UNIT 2 CRYSTAL GEOMETRY

Structure
Introduction
Geometrical Arrangement in Crystals
Bravais Lattice
Three-dimensional Bravais Lattices and Crystal Systems
Symmetries in Crystals
Rotational Symmetry
Translational Symmetry
Reflection Symmetry
Inversion Symmetry
Symmetry Groups
Cubic Crystals
Symmetry in Cubic Crystals
Types of Cubic Crystals
Miller Indices
Polycrystalline and Amorphous Solids
Summary
Terminal Questions
Solutions and Answers

2.1 INTRODUCTION
Solids occupy a prominent place in our everyday life because of their characteristic
physical properties. In order to understand these properties, we need information
about the arrangement of atoms in a solid. From your school physics you may recall
that depending on their atomic arrangement, solids can be broadly categorised as
crystalline and amorphous. In crystalline solids, atoms are arranged in a regular and
periodic pattern. Moreover, the arrangement is unique to every material. When the
periodicity of the pattern extends unbroken throughout a piece of material, it is said to
be a single crystal. Solids composed of an aggregate of such crystals are called
polycrystalline solids. Obviously in polycrystalline substances, the periodicity in
structure is interrupted. Amorphous solids lack a regular atomic arrangement.
From experience we know that various materials made of the same element can exist
in very different solid forms and exhibit vividly different physical properties. For
example, black soot, diamond and graphite are different forms of carbon. But their
mechanical, thermal, optical and electrical properties are completely different. The
diamond is hard, optically transparent and an electrical insulator; graphite is soft,
opaque, black in colour and good conductor of electricity. This is due to different
atomic arrangements in these materials. Whereas soot is amorphous, graphite and
diamond are crystalline. But even graphite and diamond have different structures.
Such observations have always fascinated human intellect. Though obtaining insight
into internal details of crystal structures is an area of active and fertile research, we
shall confine ourselves only to basic ideas.
In Sec. 2.2 you will learn about geometrical arrangement of atoms in crystals in two
and three dimensions and their classification. In Unit 1 you have learnt various types
of symmetries observed in the nature. We also briefly touch upon the symmetries
observed in molecules. J i Sec. 2.3 we will discuss these in detail with particular
reference to solid crystals. Since cubic geometry is the most widely observed structure
of crystalline solids, we take a closer look at this geometry in Sec. 2.4. The atoms in
crystal can be visualised as arranged in planes inclined at different angles and having
different spacings. The physical properties of the material can be different along
different planes. Therefore, it is important to index crystal planes. This is the subject
 Crystal Geometry
of discussion of Sec. 2.5. Polycrystalline and amorphous solids are discussed in
Sec. 2.6.

Objectives
After studying this unit, you should be able to:

classify solids as crystalline, polycrystalline or amorphous;

draw various crystal geometries and categorise them in the Bravais lattices;
,, determine symmetry elements of a crystal from its geometry;
understand the correlation between the density of material and its crystal
structure; and
depict various planes in a cubic crystal and determine their Miller indices.

2.2 GEOMETRICAL ARRANGEMENT IN CRYSTALS

All matter is composed of atoms. When we wish to determine the structure of a solid
material, we need to know the arrangement of atoms in it. In other words, we must
identify every atom and specify its respective position. It is however important to
know that atoms in a solid do not behave exactly like when they are isolated. This is
because in a solid, each atom finds itself interacting with other atoms and electrons in
the outermost shell may be exchanged or shared. You will learn the details of the
nature of interacting forces in Unit 5 of this course but it suffices to state here that
solids owe definite shape to the strength of interatomic forces. Moreover, a crystalline
solid is characterized by long-range order; a recurring pattern in the arrangement of
atoms. This pattern is called the crystal structure. Amorphous solids have a regular
but non-recurring pattern and are characterized'by short-range order.

(c) (dl
Fig.Z.l: a) The structure of a crystal can be obtained by repetition of a single unit cell; b) unit cell
after it has been moved to a new position; c) smaller hnd larger unit cells can also portray
the crystal structure; and d) a different shape of unit cell

To describe the structure of a crystal, it is not necessary to specify positions of all

atoms in it. The regularity and orderliness characterizing atomic arrangement in a
crystal simplify the problem considerably. We can describe the orderliness by
repeating some pattern. All that we have to do is to select an appropriate unit
consisting of a few atoms, which when repeated forms the entire crystal. This unit is
referred to as unit cell. Fig.%.1 shows a two-dimensional crystal structure. (Strictly
Crystal Structure speaking, there exists no two-dimensional crystal.) In this example, the entire crystal
can be built up by fitting together identical hexagonal units side by side, like a mosaic
pattern. The unit cell can be chosen in different ways; a few possible unit cells are
shown in Fig.2.1. Note that complete information about structure of a crystal is
contained in the unit cell and its constituents. You may be tempted to think that the
choice of unit cell is unique. However, the fact is that any structure can be described
by many different types of unit cells. We can obtain a different unit cell by moving it
about (Fig.2.lb) or changing the size (Fig.2.1~)or shape (Fig.2.ld). And any one of
these serves the purpose equally well. The only requirement is that the unit cell
should reproduce the crystal structure correctly when it is repeated.
Since we are only concerned with the geometrical structure of a crystal, for simplicity,
we replace each constituent ion/atom by a geometrical point located at its equilibrium
site. In a material made of one element, diamond say, we deal with only single type of
atoms (carbon). As you know, most materials consist of more than one type of atoms.
In such cases, we replace a group of atoms with a single point. For example, in the
crystal of sodium chloride, we replace both sodium and chlorine atoms together with
one point. In this way, we obtain a network of points in space as representation of the
particular crystal structure. Such a grid is known as lattice of the crystal and the
points where atoms reside are referred to as the lattice points. When we wish to
reconstruct the solid crystal, we attach the atom or the group of atoms, called basis, at
every lattice point. That is, crystal structure comprises of lattice plus basis. For
example, to form the crystal of graphite or diamond each point is replaced by a carbon
atom, but in NaCI, the group of Na' and C1- ions forms the basis and replace just one
lattice point. This is illustrated in Fig.2.2.

