U'NIT 2 THE BAND T ORY OF SOLDS

2.1 Introduction
Objectives
2.2 Electronic Struchwe
2.2.1 The Hydrogen Atoll1
2.2.2 'I'hePeriodic Table
2.2.3 The Nydpogkn Molecule
2.3 Types of Bonding in Solids
2.3.1 The Ionic Bond
2.3.2 The Covalent Bond
2.3.3 The Metallic Bond
2.3.4 The Van der Waals.Bond
2.3.5 The Hydrogen Bond
2.3.6 Mixed Bonds
2.4 Energy Levels in Solids
2.4.1 Degenerate Levels
2.4.2 Lifting of the Degeneracy in Crystals
2.4.3 Rlling up of Energy Levels by Electrons
2.4.4 Forrnatioli of Bands
2.4.5 The Band Gap
2.5 Classification of Solids (zccordi~gto their Electrical Properties
2.5.1 Conductors
2.5.2 Insubtors
2.5.3 Scmiconducto~s

2.7 Key Words

I

I 2.8 Answers to SAQs

In Unit 1 we have seen that it is very convenient to classify materials into metals, ceramics
and polyn~ers.This classification becomes very natural if one considers the atomic forces
responsible for holding together Ule nuclei and electrons in solids. The theoretical ideas of
the solid state come out nicely when we consider these Ulree classes of materials.
I To consider the main fcatures of these solids, it is not necessary to consider nuclear forces
at all as these never really come into play under normal circumstances when solids
I
experience external forces; the nuclear forces are too strong to be responsible for the
changes co~lsidered.One can thus talk of two entities in solids; the nucleus plus the core
eleclrons, i.e. the ions, and the bonding electrons.
I For bonding in a solid, the outermost eleclrons change their energy states or orbitals in
1 such a way that a ~ninimisationof the total energy results in the fonnation of a regular
1 array of ions, i.e. the lattice. The behaviour of these bonding electrons is distinctly
different in the three classes of solids mentioned in the classification. These are now
considered in some detail. But f i s t let us see why a bond gets formed at all! We know that
two atoms do come together to form a molecule as in hydrogen or oxygen. Can we expand
on the sanle ideas for the bonding in solids?
Objectives
In this mil the role of the valence electrons and ions in the formation of solids is
considered. This would help us to understand why solids behave the way they do when
subject to external forces such as an electric f i ~ l dWe
. will see that conventionally there
ate primary and secondary bonds in materials and consideration of these helps us to
classify the solids into those useful for applications such as withstanding high temperatures
or those which are very good insulators etc. We will also understand why some solids are
dense and heavy'while some are lighl; aid many simple ideas will emerge.
Structureof Materials By the end of the unit, you should be able to :
* predict the electronic structure of elements,
clearly understand the five types of bonding,
understand that most behaviour in solids can be ascribed to particular bonds or
to mixed types of bonding,
* estimate the bond energies in ionic solids,
* calculate the energies involved in the formation of simple polymers, and
predict the electrical behaviour of solids in simple cases.

2.2 ELECTRONIC STRUCTWRIE

As mentioned in the previous unit, we consider the solid to be made up of a regular array
of ions and redistributed valence electrons. Of course, you must remember that this would
only occur if this rearrangement of electrons leads to a lowering of the energy of the total
system. One must then consider whether it is possible to remove electrons from the neutral
atoms with relative ease. In order to understand this we have to consider the electronic
structure of atoms, and the energies involved. The simplest atom is Hydrogen, as it
contains just one proton and one electron. We will see the energy levels for this structure
and try and generalise our ideas to atoms with many protons and electrons.
2.2.1 The Hydrogen Atom
As mentioned above, the hydrogen atom consists of an ion containing a proton with a
charge of + e, and there is an electron, with ch'arge - e, which surrounds the ion. It is
necessary to remember the sizes of the particles involved. The proton has a mass nearly
1800 tirnes that of an electron and occupies a space of &,tenthof a picometer (Note: a
picometer is 10'12 meter or one-hundredth of an hgstrom). The hydrogen atom, however,
which is an electron and proton together, occupies a space of a hundred picometers. This
situation is pictoriallyrepresented by saying that there is an electron "cloud" that
surrounds the proton as shown in Figure 2.1.

Figure 21 t Electron Charge Clwd (Schematic)Surrounding the Nucleus d a Hydrogen Atom in the
G m ~ Sbate
d The Outer Circle o f r = 529 pm Corresponds to the Radius ot the h t Orbit
( I.-, n = 1) and Indicates the Mast Probable Region lor Finding the Electron

Can we work out the energies of this simple electron-proton system?

A simple way of solving the problem is by assuming the dktron to be in spherical orbits
of definite energy. The energy of these orbits or levels is quafitized and, for every level,
there is an angular momentum assigned which also takes on definite values in multiplesof
a universal constant known as the Ranck constant,.h, which has a value of
6.626 x J-s. We may write the equation for angular momentum as:

i.e. the angular momentum is quantized; the orbits have radius r, m is the mass and v the
velocity of the electron, with n a number which takes on integral values 1,2,... etc.
 We can calculate the radius of the orbit if we know the velocity of the electron in its orbit. The Band Theory of Solids
To do this we have to write the equation of the total energy of the electron. The results are
now well known as the Bohr theory of the hydrogen atom, giving the total energy as:

Putting the values of the constants we get the easy-to-remember values for the energy as:

En =
13.6
-- . .. (2.2a)
2
n
where En is the energy of the' n orbit expressed in electron volts. (Note : An
electron-volt or eV is a unit of energy that an electron gains when it moves up a potential
of 1 volt. 1 eV = 1.6 x 10-~~joule.)
Tlie radius of the orbit is obtained by equating the centripetal force to the electrostatic
force, i.e.

Combining this with the condition of momentum quantization (Eq. 2.1), we get for the
radius of the nh orbit as:

By putting the value of n equal to unity, we gel r = 52.9 pm (or 0.529 A). The number n is
called a quantum number, and the electron is supposed to be in specified orbits with
d
n = 1 for,th ground state, n = 2 for the fust excited state and so on. Note that the energy
difference tween successive energy levels becomes smaller as n (and therefore r )
increases. This is shown in Figure 2.2. Classically this means that as the electron moves
away from the core, it becomes more and more loosely bound.

0 IrX
n 85
- 1.5eV .
&

nm3 [ M Shell)

~ e n( One-dhnedotd Model )
Figure 2.2: Energy Levds e9r the ~ ~ d r o Atsm

The levels with the various values of n are given names following spectroscopic notations;
n = 1 is called the K shell; n = 2 the L shell, n = 3 the M shell and so on. For n equal to
inFmity the electron is supposed to be "free" as its total energy is theniem. Ac~dUng to
this picture, an electron can make a transition from one orbit to another by either absorbing
or emitting energy in quanta. The energy needed to excite an electron of an atom from its
ground state to its free state (for which E = 0) is called the ionisation energy for the atom.
From Fq. 2.2a, the ionisation energy for hydrogen is 13.6 eV.
Structure of Materials Example 2.1
Determine the wavelength of light that is emitted when an electron in a hydrogen
atom makes a transition from the state of n = 3 to (i) the ground state, and
(ii) to a state with n = 2.
Soliltion
E3= 13.619 = 1.51 eV. The ground state has n = 1 (the lowest allowable energy
level for the electron) and El = 13.6 eV. The energy of emitted radiation is
hc
quantized as IE3-El I = hv = - = 12.49 eV, where h is the wavelength of
3L
light. Putting the various values we get for change in energy:
BE = (1.242)/ h , where h is in lnicrons if E is expressed in eV.
This gives h = 99.4413111 (i.e. in the ultra-violet region).
Similarly, IE3-E2 I = 3.40 - 1.51 = 1.89 eV; and h = 657.1 nm (in the red region).
SAQ 1
By putting in the values of the quantities in Eq. 2.2, show that Eq. 2.2a results.

SAQ 2
Determine the radius of the third orbit for the electron in a hydrogerl atom.

What about the electron levels in atoms containing more than one electron? To answer this
question we would need to see how quantum mechanics alters our view of the energy
levels in the hydrogen atom. We will see that a fairly simple explanation of the periodic
chart of elements follows.
2.2.2 The PerSo&c Table
It is now well understood that, in order to describe systems at the microscopic level one
needs to use quantum mechanics. What does this mean? It means that at levels where one
cannot consider the mass to be a good representationof the body, one camlot use Newton's
laws. Why is mass undefined? It is because the m s is not uniformly distributed in space
but iS localised at the position of the ion. In between the ion positions, one cannot speak of
a mass! In order to consider such situations, the correct equation of motion is the
Schrddinger equation. As we will be considering only equilibrium energies, we choose the I
time-independent Schrtldinger equation, which can be written as:

Here y is the wave function describing the behaviour of the electron under a potential V,
and E is the total energy of the electron. To solve the hydrogen atom problem quantum .
eL
mechanically, one uses the value of Vequal to -, i.e. equal to the electrostatic
4fce.i~r
potential that an electron would experience in the electric field of the proton when it moves
in a spherical orbit of radius r. After an involved algebraic manipulation, we would get the
same value for E as in Eq. 2.2.
However, one should remember that we have solved a simple problem here by considering
spherical orbits which would give the same answer for the x, y and z axes. Generalising to
non-spherical orbits, the three dimensions would have to be considered separately and it '
would be logical to expect that E would depend on three quantum numbers. These have
been called n, the principal quantum number, 1 the orbital quantum number (to take
into account non-spherical orbit.) and m , the magnetic quantum wumber (which takes
 into account the spatial orientation o[: the orbits). Furlher, when we consider that the Thc Band Theory of Solids
electron has a mass, charge and a spin, a quantity not included in Eq. 2.4, we need a fourth
quantuni number, the spin quantum number, o.The energy state of the electron would
be given by the four quantum numbers : n, 1, m and o .
Quantum mechanics also gives the range of possible values that these quantum numbers
can Lake. The principal quantum number, n,defines the size of the electron orbit. The
larger tlie value of n, lhe larger is the space occupied by the electron cloud. By selting the
value of n, we define the electron shell of the atom, as in Figure 2.2. The shell wiffi n = 1 is
called the K shell, the one with n = 2, the L shell, n = 3, the M shell and so on. So n can
take on vahies from 1 to infinity:
n = I , 2 , 3 , ... integral values ...
The orbital quantum number 1 defines the shape of the electron cloud. For different values
of n the same value of 1 denotes orbits of the same shape. For example, 1= 0 gives an orbit
of spherical shape. To each value of 1 we assign a corresponding subslael!. It is customary
to assign spectroscopic nolations to these subshells; subshells are designated as s, p, d, f
etc. corresponditig lo 1 taking values of 0, 1 , 2 and 3 respectively. The first orbit that an
eleclron can have in the hydrogen atotn, as shown in Figure 2.1, is Lhe 1s shell, i.e. n = 0
'and I = 0. The largest value that 1 can L'ake is n -1. So:

The magnetic quantum nuniber t?t detines the spatial orientation of the electron orbit in the
group of orbits with the same shape, i s . with the same 1 value. In each subsliell there are
21 + 1 orbits wilh diffcrenl orientations ;is m can take values from 1,l-1, 1-2 ... up to - 1 :
n 1 = - 1 , - l + l , - 1 + 2 ,.+.,-I , O , + l ,...+ 1 - I , + 1
Finally, the spin quantum nuniber o denotes the orientation of tlie electron spin in a given
orbit. cr can only take two values:
o = +1/2 and -112.
2
In eacli n level, Ihen, there are 2n electrons.

