Basis and Crystal structure

The ‘basis’ of crystal can be obtained when the crystal aces are chosen. Every basis in
a crystal is identical to every other in composition, arrangement and orientation. The
basis can contain one or more atoms. The position of the centre of an atom j of basis
relative to the associated lattice is, rj=xja1+yja2+zja3.

In this case, the origin may be arranged in such a way that, 0≤xj,yj,zj≤1

a2 1 a2’
2 a2” 3
a1 a1’ a1”

a2’’’ 4
a1’’’

Fig. 4: All pairs of vectors a1 and a2 are translation vectors of the lattice. a1’’’ and a2’’’
are not primitive since one cannot form the lattice translation vector T from an integral
combination of a1’’’ and a2’’’. The parallelograms 1, 2 and 3 are equal in area and any
of them could be taken as the primitive cell. The parallelogram 4 has twice the area of
a primitive cell.

The figure 5 below shows a primitive cell of a space lattice in 3 dimensions.

Fig. 5: A 3 dimensional lattice of a primitive cell and the corresponding unit cell.

The parallelepiped defined by primitive axes a, b and c is called a primitive cell, which
is a type of a unit cell. Once more, a primitive cell is a minimum-volume cell, and there
are different ways of choosing the primitive axes and primitive cell for a given lattice.
Recall that the number of atoms in a primitive cell or primitive basis is always the
same for a given crystal structure. Also, there is always one lattice point per primitive
cell. If for example the primitive cell is a parallelepiped as in Fig. 5 with a lattice point
at each of the eight corners, each lattice point is shared among eight cells so that the
total number of lattice points in the cell is 8x1/8 = 1 and the volume of the
parallelepiped is Vc=|a.bxc|

The basis connected to a primitive cell is called a primitive basis. No basis contains
fewer atoms than a primitive basis. Fig. 6 shows another way of choosing a primitive
cell and this is known as the Wigner-Setiz cell.

Fig. 6: Wigner-Seitz cell.

To obtain a Wigner-Seitz cell, choose a reference atom, join reference atom with all
nearest neighbouring atoms, bisect these lines joining the atoms. The region enclosed
by the bisectors is the Wigner-Seitz cell. All space may be filled by the Wigner-Seitz
cell just like the cells shown in Fig. 4, through translation.
Fundamental types of lattices

Crystal lattices may be mapped onto themselves by the lattice translations denoted as
T and also through other symmetry operations such as rotation about an axis that
passes through a lattice point. In this case, lattices can be found in such a way that one,
two, three four or six fold rotations map the lattice onto itself, corresponding to
rotations of 2 , 2 /2, 2 /3, 2 /4 and 2 /6 radians and by any integral multiples of
these rotations.

Mirror reflection planes

The rotation axes corresponding to these rotations are denoted as 1, 2, 3, 4 and 6 and
it’s important to realize that no lattice can be found that goes into itself under other
rotations such as 2 /7 or 2 /5 radians. Additionally, a single molecule properly
designed can have any degree of rotational symmetry, but an infinite periodic lattice
cannot. It is important to realize that although we can make a crystal from molecules
that individually have a fivefold rotation axis, we shouldn’t expect the lattice to have
a fivefold rotational axis.

Suppose the lattice shown below has m-fold rotational symmetry about the axis
passing through the lattice point and perpendicular to the plane of the paper.

When we carry out a rotation of angle α= 2π/m we reproduce the same lattice. Thus
in the diagram, A' and B' are also lattice points just like A and B. Hence, it can be
written A’B' = n*AB where n is an integer. Thus,

A'B' = n*AB
Hence, A’X + XY+ YB' = na

acosα +a+ acosα = na

cosα = (n-1)/2 ...................(1)

Since -1 ≤ cosα ≤ 1 this mean that n cannot be any random integer, but should possess
some specific values shown in the below table.

n cosα α (degrees) Symmetry

0 -0.5 120 3-fold

1 0 90 4-fold

2 0.5 60 6-fold

3 1 360 1-fold (identity)

-1 -1 180 2-fold
So from this table, we can see 1,2,3,4,6-fold symmetry exist for crystal but 5-fold
symmetry does not exist.

For example, Fig. 7 shows what happens if one tries to construct a periodic lattice
having a fivefold symmetry. Here, the pentagons do not fit together to fill all space,
showing that we cannot combine fivefold point symmetry with the required
translational periodicity.

Fig. 7: Pentagons do not fill all space.

Another expression used in crystal lattices is lattice point group, which implies the
collection of symmetry operations which, when applied about a lattice point, carries
the lattice point into itself. For example, mirror operations about a plane through a
lattice point.

An inversion operation is composed of a rotation of followed by a reflection in a

plane normal to the rotation axis; the total effect is to replace a vector r by –r.
The symmetry planes of a cube are shown in Fig. 8.

Top panel shows plane symmetry parallel to the faces of the cube, middle and bottom
panels show a diagonal planes of symmetry.

The figure shows three tetrad axes of a cube originating from the face centres, four
triad axes originating from the cube corners and six diad axes originating from the
centres of the edges of the cube.
Two dimensional lattice types

A lattice that is drawn using arbitrary vectors a1 and a2 as shown in figure 4 is called
an oblique lattice and its invariant under rotations of and 2 about any lattice point.
However, special lattices of the oblique type can be invariant under rotations of 2 /3,
2 /4 or 2 /6 or under mirror reflection.

There are five distinct lattice types in two dimensions i.e the oblique lattice and four
other special lattices shown in figure 9.

Fig. 9: The five different types of Bravais lattices: (a) Hexagonal, (b) Square, (c)
Rectangular, (d) Centered rectangular, where the dashed lines form a rectangle, (e)
Oblique.

Bravais lattice is the phrase used for distinct lattice type and hence we have five
Bravais lattices in two dimensional space as shown in Figure 9. Bravais lattices are
named after the French crystallographer – August Bravais who showed in 1848 from
geometrical considerations that there are only 14 distinct ways in which similar points
can be arranged ina 3-Dimensional space. On the other hand, a unit cell is the smallest
portion of the crystal lattice which when repeated in different directions generates the
entire lattice.
Three dimensional lattice types

There are fourteen different (distinct) lattice types in three dimensions. These are
shown below.
Figure 10: 3-Dimensional Bravais lattices

Some unit cells contain only points at the corners and these are referred to as simple.
Others contain points at the centre of the body or at the centres of the faces. Those that
contain points at the body centre are called body centred cubic (BCC) structure while
those with atoms at the face centers are called face centred cubic (FCC) structures. The
size of the edge of a body-centred tetragonal lattice is 1/√12 times of the original
lattice. It is also important to recall that for the lattice point in space, a lattice point
doesn’t just represent a single atom but a group of atoms.

The primitive cell of a BCC lattice is shown in figure 11, while the primitive translation
vectors a1, a2 and a3 are shown in figure 12.