OK, I'm starting what I hope will be a fun topic where we can challenge each other to try and figure out what we are seeing in an image or graph.
Please post your "weird science" images or plots!
Ok, I'll start with this recent image of mapping a glass melt inclusion in olivine about 120 um in size. Conditions were 15 keV, 30 nA and 2 sec (2000 ms) per pixel. 256 x 256 pixels.
(https://smf.probesoftware.com/oldpics/i41.tinypic.com/2cy4hs3.jpg)
It is quite reproducible and also seen to a much lesser degree in the other elements (e.g., K) as might be expected.
OK, here's another weird image- but this one I can explain (see attachment for a good hint).
(https://smf.probesoftware.com/oldpics/i44.tinypic.com/15s6icj.jpg)
Polyphase saturation intergrowth following rind crystallation that occurred after melt entrappment.
Ed,
Which image does your "explanation" refer to? ???
Meanwhile, George Morgan asks what was the maximum Na count rate on the x-ray map, so here is a map of the Na raw data intensities in cps (attached).
I also attach a BSE image of the same inclusion before the map acquisition. The circular feature in the inclusion is where the carbon coat was starting to lift off but it "subsided" before the map was acquired.
rgd kileaua melt inclusion
Must be beam damage and recovery, or beam damage depleting one area and enriching an adjacent area (alhtough never seen it laterally). Is the map one really slow map or a stack of intergrated maps? Don't suppose you can do TDI on your maps can you, that's probably asking to much. I surpose the only way to make sense of it is to try different step sizes or speeds and see if the pattern changes.
I was referring to the electron images in the 3:01 post
RE Kilauea Melt Inclusion
At first look I'd agree with Ben. What's your spot size for this image? My guess is that the green regions on the raw intensity map are the spot locations with Na-depletion and you're getting enrichment in the regions between the adjacent points.
Quote from: qEd on January 13, 2014, 07:08:40 AM
Polyphase saturation intergrowth following rind crystallation that occurred after melt entrappment.
Not that I understand anything about "polyphase saturation intergrowth..." but you'll notice that this is a secondary electron image (see the FEI databar at the bottom of the image where it says Sig SE). So that should be a good hint.
Quote from: srmulcahy on January 13, 2014, 11:23:09 AM
RE Kilauea Melt Inclusion
At first look I'd agree with Ben. What's your spot size for this image? My guess is that the green regions on the raw intensity map are the spot locations with Na-depletion and you're getting enrichment in the regions between the adjacent points.
Ok, except that the Na x-ray map is a 256 x 256 pixel stage scan so the analysis positions are much, much smaller than the artifacts and the stage was moving perfectly smoothly!
You can see the individual pixels if you click on the raw data x-ray map:
http://smf.probesoftware.com/index.php?action=dlattach;topic=144.0;attach=296;image
Quote from: Ben Buse on January 13, 2014, 09:38:16 AM
Don't suppose you can do TDI on your maps can you, that's probably asking to [sic] much...
Funny you should mention this, but I've got a very cool idea on how one *might* do this, that I hope to test soon.
Quote from: Probeman on January 10, 2014, 02:47:26 PM
Ok, I'll start with this recent image of mapping a glass melt inclusion in olivine about 120 um in size. Conditions were 15 keV, 30 nA and 2 sec (2000 ms) per pixel. 256 x 256 pixels.
(https://smf.probesoftware.com/oldpics/i41.tinypic.com/2cy4hs3.jpg)
It is quite reproducible and also seen to a much lesser degree in the other elements (e.g., K) as might be expected.
Can you compare closely to other elements? If this is in fact diffusion, then you should expect a complimentary pattern. Specifically, Na diffuses much more easily than many other elements, and you may have another cation diffusing oppositely to the Na to balance charge. If so, then the patterns for all elements should not be the same, but they should be complimentary.
If they are all the same, then I'm going to put my money on an instrumental fluctuation: like (to make stuff up) the emission from your electron gun is varying, or your detector amp has some noise on it. You maybe can check this by looking at the amplitude of the signal. I notice the "Wt%" of the Na is varying by about 1/2 Wt%. So, do other nearby elements also vary by about that? Compute the variance of the concentration as a function of element Z. Maybe you get an overvoltage curve....
Quote from: Zack Gainsforth on January 13, 2014, 02:04:55 PM
Can you compare closely to other elements? If this is in fact diffusion, then you should expect a complimentary pattern. Specifically, Na diffuses much more easily than many other elements, and you may have another cation diffusing oppositely to the Na to balance charge. If so, then the patterns for all elements should not be the same, but they should be complimentary.