I The possible wave unctions 01 a11 electron in a one-clectron atom, such as hydrogen, are
called atomic orbitals. In the hydrogen alom, all orbitals with the same n value have the
1
I
same energy and so are called degenerate.
How do we apply these ideas to atonis containing morc than one electron, such as say
lithium'?A simple method would be to assume that tlie energy slates are exactly Uie same
I
as the one-electron atom and that electrons occupy states by followi~igthe Pauli exclusion
I principle, i.e. no two electro~iscrln have thc samc set of quarilum oumbers. Hence, in
lithium, which has 3 eleclrons, the first two would occupy the first shell, i,e. with n = I
(with opposile spins, i.e. o = + 112 for one eleclron aid cr = -112 for the other) md the
third electron wo~ildgo up to the second shell with n = 2. In this way we car1 co~lsider
atoms with 11iany electrons. The rules are that in each orbit I, thc electrons occupy states in
pairs, and the states are designated by the value L a I can *!te. This meals that in the s
state (hi which 1 = 0 ) there can be two electrons; in the p stdte (with 1= 1) there can be 6;
in the d state 10 aiid in the f stale 14. A shell is said to be colnplelely tilled if the electrons
oc<:upyall tlie stales. In this way it has been possible lo underst,md the rcactivity of the
elements. For noble gases, for cxample, the 1) state is always cotllpletely filled; an
incompleiely filled outcrtnost shell leads to rcactivity of the element which is the ease with
which it can form compounds. We will see lalei that the elements, in compounds, prefer Lo
have completely filled shelis as such a configuration leads to an overall reduclion in the
energy.
Is there an order in whic11 the levels fill up? You rnay recall that the way the levels fill up
(details of which are in tlie Unit on Chemistry) is in Uie following order: first the Is, then
the 2s, the 2p, followed by 3s, 3y and the 4s, followcd by 34 4p and 5s, then the 4d, 5p
and GS, then 4J 5rl, 6p and 7s ant1so on, One must exercise some caution here as the
energy differences in states with large n are small, i.e, tile 5d levels may partly fill up
before the filling up of the 4f levels. Generally, the filling up of the levels is consistent
with gettin the lowest ener ry configuration. That is why in copper the outernlost orbitals
would lx d9 0s 1 rather than d$ s2 ! This is because a fully filled inner shell would lower the
cncrgy, as Uie outer electron wottld participate in bonding, as we shall see later in this unit.
 ..
Structure of Materinls SAQ 3
The element gadolinium (Gd) has a Z value of 64. Write down its electronic
structure. Would the knowledge that the element exhibits predominantly a valence
state of 3 help you to refine the structure?

What about the ionisation energies for the elements? In Table 2.1 we give some
representative values for the first ionisationenergies in eV for some elements.
Table 2.1: Ionisation Energy of Some Elements, in eV

Note that the elements to the right in the table have higher ionisation energies than those to
the left (except hydrogen). This means that elements like Li, Na, Mg etc. give up the
outermost electrons rather readily. This is the bas@of the formation of ionic bonds in
solids as we shall see later.
Note also that as we gd down the periodic table the ionisation energies decrease. This
means that it becomes progressively easier for one to knock off the outermost electron in
the heavy elements. No wonder then that the heavy elements show greater tendency to
form compounds and even the heavy inert gases form some complexes.
We have seen how an understanding of the hydrogen atom levels has helped us to
appreciate the periodic table of elements. We have also seen how completely filled s andp
shells lead to a lowering of the energy of the structure. What would now happen if two
hydrogen atoms come together to form the hydrogen molecule? Further, why cannot three
hydrogen atoms form a stable structure? Also, how is it that many sodium atoms come
together to form a solid even though a sodium molecule does not exist? Let us consider the
answers to these questions.
When a gas of neutral atoms condenses to form a liquid and later to a solid, the atoms are
held together rigidly by mutual attraction. ?he pull between them is much stronger than
the gravitational force and we say that the atoms form bonds between one another. In a
solid each atom forms a bond with every other atom. In a solid containingp atoms, there
would then be p2 bonds. The number of atoms is typically of the order of per cubic
centimetre. This means that a full solution to h e bonding theory in solids is a formidable
task even with the most advanced computational techniques. It is easier to begin with a
system of two atoms, a diatomic molecule, and then try and generalise the theory to
include the moie complicated reality.
2.2.3 The EIydrogen Molecule
The simplest diatomic molecule is the hydrogen molecule. The hydrogen atom has been .
understood in terms of the manifestation of quantum numbers, which gave a nice
explanation of the periodic table of elements. 1n this section we will see the formation of a .
molecule of hydrogen. Consider two atoms of hydrogen, labelled as Ha and Hb , at some
distance from each other. Let us assume that there is a small force of attraction which
would bring the atoms closer as depicted in Figure 2.3(a). As they come closer the electron
orbits (clouds) around them distort (Figure 2.3(b)), inducing an electric dipole moment at
each a tom. This induced moment means an increased attractive force, bringing the atoms
even closer, till th ir electron orbits overlap (Figure 2.3(c)). Tryhg to bring the atoms even
' 1
closer encounters fficulties. There would be proton - proton repulsion as well as
 repulsion dut: to Pauli's exclusion principle, as two electrons could not occupy the same T h e Band Theory of solids
orbital unless their spins were opposite. The atoms are thus pulled together till each
electron orbital encompasses both the protons, at which stage both the electrons share the
same wave function.

la)

. .

Figure 2.3 : (a) Isolated Atoms; A Weak Attrnctive Force is Present;

(b) At Closer Appronch the Electron C l o ~ ~ are
d s Distorted by the AttrncHve
Forces, the Two Elcctron Sybts are Antipnrllel;
(c) In Equilibrium the Electrons Share the Snme Wave b n c t i o ~ ~
At this point there is an equilibrium between a repulsive force and an attractive force. This
is schematically shown in Figure 2.4. The repulsive force is due to electron overlap and the
attractive force due lo the electrostatic interaction between the dipoles. The atoms then
approach within an equilibrium distanq to form a H2molecule, with the electron spins
oppositely directed. For hydrogen this is about 74 pm. Note that the bonding results in the
ovcrlap of electron clouds. This is a universal characteristic and it means that a pair of
atoms form a bond only when their elec&onclouds overlap. Note also that a third electron
could not occupy the same molecular orbital and so a H3molecule would not form due to
Pauli repulsion. The bond is said to be saturated.

Re~ulsiveforce
(electron orbital overlap I

forcr
rostatlc )

distance t, lonic s~poration( r )

= 74pm f o r k !
Figure 2.4 : Interatomic Forces and their Dependence on Distance

When two heavier atoms come closer there is a minor difference from the case of the
hydrogen atom. Let us consider the case of two sodium atoms coming close to cach other.
The Na atomhas an electronic configuration,of 1s22s22p63s1, i.c. it consists of the Neon
core plus one 3s electron. As two sodium atoms come closer, the outermost 3s electron
orbitals would overlap first and then the 2y and 2s orbitals would ovcrlap. As there is a
repulsion term in the force, it is unlikely that the atoms would come so close that the
innermost orbitals would also overlap. It is convenient to think of the atoms as contailling ,
a core (of ions) and overlapping outer electrons. A condition of equilibrium can be attained .
with the ions coming close enough so that the electrostatic attractive force due to
interaction between the ions and the electrons is balanced by the repulsive force and it is
this assemblage that constitutes the lattice. Formation of crystals can then be considered as
the ion cores being arranged in a lattice and the electrons around them providing the
necessary bonding forces to form the solid, In sodium then, the picture of the solid would
 be that of the nucleus plus core electrons being arranged in the tlhPee-dirne~lsio~~al
lattice
and the 3s electrons of every sodium atom in the volume being free and forming a cloud
around the lattice. The distance between the ions is nearly Bike the distance bctween two
non-compressible spheres, which are glued together. Trying to pals11 them closer involves a
large increase in potential energy, while trying to draw them apart also increases the
potential energy. This picture helps us to classify solids dependimg on the nalure of the
bonding forces and the magnitude of the bondhg energy between the ions md electrons.

2.3 TYPES OF BONDING IN SOLIDS

- - - -

We have seen how the overlapping of electron orbitals of the outermost or valence
electrons can lead to a formation of a bond between two atom. 'lrtae formation of a solid
body can be considered similarly as ilrising out of the interaction of one atom or ion with
all the other atoms or ions in the solid. We must remember, however. that h e number of
atoms in question is very large and so the solution of the general problem is difficult. The
number of bonds that a single ion can have determines tl~enumber of neiglnbours for the
ion, i.e. its eo-ordination number. It is the co-ordination nunnkr, CN, which also plays a
major role in the way the ions are arranged in space, i.e. the crystal structure. For the
present we shall consider the bond orfly between tl1e ion and its neighbour, remembering
that there may be more than one such neighbour. There are converitionally five different
types of bonds, though the exact line of deinarcation m y be blurred iaa actual solids. These
are: (i) ionic bonding, (ii) covalent bonding; (iii) metallic bollding; (iv) Vaa der Waais
bonding and (v) hydrogen bonding. The fnrst three ane considered to be priulaiuy types of
bonding as the theory involves interaction between eleckically charged ions, while tlie
latter two involve interaction between electrical dipoles 'and hence are not as strong. An
easy way of judging the strength of a bond is to measure the melting temperature of the
solid. High melting temperatures generally mean a primary type of bond. Examples of
materials with the primary type of bond are MgO, SiC or the metal tungsten. Secondary
bonds are.seen in solid carbon dioxide, polymeric solids like polyelhylene and the bonds
holding the water of crystallisation in salts such as copper sulphate. k t us see the miin
features which distinguish each type of bond.
2.3.1 ,meIonic Bond
This bonding is perhaps the easiest to understand and is a characteristic type of bond found
in a great number of ceramics. Consider the formation of ehe well blown ceramic
magnesia or magnesium oxide. The chemical formula is M g 0 md is often used in
applications involving the use of high temperatures. Its popularity is because it is hard,
electrically insulating, and stable up to its melting point which is 2800 O C . M a t atomic
forces bring about such a stability? The clue is given in the molten state of the compound
which exhibits electrical conduction. This meals that the nuaterial is composed of
electrically charged particles or ions. The same is true in a sinnpler material like common
salt or sodium chloride. When dissolved in water, MaCl imparts electrical conduclivity to
the water which is supposed to arise due to the formation of sodium and chlorine ions.
Why are these ions formed?