X X X X e0 e0 e0 e0 eO

X X X X lO P 8 .0

Fig.2.2: Fprmation of a crystal by combining the lattice with basis

In 1848, Bravais put forward the concept of the space lattice to describe the structure
of crystals. You will learn about it in detail now.

2.2.1 Bravais Lattice

A space lattice comprises an infinite number of points in space such that the
arrangement about a given point is the same as that about any other point. It means
that in a Bravais lattice, all lattice points are equivalent. That is, all atoms are of the
same kind. Sometimes the basis will have more than one type of atoms in it. In that
case the surrounding of one type of atom will not be same as that of other type of
atom i.e. we get non-equivalent sites in the lattice. Such a lattice is called non-Bravais
lattice.
Now refer to Fig.2.3. It shows periodic arrangement of points in 2-D, i.e. a space
lattice restricted in the plane of the paper. This is an oblique Bravais lattice and is a
general case of two-dimensional lattice.
To specify the positions of lattice points, we choose the origin of coordinates at a
particular point 0, say. Then we can write position of lattice point P as:

RP=mal+na2,

where al and at are two vectors inclined at an angle 4 and (m, n) is a pair of integers.
 "'
Crystal Geometry

- --- - -----
a a -- ---9 a
a,
I

a a a a al (a)
a, A square lattice is defined by
a, = q and I$ = 90"

!@ 1
Fig.2.3: 2-D oblique Bravais lattice
-- - - - - - - a
The non-collinear vectors a, and az form a set of basis vectors for the lattice. All the
vectors defined by Eq. (2.1) are called the lattice vectors. By referring to Fig.2.1, you '2

will recall that the choice of the unit cell is not unique. This is also true for basis a
a,
vectors. Depending on the magnitudes of basis vectors and the angle 4 between them, (b)
we obtain special cases of 2-D Bravais lattices (Fig.2.4a-c). For example, a, = a2 and A rectangular lattice is defined
4 = 90. results in a square lattice, whereas a, # a2and 4 = 90. leads to a rectangular *
by a1 az and I$ = 90".
lattice. Similarly, for a hexagonal lattice, the unit cell is a parallelogram with a, = a2
and #I = 120".
Note that in all the cases shown in Fig 2.4, a parallelogram or a rectangle represents
the smallest unit that may be repeated to construct the complete lattice. Here, each
unit cell has one point at each of the four comers. But each point is shared by 4
adjoining cells, so that there is only one lattice point per cell. Such unit cell is called
the primitive unit cell. The vectors a, and az encompassing the primitive cell are
called primitive lattice vectors.
(c)
You may now like to answer an SAQ. A hexagonal lattice is defined
byal=a2and+=1200.,
SAQ 1
Fig.2.4: Special Bravais
Refer to Fig.2.5. The rectangle ABCD can map the complete lattice by repeating itself. lattices in 2-D
Is it a primitive cell?
Spend
2 min.

Fig.2.5: Primitive and non-primitive unit cells in a Bravais lattice

Hint: Calculate the number of lattice ~ o i n t contributed

s bv the rectangle.

On answering the above SAQ, you must have observed that the unit cell ABCD
defined by basis vectors a; and a; is not primitive. In fact, many a time we prefer a
non-primitive unit cell because it may exhibit the lattice symmetry more clearly. This
Crystal Structure is shown in Fig.2.5. Here S1 represents a non-primitive unit cell while Sz is a primitive
unit cell.
To construct a primitive unit cell for any given lattice structure, the most common
method is due to Wigner and Seitz. It consists of following steps:

choosing a point, say 0 in Fig.2.6, in the lattice as centre of the primitive cell;
joining it with all the nearest neighbouring atoms (P, Q, R, S, T, U); and
drawing perpendicular bisectors of lines joining nearest neighbours (AB, CD, EF,
GH, IJ, KL).
The area enclosed by these bisectors (MI M2 M3 M4 M5 M6) gives the W igner-Seitz
unit cell. This cell contains only one lattice point (at the centre) and hence this is a
primitive unit cell.
a

Fig.2.6: Wigner-Seitz construction of a primitive unit cell

So far we have confined ourselves to two-dimensional lattices. But as you now know,
exact understanding of actual solid crystal structures demands knowledge of atomic
arrangement in three dimensions. Though statements made for the 2-D case can be
readily extended to 3-D, the details are somewhat involved. You should concentrate
while studying the following paragraphs.
In a three-dimensional crystal lattice, the position of any point M is represented by
three non-coplanar basis vectors al, a2, a3 and its lattice vector can be expressed as
RM=pal+qaz+ra3, (2.2)

where @, q, r) is a set of integers, which can take positive, negative or null values.
Note that the basis vectors may not be orthogonal.
A typical unit cell in 3-D lattice is shown in Fig.2.7. It is a ~arallelepipedwhose sides
Fig.2.7: Unit cell in a 3-D are basis vectors a,, a2, a3 and a, p, -y are the angles between them. As in 2-D case, the
lattice choice of primitive unit cell in 3-D is also not unique. Moreover, all the Bravais
lattices may not be primitive unit cells; some of them may contribute more than one
lattice point to the lattice.

