this task?

Yes, but there's a catch for the JEOL software.

At acquisition time, the EDS spectrum for each point analysis is stored in the local PFE database obtained from the JEOL (and Thermo and Bruker) server software. At that time (unless turned off by the user during automation as described above), PFE also obtains the net intensities for each EDS element and also stores those net intensities locally.

For the Thermo and Bruker EDS software, this means reading the EDS spectrum stored locally in the PFE database, and sending that spectrum, along with the spectrum meta data, EDS elements/x-ray lines, to the OEM EDS server, and then getting the EDS net intensities back, which are then (as previously mentioned) also stored locally.

For the JEOL software however, the JEOL EDS server cannot receive an EDS spectrum and meta data and return the EDS net intensities. Instead, the JEOL EDS server needs the *name* (GUID actually) of the EDS point acquisition which is stored locally on the JEOL server. 

When this GUID is passed to the JEOL EDS server, the JEOL server software reads the EDS spectrum and meta data from its local storage and returns the net intensities for the EDS elements specified by PFE.

Why this matters for obtaining EDS net intensities, is because for the Thermo and Bruker EDS systems, one can simply install the Thermo of Bruker EDS software on any computer and when running Probe for EPMA off-line for quantitative analyses, one can simply send the EDS spectrum (and meta data) to the OEM EDS software and the OEM software will return the EDS net intensities to PFE for quantitative analyses. Even when PFE or the computer is not connected to the instrument!

However, with the JEOL EDS software, even if one installed the JEOL software on an off-line computer, one could still not obtain the EDS net intensities, because the EDS spectrum files are stored with the JEOL EDS software which is on the JEOL computer connected to the instrument!

However, there is a "work around" for this. If one wants to re-process EDS elements off-line on another computer, that is not connected to the OEM software on the OEM computer connected to the instrument, one can simply make sure that all net intensities are extracted and stored in the PFE database locally by just having the PFE software list the raw data for each sample from the Analyze! window. Then go into the Analytical | Analysis Options menu and check this checkbox:



Then PFE will only use locally stored EDS net intensities for quantitative analyses and this file can be moved to any computer for further off-line re-processing.

b. If PFE had the spectra for each pixel in a map (if this was extracted from JEOL pts and given to PFE), could it then ask the JEOL EDS to give net intensities for each spectra?

It's a great question with a very complicated answer.

The EDS API software system on your instrument did have functions for this (LibEDSIS.DLL), e.g., EDSGetNetCountMapData(), but by the time we were implementing this feature, JEOL stopped supporting this 8230/8530 EDS API, and now only support the new MEC EDS API (for the iSP100/iHP200F), which so far as I can tell does not have support for EDS net intensities from spectrum images.  I can ask JEOL Japan again, but it wouldn't help you on the 8230/8530 instrument with the old EDS API.

I can't find the email, but apparently there could also be an issue with getting the JEOL EDS hypermap net intensities when the spectrum image data is acquired from the WDS scan generator vs. the JEOL video scan generator.  I remember they mentioned that getting the EDS pixels hardware dead time corrected in the WDS hypermap was "very difficult". Apparently only the EDS hypermaps acquired using the video scan generator are corrected for dead time using hardware. Which is why we started trying to get Bruker to work with JEOL to do synchronized WDS and EDS mapping.

If you want to contact me directly by email or Zoom we can discuss what we might be able to do for your situation.

[John Donovan]

As we have mentioned previously the JEOL PC-SEM software automatically inserts the PCD (Faraday cup) after the EDS spectrum acquisition is complete (it's a feature not a bug!).   

In order to prevent the Faraday cup from being inserted before the WDS acquisition finishes (if it is using a longer acquisition time than the EDS), one must be sure to *uncheck* the "Insert PCD after the reservation execution is completed" checkbox in the JEOL software:



A slightly more subtle bug (sorry, I mean feature!), is that the JEOL EDS software also changes the beam mode from point to scanning mode once the EDS acquisition is complete.  So if you haven't specified a high magnification for scanning, say 100Kx and your WDS takes longer than the EDS acquisition, you might have an issue when your WDS quantification when the EDS completes.