@::@- @@
Sodium Chlorine
atom ,Na atom ,CI ?$,"& Chloride_
jon,CI
Atomic Atomic
radtu s
= 191pm
md~us
= 99pm
#hS
~102pm
lonlc
radius
= lelpm

I
, .
Mgure 2 5 : Formation of a Sodium Chloride Ion I%irfrom Sodium and Chlorine Atoms. 111 tile Ionization
Process A 3s' Bectron from the Sodium Atom is Transferred to a Half-empty 3p Orbital nf the
ClllorineAtom

. Sodium has atomic number 11, and tile atom has a radius of 191 pm. The eleven electrons
are arranged as Is22s2 2p63s1. This can be written as neon core plus the 3s elec*on, i.e.
 [Ne] 3s. A stable arrangement of electrons would emerge if the 3s electron is removed T ~ ~Cs n Theory
d OP solids

giving rise to the sodium ion which would have a radius of 102 pm. Chlorine, with atomic.
radius of 99 pm, on the other hand would attain the argon structure by acquiring an
additiorbl electron. In the process its size would increase to 181 pm. The Na atom
therefore donates the 3s electrom to C1, so that both ions have inert gas structures and are
thus very stable. 'This is schematicallyshown in Figwe 2.5. The ions however have
acquired a charge; Na acquiring a positive charge having donated the electron and C1
beconing negatively charged by capturing an electron. The resulting ions Na' and Cl-
Jtetnately occupy the lattice sites in solid sodium chloride, with the equilibrium distance
between the ions being 283 pm.
The number of eleclsons that an atom can give to form a completely filled shell or the
number of electrons that ma alonl needs to fill up h e octet is called the valence of the
atom. Both Na and C1 have a valency of one. Mg and oxygen, as discussed above have a
vJency of two. kt us see the concept of valency in some other cases.
Example 2.2
I Carbon is in Group IV of the periodic table and has six electrons. 11s electronic
configuration is Is22s22p2' There are 4 electrons hi the outer orbitals and so carbon
c m have a maximum valence of + 4. It also needs 4 electrons to fill up they shell.
Hence it also forms compounds by accepting 4 electrons, i.e, a valency of - 4. We
will see later tlaal (his fact as the basis of carbon forming covalent bonds. An

1
inlarrrlediate valelacy of two would also be possible with both the Is and 2s shells
being filled. hl carbon monoxide we do see the valency of 2.
Example 2.3
Sulphur, which is in Group VII laas 16 electrons arranged as 1s22r22y63s 2 3 p4. By
accepting two more electrons, Ule p shell gets completely filled. Sulphur prefers to
form conapounds by rkccepting two electrons aid so its valence is -2 as in ZnS. It
I
can also donate 6 clecbons, as in SO3 or in the ( ~ 0 ~ ) ~ - iThe
o n .valency of the
I element is then either h e group iiuinbcr in the periodic table or 8 minus the group
number.

It is necessary, for energy nniniirisation, that Na ions have only C1 ions as near neighbours
and vice-versa, otherwise tlae electrostatic repulsio~ibetween like charges would raise the
total energy. Na and Cli ions are said to form a bond, 'anionic bond.
As we have seen above, 'any bond formation in a solid is iui interplay of a repulsive and a11
attractive force (See Figure 2.4). In ionic solids the attractive force is electrostaticand its
poteiltial energy can be written as:

in which the term A would be proportional to Z l q e2 where Zl and 22 are the valencies of
the ions and r is the distance between the ions. The repulsive potential is a term which is
significant only wlleai overlap of tile electron orbitals takes place, and keeping in view its
strength at short distances, one can write it as :

where B is a constant and n is a large number of the order of 10.

The resultant potential energy is tllen given by:

To get the total energy of Uic ionic solid, we would laave to sum up h e potential energy
over all the bonds. Tlie sumnation of the first term would involve positive and negative
terms successively, but the summation would still give a net attractive terrn and depend
upon the position of the atoms, i.e. the crystal sU-ucLure. The second term, however, would
be significantonly between the closest neighbours, i.e. tlie summatio~ishould reflect the
CN. But one can see a possibility of simplification; at equilibrium, tlle net force must be
zero and the potential energy should be a minimutn. In other words, at r = r 0 the
St~cture
of Materials derivative of should be zero. This gives a relation between A and B, which can be
solved further to give:

$R (at r=ro) = --
Kl[ I: 1- -

If n is of the order of 10, by approximating it t\o ihnity, we would overestimate the energy
by just 10%. Putting n equal to infinity means that the ions cannot be brought closer than
ro and that they act as incompressible spheres! One can then speak of a ionic radius which
is characteristic of that ion, 'Ihe binding energy of the solid is essentially equal to the
electrostatic energy between the ions.
So, an easy way of understanding the ionic bond is to consider the ions to be hard spheres,
whose sizes are fixed according to the CN. This is because, as the CN of an ion increases,
the available volume for the ion also increases. Some increase in volume would allow a
larger ion to fill up the space properly, without any rattling possible in the structure. Tllis
would also mean that as rattling is to be avoided, not every ion can fit into the available
space. Consider the example of the rocksalt or sodium chloride structure. The chlorine ion
C1- , in a six fold co-ordination has a radius of 181 pm and that of a sodium ion, in the
same CN is 102 pm. Trying to compress an ion of chlorine such that its size reduces from
181 pm would need a very large force. Another way of saying the same thing is to say that
the increase in potential energy is very steep if one tries to reduce the bond length of ~ a - +
C1-ionic bond to less than (181 + 102) or 283 pm. Using X-ray diffraction analysis, one
can determine the lattice parameter of NaCl, which is 566 pm and from that the distance
between Na' and C1' which is equal to half the lattice parameter. The agreement between
the two values is indeed very good. It is then also possible to predict the lattice parameters
of ionic solids by knowing their ionic radii. In Table 2.2, some common ionic radii ate
given. Note that as one goes lower down in the periodic table the ion size increases; also as
one increases the valence state of a cation the ionic size reduces.
Table 2.2 : Ionic Radii in Picometers (pm) of Some Common Ions
Element Radius for Radius for
- - . -
A1 39 53
Ba 136
Ca 100
Cd 84 95
C1 181
Co 74
61
Cu 86 96
63 73
F 131 133
Fe 63 77
49 65
K 133
La 106
Li 60 78
Mg 49 72
Mn 67
54
Na 99 102
Nb 32 64
Ni 69
0 132 140
Sn 93
69
Sr 116
Ti 86
61
v 79
54
Y 89
Zn 60 75
Zr 72
(Note: We have given vahes for CN equal to 4 and 6 only as these occur frequently in
engineering solids. You should refer to standard tables for the other values.)
 HOW many near neighbours can Na' have? One should realise that the more the number of The Band Theory OE Solids
near neighbours, the more the bonds, and that much more energy would be required to
break up the solid and so the more stable would be the arrangement. However, there
should be a perfect three-dimensional arrangement so that the space is filled up uniformly
and there carnot be gaps in the structure. One can consider the larger ion ( C1- in this case)
tb have a spherical shape and to form a regular array, and the smaller ion (Na' in this case)
to also be spherical and to completely fill up the space in between. It is possible to
geomettically calculate the ratio of the ionic radii which would give the most close packed
mangement. If the ions have the same radius, that is the radius ratio equals unity, then
each ion would best be surrounded by 12 neighbours. As the radius of the cation becomes
smaller than that of the anion, the number of neighbours that an anion can have becomes
smaller. In Table 2.3 we give the radius ratios and the most probable CN for the cations.
1 Table 2.3 : Co-ordination Number as a Function of Cation to Anion Ratio

Range o f Radius Ratios Packing (Position o f tho Coordination Number

for Stable Coordination Anions)

< o r = 0.155 Linear 2

0.155 to 0.225 Triangular,corners of an 3
equilateral trinngle
0.225 to 0.414 Tetrahedral, comers of a 4
regular tetrahedron
0.414 to 0.732 Octahedral, comers of a 6
square
0.732 to 1 .O Cubic, corners of a cube 8
1 .O close packed hexagonal or 12
cubic

The Table gives the most probable CN. This meals that there could be some violation of
the rule. Let us apply the rule to alumina. A1203is made up of the anions (Oxygen) packed
according to the requirement of a complete1 s ace-filling arrangement. Applying the
radius ratio rules we see that the radii for A13 and
+
p 0" are 53 pm and 140 pm,
respectively. The ratio is 501140 or 0.378. From the table the most probable CN is 4. (But
being on the higher side of the range it could also be 6.) What would be the most probable
CN for Oxygen? Since each CN is really a manifestalion of the ionic bond, the CN for A1
multiplied by the number of A1 ions, which would give the total number of bonds
emanating from Al, should equal the number of bonds emanating from 0,or equal to the
CN for oxygen multiplied by the number of ions in the formula. The only solution possible
is CN[Al] = 6 and CN[O] = 4. And that is the structure for alumina. Let us take another
example, that of Si02or quartz. Though we know Uiat Si always prefers a tetrahedral bond
(it being a group IV element), let us see the radius ratios. si4+has a radius of 41 pm. The
. radius ratio is 411140 or 0.29. This is well within the range for four-fold CN. The Oxygen
on the other hand form Si - 0 - Si linkages. So, if the coordination number of Si is 4, then
the CN for 0 is 2.
SAQ 4
In ZrOa what is the CN for zirconium and oxygen?

