Is there a compound that can melt under its own weight? That is, a block of such material remains solid when put horizontally but starts to melt when put vertically or remains solid in microgravity, but melts in normal gravity?

The question is motivated by a documentary which I saw on TV some years ago where they claimed that a metal block (indium or its alloy) melts under its own weight. It is suspicious though because it was quite a crank documentary which featured alleged "red mercury" for instance.

  • $\begingroup$ Are you mixing plasticity with phase change? $\endgroup$ – Mithoron Jun 9 '15 at 21:27
  • $\begingroup$ @Mithoron I am asking exactly about melting. Plastic deformation can lead to rise in temperature though which in turn can lead to melting. I wonder whether there are such compounds. $\endgroup$ – Anixx Jun 9 '15 at 21:28
  • $\begingroup$ Then it's turning of potential energy into kinetic with temperature rise - you only need big enough amount in right shape and high enough starting temperature. $\endgroup$ – Mithoron Jun 9 '15 at 21:37
  • $\begingroup$ @Mithoron so are there compounds that exhibit this property, particularly with blocks under 20 cm in length? $\endgroup$ – Anixx Jun 9 '15 at 21:38
  • 1
    $\begingroup$ By big enough I'm thinking rather about kilometers $\endgroup$ – Mithoron Jun 9 '15 at 22:01

Sure. Even discounting non-equilibrium conditions, any substance which expands upon freezing will by symmetry melt under pressure, even if the pressure is caused by supporting its own weight. There's a simple example: water! A sufficiently tall self-supporting block of ice standing on a flat surface will create enough pressure at its bottom that it could melt in temperatures between approximately $-21.9\ ^\circ \mathrm{C}$ and $0\ ^\circ \mathrm{C}$, according to the phase diagram for water. This has very important implications for the study of glaciers, for example. However, this process requires a very large amount of pressure, and thus a very tall block of ice. Using an approximate value for the freezing point depression under pressure of $0.0072\ ^\circ \mathrm{C\ atm^{-1}}$, it would take an ice block about $7.5\ \mathrm{km}$ tall for the base to melt at a temperature of $-5\ ^\circ \mathrm{C}$ or higher.

I thought indium was among the substances which expanded upon freezing, but it actually turns out to contract by 2.5% when frozen, so this mechanism would not work (though there could be something else to consider). Some other materials which could theoretically melt under their own weight in some regime of temperatures and pressures are gallium, bismuth and silicon. There might be some very specific conditions under which this effect could be achieved in a column of around $20\ \mathrm{cm}$, though pressure usually only slightly alters the freezing point, so it's likely that the material would have to be kept only slightly under its melting point at standard pressure for this to happen.

  • 1
    $\begingroup$ Brilliant answer! It really brings out the inevitable symmetry inherent in thermodynamic equilibria. (+1) $\endgroup$ – MarcoB Jun 9 '15 at 22:20
  • $\begingroup$ Don't forget germanium and antimony as elements that densify on melting. Not that it helps for this question... $\endgroup$ – Jon Custer Jun 9 '15 at 22:23

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.