I asked JEOL Japan if they could add another check box to uncheck so the JEOL EDS software would not change the beam mode when the EDS acquisition is complete, but they said they looked into it and it would be too difficult do. I have to wonder if this same thing happens with the JEOL PC-EPMA software when acquiring WDS and EDS together?  Does anyone know?

Anyway, just a heads up so you know when performing integrated WDS and EDS in Probe for EPMA using the JEOL EDS that you make sure your EDS acquisition times are always longer than your WDS acquisition times.

[JonF]

I asked JEOL Japan if they could add another check box to uncheck so the JEOL EDS software would not change the beam mode when the EDS acquisition is complete, but they said they looked into it and it would be too difficult do. I have to wonder if this same thing happens with the JEOL PC-EPMA software when acquiring WDS and EDS together?  Does anyone know?

I tried combined WDS/EDS on our iHP200F using SEM Center and PC-EPMA, using all three options of "WDS -> EDS", "EDS -> WDS" and "WDS/EDS" in the PC-EPMA software. I set a single WDS pass of elements for 20s on peak plus 5s off peak and 10s EDS real time, and in each case it didn't change the beam conditions until after the acquisition was completed for all spectrometers (EDS and WDS).     
This is assuming the SEM Center GUI is updated whilst the analysis is ongoing, but to try and see if this is the case, I set up test points on a cathodoluminescent sample and it didn't seem to change until the acquisition is finished.

So it would seem that PC-EPMA can do what you suggested?

[John Donovan]

I asked JEOL Japan if they could add another check box to uncheck so the JEOL EDS software would not change the beam mode when the EDS acquisition is complete, but they said they looked into it and it would be too difficult do. I have to wonder if this same thing happens with the JEOL PC-EPMA software when acquiring WDS and EDS together?  Does anyone know?

I tried combined WDS/EDS on our iHP200F using SEM Center and PC-EPMA, using all three options of "WDS -> EDS", "EDS -> WDS" and "WDS/EDS" in the PC-EPMA software. I set a single WDS pass of elements for 20s on peak plus 5s off peak and 10s EDS real time, and in each case it didn't change the beam conditions until after the acquisition was completed for all spectrometers (EDS and WDS).     
This is assuming the SEM Center GUI is updated whilst the analysis is ongoing, but to try and see if this is the case, I set up test points on a cathodoluminescent sample and it didn't seem to change until the acquisition is finished.

So it would seem that PC-EPMA can do what you suggested?

Interesting.  Thanks for testing this.

It seems that JEOL MEC EDS interface API is based on the JEOL PC-SEM application GUI rather than the actual sub routines available to the PC-EPMA softare. Too bad... maybe you should have bought a Bruker!   

[Ben Buse]

Hi

Is it possible to add EDS elements after the analysis has been completed if you've acquired the spectrum?

[John Donovan]

Is it possible to add EDS elements after the analysis has been completed if you've acquired the spectrum?

Of course you can.

You can add an element by EDS before or after the sample has EDS spectrum data. Here is an example of adding an element by EDS for use in the interference correction with another element (WDS or EDS) that is already acquired:

https://smf.probesoftware.com/index.php?topic=482.msg2826#msg2826

This hasn't been tested with WDS and EDS maps, but it should be the same in CalcImage (if we ever get that working):

https://smf.probesoftware.com/index.php?topic=1524.msg11777#msg11777

[Ben Buse]

2. To avoid issues with the Faraday cup inserting before the WDS acquisitions are finished, you must uncheck the "Insert PCD after reservation execution is completed" [PFE will prompt you about this once per run]

Hi, where's the option to do this?

[John Donovan]

2. To avoid issues with the Faraday cup inserting before the WDS acquisitions are finished, you must uncheck the "Insert PCD after reservation execution is completed" [PFE will prompt you about this once per run]

Hi, where's the option to do this?

