Change in peak shape over time

[Brian Joy]

The plot below illustrates background radiation (counts per second per nanoamp) collected at the upper background offset of Si Ka measured using TAP (P-10 gas-flow counter) on either of two diopside standards, one natural (PSU Px-1) and one synthetic (crystalline).  The background offset in every case was located at an L-value 2.8 mm above that of the peak.  Accelerating potential in each case was 15 kV.  Each point on the plot represents an average of 9 or 10 measurements on the standard.

At accelerating potential = 15 kV, a large enough fraction of backscattered electrons has energies appropriate to ionize argon such that these ionizations can contribute significantly to measured background radiation.  The effect is especially pronounced at low Bragg angle, such as for Si Ka diffracted using TAP.  Using the plot below, I can state with some confidence that the period of failure of the static filter -- apparently due to cracking of a cold solder joint at the point of attachment of the power supply to the filter -- was 20 Oct 2014 to 17 Jan 2017.

I had my reservations about the modification to the objective lens cooling system.  It brought to mind an old adage, “If it ain’t broke, don’t try to fix it.”  I guess the solder joint might have failed at some point, but by then (Oct 2014) it had held for 3.5 years.

[Probeman]

The plot below illustrates background radiation (counts per second per nanoamp) collected at the upper background offset of Si Ka measured using TAP (P-10 gas-flow counter) on either of two diopside standards, one natural (PSU Px-1) and one synthetic (crystalline).  The background offset in every case was located at an L-value 2.8 mm above that of the peak.  Accelerating potential in each case was 15 kV.  Each point on the plot represents an average of 9 or 10 measurements on the standard.

This just goes to show that monitoring of standard intensities can provide the proverbial "smoking gun" for troubleshooting such issues. PFE users can utilize the Drift application for these purposes as described here:

http://probesoftware.com/smf/index.php?topic=575.msg3565#msg3565

[Brian Joy]

After speaking with the JEOL office in Peabody, Mass. and collecting some SE images with an “electron mirror,” I now have a plausible explanation for the deterioration in resolution of the three PET crystals at high Bragg angles:  the static filter is not functioning correctly, and this has probably been the case for some time.  Thus the diffracting crystals (and other parts of the spectrometers) have been bombarded on a daily basis with backscattered electrons.  This probably heated the PET crystals sufficiently that their surfaces have become damaged.  In looking at an “electron mirror” SE image that I collected in May, 2015 in order to document a problem with the BSE detector, I can clearly see the diffracting crystals.  Live and learn, I guess.

An update on this old thread:  As of this morning, JEOL has agreed to replace our three PET crystals (PETJ, PETL, and PETH) at no charge to us.  This is fantastic news and is very gracious of them.

[Changkun]

An update on this old thread:  As of this morning, JEOL has agreed to replace our three PET crystals (PETJ, PETL, and PETH) at no charge to us.  This is fantastic news and is very gracious of them.

We have the same issue on PETL and PETJ crystals in 8530F.
When the "electrostatic deflector" switch is toggled on/off, we can see the brightness change (but not so largely.)
Was there another reason to cause the peak shape change with PET crystals rather than static filter issue?
Could you let us know how and why JEOL factory decided to replace your PET crystals with new ones?

Thanks,
Changkun

[Probeman]

Are these peak shape changes due to a crystal cracking? 

We have seen these PET peak shape changes on our Cameca instrument as well, and they have a significantly larger radius of curvature than JEOL crystals, especially the JEOL "H" type crystals:

https://probesoftware.com/smf/index.php?topic=18.msg6516#msg6516

[Brian Joy]

An update on this old thread:  As of this morning, JEOL has agreed to replace our three PET crystals (PETJ, PETL, and PETH) at no charge to us.  This is fantastic news and is very gracious of them.

We have the same issue on PETL and PETJ crystals in 8530F.
When the "electrostatic deflector" switch is toggled on/off, we can see the brightness change (but not so largely.)
Was there another reason to cause the peak shape change with PET crystals rather than static filter issue?
Could you let us know how and why JEOL factory decided to replace your PET crystals with new ones?

Thanks,
Changkun

Hi Changkun,

Sorry, I didn't see your message earlier.  JEOL replaced our three PET crystals because an engineer accidentally disconnected the power supply to the static filter in October, 2014.  The repeated heating and cooling of the PETJ, PETL, and PETH crystals due to the periodic influx of backscattered electrons likely caused microscopic fractures to develop (due to the large coefficient of thermal expansion of PET).  The net effect was the development of a "parasitic peak" on the low-L side of peaks at high Bragg angle; this led to a decrease in resolution and sharp decrease in count rate at the main peak.  The effect was similar to what you show in your figures, yet it appears that your static filter is still working?  Have you checked with an "electron mirror"?

Brian

[Changkun]

Are these peak shape changes due to a crystal cracking? 

We have seen these PET peak shape changes on our Cameca instrument as well, and they have a significantly larger radius of curvature than JEOL crystals, especially the JEOL "H" type crystals:

https://probesoftware.com/smf/index.php?topic=18.msg6516#msg6516

In our case, PETH is Ok.
I don't know how to identify crystal cracks but we did not find any suspicious thing so far except this one.

