2

Nature communications article "Mission to Earth's core — a modest proposal", suggests placing a large volume of liquid iron in a crack and let it sink all the way to the Earth's core, carrying along a probe that can transmit data using seismic waves.

It is an interesting idea. Although there might be many technical problems I feel no one seems insurmountable and it is worth a try.

However, I wonder why the author suggest using a liquid iron instead of a liquid lead. I would expect lead to be cheaper and also require a smaller volume to achieve the same fracture stresses (due to higher density). In addition, a lower melting point would facilitate the initiation of the fracture propagation near the surface.

Would lead dissolve in the surrounding magma? Would it chemically interact turning into a lighter compound?

As the original article is behind a paywall I copy here the main text:

enter image description here

enter image description here

Camilo Rada
  • 17,644
  • 6
  • 64
  • 113

1 Answers1

3

I've already commented on this before here. This will not work regardless of whether this is iron or lead or anything else. The fact that it was published in Nature does not mean it is true, always remember that.

A 108 kg ball of iron would have about 30 metres in diameter, which is not a lot, compared to Earth scales.

Here is why it will not work:

  1. We have very massive objects on the surface. They are called buildings. They don't sink. Even though they are not that dense, buildings stand on foundations that funnel their mass to a small area, creating very large pressures. The point of the foundations is to hold the building on solid rock so they do not sink.
  2. The author suggests to initiate the crack with explosives. Perhaps a nuclear explosion. We had many nuclear explosions in the past - they don't make cracks. They make craters.
  3. Crack propagation is only relevant in the lithosphere. Once you get to the asthenosphere, about 200 kilometres deep, cracks don't propagate any more and the equations the author presents are meaningless. Plastic behaviour of rocks is achieved much earlier anyway.
  4. Even if there is a crack, it will have smaller cracks radiating from it. The liquid iron, being liquid, will fill those smaller cracks and lose from the mass of the main liquid blob. With time, the blob will get smaller and smaller until it will not be able to do anything any more.
  5. The melting point of iron is more than 1500 °C, this higher than the temperature of the ambient rocks. It will lose heat quicker than it gains by gravitational potential and solidify.
  6. The iron will react with the rocks around it. Particularly if it is liquid, because these things are highly reactive. It will start dissolving things into it (which will counter point 5), but it will also lose iron to the surrounding rocks. It will dissolve oxygen, carbon, sulfur, possibly alkalis like sodium, maybe silicon. These are light elements that will overall lower the density of the ball of iron, bringing it to a stop.
  7. The oxygen fugacity of the crust and upper mantle is higher than iron-wustite, which will be imposed by the equilibrium with iron. Therefore, it will leach oxygen from the surrounding rocks, essentially turning some of the liquid iron into "rock". This will also decrease the mass of the blob of iron.

You were specifically asking about lead. All points above (except 5) are equally applicable to lead.

