Could meteorites contain evidence of troilite drainage and trapped iron grains?

[snorkack]

I understand that when chondritic material is heated, the first material to melt is matte - eutectic mixture of troilite with a small excess of iron-nickel. This melts at about 990 Celsius.

At this temperature, most silicate minerals are still solid. And so is nickel-iron, which needs over 1400 Celsius to melt.

The grains of nickel-iron may be softened at 1000 degrees, but they are still solid. Therefore they should get stuck between the solid silicate grains and be unable to drain to the core with molten matte.

Are meteorites commonly found which have undergone partial melting such that troilite has drained away and iron grains remained stuck mixed with rock?

 

[Barakn]

From the phrase "drain to the core," can we assume that you are referring to some hypothetical >990 C heating event in the parent asteroid or planetisimal, not during the meteorite's fall to Earth?

 

[snorkack]

From the phrase "drain to the core," can we assume that you are referring to some hypothetical >990 C heating event in the parent asteroid or planetisimal, not during the meteorite's fall to Earth?

Yes. Parent asteroid.

So, "achondrites" are supposed to be lavas frozen from completely molten material. Chondrite type 3 is pristine, never heated dust, and chondrite types 4...6 are chondrite rocks which have been heated to various temperatures short of melting.

But considering that the different minerals of chondritic material melt over a fairly wide range of temperatures, what do rocks look like that have undergone partial melting and partial, not complete, differentiation?

 

[Barakn]

Well...

I think you are looking for primitive achondrites, meteorites which appear to be chondrites that partially melted. More specifically, Acapulcoites and Lodranites. I think the idea that the melt from troilite could completely drain away from a matrix of higher melting-point minerals is a bit far fetched because of the environment in which this is happening: a large asteroid, partially molten and partially differentiated, and therefore an object whose center is already occupied by melt (possibly molten iron-nickel, which would have a higher density than the iron-sulfur from molten troilite). Thus the Acapulcoites and Lodranites contain troilite, which recrystallized from the partially molten chondritic material as it cooled.

 

[snorkack]

More specifically, Acapulcoites and Lodranites. I think the idea that the melt from troilite could completely drain away from a matrix of higher melting-point minerals is a bit far fetched because of the environment in which this is happening: a large asteroid, partially molten and partially differentiated, and therefore an object whose center is already occupied by melt (possibly molten iron-nickel, which would have a higher density than the iron-sulfur from molten troilite).

But molten iron-nickel has much higher melting point than iron-sulphur. Over 1440 degrees, and actually over 1500 degrees for the usual nickel concentrations.
Just because iron-nickel is denser that troilite does not allow it to differentiate if it is still solid and still stuck between the also solid grains of refractory silicates.

On slow heating:
The initial chondrite material is loose dust, with grains weakly resting against each other, and vacuum in between.
As the material is heated, well below the melting point, solids would soften, they would undergo rapid solid diffusion, creep under mechanical pressure, and recrystallization to bigger crystals - all under melting point as stated. The dust would be sintered, settled and compressed, decreasing the porosity - first the grain contacts would increase, but at high temperatures they would be weak and liable to slip, then the remaining pores would be sealed shut.

When some of the solids do reach melting point, the first drops of liquid should be trapped in waterproof network of sintered crystals, right?
The typical metal iron content in chondrites is 4 to 20 % metal by weight - and because of their high density, rather less by volume. Troilite is up to 5 % (mass).

Hm... could it be that so long as it is troilite alone that is molten, it would stay trapped in solid?

Then the first rocks to melt should be feldspars - especially plagioclase, that melts under 1200 degrees. Note that at that point, pure nickel-iron is still solid - and so are most of the other rocks, like the higher melting feldspars, and the basic enstatites and magnesia-rich olivines.

At which melt volume fraction does partially melting rock become permeable for the melts to drain away under gravity?

 

[Barakn]

I see your point. It's just that in the group of meteorites of interest, the primitive achondrites, there doesn't appear to be any that match what you're looking for. These are pretty rare - the Acapulcoites and Lodranites apparently both come from the same parent body, and it's my guess that this particular object had a molten metal core. Perhaps there were other objects that didn't get quite so hot and hit the magic temperature range between 990 and 1400 C, but we just haven't gotten any samples of them.

