|
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot some of my specimens with his XRF "gun", which has a pinpoint xray source.? Several of my rocks were found in the rubble piles of the local rock shop "Tucson Mineral and Gem World", which has been open for about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the charts!? Even more significant is that Dud's results show the same thing:
Nb 101.88% Fe 35.76% Ta 15.4% Mn 2.54% Y 0.279%
Comparing these percentages to the relative heights of the same peaks on my scan one finds similar proportions.? It is very satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has a relatively high Y component.
Charles
|
BTW, the small peak at about the FeSi escape actually aligns better with TiKa1.? Dud's XRF shows 0.6% Ti.? Are we sure this is not TiKa1?
Charles
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot some of my specimens with his XRF "gun", which has a pinpoint xray source.? Several of my rocks were found in the rubble piles of the local rock shop "Tucson Mineral and Gem World", which has been open for about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the charts!? Even more significant is that Dud's results show the same thing:
Nb 101.88% Fe 35.76% Ta 15.4% Mn 2.54% Y 0.279%
Comparing these percentages to the relative heights of the same peaks on my scan one finds similar proportions.? It is very satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has a relatively high Y component.
Charles
|
BTW, Y PPM = 279 = 0.0279%
BTW, the small peak at about the FeSi escape actually aligns better with TiKa1.? Dud's XRF shows 0.6% Ti.? Are we sure this is not TiKa1?
Charles
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot some of my specimens with his XRF "gun", which has a pinpoint xray source.? Several of my rocks were found in the rubble piles of the local rock shop "Tucson Mineral and Gem World", which has been open for about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the charts!? Even more significant is that Dud's results show the same thing:
Nb 101.88% Fe 35.76% Ta 15.4% Mn 2.54% Y 0.279%
Comparing these percentages to the relative heights of the same peaks on my scan one finds similar proportions.? It is very satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has a relatively high Y component.
Charles
|
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode. ?See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering. ?For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of Charles David Young
Sent: Saturday, January 18, 2020 5:45 AM
To: [email protected]; Mike Loughlin; Steve Dubyk; Frank Roberts
Subject: [XRF] XRF - comparing Dud's with my Si-PIN
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
|
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
Charles
On Sat, Jan 18, 2020 at 10:27 AM Dude < dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
|
Charles, do you have pure metal element samples?
I can give you titanium and a few more for use as calibration points.
By the way the needed correct 5V power supply arrived for my SI-PIN, so It will be back on lone withing a short time.
Geo
----- Original Message ----- From: Charles David Young <charlesdavidyoung@...> To: [email protected]Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk <sdubyk@...>, Frank Roberts <froberts@...> Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST) Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
Charles
On Sat, Jan 18, 2020 at 10:27 AM Dude < dfemer@...> wrote: Charles, Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.? Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter. Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review Dud ? ? ? Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly: Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing: Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's. Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
|
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
Geo
----- Original Message ----- From: Charles David Young <charlesdavidyoung@...> To: [email protected]Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk <sdubyk@...>, Frank Roberts <froberts@...> Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST) Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
Charles
On Sat, Jan 18, 2020 at 10:27 AM Dude < dfemer@...> wrote: Charles, Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.? Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter. Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review Dud ? ? ? Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly: Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing: Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's. Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
|
Charles,
I have pretty much everything in terms of base metals and REE
oxides as well as base and REE metals. ?Some more dangerous than others.
I did not get the close up Ti email that Geo got. Send me a copy
?
Dud
?
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of GEOelectronics@...
Sent: Saturday, January 18, 2020 12:16 PM
To: [email protected]
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
?
Charles,
do you have pure metal element samples?
I
can give you titanium and a few more for use as calibration points.
By
the way the needed correct 5V power supply arrived for my SI-PIN, so It will be
back on lone withing a short time.
-----
Original Message -----
From: Charles David Young <charlesdavidyoung@...>
To: [email protected]
Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk
<sdubyk@...>, Frank Roberts <froberts@...>
Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST)
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So
qualitatively the results from both XRFs seem to be similar.? I will be
looking at other specimens that you analyzed as well to see if they jive with
my setup.
