Clay analysis

[BenjaminWade]

Hi all
I have been approached with an enquiry about EPMA of clay. Just wondering if anyone has performed any analysis of clay via EPMA, and if so any tips/tricks? From what I gather the customer has various smectite and kaolinite presented in an epoxy polished block (not exactly sure how well they are polished...but I think I can take a pretty good guess...). I am thinking defocussed beam and lower beam current, maybe 15kV/10nA? Any tips on how to deal with the variable OH-?

Cheers

[Malcolm Roberts]

Hi Ben
Good luck. Clays are messy things as you've gathered already. The definition of a clay is also a grain size thing, off the top of my head and don't quote me, probably <5 microns. So you're going to defocus a beam? Also, clays tend to be multilayered entities, making any analysis you get something of a bucket composition. I guess where I am going is that maybe EPMA on your samples might not be the best way to go? You might be better steering your customer towards a TEM and maybe even some raman work.
Other's however may know better than me..... But I seem to be the only one dumb enough to pipe up!!
Cheers,
malc.

[BenjaminWade]

Hi Malcolm
Yes you are correct, I am predicting a mess of platelets all smooshed together with epoxy mixed in between. They are "supposed" to be of all the same clay type in each sample but we will see. He wants to accurately measure cation exchange in some of the clays, so standardless EDS/TEM is probably out. He has done bulk chemical analysis and XPS, now he wants to use EPMA as his supervisor analysed clays many years ago using the same technique...
Sounds like I will just suck it and see how we go!

Cheers

[BenH]

Hi Ben.

I have had some luck doing analyses of fine particles embedded in epoxy.  The totals come out low due to the presence of the epoxy in the pores.  I simply set "CO2" by difference in the JEOL software as a proxy for the epoxy composition, then normalize to 100% .  The compositions that I get agree surprisingly well with bulk analyses done by ICPMS.  One thing to look out for is that the count rates will change as the beam "burns" the epoxy.  The count rates begin to stabilize after a couple of seconds so it might be best to count for a while.  Perhaps the TDI option will help.   I'm not sure how you would go about getting OH data. If there is epoxy you can't do it by difference because of the low totals. At the very least, you can get the major elements quite nicely.

Not sure what the required spatial resolution is but micro-focus XRF could be a nice option.

Ben Hanson

[jared.wesley.singer]

Ben,

Consider "firing" the clay (a.k.a. calcination) before the electron probe.  Do this in a TGA to get a very accurate determination of molecular water (<200C), organics ~250, various OH groups 250-500C, and carbonates >500C.  Then 'probe away on the anhydrous product, which will polish better and is more beam stable. 

Depending on the alkali content of your clay, it may be feasible to sinter or melt the TGA product.  Melting is advantageous if your "clay" turns out to be a heterogeneous mix of minerals.

Good luck,

Jared