Can we, then, calculate the strength of the ionic bond and hence the binding energy of
ionic solids? The binding energy is defined as the energy needed to break up the solid into
constituents that do not interact with each other and may be considered as free. This would
also be equal to the energy needed to make the distance between the ions, r to be infinity.
In order to do that we must sum up the Eq. 2.7 aver the volume under consideration. If
Structure of Materials there are N "molecules" in the considered volume, then the answer would be dose to
multiplying Eq. 2.8 by N, the number of pairs. The exact summation gives:

If Nis taken as the Avogadro number then the binding energy $0 is expressed in kilo joules
per mole. The quantity a is a numerical constant called the Madelung constant and
depends on the crystal structure. Some representativevalues for a are given in Table 2.4.
Table 2.4 : Madelung Constant for Some Simple Crystal Structures

Structure a

NaCl 1.747
CsCl 1.762
ZnS (wurtzite) 1.641
CaFz (fluorite) 5.038
Ti02 (ru tile) 4.8 16
SiOz (quartz) 4.439

The binding energy, also called the lattice energy, in ionic solids must be high as the bond
arises from the electrostaticinteraction. This means that the energy required to break up
the solid would also be high. Those energy would be connected with the melting point also
and therefore the melting points of ionic solids is high. In Table 2.5 below some lattice
energies and melting points are given.
Table 2.5 : Melting Points and Lattice Energies of Some Ionic Solids

Ionic Solid Lattice Energy Melting Point

@Jlmol) ec,
LiCl 829 613
NaCl 7 66 801
KC1 686 776
CsCl 6 49 646
MgO 3932 2800
CaO 3583 2580
SIO 331 1 2430
BaO 3127 1923

From the table, the following should be clear: that as one goes down the periodic table, tlie
lattice energies decrease and so do the melting points. (Lithium is an exception as the ion
being like helium, quantum effects cannot be ignored.) Also ionic solids with divalent ions
have higher lattice energy and so are highly refractory (i.e., they can withstand very high
temperatures).
We may summarise the main features of the ionic bond: the ions are arranged
geometrically, there is no directionality and the coordination number plays a key role;
there must be electrical neutrality and so the ions are arranged in a way that local neutrality
is maintained and one can define a formula for this neutral unit. As Na and C1 are both
monovalent the formula is NaCl; Ca and C1 would combine to give CaC12; remember that
neither NaCl nor CaC12exist as molecules.
2.3.2 The Covalent Bond
Earlier we had seen the formation of the hydrogen molecule when two hydrogen atoms
came close enough so that there was an overlap of the electron orbitals. Both the hydrogen
atoms share the two electrons to form a stable electron configuration. Essentially this is the
basis of the covalent bond, except Ulat in larger atoms a stable electron configuration is an
octet (i.e., 8 electrons occupying the s2p6 orbitals). In this bond the near neighbour atoms ,.
share the total electrons in such a way that an octet is formed on every atorn/ion. For
example consider the formation of the diamond crystal. In this, every carbon atom has a
s2p2configuration.When two atoms come close enough that their orbitals overlap the
 electrons get redistributed in the orbitals in such a way that every atom has a complete The Band Theory OE Solids
octet of electrons. For this to happen and for h e exclusion principle to be obeyed tbp, one
s electron goes up to a vacant p orbital so that the atom has a11 electron co~lfi~uiation
of
sp3. This is called hybridisation. The spuls of all the electrons can be parallel without
violatirig the exclusion principle. Then another atom with a set of four electrons with
oppositely directed spins can form a stable bond. When translated to three dimensions, this
means that every carbon atom can have four near neighbours each along the opposite
corners of a cube, 'and every cubon atom would be ru-rayed similarly in space. The CN of
each alonl would be 4 and this is characteristic of the covalent bond.
As can he readily seen, h e covalent bond arises in those elemeills whicli cannot be ionised
easily as such elemeilts would otherwise give ionic bonding. The elements in group 4 of
the periodic table are difficult to ionise and these are the ones that readily form the
covalent bond. One Izlay visualise the situation in the following way that as it is difficult to
remove all the four electrons, it is easier to share the eleclrons. One must remember that
here too it is not possible to isolate a molecule (as also was the case in NaCI), but the
whole solid behaves as a giant molecule, and nlore atoms can be added to make the
"molecule" bigger. The other group 4 elemelits sucli as silicon atid ge~~naniurn also
crystallise in tlie same structure whicli has been given die name "diamond" structure. A
closely related structure is forliied in ZnS. In ehis solid, two eleckolls of Zn and the six
outermost electrons in sulpliur get reiuranged so that each atom has an octet. Each atom
has ;i CN of 4, with tlle difference that the near neighbours of Zn are S <andthe near
neighbours of 5 are Zn. Tlle iatolns occupy the same lattice positions as i~ the diamond
structure, but tlie fact Ihal h c war neighbours are dissitnilar (in contrast lo Ule situation in
silicon or gerinanium), has interesting consequences in the interaction of the solid with
electroinagnctic ratlidion. 7 3 6 : ZnS structure has bee11 given the nallle zinc ble~ide
structure. Such structu~esa n orten found in cornpound se~niconductorssucli as GaAs,
CdTe etc.
The ionic bo~idwas seen to be non-directional. By contrast, the covalent bond is highly
directional. As tlie name suggests, covdency means sharing of tlle valence electrons and
lhis the atoms do so that Uie octet is con~pletcd.Consider the case of chlorine gas. In the
gas, C1 atom, which is a s2,v5 skucture, shares one electron wilh its neighbour. This means
that both the atoms have a s2p6 structure in the bond and we say tliat a chlorine ~noleculeis
formed and that tliere is a single boild between tl~eclilorule atoms.
Another case of the covalent bond exists in the molecule of ethylene (C2H4).In this
molecule, tliere are two H connected to each C and the two C alons have a double bond
(often written as H2C=CI-12).The double bond means n sharing of two electrons with the
neigllbour atom and if tliis bond is opened or converted to fo~nltwo single bonds, adjacent
ethylene molecules can be covalently bonded togctlier, hi tliis way a long chain can be
formed in which the repeating unit is C21-1[4.The phenonie~iollis sliowai in Figure 2.6.

(a I
--- Polyethylene molecule ---------8. --

Figure 2.6 : (a) An Ethylene Molecule ( C2F14)is Compared With (b) A PolycthyleneMoPeculo ( C2Ha)n that
Results from LCConversion of the C=C Double B O I Iinto
~ two C-C Single Bonds

This repeating unit is called a mer and the long chair~isa polymer. Such long chain
molecules are the struclural basis for polymeric materials. The long chain molecules have
sufficientflexibility to fill a three dimensional volume by coiling in a ralher complex way,
Though the bonding within a molecule is strong arid covalent, the bonding between
molecules is weak. This is the reason why most polymeric materials have rather low
st~ucturcor Materials melting points and low mechanical strengths. On the other hand, if there are no long chains
and there is a strong overlap between the atoms, as is the case in diamond, we get
incredibly high'strengths. The key word is electron orbital overlap and this is the reason
why in a covalent bond one can get small coordinalion numbers. For diamond, the radius
ratio is one (botll calion and anion are carbon) and so the CN should have been 12. What is
observed is a CN of 4.
Silicon is just below carbon in the periodic table. Does it too show covalent bonding? Yes,
and that is the basis for tnost semiconductor development. The fact that the bonding in
semiconductors is predominantly covalent means that the conductivity can be controlled
by the introduction of defects in the structure. But what about the formation of chain like
structures? Indeed Si does form long chain structures in the form of [ S ~ Otetrahedra.
~ ] ~
This repeating unit fornls sheet structures, tiamely lhe silicate struclures arid these are
found abundantly in natural rocks ruld minerals. The Si-0 bond is about half covalent
(electronsharing) and half ioi~ic(electron transfer). Tl~isgives it the unique ability to form
Si-0-Si-0-Si chains of undefined length. The three-dimensional network (SiOz),, is
found in pure silica, feldspars and clay. The [si0414 anion also forms many interesting
compounds (silicates) with the appropriate cations. One such is the garnet which is a
calcium-alutnino-silicate; another is zircon, which is zirconium silicate.
In Table 2.6 we have given the bond energies and bond lengths of a few commonly
occurring covalent bonds. Note that the bond energies are quite conlparable to ionic solids.
Note further that if atoms share more eleclrons the bond becomes stronger as is to be
expected.
Table 2.6 : Bond Energies and Bond Lengths of s few Representative Covalent Bonds

Bond Bond Energy Bond Length

(kJImole) (pn1)

C-C 370 154

C=C 680 130
C=C 890 1 20
C-H 435 110
C-N 305 150
C-0 3 60 1 40
C=O 535 I 20
C-F 450 140
C-Cl 340 180
0-H 500 100
0-0 220 150
0-Si 375 160
N-H 430 100
N-0 250 1 20
F-F 160 140
H-1-I 435 74

Example 2.4
Calculate the energy released during polymerisation of poly(viny1 chloride) PVC.
Solution
The vinyl chloride molecule is C2H3Cl.There is a double bond between the C and
C. The reaction proceeds by changing the double bond to two single bonds. The
energy released is then (2 x 370 - 680) or 60 kJImol. As this energy is released, this
is a spontaneous reaction in which the product PVC is more stable than the
individual vinyl chloride mers.
One may like to consider a valency for covalently bonded ions as the rearrangement of the
outer or valence electrons gives the stable electronic configuration. The valency can then
be considered to be the spne as the Group of the element in the periodic table or 8 N,
where N i s the number of the outermost electrons. Thus in GaAs, Ga can be considered to
have a valence of 3 as it belongs to Group 111, while' As, which belongs to tlle group V
would be considered to have a valence of 8 - 5, i.e. 3. The concept of valency in covalent
compounds is quite different from that in ionic compounds. Consider the case of copper
chloride, CuC1. In this covalent compound the valency of copper would always be 1. If
 copper and chlorine bond ionicdly, it would most probably be as CuClz , tliat is Cu would T h e Band Theoryof SON&
have a valency of 2 (the same argument holds for Ag or Au! Of course, it is another matter
mat few compounds of silver with valency 2 are stable.) The reason for talking about
valency is thal in ionic compounds, an ion can exhibit more than one valence state (Mn
exhibits vdencies of 2 , 3 , 4 , 6 and 7 , being a d 's2 configuration), in covalent compounds
the ion has a single valence state equal to the Group in the periodic table it belongs or eight
minus the Group if it belongs to Groups V, VI or VII (there are a few exceptions in the
very heavy elements such as Pb which shows a valency of 2 instead of 4). In the next kind
of bonding, the metallic bonding, we shall see that it is not possible to talk of a valence
state!
2.3.3 The Metalllie Bond
As its very n,une suggests, the metallic bond is confirned to the metallic elements, found in
Groups IA, IIA ;md IIIA of the periodic table aid in the heavy elements. Again for
bonding, the aloms should come close enough that the electron orbitals of the outermost
electrons overlap. The electrons then rearrange themselves to give a stable arrangement.
Consider tlic case of a metal sucli as copper. The electron configuration is dlo s' , and a
closed shell structure would emerge if the s electron is given up by the atom. This is the
outer~nostelectron it can attach itself, temporarily, lo a neighbouririgatom giving a
closed s2structure and then overlap with its next neighbour and so 011, thal the electron
may share some time with all the atoms in the solid. There would be as many eleclrons as
there are atoins (each atom contribuling one electron) and every electron would overlap
with every other clectron. A two-dimensional picture would be like Figure 2.7. Every
copper atom would liave given up its outern~oslelectron and the copper ions would be held
in equilibrium positions by a balance of forces: <analtractive force due to the ions and
electron sea and a repulsive force due to the exclusion principle.

