Author Topic: Troilite analysis problems  (Read 4833 times)

John Donovan

  • Administrator
  • Emeritus
  • *****
  • Posts: 2499
  • Other duties as assigned...
    • Probe Software
Re: Troilite analysis problems
« Reply #15 on: May 26, 2016, 06:09:09 am »
Just to reply to my own thread...
John in an earlier post I edited I was having a problem slowing the peak scanning down for integrated intensities:

EDIT: Hi all. I am in the process of trying the integrated intensities, and noticed that the wavescan it is doing on Fe and S is very fast. I can change the number of steps no problems in the Elements/Cations box, but not matter what count times I change for Wave/Peak scan I cant get it to slow down. Is there some way I can change the count times of the wavescan it does during the Peak integration scan?

Can this be done or is this a fixed speed? I imagine it can be done given you can change the step sizes for the integration scan.

cheers

Hi Ben,
I will need to look at this when I get back to Eugene.   
john
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

  • Administrator
  • Emeritus
  • *****
  • Posts: 2499
  • Other duties as assigned...
    • Probe Software
Re: Troilite analysis problems
« Reply #16 on: June 05, 2016, 03:38:27 pm »
Just to reply to my own thread...
John in an earlier post I edited I was having a problem slowing the peak scanning down for integrated intensities:

EDIT: Hi all. I am in the process of trying the integrated intensities, and noticed that the wavescan it is doing on Fe and S is very fast. I can change the number of steps no problems in the Elements/Cations box, but not matter what count times I change for Wave/Peak scan I cant get it to slow down. Is there some way I can change the count times of the wavescan it does during the Peak integration scan?

Can this be done or is this a fixed speed? I imagine it can be done given you can change the step sizes for the integration scan.

cheers

Hi Ben,
I fooled myself for a moment when looking into this, but it turns out not to be a bug.  The deal is, the on-peak count time is for the *full* scan.  So if you put in 10 seconds, each point will be a fraction of that time.  You can put in say 120 seconds and will see it takes longer.  It is true that this could have been clearer so I added some text to the count time dialog.  I checked both the step/count integrated scanning and also the ROM integrated scanning.  They both seem to work.

But there was a small display bug in the graph so I fixed that and also added the current count time per point to the x axis label since it varies as a function of intensity.

Download the latest PFE v. 11.51 and you should be OK.
john
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

BenjaminWade

  • Professor
  • ****
  • Posts: 176
Re: Troilite analysis problems
« Reply #17 on: June 05, 2016, 05:12:51 pm »
Awesome, thanks again John. I was trying to alter it via wavescan timing, but didn't even think of using the on peak count time to alter it.
Just FYI we conducted the troilite analysis using what we had at the time with the fast peak scanning. It was hard with the superfast integrated intensities to notice fine moves in peak shift, but some were definitely very different and some not at all. Which is strange but its a natural troilite sample so who knows what kind of variation to expect in it. In addition there were definitely peak shifts in S Ka as we might expect. Totals ranged from too low now (by 1-2 wt%), however some were very close to what we might expect.

We are hoping to rerun his samples again soon now with the slower integrated intensities and will see what happens. He is also getting his lab back in Nanjing to send us some of their synthetic troilite which we might play around with.

Cheers

Probeman

  • Emeritus
  • *****
  • Posts: 1975
  • Never sleeps...
    • John Donovan
Re: Troilite analysis problems
« Reply #18 on: June 05, 2016, 06:00:57 pm »
Awesome, thanks again John. I was trying to alter it via wavescan timing, but didn't even think of using the on peak count time to alter it.
Just FYI we conducted the troilite analysis using what we had at the time with the fast peak scanning. It was hard with the superfast integrated intensities to notice fine moves in peak shift, but some were definitely very different and some not at all. Which is strange but its a natural troilite sample so who knows what kind of variation to expect in it. In addition there were definitely peak shifts in S Ka as we might expect. Totals ranged from too low now (by 1-2 wt%), however some were very close to what we might expect.

We are hoping to rerun his samples again soon now with the slower integrated intensities and will see what happens. He is also getting his lab back in Nanjing to send us some of their synthetic troilite which we might play around with.

Cheers

Although one can do excellent quant on sulfides/sulfates using different peaked positions for each phase, with sufficient precision one should be able to get excellent quant on pyrite, pyrrhotite, sulfate and toilite, using the same standard, by using the integrated intensity scan method... without knowing the exact peak position in advance!

Some one should do that and post it here!

You'll need a decent integration time, and I'd bring in the backgrounds a little closer to the peak as we don't need to spend too much time on the tails, but not too close, so as to allow for the expected peak shifting!
The only stupid question is the one not asked!