If they are all the same, then I'm going to put my money on an instrumental fluctuation: like (to make stuff up) the emission from your electron gun is varying, or your detector amp has some noise on it. You maybe can check this by looking at the amplitude of the signal. I notice the "Wt%" of the Na is varying by about 1/2 Wt%. So, do other nearby elements also vary by about that? Compute the variance of the concentration as a function of element Z. Maybe you get an overvoltage curve....
At the risk of being wrong I'd have to say it's not an instrumental effect as the Na signal does not vary at all in the host phase (olivine)- though it is essentially zero.
Also I recall Si and Al do not seem to diffuse as much as Na and K under normal TDI measurements. Typically one will see TDI variations of Na and K several times larger than Si and Al. See here for examples:
http://smf.probesoftware.com/index.php?topic=116.msg461#msg461
On the other hand, maybe the images *are* consistent with that hypothesis- though wouldn't that mean that the ion migration is real, rather than an instrumental artifact? But then what causes the periodicity? Is it related to the size of the inclusion for some reason?
Of course, I've always assumed that it is the thermal heating of a glass which mobilizes the alkali ions (which is why we see Kearns' so called "incubation time" as seen here):
http://smf.probesoftware.com/index.php?topic=116.msg454#msg454
And subsequently those alkali ions are then attracted to the sub-surface (dynamic) primary electron charge, hence the ions are drawn *deeper* into the sample, therefore exhibiting greater absorption losses, while the Si and Al atoms are less mobile and only show a minor increase in intensity due to the reduced degree of absorption by Na and K?
But, here's some of the other elements plotted so you can see them yourself:
Quote from: Probeman on January 13, 2014, 01:10:10 PM
Quote from: Ben Buse on January 13, 2014, 09:38:16 AM
Don't suppose you can do TDI on your maps can you, that's probably asking to [sic] much...
Funny you should mention this, but I've got a very cool idea on how one *might* do this, that I hope to test soon.
Couldn't you do multiple passes over the same area similar to what many EDS softwares do? Would require very stable position though.
Quote from: Karsten Goemann on January 13, 2014, 10:44:51 PM
Quote from: Probeman on January 13, 2014, 01:10:10 PM
Quote from: Ben Buse on January 13, 2014, 09:38:16 AM
Don't suppose you can do TDI on your maps can you, that's probably asking to [sic] much...
Funny you should mention this, but I've got a very cool idea on how one *might* do this, that I hope to test soon.
Couldn't you do multiple passes over the same area similar to what many EDS softwares do? Would require very stable position though.
Maybe... stay tuned!
Quote from: Probeman on January 10, 2014, 02:47:26 PM
Ok, I'll start with this recent image of mapping a glass melt inclusion in olivine about 120 um in size. Conditions were 15 keV, 30 nA and 2 sec (2000 ms) per pixel. 256 x 256 pixels.
(https://smf.probesoftware.com/oldpics/i41.tinypic.com/2cy4hs3.jpg)
It is quite reproducible and also seen to a much lesser degree in the other elements (e.g., K) as might be expected.
When I zoom in on the Na x-ray map I count around 10 pixels between the "high points" in the map attached in this post:
http://smf.probesoftware.com/index.php?topic=144.msg595#msg595
Since the stage scan utilized 2 seconds per pixel that mean that the "frequency" of these "high points" is around 20 seconds.
Now, here is where it gets weird: has anyone else noticed an apparent "cyclic" appearance to their TDI plots that seems to be reproducible? I couldn't find a good example so I'd appreciate someone else posting a better example, but this "cyclic" appearance looks something like this plot here with 3 sigma statistics.
(https://smf.probesoftware.com/oldpics/i41.tinypic.com/2u4u8ly.jpg)
Edit by John: here's a better example of the sinusoidal variation in x-ray intensities:
http://smf.probesoftware.com/index.php?topic=40.msg4006#msg4006
Ok folks, what is this sample? (see attachment below).
A couple of hints: it is a pure element, it is in a vacuum chamber.
I thought red mercury was a myth ;)
Ha! :D
Nope, not mercury. Two more hints: The actual size is roughly fist sized. The color is from its temperature.
Let me help you with this: The isotope mass of this is 238.049553 u.
Nice
http://onlyhdwallpapers.com/high-definition-wallpaper/metal-plutonium-desktop-hd-wallpaper-331177/
Quote from: Philipp Poeml on February 11, 2014, 11:28:14 PM
Let me help you with this: The isotope mass of this is 238.049553 u.
And getting a little bit lighter all the time... ;D
Quote from: Gareth D Hatton on February 12, 2014, 12:56:26 AM
Nice
http://onlyhdwallpapers.com/high-definition-wallpaper/metal-plutonium-desktop-hd-wallpaper-331177/
I got my picture directly from Rollin at Los Alamos! I wonder where this one came from?
