Ice 1: introduction

Ice isn't usually considered to be a mineral, but it has a defined formula and 3-D structure and is naturally occurring so it is really. Ice crystallises in many different forms in snowflakes, most of which carry hexagonal symmetry. In 1951 the International Commission on Snow and Ice defined seven principal snow crystal types as plates, stellar crystals, columns, needles, spatial dendrites, capped columns, and irregular forms. These classes can be further subdivided: in 1966 Magono and Lee defined 80 different snow crystal types. Artificial snow, on the other hand, is created too quickly for intricate regular crystal structures to form. Other forms of solid precipitation have their own structures and yet more structures can be grown under carefully controlled conditions in the laboratory.

A fragment of the bulk structure of ice is shown to the left. The oxygen atoms are red and the hydrogens are white. Covalent bonds are shown in green and hydrogen bonds in blue. Each oxygen atom in the bulk is connected to four hydrogen atoms: two via covalent bonds and two via hydrogen bonds from its lone pairs. This hydrogen-bonding network therefore imposes 4-coordinate tetrahedral symmetry around each water molecule. Such a low coordination number leads to very inefficient packing - recall that there is a coordination number of 12 in close packing. This explains the anomalously low density of ice, water being one of the few materials whose solid phase is less dense than the liquid phase.

The structure is based on a hexagonal repeating unit and is closely related to the wurtzite structure. Indeed the oxygen atoms in ice occupy the positions of the zinc and sulfide ions in wurtzite. Like wurtzite there are hexagonal channels, which can easily seen by rotating the structure on the left. Chair- and boat-type rings can also similarly be observed in the structure.

Go to page 2 to look at the unit cell within the bulk structure of ice.

Page 2 Page 3