2.2.2 Three-dimensional Bravais Lattices a n d C r y s t a l Systems

All crystalline solids can be categorised into fourteen different Bravais lattices. You
may be wondering as to what makes such a small number of Bravais lattices possible.
It is the condition of translational symmetry along with other point symmetries!
 Crystal Geometry
Depending on the relative lengths of a l , a2 and a3 and the angles between them, these
lattices are grouped into seven crystal systems: triclinic, monoclinic, orthorhombic,
trigonal, hexagonal, tetragonal and cubic. The unit cells of each of these systems have
their characteristic shape and symmetry. These systems are shown in Fig.2.8 in the
increasing number of symmetry elements governing them.
Table 2.1 lists the Bravais lattices associated with the seven crystal systems and
characteristics of their unit cells in terms of relationships between a , , az, a3and a , P, y
in each case. Typical examples of elements/compounds falling in every category are
given in the last column.

Table 2.1: Seven crystal systems divided into 14 Bravais lattices

S. Crystal system Bravais lattice Unit cell Typical materials .

No. characteristics

1. Triclinic Primitive (P) al # a2# a3 Potassium

a#P#y dichromate
2 Monoclinic Primitive (P) at # a2 # a3 Gypsum, Borax
Base-centred (C) a=p=90°#y
y # 120"

3. Orthorhombic Primitive (P) al # a2 # a3 Ga, C1, I, Br

Base-centred (C) a =p = = 900
Body-centred (I)
Face-centred (F)

4. Trigonal Primitive (P) al = a2 = a3 As, Sb, Quartz,

(Rhombohedral) a =p = y # 900 Calcite (marble)

5. Hexagonal Primitive (P) a, = a2 # a3 Mg, Zn, Ti, Cd, Zr,

a = p = 90' Be, Si02
y = 120"

6. Tetragonal Primitive (P) al = a2 # a, Zircon, Ti02, In

Body-centred (I) a =p = y = 900

7. Cubic . Primitive (P) al = a2 = a3 Na, Fe, K , Cr, Ni,

Body-centred (I) a =p= y = 900 CsCl, NaCl, KBr,
Face-centred (F) MnO, MgO, Al, Cu

Note that a Bravais lattice can be simple (primitive), body-centred, base-centred or

face-centred. In a simple lattice, the unit cell has points only at the corners while in a
body-centred lattice, the unit cell has one additional point at its centre. Similarly, in
the base-centred lattice, the unit cell has additional points at the centres of one pair of
either of three opposite faces. You may note that two sets of opposite faces cannot be
centred simultaneously. This is because compatibility with translational symmetry
will be disturbed. In a face-centred lattice, the unit cell has six additional points; one
on each face.
Let us now closely examine the crystal systems and the associated Bravais lattices.
1. Triclinic: This is a general crystal system in which the angles between all the
lattice vectors are unequal, i.e. a # P # y and their lengths are atso not equal, i.e.
al # a2 # a3 (Fig.2.8a).
2. Monoclinic: In this crystal system, two of the three basis vectors are not
perpendicular to each other but the third is perpendicular to both of them, i.e.
a = p = 90° # y and y #120°. All basis vectors have unequal lengths,
Crystal Structure

(a)Triclinic (b) Simple (c) Base-centred

monoclinic monoclinic

(d) Simple (e) Base-centred (f) Body-centred (g) Face-centred

orthorhombic orthorhombic orthorhombic orthorhombic

u
(h) Trigonal (i) Hexagonal (j) Simple (k) Body-centred
tetragonal tetragonal

(I) Simple cubic (m) Base-centredl (n) Face-centred

cubic cubic
Pig.2.8: Unit cells of seven crystal systems in 3-Dwith corresponding 14 Bravais lattices

i.e. a,# a*# a3.Monoclinic lattices may be primitive or base-centred, as shown in

Fig.2.8 b & c, respectively.
3. Orthorhombic: In an orthorhombic crystal, the basis vectors are orthogonal,
but their lengths are different along all the three axes, i.e. a = P = y = 90"
and a,z a2 # q .The orthorhombic lattices may be primitive, base-centred;
body-centred or face-centred, as shown in Fig.2.8d-g, respectively.
4. Trigonal: In a trigonal, also called rhombohedral, crystal system, all three basis
vectors have the same magnitude. The angles between each pair of vectors are .
also equal but not 90°, i.e. a = P = y + 90" < 120' (Fig.2.8h).
5. Hexagonal: In a hexagonal crystal system, two crystal axes are at an angle
of 120°, i.e, y = 120°, while the third axis is perpendicular to both of them, i.e.
a = p = 90°. Moreover, the magnitudes of the basis vectors separated by I 20° are
equal but different from the third vector, i.e. a, = a2 # a3 (Fig.2.8i).
6. Tetragonal: In a tetragonal crystal system, the basis vectors are perpendicular to
each other i.e., a = P = y = 90° but the lengths of the primitive lattice vectors are
equal only along two axes, i.e, a, = a2 # as. The tetragonal lattices may be
primitive (simple) or body-centred, as shown in Fig.2.8 j & k, respectively.
7. Cubic: In a cubic crystal system, basis vectors are orthogonal to each other, i.e.
a = p = y = 90' and of the same length, i.e. a , = a2 = a3. Cubic lattices may be
simple cubic (sc), body-centred cubic (bcc) or face-centred cubic (fcc), as shown
in Fig.2.81-n, respectively. For sc structure, the unit cell has 8 atoms, each situated
at the corner of the cube. Since each corner is shared by 8 adjoining cubes, each
corner atom contributes (1/8) atom to the unit cell. SQ,the total contribution of sc
unit cell is 8 x (1/8) =1 and we can say that the unit cell of an sc lattice is a
primitive unit cell.
The unit cell of a bcc lattice has 8 atoms on its corners and one atom is situated in
its centre so that the total number of atoms per cell is two. On the other hand, the
unit cell of an fcc lattice has 8 atoms at its corners and one atom each at the centre
of its six faces. The face-centred atoms are shared by two adjoining unit cells. Can
you guess, how many atoms does each face effectively contribute to the cell? It is
only one-half. Hence, the number of atoms in the unit cell of an fcc lattice is four:
8 x (1 / 8) + 6 x (1 1 2 ) = 4 . From the above discussion we can conclude that the unit
celb for bcc andfcc lattices are nsn-primitive.
You may now like to answer an SAQ.