OK, here's a new one. What is this specimen?
See attached.
Quote from: Probeman on March 05, 2014, 10:36:06 AM
OK, here's a new one. What is this specimen? See attached.
Well it turns out that the customer had etched the surface with nitric acid, so as is often the case, if you see something weird- it probably is! ;)
A couple more weird ones (no acid etch this time!) are below... any guesses?
(https://smf.probesoftware.com/oldpics/i57.tinypic.com/5phlwn.jpg)
Hint: CL image...
(https://smf.probesoftware.com/oldpics/i62.tinypic.com/wtde9z.jpg)
Hint: BSE image...
These are from my old JEOL 6400 SEM taken many moons ago.
Occums razor might suggest that a very simple reason for the Na depletion matrix - has someone put down a matrix of analyses before you mapped it?
1. Reprep and map again.
2. If it is present, is it dose dependant? Kv or current?
re the fine structure, it could be a solid state eutectoid decomposition or a spinodal decomposition. Both are free energy minimisation reactions driven by local activity reductions. The spacing of the lamella is diffusion (in the solid state) controlled and would give some indication of the cooling rate through the reaction.
I would doubt it is a eutectic as the structure is so fine and would suggest a very high cooling rate as diffusion in liquids is so fast.
Quote from: Les Moore on November 11, 2014, 02:49:47 PM
Occums razor might suggest that a very simple reason for the Na depletion matrix - has someone put down a matrix of analyses before you mapped it?
1. Reprep and map again.
Hi Les,
I assume you are talking about the Na map of the melt inclusion at the top of the topic? If so, please use the "quote" button for clarity next time... :P
Your speculation is not unreasonable (that it was previously mapped), but no, it is a virgin material and the artifact is due to some sort of sub surface charging dynamics during the stage scan acquisition. It is also present (though less apparent), in the K image (and Si and Al). The material is essentially homogeneous. See Julie Chouinard's poster from this summer (attached below).
Quote from: Les Moore on November 11, 2014, 02:49:47 PM
2. If it is present, is it dose dependent? Kv or current?
We have not tried various conditions. If it is an ion migration artifact, and I'm pretty sure it is as these glasses show extreme time dependent intensity (TDI) effects, then beam current, size and even keV should affect the artifact "wavelength".
I have an idea for a TDI map acquisition, but so little time...
Whilst browsing the 'Standards' topic I noticed a mention of the Sulphide standards produced by Gerald Czamanske. This reminded me of some images I took a while back, and the story that went with them: I rescued two 1"dia. standard mounts from an old desiccator minutes before they were thrown in a dumpster. After doing some investigation, they turned out to be a set of the Czamanske sulphides. I put them in the SEM to check out what I had. After a quick glance at low mag I was cursing myself for not doing a better job of 'dusting-off' the surface (there appeared to be a large clump of fluff on the surface of one of the standards). I zoomed-in to find a spot to take a quick EDS measurement and this is what I found:
(https://smf.probesoftware.com/oldpics/i58.tinypic.com/x59pif.jpg) (https://smf.probesoftware.com/oldpics/i59.tinypic.com/babu0.jpg)
This is an image of the surface of the pure Ag standard. As far as I can tell, the crystals growing on the surface are all the same phase: Ag2S.
I remember being told in a geochemistry class that this is why there is always a bone spoon in a silver cutlery set!
Trivia on the silver sulphide:
This effect causes major problems on silver connectors in industrial (SO2 rich environments).
I have also heard that the telephone exchange at Rotaurua in NZ has a double air environment.
It is also of great use in sepia toning of B&W prints.
The latter is employed to see the S distribution in steel.... an unexposed (or developed for that matter) piece of acid soaked photographic paper is placed on a ground/polished steel sample. Any S rich inclusions or regions create H2S and these react with the Ag in the paper to leave brown spots. It's called a sulphur print.
This is a beam damage issue related I think to the beam damage issues in SiO2 as described here:
http://smf.probesoftware.com/index.php?topic=418.msg2269#msg2269
But this is zircon, a much more robust material I think everyone will agree. And the data was acquired at 100nA but with a 5 um defocussed beam. To me it looks like a 5 um carbon ring, with a smaller central area that appears more damaged. So what I'm wondering is whether the damage pattern seen here:
(https://smf.probesoftware.com/oldpics/i67.tinypic.com/2gvtzjo.jpg)
represents an uneven distribution of electrons in the defocussed beam or if the central area that appears damaged is simply an effect of the heat being concentrated in the center, since the outer regions are also being heated by the defocussed beam so the center area should get the hottest...
