Chemical Bonding

A Chem1 Reference Text

Stephen K. Lower • Simon Fraser University
[email protected]

1 • Bonds and molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2 • Observable properties of chemical bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Bond energies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Bond lengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
Stretching frequency and infrared absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
3 • Why do chemical bonds form? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Classical models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
Quantum mechanical models of the chemical bond . . . . . . . . . . . . . . . . . . . . . . . . . .13
4 • The shared-electron model of chemical bonding . . . . . . . . . . . . . . . . . . . . . 16
The Octet Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16
Lewis electron-dot formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
Too many structures: resonance hybrids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20
5 • Polar and nonpolar bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Electronegativity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23
Dipole moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23
Formal charge and oxidation number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25
Ionic compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26
Electron donor-acceptor bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28
6 • The Shapes of Molecules: the VSEPR model . . . . . . . . . . . . . . . . . . . . . . . . 30
Electron pair repulsion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
Digonal and trigonal coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
Tetrahedral coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
Tetrahedral coordination with lone pairs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
Atoms bonded to five atoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .34
Octahedral coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
Summary of VSEPR theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36
7 • Hybrid orbitals: the valence bond model . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
What are hybrid orbitals? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38
Digonal bonding: sp-hybrid orbitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40
Trigonal (sp2) hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41
Tetrahedral (sp3) hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42
Conjugated double bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47
The nitrate ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48
Hybrids involving d orbitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50
Transition metal complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51
Limitations of the hybrid model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53
8 • The molecular orbital Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Molecular orbitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55
The hydrogen molecule ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55
Bonding and antibonding molecular orbitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55
Molecular orbital diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57
Diatomic molecules containing second-row atoms . . . . . . . . . . . . . . . . . . . . . . . . . .58
Sigma and pi orbitals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59
9 • Bonding in coordination complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
The basics of coordination complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Structures and bonding in transition metal complexes. . . . . . . . . . . . . . . . . . . . . . . . 65
Ligand field theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Coordination complexes in biochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
10 •Bonding in metals and semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Molecular orbitals in metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Why metals are different . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Band structure in metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Metals, insulators and semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Why you need to know this stuff

The uniqueness of a particular chemical substance must ultimately lie in

the number and geometrical arrangement of the atoms of which it is com-
posed. The lower the potential energy of a particular aggregation of atoms,
the more stable the chemical substance. Chemical bonding refers to the set
of principles and theories that govern the structure and stability of chem-
ical substances, and thus of the rearrangements that occur during chemi-
cal reactions in which one substance is transformed into another. The
study of chemical bonding is thus one of the fundamental pillars of modern
chemical science.

© 2002 by Stephen Lower ([email protected])

Dept of Chemistry - Simon Fraser University - Burnaby BC V5A 1S6 Canada
This document is part of the Chem1 Virtual Textbook ( http://www.chem1.com/acad/webtext/
virtualtextbook.html) and may be reproduced for non-commercial purposes only.
Last modified 24 July 2003
A Web version of this document is available at
http://www.chem1.com/acad/webtext/chembond/

Page 2 Chem1 Chemical Bonding

1. Bonds and molecules
What is a chemical bond?
A chemical bond is so often represented as a line drawn between
atom symbols or a stick connecting two balls in a plastic molecu-
lar model that we sometimes tend to think of chemical bonds as
“things”. It is more useful, however, to regard a chemical bond as
an effect that causes the energy of two atoms close together to be
markedly lower (by about 100 kJ per mole or more) than when
they are far apart.
The forces that hold bonded atoms together are basically just the
same kinds electrostatic attractions that bind the electrons of an
atom to its positively-charged nucleus; a chemical bond occurs when these same forces
are able to act on electrons subject to the simultaneous influence of two nuclei.

When two atoms combine to form a molecule, they do so because the

presence of two nearby positive centers (the nuclei) allows the electrons
to rearrange themselves into locations (orbitals) in which more electrons
are closer to more nuclei.

This is the most important fact about chemical bonding that you should know, but it
is not of itself a workable theory of the chemical bond because it does not describe the
conditions under which bonding occurs, nor does it make useful predictions about
the properties of the bond.

What is a molecule?
All atoms attract one another at small distances; the universal attractive interactions
known as van der Waals forces exist between all matter, and play an important part in
determining the properties of liquids and solids. These attractions are extremely weak,
however, and they lack specificity: they do not lead to aggregates having any special
structure or composition.
A molecule is an Chemical bonding connotes the existence of an aggregate of
atoms that is sufficiently stable to possess a characteristic
aggregate of
structure and composition. The important thing to understand
atoms that pos-
about the definition written at the left is that it is essentially an
sesses distinctive operational one; as our ability to observe the characteristic
and distinguish- properties of loosely-bound aggregates of atoms increases, our
ing properties. ideas of what constitutes a molecule will change. This was illus-
trated quite vividly in the early 1980’s, when metal clusters—
stable arrangements of 5-20 metallic atoms— were first characterized. These had not
previously been recognized as molecules because no one know how to observe them.
More recently, advances in technology that allow chemists to observe chemical species
that can only exist for tiny fractions of a second have greatly extended the range of
what we can call “molecules”.

Page 3 Bonds and molecules

Stability and reactivity
The fall in energy when atoms join together is a measure of the stability of the new
aggregate. In order to be regarded as a molecule, the aggregate must be sufficiently sta-
ble to resist disruption by thermal motions long enough to enable the observation of
whatever distinctive properties and composition it might have.
Some molecules are stable or observable only under certain conditions: many, such as
KrF2, are so weakly bound that they decompose at all but the lowest temperatures.
Others, such as gaseous LiF, can be observed only at temperatures around 1000 °C.
There are many molecules that are energetically stable, but are so reactive that their
lifetimes are too brief to make their observation possible. The molecule CH3, methyl, is
a good example: it can be formed by electrical discharge in gaseous CH4, but it is so
reactive that it reacts with almost any molecule it strikes within a few collisions. It was
not until the development of spectroscopic methods (in which a molecule is character -
ized by the wavelengths of light that it absorbs) that methyl was recognized as a stable
molecule that is an important intermediate in many chemical processes ranging from
flames to atmospheric chemistry.

Potential energy curves

In an earlier unit of this course, you were introduced to plots in which the energy pos-
sessed by a system of two atoms is shown as a function of the distance between them.
These Morse curves, as they are sometimes called, are quite useful in defining certain
properties of a chemical bond.

Fig. 1: Bond energy and the Morse curve.

The energy of a system of two atoms depends on the dis-

tance between them. At large distances the energy is zero,
meaning “no interaction”. At distances of several atomic
diameters attractive forces dominate, whereas at very close
approaches the force is repulsive, causing the energy to rise.
The attractive and repulsive effects are balanced at the mini-
mum point in the curve. If the energy minimum is at least of
the order of RT, the two atoms will be able to withstand the
disruptive influence of thermal energy., and a chemical bond
can be said to exist between them.

The internuclear distance at which the energy minimum occurs defines the bond
length. This is more correctly known as the equilibrium bond length, because thermal
motion causes the two atoms to vibrate about this distance. In general, the stronger
the bond, the smaller will be the bond length.

Page 4 Chem1 Chemical Bonding

Fig. 2: When does attraction result in
bonding?
Attractive forces operate between all atoms, but
unless the potential energy minimum is at least
of the order of RT, the two atoms will not be able
to withstand the disruptive influence of thermal
energy long enough to result in an identifiable
molecule. Thus we can say that a chemical bond
exists between the two atoms in H2. The weak
attraction between argon atoms does not allow
Ar2 to exist as a molecule, but it does corre-
spond to the van Der Waals force that holds
argon atoms together in the liquid and solid.

Potential energy and kinetic energy. Quantum theory tells us that an electron in
an atom possesses kinetic energy K as well as potential energy P, so the total energy E
is always the sum of the two: E = P + K. The relation between them is surprisingly
simple: K = –0.5 P. This means that when a chemical bond forms (an exothermic pro-
cess with ∆E < 0), the decrease in potential energy is accompanied by an increase in
the kinetic energy (embodied in the momentum of the bonding electrons), but the
magnitude of the latter change is only half as much, so the change in potential energy
always dominates. The bond energy –∆E has half the magnitude of the fall in potential
energy.

Page 5 Bonds and molecules

2. Observable properties of chemical bonds
Chemical bonds, of course, cannot be observed directly; the best we can do is to carry
out experiments on substances containing the appropriate pair of atoms, and then try
to make inferences about the nature of the bonding force between them.
It is important to bear in mind that the exact properties of a specific kind of bond will
be determined in part by the nature of the other bonds in the molecule; thus the energy
and length of the C–H bond will be somewhat dependent on what other atoms are con-
nected to the carbon atom. Similarly, the C-H bond length can vary by as much a 4
percent between different molecules. For this reason, the values listed in tables of bond
energy and bond length are usually averages taken over a variety of environments for a
specific atom pair.
In some cases, such as C—O and C—C, the variations can be much greater, approach-
ing 20 percent. In these cases, the values fall into groups which we interpret as repre-
sentative of single- and multiple bonds: double, and triple.

2.1 Bond energies

The bond energy is the amount of work that must be done to pull two atoms completely
apart; in other words, it is the same as the depth of the “well” in the potential energy
curve in Fig. 1. This is almost, but not quite the same as the bond dissociation energy
actually required to break the chemical bond; the difference is the very small zero-point
energy as explained inFig. 3.
Bond energies are usually determined indirectly from thermodynamic data, but there
are two main experimental ways of measuring them directly:
1. The direct thermochemical method involves separating the two atoms by an electrical
discharge or some other means, and then measuring the heat given off when they
recombine. Thus the energy of the C—C single bond can be estimated from the heat
of the recombination reaction between methyl radicals, yielding ethane:
CH3 + CH3 → H3CCH3
Although this method is simple in principle, it is not easy to carry out experimentally. The
highly reactive components must be prepared in high purity and in a stream of moving gas.
2. The spectroscopic method is based on the principle that absorption of light whose
wavelength corresponds to the bond energy will often lead to the breaking of the bond
and dissociation of the molecule. For some bonds, this light falls into the green and
blue regions of the spectrum, but for most bonds ultraviolet light is required. The
experiment is carried out by observing the absorption of light by the substance being
studied as the wavelength is decreased. When the wavelength is sufficiently small to
break the bond, a characteristic change in the absorption pattern is observed.
Spectroscopy is quite easily carried out and can yield highly precise results, but this method is
only applicable to a relatively small number of simple molecules. The major problem is that the
light must first be absorbed by the molecule, and relatively few molecules happen to absorb
light of a wavelength that corresponds energetically to a bond energy.

Experiments carried out on diatomic molecules such as O 2 and CS yield unambiguous

values of bond energy, but for more complex molecules there are complications. For
example, the heat given off in the CH3 combination reaction written above will also

Page 6 Chem1 Chemical Bonding

include a small component that represents the differences in the energies of the C-H
bonds in methyl and in ethane. These can be corrected for by experimental data on
reactions such as
CH3 + H → CH4
CH2 + H → CH3
By assembling a large amount of experimental information of this kind, a consistent
set of average bond energies can be obtained. The energies of double bonds are greater
than those of single bonds, and those of triple bonds are higher still.