So
what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the
TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On
Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based
on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For
samples
with percent grade material they would be shot with the Mining Plus mode.?
See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal
as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals
are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built
using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would
have
to normalize it to the Am-241 signal.? The system needs to have a fixed
ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks
that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative
gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It
has
a relatively high Y component.
?
?
|
Geo,
I would venture to say precise quantitative analysis is beyond the
reach of the professionals as well. For a simple matrix it works fine but for a
complicated matrix like rocks and junk it doesn¡¯t and varies wildly.? A ?SEM Micro
probe properly calibrated is great, but a wide beam on a target leaves a lot to
be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software,
some good, some bad and none doing everything well and all are complicated to
set up. ?Believe nothing, verify everything and use it as an anomaly finder,
then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal. ?Soils has 3 beams with
filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is
called Geochem and you don¡¯t need to switch between the modes. It¡¯s expensive
though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike
gamma spec where the target is the source. In the XRF world we provide the
source which is going to be very different between everyone¡¯s setup, beam size,
?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system
I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will
have better resolution and no LN2 issues
Dud
?
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of GEOelectronics@...
Sent: Saturday, January 18, 2020 12:30 PM
To: [email protected]
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
?
"Based
on the reported percentages the assay you are looking at was done using the
Soils mode which is used when looking at PPM level concentrations. The Soils
mode will well over report percent level concentrations? For samples with
percent grade material they would be shot with the Mining Plus mode."
I
agree that precise quantitative analysis is beyond the reach of our amateur
efforts at the moment. My first question is what is the difference between
SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is
it a scanning technique or is it all in software?
The
price of FP software has declined by about 50% already, but that took some
years to happen. Meantime I think we can all agree that direct comparison to a
known assayed mineral is a legitimate goal, as long as the sensors are the same
type?
That's
how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by
direct comparison with calibrated samples.
On
another thread, has anyone else anywhere else reported getting one of these
DP-5 processors up and running? It seems that maybe 20 or so of them have
actually been sold, maybe a few more.
-----
Original Message -----
From: Charles David Young <charlesdavidyoung@...>
To: [email protected]
Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk
<sdubyk@...>, Frank Roberts <froberts@...>
Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST)
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So
qualitatively the results from both XRFs seem to be similar.? I will be
looking at other specimens that you analyzed as well to see if they jive with
my setup.
So
what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the
TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On
Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based
on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level
concentrations.
The Soils mode will well over report percent level concentrations? For
samples
with percent grade material they would be shot with the Mining Plus mode.?
See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal
as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals
are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built
using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would
have
to normalize it to the Am-241 signal.? The system needs to have a fixed
ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks
that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative
gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It
has
a relatively high Y component.
?
?
|
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet you can see how xenotime is shot through the zircon; these minerals are chemically
different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite has been replaced by xenotime at a late stage, hence an odd altered white thorite
that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district, and these types of alteration and exsolution examples are common in many rare mineral
districts.
Steve
toggle quoted message
Show quoted text
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine but for a complicated matrix like rocks
and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good, some bad and none doing everything well
and all are complicated to set up. ?Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal. ?Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is called Geochem and you don¡¯t need
to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide the source which is going to be very different
between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will have better resolution and no
LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used when
looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I should
ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate goal,
as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
----- Original Message -----
From: Charles David Young <charlesdavidyoung@...>
To: [email protected]
Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk <sdubyk@...>, Frank Roberts <froberts@...>
Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST)
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in
the xenotime. ?Are there other zircons in the area that still hold the
HREE without any exsolution and at what concentration? ?How many episodes of
mineralization have these materials been through or is it all one unaltered
emplacement?
Dud
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of WILLIAM S Dubyk
Sent: Saturday, January 18, 2020 3:48 PM
To: [email protected]
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
?