It is circled in red in this post here:

https://smf.probesoftware.com/index.php?topic=79.msg14055#msg14055

[Ben Buse]

It is circled in red in this post here:

https://smf.probesoftware.com/index.php?topic=79.msg14055#msg14055

Hmm, that's the software for the new machine, I wonder where it is on the previous software

[John Donovan]

It is circled in red in this post here:

https://smf.probesoftware.com/index.php?topic=79.msg14055#msg14055

Hmm, that's the software for the new machine, I wonder where it is on the previous software

I do not know when JEOL added this checkbox to their software. I wish JEOL had given us an EDS API call that just acquired an EDS spectrum and let us handle the Faraday cup and beam modes...      Both the 8230/8530 EDS API and the iSP100/iHP200F EDS API seen to be implemented as an API on top of the JEOL software GUI for some strange reason.

If you do not have this checkbox in your JEOL software you should be sure to always run your EDS acquisition longer than the WDS acquisition, in order to prevent the Faraday cup being inserted while the WDS is still running.

[John Donovan]

Ben Buse has been doing some impressive work integrating WDS and EDS for quantitative mapping (more on this next week!), but he also shared with me his PFE MDB file for the test project he's been working with:

https://smf.probesoftware.com/index.php?topic=1845.msg14222#msg14222

I was especially struck by him combining EDS for major elements (with a small aperture) and WDS for traces at 1000 nA of beam current!



Very impressive!

I wanted to point out that one can also perform interference corrections between EDS elements (and WDS/EDS elements also). As Ben points in the post linked above, the NeXL seems to deconvolve the Mn Kb interference on Fe better than the JEOL algorithm.

Nevertheless let's look at this interference in Ben's run just using the JEOL de-convolved EDS net intensity data. Here a Mn metal standard without the interference correction for Mn and Fe both as EDS elements:

St  818 Set   3 B4 Mn-metal, Results in Elemental Weight Percents
 
ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn
TYPE:     ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL
BGDS:      EDS     EDS     EDS     EDS     EDS     LIN     LIN     LIN     LIN     EDS
TIME:    40.00   40.00   40.00   40.00   40.00   10.00   10.00   10.00   10.00   40.00
BEAM:  1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn   SUM  
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (la)    (la)    (ka)
    47    .243    .056   4.019    .039    .000    .004    .000    .010    .004 100.348 104.724
    48    .159    .020   3.579    .149    .000    .002    .002    .005   -.002 100.054 103.966
    49    .063    .071   3.620    .038    .173    .007    .003    .002    .004 100.133 104.112

AVER:     .155    .049   3.739    .075    .058    .004    .002    .005    .002 100.178 104.268
SDEV:     .090    .026    .243    .064    .100    .002    .002    .004    .004    .152    .402
SERR:     .052    .015    .141    .037    .058    .001    .001    .002    .002    .088
%RSD:    58.30   54.11    6.51   84.77  173.20   49.05  109.95   78.22  165.86     .15

PUBL:     n.a.    n.a.    n.a.    n.a.    n.a.    n.a.    n.a.    n.a.    n.a. 100.000 100.000
%VAR:      ---     ---     ---     ---     ---     ---     ---     ---     ---   (.18)
DIFF:      ---     ---     ---     ---     ---     ---     ---     ---     ---   (.18)
STDS:      803     804    1051     807     824     511     511     502     502     818

STKF:    .0782   .2974   .6607   .3211   .3179   .5829   .5829  1.0000  1.0000  1.0000
STCT:     1.60    4.54    6.01    7.14    6.58 2027.59 2238.91  668.88 2450.28    9.82

UNKF:    .0006   .0005   .0383   .0002   .0001   .0000   .0000   .0001   .0000  1.0000
UNCT:      .01     .01     .35     .00     .00     .13     .05     .04     .06    9.82
UNBG:      .16     .25    1.09     .14     .10   21.55   11.89    4.59   19.85     .64

ZCOR:   2.4076   .8999   .9765  3.5173  5.2513  1.1832  1.1832   .9857   .9858  1.0018
KRAW:    .0082   .0018   .0580   .0007   .0003   .0001   .0000   .0001   .0000  1.0000
PKBG:     1.08    1.03    1.32    1.04    1.02    1.01    1.00    1.01    1.00   16.38