H C N O F Cl Br I Si
43 41 39 46 56 43 37 29 39
H
6 5 0 4 9 2 0 5 5
34 29 35 43 33 27 24 36
C
5 0 0 9 0 5 0 0
16 20 27 20 24
N
0 0 0 0 5
14 18 20 22 20 37
O
0 5 5 0 0 0
16 25 23 28 54
F
0 5 5 0 0
24 22 21 35
Cl
3 0 0 9
19 18 29
Br
0 0 0
15 21
I
0 0
23
Si
0
Table 1: Average energies of some single bonds (kJ/mol)

Use of bond energies in estimating heats of reaction

One can often get a very good idea of how much heat will be absorbed or given of f in a
reaction by simply finding the difference in the total bond energies contained in the
reactants and products.
As an example, consider the reaction of chlorine with methane to produce dichlo-
romethane and hydrogen chloride:
CH4(g) + 2Cl2(g) → CHCl2(g) + 2HCl(g)
In this reaction, two C–H bonds and two Cl–Cl bonds are broken, and two new C–H and
H–Cl bonds are formed. The net change is
2(C–H) + 2(Cl–Cl) – 2(C–Cl) –2 (H–Cl) = (830 + 486 -660 - 864) kJ
which comes to –208 kJ per mole of methane; this agrees quite well with the observed
heat of reaction, which is –202 kJ/mol.

Page 7 Observable properties of chemical bonds

2.2 Bond lengths
The bond length is the internuclear distance: the distance between the centers of the
two bonded atoms. Bond distances are customarily expressed in Angstrom units (1Å =
10-8 cm = 100 pm) and are mostly in the range 1-2 Å. Even though the bond is vibrat-
ing, equilibrium bond lengths can be determined to within 0.01Å.
Bond lengths reflect the sizes of the atoms; thus those involving
154 pm
H3C CH3 hydrogen can be quite short. The shortest, H–H, is only 0.74Å.
Multiply-bonded atoms are closer together than singly-bonded
134 pm
ones; this is a major criterion for experimentally determining the
H2C CH2 multiplicity of a bond.
The most common method of measuring bond lengths in solids is
120 pm by analysis of the diffraction or scattering of X-rays when they
CH CH pass through the regularly-spaced atoms in the crystal. For gas-
eous molecules, neutron- or electron-diffraction can also be used.

2.3 Stretching frequency and infrared absorption

When an atom is displaced from its equilibrium position in a
molecule, it is subject to a restoring force which increases with
the displacement. A spring follows the same law (Hooke’s law);
a chemical bond is therefore formally similar to a spring that
has weights (atoms) attached to its two ends. A mechanical
system of this kind possesses a natural vibrational frequency
which depends on the masses of the weights and the stiffness of the spring.

Fig. 3: A close-up view of the

bottom of the potential energy
curve.
A chemical bond is something like a 0
vibrating spring, except that the vibra-
tional motions are quantized: only
vibrations having certain discrete
energies are possible. The Uncer- energy
tainty Principle does not allow us to 5 bond dissociation energy
simultaneously observe energy and 4
position, so the minimum potential vibrational
quantum 3
energy lies somewhat above the mini-
numbers 2
mum of the potential energy curve.
1
For this reason, the bonded atoms are ground vibrational zero-point energy
never at rest and always possess at 0
state of molecule
least a zero-point energy. For this rea-
equilibrium bond length
son, the bond dissociation energy is
slightly smaller than the energy corre- 1 2 3 4
sponding to the bottom of the curve. internuclear distance, pm

On the atomic scale in which all motions are quantized, a vibrating system can pos-
sess only certain allowed vibrational frequencies, or states. These are depicted by the
horizontal lines in the potential energy curve shown here. Notice that the very bottom
of the curve does not correspond to an allowed state, as this allows for no change in
the position of the atoms at all, and would therefore violate the uncertainty principle.

Page 8 Chem1 Chemical Bonding

The lowest-allowed, or ground vibrational state is the one denoted by 0, and it is nor-
mally the only state that is significantly populated in most molecules at room temper-
ature. In order to jump to a higher state, the molecule must absorb a photon whose
energy is equal to the distance between the two states. For chemical bonds, these
energies correspond to light that falls in the infrared region of the spectrum.
For ordinary chemical bonds, these natural frequencies correspond to those of infrared
light. Each wavelength of infrared light that excites the vibrational motion of a particu-
lar bond will be absorbed by the molecule. In general, the stronger the bond and the
lighter the atoms it connects, the higher will be its natural stretching frequency and
the shorter the wavelength of light absorbed by it. Thus the C–H bond absorbs at a
shorter wavelength than does the C–C bond, and C–C bonds are easily distinguished
from C=C double bonds. Studies on a wide variety of molecules have made it possible
to determine the wavelengths absorbed by each kind of bond. By plotting the degree of
absorption as a function of wavelength, one obtains the infrared spectrum of the mole-
cule which allows one to “see” what kinds of bonds are present.
Actual infrared spectra are complicated by the presence of more complex motions
(stretches involving more than two atoms, wagging, etc.), and absorption to higher
quantum states (overtones), so infrared spectra can become quite complex. This is not
necessarily a disadvantage, however, because such spectra can serve as a “fingerprint”
that is unique to a particular molecule and can be helpful in identifying it. Largely for
this reason, infrared spectrometers are standard equipment in most chemistry labora-
tories.

Infrared absorption and global warming

The aspect of bond stretching and bending frequencies that impacts our lives most
directly is the way that some of the gases of the atmosphere absorb infrared light and
thus affect the heat balance of the Earth. Owing to their symmetrical shapes, the prin-
cipal atmospheric components N2 and O2 do not absorb infrared light, but the minor
components water vapor and carbon dioxide are strong absorbers, especially in the
long-wavelength region of the infrared. Absorption of infrared light by a gas causes its
temperature to rise, so any source of infrared light will tend to warm the atmosphere;
this phenomenon is known as the greenhouse effect.
The incoming radiation from the Sun (which contains relatively little long-wave infrared
light) passes freely through the atmosphere and is absorbed by the Earth's surface,
warming it up and causing it to re-emit some of this energy as long-wavelength infra-
red. Most of the latter is absorbed by the H2O and CO2 in the atmosphere, effectively
trapping the radiation as heat. Thus the atmosphere is heated by the Earth, rather
than by direct sunlight. Without the “greenhouse gases” in the atmosphere, the Earth's
heat would be radiated away into space, and our planet would be too cold for life.
Since the beginning of the Industrial Revolution in the 19th century, huge quantities of
additional greenhouse gases have been accumulating in the atmosphere. Carbon diox-
ide from fossil fuel combustion has been the principal source, but intensive agriculture
also contributes significant quantities of methane (CH4) and nitrous oxide (N2O) which
are also efficient far-infrared absorbers. The measurable increase in these gases is
believed by many to be responsible for the increase in the average temperature of the
Earth that has been noted over the past 50 years— a trend that could initiate wide-
spread flooding and other disasters if it continues.

Page 9 Observable properties of chemical bonds

3. Why do chemical bonds form?
The answer to this question would ideally be a simple, easily understood theory that
would not only explain why atoms bind together to form molecules, but would also pre-
dict the three-dimensional structures of the resulting compounds as well as the ener -
gies and other properties of the bonds themselves. Unfortunately, no one theory exists
that accomplishes these goals in a satisfactory way for all of the many categories of
compounds that are known. Moreover, it seems likely that if such a theory does ever
come into being, it will be far from simple.
When we are faced with a scientific problem of this complexity, experience has shown
that it is often more useful to concentrate instead on developing models. A scientific
model is something like a theory in that it should be able to explain observed phenom-
ena and to make useful predictions. But whereas a theory can be discredited by a sin-
gle contradictory case, a model can be useful even if it does not encompass all
instances of the phenomena it attempts to explain. We do not even require that a model
be a credible representation of reality; all we ask is that it be able to explain the behav-
ior of those cases to which it is applicable in terms that are consistent with the model
itself. An example of a model that you may already know about is the kinetic molecular
theory of gases. Despite its name, this is really a model (at least at the level that begin-
ning students use it) because it does not even try to explain the observed behavior of
real gases. Nevertheless, it serves as a tool for developing our understanding of gases,
and as a starting point for more elaborate treatments.
Given the extraordinary variety of ways in which atoms combine into aggregates, it
should come as no surprise that a number of useful bonding models have been devel-
oped. Most of them apply only to certain classes of compounds, or attempt to explain
only a restricted range of phenomena. In this section we will provide brief descriptions
of some of the bonding models; the more important of these will be treated in much
more detail in later parts of this chapter.

Fig. 4: Some early views of chemical

bonding
Intense speculation about “chemical affinity”
began in the 18th century. Some likened the
tendency of one atom to “close” with another
as an expression of a human-like kind of
affection. Others attributed bonding to mag-
netic-like forces (left) or to varying numbers of
“hooks” on different kinds of atoms (right). The
latter constituted a primitive (and extremely
limited) model of combining power or valence.

3.1 Classical models

By classical, we mean models that do not take into account the quantum behavior of
small particles, notably the electron. These models generally assume that electrons and
ions behave as point charges which attract and repel according to the laws of electro-

Page 10 Chem1 Chemical Bonding

statics. Although this completely ignores what has been learned about the nature of
the electron since the development of quantum theory in the 1920’s, these classical
models have not only proven extremely useful, but the major ones also serve as the
basis for the chemist’s general classification of compounds into “covalent” and “ionic”
categories.

The ionic model

Ever since the discovery early in the 19th century that solutions of salts and other elec-
trolytes conduct electric current, there has been general agreement that the forces that
hold atoms together must be electrical in nature. Electrolytic solutions contain ions
having opposite electrical charges, opposite charges attract, so perhaps the substances
from which these ions come consist of positive and negatively charged atoms held
together by electrostatic attraction.
It turns out that this is not true generally, but a model built on this assumption does a
fairly good job of explaining a rather small but important class of compounds that are
called ionic solids. The most well known example of such a compound is sodium chlo-
ride, which consists of two interpenetrating lattices of Na+ and Cl– ions arranged in
such as way that every ion of one type is surrounded (in three dimensional space) by
six ions of opposite charge.
One can envision the formation of a solid NaCl unit by a sequence of events in which
electrons are removed from one mole of Na atoms and given to Cl atoms, followed by
condensation of the resulting ions into a crystal lattice:
Na(g) → Na+(g) + e– +494 kJ(ionization energy)
Cl(g) + e–→ Cl–(g) –368 kJ(electron affinity)
Na+(g) + Cl–(g) → NaCl(s) –498 kJ(lattice energy)
Since the first two energies are known experimentally, as is the energy of the sum of
the three processes, the lattice energy can be found by difference. It can also be calcu-
lated by averaging the electrostatic forces exerted on each ion over the various direc-
tions in the solid, and this calculation is generally in good agreement with observation,
thus lending credence to the model. The sum of the three energy terms is clearly nega-
tive, and corresponds to the liberation of heat in the net reaction
Na(g) + Cl(g) → NaCl(s), which defines the Na–Cl “bond” energy. The ionic solid is more
stable than the equivalent number of gaseous atoms simply because the three-dimen-
sional NaCl structure allows more electrons to be closer to more nuclei. This is the cri-
terion for the stability of any kind of molecule; all that is special about the “ionic” bond
is that we can employ a conceptually simple electrostatic model to predict the bond
strength.
The main limitation of this model is that it applies really well only to the small class of
solids composed of Group 1 and 2 elements with highly electronegative elements such
as the halogens. Although compounds such as CuCl2 dissociate into ions when they
dissolve in water, the fundamental units making up the solid are more like polymeric
chains of covalently-bound CuCl2 molecules that have little ionic character.

