8
$\begingroup$

I was learning about how water, because of its hydrogen bonds, actually gets less dense as it goes into its solid state - I was just wondering, what other elements do this? Are they similar to water?

P.S. I want to make sure that I understand why ice is less dense than water in its liquid state, so please correct me if I am wrong.

In water, the hydrogen bonds have the ability to move around and bounce with other hydrogen bonds. They also have the ability to break and reform. Their ability to move around and bounce around allow them to be more 'closely knit'. But, as they get cooler, they, the molecules, loose their kinetic energy and thus stop moving around and become more framed and rigid. They loose the ability to come closely together. And thus, ice is less dense.

What I don't understand also, is why making water cold makes the hydrogen bonds spread out. Why don't they just freeze in the position they are in, instead of spreading out?

$\endgroup$
7
  • 2
    $\begingroup$ Elements do not form H-bonds. E.g. silicon, gallium, germanium, bismuth, and plutonium have higher liquid than solid density at standard pressure. $\endgroup$
    – aventurin
    May 16, 2016 at 15:58
  • 1
    $\begingroup$ And antimony as well. It really isn't that unusual, but many people seem to think water is unique. It isn't. $\endgroup$
    – Jon Custer
    May 16, 2016 at 16:10
  • 1
    $\begingroup$ Not an element, but "type metal", an alloy of lead, antimony, and tin, uses that property of expanding on freezing to ensure that the type fills the mold without leaving voids. $\endgroup$ Jul 17, 2016 at 17:03
  • $\begingroup$ Without checking it I would say you could look for metals and elements that do not form close-packings. Usually metals prefer known structures in their solid state like the Cu, Mg and W-type. But there are some like Ga, In, Mn, Po, ... that do not form close or dense packings but have a considerably lower density. I can imagine this changes when they melt and may perhaps lead to a higher density in their molten state. But that is just my thoughts without further verification by literature or so. $\endgroup$ Mar 28, 2018 at 8:36
  • $\begingroup$ We can add lithium to the list, at high pressure. The fcc phase tends away from a close-packed structure towards a more open structure, which may contract as it melts, because of the separation of electronic charge from the ionic cores. $\endgroup$ Apr 11, 2022 at 20:10

1 Answer 1

11
$\begingroup$

Your terminology is a little off: water isn't an element.

There isn't really a simple single explanation of why some solids are less dense than their associated liquids. The general explanation (which doesn't really explain much) is that the solid has a structure and, sometimes, that structure takes us more space than the average structure that occurs in the liquid.

In water, ice has a definite structure where each oxygen sits at the approximate centre of a tetrahedron with two bonds to hydrogen and two hydrogen bonds to the hydrogens of other water molecules. In liquid water the structure is far more fluid and averages out to be slightly more dense that the (more ordered) solid. Exactly why this is true in this case requires a lot more than could fit into an answer here. If water was frozen extremely quickly you might be able to freeze the structure in the liquid; normal cooling allows enough time for the bonding to become more organised.

But the phenomenon isn't unique to water, though the specifics will be very different for other substances. Among elements Gallium, for example, also expands on freezing (don't store it in glass bottles: see this other answer here).

Other examples of elements are germanium and silicon. Silica is also less dense as a solid than as a liquid. In all cases the substances form a relatively ordered crystal structure which just happens to be less dense than the (less ordered) liquid phases.

$\endgroup$
1
  • $\begingroup$ If we cool running water it will even reach below zero temperatures but remains liquid. What will the density be in this case? $\endgroup$
    – user85778
    Jun 15 at 3:14

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