Monday, August 27, 2007

talc and the "weak fault problem"

While I was out with my Advanced Structure class, looking at fault and shear-zone rocks, there was a fascinating article published in Nature about the surprise finding of talc in the San Andreas fault zone.

The San Andreas both makes wonderful sense – it’s a strike-slip plate boundary, THE strike-slip plate boundary that everyone knows about – and it’s mysterious. A year or two ago, the Teaching Structural Geology mailing list put out a call for current enigmas in structure and tectonics, and right there at the top of the list is “the weak fault problem.” The San Andreas fault may be capable of a pretty impressive shake, but it’s not as tough as one might expect. For one thing, the principle stress directions around it are wrong. A strike slip fault should theoretically have its most and least compressive principle stresses at an angle to it, somewhere in the neighborhood of 30 and 60 degrees in an ideal world where all rocks break like in laboratory experiments and nothing ever rotates. But even in a non-ideal world, you shouldn’t get principle stresses parallel and perpendicular to a fault. Those imply that the fault is a free surface, that it has no strength, that it can deform continuously, like water or air. Water could solve the problem, but the evidence for the necessary fluid pressures just doesn’t seem to be there.

Now, a major fault zone like the San Andreas isn’t composed so much of rock as it is of rock that has been broken, fragmented, crushed, cataclasized. But even powdered rock has frictional strength – too much strength for the stress orientations along the San Andreas. Even serpentinite, California’s slippery state rock, found along the eastern side of the fault where it follows the old Coast Range ophiolite, is too strong to explain the fault’s behavior.

So they dug a hole into the San Andreas fault, just north of Parkfield. (As an aside: I think it’s wild that the long-expected Parkfield earthquake happened while they were starting to drill.) And they sent the cuttings to Diane Moore, who studies the metamorphism of the mafic and ultramafic rocks in California. And she found...

Ok, no, not that. But she found talc. Softest mineral known. And, better than that, a mineral whose behavior in deformation seems to fit some of the behavior of the San Andreas. Talc can deform constantly, but only when deformation is slow. Push it too hard, and it becomes stronger. And it’s a common alteration product of ultramafic rocks in continental crust (at least) – just add quartz to serpentine, and you’ve got talc. (In fact, this is why American baby powder makers have switched from talc to corn starch. Talc mines also frequently contain serpentine... and serpentine is one of the minerals that can form asbestos.)

There wasn’t much talc in the cuttings, and it doesn’t solve every problem with the behavior of the San Andreas, but the thought is intriguing. And... well, could there be talc in other faults through ultramafic rock? Are there subduction zones that have managed to incorporate hydrated mantle rock and then added quartz... and become too weak for magnitude 9 earthquakes? I don’t know. I don’t even know how such ideas could be tested, or how anyone could tell if a fault is soft and slippery at depth. (I guess that maybe GPS monitoring would be able to pinpoint faults that deform without major earthquakes. But how to tell if they’ve got talc along them... drilling holes into them all probably isn’t feasible.) But it’s a neat idea.

Reference: Moore, Diane E. and Rymer, Michael J., 2007, Talc-bearing serpentinite and the creeping section of the San Andreas fault: Nature, v. 488, p. 795-797. (It's behind a paywall, so I can't directly link to the paper.)


Covenant said...

Excellent post - I remember during my masters degree discussing with my advisor and thinking about deformation characteristics of rocks and minerals and the idea of rock strength being tied to earthquake intensity.

I wondering if you can observe this on a microscopic level. in thin section?

This could be an excellent Ph. D thesis idea.

Anonymous said...

I was hoping you'd post about this...I saw it in Nature but didn't have a chance to really get into it...thanks

Kim said...

Thanks, both of you.

I remember during my masters degree discussing with my advisor and thinking about deformation characteristics of rocks and minerals and the idea of rock strength being tied to earthquake intensity.

Well, I might put it a little bit differently. Earthquake magnitude is related to the amount of fault slip and the area of the fault that slips. So, yeah, if the fault rock is weak, and slips before much elastic strain can build up (at least in places), then either the amount of slip or the area of slip (or both) would presumably be smaller.

They did observe the talc at a microscopic level - there are backscatter electron images of the talc replacing the serpentine. A couple of them are really hard to make out - talc and serpentine are both Mg-Si-hydroxides, so they've got similar average atomic numbers and look similar on a backscatter image.

I don't know how easy it would be to look at thin sections of the rock. I don't know how powdery the rock itself is - it might be hard to prepare the sections. And then the talc grains are 50 microns across, so 0.05 mm... it's hard to work with stuff that's that fine in thin section.

I wonder whether talc experiences intracrystalline deformation at very low temperatures? That could explain why it's so slippery in hand samples. I don't know of anyone who has deformed it experimentally, but I don't do a very good job of keeping up on that literature. I guess the data must be out there, at least about its aggregate behavior - it's cited in the Nature paper. (Actually, it looks like Diane Moore did some of those experiments.)

Brian - unfortunately, I will always be really slow to comment on Nature papers. We get them a week late in our library, and then I have to physically walk over and photocopy the things. (We don't have an electronic subscription.) Science, at least, I get at home, but I've only seen a handful of straight geology papers in the year I've been subscribed. Lots of good climate change stuff, though.

Covenant said...


It been a long time since I have been active in geology. My profession is in environmental, health and safety......

I understand that talc is used as a filler in the polymer industry - at least in the 70's. Several papers from the 1970's discuss the deformation properties of polymers with talc as a filler. I would like to believe that these papers have deformation characteristics of talc and, indeed, different filler materials.

This abstract provides some information on the rheology of talc.

Anonymous said...

There are actually some problems with talc as the big weakener. For instance, the 2-3% of talc found in the cuttings is not enough to weaken an entire fault (however, as Diane Moore states, some talc may have been lost during recovery). Also, if the talc is mostly present on the foliation planes of the serpentinite, the dilatation angle of the foliation should be taken in to account (which is not initially present in experimental monomineralic shear zones). But still, it is an interesting idea to explain the weakness of the SAF. We will see when the samples of the creeping section are availabe.

Kim said...

Thanks, anonymous, for pointing out some of the problems.