SAQ 2 Spend
3 min.
Determine the number of atoms in a unit cell of base centred orthorhombic lattice.

To construct the primitive cell of these lattice structures, we adopt the same procedure
as described for 2-D case of Wigner-Seitz Cell.
Before we proceed further, it is appropriate to introduce other important terms used in
the theory of solids.
One of the first things that you may like to know about a unit cell is its size. It is
expressed in terms of lattice constants, which specify the magnitude of basis vectors,
al, a2 and a3.Obviously, cubic and trigonal crystals have single lattice constant:
a (= a, = a2 = a3),and tetragonal and hexagonal crystals have two lattice constants.
How many lattice constants are required for orthorhombic, monoclinic and triclinic
crystals? Will the number be three?
In a solid, every atom/ion is surrounded by many other atoms/ions. The number of
equidistant nearest neighbours of an atom/ion is known as coordination number. The
coordination number for sc structure is 6. Similarly, you can easily verify that the
coordination numbers for bcc and fcc systems are 8 and 12, respectively.
Since atoms have finite size, they occupy some of the space available in the unit cell.
The ratio of volume occupied by the atoms in a unit cell to its actual volume is known
as atomic packing fraction (APF). As you proceed in this course, you will learn that
knowledge of APF allows us to calculate density of a material using lattice constants.

2.3 SYMMETRIES IN CRYSALS

In Unit 1 you have learnt that nature exhibits symmetries in its various forms and
naturally occurring crystals are no exception. You now also know that any crystal
structure can be obtained by repeating a unit cell. So it is natural to think that some
symmetry is inherent in them. In this section we will closely examine symmetries
observed in crystal structures. But before you proceed, it will be worthwhile to
carefully go through Unit 6 of Block 1 of PHE-14 course on Mathematical Methods in
Physics-111.
Crystal Structure
2.3.1 Rotational Symmetry
From Unit 1 you may recall that a body is said to possess rotational symmetry ifit
remains invariant under rotation about an mispassing through it. The number of
times the structure coincides with itself in one complete rotation (360") determines the
folds of rotation. In practice, a triangle, square, rectangle, parallelogram/rhombus and
hexagon are the only basic shapes. These on repetition, can fill the 2-D lattice space
completely without leaving any gap in-between. ~ i ~ . 2 . 9 a shows
-e the patterns
obtained by doing so.

Fig. 2.9: Shapes that fill in the complete 2-D lattice space by repeating themselves

We know that when we rotate a body about a vertical axis, it comes back to its
original shape in one complete revolution. This is said to be one-fold rotational
symmetry and every structure possesses it.
Youcan easily convince yourself that only I-, 2-, 3-, 4- and 6-fold rotational
symmetry operations are possible in 2-D lattices. That is to say, we can reproduce an
infinite-lattice by repetition of units having these rotational symmetries only. This
means that Sfold, 7-fold, and &fold symmetries are ruled out and it is not possible to
fill the whole lattice space, without leaving any gap, with a pentagon, a heptagon or an
octagon, though the unit cell itself can be highly symmetric. You can verify this by
performing the following activity.

Activity 1 .

Construct 10 regular pentagons on a card sheet and arrange them next to each other
with edges touching. Is the space completely covered to make a closed figure like the
ones shown in Fig.2.9?

It is also possible to prove this result mathematically. To this end, let us consider a
If rotation by I),,results into linear array of lattice points P, Q, R, S on a lattice with lattice spacing a, as shown in
invariant structure, rotation Fig.2.10. Suppose the lattice has n-fold rotational symmetry. Then, if we rotate
by - d also results into segments PQ and RS through angles - $, and $, respectively, we obtain segments P'Q
invariant structure. and S R . Note here that P' and S are also lattice points and hence the distance between
them will be an integral multiple of a, equal to say ma. We draw perpendiculars QX
and RY fiom Q and R, respectively on the line P'S'. From AP' XQ, we can write

P'X
cost$, =-
P'Q '
 Crystal Geometry

P Q R S
4 a
b
Fig.2.10: Construction for determining allowed rotation symmetries

But P'Q = a, since P'Q is simply rotated PQ. Therefore, we have

P'X =aces$,.

Similarly, we can show that

S'Y = acos4,.

From the figure you will note that

P'S' = P'X + XY + YS'.

On substituting the values of P'X and Y S in this expression, we obtain

since XY= a and P'S' = ma.

