inorganic chemistry - How do copper sulphate and the like change crystal structure on adding water of hydration?

Copper sulphate, in its hydrated form, is crystalline, whereas the anhydrous form is amorphous.

Gypsum has a similar story-- on heating the crystalline dihydrate we get an amorphous hemihydrate. (Gypsum in fact has two different crystalline forms of the dihydrate, and has an anhydrous form as well).

I've never really grasped how the water makes it crystalline. I suspect it has to do with the relative sizes of the compounds being re-compensated by ligands, but I'm not sure. How does the water affect/create the crystalline properties?

Answer

I learned something in trying to answer this question since I've never quite understood what the water was doing in hydrated crystals either. The waters of hydration help to stabilize the crystalline form by holding the crystal together with hydrogen bonding. Water incorporated into the crystal allows the negative dipole of the water to partially diffuse the positive charge and the positive dipole of the water to partially diffuse the negative charge, and would help to minimize repulsive forces between ions of like charge "close" in the crystal. Speaking of $\ce{Cu(SO4)2 .5H2O}$...

Copper is surrounded by six oxygen atoms, provided by two different sulfate groups and four molecules of water. A fifth water resides elsewhere in the framework but does not bind directly to copper.....Water of crystallization is stabilized by electrostatic attractions, consequently hydrates are common for salts that contain +2 and +3 cations as well as -2 anions. 1 (Ref)

A more interesting reference that takes some looking at to get the sense of is given below. $\ce{Fe2(SO4)3.xH2O}$, with iron with a 3+ charge, forms various hydrates with x = 6, 7, 9, 10. (Disclaimer....actually this is the crystal structure for coquimbite with has an occasional Al for Fe substitution thrown in, but the principle is the same)

In this picture, the sulfate ions are shown as tetrahedra and the $\ce{Fe^3+}$ ions are shown as octahedra (presumably because $\ce{Fe^3+}$ ions have 6-coordination sites). The crystal is formed from "clusters of 6 sulfate tetrahedra and 3 iron octahedra which share only corners" and there are two clusters per unit cell. Each of the clusters is linked to the others through hydrogen bond provided by the incorporated water, and parallel chains are held together only through hydrogen bonds.

The individual chain segments are linked through hydrogen bonds only. The geometrical arrangement of the chains gives rise to "channels," ... which are occupied by water molecules linked to the chains by hydrogen bonds.

So, you can see how removing the waters of hydration would cause the crystalline structure to break down and result in an amorphous substance.

Further information from the paper: Coquimbite has $\ce{.9H2O}$. The crystal structure shows that, of the 3 Fe found in a cluster, one Fe is associated with 4 oxygens from sulfate groups, the second Fe is surrounded by 3 oxygens from sulfate groups and 3 water molecules, and the third Fe (or Al) is surrounded by 6 water molecules, so the octahedral representation does make sense.