Shared-electron (covalent) model

This model originated with the theory developed by G.N. Lewis in 1916, and it remains
the most widely-used model of chemical bonding. The essential elements of this model
can best be understood by examining the simplest possible molecule. This is the hydro-
Page 11 Why do chemical bonds form?
gen molecule ion H2+, which consists of two nuclei and one electron.
First, however, think what would happen if we tried to make the even simpler molecule
H2+. Since this would consist only of two protons whose electrostatic charges would
repel each other at all distances, it is clear that such a molecule cannot exist; some-
thing more than two nuclei are required for bonding to occur.
In the hydrogen molecule ion H2+ we have a third
particle, an electron. The effect of this electron
will depend on its location with respect to the
two nuclei. If the electron is in the space
between the two nuclei, it will attract both pro-
tons toward itself, and thus toward each other.
If the total attraction energy exceeds the inter-
nuclear repulsion, there will be a net bonding
Fig. 5: Locations effect and the molecule will be stable. If, on the
at which an electron will bind two nuclei other hand, the electron is off to one side, it will
attract both nuclei, but it will attract the closer
one much more strongly, owing to the inverse-square nature of Coulomb’s law. As a
consequence, the electron will now help the electrostatic repulsion to push the two
nuclei apart.
We see, then, that the electron is an essential component of a chemical bond, but that
it must be in the right place: between the two nuclei. Coulomb’s law can be used to cal-
culate the forces experienced by the two nuclei for various positions of the electron.
This allows us to define two regions of space about the nuclei, as shown in the figure.
One region, the binding region, depicts locations at which the electron exerts a net
binding effect on the new nuclei. Outside of this, in the antibinding region, the electron
will actually work against binding.
This simple picture illustrates the number one rule of chemical bonding: chemical
bonds form when electrons can be simultaneously close to two or more nuclei. It
should be pointed out that this principle applies also to the ionic model; as will be
explained later in this chapter, the electron that is “lost” by a positive ion ends up being
closer to more nuclei (including the one from whose electron cloud it came) in the com-
pound.

The polar covalent model

A purely covalent bond can only be guaranteed when the electronegativities (electron-
attracting powers) of the two atoms are identical. When atoms having different elec-
tronegativities are joined, the electrons shared between them will be displaced toward
the more electronegative atom, conferring a polarity on the bond which can be
described in terms of percent ionic character. The polar covalent model is thus an gen-
eralization of covalent bonding to include a very wide range of behavior; it is discussed
in greater detail beginning on Page 23.

The Coulombic model

This is an extension of the ionic model to compounds that are ordinarily considered to
be non-ionic. Combined hydrogen is always considered to exist as the hydride ion H–,
so that methane can be treated as if it were C4+ H–4. This is not as bizarre as it might
seem at first if you recall that the proton has almost no significant size, so that it is
essentially embedded in an electron pair when it is joined to another atom in a covalent
Page 12 Chem1 Chemical Bonding
bond. This model, which is not as well known as it deserves to be 1, has considerable
predictive power, both as to bond energies and structures.

VSEPR model
The valence shell electron repulsion model is not so much a model of chemical bonding
as a scheme for explaining the shapes of molecules. It is based on the quantum
mechanical view that bonds represent electron clouds— physical regions of negative
electric charge that repel each other and thus try to stay as far apart as possible. We
will explore this concept in much greater detail beginning on Page 30.

3.2 Quantum mechanical models of the chemical bond

These models of bonding take into account the fact that a particle as light as the elec-
tron cannot really be said to be in any single location. The best we can do is define a
region of space in which the probability of finding the electron has some arbitrary value
which will always be less than unity. The shape of this volume of space is called an
orbital and is defined by a mathematical function that relates the probability to the
(x,y,z) coordinates of the molecule.
Like other models of bonding, the quantum models attempt to show how more elec-
trons can be simultaneously close to more nuclei. Instead of doing so through purely
geometrical arguments, they attempt this by predicting the nature of the orbitals which
the valence electrons occupy in joined atoms.

The hybrid orbital model

This was developed by Linus Pauling in 1931 and was the first quantum-based model
of bonding. It is based on the premise that if the atomic s, p, and d orbitals occupied by
the valence electrons of adjacent atoms are combined in a suitable way, the hybrid
orbitals that result will have the character and directional properties that are consis-
tent with the bonding pattern in the molecule. The rules for bringing about these com-
binations turn out to be remarkably simple, so once they were worked out it became
possible to use this model to predict the bonding behavior in a wide variety of mole-
cules. The hybrid orbital model is most usefully applied to the p-block elements the
first two rows of the periodic table, and is especially important in organic chemistry;
see Page 37.

The molecular orbital model

This model takes a more fundamental approach by regarding a molecule as a collection
of valence electrons and positive cores. Just as the nature of atomic orbitals derives
from the spherical symmetry of the atom, so will the properties of these new molecular
orbitals be controlled by the interaction of the valence electrons with the multiple posi-
tive centers of these atomic cores. These new orbitals, unlike those of the hybrid model,
are delocalized; that is, they do not “belong” to any one atom but extend over the entire
region of space that encompasses the bonded atoms. The available (valence) electrons
then fill these orbitals from the lowest to the highest, very much as in the Aufbau prin-
ciple that you learned for working out atomic electron configurations. For small mole-

1. For more information on the coulombic model, see the articles by Lawrence J. Sacks
in J. Chemical Education, 1986: 288-297, and 373-378.

Page 13 Why do chemical bonds form?

cules (which are the only ones we will consider here), there are simple rules that govern
the way that atomic orbitals transform themselves into molecular orbitals as the sepa-
rate atoms are brought together. The real power of molecular orbital theory, however,
comes from its mathematical formulation which lends itself to detailed predictions of
bond energies and other properties. This model is developed (and extended to metals)
beginning on Page 54.

The electron-tunneling model

A common theme uniting all of the models we have discussed is that bonding depends
on the fall in potential energy that occurs when opposite charges are brought together.
In the case of covalent bonds, the shared electron pair acts as a kind of “electron glue”
between the joined nuclei. In 1962, however, it was shown that this assumption is not
strictly correct, and that instead of being concentrated in the space between the nuclei,
the electron orbitals become even more concentrated around the bonded nuclei. At the
same time however, they are free to “move” between the two nuclei by a process known
as tunneling. This refers to a well-known quantum mechanical effect that allows elec-
trons (or other particles small enough to exhibit wavelike properties) to pass (“tunnel”)
through a barrier separating two closely adjacent regions of low potential energy. One
result of this is that the effective volume of space available to the electron is increased,
and according to the uncertainty principle this will reduce the kinetic energy of the
electron.
According to this model, the bonding electrons act as a kind of fluid that concentrates
in the region of each nucleus (lowering the potential energy) and at the same time is
able to freely flow between them (reducing the kinetic energy). A summary of the con-
cept, illustrating its application to a simple molecule, is shown on the next page.
Despite its conceptual simplicity and full acknowledgment of the laws of quantum
mechanics, this model is less known than it deserves to be and is unfortunately absent
from most textbooks.
A more detailed summary of this model and examples of its application is shown on the
next page.

Page 14 Chem1 Chemical Bonding

Fig. 6: Electron-tunneling model of chemical bonding
These diagrams are adapted from those in the article by Sture Nordholm: Delocalization– the key concept of
covalent bonding. 1986: J. Chem. Education 581-584.

Page 15 Why do chemical bonds form?

4. The shared-electron model of chemical bonding
Prior to the discovery of nuclei and electrons around 1900, theories of chemical bond-
ing were based mainly on the rather vague concept of “chemical affinities” between dif-
ferent kinds of atoms. Beyond this, the primary need was to explain the dif ferent
combining capacities, or “valences”, of the different elements. These valences tended to
vary with the periodic group of the element, and were therefore thought to encompass a
range of from zero to eight; once the electron shell arrangement of atoms was worked
out, the connection between outer -shell electron occupancy and bonding became
apparent.

4.1 The Octet Rule

The first successful theory of chemical bonding
was formulated by G.N.Lewis in 1916.
G.N. Lewis (1875-1946; right) created the Chemis-
try Department at the University of California, Ber-
keley, and made it into one of the world’s best. His
other notable work included the first isolation of
heavy water (D2O), the thermodynamics of solu-
tions, and the phosphorescence and magnetic
properties of molecules.
Although Lewis originated the idea of the electron-
pair bond, much of the credit for its early accep-
tance must go to Irving Langmuir, who extended it
somewhat and enthusiastically popularized it to the
extent that it began to be known as the Lewis-Lang-
muir theory, and even as the “Langmuir theory”
(much to Lewis’ annoyance!)
Irving Langmuir (1881-1967, Nobel Prize 1932)
was an industrial scientist employed by the General Electric Co. His most notable
work was on the chemistry of surfaces and monomolecular layers. Lewis and Lang-
muir were probably the two greatest American chemists of the first half of the twenti-
eth century.
At the time Lewis began developing his ideas in 1902, it was widely believed that
chemical bonding involved electrostatic attraction between ion-like entities. This
seemed satisfactory for compounds such as NaCl that were known to dissociate into
ions when dissolved in water, but it failed to explain the bonding in non-electrolytes
such as CH4. Atomic orbitals had not yet been thought of, but the concept of
“valence” electrons was known, and the location of the noble gases in the periodic
table suggested that all except helium possess eight valence electrons. It was also
realized that elements known to form simple ions such as Ca2+ or Cl– do so by losing
or gaining whatever number of electrons is needed to leave eight in the valence shell
of each. Lewis sought a way of achieving this octet in a way that did not involve ion
formation, and he found it in his shared electron-pair theory published in 1916.
Present-day shared electron-pair theory is based on the premise that the s2p6 octet in
the outermost shells of the noble gas elements above helium represents a particularly
favorable configuration. This is not because of any mysterious properties of octets (or of
noble gas elements); by allowing each nucleus to claim half-ownership of a shared elec-
tron, more electrons are effectively “seeing” more nuclei, leading to increased electro-

Page 16 Chem1 Chemical Bonding

static attractions and a lowering of the potential energy.
.
Fig. 7: shared electron-
pair bond
Sharing of one or more
valence electrons between
two atoms allows each to
possess a noble gas (s2p6)
valence shell configuration.
Each atom can claim half-
ownership of each shared
electron.

Lewis’ idea that the electrons are shared in pairs stemmed from his observation that
most molecules possess an even number of electrons. This paired sharing also allows
the formulas of a large number of compounds to be rationalized and predicted— a fact
that led to the widespread acceptance of the Lewis model by the early 1920s.
For the lightest atoms the octet rule must be modified, since the noble-gas configura-
tion will be that of helium, which is simply s2 rather than s2p6. Thus we write LiH as
Li:H, where the electrons represented by the two dots come from the s orbitals of the
separate atoms.
The octet rule applies quite well to the first full row of the periodic table (Li through F),
but beyond this it is generally applicable only to the non-transition elements, and even
in many of these it cannot explain many of the bonding patterns that are observed. The
principal difficulty is that a central atom that is bonded to more than four peripheral
atoms must have more than eight electrons around it if each bond is assumed to con-
sist of an electron pair. In these cases, we hedge the rule a bit, and euphemistically
refer to the larger number of electrons as an “expanded octet”.
In spite of the octet rule’s many exceptions and limitations, the shared electron-pair
model is the one that chemists most frequently employ in their day-to-day thinking
about molecules. It continues to serve as a very useful guiding principle and can be a
good starting point for more sophisticated theories.

