Looks of komatiite flows

On 6 nov, 12:03, "John Kepler" wrote:
Fluidity/viscosity has more to do with the chemistry of the

material than the temperature!

A clarification. This is NOT to imply that komatiite flows weren't
hot.....they were hotter than a jalapeno fart, but just that their
ultramafic chemistry was responsible for their water-like fluidity. You
could heat more felsic dacitic or rhyolitic magma to the same temperature
and NOT get the same fluidity as the ultramafic komatiites.

BTW, all of this is stretch for an old man living on the Craton digging coal
for a living....but I have whacked a couple of komatiite flows up on the
Shield....damn laterites!

Does it mean that unlike the basalt, let alone rhyolite lavas, which
freeze like amorphous substances - get increasingly viscous and pasty
on cooling, till they solidify first on both extremities and finally
in the interior - the komatiite lavas ought to be fluid at all
temperatures down to a sharp freezing point?

Then, if we have a freezing komatiite flows, the komatiite crystals
ought to sink through the still fluid liquid to the bottom, so the
komatiite flow should freeze from bottom up, till the last remaining
fluid on the very top freezes - and only then should the surface start
cooling below the melting point.

How close has anyone got to the active komatiite flows of Io?

Crown-Horned Snorkack wrote:


Then, if we have a freezing komatiite flows, the komatiite crystals
ought to sink through the still fluid liquid to the bottom, so the
komatiite flow should freeze from bottom up, till the last remaining
fluid on the very top freezes - and only then should the surface start
cooling below the melting point.

All freezing points are sharp - for a given pressure *and* chemically
homogeneity. The problem with your scenario above is that as the
komatiite crystallises out, the chemical composition of the remaining
melt changes. It's been a long time since I did any igneous petrology...
but here goes.

If the Mg-rich olivines crystallise first, the remaining melt will
become more Si-rich. The crystals, given the v. low viscosity of the
magma, will sink, trapping some of the remaining melt in the matrix, but
essentially you'll have a liquid sitting on a crystal slush. The
Si-rich(er) melt (given the very low initial concentrations of K, Al, Ca
and Na) will still be high in Mg - pyroxenes and Mg-feldspars will form.
The liquid itself will become increasingly viscous due to not just
cooling but to the Si content.

I also note that komatiites come out bloody hot - way hotter than flood
basalts even. At 1 atm, pure forsterite freezes at 1900oC. So as soon as
the temperature drops below this, olivine crystals will start to form.
This changes the chemical composition as noted: it also changes the
freezing point of the melt, so what remains will stay liquid at lower
temperatures.


How close has anyone got to the active komatiite flows of Io?

rofl


Simon Morden
--
Visit the *all new* Book of Morden (www.bookofmorden.co.uk)
"I haven't had that much fun with a novel for a while." - Bookbag
The Lost Art - from David Fickling Books

On 7 nov, 16:15, Simon Morden
wrote:
Crown-Horned Snorkack wrote:

Then, if we have a freezing komatiite flows, the komatiite crystals
ought to sink through the still fluid liquid to the bottom, so the
komatiite flow should freeze from bottom up, till the last remaining
fluid on the very top freezes - and only then should the surface start
cooling below the melting point.

All freezing points are sharp - for a given pressure *and* chemically
homogeneity. The problem with your scenario above is that as the
komatiite crystallises out, the chemical composition of the remaining
melt changes.

Some melts freeze into one solid phase with no change of composition
in the process. Some melts freeze into two or more solid phases with
no composition change at a sharp eutectic melting point. and some
melts get increasingly viscous on cooling and freeze into a glass
without forming any crystals.

It's been a long time since I did any igneous petrology...
but here goes.

If the Mg-rich olivines crystallise first, the remaining melt will
become more Si-rich. The crystals, given the v. low viscosity of the
magma, will sink, trapping some of the remaining melt in the matrix, but
essentially you'll have a liquid sitting on a crystal slush. The
Si-rich(er) melt (given the very low initial concentrations of K, Al, Ca
and Na) will still be high in Mg - pyroxenes and Mg-feldspars will form.
The liquid itself will become increasingly viscous due to not just
cooling but to the Si content.

