Quantifying lead in thorianite Th-U-Pb dating

[anenburg]

Hi all,

We recently measured a thorianite (ideally ThO₂) in an attempt to calculate an age for it. We measured everything that appeared on a WDS scan: Th, U, Pb, Ca, Nd, Ce and La. Totals were ~97 and FTIR showed OH so we conservatively guesstimated 1.5% H₂O, and 0.1% Pr₂O₃ from LAICPMS data.
Our age is pretty good and consistent with other ages of the same rocks from similar studies.

However, I am trying to eliminate systematic errors as much as possible. Several questions:

• I am getting results in wt%. But, my lead is a bit heavier than "normal" lead. It's almost purely ²⁰⁸Pb, and not much of all other lighter isotopes. Should this be taken into consideration somewhere?
• The lead was calculated as divalent (PbO) - but this could be underestimating the amount of oxygen. After all, it started its life as ThO₂, and the oxygen shouldn't have left the lattice upon decay. Maybe. Is there any argument to use PbO vs PbO₂? At ~6 wt% Pb this is a non-negligible difference, when propagated to the actual age.
• Using two different corrections in the JEOL stock software (Armstrong vs XPP) we get an age difference of ~1.5% (almost 20 million years!). One of them is more consistent with previous data,  but that's not a good rationale for choosing this correction over the other. Is there a better way to select one or the other?

[Probeman]

Hi all,

We recently measured a thorianite (ideally ThO₂) in an attempt to calculate an age for it. We measured everything that appeared on a WDS scan: Th, U, Pb, Ca, Nd, Ce and La. Totals were ~97 and FTIR showed OH so we conservatively guesstimated 1.5% H₂O, and 0.1% Pr₂O₃ from LAICPMS data.
Our age is pretty good and consistent with other ages of the same rocks from similar studies.

However, I am trying to eliminate systematic errors as much as possible. Several questions:

• I am getting results in wt%. But, my lead is a bit heavier than "normal" lead. It's almost purely ²⁰⁸Pb, and not much of all other lighter isotopes. Should this be taken into consideration somewhere?
• The lead was calculated as divalent (PbO) - but this could be underestimating the amount of oxygen. After all, it started its life as ThO₂, and the oxygen shouldn't have left the lattice upon decay. Maybe. Is there any argument to use PbO vs PbO₂? At ~6 wt% Pb this is a non-negligible difference, when propagated to the actual age.
• Using two different corrections in the JEOL stock software (Armstrong vs XPP) we get an age difference of ~1.5% (almost 20 million years!). One of them is more consistent with previous data,  but that's not a good rationale for choosing this correction over the other. Is there a better way to select one or the other?

Evaluating accuracy is always problematic.  That said, you might take a look at this topic where we discuss using CalcZAF to run all 10 matrix corrections methods to obtain a "distribution" of the results:

https://probesoftware.com/smf/index.php?topic=508.0

Of course a consensus isn't necessarily equivalent to absolute accuracy, but at least it's a start.  And then there are 6 different tables for mass absorption coefficients used by CalcZAF, so 10 * 6 = 60 results!  Also you didn't mention which emission line for Pb you are using, but if you are using the M line, the choice of matrix corrections is probably significant.

As for the atomic weight issue, that is something to consider for sure.  It's certainly true that mass does not significantly affect the generation, emission or absorption of x-rays. See this post and its attachments for more details:

https://probesoftware.com/smf/index.php?topic=92.msg3010#msg3010

But the calculation of relative weight concentrations will be affected by the values selected for atomic weight.  I would start by seeing how much of an effect the difference in atomic weights could have. I suspect this is only going to be a problem for major concentrations, as a few percent of a few percent isn't going to amount to all that much.

In CalcZAF one can easily edit the values in the Elements.dat file and try different atomic weights to see.

[anenburg]

That's interesting. I always knew that heavier elements absorb more, but that's only because they have more electrons (as I've discovered by the post you linked to).

Now the question is how do I find what's the mass they use for Pb in the JEOL software...

[Probeman]

I would assume they use the natural abundance isotopes (which does vary globally depending on the geological setting) atomic weight.

A quick search shows 207.20, 207.21 (latest IUPAC) and 207.19 (the default value in CalcZAF/PFE which can be edited).