Figure 2.7 : A Metallic Crystal, Pich~rcdes a Lattice of Positive Ions Ernberldcd in a '§cay of Electrons

What about lhe nlerallic bond between two dissi~nilarelements? This would be an alloy
and let us consider a "mixture" of copper aid zinc. Zinc is the next element to copper in
the periodic table and there are two outernlost electrons. Thcre should be two electrons per
zinc atom mixing with the onc electron per copper atom to form an arrangement which is
stable. As both copper aid zinc liave "donaled" electrons to thc structure, both would have
acquired a positive charge. The bond cannot be considered as similar to tile ionic bond as
there are cations and no anions. It is customary to think of tlie electrons as free to move
inside the solid 'and they are pictured as for~llinga cloud or a sea in which [lie cations x e
fixed (forming a regular array or the lattice) and there is an electrostatic force bctween Uie
electrons and the ions. Conversely, one may visualise the metallic bond as electroll sharing
and hence more akin to tllc covalent borid; except that no liybridisatio~itlakes place and the
bond is not directional. This means that the atom rearrangemen1 would favour large CN as
each bond lowers the free energy of the solid. This also meals thal there is no fixed ratio in
which the aloms of Cu arid Zn would combine to form a solid. There is no compound
formation; tltere is Uie formation of an alloy. One can also talk of the alloy as a solid
solution and talk al~outthe solubility of one element in anotlier.
One could still calculate the energy by considering the equilibrium between an attractive
potential and a repulsive potential energy term as was done in Eq;2.&-The major
difference would be that the constants A and n ould be different; Some representative
i"
values from literature for lattice energy and n &ing point are given in Table 2.7.
The iollic, covalent and metallic bonds give rise to strong solids as the bonds atc very
strong. There are possibilities of slightly weaker bonds, arising not out of electrostatic but
Structure of MateriaIs' out of dipolar forces. These are generally prevalent in molecules and electric dipoles are
induced in them due to the molecules coming sufficiently close to each other. Such solids
have low melting points and rarely have engineering applications. But it is necessary to
complete ow picture of bonding in solids. One such bond is the van der Waals bond and
the other is a special case of bonding found'in hydrogen. Let us consider these.
Table 2.7 : Bonding Energies and Melting Points of Some Metallic Solids

Metal Electron Binding Melting

Configuration Energy Point
kJ/rnole eC)

Na 3s' 109 97.8

K 4s' 90 63.7
Ca 4s' 177 ' 838
Sc 3d'4s2 342 1540
Ti 3d2 4s2 473 1670
2
V 3d34s 515 1860
Cr 3d5 4s' 398 1875
Mn 3d5 4s2 279 1245
Fe 3 8 4s2 418 1536
Co 3d7 4s2 383 1495
Ni 3 8 4s2 423 1453
Cu 3dL04s' 339 1083
Zn 4s2 131 419.5
Ga 4s2 4p 272 , 29.8 ,

2.3.4 The Van der Waals Bond

This is a type of bond which does not lead to solids with sufficient strength for structural
applications but we come across these bonds in some plastics and waxes. The idea stems
from the fact that at very low temperatures the inert gases also condense to form solids.
There are no unfilled shells or atoms that can be ionised and no spare electrons available to
form bonds. Yet there must be bonds as the solids do get forme . There must then be
f
another kind of bonding, called secondary bonding, responsib e for these bonds in which
no major modifications of the electronic structure of the atoms takes place.
The idea that the elec&ns move in orbits around the atom means that it can have no
permanent dipole moment, averaged over a sufficiently long time. However, one can
consider this motion of electrons as giving rise to a rapidly fluctuating dipole moment, so
that at any particular instant the centre of the negative charge distribution doesnol
coincide with the nucleus but rapidly fluctuates around it. When two such atoms come
close together, the rapidly fluctuating dipole moment of each atom can influence the
rapidly fluctuating dipole moment of the other and a lower energy structure would result if
they fluctuate in phase. This can be seen from Figure 2.8 (a) and (b). In Figure 2.8(a) there
are some situations shown that would always lead to a repulsion and energy would be 1

[a) Ibl

Figure 2,s : Illustrating the Origins of Van der Waals ~ o r c e

increased if the atoms come together. But in Figure 2.8(b) we show situations in which the
energy gets lowered when the atoms come together as there exists an attractive force
between the atoms. This is the basis for the van der Waals bond. As the bonds are not very
strong, such solids have low melting tepperatures and are easily .deformable. Notice that
I
 the attractive forces act at sufficiently large distances so that overlap, and so repulsive
forces, become appreciable only when the atoms come very close together. Analogous to
Eq. 2.7 we can write, for the potential energy:

where r is the distance between the atoms. In literature this form of potential is known as
the Lennard - Jones potential. By sunlming this potential over all atom pairs one can
calculate the binding energy. It should be noted that, as the bond does not have a
directional character, an atom would like to have as many near neighbours as possible.
'Illis is seen in solids formed by cooling the inert gases; Ne, Ar, Kr and Xe all solidify in
the FCC structure, in which each atom has a coordination number of 12. As we shall see
later, though the van der Waals bond does not give strong solids it coexists with the
covalent bond in graphite and in many polymers giving them some very useful properties.
Within the planes, in graphite, the bonds are covalent. But between the planes the bond is
of the v,m der Waals type. Graphite acts as a good lubricant whe~ithe planes can easily
slide over one another.
2.3.5 The Hydrogen Bond
To complete our study on bonding we have lo consider llie special case of hydrogen as in
the, extreme ionic case, hydrogen forms a bare proton and can consequently attract two
atoms with negative polarity. In this way the hydrogen ion can form a bridge between two
electronegative atoms. This bonding can be seen in water, hydrogen fluoride,
acid etc. In Figure 2.9 we show how Uie hydrogen bond is formed. In the extreme
water can be considered as H+(o~-H+);the proton, can
then attract
another 02-forming a larger molecule and so on. III formic acid, tlie H in COOH can also
form a bridge with another COOH iieighbour. The bridging behaviour of the hydrogen
atom gives rise to the hydrogen bond. This bond is not as strong as the ionic or covalent
bond, but plays a inajor role in biological molecules,

Dipole

Figure 2.9 : 'mydrogen Bridge". This Secondary Bond is Formed bctwew~Two Pcrrunncnt Dipoles in
Adjacent Water Molecules -

2.3.6 Mixed Bonds

Though, at first sight, the primary bonds appear distinctly different, in reality it is often
very difficult to classify solids as belonging to one type or anotlier. lliis is very true in the
case of Uie ionic and covalent bonds as b o h need some rearrangement of the outermost
bonding electrons. Lel us consider Ule case of solids formed in a particular row of the
periodic table as we begin from the outer columns and proceed to the inner columns. That
is, consider the following set of comnpounds: LiF, BeO, BN and C. In LiF, the bond is
unquestionably ionic and that in C (diamond) definitely covale~it.Be0 and BN must have
bonds which would be a mixture of ioklic and covalent as there is a finite probability of
electron transfer as well as electron sharing. That boron nitride has a large amount of
covalency in the bond arises from the fact that the properties of BN arc very diamond like.
BN has a hardness next only lo diamond.
How then should one classify the bonds; or can one determine the amount of ionicity or
covalency of a bond? This can be done by assigning a quality luiown as electronegativity
for the elements, the values for which are given in Figure 2.10. Electronegativityis defined
Stmcture ot Materlds a's the degree to which an atom attracts electrons to itself. It is measured on a scale from 0
to 4.1, and each element is assigned a value on this scale. The most electropositive
elements are those which readily give up their electrons, i.e. the alkali elements; the most
electronegativ'r:elements &e fluorine,oxygen and nikogen.

H
2.2
Li Be B C N O F
1.0 1.5 ,
2.0 25 3.1 3.5 4.1

Na Mg A1 Si P S C
1.0 1.3 1.5 1.8 2.1 2.4 2.9

K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br
0.9 1.1 1.2 1.3 1.5 1.6 1.6 1.7 1.7 1.8 1.8 1.7 1.8 23 2.2 2.5 2.8

Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I
0.9 1.0 1.1 1.2 1.3 1.3 1.4 1.4 1.5 1.4 1.4 1.5 1.5 1.7 '1.8 2.0 2.2
Cs Ba La Hf Ta W Re 0s Ir Pt Au Hg TI Pb Bi Po At
0.9 0.9 1.1 1.2 1.4 1.4 1.5 1.5 1.6 1.5 1.4 1.5 1.5 1.6 1.7 1.8 2.0

I t; I Acf 1 Lan,~ayis 1.0 - 1.2

1 -0Actini es 1.0 - 1.2
Figure 2.10: Electronegativitiesfor the Elements
The electronegativity concept lielps us in understanding the bonding behaviour in solids
ernphasising the fact that the bonds are of a mixed type.
We can then state some rules of bonding:
1) Two atoms of similar electronegativity form a metallic bond if they can donate
electrons; form a covalent bond if they can accept electrons.
2) When the electronegativities differ, the bond is ionic; the ionicity of the bond
increases with the increasing electronegativity difference.
The percentage ionic character of bonding in a compound XY is often written as:
% ionic character = (1 - exp [(- 1/4)(tx - ty)2])x 100 ... (2.11)
where cx and tr are the electronegativities of elements X and Y respectively.

SAQ 5
Calculate the percentage ionic character of the compounds NaCl and Gap.

The mixed bonds mentioned above are mixed ionic-covalent bonds. Can we have mixed
metallic-covalent bonds and metallic-ionic bonds also? The answer is yes.
Metallic-covalent bonds are quite common and one can see a strong covalent cliaracler in
diamond. But as one goes down the periodic table the bonds in Si and Ge become
progressively more metallic in character till we come to Sn and Pb in which the bond is
predominantly metallic. In the transition metals, too, one sees evidence of mixing up of the
orbitals of the d, s and p electrons. The high melting points of these elements is ascribed to
hixed metallic-covalent bonding. Metallic-ionic bonding is rarer because it requires a
large change in electronegativities; but evidence of such mixed bonds are seen in
intermetallic compounds such as NaZnl , Mg17Al12or Fe5Zn21.