Here
are a couple of examples of how messy these minerals can be. The zircon from
the Coats is a variety called cyrtolite. If you look at the photos at the
bottom of the sheet you can see how xenotime is shot through the zircon; these
minerals are chemically different but structurally the same. As the zircon
crystallized, the xenotime exsolved. Next example is a very strange powdery
looking thorite. The normally dark glassy thorite has been replaced by xenotime
at a late stage, hence an odd altered white thorite that is about 15% REE. Note
how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late
stage xenotime alteration of minerals is very common in the district, and these
types of alteration and exsolution examples are common in many rare mineral
districts.
Geo,
I would venture to say precise quantitative analysis is beyond the
reach of the professionals as well. For a simple matrix it works fine but for a
complicated matrix like rocks and junk it doesn¡¯t and varies
wildly.? A ?SEM Micro probe properly calibrated is great, but a wide
beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences.
Then there¡¯s the software, some good, some bad and none doing everything
well and all are complicated to set up. ?Believe nothing, verify
everything and use it as an anomaly finder, then send it out for a real ICP/MS
analysis.
Soils is FP the other is a element cal. ?Soils has 3 beams
with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new
software is called Geochem and you don¡¯t need to switch between the
modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m
looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike
gamma spec where the target is the source. In the XRF world we provide the
source which is going to be very different between everyone¡¯s setup, beam
size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus
system I¡¯d like to get running with it. I can use the LE ?HPGe
detectors but a CdTe will have better resolution and no LN2 issues
Dud
?
?
"Based
on the reported percentages the assay you are looking at was done using the
Soils mode which is used when looking at PPM level concentrations. The Soils
mode will well over report percent level concentrations? For samples with
percent grade material they would be shot with the Mining Plus mode."
I
agree that precise quantitative analysis is beyond the reach of our amateur
efforts at the moment. My first question is what is the difference between
SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is
it a scanning technique or is it all in software?
The
price of FP software has declined by about 50% already, but that took some
years to happen. Meantime I think we can all agree that direct comparison to a
known assayed mineral is a legitimate goal, as long as the sensors are the same
type?
That's
how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by
direct comparison with calibrated samples.
On
another thread, has anyone else anywhere else reported getting one of these
DP-5 processors up and running? It seems that maybe 20 or so of them have
actually been sold, maybe a few more.
-----
Original Message -----
From: Charles David Young <charlesdavidyoung@...>
To: [email protected]
Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk <sdubyk@...>,
Frank Roberts <froberts@...>
Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST)
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So
qualitatively the results from both XRFs seem to be similar.? I will be
looking at other specimens that you analyzed as well to see if they jive with
my setup.
So
what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the
TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On
Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level
concentrations.
The Soils mode will well over report percent level concentrations? For
samples
with percent grade material they would be shot with the Mining Plus mode.?
See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very
good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built
using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would
have
to normalize it to the Am-241 signal.? The system needs to have a fixed
ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks
that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud
was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan)
the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to
the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from
Little Patsy.? It has
a relatively high Y component.
?
?
|
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in
the xenotime.? Are there other zircons in the area that still hold the
HREE without any exsolution and at what concentration?? How many episodes of
mineralization have these materials been through or is it all one unaltered
emplacement?
Dud
?
Here
are a couple of examples of how messy these minerals can be. The zircon from
the Coats is a variety called cyrtolite. If you look at the photos at the
bottom of the sheet you can see how xenotime is shot through the zircon; these
minerals are chemically different but structurally the same. As the zircon
crystallized, the xenotime exsolved. Next example is a very strange powdery
looking thorite. The normally dark glassy thorite has been replaced by xenotime
at a late stage, hence an odd altered white thorite that is about 15% REE. Note
how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late
stage xenotime alteration of minerals is very common in the district, and these
types of alteration and exsolution examples are common in many rare mineral
districts.
Geo,
I would venture to say precise quantitative analysis is beyond the
reach of the professionals as well. For a simple matrix it works fine but for a
complicated matrix like rocks and junk it doesn¡¯t and varies
wildly.? A ?SEM Micro probe properly calibrated is great, but a wide
beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences.
Then there¡¯s the software, some good, some bad and none doing everything
well and all are complicated to set up.? Believe nothing, verify
everything and use it as an anomaly finder, then send it out for a real ICP/MS
analysis.