Probeman

  • Emeritus
  • *****
  • Posts: 1975
  • Never sleeps...
    • John Donovan
Re: Troilite analysis problems
« Reply #19 on: June 06, 2016, 01:06:30 pm »
Awesome, thanks again John. I was trying to alter it via wavescan timing, but didn't even think of using the on peak count time to alter it.

I guess one way to look at it is, that although we are really just running a tight wavescan on the peak, because it is for quantitative analysis of unknowns and standards, I decided to use the on-peak time (and the off-peak times as well for the background correction which is subtracted from the integrated intensity to get a net intensity for quantification). Hence why we specify the total integration time rather than the count time per point as in a traditional wavescan.
john
The only stupid question is the one not asked!

Probeman

  • Emeritus
  • *****
  • Posts: 1975
  • Never sleeps...
    • John Donovan
Re: Troilite analysis problems
« Reply #20 on: June 08, 2016, 03:52:51 pm »
So I ran some integrated intensity scans on pyrite, pyrrhotite and anhydrite and the results are interesting.  I ran both with the step/count method which utilizes a variable step size to concentrate more points on the peaks and also ran a straight ROM scan which uses a fixed step size.

Attached below (you must be logged in to see them), are the scans using the step/count method (variable step size). Here are the quant running pyrite against itself:

ELEM:        S       S       S       S       S      Fe      Ti   SUM 
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)      ()      ()
   235  51.565  51.655  51.492  51.731  51.550  46.550    .058 304.601
   236  51.432  51.428  51.512  51.306  51.520  46.550    .058 303.807

AVER:   51.499  51.542  51.502  51.518  51.535  46.550    .058 304.204
SDEV:     .094    .160    .014    .301    .021    .000    .000    .561
SERR:     .067    .113    .010    .213    .015    .000    .000
%RSD:      .18     .31     .03     .58     .04     .00     .00

PUBL:   53.450  53.450  53.450  53.450  53.450  46.550    .058 100.058
%VAR:  (-3.65) (-3.57) (-3.64) (-3.61) (-3.58)     .00     .00
DIFF:  (-1.95) (-1.91) (-1.95) (-1.93) (-1.91)    .000    .000
STDS:      730     730     730     730     730     ---     ---


Note that the concentrations of sulfur are low, because we measured sulfur on all 5 channels. This results in a non-physical matrix correction. Turning on the "aggregate" mode we now get this for the quant:

ELEM:        S       S       S       S       S      Fe      Ti   SUM 
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)      ()      ()
   235  53.488    .000    .000    .000    .000  46.550    .058 100.096
   236  53.365    .000    .000    .000    .000  46.550    .058  99.973

AVER:   53.427    .000    .000    .000    .000  46.550    .058 100.035
SDEV:     .087    .000    .000    .000    .000    .000    .000    .087
SERR:     .062    .000    .000    .000    .000    .000    .000
%RSD:      .16   .0000   .0000   .0000   .0000     .00     .00

PUBL:   53.450    n.a.    n.a.    n.a.    n.a.  46.550    .058 100.058
%VAR:   (-.04)   (.00)   (.00)   (.00)   (.00)     .00     .00
DIFF:   (-.02)     ---     ---     ---     ---    .000    .000
STDS:      730       0       0       0       0     ---     ---


As expected since it is assigned to itself.  What about pyrrhotite?

ELEM:        S       S       S       S       S      Fe   SUM 
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)      ()
   238  39.518    .000    .000    .000    .000  59.808  99.326
   239  39.428    .000    .000    .000    .000  59.808  99.236

AVER:   39.473    .000    .000    .000    .000  59.808  99.281
SDEV:     .063    .000    .000    .000    .000    .000    .063
SERR:     .045    .000    .000    .000    .000    .000
%RSD:      .16   .0000   .0000   .0000   .0000     .00

PUBL:   39.580    n.a.    n.a.    n.a.    n.a.  59.808  99.388
%VAR:     -.27     .00     .00     .00     .00     .00
DIFF:    -.107     ---     ---     ---     ---    .000
STDS:      730       0       0       0       0     ---


Nice. And now for anhydrite:

ELEM:        S       S       S       S       S      Ca       O      Sr   SUM 
XRAY:     (ka)    (ka)    (ka)    (ka)    (ka)      ()      ()      ()
   203  23.157    .000    .000    .000    .000  29.380  46.881    .300  99.718
   204  23.781    .000    .000    .000    .000  29.380  46.881    .300 100.342
   205  23.435    .000    .000    .000    .000  29.380  46.881    .300  99.996
   206  23.283    .000    .000    .000    .000  29.380  46.881    .300  99.844
   207  23.464    .000    .000    .000    .000  29.380  46.881    .300 100.025