That's close to 4 wt% of an artifact.  Now with the interference correction:

St  818 Set   3 B4 Mn-metal, Results in Elemental Weight Percents
 
ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn
TYPE:     ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL
BGDS:      EDS     EDS     EDS     EDS     EDS     LIN     LIN     LIN     LIN     EDS
TIME:    40.00   40.00   40.00   40.00   40.00   10.00   10.00   10.00   10.00   40.00
BEAM:  1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00 1004.00

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn   SUM  
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (la)    (la)    (ka)
    47    .242    .055    .273    .039    .000    .004    .000    .010    .004 100.253 100.881
    48    .158    .020   -.156    .148    .000    .002    .002    .005   -.002  99.957 100.134
    49    .063    .071   -.118    .037    .172    .007    .003    .002    .004 100.037 100.277

AVER:     .154    .049    .000    .075    .057    .004    .002    .005    .002 100.083 100.431
SDEV:     .090    .026    .238    .063    .099    .002    .002    .004    .004    .153    .397
SERR:     .052    .015    .137    .037    .057    .001    .001    .002    .002    .088
%RSD:    58.29   54.11    ----   84.77  173.20   49.05  109.95   78.22  165.86     .15

PUBL:     n.a.    n.a.    n.a.    n.a.    n.a.    n.a.    n.a.    n.a.    n.a. 100.000 100.000
%VAR:      ---     ---     ---     ---     ---     ---     ---     ---     ---   (.08)
DIFF:      ---     ---     ---     ---     ---     ---     ---     ---     ---   (.08)
STDS:      803     804    1051     807     824     511     511     502     502     818

STKF:    .0782   .2974   .6607   .3211   .3179   .5829   .5829  1.0000  1.0000  1.0000
STCT:     1.60    4.54    6.01    7.14    6.58 2027.59 2238.91  668.88 2450.28    9.82

UNKF:    .0006   .0005   .0000   .0002   .0001   .0000   .0000   .0001   .0000  1.0000
UNCT:      .01     .01     .00     .00     .00     .13     .05     .04     .06    9.82
UNBG:      .16     .25    1.09     .14     .10   21.55   11.89    4.59   19.85     .64

ZCOR:   2.3967   .8974   .9757  3.4994  5.2234  1.1818  1.1818   .9831   .9831  1.0008
KRAW:    .0082   .0018   .0000   .0007   .0003   .0001   .0000   .0001   .0000  1.0000
PKBG:     1.08    1.03    1.00    1.04    1.02    1.01    1.00    1.01    1.00   16.38
INT%:     ----    ---- -100.27    ----    ----    ----    ----    ----    ----    ----

Zero of course since this is the standard for the interference correction. Though that's a pretty high variance, but he is running the EDS at a very low count rate with only ~8% dead time.

Now let's look at an unknown sample, a Mn rich garnet from Ben's run, first without the interference correction:

Un   24 6 SA-1 Alm/spess garnet, Results in Elemental Weight Percents
 
ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn       O
TYPE:     ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    CALC
BGDS:      EDS     EDS     EDS     EDS     EDS     LIN     LIN     LIN     LIN     EDS
TIME:   100.00  100.00  100.00  100.00  100.00   40.00   40.00   40.00   40.00  100.00     ---
BEAM:   999.13  999.13  999.13  999.13  999.13  999.13  999.13  999.13  999.13  999.13     ---

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn       O   SUM  
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (la)    (la)    (ka)      ()
    65  16.489    .241   9.972  10.308    .000    .008    .009    .003    .003  25.716  38.406 101.156
    66  16.374    .287   9.963  10.332    .073    .009    .010    .006    .004  25.564  38.317 100.939
    67  16.284    .250   9.993  10.392    .000    .009    .009    .007    .000  25.680  38.247 100.871

AVER:   16.382    .259   9.976  10.344    .024    .009    .009    .005    .002  25.654  38.323 100.989
SDEV:     .103    .025    .015    .043    .042    .001    .000    .002    .002    .080    .080    .149
SERR:     .059    .014    .009    .025    .024    .000    .000    .001    .001    .046    .046
%RSD:      .63    9.48     .16     .42  173.20    6.86    3.81   40.12  106.18     .31     .21
STDS:      803     804    1051     807     824     511     511     502     502     818     ---