2.4 ENERGY LEVELS IN SOLIDS

The determination of the electron energy levels in a solid is complicated by the fact that
the number of electrons per cubic meter of a solid is very large, of the order of in
simple metals. We have seen that the electrons obey Schrtidinger's equation; but the
estimation of the potential energy V is quite difficult as we would have to take into
account the electron-electron interactions, the eleclron-ion interactions and so on. One way
 to get out of this difficulty is to consider the electron levels in a single atom and see how T h e Band Theory of Solids
&Jeseare affected when we bring another atom close by, and then a third atom and a fourth
and so on. The actual solution of Scllrtidinger's equation is rather messy and we will
obtain the results using qualitative arguments only. Whether our conclusions are right or
olherwise would be assessed by taking recourse to experimental findings. We of course
assume that the experimental results have been taken with the utmost care. Needless to say
&Jatmany a times new experimental results have resulted in new theories!

2.4.1 Degenerate Levels

For a single electron in an atom (like hydrogen), we can solve the equation of motion by
taking the potential energy V = -
zc2 . where r is the distance between the electron and
47t~{)I
: the nucleus, Ze the nuclear charge and Q the permittivity of free space. The result of the
, calculation is that Ule electron energy is given by:

exactly as in the Bohr model discussed in sub-section 2.2.1, witli R a constant. In this
calculation the eleclron orbit is taken as spherical and so only one quantum number, n,
comes into play. If.we include non-spherical orbits then the energy would depend on two
quantum numbers n and I. Each orbit is denoted by these two quantities as tlie Is levels or
the 3d levels and so on. Including spin, the Is levels can acconlrnodate 2 electrons, the 2p
levels 6 eleclrons , the 3d levels 10 electrons and so on. This is on account of different
values of the quantum numbers m 'and o , consistent with the Pauli exclusion principle.
Electron energy levels or states descrilwd by dii'ferent quantum numbers but with the same
energy are called degenerate.
The degeneracy in the levels can be removed or lifted when the free atom is placed in an
extemal electric field. Under lhis field the electrons with different m will have different
energies. This is shown in Figure 2.1 1 for thc 3d level. Note that the d level has split up
into 5 levels. As the interaction does not depaid upon tl~espin, here is still spin
degeneracy, i.e. eadh level can accommodate two electrons with opposite spins.

External External
field field
b switched switched
off m =- 2
I /

Total number
of levels = 2 l +1

The spin degeneracy can be lifted under UI external magnetic field. Eleclrons then ccul be
excited from one spin state to another. This is the basis of the magnetic (spin) resonance,
which has gained importance of late in the field of medicine.
2.4.2 Lifting of the Degeneracy in Crystals
Consider the building up of the cry. ,a1 one atom at a time. For a single atom the electron
levels are as discussed in the previous section. Now, when anolher atom comes close
enough for the atomic orbitals to overlap, it becomes possible for electrons to be shared by
the atoms. The electron energy levels must change. In sub-section 2.2.3, we had seen that
what happen when two hydrogen atoms can come together to form a stable molecule. This
is because the energy gets lowered when the electrons, with opposite spins can occupy the
s m e orbit. There are two possibilities, then: one with both electron spins parallel (and the
electrons occupying orbitals wilh different n, 1 or m) and the second in which Ule spins are
 anti-parallel (and the electrons have the same n, 1 and sn ). The latter has a lower energy
than the former as we had seen in the hydrogen molecule formation. This means that the
degeneracy of the level has been removed and one m y say that the magnetic quantum
number m has split up exactly as if an internal eleckic field exists. Each of the m levels
can accommodate at the most two electrons with opposite spins. When a third atom comes
close for electrons to overlap, the electron levels split into two sets of three levels each,
one set of lower energy and the other of higher energy. Similarly, if there are N atoms with
overlapping electron orbitals, the electron energies would split up into two sets, each set
consisting of M energy levels. The two sets would be sepcu'ated in energy, and in the
ground state only the lower energy levels would be occupied. If the lattice is made up of
many-electron atoms, like say silicon which has an electronic configuration of
Is22s22p63s23p2 , the outermost orbitals (of n =3) would begin to overlap first as the atoms
come closer and the overlap of the inner levels would be smaller. In any case each level
would broaden out into a band containing as many levels as overlappigg electrons. If the
overlap is more the band spreads more in term of energy. Now, energy minimisation
occurs when electron can get paired due to overlap. In silicon this is possible by sp3
hybridisation and in these orbitals all electrons have parallel spins. If the neighbouring
atom has spins which are anti-parallel,close overlap is possible (otherwise Pauli exclusion
would prevent the atoms from coming closer) and very stable bonds can be formed. The
broadening of the single atom electron energy levels into bands due to another atom
coming close is shown schematically in Figure 2.12. One must note that in a crystal of
volume V which 1x1s Nelectrons, each band has Nlevels and as each level can take two
electrons, the band would get filled when 2N electrons occupy it. The formation of bands
is then a direct consequence of the interaction of the electrons with the ions. This
interaction lifts the degeneracy of the free atom electron levels, and causes the spreading
out of single levels into b'mds. Another way of looking at this electron-lattice interaction
will be discussed in sub-section 2.4.4.
2.4.3 filling up of the Energy Levells by Electrons
It is now well understood that, in order to describe systems at the microscopic level one
needs to use quantum mechanics. In quantum mechanics, to every particle there is an
associated wave and one cannot distinguish between the wave-like or particle-like
behaviour of the quantum particle. So, to consider the electron only as a particle is not
altogether correct as it does show wavelike behaviour; electrons are diffracted by solids
and this fact is used in some modern day instruments such as the electron nlicroscope. In
order to consider such situations, the correct equation of motion is the Schrodinger
equation. As we will be considering only equilibrium energies, we clloose the
time-independent Schrddingerequation, which, in one-dimension is:

In this equation, y is a wave-function which represents the behaviour of an electron in a

solid, which has a total energy E and moves in a potential V. Again we have assumed a
one-dimensional solid for simplicity. To understand the response of electrons to, say, an

of Single Elcctmn Level bcing Brodened into Bands due to the A t m s Coming
Figure 2.12: Illustratio~~
Closer
 external field, we have to choose a value of which is representative of the system. If .we The Band Theory of solids
consider the electrons to be free, i.e. the potential energy V to be zero, then the electrons
can be represented as plane waves. yr is then proportional to exp(- i a). It is easy, and ,

instructive,tosolve this equation. The first term is equal to - K%. The total energy is then:

We should notice here that the result can be interpreted as E = yL /2m (where p is the
momentum), which is just the kinetic energy (as the potelitid energy has been taken as
zero, this is also tlie total energy). This is a classical result, one which also arises from
hk
Newton's second law. The relationship between p ar~dk is p = - ,just the statemsnt of
2a
the de Broglie equation, sllowing wave-particle duality. This then is clear: that the
Schri3dinger equation is an equation of motion, like Newton's laws, and gives an adequate
picture of the behaviour of electrons, endowing dlem with boll1 arl energy and a
mornenturn, which are classically equiv@entto frequency and wavelength.
How does one use this for electrons in metals? Well, we laow that eleclrons normally do
not escape spontaneously from a metal. But electrons do get elizittcd when a metal is
heated m~ldwhen a high electric tkld is applied to it. Phis is the pri~~ciple
(thermo-emission) on which devices like cathode ray tubes and television tubes work. So
one can assu~nethal Uie electron in a rnetal is in a pote~itialwell; i.e. inside the inetal it
experiences a consca~~t potential but for it to come out it has to cross a potential barrier as
outside tlic potential is very high. This idea can bc simplifietl by writing tlhc conditions for
V as for x between 0 and L, V =0; aid for other valucs of x it is infinity. The sdution to
Eq.2.13 would then give Uiat yr is finite f(w x between - 0 iu~dL and zero outside Chis rmge.
A

/lL lCL
The energy E, betwecri n = O and L would be -. But what values could K take? By its
8 n2rn
fundamental nature, ty is continuous will] x; i.e. yr (x = 0) = yr (x = E) = 0.
This would mean that ex]) (- i KL)= 0, and SO:

where n is a]integer. The lowest value for

x 2n 3n
is - , Ule next value is -- , the next -
K
L L L
and so on. No intermediate K viilues are allowed ax~dthat mcnlw that the enerf;y valucs are
also discrete. We then have:

with n taking integral values of 1,2,3, ... etc. Figure 2.13 gives Ule wavc hnction, tlie
probability of finduig the electron, i.e. y * q~ and the energy values for (he first tluee
states.

Figure 2.13 : (a) h ~ t c t i o nand (b) Probability Fupction y * w for an Electron in n Potential Well for
1)rneaent N-values

When we generalise the problern lo three dimensions, the quantunl number n should be
replaced by nx , n, w d n, witli n2 = n,2 + ny2 + n: . Each elprgy state is then
specified by a set of three quantum immbers (because of the ~ree-dimensionalnature o f
the problem) and we have to add aioher q~iantumnumber, the spin. You can see that the
Structure of Materinls ' energy state with n, , n,, and n, .respectively equal to 5,1 and 1 has the same energy
value as the state 3,3 and 3:Also the state 3 , l and 1 is degenerate with the state 1,3,3.
SAQ ti
In a cubic crystal, what is the degeneracy of the state 3,2,1 ?

SAQ 7
Wh3t is the energy, in eV, of the ground state of an electron confined to a cube of
side 10 nm?
[The ground state is the energy state with the lowest allowable energy.]