Soils is FP the other is a element cal.? Soils has 3 beams
with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new
software is called Geochem and you don¡¯t need to switch between the
modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m
looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike
gamma spec where the target is the source. In the XRF world we provide the
source which is going to be very different between everyone¡¯s setup, beam
size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus
system I¡¯d like to get running with it. I can use the LE ?HPGe
detectors but a CdTe will have better resolution and no LN2 issues
Dud
?
?
"Based
on the reported percentages the assay you are looking at was done using the
Soils mode which is used when looking at PPM level concentrations. The Soils
mode will well over report percent level concentrations? For samples with
percent grade material they would be shot with the Mining Plus mode."
I
agree that precise quantitative analysis is beyond the reach of our amateur
efforts at the moment. My first question is what is the difference between
SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is
it a scanning technique or is it all in software?
The
price of FP software has declined by about 50% already, but that took some
years to happen. Meantime I think we can all agree that direct comparison to a
known assayed mineral is a legitimate goal, as long as the sensors are the same
type?
That's
how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by
direct comparison with calibrated samples.
On
another thread, has anyone else anywhere else reported getting one of these
DP-5 processors up and running? It seems that maybe 20 or so of them have
actually been sold, maybe a few more.
So
qualitatively the results from both XRFs seem to be similar.? I will be
looking at other specimens that you analyzed as well to see if they jive with
my setup.
So
what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the
TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On
Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level
concentrations.
The Soils mode will well over report percent level concentrations? For
samples
with percent grade material they would be shot with the Mining Plus mode.?
See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very
good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built
using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would
have
to normalize it to the Am-241 signal.? The system needs to have a fixed
ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks
that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud
was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan)
the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to
the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from
Little Patsy.? It has
a relatively high Y component.
?
?
|
Hey Dud, it is due to the ionic radius and crystal structure that this occurs. Monazite is monoclinic and incorporates the LHEE, xenotime is tetragonal and incorporates the HREE; both are phosphates. Zircon is a silicate and does not incorporate HREE
as much. There are at least two generations of zircon at Petaca, and the first is poorer in REE. Pegmatites crystallize very quickly, with late stage fluids corroding the first crystallized minerals. Real common in these systems.?
Steve
toggle quoted message
Show quoted text
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime. ?Are there other zircons in the area that still hold the HREE without any exsolution and at what concentration?
?How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet you can see how xenotime
is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite has been replaced by
xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district, and these types of
alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine but for a complicated matrix like rocks
and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good, some bad and none doing everything well
and all are complicated to set up. ?Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal. ?Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is called Geochem and you don¡¯t need
to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide the source which is going to be very different
between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will have better resolution and no
LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used when
looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I should
ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate goal,
as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
----- Original Message -----
From: Charles David Young <charlesdavidyoung@...>
To: [email protected]
Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk <sdubyk@...>, Frank Roberts <froberts@...>
Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST)
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
Thanks Steve. That slide and your explanation ?is most
useful.
Dud
?
toggle quoted message
Show quoted text
From: [email protected]
[mailto:[email protected]] On Behalf Of WILLIAM S Dubyk
Sent: Monday, January 20, 2020 6:02 PM
To: [email protected]
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
?
Hey
Dud, it is due to the ionic radius and crystal structure that this occurs.
Monazite is monoclinic and incorporates the LHEE, xenotime is tetragonal and
incorporates the HREE; both are phosphates. Zircon is a silicate and does not
incorporate HREE as much. There are at least two generations of zircon at
Petaca, and the first is poorer in REE. Pegmatites crystallize very quickly,
with late stage fluids corroding the first crystallized minerals. Real common
in these systems.?
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in
the xenotime. ?Are there other zircons in the area that still hold the
HREE without any exsolution and at what concentration? ?How many episodes
of mineralization have these materials been through or is it all one unaltered
emplacement?
Dud
?