I guess my question is: has anyone attempted to profile the electron distribution in a defocussed beam spot? In other words, is the electron density distribution profile across the defocussed beam like A or B:
(https://smf.probesoftware.com/oldpics/i63.tinypic.com/15qcl6s.jpg)
I assume we all hope it is like A, but I suspect it is a lot more like B...
Quote from: Probeman on November 04, 2015, 03:27:54 PM
This is a beam damage issue related I think to the beam damage issues in SiO2 as described here:
http://smf.probesoftware.com/index.php?topic=418.msg2269#msg2269
But this is zircon, a much more robust material I think everyone will agree. And the data was acquired at 100nA but with a 5 um defocussed beam. To me it looks like a 5 um carbon ring, with a smaller central area that appears more damaged. So what I'm wondering is whether the damage pattern seen here:
(https://smf.probesoftware.com/oldpics/i67.tinypic.com/2gvtzjo.jpg)
represents an uneven distribution of electrons in the defocussed beam or if the central area that appears damaged is simply an effect of the heat being concentrated in the center, since the outer regions are also being heated by the defocussed beam so the center area should get the hottest...
I guess my question is: has anyone attempted to profile the electron distribution in a defocussed beam spot? In other words, is the electron density distribution profile across the defocussed beam like A or B:
(https://smf.probesoftware.com/oldpics/i63.tinypic.com/15qcl6s.jpg)
I assume we all hope it is like A, but I suspect it is a lot more like B...
I recently ran across this issue again on a very beam sensitive sample. This was a thin film of ZnSn oxide on a polyester material that showed significant damage in the center of the beam spot, even when the beam was defocused to over 20 microns. Even when the beam current was reduced to 15 nA!
I ended up having to use a scanning mode at 10,000x to avoid damaging the sample. The difference was that the beam distribution seems much more uniform with the scanning beam than the defocused beam.
I wonder if anyone has tried to measure the beam flux profile on our EPMA instruments... perhaps using an imaging CCD array as a sample? Or would that just cook the CCD?
Quote from: Probeman on August 27, 2016, 09:20:41 AM
I recently ran across this issue again on a very beam sensitive sample. This was a thin film of ZnSn oxide on a polyester material that showed significant damage in the center of the beam spot, even when the beam was defocused to over 20 microns. Even when the beam current was reduced to 15 nA!
I ended up having to use a scanning mode at 10,000x to avoid damaging the sample. The difference was that the beam distribution seems much more uniform with the scanning beam than the defocused beam.
I wonder if anyone has tried to measure the beam flux profile on our EPMA instruments... perhaps using an imaging CCD array as a sample? Or would that just cook the CCD?
I just chatted with Ed Vicenzi and he pointed out that by looking carefully at a defocused electron beam on a fluorescent sample, we should be able to see if the electron flux is uniform over the defocused area... and now that I think of it, this is what I think we in fact see.
So that might mean that the damage in the center of the defocused beam area must be due to differential heating of the inner area by the outer area. That is consistent with seeing less sample damage when we scan the beam over the same size area since we are distributing the heating over the scan area...
He also mentioned that any finite element modeler should be able to calculate the temperature distribution in a material given the electron flux, area and the thermal conductivity of the material. Anyone out there interested?
Is the image an otolith?
Quote from: Nick Bulloss on August 28, 2016, 12:25:55 PM
Is the image an otolith?
Hi Nick,
How's it going?
Which image are you asking about? This one?
http://smf.probesoftware.com/index.php?topic=144.msg871#msg871
john
Quote from: Probeman on August 28, 2016, 12:50:31 PM
Quote from: Nick Bulloss on August 28, 2016, 12:25:55 PM
Is the image an otolith?
Hi Nick,
How's it going?
Which image are you asking about? This one?
http://smf.probesoftware.com/index.php?topic=144.msg871#msg871
john
Hi John,
I was looking at the image in the third post on page one, now I look more at the attachment I don't think it is an otolith.
Cheers,
Nick
Quote from: Nick Bulloss on August 29, 2016, 04:16:55 PM
Hi John,
I was looking at the image in the third post on page one, now I look more at the attachment I don't think it is an otolith.
Cheers,
Nick
Ah, OK.
That is a SIMS etch pit in a zircon grain. The area around the hole is coated with gold, but the etch pit is uncoated and somehow creates a charging cavity of some type.
The thing that is really weird is that this charging artifact is quite stable and persistent.
john
Regarding the odd spots.
The C rich contamination spot deposited at high currents on a steel sample is 'as expected' nicely Gaussian.
The C rich contamination spot on a refractory mineral under the same conditions is a donut.
Well, a Gaussian donut made by spinning the distribution about one tail on the sample.
I have some maps showing this behaviour somewhere (a long time ago).