We can rewrite this result as

Here N = m - 1 is an integer. Since cos 4, can take values between -1 and 1, N can
take values -52, -1,0, 1 and 2 only. The angles 4, corresponding to these values gf N
are 180°, 120°, 90°, 60" and 360°, respectively. What can you conclude about the
folds of rotational symmetry from these values of $,? These values of 4, lead to 2,3,
4 , 6 and 1-fold symmetry. Hence it is clear that any rotational symmetry other than
1,2,3,4 and 6-fold is forbidden, ifa lattice has translational periodicity a.
In 3-D Bravais lattices, the rotational axis passes through the crystal. For example, in
bcc structure, the body diagonal is a 3-fold rotational axis, i.e. rotation around body
diagonal by 120" leaves the structure invariant.
2.3.2 Translational Symmetry
In the last section we have said that a crystal lattice could be obtained by repeating a
small geometrid pattern. This definition has built within it the concept of translational
symmetry. We can look at this aspect in another way. All atoms in a crystal are
equivalent and their surroundings are exactly the same in an infinite lattice space. If
you concentrate on one of these atoms, it is not possible to tell where you are by
looking at the neighbouring atoms. And even if you move to,the next atomic site or
Crystal Structure the next, you cannot make out the difference. This invariance to translation within the
crystal constitutes\translationalsymmetry. Any point in the lattice can be mapped by a
translation defined bypal + qa2+ ra3, where a,, a2, a3 are basis vectors andp, q, r are
integers taking positive, negative or null value. Note that all lattices necessarily
possess translational symmetry. In other words, an array of points in space can be
A6 called lattice only if it possesses translational symmetry.

i,,' y reflection symmetry operation, the object remains invariant under reflection in a
D c plane, called, mirror plane. For example, in the orthorhombic crystal ABCDEFGH
shown in Fig. 2.1 1, the reflection o'f upper part of the structure in the plane leaves it
Fig. 2-11: Reflection symmetry invariant. You can identify many such planes for any given structure.
in an orthorhombic
crystal structure Before proceeding further, you may like to answer an SAQ.

3 min.
Draw all the mirror planes of symmetry for the crystal system shown in Fig.2.11

2.3.4 Inversion Symmetry

Yet another symmetry operation is inversion symmetry. A system is said to possess
inversion symmetry if it remains invariant under inversion around a point. The cubic
structure of Fig.2.12 exhibits this property about point 0. It means that if we choose a
frame of reference with centre at 0 , then for every point (x, y,'z),there exists an
equivalent point defined by (-x,- y,- 2 ) .
1

Inversion

Fig.2.12: Inversion symmetry o f a cube

Note that points A, B, C,D, E, F, G and H in Fig.2.12 are equivalent. With inversion
operation, every point in the structure is translated along the body centre. Only the
Fig.2.13: A hexagonal point 0 does not move in space. For this reason, it is called the centre of inversion
Bravais lattice symmetry.

Spend
2 rnin.
Refer to Fig.2.13. The upper and lower faces A and B are regular hexagons. Does the
crystal have inversion centre? Justify your answer.

So far we have confined ourselves to individual symmetry operations. But many a

 Crystal Geometry
2.3.5 Symmetry G r o u p s
In the preceding sections we have used simple examples to demonstrate various
symmetry operations. The individual operations or their combination that leave an
object invariant is called the symmetry group of the object. The symmetry elements
such as axes (in case of rotation), planes (in case of reflection) and centre (in case of
inversion) constitute macroscopic symmetry elements. These occur singly or in a
group in crystals and define symmetry and physical properties of the crystal.
The symmetry groups can be classified on the basis of the nature of operations as
point group and space group. A point symmetry operation (rotation, inversion and
reflection) is a symmetry operation carried out with respect to a fixed point in space.
In other words, the point does not move during the operation. The space group is
attributed to crystals where translational symmetry plays an important role, as in
constructing the entire crystal by translating the basis vectors in relevant directions.
In 2-D space, there are five point groups consistent with allowed n-fold rotational
symmetries in crystals C, (viz. C1, C2, C3, C4,and C6) and five point groups with n-
fold rotational symmetry and reflection denoted by C,,. As there is no separate
inversion operation in 2-D, the number of point groups in 2-D Bravais lattices is
limited to 10. However, in 3-D, the combinations of inversion, rotation and reflection
lead to 32 different point groups.
The details of the point groups
When the translational symmetry is taken into account, the space groups replace the associated with 3-D Bravais
point groups. In 2-D there are in all 17 space groups while in 3-D there are 230 space lattice are discussed in detail
groups. in Unit 6 of PHE-14 course.

Having learnt about the symmetries involved in crystal structures in general, you may
now like to know more about the most commonly occurring crystal system viz. cubic
crystal. This is the subject of discussion of the following section.

2.4 CUBIC CRYSTALS

The cubic structure is the most familiar, highly symmetric and simple to visualise. We
now take a closer look at it.

2.4.1 Symmetry in C u b i c Crystals

In Sec. 2.3 you have learnt about different symmetry operations observed in crystal
structures. A cubic structure exhibits all possible symmetry operations: reflection,
inversion, rotation, inversion-reflection and inversion-rotation symmetries. You will
learn about these now in details.

a) Reflection symmetry: An object is said to be mirror-symmetrical if one-half of it

is mirror reflection of the other half about a plane. From Unit 1, you may recall
that a cube has two types of mirror planes (Fig.l.5):

three planes passing through the body centre of the cube and parallel to two s f
its faces. One such plane is shown in Fig. 1.5a.
six planes passing through the body centre and diagonally opposite edges.
One such plane is shown in Fig.l.5b.

b) Inversion symmetry: Like reflection symmetry, you have learnt about inversion
symmetry of a cube in Unit 1. You may recall from Fig.2.12 that when a cube is
inverted around its body centre, it remains invariant.
c) Rotational symmetry: When an object coincides with itself exactly on being
rotated through an angle 8, (= 2 d n ; n = 1,2,3 ...) about an axis passing through it,
Crystal Structure it is said to have n-fold symmetry. A cube has three types of rotational symmetry
axes as shown in Fig. 2.14:

Identity operation is a ~ x e passing

s through the centres of opposite faces exhibit 4-fold symmetry
rotation through 360°, which and there are three such axes (Fig.2;14a).
leaves every structure
inuariant. All structures 3-fold axes passing through the diagonally opposite comers and body centre
possess identity operation. of the cube. There are in all four such axes (Fig.2.14b).