STKF:    .0782   .2974   .6607   .3211   .3179   .5829   .5829  1.0000  1.0000  1.0000     ---
STCT:     1.60    4.54    6.01    7.14    6.58 2027.59 2238.91  668.88 2450.28    9.82     ---

UNKF:    .0726   .0023   .0894   .0406   .0001   .0001   .0001   .0000   .0000   .2240     ---
UNCT:     1.48     .04     .81     .90     .00     .24     .29     .03     .04    2.20     ---
UNBG:      .25     .15     .33     .17     .11   13.17    7.24    2.49   10.65     .21     ---

ZCOR:   2.2574  1.1152  1.1164  2.5456  3.6910  1.2446  1.2446  1.1998  1.1998  1.1454     ---
KRAW:    .9280   .0078   .1353   .1265   .0002   .0001   .0001   .0000   .0000   .2240     ---
PKBG:     6.85    1.24    3.44    6.24    1.01    1.02    1.04    1.01    1.00   11.60     ---

And now with the interference correction:

Un   24 6 SA-1 Alm/spess garnet, Results in Elemental Weight Percents
 
ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn       O
TYPE:     ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    CALC
BGDS:      EDS     EDS     EDS     EDS     EDS     LIN     LIN     LIN     LIN     EDS
TIME:   100.00  100.00  100.00  100.00  100.00   40.00   40.00   40.00   40.00  100.00     ---
BEAM:   999.13  999.13  999.13  999.13  999.13  999.13  999.13  999.13  999.13  999.13     ---

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn       O   SUM  
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (ka)    (la)    (la)    (ka)      ()
    65  16.453    .242   9.019  10.251    .000    .008    .009    .003    .003  25.735  38.047  99.769
    66  16.339    .288   9.015  10.275    .073    .009    .010    .006    .004  25.582  37.960  99.560
    67  16.249    .250   9.041  10.335    .000    .009    .009    .007    .000  25.699  37.888  99.486

AVER:   16.347    .260   9.025  10.287    .024    .009    .009    .005    .002  25.672  37.965  99.605
SDEV:     .102    .025    .014    .043    .042    .001    .000    .002    .002    .080    .079    .147
SERR:     .059    .014    .008    .025    .024    .000    .000    .001    .001    .046    .046
%RSD:      .63    9.48     .15     .42  173.20    6.86    3.81   40.12  106.18     .31     .21
STDS:      803     804    1051     807     824     511     511     502     502     818     ---

STKF:    .0782   .2974   .6607   .3211   .3179   .5829   .5829  1.0000  1.0000  1.0000     ---
STCT:     1.60    4.54    6.01    7.14    6.58 2027.59 2238.91  668.88 2450.28    9.82     ---

UNKF:    .0726   .0023   .0808   .0406   .0001   .0001   .0001   .0000   .0000   .2240     ---
UNCT:     1.48     .04     .73     .90     .00     .24     .29     .03     .04    2.20     ---
UNBG:      .25     .15     .33     .17     .11   13.17    7.24    2.49   10.65     .21     ---

ZCOR:   2.2526  1.1165  1.1171  2.5315  3.6680  1.2445  1.2445  1.2011  1.2011  1.1462     ---
KRAW:    .9280   .0078   .1223   .1265   .0002   .0001   .0001   .0000   .0000   .2240     ---
PKBG:     6.85    1.24    3.21    6.24    1.01    1.02    1.04    1.01    1.00   11.60     ---
INT%:     ----    ----   -9.60    ----    ----    ----    ----    ----    ----    ----     ---

I don't know what implications a 10% error in Fe has for garnet thermo-barometry, but it can't be good!

I like that the duplicate WDS Ga and Sn elements agree quite well (90 PPM for the Ga channels and within a std deviation of zero for the Sn channels).

I've never run analyses this high a beam current. Has anyone else?