4.2 Lewis electron-dot formulas

The shared electron-pair is such a fundamental concept of chemical bonding that it is
important to have a simple way of writing out a formula that shows the disposition of
the shared pairs between the different atoms. This is commonly accomplished by
depicting the valence (outermost) electrons of an atom as dots that are written around
the atom symbols. Sometimes it is convenient to represent the electrons that are con-
tributed by different atoms by different symbols. For example, the formation of H2 can
be depicted as
H• + H → H H
The Lewis theory makes no prediction about molecular shapes, so it is permissible to
arrange the dot pairs and other atoms around the central atom in an arbitrary way.
Usually, a more or less symmetrical arrangement is written.

Page 17 The shared-electron model of chemical bonding

It is sometimes preferable to represent an elec-
tron pair by a line; this is most commonly done
with shared electron pairs, which correspond to
chemical bonds which have traditionally been
represented in this way.

The Lewis theory makes no prediction about molecular shapes, so it is permissible to

arrange the dot pairs and other atoms around the central atom in an arbitrary way.
Usually, a more or less symmetrical arrangement is written.

Bonding and nonbonding electrons: lone pairs

In many molecules, not all of the electron pairs comprising the octet are
shared between atoms. The unshared electron pairs are often called lone H N H
pairs. Although lone pairs are not directly involved in bond formation,
they should always be shown in Lewis formulas; we will see later that H
they have an important role to play in determining the shape of the mol-
O H
ecule.
H
Lewis formulas of ions
The Lewis formulas of charged molecules are no different from those
– of neutral molecules. It is customary to enclose the formulas in
O H brackets so that the charge can be depicted as belonging to the mole-
cule as a whole, rather than being localized on a particular atom.
2–
S Notice that the Lewis formulas of monatomic cations such as H+ and
Ca2+ contain no dots at all, since these ions have no valence elec-
trons.

Multiple bonds
If one pair of electrons shared between two atoms constitutes a
chemical bond, it seems logical that two or three pairs could be C O
shared to produce double and triple bonds. Such formulations
appear quite naturally when the octet rule is applied to elements H –C C–H
such as C, O, S, and N.
Since multiple bonds place more electron density between the two N N
nuclei, the latter are held toward each other more closely and tightly;
multiple bonds are therefore shorter and stronger than single bonds. –
C N
Expanded octets
As mentioned previously, the octet rule works best for the elements in the first period
(Li through F) of the periodic table. The reason for this is that electrons, whether
shared or not, must be contained in orbitals, and the energies of electrons in such
orbitals must be relatively low; otherwise, there would be no energetic advantage in
forming a bond in the first place-- the atoms would be better off by themselves.
For the first-period elements, the only available orbitals are the s2p6 set that, when
filled, comprise the octet. In order to place more than four other atoms around any one
of these elements, the additional shared electrons would have to go into the much-

Page 18 Chem1 Chemical Bonding

higher-energy 3s23p6 set.
With elements in the fourth period (K through Cl) and beyond,
Cl Cl these higher empty orbitals can sometimes be used to accommo-
P date additional shared pairs beyond the octet. If you review a dia-
Cl Cl gram showing the relative energies of the different kinds of atomic
Cl orbitals, you will notice that all the energy gaps become smaller as
F F the principal quantum number increases. Also, the gap between
the higher s- and p- orbitals is bridged by the d orbitals that begin
F S F at n=3.
F F
For these reasons, it is common for elements such as S, Cl, P, and
Si to form compounds in which five or six electron pairs surround
the central atom. These pairs may all be shared with other atoms, or some of them may
remain as lone pairs.
Examples of molecules in which the central atom contains an expanded octet are the
phosphorus pentahalides and sulfur hexafluoride.

Electron structures without molecules

Although there are many violations of the octet rule, most electron dot structures that
one can write down in accordance with this rule and its general scope of validity corre-
spond to molecules that actually exist. Sometimes, however, we are surprised to find
that the molecules corresponding to an apparently reasonable Lewis formula are not
known.
In some cases, this has been shown to be a consequence of the very high chemical
reactivity of the molecules. Thus hypofluorous acid, HOF, has never been isolated. It
was not until 1967 that its short-lived presence was detected spectroscopically. It is
now believed that the molecule is stable, but that the products obtained when it reacts
with itself are so much more stable that it decomposes almost as fast as it is formed:
2HOF(g) → 2HF(g) + O2(g)
Other molecules having proper Lewis structures but no apparent existence may be sta-
ble only at very low temperatures; examples are O4 and H2O4.
The fluorate ion, FO3–, has also never been detected, even though analogs containing
the other halogen elements are well known. The problem here may well lie with the very
small fluorine atom, which would allow the oxygens to approach so closely that they
would repel each other.
Small size is also suggested as the reason for the non-existence of the nitrogen analogs
of the sulfate and sulfite ions. These would have the formulas NO43– and NO33–. Here,
the problem is believed to be the high charge density: it costs a lot of energy to squeeze
this much electric charge in such a small volume. Sulfur, having a larger radius, forms
larger ions having lower charge densities, and the total charge would also be only –2
instead of –3.

Molecules without electron structures

There are also examples of molecules whose existence is beyond question, but for
which no satisfactory Lewis structures can be written. Two examples are the triiodide
ion I3–, and the bifluoride ion HF2–. The triiodide ion is a well known species found in
aqueous solutions containing iodine and iodide ions:

Page 19 The shared-electron model of chemical bonding

The bifluoride ion is formed in a rather similar way in hydrofluoric acid solutions con-
taining fluoride ion:
Try writing electron-dot structures for these two species, and you will see the problem!

Paramagnetic molecules
A somewhat different situation is exemplified by the common oxygen
molecule. It is easy to write a proper Lewis structure for O2 that places O O
an octet around each oxygen atom and a double bond between them.
However, it takes only a simple experiment to show that the electrons in
dioxygen cannot all be arranged in pairs: if you place a magnet near some liquid oxy-
gen, the liquid will be drawn to the magnet. This can only mean one thing: there are at
least two unpaired electrons in the O2 molecule. A more exact experiment shows that
this number is exactly two. Are they in the bond or are they non-bonding electrons?
You can decide this by sketching out a few possible structures.
The paramagnetism of oxygen is an anomaly in terms of the Lewis theory, although it is
predicted by a more comprehensive theory that we will look at later. There are, how-
ever, a few other molecules that we would expect to be paramagnetic simply because
they contain an odd number of valence electrons. The most well known example is
nitric oxide, NO. Since oxygen has four and nitrogen has five outer electrons, the total
number of valence electrons is nine, and magnetic measurements show that one of
these is unpaired.

4.3 Too many structures: resonance hybrids

There is a rather large class of molecules for which one has no difficulty writing Lewis
structures; in fact, we can write more than one valid structure for a given molecule.
This raises a new problem: how do we decide which one to use?
The answer is simple: if the structures are all equivalent, we use them all. As an exam-
ple, consider the nitrate ion NO3–, for which three Lewis structures can be written:
O O – O O – O O –
N N N
↔ ↔
O O O

These structures differ only in which oxygen atom is attached by the double bond.
Since there is no reason to prefer one over another, the NO3– ion is regarded as a
superposition, or hybrid of these three structures.
The term resonance has been used to describe this phenomenon, which is indicated in
the above structures by the double-ended arrows. The choice of this word was unfortu-
nate, because it connotes the existence of some kind of dynamic effect that has led to
the mistaken idea that the structure is continually alternating between the three possi-
bilities. The correct interpretation is simply that none of the hybrid structures com-
pletely represents the molecule; the “real” structure is a superposition of the individual
contributing structures.

Page 20 Chem1 Chemical Bonding

Thus there is no true double bond in NO3–; what we have
instead are three “4/3 bonds”. A more realistic way of repre-
O O –
senting the structure might be to show the fractional bonds
N
as dotted lines:
This interpretation is supported by structural studies on
O crystalline nitrate compounds; all the N-O bond distances are
bond order 4/3 the same, and they are intermediate in length between the
values expected for single and double bonds.
Since electrons in molecules tend to arrange themselves into configurations that give
the lowest possible energy, it is not surprising that the resonance hybrid represents a
more stable (i.e., strongly bound) molecule than does any one of the contributing struc-
tures.
There is a good quantum-mechanical reason for this; according to the Heisenberg
uncertainty principle, the energy of the electron will be more uncertain as its position is
more exactly specified. Since energies cannot be negative, the more “uncertain” the
energy can be, the higher it can be. If an electron is spread out over three bonds as in
the NO3– hybrid instead of being confined to the space between only two atoms, its
exact location is much less exactly known, and so its energy will be less uncertain, and
therefore lower.
This is all that is meant by the statement you will often see,
particularly in the older literature, that a given structure “is O –
stabilized by resonance”. This jargon is used, for example, to
explain the acidity of the -COOH group found in organic acids. H C
The idea is that resonance is only possible when the proton O
has been lost, and that the lower energy of the “resonating”
structure provides the driving force for the loss of the proton,
and thus is the source of the acidity carboxylic group.

The Benzene ring

Perhaps the most well known molecule whose structure must be represented as a reso-
nance hybrid is benzene, C6H6. The structure of this molecule had long been some-
thing of a mystery, in that it was difficult to understand how this formula could be
consistent with the well-established tetravalence of carbon in organic compounds. The
breakthrough came when the German chemist August Kekulé proposed that the mole-
cule is based on a hexagonal ring of carbon atoms as shown at the left below.
However, this structure is not consistent with the chemistry of benzene, which does not
undergo the reactions that would be expected of a molecule containing carbon-carbon
double bonds. Such compounds are normally quite reactive, while benzene, in con-
trast, tends to be rather inert to all but the most powerful reagents.

Page 21 The shared-electron model of chemical bonding

This apparent discrepancy disappeared after the resonance hybrid theory was devel-
oped; benzene is regarded as a hybrid of the two structures shown above. The struc-
ture of benzene is often represented by either of the figures shown at the left, in which
the dashed or solid circle represents a “half bond”, so that the bond order of each C–C
bond is 1.5. Bond length measurements are entirely consistent with this interpretation;
they are almost exactly halfway between the values found in compounds known to con-
tain single and double bonds.
In writing out resonance structures, it is important to bear in mind that only the elec-
tron pairs can be moved around; the atoms must all be kept in the same place. In some
cases it is necessary to move electrons to locations that would produce a positive
charge on one atom and a negative charge on the other. Since the separation of electric
charge always costs energy, such resonance forms will tend to be less stabilizing and
will not be as important contributors to the overall structure as those in where there is
no charge separation.

Page 22 Chem1 Chemical Bonding

5. Polar and nonpolar bonds

5.1 Electronegativity
The electrons constituting a chemical bond are simultaneously attracted by the electro-
static fields of the nuclei of the two bonded atoms. In a homonuclear molecule such as
O2 the bonding electrons will be shared equally by the two atoms. In general, however,
differences in the sizes and nuclear charges of the atoms will cause one of them to exert
a greater attraction on the bonding pair, causing the electron cloud to be displaced
toward the more strongly-attracting atom.
The electronegativity of an atom denotes its relative electron-attracting power in a chem-
ical bond.
It is important to understand that electronegativity is not a measurable property of an
atom in the sense that ionization energies and electron affinities are, although it can
be correlated with both of these properties. The actual electron-attracting power of an
atom depends in part on its chemical environment (that is, on what other atoms are
bonded to it), so tabulated electronegativities should be regarded as high-precision
predictors of the behavior of electrons in more complicated molecules.
There are several ways of computing electronegativities, which are expressed on an
arbitrary scale. The concept of electronegativity was introduced by Linus Pauling and
his 0-4 scale continues to be the one most widely used.
.
Fig. 8: Electronegativities

The 0-4 electronegativity scale of Pauling is the best

known of several arbitrary scales of this kind. Electroneg-
ativity values are not directly observable, but are derived
from measurable atomic properties properties such as
ionization energy and electron affinity. The place of any
atom on this scale provides a good indication of its ability
to compete with another atom in attracting a shared elec-
tron pair to it, but the presence of bonds to other atoms,
and of multiple- or nonbonding electron pairs may make
predictions about the nature a given bond less reliable.