Six 2-fold symmetric axes passing through the centres of diagonally opposite
edges (Fig.2.14~).
, I

(a) (b) (c)

Fig.2.14: Axes of rotational symmetry in a cube

d) Inversion-reflection symmetry: As the name suggests, this symmetry combines

two operations. All the reflection operations followed by inversion leave the cubic
structure invariant.
e) Inversion-rotation symmetry: Like inversion-reflection symmetry, all rotational
operations followed by inversion leave the cube invariant.
Having learnt about the symmetries characterizing a cubic crystal, it is important to
develop appreciation for actual crystals. To this end, you must be familiar with at least
someof the better known crystals. Since cubic crystals are most common, let us now
learn more about various types of cubic structures.

2.4.2 Types of Cubic Crystals

In a cubic crystal structure, three types of atomic arrangements are possible: simple
cubic, body centred cubic and face centred cubic. Do you know what differentiates
them? Let us learn about these now.

a) In a simple cubic (sc) structure, the atoms are situated at every comer of a cube
(Fig.2.15a). Though this structure is very simple, it is not very common in natural
crystals. Only polonium and that too in a certain temperature range exhibits this
structure.

Fig.2.15: a) Simple cubic lattice; and b) nearest neighbour atoms in sc structure

 Crystal Geometry
You know that in a sc structure, every cube contributes only one atom per unit
cell. The basis vectors defining a simple cubic structure are:

al = a i ,

and

- a a

where i, j, k are unit vectors.

Since each edge of a cube has the same length, say a, the volume of the unit cell is
a3. Suppose the neighbouring atoms touch each other (Fig.2.15b). Then by

occupied by atoms in a unit cell is

3
xa 3
Vs = ( 8 x ~ ) ~ u 3 = ~ n =( T~ . ) (2.5)

And atomic packing fraction, defined as the ratio of the volume occupied by
atoms in a unit cell to the volume of unit cell, is

xa 3
-

-
A P F =a ~ - -6 --0.52. (2.6)

That is, nearly hawthe volume of unit cell of an sc lattice is occupied by its atoms.

b) A body centred cubic (bcc) structure has 8 comer atoms and one atom is situated
at the centre of the cube. The bcc unit cell contributes two atoms. Sodium, iron,
lithium, potassium, chromium, cesium chloride etc. are typical examples of a bcc
crystal.

Fig.2.16: a) Basis vectors of body centred cubic lattice; and b) nearest neighbouring atoms in bcc
structure situated across the body diagonal

The basis vectors defining the unit cell of bcc structure are shown in Fig.2.16a.
Mathematically, we can write
a

a, =ai,

and
a3 = E ( i + j + i ) .
2
Crystal Structure In this case, every atom has 8 nearest neighbours situated across the body
diagonals. From Fig.2.16b, you will note that the touching nearest neighbours lead
to the result

J5
r=--a. (2.8)
4

The volume occupied by the atoms in a bcc unit cell is

and the atomic packing fraction is

JTna3
--
&n
APF= =-=0.68. (2.9)
a 8

On comparing the results contained in Eqs. (2.6) and (2.9), you will note that a
bcc structure is more densely packed.
A more symmetrical mathematical formula is obtained if we define the unit
vectors of bcc lattice as shown in Fig.2.17:

al = E2( j + k - i ) ,

a, ="(i+i-j),
2

and
a
a3=-(:+;-k)
2 (2.10)

The primitive unit cell of bcc structure, also shown in Fig.2.17, is constructed by
joining the four neighbouring body centred atoms to their nearest three corner
atoms. This construction results in a rhombohedron.

Fig.2.17: Primitive unit cell of bcc lattice

c) A face centre cubic (fcc) structure has, apart from the 8 corner atoms, 6
additional atoms, each at the centre of every face of the cube, as shown in
Fig.2.18a. NaCI, KBr, MnO, MgO, Al, Ni, Cu are some examples of fcc lattice.
 Crystal Geometry

Fig.2.18: a) Basis vectors; and b) nearest neighbouring atoms in fcc lattices

Every atom in fcc lattice has 12 nearest neighbours as shown in Fig.2.18b. The
basis vectors for fcc structure can be defined by joining a comer atom with the
three neighbouring face centred atoms. Mathematically, we can write

and
(2.11)

The primitive unit cell for an fcc lattice is1 shown in Fig.2.19. If you calculate the.
APF of fcc structure, and compare it with sc and bcc lattices, you will find that fcc x
is the most densely packed stricture (TQ 3). Fig. 2.19: Primitive unit cell
of fcc lattice
d) The diamond structure is observed in group IV elements (C, Si, Ge etc). The
compound semiconductors like GaAs, InP, ZnSe, CdTe also possess the same
lattice structure. Fig.2.20a shows a typical diam&d structure. This structure cah
be built by considering two fcc lattices inter-penetrating each other. The second
fcc lattice is displaced by
(a4,
-I,-
:-I
:=
J,-k along the body diagonal of the first

lattice, i.e. it is displaced along the body diagonal by one fourth of its length. As
may be seen, the atoms in this structure are bonded to 4 nearest neighbours in the
tetrahedral structure with typical bond angles of 109'28'. For simplicity in
understanding the positioning of atoms in diamond structure, we divide the cube
into 8 smaller cubes. Then the second lattice atoms lie in the centre of 4
diagonally opposite cubes, as depicted in Fig.2.20b.
Crystal Structure Before proceeding further, you may like to answer an SAQ.

Spend SAQ 5
2 min.
Determine the number of atoms contributed to the lattice by a unit cell of diamond
structure.