[John Donovan]

I like that the duplicate WDS Ga and Sn elements agree quite well (90 PPM for the Ga channels and within a std deviation of zero for the Sn channels).

I've never run analyses this high a beam current. Has anyone else?

Because Probe for EPMA acquires standards and unknowns in the same data file, it's pretty cool that Ben can just change the EDS aperture, and get quantitative results immediately after standardizing.  Note also that I am able to re-process his EDS data off-line without access to the JEOL server because he checked the "Only Use Stored EDS intensities..." checkbox as shown here:

https://smf.probesoftware.com/index.php?topic=79.msg13471#msg13471

This means one can re-process their WDS and EDS data off-line in your office, at home or when traveling!

I should also point out a couple more interesting aspects to Ben's run.  Ga and Sn trace elements were run at 30 keV, so really "cooking". Count times were 40 sec on-peak and 40 sec off-peak, the calculated (single channel) detection limits are shown here:

Detection limit at 99 % Confidence in Elemental Weight Percent (Single Line):

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn
    65    .052    .027    .067    .045    .055    .002    .001    .004    .002    .050
    66    .053    .026    .067    .045    .054    .002    .001    .004    .002    .051
    67    .053    .027    .067    .045    .055    .002    .001    .004    .002    .050

AVER:     .053    .027    .067    .045    .055    .002    .001    .004    .002    .050
SDEV:     .001    .000    .000    .000    .001    .000    .000    .000    .000    .001
SERR:     .000    .000    .000    .000    .000    .000    .000    .000    .000    .000

Detection Limit (t-test) in Elemental Weight Percent (Average of Sample):

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn
  60ci     ---    .001     ---     ---    .001    .000    .000    .001    .001     ---
  80ci     ---    .002     ---     ---    .002    .000    .000    .002    .002     ---
  90ci     ---    .003     ---     ---    .003    .000    .001    .002    .003     ---
  95ci     ---    .004     ---     ---    .005    .000    .001    .003    .005     ---
  99ci     ---    .009     ---     ---    .011    .001    .002    .008    .011     ---

So 20 to 40 PPM for single points or 10 to 50 PPM for 99% confidence t-test for the average! I guess we need to add some more significant digits as the first Ga channel has a 95% confidence t-test detection limit of less than 10 PPM! Though one can export detection limits with more precision using the User Specified output shown here:

https://smf.probesoftware.com/index.php?topic=93.msg8035#msg8035

But remember, we can also aggregate these duplicate channels to get even better statistics:

Detection limit at 99 % Confidence in Elemental Weight Percent (Single Line):

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn
    65    .052    .027    .067    .045    .055    .001    .000    .002    .000    .050
    66    .053    .026    .067    .045    .054    .001    .000    .002    .000    .051
    67    .053    .027    .067    .045    .055    .001    .000    .002    .000    .050

AVER:     .053    .027    .067    .045    .055    .001    .000    .002    .000    .050
SDEV:     .001    .000    .000    .000    .001    .000    .000    .000    .000    .001
SERR:     .000    .000    .000    .000    .000    .000    .000    .000    .000    .000

Detection Limit (t-test) in Elemental Weight Percent (Average of Sample):

ELEM:       Si      Ca      Fe      Al      Mg      Ga      Ga      Sn      Sn      Mn
  60ci     ---    .001     ---     ---    .001    .000     ---    .001     ---     ---
  80ci     ---    .002     ---     ---    .002    .000     ---    .002     ---     ---
  90ci     ---    .003     ---     ---    .003    .000     ---    .003     ---     ---
  95ci     ---    .004     ---     ---    .005    .000     ---    .004     ---     ---
  99ci     ---    .009     ---     ---    .011    .001     ---    .010     ---     ---

Pretty impressive for Ga Ka and Sn La emission lines.

And finally here is the automatic Report output for this sample:

Un   24 6 SA-1 Alm/spess garnet
TakeOff = 40.0  KiloVolt = 30.0  Beam Current = 1000.  Beam Size =    5
(Magnification (analytical) =  20000),        Beam Mode = Analog  Spot
(Magnification (default) =     1000, Magnification (imaging) =    100)
Image Shift (X,Y):                                         .00,    .00

Formula Based on Sum of Cations = .000   Oxygen Calc. by Stoichiometry

Compositional analyses were acquired on an electron microprobe (JEOL 8230/8530 (TCP/IP Socket and EIKS)) equipped with 5 tunable wavelength dispersive spectrometers.

