A question about liquids and solids, metals vs non metals

I was wondering, why is it that metals, when being heated from say room temperature to their melting point, become malleable, and sort of change phase slowly, but non-metals (take water for example), simply go from being ice to water, with no change in malleability. Why is it that ice near freezing point isn’t a lot more solid than ice that’s much colder than freezing (or is it?).

Basically, some substances seem to change phase between liquid and solid smoothly, and others seem to change almost instantly. What is the reason for this?

One reason is pure compounds, like water and pure metals, have a distinct melting point and sharp changes in properties. Mixtures, like alloys, and especially glass and plastics, soften gradually before melting.

That’s the change in matter; physical and chemical change. What happened to the metal and water is an example of physical change. This happens when energy is present or applied.

So pure gold doesn’t become softer and more liquid-like as it approaches melting/freezing point?

Is it really just a question of purity, and not metal vs non-metal?

If it is a question of purity, then why do non-pure substances change phase more gradually than pure substances? Is it because some of the compounds are reaching the liquid phase, but others are still solid, so we see a combination of the two?

Is it really only alloys that change phase more slowly?

Metals soften at high temperature, because they are made of metallic bonds that shift more easily. (I could get a lot more complicated than that, but that’s basically it.) Ceramics generally stay hard until they reach their melting point, even “alloys.” (There is also the range of melting temperatures associated with alloys, but I think that the softening of the metallic bonds is the main thing that the OP is noticing.) This does not mean that they are “pre-melting” below the melting point - they are just softer at higher temperatures (but not “more liquid”).

ENugent, PhD metallurgist

Pure metals also soften slightly. Actually, the complex answer is - it also depends on how the material deforms. Certain types of materials soften more than others.
Metals are crystalline with their atoms stacked in an ordered arrangement (think a stack of oranges). This arrangement is not perfect and there are faults in it (technical term is “dislocations”). These faults move when you apply force on the metal, resulting in deformation. As temperature increases, it gets easier for them to move, so the material weakens.
In materials like ceramics, dislocations can’t move easily, so their strength doesn’t decrease much with temperature.
I’m not sure about ice, but since the bonding isn’t the same as in metals, I don’t think it has mobile dislocations, so it doesn’t soften very much.

In non-pure substances, yes, it’s something like that. For example, plastics are like a bunch of tangled chains (think cold spaghetti) with different lengths. As temperature increases, the shorter chains untangle first, then the longer, so the material softens gradually.

Ok. Trying to explain better, the reason some materials soften before melting and some don’t is it depends on how they deform. E.g.:
Ceramics (soften very little) - deforms by cracking, no way to gradually deform
Metals (soften) - deforms by dislocation motion
Polymers (soften a lot) - deform by chains sliding past one another
These deformation mechanisms are affected differently by temperature, so materials soften differently. There is a rough correlation between hardness and deformation mechanism, so harder materials soften less.

Purity is a simpler reason, that can be used to compare materials that deform in the same way.

Another followup question (thanks guys you are doing great). Could you make “malleable” ice by adding certain impurities to it, and raising it to near-melting temperature?

If it really is just a matter of certain substances containing dislocations, why is it that certain things like ice, no matter how impure, never seem to be able to become “softer” no matter what impurities are introduced. For example, if you take mud and freeze it well below the melting point, and slowly warm it up, it doesn’t become soft and malleable like metals do, I don’t think. (maybe I’m wrong?)

I’m also not set on using water as the poster child. I’m sure there are many other substances out there that simply just go from being a very hard solid to being a liquid.

Why is it that pure ice doesn’t contain dislocations, but pure gold or pure silver does? What about things like diamond, which is pure carbon. Does diamond become more maleable as it heats up, being pure carbon?

Does it have to do with the fact that water has hydrogen bonds?

What if I had an ionic compound, such as a solid brick of salt. Does it become more malleable as heated, or will it go pretty much from hard solid to liquid like water does?

Here’s what I remember from the chemistry classes I took many years ago:

A substance melts when the energy from heat gets high enough to break the bonds between adjacent molecules. These bonds are the result of adjacent atoms sharing electrons. In some substances, the electrons don’t have much freedom to shift from atom to atom. In these cases, the atomic bonds are pretty much fixed in place until the energy gets high enough to break them. In metals, however, the electrons can move from atom to atom fairly easily, which means the bonds are not in any fixed places. As a metal gets hotter, the electrons move around more even before the energy becomes high enough to break the bonds. In many non-metals, though, the electrons aren’t free to move around as much, which means the substance doesn’t soften much before it melts.