Hexagonal close-packed structure (hcp) does not fall under the category of cubic
crystals but this structure is observed in several crystals like Be, Cd, Mg, Ti, Zn, Zr,
etc. and it is worthwhile to spend some time studying it.
As the name suggests the atoms in hcp lattice have hexagonal arrangement in a plane
(Fig.2.2 la). Many such planes.are stacked on one another to form the 3-D lattice as '
shown in Fig.2.2 1b.
In a hexagonal close-packed structure, atoms in two consecutive planes are not
situated one above the other; they touch each other in a horizontal plane (A) forming
closely arranged triangles and the atoms in the next plane (B) sit in the void at the
centre point of these triangular units. The third plane (15') atoms are again placed on
the voids in the second plane. Effectively, the atoms in the first and the third plane are
aligned vertically.

Fig. 2.21: a) Atomic arrangement in a plane in hcp structure; and b) hcp structure in 3-D

The basis vectors defining the hcp lattice shown in Fig.2.21 b are:
5

and

where c denotes the distance between planes A and C. The hexagonal close-packed
structure suggests a definite relationship between a and c: the ratio cla has
a
a vdue more than this, the atoms are stacked loosely above one another. It is a normal
practice to state the value of a and cla for defining the lattice constants of hcp lattices.

After learning about the lattices, you can visualize the atoms in a crystal to be
arranged in planes inclined at different angles and having different spacing. In many
crystals, different physical properties like thermal conductivity, electronic conduction,
X-ray reflectivity etc. can vary along different planes. Hence, it is necessary to
distinguish various crystal planes. For doing this, let us now learn to index different
 Y Crystal GeomcWy
2.5 MILLER INDICES
A crystal has various smooth faces with some fixed angle between them. Each of
The choice
these faces represents a particular plane in the crystal lattice. To identify these index the pl
planes, we index them with 3 numbers. These numbers are called the Miller indices of kee-dimensl
the plane. The Miller indices of any crystal plane that intersects three axes along the of the crystals.
basis vectors, say a,, a*, a3 at points such as H, K and L respectively is denoted as
(hkl) and consists of 3 integers proportional to inverse of intersects OH, OK and OL; 1
i.e. h: k:l::- --a2 :-a3 . For a lattice with a, = a* = a3 = I, if we draw the plane
OH 'OK OL
with H= 3, K = 4 and L = 4 as shown in Fig.2.22, the indices (hkl) are given
1 1 1
by h : k : I :: - :- :- = 4 :3 :3. Hence (433) are the Miller indices of this plane. In the
3 4 4
next unit you will learn that these indices represent the reciprocal lattice vector, which
is along the perpendicular direction to the plane.
In general terms, the Miller indices can be defined to be integers such that

Fig.2.22: Defining Miller

indices of a cryrhl
plane
The faces of a cubic unit cell are indexed as (loo), (0 10) and (00 1) as shown in
Fig.2.23a. Note that Miller index 0 corresponds to a plane assumed to intersect the
particular axis at infinity. Fig.2.23b depicts another plane in the cube with Miller
indices (1 10). To familiarise yourself with these constructions, you should draw
planes correspondingto Miller indices (01 1) and (10 1).

(a)
Fig.2.23: Miller indices of a) faces of a unit cube; and b) another plane in the cube
(711)
If a plane intersects on the negative side of any axis, the corresponding index is *<--fl-;

written with a bar on top of it. The plane shown in Fig.2.24 has Miller index(i1 I ) . .' -1
Since a crystal has translational symmetry, it is not possible to distinguish between
any two parallel planes. For this reason, Miller indices of a plane represent the entire
)yj Y

family of parallel planes. Fig.2.24: Mi

pk
Once the Miller indices of the family of parallel planes are known, it is easy to on t ~ -x-.mis
e
determine interplanar distances. The interplanar distance of any family of planes
having Miller indices (hkl), denoted by dm, is given by

where a is lattice constant (cube edge).

 Crystal Structure Now we would like you to answer the following SAQ.

1 Spend
4 min.
SAQ 6
Determine the Miller indices of the planes shown in Fig.2.25. Also determine the
interplanar distance, if lattice constant a = 3.58L.

2.6 POLYCRYSTALLINE AND AMORPHOUS SOLIDS

So far you have learnt about the structure of single crystals. However, all the solids do
not exist as single crystals. This is because the growth of a crystal occurs around a
nucleation site commonly known as seed, which is a small cluster of atoms. If
physical conditions provide more than one nucleation site, crystals start growing
around each such site. As the seeds can be randomly oriented with respect to each
other, the crystals growing around these are oriented in different directions. When
such crystals grow large enough and begin to touch neighbouring crystals, they share
a common face or boundary. These crystals are much smaller in size as compared to
the single crystal. For this reason, they are termed as crystallites. Many such
crystallites come together to form the solid, which is obviously apolycrystal.
In polycrystalline solids, such as quartz and rock salt, each crystallite is a small replica
of the single crystal and has the same lattice periodicity. However, this periodicity is
broken at the boundary of the crystallites as shown in Fig.2.26. The properties of such
polycrystalline solids are much different from those of the single crystal of the same
material. For example, when light beam is made to fall on a polycrystalline solid,
randomly oriented crystallite boundaries scatter this light rather than reflecting or
refracting it uniformly, as in the case of single crystals. Hence even the colourless
pclycrystalline solids appear to be translucent or opaque, though their single crystals
Flg.2.26: Crystallite boundaries art. transparent. The electrical conduction is also affected by the crystallite boundaries.
In polycrystals
In amorphous solids, also called non-crystalline solids, the ordering is normally
limited to few neighbours only. They lack periodicity in their lattice and have
Rapid quenching means very
fast or super cooling of the completely different physical properties. Window glass is an example of amorphous
melt. This arrests any solid. The conventional fused silica glass consists of randomly orientated SiOz
movement of atoms in the molecules. But we also come across metallic glasses that are formed by rapid
melt. Hencethestmcturejust quenching of melt of the mixture of metals. These are characterized by very peculiar
freezes without attaining properties like corrosion resistivity. These find use in constructing the outer bodies of
crystalline periodicity.
aircrafts, ships etc. Due to their metallic nature, they are good thermal conductors,
unlike the conventional silica glass. There has been growing interest in amorphous
solids in recent years. This is particularly due to the discovery of special electrical
properties of amorphous semiconductors.
Let us now summarise what you have learnt in this unit.