So that once a large amount of olivine has frozen and settled out of
the initially fluid komatiite, the remaining melt is more silica-rich
and more viscous, more like ordinary lava and ought to form a crust on
top like ordinary lava does?

I also note that komatiites come out bloody hot - way hotter than flood
basalts even. At 1 atm, pure forsterite freezes at 1900oC. So as soon as
the temperature drops below this, olivine crystals will start to form.
This changes the chemical composition as noted: it also changes the
freezing point of the melt, so what remains will stay liquid at lower
temperatures.

But what about iron and fayalite? Are komatiites rich in iron? And
does iron freeze out along with magnesium or remain in melt while
forsterite freezes out?


How close has anyone got to the active komatiite flows of Io?

rofl

You are right - no one has been much further than Moon, and therefore
everyone has been exactly as close as Earth is to Io.

However, a fair number of probes have passed Jupiter and satellites,
carrying instruments and sending back photos. Pioneer, Voyager,
Galileo... How thoroughly is Io known, between those probes?

Also, komatiite is supposed to be very hot and fluid. If so, it would
be expected to erupt quietly, spread in smooth floods filling holes
and valleys, like the gentle basalt shields of Hawaii and yet gentler
flood basalts.

Io seems to have violent volcanic explosions. Even more violent than
on Earth.

How do those happen, and what is the petrology of those explosion
products?

Crown-Horned Snorkack wrote:
On 7 nov, 16:15, Simon Morden
wrote:

Crown-Horned Snorkack wrote:


Then, if we have a freezing komatiite flows, the komatiite crystals
ought to sink through the still fluid liquid to the bottom, so the
komatiite flow should freeze from bottom up, till the last remaining
fluid on the very top freezes - and only then should the surface start
cooling below the melting point.

All freezing points are sharp - for a given pressure *and* chemically
homogeneity. The problem with your scenario above is that as the
komatiite crystallises out, the chemical composition of the remaining
melt changes.


Some melts freeze into one solid phase with no change of composition
in the process. Some melts freeze into two or more solid phases with
no composition change at a sharp eutectic melting point. and some
melts get increasingly viscous on cooling and freeze into a glass
without forming any crystals.

It's a rule of thumb that the faster the cooling, the smaller the
crystal size - obsidian cools too quickly for nucleation points (though
there might be some post-freezing ie snowflake obsidian). A change in
bulk composition will occur if, frex, a biphase melt AB freezes out
crystals of A which then get removed from the remaining melt B - either
by gravity or by mechanical means.



It's been a long time since I did any igneous petrology...
but here goes.

If the Mg-rich olivines crystallise first, the remaining melt will
become more Si-rich. The crystals, given the v. low viscosity of the
magma, will sink, trapping some of the remaining melt in the matrix, but
essentially you'll have a liquid sitting on a crystal slush. The
Si-rich(er) melt (given the very low initial concentrations of K, Al, Ca
and Na) will still be high in Mg - pyroxenes and Mg-feldspars will form.
The liquid itself will become increasingly viscous due to not just
cooling but to the Si content.


So that once a large amount of olivine has frozen and settled out of
the initially fluid komatiite, the remaining melt is more silica-rich
and more viscous, more like ordinary lava and ought to form a crust on
top like ordinary lava does?

For some values of ordinary, yes. From the New Zealand simulation, they
suggest that the lava tubes will stretch a long way. I imagine that for
a large, turbulant flow, the crust will be continually reabsorbed until
the eruption has all but ceased. Komatiite is *so* rich in Mg, so poor
in other metals, that left to its own devices, it'll be almost all olivine.

Wiki tells me that the lavas often have a super-cooled appearance -
densely packed small crystals, whilst magma chamber deposits have
massive aggregations of olives and pyroxenes. It does ultimately depend
on the cooling rate. And probably the pressure, too - I don't have a
phase diagram of P-T variations for olivines to hand...

The nearest analogue I can think of is the cement product called a
levelling compound: it's a cement slurry that flows like water (finding
its own level). Then it sets solid to produce an exactly flat surface.



I also note that komatiites come out bloody hot - way hotter than flood
basalts even. At 1 atm, pure forsterite freezes at 1900oC. So as soon as
the temperature drops below this, olivine crystals will start to form.
This changes the chemical composition as noted: it also changes the
freezing point of the melt, so what remains will stay liquid at lower
temperatures.