The only idea that makes sense is that the sample is becoming so hot that it stops contaminating right under the beam but deposits on the cooler regions immediately adjacent. The high conductivity of metals does not allow sufficient elevation in temp for this to occur.
Cool.
Quite what this would do to the C yield is a worry as the C is deposited right where you don't want it re secondary fluorescence.
I just finished re-reading Victor's Weisskopf's most excellent book "The Privilege of Being a Physicist" (1990):
https://www.amazon.com/Privilege-Being-Physicist-Victor-Weisskopf/dp/0716721066
And this isn't really an "Explain This if You Can" but it does make one wonder what explanation Victor Weisskopf could have for leaving out energies of 10^4 eV in his "Quantum Ladder":
(https://smf.probesoftware.com/gallery/395_25_08_19_2_52_07.jpeg)
This is after all, the realm of microanalysis! He even says in the text that atomic physics deals with energies up to several thousand eV, but then jumps to 10^5 eV, calling that the beginning of the nuclear realm. He must have heard of inner shell ionizations!
Here's a fun science trivia quiz question: there are three places in the periodic table where the atomic number increases, yet the atomic weight decreases. Can you name them without looking at the table?
But the more interesting question is: why does this occur in those three places? Explain this if you can! :)
Quote from: Probeman on December 19, 2019, 10:45:48 AM
Here's a fun science trivia quiz question: there are three places in the periodic table where the atomic number increases, yet the atomic weight decreases. Can you name them without looking at the table?
But the more interesting question is: why does this occur in those three places? Explain this if you can! :)
OK, I guess you all better check the periodic table if you don't have it memorized! ;)
The element pairs with an increasing atomic number and a decreasing atomic weight are:
Ar-K
Co-Ni
Te-I
Did anyone know this bit of science trivia? In fact the Te-I pair is historically interesting because Mendeleyev, who arranged his periodic table by atomic *weight* (because the concept of atomic number had not yet been invented!), made a not often remembered (and wrong) prediction as seen here in a slide I prepared some years ago:
(https://smf.probesoftware.com/gallery/395_23_12_19_12_55_36.png)
But now the question is why does this decrease in atomic weight occur in these three places only in the periodic table? Seriously. I'm asking for help here!
The atomic weight is the sum of the nucleons minus the amount of 'mass' that's been taken up as binding energy in the nucleus (it's that e=mc2 equation) so I presume there must be a larger relative increase in binding energy at these points.
Quote from: Mike Matthews on December 25, 2019, 05:21:55 AM
The atomic weight is the sum of the nucleons minus the amount of 'mass' that's been taken up as binding energy in the nucleus (it's that e=mc2 equation) so I presume there must be a larger relative increase in binding energy at these points.
Hi Mike,
That's what I thought also. But I wrote to my two "card carrying" physicist friends, Andrew Westphal and Zack Gainsforth at Berkeley, and asked about these binding energies, and Zack responded that the binding energies are not that different at all, but that is has more to do with the "magic" shells number, or as Andrew subsequently confirmed:
QuoteZack is right, it's because of the nuclear closed shells for both protons and neutrons, at the numbers that he gave: 2, 8, 20, 28, 56, 82, 126, ...
This why there are s-process abundance peaks at 56Ni (which decays to 56Fe), Ba, Pb, and r-process peaks at Sn and Pt (for these there is pile-up at Z=56 and 82, then the beta-decay after the r-process is over shifts the peak down by a few elements.
He also adds:
QuoteIn the case of K and Ar, it's not because of binding energy, but just because 40Ar is the most abundant isotope of Ar in the Earth's atmosphere and 39K is the most abundant isotope of K. The reversal is not universal, it's just because most of the Ar in the atmosphere comes from the decay of 40K. If you used the average atomic weight of the elements based on the *cosmic* abundances, e.g., in interstellar gas, you would not see a reversal -- that is, the *cosmic* atomic weight of Ar is close to 36, not 40. So the sequence starting from S would be 32, ~35.4, 36, 39, ...
So the atomic weight is not a universal number, but varies according to the isotopic composition of the material that you're considering.
I have to say, I'm still not sure I understand completely, but I'm thinking about it.
Explain this if you can!
Recently I made some simultaneous k-ratio measurements on our PET crystals at high sin theta using F Ka on CaF2 (natural) and BaF2 (synthetic) in order to look at the effective takeoff angles as described here:
https://smf.probesoftware.com/index.php?topic=1569.msg12191#msg12191
But I found that were some significant peak shape issues that fortunately could be corrected for quite well using the area peak factor (APF) method, as described here:
https://smf.probesoftware.com/index.php?topic=536.msg12187#msg12187
What I did not mention was that there were also some significant time dependent intensity (TDI) issues during the measurement that also needed to be corrected for, even with a 20 um diameter beam and at 30 nA!