EDS spectra were acquired and processed using a JEOL OEM EDS system.

Operating conditions were 40 degrees takeoff angle, and a beam energy of 30 keV.
The beam current was 1000 nA, and the beam diameter was 5 microns.

Elements were acquired using analyzing crystals EDS for Si ka, Ca ka, Fe ka, Al ka, Mg ka, Mn ka, LIFL for Ga ka, LIFH for Ga ka, PETH for Sn la, and PETJ for Sn la.

The standards were CSTD1 Sn-metal for Sn la, Sn la, CSTD1 GaAs for Ga ka, Ga ka, B4 SJIO for Si ka, B4 Wollastonite 1 for Ca ka, B4 Al2O3 for Al ka, B4 Mn-metal for Mn ka, B4 MgO for Mg ka, and Hematite for Fe ka.

B4 SJIO
Standard Description

B4 Wollastonite 1
Standard Description

Hematite
Standard Description

B4 Al2O3
Standard Description

B4 MgO
Standard Description

B7 Neodymium glass
Univ Edinburgh synthetic

SiO2 glass
Standard Description

B4 Diopside
Loch Shin

CSTD1 GaAs
Standard Description

CSTD1 Sn-metal


B4 Mn-metal
Standard Description

The counting time was 40 seconds for Sn la, Ga ka, and 100 seconds for Mg ka, Si ka, Ca ka, Fe ka, Al ka, Mn ka.

The off peak counting time was 40 seconds for Ga ka, Sn la.

Off Peak correction method was Linear for Ga ka, Sn la, Sn la, Ga ka.

Unknown and standard intensities were corrected for deadtime using the Logarithmic (Moy) correction method.

Donovan, J. J., Moy, A., von der Handt, A., Gainsforth, Z., Maner, J. L., Nachlas, W., & Fournelle, J. (2023). A New Method for Dead Time Calibration and a New Expression for Correction of WDS Intensities for Microanalysis. Microscopy and Microanalysis, 29(3), 1096-1110.

Standard intensities were corrected for standard drift over time on an element by element basis.

Interference corrections were applied to Fe for interference by Mn.

Donovan, J. J., Snyder, D. A., & Rivers, M. L. (1992). An improved interference correction for trace element analysis. In Proceedings of the Annual Meeting-Electron Microscopy Society of America (pp. 1646-1646). San Francisco Press.

Results are the average of 3 points and detection limits ranged from .001 weight percent for Ga ka to .027 weight percent for Ca ka to .053 weight percent for Si ka to .067 weight percent for Fe ka.

Analytical sensitivity (at the 99% confidence level) ranged from 6.158 percent relative for Ga ka to 46.960 percent relative for Sn la.

Goldstein, J. I. (1992). Scanning electron and x-ray microanalysis. A text for biologists, materials scientists, and geologists, 395-416.

Oxygen was calculated by cation stoichiometry and included in the matrix correction. 

Moy, A., Fournelle, J., Nachlas, W., Dungan, M., Locock, A., Bullock, E., ... & Handt, A. V. D. (2023). On the Importance of Including All Elements in the EPMA Matrix Correction.


The aggregate intensity option was selected.

Donovan, J. J., Lowers, H. A., & Rusk, B. G. (2011). Improved electron probe microanalysis of trace elements in quartz. American Mineralogist, 96(2-3), 274-282.

The matrix correction method was ZAF or Phi-Rho-Z Calculations and the mass absorption coefficients dataset was FFAST    Chantler (NIST v 2.1, 2005).

The ZAF or Phi-Rho-Z algorithm utilized was Pouchou and Pichoir-Full (PAP).

Armstrong, J. T. (1988). Quantitative analysis of silicate and oxide minerals: comparison of Monte Carlo, ZAF and phi-rho-z procedures. Analysis microbeam.