Standards Which Should Be Developed For EPMA

[UofO EPMA Lab]

We probed the first trial of the Cs Zr phosphate on the weekend and it appears to have a different stoichiometry than anticipated.

wt%:
ELEM:       Cs      Zr       P      Hf      Rb      Si       O   SUM 
   116    20.9    30.1    16.1     .44    n.d.    .000  32.688   100.2
   117    20.8    29.5    15.9     .50    n.d.    .000  32.211    98.9
   118    20.9    29.8    16.2     .48    n.d.    .000  32.784   100.2
   119    20.8    30.5    15.7     .45    n.d.    .000  32.352    99.8
   120    20.9    30.2    16.0     .48    n.d.    .000  32.538    99.9
   121    21.0    29.8    16.2     .50    n.d.    .000  32.733   100.1
   122    21.1    30.1    15.9     .46    n.d.    .000  32.503   100.1

AVER:     20.9    30.0    16.0     .47    n.d.    .000  32.544   99.88
SDEV:     .116    .321    .183    .024    .008    .000    .209    .453


Best match in terms of stoichiometry appears to be CsZr2(PO4)3
formula units calculated on basis of 12 oxygens:
ELEM:       Cs      Zr       P      Hf      Rb      Si       O   SUM 
   116    .922    1.94    3.05    .014    n.d.    .000  12.000    17.9
   117    .935    1.93    3.06    .017    n.d.    .000  12.000    17.9
   118    .921    1.91    3.07    .016    n.d.    .000  12.000    17.9
   119    .928    1.98    3.02    .015    n.d.    .000  12.000    17.9
   120    .926    1.95    3.04    .016    n.d.    .000  12.000    17.9
   121    .925    1.91    3.07    .016    n.d.    .000  12.000    17.9
   122    .940    1.95    3.04    .015    n.d.    .000  12.000    17.9

AVER:     .928    1.94    3.05    .016    n.d.    .000  12.000   17.93
SDEV:     .007    .024    .020    .001    .001    .000    .000    .009


The stoichiometry doesn't work out perfectly, e.g. Cs should be 1 formula unit but is only 0.93. But we only had pollucite as Cs standard for calibration so if there's a problem with that standard this will also affect oxygen calculation by stoihiometry etc.

[Malcolm Roberts]

John
What is your preferred method to get a hundred bucks to you?
Cheers,
Malc.

[John Donovan]

John
What is your preferred method to get a hundred bucks to you?
Cheers,
Malc.

Hi Malc,
I don't want your hundred bucks because I don't have the RbTiOPO4, but Marc Schrier does:

http://calchemist.com/contact.htm

Please contact him directly.
john

[Malcolm Roberts]

Anyone have any thoughts on alternative for Br? Something other than TlBr? We have just been pondering a synthetic scapolite with Br only. However, the thoughts are that this would probably need to be the Na end-member and may be a bit on the beam sensitive side. Others?

[Probeman]

Anyone have any thoughts on alternative for Br? Something other than TlBr? We have just been pondering a synthetic scapolite with Br only. However, the thoughts are that this would probably need to be the Na end-member and may be a bit on the beam sensitive side. Others?

Malcolm,
I completely concur with you.   The community really needs some non water soluble beam stable halogen standards.  I personally am looking for an iodine standard.

Maybe Marc Schrier can think of some such halogen compounds that might be synthesized for us?
john

[Malcolm Roberts]

Hi John,
I've been working on the Br "problem" with specific reference to scapolite and am going to go the 25kV route and use the ka line. It looks really good. Our Astimex block has TlBr/TlI in it so we can manage both of these halogens, however, I'd much rather have something better matrix matched. If Marc can come up with something, it would be great!!
Cheers,
Malc.

[Marc Schrier]

John,
    In reply to your 8/10 post on the Cs/Zr/PO4 I prepared and sent up, sorry I've been such a flake; I've been swamped!  The hope was that I would prepare an analog to the KTiOPO4 family, so just like the RbTiOPO4, but CsZrOPO4.  I've not looked too hard, but it does not appear the phase has been prepared before.  The two know Cs/Zr/PO4 phases I see in JCPDS are CsZr2(PO4)3 and Cs2Zr(PO4)2.  We'd have to confirm by PXRD, but I think your probe analysis was correct, and the material is likely CsZr2(PO4)3.  How did the material stand up to the beam?  I'm pretty sure the sample I sent you was just powder, but if it was okay in the beam, then maybe that's just an easier target than going after an unknown phase.  Is the Cs content getting too low?  The Cs content in the other known phase, Cs2Zr(PO4)2, would be higher.  I need to get the papers and see if they have already outlined a clear synthesis for crystals.
-Marc

[Probeman]