Gimelist
  • 23,122
  • 4
  • 68
  • 133
  • I disagree with your points 1-3. And the others can be solved. The whole point of crack propagation is the extreme accumulation of stress in the tip of the crack. Therefore, your analogy with buildings is incorrect, both for the area and because is a solid, and here the key is that the stress is provided by hydrostatic pressure. In ice sheets, crack propagation can make a a crack under a surface pond propagate a couple kilometers in solid ice until it reaches the bed. That happen often in Greenland every summer, and the density contrast is much less than iron/crust. – Camilo Rada Apr 14 '19 at 02:14
  • Then, once the blob pass the layers with brittle behaviour, it don't needs cracks. It would sink just by density difference, as subducted slabs do, or similarly to the way magma diapirs rise. For the other points, you would just need a larger initial volume. – Camilo Rada Apr 14 '19 at 02:17
  • @CamiloRada what you just said just addresses point 1. But regardless of that, cracks only exist in the lithosphere which is the top ~200 km. Even if you could magically make a crack that goes down 200 km, it stops there. – Gimelist Apr 14 '19 at 02:17
  • @CamiloRada the original paper talks about how the crack propagates at 5 metres per second, making the mission take about a week. Once you're in the asthenosphere, timescales become hundreds of thousands of years. Which is more than enough time to chemically equilibrate the iron (or lead, or whatever) with the ambient mantle silicate and dissolve it away. – Gimelist Apr 14 '19 at 02:19
  • I agree, that in in the astenosphere the sinking would be slower, but I would not be so quick in saying it will take hundreds of thousands of years, it could be much quicker than that. – Camilo Rada Apr 14 '19 at 02:24
  • @CamiloRada or in fact - never. Remember chemical equilibrium. Metallic iron will react away to make things that are not metallic iron, and are much less dense. – Gimelist Apr 14 '19 at 02:25
  • Maybe, but I'm not so sure about that chemical equilibrium argument. In that case the iron catastrophe would have not taken place and the core would not be made of iron. This make me think that because of the iron catastrophe, maybe the mantle is chemically saturated of iron, and that's why he picked iron insted of lead. – Camilo Rada Apr 14 '19 at 02:29
  • The iron "catastrophe" occurred because oxygen fugacity was buffered between roughly equal amounts and somewhat well mixed portions of metallic iron+nickel and silicate. This is not the case here. You're dropping a 30 m ball of iron into a 1000000 m ball of silicate. Some people might argue that deep enough in the mantle metallic iron is stable, but (1) you first have to get there through hundreds of km of relatively oxidised rock and (2) I am still not convinced this is the case, because reduced iron will be in sulfides, not metal. – Gimelist Apr 14 '19 at 02:52
  • So in that case: why the liquid iron and nickel of the outer core have not dissolved away into the mantle? It have definitely had enough time to do so. At the end the outer core is also a blob of liquid iron in a sea of silicates. I would naively think that is doesn't dissolve because it is already in chemical equilibrium, and in that case, any other blob of liquid iron would be in equilibrium as well. – Camilo Rada Apr 14 '19 at 15:53
  • @CamiloRada excellent question. You'd think that the silicate earth would be in equilibrium with the metallic core because they were in equilibrium when the Earth formed, but the fact is that they're not. Figure 12 from Foley (2011) shows it pretty well. It only gets to QFM-5 (≈IW=iron-wustite) at a depth of 500 km. There is still some disagreement to why this is the case, but the fact is that the upper mantle is more oxidised than iron-silicate equilibrium. – Gimelist Apr 14 '19 at 23:24
  • I'm pretty lost in the geochemistry technicalities and lingo here. I'll post a new question to see if you or somebody else can clarify this for me, building up some of the concepts your refer to from basic principles. I feel your answer is a good one, but doesn't really answer the question, anyway I understand much better the problem now and have some guesses about the answer. – Camilo Rada Apr 15 '19 at 03:09
  • Question posted: https://earthscience.stackexchange.com/q/16744/11908 – Camilo Rada Apr 15 '19 at 03:20
  • 1
    @CamiloRada yea - it doesn't really answer the question of why lead and not iron. No idea. My answer was mostly pointing out that it doesn't matter, both will fail. – Gimelist Apr 15 '19 at 03:58
  • Look page 57 of "Formation of the Earth’s Core" (https://websites.pmc.ucsc.edu/~fnimmo/website/Rubie_Vol9.pdf) "Liquid iron ponded at the base of the magma ocean may also, under the right conditions, sink rapidly toward the Earth’s core by diking. Although it may be supposed that the hot, but nevertheless crystalline, mantle underlying the magma ocean cannot support brittle cracks, numerical studies summarized in Rubin (1995) indicate that dikes can still form, so long as the contrast in viscosity between the fluid in the dike and the surrounding host rocks is greater than $10^{11}-10^{14}$". – Camilo Rada Apr 15 '19 at 21:31
  • @CamiloRada Keep on reading - it says that therefore it will not equilibrate with the surrounding mantle. It will not oxidise. This is reasonable for a magma ocean and huge kilometre-scale blobs of iron. The metal blob proposed in the OP here is only a few metres across. It will equilibrate with the mantle, and cease being a blob of iron. – Gimelist Apr 16 '19 at 00:55