Re: Si-PIN of microlites from Hollister and Oak Leaf in CT
|
You¡¯ll need a really long run to pull anything out me thinks.
Attached is the actual AM241 Okaklef.mca file spectra blown up and
what could hardly be called peaks in ROI¡¯s
There is no Ta or Hf shown at the cursor location 56.27 keV of the
same rock. Are we looking at the same spectra here? One is natural (I assume)
and this one is Am excited.
This is noisy under sampled spectra that should not be used for
picks as it needs a lot more count time before any attribution should be
assigned to anything.
Why the difference between Thermo and Amptek peaks? There is
something wrong here.? Why are you using a 17.74 Np Lb when a La at 13.9 would
be stronger?
Do your own calibration and ALWAYS check the calibration. Things drift
and it doesn¡¯t take much with high res detectors.?
Send the natural .mca file
Dud
?
?
?
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of Charles David Young
Sent: Sunday, January 12, 2020 2:29 PM
To: [email protected]
Cc: Mike Loughlin; Steve Dubyk; albinitony@...
Subject: Re: [XRF] Si-PIN of microlites from Hollister and Oak Leaf in
CT
?
Here is the .mca.? I'll also do a longer run to get at
least twice the counts.
Regarding Ecal, Theremino is not doing that.? The Ecal
was done by Geo and I am using the two points that he provided:
I just adjusted the master energy trimmer on Theremino until
those 2 points line up.? From all the other testing I have done this seems
to make all the known peaks line up very well over the entire range.? I
have not tried to calibrate the detector or make any other changes to the
adjustments that Geo did.
Regarding Hf actually being Ta Ka2, I considered that but they are
really not very close:
No doubt the longer count will help clarify that but it sure looks
like a strong, centered Hf peak to me.
The PtLa1=9.44 is also very strong and centered.? It is hard
to understand how it could be anything else given that the BiLa and TaKa1 that
bracket it seem to be dead on.? I looked at the PtKa1 and it is pretty
small but I was thinking that was perhaps due to the rolloff at the high
energy.? A longer count may help clarify this as well.
So let's revisit this once I get a longer run.
?
On Sun, Jan 12, 2020 at 1:27 PM Dude <dfemer@...> wrote:
Charles
,
The
spectra needs a lot more count time before one can say much about what¡¯s in
it.? Please don¡¯t overlay spectra it makes it hard to ?interpret, complicates
everything and you can¡¯t see the baseline which is important.? The Hf is
actually the Ta Ka2.? I don¡¯t see any Ce in this that couldn¡¯t be
accounted for by counting noise The Ce and Yb are really too noisy single lines
to say anything about.? There¡¯s also no Pt in there either as its missing
the Ka. The? Rb Ka1 is actually the Th La1.? The Sn has something
forming there but it is too noisy to interpret. I might add that Sn shows up a
commonly as instrument hosted rather than in the target so always be careful of
that one.
Look at
the Fe. The Ka line doesn¡¯t match the ¡°peak¡±, Fe should fall? at 6.3 and
6.4. This peak is below that around 5.9 which could be Mn Ka at 5.89.? I
would not interpret that as a peak unless there were many more counts to define
it.? This spectra is really too noisy to do a good interp by just looking
at the Theromino plots.? A single line where one thinks something should
be does not define a peak.?
I would
suggest doing some ?things:
1.?????? Don¡¯t
preconceive anything. Don¡¯t look up what¡¯s suppose to be in it and fool
yourself into seeing what¡¯s not there ¨C prove it.? Shoot it first, then do
an interpretation of the clear peaks.? If you need more count time to
define something then shoot it again and add the spectra together.? Then
do the look up and see if it fits.
2.?????? Thermino
is not the right tool to use for a high resolution system.? It looks like
the Ecal is off somewhat and picking peak centroids is problematic. Peaks are
gaussian not one single line. A NaI detector is pretty broad and there is slop
as to what the actual energy really is. A SDD or Si-Pin has much tighter
resolution and the actual peak is going to be right where it belongs and well
defined with the proper ECal.? If it isn¡¯t, it ain¡¯t - ?close doesn¡¯t
count.? You have a profession MCA and software use it.? When posting
pictures also post the .mca file so everyone can play with it.?
3.?????? Use the
AmpTek software and do ROI¡¯s on the peaks. Run the Peak ID and Energy table to
?see the peak energy and the computers best idea of L and K ?ID, but
don¡¯t believe it, double check it. Show a copy of that table so everyone can
see the ROI¡¯s energy peak.
4.?????? Get an
active source, count times well be faster, data better and it will help to
minimize the natural radiation interferences.
?
If this
is a Microlite where is the Ca? why is there more Th and no U?. Count it longer
to see if the Yb comes out, if so it may? be a xenotime
Dud
?
I
bought a couple of nice microlites from Tony Albini a while back.? One was
actually identified as xenotime but my NaI scan showed a strong Ta presence so
I proposed that it was actually microlite and Tony agreed.? The Si-PIN
scans confirm these ids, which are remarkable for the number of elements
identified that can be in microlite.
Microlite
is a complex mineral with many possible elements.? One constant though is
the presence of Ta, which really stands out in my attached NaI scans.? In
the Si-PIN scans the Ta up at 57.5 is not as prominent but I believe that is
because the Si-PIN is not as sensitive in the higher energies.
microlite
formula: A2-mTa2X6-wZ-n
The A site may host Na, Ca, Sr, Pb2+, Sn2+, Sb3+, Y, U, or H2O or Ag, Mn, Ba,
Fe2+, Bi3+, Ce, Sc or Th.
elements identified by Si-PIN:
Ta Hf Pt Sn Rb Ce Fe Yb
|