Here
are a couple of examples of how messy these minerals can be. The zircon from
the Coats is a variety called cyrtolite. If you look at the photos at the
bottom of the sheet you can see how xenotime is shot through the zircon; these
minerals are chemically different but structurally the same. As the zircon
crystallized, the xenotime exsolved. Next example is a very strange powdery
looking thorite. The normally dark glassy thorite has been replaced by xenotime
at a late stage, hence an odd altered white thorite that is about 15% REE. Note
how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late
stage xenotime alteration of minerals is very common in the district, and these
types of alteration and exsolution examples are common in many rare mineral
districts.
Geo,
I would venture to say precise quantitative analysis is beyond the
reach of the professionals as well. For a simple matrix it works fine but for a
complicated matrix like rocks and junk it doesn¡¯t and varies
wildly.? A ?SEM Micro probe properly calibrated is great, but a wide
beam on a target leaves a lot to be desired as you¡¯ll get a mix of
interferences. Then there¡¯s the software, some good, some bad and none
doing everything well and all are complicated to set up. ?Believe nothing,
verify everything and use it as an anomaly finder, then send it out for a real
ICP/MS analysis.
Soils is FP the other is a element cal. ?Soils has 3 beams
with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new
software is called Geochem and you don¡¯t need to switch between the
modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m
looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike
gamma spec where the target is the source. In the XRF world we provide the
source which is going to be very different between everyone¡¯s setup, beam
size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus
system I¡¯d like to get running with it. I can use the LE ?HPGe
detectors but a CdTe will have better resolution and no LN2 issues
Dud
?
?
"Based
on the reported percentages the assay you are looking at was done using the
Soils mode which is used when looking at PPM level concentrations. The Soils
mode will well over report percent level concentrations? For samples with
percent grade material they would be shot with the Mining Plus mode."
I
agree that precise quantitative analysis is beyond the reach of our amateur
efforts at the moment. My first question is what is the difference between
SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is
it a scanning technique or is it all in software?
The
price of FP software has declined by about 50% already, but that took some
years to happen. Meantime I think we can all agree that direct comparison to a
known assayed mineral is a legitimate goal, as long as the sensors are the same
type?
That's
how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by
direct comparison with calibrated samples.
On
another thread, has anyone else anywhere else reported getting one of these
DP-5 processors up and running? It seems that maybe 20 or so of them have actually
been sold, maybe a few more.
-----
Original Message -----
From: Charles David Young <charlesdavidyoung@...>
To: [email protected]
Cc: Mike Loughlin <loughlin3@...>, Steve Dubyk
<sdubyk@...>, Frank Roberts <froberts@...>
Sent: Sat, 18 Jan 2020 13:21:26 -0500 (EST)
Subject: Re: [XRF] XRF - comparing Dud's with my Si-PIN
So
qualitatively the results from both XRFs seem to be similar.? I will be
looking at other specimens that you analyzed as well to see if they jive with
my setup.
So
what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the
TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On
Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level
concentrations.
The Soils mode will well over report percent level concentrations? For
samples
with percent grade material they would be shot with the Mining Plus mode.?
See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very
good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built
using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would
have
to normalize it to the Am-241 signal.? The system needs to have a fixed
ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks
that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud
was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue
scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to
the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite
from Little Patsy.? It has
a relatively high Y component.
?
?
|
I'm sure they are Charles, for sure the ones from the South Platte peg area, those are probably intergrown with thorite, both are tetragonal and exsolve after a certain point. But it is hard to find detailed analyses of those. I suspect the Llano cyrtolites
are like that too, maybe Frank has some good analyses of those. Cyrtolites are formed during a late stage so they likely are all pretty messy, a conglomeration of several rare minerals.?
Steve
toggle quoted message
Show quoted text
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime.? Are there other zircons in the area that still hold the HREE without any exsolution
and at what concentration?? How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet
you can see how xenotime is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite
has been replaced by xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district,
and these types of alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine
but for a complicated matrix like rocks and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good,
some bad and none doing everything well and all are complicated to set up.? Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal.? Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software
is called Geochem and you don¡¯t need to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide
the source which is going to be very different between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors
but a CdTe will have better resolution and no LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between
SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison
to a known assayed mineral is a legitimate goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have
actually been sold, maybe a few more.