We have thus seen that the electron obeys Schriidinger's equation and has the energies E
given by Eq. 2.16 for the case of an electron which is confined to a length L, but is free,
3.9. V = 0.This means that an electron which is confined to remain within a definite region
in space should have energy levels that are discrete. These are the one-electron energy
levels. How can we use this result for a solid which contains rnany electrons? Let us say
that there are N electrons which occupy the solid of volume V.One way is to strip the solid
of all the N electrons and then fill in one electron at a time such that each new electron
occupies the appropriate level. We can introduce electron-electron interaction in the form
of the Pauli exclusion principle so that each energy level can be occupied at most by two
electrons with opposing spins. The lowest energy level would be occupied by the first two
electrons, the next higher level by two more and so on. Remember that n = N /V is a very
large number. It makes a good approximation to assume that the elecgons. would occupy
levels so that all the occupied K levels lie within.zsphere of volume
L 3j u$ , where NP
is the highest occupied K level. This levei is called the Ferrni level. s e have restricted
K to only positive values, we should only consider the positive 1/8th rcgion of the sphere.
The volume in questioh would then be -
($1 r . This volume then contains all the N

electrons. But each electron occupies a region in tc space equal to

2
-
v ; we get this by
recalling thatin one dimension, there is a single K state in a linear u region of nJL, and

then easy to see that :

rzT "
so in three dimensions the region occupied by a single r state would be - , or
\ 1
It is

This gives us:

!4 ... (2.18)

&id the Fermi energy

 All the electron states from E = 0 to E = EFwill be filled up by the N electrons. All the '~%eBand T~ISIYOP&E~P

levels above EF will be unoccupied. As each occupied level coiltainh two electrons, an
electron can be excited from a filled level only to a vacant level. As ffie normal excitation .,

energies are small compared to Ep, the number of electrons near Ep assurnegeat
importance. Can we determine this number?
To calculate the nunlber of levels per unit energy, also called the density of energy states,
we can use Eq. 2.18 and determine dN as the fraction of energy states as a function of the
energy by differentiation as follows:

, Differentiating iV with E as a variable, we get:

. .
The quantity D(E) is known as the dendtgr off energy states and is proportional to E In
By definition it is easy to see that:

With the help of D(E), we can determine the number of electrons that would be excited
when a solid is subjected to an external field such as an electric field. All ffiat we would
need to estimate is the change in the energy of the electron under the external influence.
How would this nulrlber change wit11 temperature? Well, to answer this question we would
have to consider the probability that a state with energy E is excited at a given temperalure
T. For electrons we have to use Fenni-Dirac statistics to estimate [his probability. The
probability of occupation of a state E is Uleli given as:

In this equation k is the B o l t z w u ~constant, Ep is the Fenni energy and T is the

temperature in question in K.If we include the probability in our calculation, then the
equation for iV would be:
EF ... (2.24)
I D(E)J(E)dE = N
0
The Fermi-Dirac function has ffie property that for E < Ep its value is uility and for E > EF
it goes to zero. For ( E . EF ) > kT, the functionf (E) rcduces to the well known
Maxwell-Boltzmann probability function. For i~onnalmetals, Ep is very large compared
to h e thermal energy, So, increasing the tcinperaturcof tlle mctal by a few hundred
I degrees does not greatly alter the number of electrons that participate in the conduction
process. Another way of saying the same thing is that the change in the cnergy of tile solid
is not due to the excitation of electrons when the temperature is raised; therefore the
contribution to the specific heat of the solid by electrons is very small. The major
contribution comes from the lattice vibrations. As this is corroborated by experimental
observation, we feel confident that our understanding of the processes in solids is valid. So
the picture of the solid is that the electrons occupy all the states from 0 lo Ep ,uld only
those electrons are resparuible for conduction that can be excited to vacant states by the
electric field. The Fermi level is the highest energy level, at zero K, that is occupied by
electrons in a metal. 11e number f states from 0 to EF is N p d these are discrete with
f
each state being occupied by 2 ele &on(. We mugdow consider the formation of bands.
2.4.4 Formaedn of Bands
In the last section we saw how all the levels from E = 0 to alarge value of energy, i.e. the
Fermi energy are all occupied. The E versus K curve is a parabola as she? in
Figure 2.14. As K also gives the direction in which the electron wave is travelling, for
generality we have plotted both positive and negative values for K . We have to now
Stm- d Materials include electron lattice interactions so that as the wavelength of the electron wave becomes
smllei it begins to see the periodicity of the lattice in such a way that the lattice acts as a
grating and the wave gets diffracted. Such diffraction is called Bragg diffraction Bnd
occurs when n h = 2a sin 8. Ln this equation n is an integer, h the wavelength of the
2K
electron wave, which is really -, and a is the distance between the atoms (or the near
K
neighbour distance). 8 is the angle of incidence of the wave, which for our
one-dimensionalcalculation is 90°.
e
-
\ I

Figure 214: Extended Zone Scheme. Tl;efirst and Sccond B d l o d n Zone are Shown

n 7c
This means that diffraction takes place whenever K = k -. In one dimewion, difiraction
a
is really reflection. Now the ongoing wave would interact with the reflected wave to give,
rise to standing waves. In such situations, no wave propagation would take place. Further
as there would be two wave functions possible for the electrons, i.e. exp(+i K x ) and
exp(4 K x ) a linear combination of these wave functions should be taken to correctly
represent the state of affairs. The positive combination would give cos( K x ) and the
negative combination a sin ( K x ) wave function. The two functions would give two
different energies, the difference being representative of tbe strength of the interaction that
an electron would have with the ion. The E vs. K curve would then break up into bands
so that the fist band would comprise the states between - 5 r 5 + -. This region in k
K
a a
states is called the First Brillouin Zone. The second Brillouin Zone would be the region
2.n n n 2a
from - - up to - - and from + - to + -. How many states ake there in each Brillouin
a a a . a
zone? In one dimension we had taken a length L for consideration. This length, if it
-contained N atoms would really be equal to Na. Again only considering positive values for
K
K in our one dimensional calculation, the number of states from 0 to -
a
would be N, i.e.
!

equal to tbe number of atoms being considered. Generalising then, the single energy
spread breaks up into bands due to electron - lattice intera tion and in each band there are
9e
N states, with each state being able to accommodate two' el ctrons. The density of states
within a band is plotted schematically in Figure 2.15.
\

DIE
_.I_

Figure 2.1$ : Schematie Reprwhtation of the Density of States Function within a Band
 Note that as the number of states becomes filled the density of available states becomes The~ ~ ~ hn e odOP
~ ysolids
less and when the band is completely filled the density of states becomes zero. Somewhere
in the middle of the band the density of states goes through a maximum. We shall see that
this is of imporlance in the classification of solids.
2.4.5 The Band Gap
nn
As seen in the earlier section, at every Brillouin zone edge, i.e. when K = It -, a standing
a
wave forms. An electron wave cannot propagate (its velocity is zero) and the electron
energy splits up as shown in Figure 2.14. Between the two energy levels, electrons cannot
have those energies and these states are said to form an energy gap. The extent of the
energy gap, as again seen in the previous section, is a measure of the electron-ion
interaction. As the electron in question moves further away from the nucleus, the
interaction strength should decrease. That is why the energy gap reduces as we go down
the periodic table. Lighter elements show energy gaps which are quite large. Remember
that the band gap separates energy levels which may or may not be occupied. To determine
which states are occupied we would have to bring into consideration the Fermi-Dirac
distribution function. We shall see, in the next section that this leads us to a natural
classification of solids into conductors and insulators.

2.5 CLASSIFICATION OF SOLIDS ACGORIPING TO

THEIR ELECTRICAL PROPERTIES
We hav seen that the electrons occupy specific energy levels in solids; that under normal
ternpenftures, the difference in energy levels is smaller than the average thermal energy of
the electrons and so the energy levels can really be viewed as bands. Further, the number
of erectrons that can occupy a band is 2 N , where N is the number of atoms in the volume
under consideration. It should be easy for you to see that a filled band does not contribute
to conductivity as the electrons must be excited to vacant states as Pauli's exclusion
principle is the guiding rule for electron occupation of the levels. These ideas give rise to a
simple understanding of the phenomenon of electrical conduction in solids.
2.51.1 Conductors
s good conductors of electricity. In the periodic table the metals occupy groups I,
~ e \ a lare
I1and HI; also most heavy elements show metallic behaviour. The combination of two or
more metals, i.e. the alloys are also generally good conductors. As the crystal structure of
metals are body-centred cubic, face-centred cubic or hexagonal close-packed (discussed at
length in the next unit), conductivity does not depend strongly on the crystal structure. It
depends only on the number of electron levels that are occupied and the number available
to be occupied. Group I elements have only one electron in the outermost shell which form
a band with twice as many states. This means that in this group of elements the electrons
only fill up half the number of states in the outermost band. Electrons can be easily excited
to higher unoccupied states by the application of a small electric field. This leads to good
-

Figure 2.16 :Energy Bands for (a) Insulators (b) A I M Met& (c) Bivalent Metals and (d) Semi
Conductors. Note the Position of the F e d Energy, Ev. I

conductivity. A similar argument would explain the good conductivity of Group 111
Structure of Materials elements, in which two electrons would fill up one band and the third electron would half
fill the outermost band. Examples of good conductors in Group I are copper, silver and
gold; examples of Group III conductors are aluminium and gallium. You must have come
across the ust?of all these elements in the' making of contacts either as connecting wues or
as cogducting surface coatings. In all these cases, the Fend level is near the middle of the
topmost band, as shown in Figure 2.16. The density of avdlable states is very high and
these metals are very good conductors.
If our argument is true, the elements of Group 11, which have two valence atoms per atom,
should be non-conductors. But we hiow that magnesium and calcium show metallic
conduction. This can be explained as follows. In these metals the energy gap between the
highest occupied band and the next higher band does not exist as the two bands overlap. h
fact, before the first band is completely filled some states of the next band get occupied.
The Fermi level lies in the region of overlap. This is shown in Figure 2.1 6. One can see
that the density of available states is small and so, though these materials show metallic
conduction, they are not as good conductors as the Group I and 111 metals.
2.5.2 Insrllators
As can be easily followed from the discussion in the previous section, those materials
which contain an even number of electrons per atom and in which band overlap does not
take place, should be insulators. Excitation of electrons from one band to the next should
be very difficult in good insulators and the band gap should be more than a hundred times
the thermal energy. The thermal energy is 0.26 eV and material which have band gaps of
3 eV or more are considered to be good insylators.
What kind of bonds would give rise to insulating materials? Well, any bond that leads to
completely filled electron shells! This means that both ionic and covalent bonding would
give rise to insulators. Ceramics, which are ionic, such as MgO or alumina are very good
insulators with electrical resistivity 1016times that of metals, typically. Historically,
ceramics were used for their insulating properties, to act as supports for high tension wires
and now as substrates for integrated circuits. Covalently bonded solids such as the
polymers would also show high resistance. Rubber, polyethylene and polystyrene are well
known insulators with resistivity even higher than that in ceramics. It is for their good
insulation properties that household copper wires are coated with rubber or polyethylene to
prevent shorting. Also, an electrician uses cellophane tape to cover open wires in order to
prevent inadvertent shocks.
In the band diagram for these materials, where would the Fermi energy lie? As shown in
Figure 2.1 6, the topmost band is completely empty, the lower band is completely filled.
The density of states above the highest filled level is zero. The Fermi level could have
been taken at the highest filled state or the lowest available state. It is customary to
reinterpret the idea of the Fermi level to take the probability of occupation of the states and
place it in the middle of the band gap signifying that the probability of a state being
occupied in the empty band should be equal to the probability of a corresponding state
becoming empty in the filled band. This point is discussed in more detdl in Unit 5.
2.5.3 Semiconductors
This is the case in materials in which the energy gap between the completely filled and
completely empty band, at zero K, is about 1 eV. In this case there would be some
electrons which would be excited from the filled band to the empty band due to t h e m 1
energy at room temperature. Such materials would have been insulators at zero K, or at
very low temperatures, but show appreciable conduction at room temperature. The most
important of such compounds are silicon and germanium. Let us see the band formation in
this case.
Silicon and gennanium belong to Group IV of the periodic table. Each3atom has 2s and 2p
electrons in the outermost shells. These give rise to hybrid orbitals (sp hybridisation) and
four bands get formed. The four electrons per atom completely fill up two bands and two
bands remain completely empty. This is shown in Figure 2.16, where for the s&e of
simplicity we have only qhown one filled and one empty band. The band gap decreases as
we go down the periodic table. That is why, at room temperature, carbon in the diamond
structure, is really an insulator with a band gap of mbre than 5 eV. Silicon comes next with
a band gap of 1.1 2 eV, followed by germanium in which the band gap is 0.67 eV. The
heavier elements in the same group, tin and lead, show metallic behaviour, with zero band
46 gaps .
 The Bawl Theory &Solids

