Looking at this website, I can see the melting point of the following minerals: http://hyperphysics.phy-astr.gsu.edu/hbase/Geophys/meltrock.html

Apparently at a temperature of 1200 °C, all the minerals exist in a molten state. But, looking at the melting point for quartz on Wikipedia I find that the melting point is close to 1700 °C. What’s going on here?

It can’t be an effect of the pressure. In the interior, the pressure will be higher and thus further increase the melting point.


Start with my answer to this very highly related question here:


The melting point of minerals in isolation, or a pure substance is higher than mixtures of minerals. For example - a (well-mixed) mixture of quartz and pyroxene will melt at a lower temperature than pure quartz or pure pyroxene. The exact temperature is dependent on the proportions.

In general, the more different kind of components (i.e. "things": minerals, elements, molecules, etc) you add to a mixture, the lower the melting point of the mixture as a whole.

For example, sea water melts at -2 °C because it's a mixture of water (melts at 0 °C) and all kinds of salts. Steel melts at a lower temperature than pure iron because it's a mixture of iron and carbon.

  • 1
    $\begingroup$ It's also possible that some components of a mixture might dissolve in the first to become liquid, rather than melt themselves. For example, you can dissolve quite a bit of common salt (among many other things) in water, even though the water is nowhere near the salt's 801 C metlting point. $\endgroup$ – jamesqf May 20 '18 at 19:54
  • $\begingroup$ i see... So Bowen's reaction series depends on composition of magma and not just the individual minerals in isolation. That's why it's a purely experimental result. $\endgroup$ – Neeraj Kulkarni May 20 '18 at 21:29
  • $\begingroup$ @NeerajKulkarni exactly. It's an experimental result that has strong relevance for natural systems. Bowen's experiments were conducted about 100 years ago, so they miss a lot of complexities that exist in natural rocks. But, they are generally applicable for most eruptive rocks (such as basalt, andesite and rhyolite). $\endgroup$ – Gimelist May 20 '18 at 23:22
  • $\begingroup$ @NeerajKulkarni It's not purely an experimental result. You can predict the order of mineral crystallization from thermodynamics as well. $\endgroup$ – bon May 21 '18 at 8:09
  • 1
    $\begingroup$ @bon sorry to nitpick, but thermodynamic data used to make these predictions is obtained using experiments. $\endgroup$ – Gimelist May 21 '18 at 9:01

The eutectic is the minimum melting point of a mixed material, like soda lime glass made from sodium hydroxide, calcium oxide, and silicon dioxide. That mixture, in the proper proportions, lowers the melting point of quartz sand (silicon dioxide) from 1700 to under 1000 Celsius. In the mantle the mixture is far more complex but if I remember my geochemistry Aluminium, Iron, Magnesium, and Calcium are the main elements that lower the melting point of pure silica in magmatic melts. Water can be very important as it not only lowers the melting point of saturated rocks but also carries heat into lower temperature melting materials, especially in subduction zones.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.