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with
my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
So would I be able to see anything one way or another with my Si-PIN?
El El lun, ene. 20, 2020 a la(s) 9:06 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
I'm sure they are Charles, for sure the ones from the South Platte peg area, those are probably intergrown with thorite, both are tetragonal and exsolve after a certain point. But it is hard to find detailed analyses of those. I suspect the Llano cyrtolites
are like that too, maybe Frank has some good analyses of those. Cyrtolites are formed during a late stage so they likely are all pretty messy, a conglomeration of several rare minerals.?
Steve
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime.? Are there other zircons in the area that still hold the HREE without any exsolution
and at what concentration?? How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet
you can see how xenotime is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite
has been replaced by xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district,
and these types of alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine
but for a complicated matrix like rocks and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good,
some bad and none doing everything well and all are complicated to set up.? Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal.? Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software
is called Geochem and you don¡¯t need to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide
the source which is going to be very different between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors
but a CdTe will have better resolution and no LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between
SOILS mode and Mining Mode? Or perhaps I should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison
to a known assayed mineral is a legitimate goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have
actually been sold, maybe a few more.
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with
my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
You would have to get the tightest beam focus possible then zap the mineral from a variety of angles. Try that and see what you get, that might be interesting.
toggle quoted message
Show quoted text
So would I be able to see anything one way or another with my Si-PIN?
El El lun, ene. 20, 2020 a la(s) 9:06 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
I'm sure they are Charles, for sure the ones from the South Platte peg area, those are probably intergrown with thorite, both are tetragonal and exsolve after a certain point. But it is hard to find detailed analyses of those. I suspect the Llano cyrtolites
are like that too, maybe Frank has some good analyses of those. Cyrtolites are formed during a late stage so they likely are all pretty messy, a conglomeration of several rare minerals.?
Steve
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime.? Are there other zircons in the area that still hold the HREE without any exsolution and at what concentration??
How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet you can see how
xenotime is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite has been replaced
by xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district, and these types
of alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine but for a complicated matrix like rocks
and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good, some bad and none doing everything well
and all are complicated to set up.? Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal.? Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is called Geochem and you don¡¯t need
to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide the source which is going to be very different
between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will have better resolution and
no LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used
when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I
should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate
goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
Well I don't have a beam, just Am241 buttons. So I guess that will have to wait until I get a real x-ray source.
El El lun, ene. 20, 2020 a la(s) 9:19 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
You would have to get the tightest beam focus possible then zap the mineral from a variety of angles. Try that and see what you get, that might be interesting.
So would I be able to see anything one way or another with my Si-PIN?
El El lun, ene. 20, 2020 a la(s) 9:06 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
I'm sure they are Charles, for sure the ones from the South Platte peg area, those are probably intergrown with thorite, both are tetragonal and exsolve after a certain point. But it is hard to find detailed analyses of those. I suspect the Llano cyrtolites
are like that too, maybe Frank has some good analyses of those. Cyrtolites are formed during a late stage so they likely are all pretty messy, a conglomeration of several rare minerals.?
Steve
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime.? Are there other zircons in the area that still hold the HREE without any exsolution and at what concentration??
How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet you can see how
xenotime is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite has been replaced
by xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district, and these types
of alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine but for a complicated matrix like rocks
and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good, some bad and none doing everything well
and all are complicated to set up.? Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal.? Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is called Geochem and you don¡¯t need
to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide the source which is going to be very different
between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will have better resolution and
no LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used
when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I
should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate
goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
If you can see the Zr with the Am buttons, you can always see the Th with gamma spec. That would help define the mineral assemblage. That in itself would be pretty neat! Can you identify Zr with your set up?
toggle quoted message
Show quoted text
Well I don't have a beam, just Am241 buttons. So I guess that will have to wait until I get a real x-ray source.
El El lun, ene. 20, 2020 a la(s) 9:19 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
You would have to get the tightest beam focus possible then zap the mineral from a variety of angles. Try that and see what you get, that might be interesting.
So would I be able to see anything one way or another with my Si-PIN?