An atom that has a small electronegativity is said to be

electropositive. As the diagram shows, the metallic ele-
ments are generally electropositive. The position of
hydrogen in this regard is worth noting; although physi-
cally a nonmetal, much of its chemisry is metal-like.

5.2 Dipole moments

When non-identical atoms are joined in a covalent bond, the electron pair will be
attracted more strongly to the atom that has the higher electronegativity. As a conse-
quence, the electrons will not be shared equally; the center of the negative charges in
the molecule will be displaced from the center of positive charge. Such bonds are said
to be polar and to possess partial ionic character, and they may confer a polar nature on
the molecule as a whole.

Page 23 Polar and nonpolar bonds

A polar molecule acts as an electric dipole which can interact with electric fields that
are created artificially or that arise from nearby ions or polar molecules. Dipoles are
conventionally represented as arrows pointing in the direction of the negative end. The
magnitude of interaction with the electric field is given by the permanent electric dipole
moment of the molecule. The dipole moment corresponding to an individual bond (or to
a diatomic molecule) is given by the product of the quantity of charge displaced q ≡ δ±
and the bond length r:
µ=q×r
In SI units, q is expressed in coulombs and r in meters, so m has the dimensions of C-
m. If one entire electron charge is displaced by 100 pm (a typical bond length), then
m = (1.6022 × 10–19 C) × (10–10 m) = 1.6 × 10–29 C-m = 4.8 D
The quantity at the right, the Debye unit, is still commonly used to express dipole
moments. It was named after PETER DEBYE (1884-1966), the Dutch physicist who pio-
neered the study of dipole moments and of electrical interactions between particles,
and won the Nobel Prize for Chemistry in 1934.
Measurement of dipole moments. When a solution of polar molecules is placed between
two oppositely-charged plates, they will tend to align themselves along the direction of
the field. This process consumes energy which is returned to the electrical circuit when
the field is switched off, an effect known as electrical capacitance. Measurement of the
capacitance of a gas or solution is easy to carry out and serves as a means of determin-
ing the magnitude of the dipole moment of a substance.

Problem Example: Estimate the percent

ionic character of the bond in hydrogen
fluoride from the experimental data
shown at the right.
Solution:

Dipole moments of more complicated molecules.

In molecules containing more than one polar bond, the molecular dipole moment is
just the vector combination of the individual bond dipoles. In some cases this can
result in a molecule containing polar bonds to be nonpolar, as in the example of carbon
dioxide shown in Fig. 9. In molecules containing nonbonding electrons or multiple
bonds, the elecronegativity difference may not correctly predict the bond polarity. A
good example of this is carbon monoxide, in which the partial negative charge resides
on the carbon (Fig. 11), as predicted by its negative formal charge (see next page.)

Page 24 Chem1 Chemical Bonding

Fig. 9: Dipole moments of H2O and CO2

5.3 Formal charge and oxidation number

Although the total number of valence electrons in a molecule is easily calculated, there
is sometimes no simple and unambiguous way of determining how many reside in a
particular bond or as non-bonding pair on a particular atom. For example, one can
write valid Lewis octet structures for carbon monoxide showing either a double or triple
bond between the two atoms, depending on how many nonbonding pairs are placed on
each:
::C::O: and C:::O:: . The choice between structures such as these is usually easy to
make on the principle that the more electronegative atom tends to surround itself with
the greater number of electrons. In cases where the distinction between competing
stsructures is not all that clear, an arbitrarily-calculated quantity known as the formal
charge can often serve as a guide. The formal charge on an atom is the electric charge it
would have if all bonding electrons were shared equally with its bonded neighbors.
The formal charge on an atom is calculated by the following algorithm:

FC = core charge – number of unshared electrons – number of bonding pairs

in which the core charge is the electric charge the atom would have if all its valence
electrons were removed.
In simple cases, the formal charge can be worked out visually directly from the Lewis
structure, as is illustrated farther on.

Problem Example: Lewis structures for carbon monoxide can be written in which the
C–O bond is either double or triple. Calculate the formal charges on carbon and oxy-
gen in the two alternative structures.
Solution: For ::C::O: :
C: 4 – 4 – 2 = –2 O: 6 – 2 – 2 = +2
For C:::O:::
C: 4 – 0 – 3 = +1 6 – 4 – 3 = +1

The general rule for choosing between alternative structures is that the one involving

Page 25 Polar and nonpolar bonds

the smallest formal charges is most favored, so we would assign carbon monoxide the
triple-bonded structure. In a species such as the thiocyanate ion (below) in which two
structures having the same minimal formal charges can be written, we would expect
the one in which the negative charge is on the more electronegative atom to predomi-
nate.

Fig. 10: Alternative Lewis structures for the thiocyanate ion SCN–
The electrons in the structures of the top row are the valence electrons for each atom; an additional
electron (purple) completes the nitrogen octet in this negative ion. The electrons in the bottom row
are divided equally between the bonded atoms; the difference between these numbers and those
above gives the formal charges.

Formal charge can also help answer the question “where is the charge located?” that is
frequently asked about polyatomic ions. Thus by writing out the Lewis structure for the
ammonium ion NH4+, you should be able to convince yourself that the nitrogen atom
has a formal charge of +1 and each of the hydrogens has 0, so we can say that the pos-
itive charge is localized on the central atom.
Oxidation number is another arbitrary way of characterizing atoms in molecules. In
contrast to formal charge, in which the electrons in a bond are assumed to be shared
equally, oxidation number is the electric charge an atom would have if the bonding elec-
trons were assigned exclusively to the more electronegative atom. Oxidation number
serves mainly as a tool for keeping track of electrons in reaction in which they are
exhanged between reactants, and for characterizing the “combining power” of an atom
in a molecule or ion.

Fig. 11: Comparison of electron assignments for formal charge and oxidation number.

5.4 Ionic compounds

The shared-electron pair model introduced by G.N. Lewis showed how chemical bonds
could form in the absence of electrostatic attraction between oppositely-charged ions.
As such, it has become the most popular and generally useful model of bonding in all
substances other than metals. A chemical bond forms when electrons are simulta-
neously attracted to two nuclei, thus acting to bind them together in an energetically-
stable arrangement. The covalent bond is formed when two atoms are able to share a

Page 26 Chem1 Chemical Bonding

pair of electrons:

In general, however, different kinds of atoms exert different degrees of attraction on

their electrons, so in most cases the sharing will not be equal. One can even imagine an
extreme case in which the sharing is so unequal that the resulting “molecule” is simply
a pair of ions:

The resulting substance is sometimes said to contain an ionic bond. Indeed, the proper-
ties of a number of compounds can be adequately explained using the ionic model. But
does this mean that there are really two kinds of chemical bonds, ionic and covalent?

Bonding in ionic solids

According to the ionic electrostatic model, solids such as NaCl

consist of positive and negative ions arranged in a crystal lat-
tice. Each ion is attracted to neighboring ions of opposite
charge, and is repelled by ions of like charge; this combination
of attractions and repulsions, acting in all directions, causes
the ion to be tightly fixed in its own location in the crystal lat-
tice.
Since electrostatic forces are nondirectional, the structure of
an ionic solid is determined purely by geometry: two kinds of
ions, each with its own radius, will fall into whatever repeating pattern will achieve the
lowest possible potential energy. Surprisingly, there are only a small number of possi-
ble structures; the very common simple cubic lattice of NaCl is illustrated here.

Is there such as thing as an ionic bond?

When two elements form an ionic compound, is an electron really lost by one atom and
transferred to the other one? In order to deal with this question, consider the data on
the ionic solid LiF. The average radius of the neutral Li atom is about 2.52Å. Now if this
Li atom reacts with an atom of F to form LiF, what is the average distance between the
Li nucleus and the electron it has “lost” to the fluorine atom? The answer is 1.56Å; the
electron is now closer to the lithium nucleus than it was in neutral lithium. So the
answer to the above question is both yes and no: yes, the electron that was now in the
2s orbital of Li is now within the grasp of a fluorine 2p orbital, but no, the electron is
now even closer to the Li nucleus than before, so how can it be “lost”? The one thing
that is inarguably true about LiF is that there are more electrons closer to positive
nuclei than there are in the separated Li and F atoms. But this is just the rule we
stated at the beginning of this unit: chemical bonds form when electrons can be simulta-
neously near two or more nuclei.

Page 27 Polar and nonpolar bonds

Fig. 12: Polar covalence in an “ionic” solid
After the 2s electron has been “lost” by lithium, it is even closer to the Li nucleus than before!

It is obvious that the electron-pair bond brings about this situation, and this is the rea-
son for the stability of the covalent bond. What is not so obvious (until you look at the
numbers such as were quoted for LiF above) is that the “ionic” bond results in the same
condition; even in the most highly ionic compounds, both electrons are close to both
nuclei, and the resulting mutual attractions bind the nuclei together. This being the
case, is there really any fundamental difference between the ionic and covalent bond?
The answer, according to modern chemical thinking1 is probably “no”; in fact, there is
some question as to whether it is realistic to consider that these solids consist of “ions”
in the usual sense. The preferred picture that seems to be emerging is one in which the
electron orbitals of adjacent atom pairs are simply skewed so as to place more electron
density around the “negative” element than around the “positive” one.
This being said, it must be reiterated that the ionic model of bonding is a useful one for
many purposes, and there is nothing wrong with using the term “ionic bond” to
describe the interactions between the atoms in “ionic solids” such as LiF and NaCl.

Polar covalence
If there is no such thing as a “completely ionic” bond, can we have one that is com-
pletely covalent? The answer is yes, if the two nuclei have equal electron attracting
powers. This situation is guaranteed to be the case with homonuclear diatomic mole-
cules-- molecules consisting of two identical atoms. Thus in Cl2, O2, and H2, electron
sharing between the two identical atoms must be exactly even; in such molecules, the
center of positive charge corresponds exactly to the center of negative charge: halfway
between the two nuclei.

Fig. 13: Mixed bond types

Categorizing all chemical bonds as either
ionic, covalent, or metallic is a gross oversim-
plification; as the diagram shows, there are
examples of compounds that exhibit varying
degrees of all three bonding characteristics.

1. See R.T. Sanderson: Chemical bonds and bond energy, Chapter 9 (Academic Press)

Page 28 Chem1 Chemical Bonding

5.5 Electron donor-acceptor bonds
In most covalent bonds, we think of the electron pair as having a dual parentage, one
electron being contributed by each atom. There are, however, many cases in which
both electrons come from only one atom. This can happen if the donor atom has a non-
bonding pair of electrons and the acceptor atom has a completely empty orbital that
can accommodate them.
This is the case, for example, with boron trifluoride and ammonia:
F H F H
F B + N H → F B N H
F H F H

In BF3, one the 2p orbitals is unoccupied and can accommodate the lone pair on the
nitrogen atom of ammonia. The electron acceptor, BF3, acts as a Lewis acid here, and
NH3 is the Lewis base.
Bonds of this type (sometimes known as coordinate covalent or dative bonds) tend to be
rather weak (usually 50-200kJ/mol); in many cases the two joined units retain suffi-
cient individuality to justify writing the formula as a molecular complex or adduct.