Or maybe we are not thinking about the problem the right way. One factor we've neglected is surface tension. Iron at 1580 K (melted using a gas mixture containing H2 to remove oxygen from the iron) has a surface tension of ~ 2000 milliNewtons per meter (mN/m), which is 28x the surface tension of water at 25 C. http://iopscience.iop.org/1742-6596/327/1/012020/pdf/1742-6596_327_1_012020.pdf It's a bit harder to find data on iron-sulfur systems in the right temperature range and sulfur concentration, but here http://www2.matse.psu.edu/modeling/597K/surface_tension1.PDF we find that at 1930 K, increasing amounts of sulfur decrease the surface tension, but at the largest concentration of sulfur measured, was still a respectable 1300 mN/m. It's quite likely that Fe-S melt has a considerable surface tension, much higher than water, and combined with the extremely weak gravity field of an asteroid, it would have stuck where it melted rather than draining, like a wet sponge.

 

[snorkack]

Or maybe we are not thinking about the problem the right way. It's quite likely that Fe-S melt has a considerable surface tension, much higher than water, and combined with the extremely weak gravity field of an asteroid, it would have stuck where it melted rather than draining, like a wet sponge.

Maybe - so long as there is a sponge to stick to.

We know that there were many parent bodies that DID differentiate all the way to molten metal core, because the iron-nickel meteorites we see originate from many parent bodies AND there must be more parent bodies with iron-nickel cores that never got shattered to meteorites.

How?

When chondrite material is heated above 1200 degrees, rock starts melting. First plagioclases, then the rest of feldspars.
And then the refractory rocks remain.
And these refractory rocks would sink and compact.
The density range of feldspars, at room temperature, is 2,55-2,76

For the refractory rocks, pyroxene is 3,2-3,3 at the enstatite end, and hypersthenes go up to 3,9. Olivine is 3,21-3,33 at the forsterite end, going to 4,39 at fayalite end.

On the other end, troilite is 4,67-4,79.

The rocks are usually much richer in magnesium than iron. Pure enstatite pyroxene would melt at 1557 degrees, incongruently, leaving refractory olivine residue and silica-rich melt. And pure forsterite, in the complete absence of fayalite solution and silica-rich melts to dissolve it, would melt congruently at 1890 degrees.

So above 1200 degrees, you´d have refractory, magnesia-rich residue and two different types of melts. Silica-rich feldspar melts that drain UP and iron-rich troilite and iron-nickel melts that drain DOWN if they can.

Silicate melts are miscible with each other - therefore they should ALSO wet the still solid refractory silicates, allowing them to seep through the pores, while the immiscible troilite and iron melts should wet rock poorly and therefore have problems getting through pores. (Is wet sponge also an impassable barrier for oil?)

Thus, it is not implausible if the light silica-rich melts drain up BUT the heavy metal/sulphide droplets remain trapped in the heavy olivine residues.

Of course, in the presence of hot silica-rich melts, olivine would not endure all the way to melting point at 1890 degrees. But on Earth, the komatiite lavas, of mostly forsterite olivine, certainly did erupt at over 1600 degrees.

So, there would be a refractory olivine sponge unmolten at least over 1600 degrees!
Sure, the volume fraction of the refractory residue from the original volume would progressively decrease before complete melting, as first the refractory end feldspars melt and then the pyroxenes undergo incongruent melting to olivine. This would upset the pore structure holding up the iron/sulphide melts. BUT, it is under pressure and quite soft near melting point. Therefore, I suspect that the sponge would compress under its weight as silica-rich melts drain up - and this settlement would narrow the pores again, closing them for the iron/sulphide drops before they manage to travel far.

Thus, below the final melting of olivine over 1600 degrees where the molten iron would finally be released and settle to completely differentiated iron core, I should expect a sponge of olivine and concentrated iron/sulphide - concentrated because silicon/alumina rich melts have drained out but olivine has been left behind together with the original iron.

Are such sponges seen?