Here is what spectrometer 1 (TAP) intensities looked like:
(https://smf.probesoftware.com/gallery/395_24_11_23_10_34_17.png)
And spec 2 (LTAP):
(https://smf.probesoftware.com/gallery/395_24_11_23_10_34_35.png)
And spec 4 (TAP):
(https://smf.probesoftware.com/gallery/395_24_11_23_10_34_53.png)
You have probably noticed that that the TDI effects were largest in spec 1, less in spec 2 and essentially nonexistent in spec 4... now why would that be? Since all three spectrometers were measuring the same points at the the same time!?
And these effects were repeated in both BaF2 and CaF2. Here are the (unaggregated) statistics for BaF2:
St 835 Set 2 BaF2 (barium fluoride), Results in Elemental Weight Percents
ELEM: F F F F Ba
TYPE: ANAL ANAL ANAL ANAL SPEC
BGDS: LIN LIN LIN EDS
TIME: 100.00 100.00 100.00 80.00 ---
BEAM: 29.95 29.95 29.95 29.95 ---
ELEM: F F F F Ba SUM
XRAY: (ka) (ka) (ka) (ka) ()
215 22.674 22.297 21.931 23.055 78.330 168.286
216 23.205 22.544 22.081 24.294 78.330 170.454
217 22.329 22.308 22.248 24.241 78.330 169.456
218 22.605 22.285 22.165 21.838 78.330 167.223
219 22.816 22.459 22.380 21.856 78.330 167.841
220 22.167 22.280 22.000 21.100 78.330 165.877
221 23.259 22.438 22.447 21.617 78.330 168.092
222 23.287 22.043 22.171 25.752 78.330 171.583
AVER: 22.793 22.332 22.178 22.969 78.330 168.601
SDEV: .429 .152 .177 1.646 .000 1.825
SERR: .152 .054 .063 .582 .000
%RSD: 1.88 .68 .80 7.17 .00
PUBL: 21.670 21.670 21.670 21.670 78.330 100.000
%VAR: 5.18 3.05 2.34 5.99 .00
DIFF: 1.123 .662 .508 1.299 .000
STDS: 831 831 831 831 ---
STKF: .1545 .1545 .1545 .1545 ---
STCT: 15.90 47.06 20.35 18.22 ---
UNKF: .1959 .1919 .1906 .1974 ---
UNCT: 20.16 58.48 25.11 23.29 ---
UNBG: .25 .67 .31 .00 ---
ZCOR: 1.1636 1.1636 1.1636 1.1636 ---
KRAW: 1.2682 1.2426 1.2340 1.2781 ---
PKBG: 82.44 88.08 83.15 .00 ---
TDI%: 9.584 1.675 .140 .000 ---
DEV%: .4 .2 .2 .0 ---
TDIF: HYP-EXP HYP-EXP LOG-LIN ---- ---
TDIT: 130.13 131.75 132.38 .00 ---
TDII: 24.3 72.1 32.3 ---- ---
TDIL: 3.19 4.28 3.47 ---- ---
Note that the TDI effects are *not* correlated with intensity!
And here (again unaggregated) for CaF2:
St 831 Set 4 Fluorite U.C. #20011, Results in Elemental Weight Percents
ELEM: F F F F Ca
TYPE: ANAL ANAL ANAL ANAL SPEC
BGDS: LIN LIN LIN EDS
TIME: 100.00 100.00 100.00 80.00 ---
BEAM: 29.95 29.95 29.95 29.95 ---
ELEM: F F F F Ca SUM
XRAY: (ka) (ka) (ka) (ka) ()
223 33.512 32.461 32.533 32.362 51.200 182.067
224 31.706 32.161 32.409 32.734 51.200 180.209
225 31.520 32.164 32.536 32.728 51.200 180.149
226 31.602 32.846 32.228 32.132 51.200 180.008
227 32.700 32.635 32.333 32.447 51.200 181.314
228 32.328 33.019 32.625 32.045 51.200 181.217
229 33.551 32.394 32.476 32.702 51.200 182.323
230 32.759 32.057 32.537 32.546 51.200 181.100
AVER: 32.460 32.467 32.460 32.462 51.200 181.048
SDEV: .815 .345 .129 .268 .000 .875
SERR: .288 .122 .046 .095 .000
%RSD: 2.51 1.06 .40 .83 .00
PUBL: 48.800 48.800 48.800 48.800 51.200 100.000
%VAR: (-33.48)(-33.47)(-33.48)(-33.48) .00
DIFF: (-16.34)(-16.33)(-16.34)(-16.34) .000
STDS: 831 831 831 831 ---
STKF: .1545 .1545 .1545 .1545 ---
STCT: 15.87 47.06 20.31 18.20 ---
UNKF: .1544 .1544 .1544 .1544 ---
UNCT: 15.87 47.06 20.31 18.20 ---
UNBG: .13 .34 .16 .00 ---
ZCOR: 2.1022 2.1022 2.1022 2.1022 ---
KRAW: .9998 1.0000 .9998 .9998 ---
PKBG: 121.53 141.66 125.91 .00 ---
TDI%: 11.326 6.243 4.077 .000 ---
DEV%: .6 .2 .4 .0 ---
TDIF: HYP-EXP HYP-EXP LOG-LIN ---- ---
TDIT: 128.63 130.00 132.50 .00 ---
TDII: 15.9 47.4 20.5 ---- ---
TDIL: 2.77 3.86 3.02 ---- ---
Again, very similar trends in the CaF2 and also uncorrelated with intensity...