AVER:   23.424    .000    .000    .000    .000  29.380  46.881    .300  99.985
SDEV:     .235    .000    .000    .000    .000    .000    .000    .000    .235
SERR:     .105    .000    .000    .000    .000    .000    .000    .000
%RSD:     1.00   .0000   .0000   .0000   .0000     .00     .00     .00

PUBL:   23.499    n.a.    n.a.    n.a.    n.a.  29.380  46.881    .300 100.060
%VAR:     -.32     .00     .00     .00     .00     .00     .00     .00
DIFF:    -.075     ---     ---     ---     ---    .000    .000    .000
STDS:      730       0       0       0       0     ---     ---     ---


The anhydrite scan was run using a shorter count time than the sulfides to avoid sample damage. But it is clear that using the integrated intensity method provides quantitative results when one does not know in advance the emission peak position.
The only stupid question is the one not asked!

Probeman

  • Emeritus
  • *****
  • Posts: 1975
  • Never sleeps...
    • John Donovan
Re: Troilite analysis problems
« Reply #21 on: June 08, 2016, 04:01:17 pm »
Before someone mentions it, yes there is something funky about my PET crystal on spectro 1. Below are scans from spectro 2 and 4 which look much more normal...

Anyone know what might be the problem with spectro 1?
The only stupid question is the one not asked!

Malcolm Roberts

  • Professor
  • ****
  • Posts: 131
Re: Troilite analysis problems
« Reply #22 on: June 08, 2016, 09:25:42 pm »
Hi John
This is interesting. I am keen to give it a go myself at a suitable opportunity. Just for clarification......... The on-peak interference correction procs work in usual way when the integrated intensity technique is used?
As a MAN aficionado, I am wondering whether this technique couldn't be expanded to include rather than off-peak positions, high and low peak limits that could be integrated and used for MAN. Might sound mad, but what do you think? Obviously choice of positions would need to be such that any peak shifts would be grabbed.
Cheers,
Malc.

Probeman

  • Emeritus
  • *****
  • Posts: 1975
  • Never sleeps...
    • John Donovan
Re: Troilite analysis problems
« Reply #23 on: June 08, 2016, 09:36:56 pm »
Hi John
This is interesting. I am keen to give it a go myself at a suitable opportunity. Just for clarification......... The on-peak interference correction procs work in usual way when the integrated intensity technique is used?
As a MAN aficionado, I am wondering whether this technique couldn't be expanded to include rather than off-peak positions, high and low peak limits that could be integrated and used for MAN. Might sound mad, but what do you think? Obviously choice of positions would need to be such that any peak shifts would be grabbed.
Cheers,
Malc.

Hi Malcolm,
Maybe crazy, but it could work, I guess, but the MAN standards would have to be acquired using integrated intensities as well...  worth a try I guess.

But yes, the interference corrections should work with integrated intensities also. Frankly I was just pleased that the aggregate mode worked fine for these integrated intensities!
john
The only stupid question is the one not asked!

Malcolm Roberts

  • Professor
  • ****
  • Posts: 131
Re: Troilite analysis problems
« Reply #24 on: June 13, 2016, 06:47:10 pm »
Hi John
Another question..... Does the interference correction have to be applied using an equivalent counting technique? By that I mean, if I want to measure element A using the integrated method, and element A has an interference from element B, do I need to acquire element B using the integrated method too or are the on-peak methods fine to go with? My gut-feeling is to use the same approach with both.........
Cheers,
malc.

Probeman

  • Emeritus
  • *****
  • Posts: 1975
  • Never sleeps...
    • John Donovan
Re: Troilite analysis problems
« Reply #25 on: June 13, 2016, 08:52:33 pm »
Hi John
Another question..... Does the interference correction have to be applied using an equivalent counting technique? By that I mean, if I want to measure element A using the integrated method, and element A has an interference from element B, do I need to acquire element B using the integrated method too or are the on-peak methods fine to go with? My gut-feeling is to use the same approach with both.........
Cheers,
malc.

Hi Malcolm,
Great question but I think you'll be fine for the interference correction between integrated and peak intensities.  Why do I think so?  Because this:

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

Think of it as though one were performing an interference correction between a normal sized WDS analyzing crystal and a large WDS analyzing crystal.  As long as the interference standard and the unknown are both measured using the same spectrometer (and element and x-ray), "it all comes out in the wash" as they say!
john
« Last Edit: June 13, 2016, 10:03:05 pm by Probeman »
The only stupid question is the one not asked!

BenjaminWade

  • Professor
  • ****
  • Posts: 176
Re: Troilite analysis problems
« Reply #26 on: June 22, 2016, 06:50:00 pm »
Hi John
Many thanks for doing that test, interesting. The student is yet to be back to reanalyse them but will hopefully get to try out the integrated intensities again soon.

Cheers