# Does granite become soft as modelling clay at 2000 °C?

As Wikipedia tells us, it consists of $\ce{Al2O3}$ ($14.42\%$) (alumina) whose melting point is $\pu{2072^\circ C}$, and $\ce{SiO2}$ ($72.04\%$) (silica) whose melting point is $\pu{1713^\circ C}$. So does it soften to modelling clay (or at least partially soft, modelable via tools structure) at $\sim\pu{2000^\circ C}$, or does it stay hard and why?

• There is no granite as a compound; it's chemical composition deviates quite a lot, and so do the thermal properties. This reminds me of the average hospital temperature of 36.6 °C (averaged from the patients with fever, 40 °C, and morgue, 4 °C) – andselisk Dec 27 '17 at 20:54
• Granite at this temp. is magma, not exactly modelling stuff... – Mithoron Dec 27 '17 at 22:53
• One of the reasons why granite looks the way it does (some large crystals of feldspars embedded in a matrix of smaller crystals of quartz or mica) is because the components of granite don't crystallise at the same temperatures. Also granite is formed by slow cooling deep in the earth not fast cooling on the surface. – matt_black Dec 28 '17 at 12:06

Mixtures of different compounds do not usually work like that. They generally melt at lower temperatures than the constituent pure compounds. So I would expect your granite to be not like modeling clay, but molten even before getting to the melting point of silica.

If you want modeling clay, try 1300 °C where the Earth's lithosphere gives way to the asthenosphere. See the Characteristics section of https://en.wikipedia.org/wiki/Asthenosphere.

Granite is not composed of the oxides listed, they are just a way of expressing the weight percentage of elements in the sample. Granite (and other rocks) are actually composed of minerals.

All granites contain quartz + alkali feldspar + plagioclase feldspar. Biotite is also pretty common and a variety of other minerals can also be less commonly present.

This is a phase diagram of the quartz - alkali feldspar - plagioclase feldspar system at 5 kbar pressure. The first melt forms in this system at the eutectic point at around 650 °C, but complete melting would not occur until a much higher temperature. I expect the last thing to melt would be left over quartz although it would first transform to tridymite and then cristobalite before melting finally melting.

• So... when it gets to be "soft" so that one could model with it like with clay (*having tools that do not melt at that temp)? – Blender Jan 27 '18 at 21:17
• @Blender Rocks never get that soft. They melt before they get to that stage. – bon Jan 27 '18 at 22:28