So I have a hypothesis, but I'm asking seriously: can anyone explain these observations?
Quote from: Probeman on November 24, 2023, 10:43:10 AM
Explain this if you can!
Recently I made some simultaneous k-ratio measurements on our PET crystals at high sin theta using F Ka on CaF2 (natural) and BaF2 (synthetic) in order to look at the effective takeoff angles as described here:
https://smf.probesoftware.com/index.php?topic=1569.msg12191#msg12191
But I found that were some significant peak shape issues that fortunately could be corrected for quite well using the area peak factor (APF) method, as described here:
https://smf.probesoftware.com/index.php?topic=536.msg12187#msg12187
What I did not mention was that there were also some significant time dependent intensity (TDI) issues during the measurement that also needed to be corrected for, even with a 20 um diameter beam and at 30 nA!
Here is what spectrometer 1 (TAP) intensities looked like:
(https://smf.probesoftware.com/gallery/395_24_11_23_10_34_17.png)
And spec 2 (LTAP):
(https://smf.probesoftware.com/gallery/395_24_11_23_10_34_35.png)
And spec 4 (TAP):
(https://smf.probesoftware.com/gallery/395_24_11_23_10_34_53.png)
You have probably noticed that that the TDI effects were largest in spec 1, less in spec 2 and essentially nonexistent in spec 4... now why would that be? Since all three spectrometers were measuring the same points at the the same time!?
...
So I have a hypothesis, but I'm asking seriously: can anyone explain these observations?
No speculations on what could be causing these different TDI trends on 3 simultaneous spectrometers...?
Quote from: Probeman on November 28, 2023, 12:38:21 PM
No speculations on what could be causing these different TDI trends on 3 simultaneous spectrometers...?
speculations... there is a bit too little information on other stuff to exclude the hardware effects. In case there is no hardware effect, I would speculate that maybe, maybe, this is crystallographic orientation thing. I cant remember now where I had read this, but as far I could remember it was somewhere shown that analyzing apatite-F at different orientations shows sever F loss where at other orientations shows nearly no F loss. I could get from that experiment the mind-shortcut that it is e-beam--crystallographic orientation controlled F loss. But Your experiment makes me start to think that maybe there is no loss at all (it was always for me illogical that negative (-1 valence) F ion would move toward negatively charged sample surface! There is probably only crystallographic repositioning of F, where for some spectrometer it gets hidden behind other atoms in lattice and thus absorption significantly goes up (decreasing intensity), while for different angle spectrometer they stay completely unhindered.
I think it would be easy to throw this hypothesis out the window If same behavior would be observed after redoing experiment with rotated sample.
Quote from: sem-geologist on November 28, 2023, 04:04:25 PM
Quote from: Probeman on November 28, 2023, 12:38:21 PM
No speculations on what could be causing these different TDI trends on 3 simultaneous spectrometers...?
https://smf.probesoftware.com/index.php?topic=144.msg12205#msg12205
speculations... there is a bit too little information on other stuff to exclude the hardware effects. In case there is no hardware effect, I would speculate that maybe, maybe, this is crystallographic orientation thing. I cant remember now where I had read this, but as far I could remember it was somewhere shown that analyzing apatite-F at different orientations shows sever F loss where at other orientations shows nearly no F loss. I could get from that experiment the mind-shortcut that it is e-beam--crystallographic orientation controlled F loss. But Your experiment makes me start to think that maybe there is no loss at all (it was always for me illogical that negative (-1 valence) F ion would move toward negatively charged sample surface! There is probably only crystallographic repositioning of F, where for some spectrometer it gets hidden behind other atoms in lattice and thus absorption significantly goes up (decreasing intensity), while for different angle spectrometer they stay completely unhindered.
I think it would be easy to throw this hypothesis out the window If same behavior would be observed after redoing experiment with rotated sample.