2.7 SUMMARY
Lattice is a mathematical abstraction. It is obtained by replacing the atoms in a
solid by massless points, called lattice points.
Solids crystals can be categorised in 7 crystal systems and 14 Bravais lattices.
A unit cell, when translated in the 3-D space, can construct the entire crystal.

A unit cell contributing single lattice point per cell is called the primitive unit
cell. It is obtained using Wigner-Seitz construction.
A lattice may exhibit symmetry operations such as rotation, translation,
reflection, inversion or their combination.
 Cry8tal Geometry
Cubic lattices are the most familiar of all crystal systems. These have the most
symmetric lattice structure. Cubic crystal can occur in simple cubic, fcc, bcc or
diamond structure.
The atomic packing fraction (APF) defines the density of crystal structure (or its
looseness).
Miller indices are used to index the planes in the crystal and these 3 numbers
define a family of parallel planes.
- .-

2.8 TERMINAL QUESTIONS Spend 30 min.

1. For the hexagonal crystal shown in Fig.2.27, determine all the reflection symmetry
planes.
2. Determine the rotational axes observed in case of simple tetragonal and face-
centred orthorhombic structures. Indicate the folds of each axes.

3. Determine the atomic packing fraction for fcc structure.

5. A metallic element has density 8.91 g cmm3,lattice constant a = 3.52 A and atomic
weight of 58.71 u. Determine the number of atoms per unit cell of this element and
predict its crystal structure. Also calculate the nearest neighbour distance.
Fig. 237: Hexagonal cry8tal
Hint: The number of atoms in 1g mole of a substance is equal to Avogadro's
number.

6. Draw the unit cell represented by the vectors a, = 2i + j, a2 = 2j and a3= 31;.

2.9 SOLUTIONS AND ANSWERS

Self-Assessment Questions
1. Rectangle ABCD has a lattice point in the centre and one point each at the comers
A, B, C and D. Hence total contribution of this cell to lattice will be 2 lattice
points. By definition, this,does not qualify as primitive unit cell.
2. The base-centred orthorhombic unit cell (Ref. Fig.2.8e) has 8 atoms at the comer,
of the unit cell and two atoms on the face centres of opposite faces. The comer
atoms are shared by 8 adjoining cells while base centre atoms are shared by two
neighbouring cells. Hence total atoms in this unit cell are
=Bx(1/8)+2~(1/2)=2.

3. The mirror planes of an orthorhombic structure are shown in Fig.2.28.

--

4. Yes, the body centre point of the given structure is the inversion centre.
 ure 5. As seen in Fig.2.20a, each unit cell of diamond structure has 8 comer atoms each
shared by 8 neighbouring cubes; 6 face centred atoms, each shared by two
neighbouring cubes and 4 atoms completely con'tained within the unit cell. Hence
total contribution of diamond unit cell to the lattice is

I
6. Planes 1 and 2 in Fig.2.25 belong to the same family of parallel planes. Hence they
have the same Miller indices. Plane 2 is clearly showing intersection at 1, -1 and oo
with x, y, z-axes respectively. Hence the Miller indices of the planes are (1 TO).
5 ,6 1'
The interplanar distance
a 3.5A

I 1. In all 7 reflection symmetry planes are possible for a hexagonal crystal. These are

planes of a hexagonal 3,. From Table 2.1, you will note that for a tetragonal crystal system, a1 = a* # a3.

c4F0
crystal Now refer to Fig. 2.30. Out of the axes passing through face centers, PQ is 4-fold
P
I
." symmetric while RS and TU have 2-fold symmetry.
For a face-centred orthorhombus, a, # a2 z a3as shown in Fig. 2.3 1 and PQ, RS
and TUaxes
R.---- --
a, ,dl
____ ___--- ---- s are 2-fold rotational symmetric.
3. As seen from Fig. 2.32, the nearest neighbours in fcc lattice are situated across the
,' I
face diagonals. Hence the touching nearest neighbours result into

I
U

Fig. 2
m Simple tetragon
4
,, Each fcc unit cell contributes 4 atoms, so the volume occupied by atoms in the

", ..."...",..""-
I 4. For close-packed plane stacking, the spherical atom sits on a void site in the I
fashion. In this case the atoms in first and third plane are exactly aligned and the
distance between the centres of first and third plane atoms is lattice constant c.
This situation is depicted in Fig.2.33a. The diameter of each atom is a.

I L - - - a 4
Our aim is to find distance QA = c 12. From Fig.2.33b, we note that the atoms in
each plane are closely packed and MCD is equilateral.
 Third plane --( i r X r )
r'%-n
Second plane -,

First plane+ (&!yyJ

Fig.2.33: A hexagonal close-packed structure

Hence,

and

From F i g . 2 . 3 3 we
~ ~ note that

(ii)

Moreover,

(iii)

On substituting these results in (ii), we obtain

Same treatment is applicable to second and third layer:

so that