John,
    In reply to your 8/10 post on the Cs/Zr/PO4 I prepared and sent up, sorry I've been such a flake; I've been swamped!  The hope was that I would prepare an analog to the KTiOPO4 family, so just like the RbTiOPO4, but CsZrOPO4.  I've not looked too hard, but it does not appear the phase has been prepared before.  The two know Cs/Zr/PO4 phases I see in JCPDS are CsZr2(PO4)3 and Cs2Zr(PO4)2.  We'd have to confirm by PXRD, but I think your probe analysis was correct, and the material is likely CsZr2(PO4)3.  How did the material stand up to the beam?  I'm pretty sure the sample I sent you was just powder, but if it was okay in the beam, then maybe that's just an easier target than going after an unknown phase.  Is the Cs content getting too low?  The Cs content in the other known phase, Cs2Zr(PO4)2, would be higher.  I need to get the papers and see if they have already outlined a clear synthesis for crystals.
-Marc

Hi Mark,
The material you sent, let's call it CsZr2(PO4)3, was small crystals and seems very stable under the beam, just very small crystals.

If it is not water soluble I'd say we have our candidate Cs standard for bulk synthesis! 

Do you have enough money around for starting materials, to try making some bigger crystals that I can better test for homogeneity, beam stability, etc.?

[Marc Schrier]

Malcolm and John,  I have crystals of a series of Ti/IO3 compounds I prepared as an undergraduate that might work as an iodine standard; they are water insoluble.  I'm not sure if they will stand up to the beam.  John, I can send you some samples if you want to take a peek.  There were a number of materials including H2Ti(IO3)6•2H2O, Ti(IO3)4, and TiO(IO3)2 and I should still have small samples of each.
-Marc

[Probeman]

Malcolm and John,  I have crystals of a series of Ti/IO3 compounds I prepared as an undergraduate that might work as an iodine standard; they are water insoluble.  I'm not sure if they will stand up to the beam.  John, I can send you some samples if you want to take a peek.  There were a number of materials including H2Ti(IO3)6•2H2O, Ti(IO3)4, and TiO(IO3)2 and I should still have small samples of each.
-Marc

The ones without water sound very interesting.   Please send them to my UofO address.  Thanks!
john

[Brian Joy]

Hi Mark,
The material you sent, let's call it CsZr2(PO4)3, was small crystals and seems very stable under the beam, just very small crystals.

If it is not water soluble I'd say we have our candidate Cs standard for bulk synthesis! 

Do you have enough money around for starting materials, to try making some bigger crystals that I can better test for homogeneity, beam stability, etc.?

But CsZr2(PO4)3 only contains about 23.5 wt% Cs2O.  Natural pollucites contain up to about 40 wt% Cs2O, and it would be nice to have a standard that could be used to analyze for Cs in them (other than another pollucite).

[Probeman]

Hi Mark,
The material you sent, let's call it CsZr2(PO4)3, was small crystals and seems very stable under the beam, just very small crystals.

If it is not water soluble I'd say we have our candidate Cs standard for bulk synthesis! 

Do you have enough money around for starting materials, to try making some bigger crystals that I can better test for homogeneity, beam stability, etc.?

But CsZr2(PO4)3 only contains about 23.5 wt% Cs2O.  Natural pollucites contain up to about 40 wt% Cs2O, and it would be nice to have a standard that could be used to analyze for Cs in them (other than another pollucite).

Hi Brian,
You are correct that we would have a raw k-ratio of somewhat under 2 for the matrix correction if we utilized this material as a Cs standard for a high Cs pollucite characterization (actually k-raw = 1.75 in the example below, which I feel is an extrapolation within reason).  I do not know how valuable a number the Cs content of a pollucite mineral is geologically and I defer to your expertise on this matter of course.

In any case this Cs standard was/is primarily intended for use as a trace element standard, and of course for this purpose, it almost doesn't matter what we use as a standard, so long as it is stable under the beam (and not water soluble would be nice).   

That said, we can examine the matrix correction magnitudes for these two materials and here is our proposed synthetic:

ELEMENT  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR STP-POW BKS-COR   F(x)u      Ec   Eo/Ec    MACs
   Cs la   .9985  1.0000  1.2158  1.2140  1.4193   .8567   .9064  5.0120  2.9928 385.778
   Zr la  1.0414   .9984  1.0941  1.1377  1.2284   .8907   .7609  2.2230  6.7476 964.694
   P  ka  1.2197   .9988   .9322  1.1357   .8687  1.0731   .7529  2.1460  6.9897 1000.49
   O  ka  3.4599   .9998   .8284  2.8655   .7452  1.1116   .2025   .5317 28.2114 7450.29

 ELEMENT   K-RAW K-VALUE ELEMWT% OXIDWT% ATOMIC% FORMULA KILOVOL                                       
   Cs la  .00000  .18238  22.142   -----   5.556    .167   15.00                                       
   Zr la  .00000  .26715  30.393   -----  11.111    .333   15.00                                       
   P  ka  .00000  .13630  15.480   -----  16.667    .500   15.00                                       
   O  ka  .00000  .11162  31.986   -----  66.667   2.000   15.00                                       
   TOTAL:                100.000   ----- 100.000   3.000

And here is a natural pollucite from Rossman with a high Cs concentration:

ELEMENT  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR STP-POW BKS-COR   F(x)u      Ec   Eo/Ec    MACs
   Cs la   .9648  1.0000  1.2478  1.2038  1.4785   .8439   .9381  5.0120  2.9928 250.373
   Si ka  1.5404   .9989   .9203  1.4161   .8678  1.0606   .5879  1.8390  8.1566 1974.48
   Al ka  1.7391   .9925   .9451  1.6313   .8841  1.0690   .5102  1.5600  9.6154 2557.00
   Na ka  2.8865   .9977   .9325  2.6853   .8611  1.0829   .2916  1.0730 13.9795 5371.51
   O  ka  2.0263   .9996   .8511  1.7239   .7787  1.0931   .3458   .5317 28.2114 4171.30

 ELEMENT   K-RAW K-VALUE ELEMWT% OXIDWT% ATOMIC% FORMULA KILOVOL                                       
   Cs la  .00000  .32090  38.630   -----   8.000    .167   15.00                                       
   Si ka  .00000  .11529  16.326   -----  16.000    .333   15.00                                       
   Al ka  .00000  .04807   7.842   -----   8.000    .167   15.00                                       
   Na ka  .00000  .02488   6.682   -----   8.000    .167   15.00                                       
   O  ka  .00000  .17534  30.227   -----  52.000   1.083   15.00                                       
   H                        .293   -----   8.001    .167
   TOTAL:                100.000   ----- 100.000   2.083


As you can see, the difference in the matrix corrections is under 1%, so I don't think it is a significant concern even for major element analytical work.  In fact the difference in the matrix corrections is probably smaller than the uncertainty in the Cs concentration in existing natural pollucite standards...  I'm guessing.

Of course there is also the statistics of the extrapolation, but a factor of less than 2 should still provide reasonable analytical sensitivity.

Please feel free to explain further if I have missed your point.
john

[Brian Joy]

In any case this Cs standard was/is primarily intended for use as a trace element standard, and of course for this purpose, it almost doesn't matter what we use as a standard, so long as it is stable under the beam (and not water soluble would be nice).   

Hi John,

As far as I know, this is the first you've mentioned that this material was meant for use primarily as a trace element standard.  I thought the idea was to produce standards that could be used for a variety of purposes?  Pollucite is a particularly important mineral to characterize accurately because it is the major commercial source of Cs, and its Cs2O content is quite variable.

I am never comfortable with compositions that are significantly extrapolated.  In the example you give, Cs La k-raw = 1.76, and this is excessive, especially for a major element.  Also, while the total matrix correction is less than 1% (assuming that it's accurate), the component Z and A corrections are larger (~3%).

In my opinion -- for what it's worth -- it seems like a good idea to explore other possibly feasible candidates, such as Cs2Zr(PO4)2, before settling on CsZr2(PO4)3 for mass production.  Respectfully, I think you're jumping the gun on this.

Brian

[John Donovan]

In any case this Cs standard was/is primarily intended for use as a trace element standard, and of course for this purpose, it almost doesn't matter what we use as a standard, so long as it is stable under the beam (and not water soluble would be nice).   

Hi John,

As far as I know, this is the first you've mentioned that this material was meant for use primarily as a trace element standard.  I thought the idea was to produce standards that could be used for a variety of purposes?  Pollucite is a particularly important mineral to characterize accurately because it is the major commercial source of Cs, and its Cs2O content is quite variable.

I am never comfortable with compositions that are significantly extrapolated.  In the example you give, Cs La k-raw = 1.76, and this is excessive, especially for a major element.  Also, while the total matrix correction is less than 1% (assuming that it's accurate), the component Z and A corrections are larger (~3%).

In my opinion -- for what it's worth -- it seems like a good idea to explore other possibly feasible candidates, such as Cs2Zr(PO4)2, before settling on CsZr2(PO4)3 for mass production.  Respectfully, I think you're jumping the gun on this.

Brian

Hi Brian,
I got you.  Well let's not argue about what constitutes "excessive extrapolation".

I really don't care what the Cs standard ends up being (it could even be more than one compound!), so long as it's stoichiometric, beam stable, and not water soluble.   

Marc was trying to grow CsZrOPO4 but got CsZr2(PO4)3 instead.  That means to me that CsZr2(PO4)3 is "easier" to synthesize than the other.  But what do I know.

Let's see what Marc Schrier says...

[Marc Schrier]

Hi Guys,  I've not had a chance to do anything more on the Cs standard front, so there is not much to report.  I will try to get the papers on the two Cs/Zr/PO4's, and I hope it sheds some light on the Cs:PO4 flux ratios and the product obtained.  CsZr2(PO4)3 is 22.1% Cs while Cs2Zr(PO4)2 is 48.6% Cs, so I'm liking the higher Cs content of the latter.  Antony sent me the paper for the CsAlTiO4 (48.9% Cs) and it's doable, but it sounds like it requires long run times at high temperatures in sealed platinum tubes.  I'll have to see if someone has come up with a flux route to CsAlTiO4 as that would be much more amenable to crystal growth.  -Marc