This ability of electrons in metals to move freely from atom to atom is also why metals are good conductors of electricity and heat, and why they are malleable and ductile.

As I remember, ionic substances tend to melt suddenly, not gradually.

A metallic element has few electrons in its outer shells (aka valence electrons). For example, copper has only one electron in an outer shell that can take eight electrons. This means that each copper atom has a lot of spaces for free electrons, so the outer electrons in pure copper can drift pretty easily from atom to atom.

Salt consists of sodium and chlorine. Each sodium atom has a single valence electron in a shell that can take eight, while each chlorine atom has seven valence electrons in a shell that can take eight. The “hole” in the chlorine atom’s valence shell attracts the sodium atom’s single valence electron, forming a strong bond. The electron tends to want to stay with the chlorine atom, which is what makes it an ionic bond. The fact that the electron wants to stay with the chlorine atom means that it isn’t free to drift from place to place the way it would in a pure metal. So, when you heat a salt crystal, the bonds tend to stay in place until the heat energy gets high enough to break them. Since all the ionic bonds in a salt crystal are similar to each other, the crystal will melt suddenly.

Ice is a crystalline inorganic solid, with the molecules held together by hydrogen bonds and Van Der Waals forces. These bonds are only semi-rigid, so they do not transmit energy via conduction. As thermal energy is added to the ice, the hydrogen bonds break, and the ice melts at the point of application, rather than energy being distributed through the entire solid phase.

Diamond is a network covalent solid - the entire structure is made up of covalent bonds between carbon atoms. This provides rigidity (diamond is an excellent thermal conductor) but has a specific binding energy, and so does not soften until it melts or vaporises. Of course, in air the diamond will burn first. Graphite, on the other hand, has rigid planes of covalently bonded carbon with weak van der Waals binding between the planes, it has a different behavior as the layers separate/slip with heating but the planes hold together until they fall apart.

Metal bonds are rigid, and thus conduct thermal energy. They also bind over a wider range of energy (due to free electron movement). This means that as the metal has more thermal energy, the bonds are further apart but are still binding to some extent. This leads to the metal softening before melting.

Ionic solids are rigid, so they conduct, but the binding energy is very specific - ionic solids don’t generally get soft, they get hot and then melt.

There are other bonds for the solid phase - non-crystalline nonionic solids often rely on weak van der Waals forces, and behave somewhat like water, but may have higher conduction so soften somewhat. It can be a bit of a balance between melting and softening.

The wiki on Bonding in Solids is a good place to start a wikiwalk.


Once water has frozen it will increase in volume by around 8%, and as temperature further decreases the ice will contract, reduce in volume and become denser. In this respect it becomes more ‘solid’. Ice is very malleable with the addition of pressure, which allows the melting point to occur at <0 degree temperatures and makes it easily deformable.

water has its most dense at 39F. solid water will float on very cold liquid water.

if there are impurities in water they will tend to come out as the solid forms. the impurities are pushed out of the way as the crystal forms. other substances also behave this way.

if you watch thin crystals of many substances the melting is instantaneous.

solids that have metallic bonds will behave differently than covalent or ionic bonds. bonds can have different strengths depending on the substance. bonds that cross link large molecules give solidity though the number and types of them will determine how easily deformed that solid is. all these bonds are affected by temperature in different ways.

Thanks everyone! That’s all a lot of great information :slight_smile:

As noted above, ice at -1 C is a different beast from ice at -200 C. Very cold ice is harder and denser. Ice does, in fact, show the same changes that you note in metal, it’s just that we don’t get to see such a thing in our daily lives.

They do?
Then how is Dairy Queen able to sell me a slushy?

I’ve thought more about this, and if I were to make a flowchart to determine if a material softened or not, it would be like this:

Is it amorphous or crystalline?
Amorphous materials have no regular arrangement of atoms (glass and plastics) - soften on heating. (thermosets and wood don’t soften because they decompose first. What would happen if you heated a thermoset in a vacuum?) Because of the irregular arrangement, there is a variation in bond lengths, so the longer bonds melt first.

Crystalline - what kind of bonding/deformation mechanism does it have?
Metallic (dislocations) - soften slightly
Ionic/covalent (ceramics) - don’t soften
Van der waals / hydrogen bonds (water, iodine) - don’t soften?

A slushy is multi phase - small ice crystals in a liquid phase (with a lowered melting point due to solutes). The agitation stops the ice crystals from growing/fusing.

Mixtures like that can have complex behaviour based on a variety of factors.


And taste real good on a hot day, too.