Most efficient platform for standardized combined EDS and WDS

[orlandin]

Hello, all!

It is getting to the NSF MRI time of year here, and I'm starting to think that my dream analytical instrument is a sort of hybrid 'microprobe' and 'SEM' with rigorous high-volume >1,000ppm quant capability provided mostly by EDS, with 1-2 WDS detectors providing backup for the trace concentrations, plus variable pressure capability (not for quant, obviously) and and EBSD plus CL. Definitely an airlock and an in-column faraday cup. Probably should be a Schottky source. I'd prefer to use PFE for the collection and post-processing given the ease with which one can collect and organize huge amounts of data.

So, this hypothetical instrument has the stability and fine stage control of what would normally be called a 'microprobe', but I just don't think I need more than one or two WDS detectors to meet the needs of researchers now or in the future - standardized EDS is able to meet the vast majority of quant demands now. Plus, an instrument with EDS, WDS, EBSD, and CL would replace two aging instruments here, and I would rather have one overbooked instrument that everyone loves than two more specialized instruments that struggle to break even. What do you all think? Does anyone have such a versatile instrument? Can you think of why this would be stupid/impossible? What manufacturers would you recommend for the platform and detectors?

One of the biggest problems I can see is the sensitivity of WDS to surface Z position. Is the only way to solve this reliably with a reflected light scope? Why can't you use a capacitive position sensor like these: http://www.microsense.net/products-position-sensors-vacuum.htm

[Probeman]

Perfect microanalysis instrument for me would be:

1. JEOL FEG electron column

2. Cameca WDS spectrometers with linear optical encoding

3. Bruker or Thermo SDD EDS

4. Cameca stage with linear optical encoding

5. Cameca light optics

6. JEOL electronics

7. Polycold cryo-pumped vacuum baffle (100 Kelvin) with turbo pump

And of course Probe Software EPMA applications...   

[Jacob]

The only thing I believe would be impractical is the variable pressure.  That requires a differentially pumped space in the column as well as a pressure limiting aperture of some kind.  It wouldn't be compatible with the inline light optics.

Distance to the objective would definitely require an optical probe of some kind.  A contact sensor could only measure the stage position, not the actual sample height to the level of accuracy required.

[sem-geologist]

1. JEOL FEG electron column

What are advantages of JEOL FEG over CAMECA FEG? Is 3 year old FEG tip on JEOL able to keep beam stable within 0.2% of initial value (with clean apertures) within 24h+? Because it is daily possible with Cameca FEG, Also Cameca FEG control is not Locked out from power users, which allows all kind of clever manipulations of FEG (prolonging its lifetime, or temporary increasing the brightness), where JEOL AFA I heared Hides away any possibility to manipulate any FEG settings (maybe except on, off, standby).

6. JEOL electronics

What so special about that?
Stage with linear optical encoding (2.), how that will work without Cameca VME board?
Cameca WDS spectrometers, the same useless without Cameca electronics in VME board.
Cameca Light optics... these again is quite integrated with Cameca electronics, how without it?

Personally I find Cameca electronics logically laid out and not so hard to troubleshoot when things stops working as intended. I find VME as backbone for electronics to be best chosen platform. It is not a coincidence that VME is still used in mil/air/space and industrial env.
It also is quite future proof and I think there are possibilities to Hack-around and make own replacements or replace cards in the future for improved functionality of some parts. The only complaint about Cameca electronics from me is that Communication between PowerQuiccII and PC could be stepped-up to 100Mbit from currently used 10Mbit.

What so special about JEOL electronics?

[Nicholas Ritchie]

Alternatively, you might consider an SEM with a couple of EDS and a single WDS (like a Thermo Fisher MagnaRay or an Oxford AztecWave.)
This would open up options for VP, EBSD & CL.  I don't know how trace element sensitivity is with either of these WDS units but I'd suspect that it is at least an order-of-magnitude worse than a dedicated EPMA unit.  Maybe this is acceptable, maybe it isn't.
I'd be a little careful about over-spec'ing your detectors.  Each additional detector makes the instrument just that much more fragile and less usable for its primary purpose.  An instrument that does everything is rarely ideal for anything.
I find the X-Y-Z-R-T stage in my SEM to be frustrating.  It is very hard to maintain the beam perpendicular to the sample.  I end up using a construction laser vertical level (retro-reflected) to evaluate the sample tilt.  My optimal stage would be X-Y-Z like a microprobe but this wouldn't work for EBSD.  Like a microprobe, I focus using the Z stage and leave the objective current fixed.  The reproducibility is good enough for EDS and likely also one of the 90-degree rotated WDS units.
I would like an insertable faraday cup on my SEM but the stage mounted one works fine.   As for stability, my Schottky field emitter SEM is as current stable as you could ever want.

[Probeman]

Alternatively, you might consider an SEM with a couple of EDS and a single WDS (like a Thermo Fisher MagnaRay or an Oxford AztecWave.)
This would open up options for VP, EBSD & CL.  I don't know how trace element sensitivity is with either of these WDS units but I'd suspect that it is at least an order-of-magnitude worse than a dedicated EPMA unit.  Maybe this is acceptable, maybe it isn't.

Can one purchase an SEM that has multiple WDS spectrometers?

That way, one could aggregate the photons (a la Probe for EPMA) to obtain better trace sensitivity... for points:

https://probesoftware.com/smf/index.php?topic=1316.0

and quant x-ray maps:

https://probesoftware.com/smf/index.php?topic=41.msg2180#msg2180