EPMA and large area SDD detectors

[UofO EPMA Lab]

On our SEM we have an SDD EDS detector which is 30 sq. mm. On our EPMA we have a 10 sq. mm. The idea being that the SEM tends to operate at lower beam currents, while the EPMA tends to operate at higher beam currents. And we don't want to saturate the detector on our EPMA when running at high beam currents.

Though it's worth mentioning that we have modified the EDS aperture wheel on our Cameca SX100 to reduce the geometric efficiency of the 10 sq. mm detector for very high beam currents, as discussed here:

https://probesoftware.com/smf/index.php?topic=342.0

Though generally we leave it "wide open". It should also be mentioned that utilizing a smaller aperture works just fine when performing standards based EDS analysis, because both the standards and unknowns are measured using the same EDS aperture.

But here's my question: does anyone have a 30 sq. mm SDD detector on their EPMA instrument?  If so, at what beam currents (with a fully open aperture) does one see deadtimes increase beyond 60-80%?

That is, is it a good idea to install a 30 sq. mm SDD EDS detector on an EPMA instrument?

[Karsten Goemann]

We have a Thermo UltraDry Extreme 30mm2 on our 8530F plus. I think it's great to have. I was even considering 60mm2 but that's probably too large for our current setup. The 30mm2 gives us great EDS count rates when we use lowish beam currents like 10nA or less for beam sensitive materials. If the dead time gets too high at higher beam currents we can always switch to a smaller aperture.
Count rates obviously also depend a lot on the distance of the detector chip to the sample surface. Haven't tried to measure/compare that for different EPMA/EDS combinations.

[Probeman]

I contacted the guys at NIST and Dale Newbury sent me these documents which characterize their 8530 for use with the Bruker 30 sq. detector.  See attachments below for the Powerpoint and Excel documents.

Karsten: That is useful information, thank-you!  I didn't realize that the JEOL has a built in aperture for the EDS on their 8230/8530 instruments.  Is this EDS aperture also available on older JEOL instruments such as the 8200/8500?
john

[Probeman]

OK, interesting. I should have explained this is in reference to a Bruker detector going on a JEOL 8500.

So apparently at the moment Bruker doesn't have availability for the 10 sq. mm detectors, so Ben Hanson is considering getting the 30 sq. mm detector. But he was concerned about driving up his deadtime at high beam currents without an aperture.  He's got an aperture on the JEOL supplied detector, but since he's keeping his JEOL detector and putting the Bruker detector on the other port, we were looking at other solutions for EPMA instruments. For example retracting the detector position for less geometric efficiency.

[Karsten Goemann]

Our 8530F+ is the same as Owen described for the 8500. The 3-way aperture is part of the Thermo EDS.
A JEOL EDS I saw on a recently installed 8530F+ has a motorised aperture included (in the JEOL EDS port)
Regarding the Bruker on JEOL, CSIRO in Melbourne have 2 Bruker detectors each on their 8500 and 8530 (in the EDS and CL detector port) and they're retractable like the SEM versions to adjust count rates.

[Jacob]

He's got an aperture on the JEOL supplied detector, but since he's keeping his JEOL detector and putting the Bruker detector on the other port, we were looking at other solutions for EPMA instruments. For example retracting the detector position for less geometric efficiency.

The Bruker detectors have a motorized slide with configurable inserted and parked positions.  Setting the parked position to a distance chosen to optimize the count rate at high beam currents and the inserted position to one optimized for lower current would work pretty well.  Changing positions is just a button click away in the software.

[Probeman]

Karsten and Jacob: this is awesome information.  Thank-you so much! I've only had a Bruker Esprit SDD on my Cameca SX50 (now gone to that microprobe heaven in the sky), but it didn't have a motorized slide.

Any idea on what percent change in count rate is observed, between the fully retracted and fully extended positions, for the 30 sq. Bruker detector?
john

[Jacob]

Yeah, I have some old solid angle calculations.  On my XL-30, there's about 100mm of travel between the inserted and parked positions.  That gives a 93% drop in solid angle and count rate. 

[Probeman]

Hi Jacob,
Thank-you for this information.  Seems to me that should be enough to handle the high beam currents utilized on an EPMA.
john