Page 29 Polar and nonpolar bonds

6. The Shapes of Molecules: the VSEPR model
Molecules are three-dimensional objects that occupy a three-dimensional world; it is
easy to forget this after seeing so many depictions of molecular structures on a two-
dimensional page. In general, only the smallest molecules can be said to have a fixed
geometrical shape; the icosahedral C60 “soccer ball” is a rare exception. In most mole-
cules, those parts joined by single bonds can rotate with respect to each other, giving
rise to many different geometric forms. However, the local geometry surrounding a
given atom that is covalently bound to its neighbors is constant. Being able to under -
stand and predict these bond angles is an important part of chemistry and is the sub-
ject of this section.
The Lewis electron-dot structures you have learned to draw have no geometrical sig-
nificance other than depicting the order in which the various atoms are connected to
one another. Nevertheless, a slight extension of the simple shared-electron pair con-
cept is capable of rationalizing and predicting the geometry of the bonds around a
given atom in a wide variety of situations.

6.1 Electron pair repulsion

The valence shell electron pair repulsion (VSEPR) model focuses on the bonding and
nonbonding electron pairs present in the outermost (“valence”) shell of an atom to
which are connected two or more other atoms. Like all electrons, these occupy orbitals
whose precise character can be left to more detailed theories. All you have to remember
at this stage is that these orbitals represent regions of negative electric charge, and
because like charges repel, these valence-shell orbitals will be energetically most stable
when they are as far from each other as possible.
The covalent model of chemical bonding assumes that the
electron pairs responsible for bonding are concentrated into
the region of apace between the bonded atoms. The funda-
mental idea of VSEPR theory is that these regions of negative
electric charge will repel each other, causing them (and thus
the chemical bonds) to stay as far apart as possible. If the
central atom also contains one or more pairs of nonbonding
electrons, these additional regions of negative charge will
behave very much like those associated with the bonded
atoms. The orbitals associated with the various bonding and
nonbonding pairs in the valence shell will extend out from the
central atom in directions that minimize their mutual repul-
sions.
If the central atom possesses partially occupied d -orbitals, it may be able to accommo-
date five or six electron pairs, forming what is sometimes called an “expanded octet”.
6.2 Coordination geometry and molecular geometry
Coordination number refers to the number of electron pairs that surround a given atom;
we often refer to the atom of interest as the central atom even if this atom is not really
located at the center of the molecule. If all of the electron pairs surrounding the central
atom are shared with neighboring atoms, then the coordination geometry is the same as
the molecular geometry. The application of VSEPR theory then reduces to the simple

Page 30 Chem1 Chemical Bonding

problem of naming (and visualizing) the geometric shapes associated with various
numbers of points surrounding a central point (the central atom) at the greatest possi-
ble angles. Both classes of geometry are named after the shapes of the imaginary geo-
metric figures (mostly regular solid polygons) that would be centered on the central
atom and would have an electron pair at each vertex.
If one of more of the electron pairs surrounding the central atom is not shared with a
neighboring atom (that is, if it is a lone pair), then the molecular geometry is simpler
than the coordination geometry, and it can be worked out by inspecting a sketch of the
coordination geometry figure.

6.3 Digonal and trigonal coordination

Linear molecules
The simplest possible case will be a triatomic molecule of the type AX 2 in which the
coordination number (number of bonded atoms X) around the central atom A is 2. What
angle between the two A-B bonds will give the greatest possible separation between
them, thus minimizing the repulsion between the associated electron clouds? It is not
very difficult to see that the answer is 180°; the two bonds must extend out from the A
atom in opposite directions.
Examples of triatomic molecules for which VSEPR theory
predicts a linear shape are BeCl2 (which, you will notice,
does not conform to the octet rule) and CO2. If you write
out the electron dot formula for carbon dioxide, you will see
that the C-O bonds are double bonds. This makes no difference to VSEPR theory; the
central carbon atom is still joined to two other atom, and the coordination number is
still only 2.

Plane-trigonal molecules
In the molecule BF3, there are three regions of elec-
tron density extending out from the central boron
atom. The repulsion between these will be at a mini-
mum when the angle between any two is 120°. This
requires that all four atoms be in the same plane;
the resulting shape is called trigonal planar, or sim-
ply trigonal.

6.4 Tetrahedral coordination

Methane, CH4, contains a carbon atom to which are connected four hydrogens. What
bond angle would lead to the greatest possible separation between the electron clouds
associated with these bonds? In analogy with the preceding two cases, where the bond
angles were 360°/2=180° and 360°/3=120°, you might guess 360°/4=90°; if so, you
would be wrong. The latter calculation would be correct if all the atoms were con-
strained to be in the same plane (we will see cases where this happens later), but here
there is no such restriction. Consequently, the four equivalent bonds will point in four
equivalent directions in three dimensions. The angle between any two bonds will be
109.5°.

Page 31 The Shapes of Molecules: the VSEPR model

This is called tetrahedral coordination; the four hydrogen atoms will be located at the
corners of a tetrahedron that has the carbon at its center. This is the most important
coordination geometry in Chemistry: it is imperative that you be able to sketch at least a
crude perspective view of a tetrahedral molecule.
It is interesting to note that the tetrahedral coordi-
nation of carbon in most of its organic compounds
was worked out in the nineteenth century on
purely geometrical grounds and chemical evidence.
For example, it was noted that there is only one
dichloromethane, CH2Cl2. If the coordination
around the carbon were square, then there would have to be two isomers of CH2Cl2,
as shown in the pair of structures above. The distances between the two chlorine
atoms would be different, giving rise to differences in physical properties would allow
the two isomers to be distinguished and separated.
The existence of only one kind of CH2Cl2 molecule means that all four positions sur-
rounding the carbon atom are geometrically equivalent, which requires a tetrahedral
coordination geometry. Unfortunately, this fact can only really be made convincing by
inspecting a three-dimensional mechanical model of the molecule.

Tetrahedrally-coordinated carbon chains

Carbon atoms are well known for their tendency to link together to form the millions of
organic molecules that are known. We can work out the simpler hydrocarbon chains by
looking at each central atom separately. Thus ethene consists of two plane-trigonal
CH2 units joined together, whereas ethane is essentially two CH3 tetrahedra joined
end-to-end.

Ethene, CH2CH2
Ethane, CH3CH3

Both of these structures can be extended indefinitely into very long hydrocarbon
chains.

6.5 Tetrahedral coordination with lone pairs

In the examples we have discussed so far, the shape of the molecule is defined by the
coordination geometry; thus the carbon in methane is tetrahedrally coordinated, and
there is a hydrogen at each corner of the tetrahedron, so the molecular shape is also
tetrahedral.

Page 32 Chem1 Chemical Bonding

This will not be the case when the valence shell of the central atom contains nonbond-
ing electrons. The reason is that the nonbonding electrons are also in orbitals that
occupy space and repel the other orbitals. This means that in figuring the coordination
number around the central atom, we must count both the bonded atoms and the non-
bonding pairs.

The water molecule

You will recall that CO2, lacking any lone pair electrons on the central atom, is a linear
molecule; there two bonded atoms and the coordination number around the carbon is
also two.
In the water molecule, the central atom is O, and the Lewis electron dot formula pre-
dicts that there will be two pairs of nonbonding electrons. The oxygen atom will there-
fore be tetrahedrally coordinated, even though the number of atoms bound to it is only
two. The two nonbonding orbitals are very much like the bonding ones, except that
there are no other atoms at their far ends.

Fig. 14: three views of the H2O molecule

Notice the two non-bonding electron pairs in the schematic diagrams at right and center. The computer-
generated picture on the right shows the higher electron density around the oxygen atom.

The O-H bond angle in water is 104.5°. This is somewhat less than the tetrahedral
angle of 109.5°, and it means that the four orbitals are not entirely equivalent. This is
not at all surprising, considering that two of them are nonbonding. Because a non-
bonding orbital has no atom at its far end to draw the electron cloud to it, the charge in
such an orbital will be concentrated closer to the central atom. As a consequence, non-
bonding orbitals exert more repulsion on other orbitals than do bonding orbitals. Thus
in H2O, the two nonbonding orbitals push the bonding orbitals closer together, closing
the angle somewhat.
The bent shape of the water molecule can be a source of confusion if you are not care-
ful: the oxygen atom in water is tetrahedrally coordinated, but the molecule itself has
a shape defined by its atoms rather than its orbitals. The two hydrogen atoms are sit-
uated near the corners of a tetrahedron that is centered on the oxygen atom, but
these three points define only a bent shape, not a complete tetrahedron.

Page 33 The Shapes of Molecules: the VSEPR model

Ammonia
The electron-dot structure of NH3 places one pair of nonbonding
electrons in the valence shell of the nitrogen atom. This means that
there are three bonded atoms and one lone pair, for a coordination
number of four around the nitrogen. We can therefore predict that
the three hydrogen atom will lie at the corners of a tetrahedron cen-
tered on the nitrogen atom. The lone pair orbital will point toward
the fourth corner of the tetrahedron, but since that position will be
vacant, the NH3 molecule itself cannot be tetrahedral. Instead, it assumes a pyramidal
shape. More precisely, the shape is that of a trigonal pyramid (i.e., a pyramid having a
triangular base). The hydrogen atoms are all in the same plane, with the nitrogen above
(or below, or to the side; molecules of course don’t know anything about “above” or
“below”!)

6.6 Atoms bonded to five atoms

Compounds of the type AB5 are formed by some of the elements in Group 15 of the
periodic table; PCl5 and AsF5 are examples.
In what directions can five electron pairs arrange themselves in space so as to minimize
their mutual repulsions? In the cases of coordination numbers 2, 3, 4, and 6, we could
imagine that the electron pairs distributed themselves as far apart as possible on the
surface of a sphere the resulting shapes correspond to the regular polyhedron whose
number of vertices is equal to the coordination number.
The problem with coordination number 5 is that there is no such thing as a regular
polyhedron with five vertices
In 1758, the great mathematician EULER proved that there are only five regular con-
vex polyhedra: tetrahedron (4 triangular faces), octahedron (6 triangular faces), icosa-
hedron (20 triangular faces), cube (6 square faces), and dodecahedron (12 pentagonal
faces). Chemical examples of all are known; the first icosahedral molecule, LaC60 (in
which the La atom has 20 nearest C neighbors) was prepared in 1986.
Besides the five regular solids, there can be 15 semi-regular isogonal solids in which
the faces have different shapes, but the vertex angles are all the same. These geomet-
rical principles are quite important in modern structural chemistry.

Page 34 Chem1 Chemical Bonding

The shape of PCl5 and similar molecules is a trigonal bipyra-
mid. This consists simply of two triangular -base pyramids
joined base-to-base. Three of the chlorine atoms are in the
plane of the central phosphorus atom, while the other two
atoms are above and below this plane.
Since the trigonal bipyramid is not a regular Eulerian poly-
hedron, the atoms in the axial positions above and below the
plane are not equivalent to those in the equatorial positions
in the plane. Thus in PCl5, there are two sets of Cl-P-Cl bond
angles (what are their values?), and the axial and equatorial
chlorine atoms will behave differently in chemical reactions.
The non-equivalence of the axial and equatorial positions
also affects the disposition of any lone pair electrons.
Because the nonbonding orbitals tend to concentrate elec-
tron density closer to the central atom, they exert more
repulsive force on other orbitals. This repulsion (and the
energy it costs) will be smaller when the nonbonding orbitals
occupy equatorial positions; this can be seen by comparing
the numbers neighboring orbitals at 90° and 120° to an axial
and an equatorial location.
Using this reasoning, we can predict that an AX4E molecule (that is, a molecule in
which the central atom A is coordinated to four other atoms “X” and to one nonbonding
electron pair) such as SF4 will have a “see-saw” shape; substitution of more nonbond-
ing pairs for bonded atoms reduces the triangular bipyramid coordination to even sim-
pler molecular shapes, as shown below. (See the examples on the next page.)