This is a reasonable explanation I think.
The only thing that bothers me with this idea is that this would assume that the orientations of both the BaF2 and the CaF2 would be similar, yes? Also I note that these are both cubic compounds, so not sure how orientation would affect ion migration... but it's an interesting idea.
Here is a drawing of the std mount with the positions of CaF2 and BaF2 outlined and the locations of the Cameca spectometers 1, 2 and 4 shown:
(https://smf.probesoftware.com/gallery/395_29_11_23_11_15_55.png)
The thing that makes a little more sense to me would be that perhaps the carbon coat connection to the mount might be only in one spot (towards spec 1?), so the sub surface negative charge would flow towards that point, and therefore "push" the F- ions away from spec 1, causing the large intensity drop over time, whereas spec 4 might see the least reduction in F- ions because it was getting more F- ions "pushed" towards it over time?
How does that sound as an explanation?
easy to check (both hypothesis), as it is 1 inch round mount. I suggest to rotate it about 45 degrees and retry the experiment. If result will be the same – then our both hypotheses are failed, and the reason is buried somewhere in hardware.
Yeah I thought of that, but I would have to ensure that conductive contact is made at specific points relative to the spectrometer orientation. I think it might be easier to try using a thin insulator on the top of the mount and then bridge that insulator in specific places using a small drop of carbon. Maybe this weekend...
This is a close up of the 4096 x 4096 quant x-ray maps that Radek acquired:
(https://smf.probesoftware.com/gallery/395_29_11_25_9_14_50.png)
https://smf.probesoftware.com/index.php?topic=1791.msg13743#msg13743
It's slightly amazing to me how one can obtain reasonable sensitivity at only 15 millisec per pixel at 100nA. Accuracy for major elements is maintained by using the logarithmic dead time correction which ensures accuracy up to 300 to 400 kcps.
The image reminds me a bit of van Gogh's "Starry Night"...
4K image eh, really nice.
You'll have to upload it to your home TV to see the full resolution. Not many computer monitors have 4K resolution and within a normal business, none.
The danger here is that smaller pixels accurately capture the interaction volume issues:
1. Are the turquoise rings another phase or just a representation of sharp transition smoothed out by electron volume effects. (MC X-Ray, DTSA etc)
2. Are the turquoise blobs yellow blobs just under the surface. (MC X-Ray, DTSA, Casino etc)
3. Are the edges due to a substrate fluoresence issue (Penepma)
4. More I haven't thought about....
So you map at lower and lower kV's and get better resolution but poorer S/N
Paradoxically, a mental image of the best spatial resolution equaling the electron beam spot size occurs right at the excitation voltage as there is no excitation from deeper in the sample; but you get almost no counts.
Quote from: Les Moore on December 17, 2025, 03:14:51 PMThe danger here is that smaller pixels accurately capture the interaction volume issues:
1. Are the turquoise rings another phase or just a representation of sharp transition smoothed out by electron volume effects. (MC X-Ray, DTSA etc)
2. Are the turquoise blobs yellow blobs just under the surface. (MC X-Ray, DTSA, Casino etc)
3. Are the edges due to a substrate fluoresence issue (Penepma)
4. More I haven't thought about....
So you map at lower and lower kV's and get better resolution but poorer S/N
Paradoxically, a mental image of the best spatial resolution equaling the electron beam spot size occurs right at the excitation voltage as there is no excitation from deeper in the sample; but you get almost no counts.
Sure, the same thing can happen at "normal" spatial resolution at high accelerating voltages. Interpretation of quantitative results in multi-phase materials is always a challenge.
As you point out, the secondary fluorescence effects from nearby phases is particularly challenging. Though we have implemented a boundary fluorescence correction in Probe for EPMA for point analyses as seen here:
https://smf.probesoftware.com/index.php?topic=1545.0
we have not tested it for x-ray maps though the same code should work for both points and pixels (in CalcImage). But either way, it's important to keep in mind that the location of the boundary relative to the points (or pixels), is critical for defining the SF boundary correction, as one must define one phase as the beam incident phase (the measured phase), and a single other phase as the boundary phase. See here for more discussion on this important aspect:
https://smf.probesoftware.com/index.php?topic=1545.msg12941#msg12941
Either way, it's a challenge not only because of the physics, but also the geometry, e.g., what is the phase boundary angle:
https://smf.probesoftware.com/index.php?topic=1545.msg13373#msg13373
and also from other artifacts such as Bragg defocus effects:
https://smf.probesoftware.com/index.php?topic=1545.msg13222#msg13222
In any case, the purpose of this 4K mapping attempt was simply to test the memory limitations of our mapping software and subsequent quantification, so hopefully there will be no attempt to interpret these quantitative maps.