Why does iron sink in molten iron whereas ice floats on water?

Both are solid states of their own form, so why is one floating and the other sinking?


Solid water (ice) is one of the few known solids whose density is lower than that of its liquid form. Yes water is special! (but very much so in its chemical properties too) Due to the crystal structure of the solid phase of water, the molecules arrange themselves in a rigid, ordered fashion and end up being, on average, farther apart from each other (than they are in the liquid phase), and thus less dense. Less dense things float because of buoyancy. Thus, in most solids, such as iron, the molecules arrange themselves in a rigid, ordered fashion that ends up being, on average, closer from each other than they are in the liquid phase, and thus more dense. that one makes it closer and the other farther apart than liquid. Why water atoms separate and iron atoms get closer is more difficult to explain, to be able to show that, you have to actually calculate it using QM models. There is no intuitive argument, as far as I know, that will show you why one gets denser and the other less so.

  • $\begingroup$ Could you perhaps explain as to why water has lower density in solid form but iron doesn't $\endgroup$ – Gaurav Nov 25 '14 at 4:57
  • $\begingroup$ I updated my answer $\endgroup$ – Wolphram jonny Nov 25 '14 at 5:12
  • $\begingroup$ you are missing a concluding sentence to the effect that iron in solid form is denser than in liquid, to complete the answer. $\endgroup$ – anna v Nov 25 '14 at 5:46
  • $\begingroup$ Both water and iron form crystal lattice in their solid state. But what makes the lattice structure more dense than the liquid state in iron and less dense in water ? $\endgroup$ – Gaurav Nov 25 '14 at 6:01
  • 2
    $\begingroup$ that one makes it closer and the other farther apart than liquid. But to be able to show that, you have to actually calculate it using QM models, there is no intuitive argument that will show you why one does and the other does not. – $\endgroup$ – Wolphram jonny Nov 25 '14 at 6:17

Basically, it has to do with the density of the material as a function of temperature.

The density of iron increases as it cools, that is, solid iron is more 'packed tight' than when it is melted. This is understandable, since the kinetic energy of the iron atoms decreases as the temperature drops (ie: the average velocity of the atoms decreases), allowing them to pack more tightly in a given volume. This would be the 'expected' behaviour of most matter.

As you might expect, liquid water is generally similar, in that, as it is cooled, the kinetic energy of the molecules decreases and the density of the water increases, until we get down to $4^o C$, when something interesting starts to happen.

When the temperature drops below $4^o C$, the hydrogen bonds effectively 'hold' the water molecules further apart as a hexagonal crystal lattice starts to form.

The effect of the hydrogen bonding can be thought of as a 'relatively weak ionic' bond, in that at low temperatures the molecules begin to arrange themselves in a crystalline structure (the hexagonal crystal lattice). Basically, below $0^o C$ water becomes ice (an open hexagonal crystal) in which the water molecules are 'locked' in to an arrangement where there is more 'space' than the liquid state.

Ice floats because it is less dense as ice than it is as liquid. The water in ice expands by about 9%.

Density of water at $0^o C$ is 0.9998 g/mL

Density of ice at $0^o C$ is 0.9167 g/mL

That's why icebergs float with about 9% of their mass above the surface of the water.

enter image description here

Further cooling of ice, that is, below $0^o C$, results in an increase in density with respect to ice at $0^o C$, similar to solid metal.

So the 'anomalous' behaviour of water happens between $0$ and $4^o C$.

To learn more about molecular bonding (and the hydrogen bond in particular), refer to Chapter 12 of "The Nature of the Chemical Bond and Structure of Molecules and Crystals", by Linus Pauling.

Here's a quick summary.

Normally, hydrogen only forms one covalent bond, eg: the bonds between $H$ and $O$ within each $H_2O molecule.

However, when hydrogen forms such a bond with a highly electronegative atom (such as oxygen, nitrogen or fluorine), it leaves a slightly positive charge around the hydrogen atom. The presence of unbonded electrons in the outer shell of the oxygen atom give it a localised 'positive' charge, so that the water molecule is 'polar'. Adjacent molecules of water are thereby attracted to each, which is why water is a liquid at atmospheric conditions, something quite rate amongst molecules with similar atomic weight. It also gives water the ability to dissolve ionic solids.

enter image description here

So, since iron does not have such hydrogen bonds, it behave more like 'regular' matter (with respect to its density variation as a function of temperature).

  • $\begingroup$ Incorrect. Iron forms a BCC lattice structure in the solid state, which all 'regular' matter do not. $\endgroup$ – Gaurav Nov 25 '14 at 8:55
  • $\begingroup$ What I meant by 'regular' was in respect to density varying inversely to temperature. $\endgroup$ – theo Nov 25 '14 at 9:12
  • $\begingroup$ That is, iron conforms with the 'general' observation of thermal expansion of matter (ie: matter 'generally' expands at higher temperatures), even at the solid-liquid phase transition. Water, particularly near the ice-liquid temperature range between 0 and 4 degrees C, does not. $\endgroup$ – theo Nov 25 '14 at 10:19
  • $\begingroup$ I think your densities are wrong. At present you have a higher density of ice at 0 C, which would make it sink. $\endgroup$ – Phil H Nov 25 '14 at 11:21
  • $\begingroup$ @PhilH My mistake... it's the other way round *now fixed $\endgroup$ – theo Nov 25 '14 at 12:54

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.