The unit cell. The unit cell of a mineral is the smallest divisible unit of a mineral that possesses the symmetry and properties of the mineral. It is a small group of atoms, from four to as many as 1000, that have a fixed geometry relative to one another. The atoms are arranged in a "box" with parallel sides called the unit cell which is repeated by simple translations to make up the crystal. The atoms may be at the corners, on the edges, on the faces, or wholly enclosed in the box, and each cell in the crystal is identical. This is what was meant by an "ordered internal arrangement" in our definition of a mineral. It is the reason why crystals have such nice faces, cleavages, and regular properties.
The box of the unit cell is, in general, a parallel-piped with no constraints on the lengths of the axes or the angles between the axes. The box is defined by three axes or cell edges, termed a, b, and c and three inter-axial angles alpha, beta, and gamma, such that alpha is the angle between b and c, beta between a and c, and gamma between a and b.

The presence of internal symmetry in the unit cell may place constraints on the geometry of the unit cell. The different kinds of symmetry possible place different constraints on the unit cell geome tries giving rise to characteristic cell geometries for each of the seven Crystal Systems. These are outlined in Table 2.1.
A symmetry operation is a transposition of an object These may be of three distinct types: rotations, inversions (including roto-inversions i.e. improper rotations), or translations, or combinations thereof. We will discuss symmetry groups made up of rotation and inversion operations only which are called the point groups, each of which is one of the 32 crystal classes. We will also discuss the groups made up from all three types of opera tion which give rise to the 230 space groups.
Permissible translations are unit cell translations or fractions thereof that are consistent with the rotational symmetry (e.g. 1/2, 1/3, 1/4, and 1/6), plus combinations.
There are three common methods to graphically display orientations of vectors in three dimen sions. In order to describe faces of crystals we would like to plot the vectors normal (perpendicular) to crystal faces.
2.3.1. Spherical. The simplest to visualize is the spherical projection. In this method, the vector is merely projected vertically onto the equatorial plane of the sphere. However, this method has the serious disadvantage of compressing the low-angle vectors onto the outside of the plot.
2.3.2. Stereographic. This disadvantage
can be
avoided if, instead of projecting vertically, one projects radially to the pole of the opposite
hemisphere.
This is the standard stereographic or equal-area plot
that we will use to plot poles (perpendiculars) to faces
of crystals. The plot is also sometimes called the
"Wulff net". If the angle made by a vector with the
vertical is r, then the distance from the center of the
plot is R tan r/2, where R is the radius.

2.3.3. Gnomonic. There is also a third type of plot called the gnomonic projection in which the vector is extended till it intersects a plane tangent to the sphere at the north pole. This has the disadvantage of not being able to display horizontal vectors. It is, however, what arises naturally from x-ray and electron diffraction experiments where each node corresponds to a lattice plane in real space and results in a reciprocal lattice.
However, for our purposes of displaying the orientations of crystal faces we will use the stereo graphic projection exclusively.
Illustrated below are stereographic projections of general crystal forms that have the allowable rotation operations that are consistent with translation symmetry.
Proper Rotations:

Improper Rotations:

Each of these ten allowable rotations generates, by itself, a unique point group. In addition, there are 22 possible combinations of rotation operations, giving a total of 32 possible 3-dimensional point groups. Each point group corresponds to different crystal class. Each crystal class places constraints on the axial geometry such that each of these 32 classes may be associated into one of the 7 crystal systems, each having different constraints on the axial lengths and inter-axial angles.
In determining a point group, one must have diagnostic faces
such as the general form. For example, if you have a cube, it can occur in several point groups as
a special form. Thus there is no way
to uniquely determine the point group. If (when) you encounter this
situation in the lab, assign the
highest point group symmetry (i.e. 4/m -3
A set of elements (operations) is a group if the following properties
hold:
1. Closure: combining any two elements of the group
gives a third element of the group.
2. Association: For any three elements of the group
(ab)c = a(bc). Note: not necessarily commutative (ab = ba). If it is true for all members of the group, the group is called Abelian.
3. Identity: There is an element of the group, I,
such that aI = Ia = a for each element of the group.
4. Inverses: For each element, a, there is another
element, b, such that ab = I = ba.
The order of the group is the number of elements of
the group. We will first consider groups made
up of all allowable combinations of rotation and inversion operations
to make up the point groups in
two dimensions and in three dimensions. There are ten possible 2-dimensional
point groups and 32
possible 3-dimensional point groups. Each of these 32 3-D point groups
corresponds to one of the
crystal classes. We will then combine these with
the possible translation operations to form the 17
2-dimensional space groups and 230 3-D space groups.
At this point it is useful to develop some of the formalism of crystal
morphology (shapes). The
morphology of a perfect crystal (i.e., our wooden blocks), in general,
reflects the maximum symme
try that a crystal can have. That is, there may be portions of the
crystal structure that violate some of
the apparent symmetry, but if high-symmetry forms (crystal faces)
are present, the crystal is likely to
have high symmetry. (e. g., if the crystal is a cube, it is most probably
isometric.)
A crystal form is a crystal face plus its symmetric
equivalents. For example, a cube is a crystal
form made up of six symmetrically equivalent faces.
A special form is a crystal form that is repeated
by the symmetry operations onto itself so that
there are fewer faces than the order of the point group. The projections
of special forms or special
faces will lie on symmetry operations in our stereographic projections.
A general form is one that is not repeated onto itself
by the symmetry operations so that it has the
same number of faces as the order of the group.
Forms are either general or special. In our stereographic projections,
we will plot only the general
form because this defines the point group. In addition to being special
or general, forms may also be
open or closed.
A closed form is one that encloses a volume;
(e.g., a cube, tetrahedron, octahedron, etc). A closed
form may then be the only form present on a perfect crystal.
An open form is one that does not enclose a volume;
(e.g., prism, pinacoid, etc.). A crystal that
has an open form must have more than one form present.
2.7.1. Planes
A crystal face (or plane) cuts the crystallographic axes at
, 2, and 1. These intersections are called intercepts.
Because
symbols are cumbersome, these intercepts are in
verted and all fractions are cleared, as shown below.
, 1/2, 1/1 = 0, 1/2, 1 = (0 1 2)
These operations give us the Miller indices of any
plane. These planes may be a cleavage plane, a
crystal face, or any diffracting X-ray plane. Thus, a cube face is
(0 0 1), the octahedron (1 1 1), and a
dodecahedron (1 1 0). There may also be negative (0 0 -1) Miller
indices. Miller indices are always
in relation to the crystallographic axes, not any orthogonal system
of convenience. The general form
for Miller indices is (h k l).
For hexagonal axes the general form is (h k i l). However, you will
note in the example below that
h + k + i = 0. This is always the case so the i index is superfluous.
Hence, we can merely use (2 1 0).
-1/2, 1, 1, 0 = (-2 1 1 0) = (-2 1 0)
In general, crystal faces, diffracting X-ray planes,and cleavages
will be denoted with simple parentheses, e.g. (2 1 0). However, a crystal
form (a face plus its symmetric equivalents will be denoted
with curly brackets, e.g. {2 1 0}. Hence the cube, {1 0 0} is made
up of faces (1 0 0), (0 1 0), (0 0 1), (-1 0 0), ( 0 -1 0), and (0 0 -1).
2.7.2. Directions
Directions in a crystal are merely the vector components with respect
to the crystallographic axes
that have been reduced to the smallest whole numbers. These are given
in square brackets [1 3 0], [0
1 0], etc. In general, the [1 1 1] is not normal
to the (1 1 1), except for isometric (cubic) crystals.
Following the rules of groups, there is a limited number of ways in
which the 10 proper and im
proper rotations can be combined to form groups, that is, there are
32 possible combinations to form
groups. These are the 32 3-dimensional point groups which
correspond to the 32 Crystal Classes.
Each of the 32 crystal classes can be ascribed to one of the 7
crystal systems. The various crystal
classes are outlined in the table below.
2.5. GROUPS
2.6. Crystal Morphology
2.7. Miller Indices

2.8. The 32 Crystal Classes