In this unit we have taken a look at the way in which solids get formed and the role that the
valence electrons play in this. We have worked out a rather simple theory starting from the
I
model of a free atom like hydrogen and using quantum mechanical ideas. It was seen that
I
I
electron orbits are characterised by four quantum numbers, three because of the three-
I
dimensional nature of our world and one more due to the electron spin. Then we used the
I s m e ideas to consider atoms with many electrons, filling the available energy levels
according to the Pauli exclusion principle, so that no two electrons could occupy states
with the same set of quantum numbers. The states were designated by the principal
quantum number, followed by the letter signifying the value of the orbital quantum
number, hence the states would be say 1s or 2p or 4f etc. Each state, or shell would be
composed of sub-shells according to the magnetic quantum number, in. When a shell is
completely filled, the reactivity of the atom is the least and this explains the nature of the
inert gases. The periodic table was understood as a natural consequence of the ideas. Then
starting from the formation of the hydrogen molecule we saw the various bondings in
solids. Bonding could be easily understood as a tendency for an atom to assume an inert
gas structure by giving or receiving the appropriate number of electrons. By giving the
excess electrons (excess over the inert gas structure) the atom would acquire a positive
charge and become a cation; by receiving electrons the atom would become an anion.
Cations and anions would come close together to form a solid by ionic bonding. In such
A
solids one could describe the bond length, i.e. distance between the atoms as equal to the
sum of the radii of the cation and anion. The strength of the solid was also a reflection of
the electrostatic energy, so that compounds with ions of a higher charge would be more
stable and have very high melting points. Inert gas structure could also be attained by
sharing of electrons. This was the key to the covale~ittype of bonding. This bond too gave
rise to very strong solids. Both the ionic and covalent bond did not lead to free electrons
and so the solids that were generated were inherently electrical insulators. Solids could
also be formed by atoms being stripped of the excess valence electrons so that the valence
electrons could be considered to be shared by d i 4 the atoms. This was the metallic bonding
and gave a simple feel for the high density in metals as the ions tended to be packed as
close to each other as possible. We then considered the energy bands in solids arising out
of the overlapping of atomic orbitals. We saw how balds cni be completely filled by
electrons to yield insulators or incompletely filled by them to yield metallic behaviour. We
saw that the density of electron states varied with energy and showed a maximum
somewhere near the middle of a band. Also, a completely filled band would not give rise
to electrical conduction. Hence, for practical purposes we needed to consider only those ,
electrons near the Ferrni level as well as the density of states at the Ferrni level. The idea of
the band gap, or the energy needed by an electron to go from an occupied state to an
unoccupied state was developed. If the excitation was possible by thermal means, the solid
was termed as a semiconductor, a special case of the insulator, i.e. an insulator with a
small band gap. !

2.7 KEY WORDS

Alloy : A mixture of two or more ~netals;it can also be a mixture of
metals and non-metals.
Atomic Orbital : The region in space about the nucleus of an atom in which
the electron is most likely to be found.
Bragg Diffraction : Constructive interference of waves reflected from atomic
planes. The waves may be X-rays, electrons or neutrons. The
condil' In is that the emergent waves have a phase difference
equal to an integral multiple of the wavelength.
Brillouin Zone : The region in momentum space that encompasses the energy
levels in a band. The region with the lowest momentum
states is called the First Brillouin zone. At the edge of the
zone the momentum values satisfy the condition for Bragg
diffraction.
Structure of Materials Co-ordination Number : The number of anions that surround a cation in an ionic solid.
It is in a general case the number of equidistant nearest
neiglibours to an atom or ion in a unit cell of a crystal
structure.
Covalent Bond : A primary bond resulting from a sharing of electrons. It is a
directional bond and frequently involves the overlapping of
half filled orbitals of two alo~ns.
de Broglie Relation : The relation between the wavelength and momentum of a
quantum particle. The relation is h = hlp, where h is the
wavelength of the wave andp is the associated momeliturn.
Degenerate States : Electron energy states with the sane energy.
Density of Energy The number of discrete energy states per energy unit in an
States energy band that electrons can occupy in a solid.
Etectric Dipole Created when two equal and opposite charges are separated
Moment by a finite but small distance. The dipole moment is the value
of the charge multiplied by the distance between them.
Electron Orbit A region in space, associated with an atom, in which an
electron with a given set of quantum numbers is most likely
to be found.
Electron Shell A group of electrons with the same value of the principal
quantum number. This group is further divided into subshells
in which electrons have the same orbital quantum number.
Electronegativity The degree to which angatom attracts electrons to itself.
Energy Band Closely spaced group of discrete energy levels in a solid. The
difference hi energy levels is much smaller than the thermal
energy 'and very little energy is needed to excite an electron
from one level to another.
Energy Band Gap : The difference in energy between the top of the highest
occupied band and the bottom of the first unoccupied band.
The band gap is usually much larger than the thermal energy.
Fermi Level The highest occupied electron state in a metal at zero K It is
also the electron state, at a finite temperature, for which the
probability of occupation is exactly half.
Permi-Disac Statistics : The statistics giving the probability of occupation of an
energy state taking into account the Pauli exclusion principle.
Ground State The quantum state with the lowest energy.
Hydrogen Bond A special case of the permanent dipole-dipole interaction
between hydrogen and another electconegative atom to form
a molecule.
Hybrid Orbital An atomic orbital formed when non-equivalent orbitals of an
atom combine. The process of rearrangement of orbitals is
called hybridisation.
Ionic Bond A primary bond resulting from the electrostatic attraction of
oppositely charged ions (cations and anions).
Ionisation Energy : The energy required to remove an electron from its ground
state in an atom to infinity.
Madelung Constant : A number which takes into account the force of attraction
between distant neighbours in an ionic lattice. It is a constant
for a given crystal structure.
Mekalltc Bond A primary bond formed when outer electrons of an aggregate
of metal atoms are delocalised to form an electron charge
cloud. This cloud interact. with the stripped metal ions to
form a non-directional bond.
 Octet : The eight electrons in an atom that form the s2p6structure in The and ~ h e o r yof S ~ d~Y
an atom.
Pauli Exclusion : The statement that no two electrons can have the same four
Principle quantum numbers.
Quantum Numbers : The set of four numbers necessary to characterise an electron
in an atom These are the principle quantum number, n, the
orbital quantum number, I, the magnetic quantum number,
m, and the spin quantum number o .
'Solid Solution : An alloy which is a single phase atomic mixture.
Valence Electrons : Electrons in the outermost shells which are most often
involved in bonding.
Van der Waals Bond : A secondary bond resulting from an interaction of the
fluctuating dipoles in a solid. The bond is weak and
non-directional.

SAQ 1
This is really an exercise in the use of S. I. units. e = 1.6 x 10;'~ C .
m = 9.1 x kg; eo = 8.85 x loeL2Flm; and h = 6.63 x ' ~ o - ~ ~ ';Jhence
. s the
result.
SAQ 2
The radius of the orbit is proportional to n2. The first orbit has a radius of 52.9 pm;
hence the third orbit would have a radius of 9 x 52.9 or 476.1 pm.
SAQ 3
The electronic structure would be [ls22s22p63s23p63d104~24p64d105~25p6 ]4$6s2*
As the structure in the square bratkets is that of the inert gas Xe with Z= 54, the
structure is often abbreviated as [Xe] 4f '6s2.In many books, however, the structure
of Gd is given as [Xe] 4f 75$s2. TO determine which of the two is correct, we need
some more information about the element. Knowing that the element prefers a
valence of 3 and does not exhibit a valence of 2, would lead us to prefer the latter
electronic structure.
SAQ 4
The cation to anion radius ratio is 721140 or 0.52. From Table 2.3 the most probable
CN for Zr is then 6 . But this would give the CN for 0 to be 3. The other possibility
is for Zr to have a CN of 8 and 0 then to have a CN of 4 . We shall see in the next
unit that the latter is the correct answer. This problem should teach us to be wary of
the detexminatjon of ionic crystal structures based only on the radius rules!
SAQ 5
NaCI: The electronegativitiesfor Na and C1 are 1.0 aid 2.9 respectively. As the
electronegativities are quite different the bonding would be predominantly
ionic.
% ionic character = ( 1 - exp{-1M x 1.g2 }) x 100 = (1- 0.406) x -.100 =
59.4%
GaP: The electronegativitiesfor Ga and P are 1.8 and 2,l respectively. As the
values are close the bonding is predominantly covalent.
9% ionic character = ( 1- exp {-I14 x 0 . 3 ~)) x 100 = (1- 0.978) x 100 =
2.22% ,
Structure of Materiala SAQ 6
In a cubic crystal, the symmetry is such that interchanging'theX, y and z axis does
not change the energy of the state. Also note that as,theenergy has squares of the
quantum numbers, changing their sign does not change the energy. So the energy of
3,2,1 is the same as 1,2,3or -1,2,3 and so on. The numbers 1,2,3 can be rearranged
in 6 different ways and as each number can take +ve and - ve. values, that is change
in two ways, the degeneracy works out to be 6 X 2 X 2 X 2, i.e. 48.

The ground state is the state with the lowest value for $E quanpm numbers n%i ny
and n, . The lowest values are 1,l and 1. So, E = (3 h )/(2mL ), with L = 10 m.
hlting the values for the constants we get: E = 0.072 eV.