El El lun, ene. 20, 2020 a la(s) 9:06 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
I'm sure they are Charles, for sure the ones from the South Platte peg area, those are probably intergrown with thorite, both are tetragonal and exsolve after a certain point. But it is hard to find detailed analyses of those. I suspect the Llano cyrtolites
are like that too, maybe Frank has some good analyses of those. Cyrtolites are formed during a late stage so they likely are all pretty messy, a conglomeration of several rare minerals.?
Steve
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime.? Are there other zircons in the area that still hold the HREE without any exsolution and at what concentration??
How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet you can see how
xenotime is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite has been replaced
by xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district, and these types
of alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine but for a complicated matrix like rocks
and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good, some bad and none doing everything well
and all are complicated to set up.? Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal.? Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is called Geochem and you don¡¯t need
to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide the source which is going to be very different
between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will have better resolution and
no LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used
when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I
should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate
goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
Charles, take a look at this. This was identified by xrf by Virgil Leuth at NM Tech; you could not have guessed this is zircon. Mel took this in because it was very odd looking. So xrf looks like a very useful tool.
I took the micro photo of the crystal terminations.
Steve
toggle quoted message
Show quoted text
Well I don't have a beam, just Am241 buttons. So I guess that will have to wait until I get a real x-ray source.
El El lun, ene. 20, 2020 a la(s) 9:19 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
You would have to get the tightest beam focus possible then zap the mineral from a variety of angles. Try that and see what you get, that might be interesting.
So would I be able to see anything one way or another with my Si-PIN?
El El lun, ene. 20, 2020 a la(s) 9:06 p.?m., WILLIAM S Dubyk < sdubyk@...> escribi¨®:
I'm sure they are Charles, for sure the ones from the South Platte peg area, those are probably intergrown with thorite, both are tetragonal and exsolve after a certain point. But it is hard to find detailed analyses of those. I suspect the Llano cyrtolites
are like that too, maybe Frank has some good analyses of those. Cyrtolites are formed during a late stage so they likely are all pretty messy, a conglomeration of several rare minerals.?
Steve
Steve, I don't understand everything you said but are any of my cyrtolites examples of this?
Charles
El El s¨¢b, ene. 18, 2020 a la(s) 10:09 p.?m., Dude < dfemer@...> escribi¨®:
Steve,
Zircon HREE¡¯s are interesting. Why do the heavies kick up in the xenotime.? Are there other zircons in the area that still hold the HREE without any exsolution and at what concentration??
How many episodes of mineralization have these materials been through or is it all one unaltered emplacement?
Dud
?
Here are a couple of examples of how messy these minerals can be. The zircon from the Coats is a variety called cyrtolite. If you look at the photos at the bottom of the sheet you can see how
xenotime is shot through the zircon; these minerals are chemically different but structurally the same. As the zircon crystallized, the xenotime exsolved. Next example is a very strange powdery looking thorite. The normally dark glassy thorite has been replaced
by xenotime at a late stage, hence an odd altered white thorite that is about 15% REE. Note how high Dy is in these analyses. Beam sizes varied from 3 to 20 microns. Late stage xenotime alteration of minerals is very common in the district, and these types
of alteration and exsolution examples are common in many rare mineral districts.
Geo,
I would venture to say precise quantitative analysis is beyond the reach of the professionals as well. For a simple matrix it works fine but for a complicated matrix like rocks
and junk it doesn¡¯t and varies wildly.? A ?SEM Micro probe properly calibrated is great, but a wide beam on a target leaves a lot to be desired as you¡¯ll get a mix of interferences. Then there¡¯s the software, some good, some bad and none doing everything well
and all are complicated to set up.? Believe nothing, verify everything and use it as an anomaly finder, then send it out for a real ICP/MS analysis.
Soils is FP the other is a element cal.? Soils has 3 beams with filters at 50, 40 and 10 Kv Mining uses 50 and 10 Kv beams. The new software is called Geochem and you don¡¯t need
to switch between the modes. It¡¯s expensive though.? I mostly use soils mode as I¡¯m looking for ppm anomalies.
?