6.7 Octahedral coordination

Just as four electron pairs experience the minimum repul-
sion when they are directed toward the corners of a tetrahe-
dron, six electron pairs will try to point toward the corners
of an octahedron. An octahedron is not as complex a shape
as its name might imply; it is simply two square-based pyr-
amids joined base to base. You should be able to sketch this
shape as well as that of the tetrahedron.
At first, you might think that a coordination number of six
is highly unusual; it certainly violates the octet rule, and
there are only a few molecules (SF6 is one) where the central atom is hexavalent. It
Page 35 The Shapes of Molecules: the VSEPR model
turns out, however, that this is one of the most commonly encountered coordination
numbers in inorganic chemistry. There are two main reasons for this:
• Many transition metal ions form coordinate covalent bonds with lone-pair electron donor
atoms such as N (in NH3) and O (in H2O). Since transition elements can have an outer
configuration of d10s2, up to six electron pairs can be accommodated around the central
atom. A coordination number of 6 is therefore quite common in transition metal hydrates,
such as Fe(H2O)63+.
• Although the central atom of most molecules is bonded to fewer than six other atoms,
there is often a sufficient number of lone pair electrons to bring the total number of elec-
tron pairs to six.
There are well known examples of 6-coordinate central atoms with 1, 2, and 4 lone
pairs. Since the corners of an octahedron are geometrically equivalent, the distinction
between axial and equatorial atoms that we had to consider in the 5-coordinate case no
longer exists; thus the horizontal rectangles you see in the sketches below have no spe-
cial significance and could just as well be placed in either of the two possible vertical
planes.

6.8 Summary of VSEPR theory

The VSEPR model is an extraordinarily powerful one, considering its great simplicity.
Its application to predicting molecular structures can be summarized as follows:
1. Electron pairs surrounding a central atom repel each other; this repulsion will be
minimized if the orbitals containing these electron pairs point as far away from each
other as possible.
2. The coordination geometry around the central atom corresponds to the polyhedron
whose number of vertices is equal to the number of surrounding electron pairs (coor-
dination number). Except for the special case of 5, and the trivial cases of 2 and 3,
the shape will be one of the regular polyhedra.
3. If some of the electron pairs are nonbonding, the shape of the molecule will be sim-
pler than that of the coordination polyhedron.
4. Orbitals that contain nonbonding electrons are more concentrated near the central
atom, and therefore offer more repulsion than bonding pairs to other orbitals.
While VSEPR theory is quite good at predicting the general shapes of most molecules, it
cannot yield exact details. For example, it does not explain why the bond angle in H2O
is 104.5°, but that in H2S is about 90°. This is not surprising, considering that the
emphasis is on electronic repulsions, without regard to the detailed nature of the orbit-
als containing the electrons, and thus of the bonds themselves.

Page 36 Chem1 Chemical Bonding

7. Hybrid orbitals: the valence bond model
As useful and appealing as the concept of the shared-electron bond is, it raises a some-
what troubling question that we must sooner or later face: what is the nature of the
orbitals in which the shared electrons are contained? Up until now, we have been tac-
itly assuming that each valence electron occupies the same kind of atomic orbital as it
did in the isolated atom. As we shall see below, his assumption very quickly leads us
into difficulties.

Bonding in beryllium hydride

Consider how we might explain the bonding in a compound of
divalent beryllium, such as beryllium hydride, BeH2. The beryl-
lium atom, with only four electrons, has a configuration of 1s22s.
This means that there are only two electrons in the outer shell of
beryllium; these electrons are paired up together in the 2s orbital.
Now according to covalent bond theory, each beryllium-hydrogen
bond must consist of two electrons, shared between the two different atoms. One of
these electrons is contributed by the hydrogen, while the other comes from the beryl-
lium atom.
The question then arises: how can the two outer -shell electrons of beryllium interact
with the hydrogen electrons? Two electrons in the same orbital have opposite spins,
and constitute a stable pair that has no tendency to interact with unpaired electrons
on other atoms. Electron sharing, as we have seen so far, can only take place when
UNpaired electrons on adjacent atoms interact.
Fig. 15: Electron configurations of Be
in its ground- and excited states
The ground state of an atom is its normal,
lowest-energy state. The excited state
shown accounts for the bivalency of Be, but
not for the equivalence of the two bonded
atoms.

The only way that we can obtain two unpaired electrons for bonding in beryllium is to
promote one of the 2s electrons to the 2p level. However, the energy required to carry
out this promotion would be sufficiently great to discourage bond formation. It is
observed that Be does form reasonably stable bonds with other atoms. Moreover, the
two bonds in BeH2 and similar molecules are completely equivalent; this would not be
the case if the electrons in the two bonds shared Be orbitals of different types.
In fact, there is little reason to believe that s, p, and d orbitals really do exist in the
outer shells of many bonded atoms. Remember that these different orbitals arise in the
first place from the interaction of the electron with the central electrostatic force field
associated with the positive nucleus. An outer -shell electron in a bonded atom will be
under the influence of a force field, emanating from two positive nuclei rather than one,
so we would expect the orbitals in the bonded atoms to have a somewhat dif ferent
character from those in free atoms. We can, in fact, throw out the concept of atomic
orbital altogether and reassign the electrons to a new set of molecular orbitals that are
characteristic of each molecular configuration. This approach is indeed valid, but we
will defer a discussion of it until later. For now, we will look at a less-radical model that

Page 37 Hybrid orbitals: the valence bond model

starts out with the familiar valence-shell atomic orbitals, and allows them to combine
to form hybrid orbitals whose shapes conform quite well to the bonding geometry that
we observe in a wide variety of molecules.

7.1 What are hybrid orbitals?

Orbitals of all types are simply mathematical functions that describe particular stand-
ing-wave patterns that can be plotted on a graph but have no physical reality of their
own. Because of their wavelike nature, two or more orbitals (i.e., two or more functions
ψ) can be combined both in-phase and out-of-phase to yield a pair of resultant orbitals
which, to be useful, must have squares that describe actual electron distributions in
the atom or molecule.
The s,p,d and f orbitals that you are familiar with are the most convenient ones for
describing the electron distribution in isolated atoms because assignment of electrons
to them according to the usual rules always yields an overall function ψ2 that predicts
a spherically symmetric electron distribution, consistent with all physical evidence that
atoms are in fact spherical. For atoms having more than one electron, however, the
s,p,d, f basis set is only one of many possible ways of arriving at the same observed
electron distribution. Because it happens to be the simplest description of the atom we
use it, but it is not unique.
In the case of a molecule such as BeH2, we know from experimental evidence that the
molecule is linear and therefore the electron density surrounding the central atom is
no longer spherical, but must be concentrated along two directions 180° apart, and we
need to construct a function ψ2 having these geometrical properties. There are any
number of ways of doing this, but the simplest is to use a new set of functions ψ (which
we call hybrid orbitals) that are constructed by combining the atomic s,p,d, f functions
that are already familiar to us.
You should understand that hybridization is not a physical phenomenon; it is merely a
mathematical operation that allows us to describe the electron distribution about a
bonded atom in terms of one particular set of functions that we prefer to use because
it is convenient to do so.
This approach, which assumes that the orbitals remain more or less localized on one
central atom, is the basis of the valence-bond theory which was developed in the early
1930s, mainly by LINUS PAULING1. Although it has proven very powerful (especially in
organic chemistry), it does have its limitations. For example, it predicts that both H2O
and H2S will be tetrahedrally coordinated bent molecules with bond angles slightly
smaller than the tetrahedral angle of 109.5° owing to greater repulsion by the non-
bonding pair. This description fits water (104.5°) quite well, but the bond angle in
hydrogen sulfide is only 92°, suggesting that atomic p orbitals (which are 90° apart)
provide a better description of the electron distribution about the sulfur atom than do

1. Linus Pauling (1901-1994) was the most famous American chemists of the 20th cen-
tury and the author of the classic The nature of the chemical bond. His early work pio-
neered the application of X-ray diffraction to determine the structure of complex
molecules; he then went on to apply quantum theory to explain these observations and
predict the bonding patterns and energies of new molecules. Pauling, who spent most of
his career at Cal Tech, won the Nobel Prize for Chemistry in 1954 and the Peace Prize in
1962.

Page 38 Chem1 Chemical Bonding

sp3 hybrid orbitals.
The valence bond approach tends to be much better at explaining or rationalizing the
structure of a known molecule, than it is at predicting a structure or shape. In other
words, it works for some molecules, but not for others. This model is fairly simple to
apply and understand, but it is best regarded as one special way of looking at a mole-
cule that can often be misleading. Another viewpoint, called the molecular orbital the-
ory, offers us a complementary perspective that it is important to have if we wish to
develop a really thorough understanding of chemical bonding in a wider range of mole-
cules.

Understanding hybrid orbitals

Hybrid orbitals are formed from “constructive” and “destructive” combinations of 2p
and 2s atomic wave functions, as illustrated by the line plots below. The solid figures
depict the corresponding probability functions ψ2 which describe the electron density
in the various directions around the bonded atm.

atomic orbitals hybrid orbitals First, recall that the electron, being a
quantum particle, cannot have a dis-
2p tinct location; the most we can do is
2s sp
define the region of space around the
nucleus in which the probability of find-
ing the electron exceeds some arbi-
trary value, such as 90% or 99%. This
region of space is the orbital. Because
in-phase combination of the wavelike character of matter, the
orbital corresponds to a standing wave
pattern in 3-dimensional space which
2s we can often represent more clearly in
2-dimensional cross section. The
p quantity that is varying (“waveing”) is a
number denoted by ψ whose value
varies from point to point according to
the wave function for that particular
orbital.

Hybrid orbitals are constructed by

combining the ψ functions for atomic
orbitals. Because wave patterns can
combine both constructively and
destructively, a pair of atomic wave
functions such as the s- and p- orbitals
sp shown at the left can combine in two
out-of-phase combination ways, yielding the sp hybrids shown.

The probability of finding the electron

2s 2p sp at any location is given not by ψ, but by
ψ2, whose form is roughly conveyed by
the solid figures in this illustration.

Page 39 Hybrid orbitals: the valence bond model

7.2 Digonal bonding: sp-hybrid orbitals
To see how the concept of hybrid orbitals helps us understand bonding, let’s go back to
the example of BeH2 that we mentioned on Page 37. Instead of assuming that the two
bonding electrons occupy s- and p- orbitals characteristic of the isolated atoms, we
place them in sp hybrid orbitals of identical character which arise from in- and out-of-
phase combinations of the atomic orbitals as described above.

Each of the two sp-hybrid orbitals contains a single

unpaired electron which is readily shared with a similar
electron on another atom such as hydrogen. Notice how
the sp hybrid orbital is intermediate in energy between
the 2s and 2p orbitals of atomic beryllium. This means
that it doesn’t cost the beryllium atom much in energy to
get its two outer electrons into the sp hybrid- what the
beryllium atom must “pay” in order to promote its 2s
electron up to the level of the sp hybrid, the atom more
than gains in the energy lost when the previously promoted 2p electron now drops
down to the sp-hybrid level.

