Saturday, February 20, 2010

It’s rock, Jim, but not as we know it

I have been working with uranium, either in U/Pb geochronology (small scale) or in mineral exploration (large scale), for a long time now. So some years back, I was googling around for information on the Oklo deposits- an old, rich African uranium deposit that became critical and lost much of its 235U through hydrothermally moderated fission.

What I found instead was this. This is a blog that has nothing at all to do with uranium, but instead describes the modeling and characterization of extrasolar planets. I thought that was pretty neat, and when they started discussing the possibility of extrasolar moons, I did a simple mass calculation to show that the chemistry was potentially interesting. The result of that dabble can be found here.

Since then, I have been working on a more formal description of extrasolar petrology, with an eye towards figuring out what bulk compositional variations will produce non-peridotitic condensates. This has meant more recreational science and less blogging, and is reason number two (after the expanding earth) that this place has been light-on.

And, as a reader pointed out fairly recently, I am not alone in this endeavor. A paper by Bond et al. was recently lofted into the arXiv on the very same subject, but with a somewhat different focus. Since the same reader asked my opinion, I wanted to say this:

Congratulations, Dr. Bond.

However, my main interests are in an area not covered by her team. This is not surprising, since she is an astronomer and I’m a geochemist. My approach has been petrological, and has mainly focused on the potential for nebular condensation of water-friendly phases, particularly cordierite.

As it turns out, I’ve been planning on attending the Goldschmidt conference as an exhibitor anyway. Ideally I would talk about SHRIMP related stuff there, but the developmental project I’ve been working on is not mature enough to present, and I don’t have the diplomatic skills necessary to convince laser ICPMS people that it is in their interest to let me use their data to show how shonky their techniques are. So I’ve submitted an abstract on the petrology of stellar condensation instead.

Normally, I would submit such an abstract a few minutes before the deadline, but the other night Mrs. Lemming pointed out that I had better finish the abstract immediately, as tweaking a figure from the birth suite was not an option. So it’s in.

The text, written at 2 in the morning, may be completely unintelligible. Y’all can see for yourselves- I’ve copied and pasted it below. If there are any egregious errors, please point them out before Sunday night American time- as abstracts are editable up until then.

The Petrology of Extrasolar Chondrites

The cosmochemical principle that the bulk composition of chondritic meteorites is similar to the spectrally-derived composition of the sun is extended to other stars. Extrasolar chondritic mineralogy is first derived via simple four-component CMAS normative calculations for 476 stellar chemical compositions from a variety of sources.
The most common exochondritic mineral assemblage in the CMAS system is the solar one: anorthite (an) - fosterite (fo) - enstatite (en) – diopside (di). It is closely followed by an-fo-en-cordierite (cd). Other assemblages combine for less than 10% of stellar compositions, and include such assemblages as: An-en-di-quartz (qz), an-en-qz-cd, an-fo-cd-spinel (sp), and fo-sp-various silica-poor Ca/Mg phases.
Equilibrium condensation calculations for major elements show that stars with higher C/O or Si/O ratios than solar can have insufficient H2O in the nebular gas to oxidize Mg, Si, Al, Ca during silicate condensation. Reduction of the nebular gas via silicate condensation leads to lower condensation temperatures for silicates, sulfide, silicide and nitride condensates, and CO breakdown into graphite and/or methane. Below 800K, this last reaction destabilizes the reduced condensates.
The percentage of stars with non-solar condensation sequences in a given survey ranges from fewer than 5% to more than 70%, depending on the study. Studies of stars from different research groups often lie in different mineralogical fields, and are not always within error of each other.

Figure 1: Equal-O projection of stellar compositions from anorthite into (truncated) fosterite-quartz-diopside and fosterite-quartz-spinel ternary diagrams. Circles: stars with planets; squares: stars without planets. Grayscaled by study.

(since this is the internet, grayscale is colored)


Lockwood said...

OK, I admit to not understanding every word, but that is pretty fascinating. Maybe I'm missing the obvious, but where would the earth fall on the diagram?

Kim said...

That is really cool.

(Oh, and congratulations on the expanding earth, too!)

Weekend_Viking said...

Nice abstract. Been a while since I did that sort of thing. Mostly smashing rocks in engineering geo stuff now. (Zane)

C W Magee said...

Oops. Blue diamonds are Bulk Silicate Earth and Sun, respectively.

Chris R said...

This is great! Different stellar chemistry producing different planetary mineral phases is both completely obvious and something I'd never really considered before. It will be interesting to see how this intersects with the results from Kepler...