But what about iron and fayalite? Are komatiites rich in iron? And
does iron freeze out along with magnesium or remain in melt while
forsterite freezes out?

I don't know about the Fe content. But with the Mg content so
unspeakably high, there's not much room for Fe. If they are, as
suspected, formed by partial melting and differentiation elsewhere
before accumulating, the Fe has probably been excluded in the first
stage melt.



How close has anyone got to the active komatiite flows of Io?

rofl


You are right - no one has been much further than Moon, and therefore
everyone has been exactly as close as Earth is to Io.

However, a fair number of probes have passed Jupiter and satellites,
carrying instruments and sending back photos. Pioneer, Voyager,
Galileo... How thoroughly is Io known, between those probes?

Also, komatiite is supposed to be very hot and fluid. If so, it would
be expected to erupt quietly, spread in smooth floods filling holes
and valleys, like the gentle basalt shields of Hawaii and yet gentler
flood basalts.

Io seems to have violent volcanic explosions. Even more violent than
on Earth.

How do those happen, and what is the petrology of those explosion
products?

Violence in volcanic eruptions is almost always due to the presence of
volatiles. Gas-rich (including steam) magmas which get trapped beneath
ground or behind a plug of cool lava are spectacularly bad news.

What volatiles Io has entrained in its magmas is a matter for
speculation - I would guess at compounds of sulphur.


Simon Morden
--
Visit the *all new* Book of Morden (www.bookofmorden.co.uk)
"I haven't had that much fun with a novel for a while." - Bookbag
The Lost Art - from David Fickling Books

obsidian cools too quickly for nucleation points (though
there might be some post-freezing ie snowflake obsidian).

Actually it isn't even a solid, but a super-cooled liquid, and it never
stops flowing!

John

John Kepler wrote:
obsidian cools too quickly for nucleation points (though

there might be some post-freezing ie snowflake obsidian).


Actually it isn't even a solid, but a super-cooled liquid, and it never
stops flowing!

John


Also true - but it does shatter into a whirling cloud of razor-sharp
shards when struck with sufficient force, frex with a geology hammer...*

Simon Morden

*not me, fortunately.
--
Visit the *all new* Book of Morden (www.bookofmorden.co.uk)
"I haven't had that much fun with a novel for a while." - Bookbag
The Lost Art - from David Fickling Books

Also true - but it does shatter into a whirling cloud of razor-sharp
shards when struck with sufficient force, frex with a geology hammer...*


Been there, done that, got the scars......a flow north of Las Vegas!

John

John Kepler skreiv:
obsidian cools too quickly for nucleation points (though
there might be some post-freezing ie snowflake obsidian).

Actually it isn't even a solid, but a super-cooled liquid, and it never
stops flowing!

Sorta like glass ?

/ducks and covers/


Eivind

Sorta like glass ?

No, EXACTLY like glass, which is what it is....so is pyroclastic ash!

John

"John Kepler" wrote in message
...
obsidian cools too quickly for nucleation points (though
there might be some post-freezing ie snowflake obsidian).

Actually it isn't even a solid, but a super-cooled liquid, and it never
stops flowing!

John

Can you point to a cite describing the actual observation of obsidian flow
at condtions near STP?

Recall of course that even crystalline materials (solid by any definition)
can be found to flow (deform) in geologic formations since they are
subjected to high (but sub-melting) temperatures and pressures over long
periods of time.

Silica glasses near room temperature are amorphous solids and not liquids.
They are rigid and do not flow at all, their atoms locked into place even
though they are not in crystals. There is a glass transition temperature
below which a fundamental thermodynamic process begins to take place as it
cools further, which is reflected in the change of the specific heat and
density of the glass just like the true phase change of solidification that
occurs in crystalline solids.

It is true that the glass transition to a solid is not abrupt like
crystallization solidification, and that there is a zone below the glass
transition temperature where viscosity increases to very high levels and the
physical state is ambiguous. But the glass does not remain in this state of
ambiguity regardless of how low the temperature goes, there is a point where
the transition is complete.