A direct inter-comparison is not going to work very well unlike gamma spec where the target is the source. In the XRF world we provide the source which is going to be very different
between everyone¡¯s setup, beam size, ?intensity and to a lesser extent detector characteristics.
?
I¡¯m still looking for a CdTe rig . I have a 60kV microfocus system I¡¯d like to get running with it. I can use the LE ?HPGe detectors but a CdTe will have better resolution and
no LN2 issues
Dud
?
?
"Based on the reported percentages the assay you are looking at was done using the Soils mode which is used
when looking at PPM level concentrations. The Soils mode will well over report percent level concentrations? For samples with percent grade material they would be shot with the Mining Plus mode."
I agree that precise quantitative analysis is beyond the reach of our amateur efforts at the moment. My first question is what is the difference between SOILS mode and Mining Mode? Or perhaps I
should ask WHERE is the difference? Is it a scanning technique or is it all in software?
The price of FP software has declined by about 50% already, but that took some years to happen. Meantime I think we can all agree that direct comparison to a known assayed mineral is a legitimate
goal, as long as the sensors are the same type?
That's how I do uranium ore %, Cs-137 in soil,? and radium sample estimate , by direct comparison with calibrated samples.
On another thread, has anyone else anywhere else reported getting one of these DP-5 processors up and running? It seems that maybe 20 or so of them have actually been sold, maybe a few more.
So qualitatively the results from both XRFs seem to be similar.? I will be looking at other specimens that you analyzed as well to see if they jive with my setup.
So what about the
FeSi escape vs TiKa1 issue?? Attached is a closeup of that region for the TMG columbite (was samarskite Tub #2).? It sure seems like the
TiKa1 lines up better and it would be consistent with the Ti that you detected.
On Sat, Jan 18, 2020 at 10:27 AM Dude <dfemer@...> wrote:
Charles,
Based on the reported percentages the assay you are looking at was
done using the Soils mode which is used when looking at PPM level concentrations.
The Soils mode will well over report percent level concentrations? For samples
with percent grade material they would be shot with the Mining Plus mode.? See
if we shot that sample with both modes and use the Mining Plus mode to try a
cross calibrate with.? It¡¯s doubtful that you¡¯ll get a very good cross cal as its
dependent on the element , matrix effects and detector efficiency over energy
as well as detector / x-ray geometry and x-ray flux.?
Cals are done using the Fundamental Parameter approach or by
running each element using a calibration curve.? A cal curve can be built using
an element oxide and blending it down to PPM levels. That curve is still
dependent on distance of the target to the detector / X-ray geometry (also air
temp and moisture content and air density if you want to get picky) but most
importantly is the x-ray flux and beam filtering.? For your case you would have
to normalize it to the Am-241 signal.? The system needs to have a fixed ROI for
each element and you would use the baseline corrected area in that ROI as the
measured parameter. Complicating that are matrix effects where any other peaks that
are intruding in that ROI must be de-convolved from it and the area corrected
for it. A Gaussian correction could be used The most important thing you need
to do is get a handle on the detector/X-ray/ target geometry. It must be the
same all the time. The beam angle to the target and the detector should be
changed from your dual illumination set up to a single beam path and keep the
angle around 30 deg or so to minimize backscatter.
Quantitative gamma ray spectroscopy is much simpler compared to
quantitative XRF and I never trust my XRF reported values without a great deal
of skepticism and review
Dud
?
?
?
Nearly 2 years ago (5/23/18) Dud was kind enough to shoot
some of my specimens with his XRF "gun", which has a pinpoint xray
source.? Several of my rocks were found in the rubble piles of the local
rock shop "Tucson Mineral and Gem World", which has been open for
about 60 years.? One piece is large and ugly:
Looking at my results (blue scan) the Nb peak is off the
charts!? Even more significant is that Dud's results show the same thing:
Comparing these percentages to the relative heights of the
same peaks on my scan one finds similar proportions.? It is very
satisfying to know that my results largely corroborate those of Dud's.
Also plotted is a samarskite from Little Patsy.? It has
a relatively high Y component.
?
?
|