[Changkun]

Hi Changkun,

Sorry, I didn't see your message earlier.  JEOL replaced our three PET crystals because an engineer accidentally disconnected the power supply to the static filter in October, 2014.  The repeated heating and cooling of the PETJ, PETL, and PETH crystals due to the periodic influx of backscattered electrons likely caused microscopic fractures to develop (due to the large coefficient of thermal expansion of PET).  The net effect was the development of a "parasitic peak" on the low-L side of peaks at high Bragg angle; this led to a decrease in resolution and sharp decrease in count rate at the main peak.  The effect was similar to what you show in your figures, yet it appears that your static filter is still working?  Have you checked with an "electron mirror"?

Brian

Thank you, Brian.
As you noted before, we tested SEI brightness change by Electrostatic Deflector On/Off.
Please take a look at the images. Does the change mean "static filter" working properly?
(BTW, what is static filter? Where is the option in FT to turn on and off?)


And we also see the weird peak shape of Mo with PET crystal, as seen in Si.


JEOL engineer will visit and see inside with mirror tomorrow.
Brian, could you share the mirror image you had taken before? It will be good to show the image to JEOL engineers.

Thanks,
Changkun

[Jacob]

Does anybody know why JEOL uses an electrostatic deflector rather than a more conventional passive magnetic electron trap?

[Brian Joy]

Hi Changkun,

Sorry, I didn't see your message earlier.  JEOL replaced our three PET crystals because an engineer accidentally disconnected the power supply to the static filter in October, 2014.  The repeated heating and cooling of the PETJ, PETL, and PETH crystals due to the periodic influx of backscattered electrons likely caused microscopic fractures to develop (due to the large coefficient of thermal expansion of PET).  The net effect was the development of a "parasitic peak" on the low-L side of peaks at high Bragg angle; this led to a decrease in resolution and sharp decrease in count rate at the main peak.  The effect was similar to what you show in your figures, yet it appears that your static filter is still working?  Have you checked with an "electron mirror"?

Brian

Thank you, Brian.
As you noted before, we tested SEI brightness change by Electrostatic Deflector On/Off.
Please take a look at the images. Does the change mean "static filter" working properly?
(BTW, what is static filter? Where is the option in FT to turn on and off?)


And we also see the weird peak shape of Mo with PET crystal, as seen in Si.


JEOL engineer will visit and see inside with mirror tomorrow.
Brian, could you share the mirror image you had taken before? It will be good to show the image to JEOL engineers.

Thanks,
Changkun

Hi Changkun,

Static filter = electrostatic deflector (four syllables versus eight).  Yes, that change in brightness suggests that it is working.  Also, the fact that PETH does not show the "parasitic peak" suggests that the static filter is not the problem.

If the static filter/electrostatic deflector is off or not working, then you will be able to see inside the spectrometers with the electron mirror.  Note that some of the diffracting crystals are visible in this image:



If the static filter is working properly, then it should appear as if a gray ring is present above the filter:

[Changkun]

Dear Brian,

Thank you so much!

We took electron mirror images when electrostatic deflector turning off and on.

Electrostatic Deflector OFF (You see L-type, normal, H-type crystals from 12 o'clock in clockwise direction.)

Electrostatic Deflector ON


So the problem of our PETJ and PETL crystals is not due to static filter, as Brian said.
Also as Brian noted earlier, the peak count is about 30% lower than that obtained from other lab.

Next try is to replace counters and get the Si spectra again.
=> As expected, this issue is not related to counter.

[Brian Joy]

Hi Changkun,

Have you made any progress in finding the cause of the apparent damage to your PETJ and PETL crystals?

Brian

[Changkun]

Hi Changkun,

Have you made any progress in finding the cause of the apparent damage to your PETJ and PETL crystals?

Brian

Dear Brian,

JEOL Korea engineer did not find any possible reason and he said he reported this issue to JEOL Japan factory.
Before JEOL Japan responds this issue, JEOL Korea will test whether the crystals themselves are originally fractured, by replacing current PETJ with another one they have.
If they do not answer me with proper explanation, I will inspect all PET crystals in Korea as many as I can, which may help us understand why this problem occurred.

Anyway, I will update this thread whenever something changes.

[Changkun]

We temporarily replaced PETJ crystal with used one just for checking malfunction of the current PETJ crystal.
The red line in the attached figure is obtained by current PETJ crystal whereas the blue line by used PETJ crystal brought by a JEOL Korea engineer.
Since the same counter was used for PETJ crystal comparison, JEOL Korea engineer concluded that there may be a "contamination" issue on the our current PETJ crystal.
I could not disagree so the crystal will be replaced with new one next year.

Just curiosity. Is there anyone who knows how to clean PET crystals?? Is it a stupid thing?

[Probeman]

We temporarily replaced PETJ crystal with used one just for checking malfunction of the current PETJ crystal.
The red line in the attached figure is obtained by current PETJ crystal whereas the blue line by used PETJ crystal brought by a JEOL Korea engineer.
Since the same counter was used for PETJ crystal comparison, JEOL Korea engineer concluded that there may be a "contamination" issue on the our current PETJ crystal.
I could not disagree so the crystal will be replaced with new one next year.

Just curiosity. Is there anyone who knows how to clean PET crystals?? Is it a stupid thing?

Looks like a cracked crystal to me (as opposed to dirty).  Sometimes these cracks are barely visible to the eye.

But I do remember a service engineer (Jim Niccolino in Florida) who told me how to "dip" my TAP crystals in distilled water to remove/dissolve away the damaged surface of these crystals (which are very water soluble).