News:

:) You cannot see "members only" boards if you are not a member, so please join the forum!

Main Menu

Sulfur Peak Shifts

Started by Probeman, December 02, 2013, 01:30:26 PM

Previous topic - Next topic

alerner

One additional possibility is that EPMA analysis induced redox changes in the S in glasses, so that you have an evolving peak position during measurement to some more intermediate state.

S photo-oxidation from S2- was observed by Wallace and Carmichael 1994 and Rowe et al 2007. It is less clear what happens if you have originally S6+ in glasses, as perhaps a photoreduction may occur as in X-ray synchrotron measurements (Wilke et al 2008). The integrated results of EPMA induced beam damage might be S transforming to some intermediate valence charge between the S6+ and S2-. It would be interesting to see time dependent measurements for each of these peak positions during the course of analysis, to track any peak decreases and growth over time.

Probeman

#16
I just wanted to add more data to this topic because I will be referencing it for a subsequent post in the "The Limits of EPMA Accuracy" topic:

https://smf.probesoftware.com/index.php?topic=1831.0

Here are some sulfur analyses I did back in 2004 (yikes!). I was attempting to characterize the peak shift between the oxidized and reduced states of sulfur and found that if I peaked my sulfur spectrometer separately on the anhydrite standard and on the pyrite standard, I could obtain excellent accuracy analyzing one against the other, just by using the proper peak positions for each material. Here is the sulfur peak position for anhydrite:

On and Off Peak Positions:
ELEM:    ca ka   si ka   al ka   na ka    k ka   fe ka   ti ka    s ka    p ka   ni ka   mn ka   mg ka
CRYST:     PET     TAP     TAP     TAP     PET     LIF     LIF     PET     PET     LIF     LIF     TAP
ONPEAK 38217.0 27657.0 32393.0 46276.0 42590.0 48383.0 68504.0 61185.0 70150.0 41478.0 52490.0 38422.0

and here is the sulfur peak position for pyrite:

On and Off Peak Positions:
ELEM:    ca ka   si ka   al ka   na ka    k ka   fe ka   ti ka    s ka    p ka   ni ka   mn ka   mg ka
CRYST:     PET     TAP     TAP     TAP     PET     LIF     LIF     PET     PET     LIF     LIF     TAP
ONPEAK 38217.0 27657.0 32393.0 46276.0 42590.0 48383.0 68504.0 61220.0 70150.0 41478.0 52490.0 38422.0

That's a difference of 35 units on the Cameca spectrometer! Now using the anhydrite as our primary standard (#327) and analyzing pyrite as a secondary standard we obtain these analyses:

St  730 Set   1 Pyrite UC # 21334, Results in Elemental Weight Percents
 
ELEM:       Ca      Si      Al      Na       K      Fe      Ti       S       P      Ni      Mn      Mg       O
TYPE:     ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    SPEC
BGDS:      MAN     MAN     MAN     MAN     LIN     MAN     MAN     LIN     LIN     LIN     LIN     MAN
TIME:    10.00   10.00   10.00   10.00   10.00   10.00   10.00   10.00   10.00   10.00   10.00   10.00     ---
BEAM:    20.03   20.03   20.03   20.03   20.03   20.03   20.03   20.03   20.03   20.03   20.03   20.03     ---

ELEM:       Ca      Si      Al      Na       K      Fe      Ti       S       P      Ni      Mn      Mg       O   SUM 
   285   -.002   -.011    .009    .019   -.012  47.493    .047  53.344    .000   -.024   -.014    .001    .000 100.851
   286    .002    .002    .007   -.001   -.011  45.925    .018  53.140   -.028   -.031   -.024   -.012    .000  98.987
   287    .007   -.003    .002   -.010   -.008  46.280    .013  54.287    .037   -.082   -.006   -.009    .000 100.509

AVER:     .002   -.004    .006    .003   -.010  46.566    .026  53.590    .003   -.046   -.015   -.006    .000 100.116
SDEV:     .005    .006    .004    .015    .002    .822    .018    .612    .033    .032    .009    .007    .000    .992
SERR:     .003    .004    .002    .009    .001    .475    .010    .353    .019    .019    .005    .004    .000
%RSD:   199.96 -168.32   67.81  488.15  -21.21    1.77   69.02    1.14 1140.72  -70.62  -61.45 -108.26     .00

PUBL:     n.a.    n.a.    n.a.    n.a.    n.a.  46.550    .058  53.450    n.a.    .000    n.a.    n.a.    n.a. 100.058
%VAR:      ---     ---     ---     ---     ---     .03  -55.16     .26     ---     .00     ---     ---     ---
DIFF:      ---     ---     ---     ---     ---    .016   -.032    .140     ---    .000     ---     ---     ---
STDS:      358     298     374     336     374     395      22     327     285      28      25      12     ---

Yup, that's sulfur accuracy within 0.26% relative extrapolating from anhydrite to pyrite. How is that possible?  It's because the change in the sulfur peak for these two materials (oxidized vs. reduced) is a purely a "peak shift change", not a "peak shape change".  And because we adjusted our peak positions for each material, we compensated for that.

And please note that these "peak shift" effects will only get smaller at higher atomic number K lines:

https://smf.probesoftware.com/index.php?topic=1423.msg11809#msg11809

Of course these "peak shift" effects will get larger at lower atomic number K lines, and for the very low atomic number elements (e.g., O Ka, N Ka, etc.), they will get still larger and then we will see some "peak shape" changes in our wavescans.
The only stupid question is the one not asked!