Shape of the sp hybrid orbital

What is the shape of the electron cloud corresponding to the sp hybrid orbital? As we
might expect, the characteristics of the parents can be seen in the progeny: the p-
orbital is elongated, and so is the hybrid, but since the s-orbital is spherical, the hybrid
is not as drawn out as a pure p orbital. Notice that whereas a single p-orbital has lobes
on both sides of the atom, a single sp-hybrid has most of its electron density on one
side.The two sp hybrid orbitals in the bonded beryllium atom point away from the cen-
tral atom in opposite directions. We can expect any central atom that uses sp-hybrid-
ization in bonding to exhibit linear geometry when incorporated into a molecule.

Fig. 16: Formation of a sp-hybrid orbital

Mixing of an s- and a p-orbital yields the two sp-hybrid orbitals shown in the middle picture. At the far right
the two hybrid orbitals are centered on the same atom, giving rise to the linear geometry.
[From www.wellesley.edu/Chemistry/chem120/hybrids.html]

Sigma bonds

Page 40 Chem1 Chemical Bonding

The figure depicts the electron clouds in a complete beryl-
lium hydride molecule. Notice that the hydrogen 1s orbit-
als are not hybridized here; they are still spherical. The two
“bonds” in beryllium hydride are the regions of overlap
between the hydrogen 1s and beryllium sp-hybrid orbitals.
Since this bonding region is symmetrical about the line of
centers joining the three atoms, the resulting bond is said
to be of the (sigma) type. A σ bond is one whose electron
density is concentrated around the line-of-centers between
two or more atoms.

7.3 Trigonal ( sp2) hybridization

We can now go on to apply the same ideas to some
other simple molecules. In boron trifluoride, for exam-
ple, we start with the boron atom, which has three
outer-shell electrons in its normal or ground state,
and three fluorine atoms, each with seven outer elec-
trons. As shown in the upper diagram, one of the three
boron electrons is unpaired in the ground state. In
order to explain the trivalent bonding of boron, we
postulate that the atomic s- and p- orbitals in the
outer shell of boron mix to form three equivalent hybrid orbitals. These particular
orbitals are called sp2 hybrids, meaning that this set of orbitals is derived from one s-
orbital and two p-orbitals of the free atom.
.

Boron trifluoride has a plane trigonal shape; a 2p orbital

on each fluorine atom overlaps with a boron sp2 hybrid. In
general, we can expect that all molecules in which a cen-
tral atom uses three equivalent sp2 hybrid orbitals will
exhibit plane trigonal geometry, since this represents the
most symmetrical, and hence “equivalent”, arrangement of
the three bonds.

Page 41 Hybrid orbitals: the valence bond model

7.4 Tetrahedral ( sp3) hybridization
Let us now look at several tetravalent molecules, and see what kind of hybridization
might be involved when four outer atoms are bonded to a central atom. Perhaps the
commonest and most important example of this bond type is methane, CH 4.

In the ground state of the free carbon atom, there are two unpaired electrons in sepa-
rate 2p orbitals. In order to form four bonds (tetravalence), need four unpaired elec-
trons in four separate but equivalent orbitals. We assume that the single 2s, and the
three 2p orbitals of carbon mix into four sp3 hybrid orbitals which are chemically and
geometrically identical; the latter condition implies that the four hybrid orbitals extend
toward the corners of a tetrahedron centered on the carbon atom.

Lone pair electrons

If lone pair electrons are present on the central atom, these can occupy one or more of
the sp3 orbitals. This causes the molecular geometry to be different from the coordina-
tion geometry, which remains tetrahedral.
In the ammonia molecule, for example, the nitrogen atom nor-
mally has three unpaired p electrons, but by mixing the 2s
and 3p orbitals, we can create four sp3-hybrid orbitals just as
in carbon. Three of these can form shared-electron bonds with
hydrogen, resulting in ammonia, NH3. The fourth of the sp3
hybrid orbitals contains the two remaining outer-shell elec-
trons of nitrogen which form a non-bonding lone pair. In acidic
solutions these can coordinate with a hydrogen ion, forming
the ammonium ion NH4+.
Although no bonds are formed by the lone pair in NH3, these
electrons do give rise to a charge cloud that takes up space
H H just like any other orbital, and thus the nonbonding electrons
H occupy one of the four tetrahedral regions. The other three
corners of the tetrahedron are occupied by hydrogen atoms,
and these three points define the base of the pyramid that
describes the shape of molecules containing three bonding
pairs and one lone pair of electrons around the central atom.
In the water molecule, the oxygen atom can form four sp3 orbitals. Two of these are
occupied by the two lone pairs on the oxygen atom, while the other two are used for
bonding. The observed H-O-H bond angle in water (104.5°) is less than the tetrahedral
angle (109.5°); one explanation for this is that the non-bonding electrons tend to
remain closer to the central atom and thus exert greater repulsion on the other orbit-
als, thus pushing the two bonding orbitals closer together.

Page 42 Chem1 Chemical Bonding

Hybridization can also help explain the existence and structure of many inorganic
molecular ions. Consider, for example, the zinc compounds shown here. At the top is
shown the electron configuration of atomic zinc, and just below it, of the divalent zinc
ion. Notice that this ion has no electrons at all in its 4-shell. In zinc chloride, shown in
the third row, there are two equivalent chlorine atoms bonded to the zinc. The bonding
orbitals are of sp character; that is, they are hybrids of the 4s and one 4p orbital of the
zinc atom. Since these orbitals are empty in the isolated zinc ion, the bonding electrons
themselves are all contributed by the chlorine atoms, or rather, the chlor ide ions, for it
is these that are the bonded species here. Each chloride ion possesses a complete octet
of electrons, and two of these electrons occupy each sp bond orbital in the zinc chloride
complex ion. This is an example of a coordinate covalent bond, in which the bonded
atom contributes both of the electrons that make up the shared pair.

Fig. 17: Multiple hybrid-

ization schemes in zinc
compounds
The empty 4s and 4p orbit-
als of Zn2+ allow it to
accept electron pairs from
two chlorine atoms in solid
ZnCl2, or from four in the
complex ion tetrachlo-
rozinc(II).

The tetrachloro zinc ion is another structure derived from zinc and chlorine. As we
might expect, this ion is tetrahedral; there are four chloride ions surrounding the cen-
tral zinc ion. The zinc ion has a charge of +2, and each chloride ion is -1, so the net
charge of the complex ion is -2.
7.5 Hybrid types and multiple bonds
Up to now we have dealt with molecules in which the orbitals on a single central atom
are considered to be hybridized. Let us now see how these various models of hybridiza-
tion can help us understand bonding between pairs of such atoms. Because of its

Page 43 Hybrid orbitals: the valence bond model

extreme importance, we will use the element carbon as an example.

Fig. 18: Bonding in ethane

The carbon-carbon bond is formed by overlap of an sp3 orbital from each carbon, creating a space in
which two electrons can be simultaneously close to two nuclei as in any covalent bond.

We have already seen how sp hybridization in carbon leads to a combining power of

four in the methane molecule. Two such tetrahedrally coordinated carbons can link up
together to form the molecule ethane C2H6. In this molecule, each carbon is bonded in
the same way as the other; each is linked to four other atoms, three hydrogens and one
carbon, by means of straight-line σ bonds.The ability of carbon-to-carbon linkages to
extend themselves indefinitely and through all coordination positions accounts for the
millions of organic molecules that are known.

Trigonal hybridization in carbon: the double bond

But carbon and hydrogen can also form a compound, ethyl-
ene, in which each carbon atom is linked to only three other
atoms. Here, we can regard carbon as being trivalent. We
can explain this trivalence by supposing that the orbital
hybridization in carbon is in this case not sp3, but is sp2
instead; in other words, only two of the three p orbitals of
carbon mix with the 2s orbital to form hybrids; the remain-
ing p-orbital, which we will call the pz orbital, does not take
part in bond formation. Each carbon is bonded to three other atoms in the same kind
of plane trigonal configuration that we saw in the case of boron trifluoride, where the
same kind of hybridization occurs. Notice that the bond angles around each carbon are
all 120°.

Page 44 Chem1 Chemical Bonding

Fig. 19: bond hybridization in ethylene
These three views of the ethylene molecule emphasize different aspects of the disposition of
shared electron pairs in the various bonding orbitals of ethene (ethylene). (a) The “backbone”
structure consisting of sigma (σ) bonds formed from the three sp2-hybridized orbitals on each
carbon. (b) The π (pi) bonding system formed by overlap of the unhybridized pz orbital on
each carbon. The pi orbital has two regions of electron density extending above and below the
plane of the molecule. (c) A cutaway view of the combined sigma and pi system.

This alternative hybridization scheme explains how carbon can combine with four
atoms in some of its compounds and with three other atoms in other compounds. You
may be aware of the conventional way of depicting carbon as being tetravalent in all its
compounds; it is often stated that carbon always forms four bonds, but that some-
times, as in the case of ethylene, one of these may be a double bond. This concept of
the multiple bond preserves the idea of tetravalent carbon while admitting the exist-
ence of molecules in which carbon is clearly combined with fewer than four other
atoms.
As shown above, ethylene can be imagined to form when two -CH2 fragments link
together through overlap of the half-filled sp2 hybrid orbitals on each. Since sp2 hybrid
orbitals are always in the same plane, the entire ethylene molecule is planar. However,
there remains on each carbon atom an electron in an unhybridized atomic pz orbital
that is perpendicular to the molecular plane. These two parallel pz orbitals will interact
with each other; the two orbitals merge, forming a sausage-like charge cloud (the p
bond) that extends both above and below the plane of the molecule. It is the pair of
electrons that occupy this new extended orbital that constitutes the “fourth” bond to
each carbon, and thus the “other half” of the double bond in the molecule.

More about sigma and pi bonds

The σ (sigma) bond has its maximum electron density along the line-of-centers joining
the two atoms (below left). Viewed end-on, the σ bond is cylindrically symmetrical
about the line-of-centers. It is this symmetry, rather than its parentage, that defines
the sigma bond, which can be formed from the overlap of two s-orbitals, from two p-
orbitals arranged end-to-end, or from an s- and a p-orbital. They can also form when

Page 45 Hybrid orbitals: the valence bond model

sp-hybrid orbitals on two atoms overlap end-to-end.

Fig. 20: Sigma- and pi bonds.

The sigma bond, being symmetric about the x-axis of the molecule, can allow free rotation of the two joined atoms.
Pi bonds prevent this rotation, giving rise to the possibility of geometrical isomers in multiply-bonded compounds.

Pi orbitals, on the other hand, require the presence of two atomic p orbitals on adja-
cent atoms. Most important, the charge density in the π orbital is concentrated above
and below the molecular plane; it is almost zero along the line-of-centers between the
two atoms.It is this perpendicular orientation with respect to the molecular plane (and
the consequent lack of cylindrical symmetry) that defines the π orbital. The combina-
tion of a σ bond and a π bond extending between the same pair of atoms constitutes the
double bond in molecules such as ethylene.

Triple bonds
We have not yet completed our overview of multiple bonding, however. Carbon and
hydrogen can form yet another compound, acetylene, in which each carbon is con-
nected to only two other atoms: a carbon and a hydrogen. This can be regarded as an
example of divalent carbon, but is usually rationalized by writing a triple bond between
the two carbon atoms.
We assume here that since two geometrically equivalent bonds are formed by each car-
bon, this atom must be sp-hybridized in acetylene. On each carbon, one sp hybrid
bonds to a hydrogen and the other bonds to the other carbon atom, forming the σ bond
skeleton of the molecule.
In addition to the sp hybrids, each carbon atom has two half-occupied p orbitals, ori-
ented at right angles to each other, and to the interatomic axis. These are two sets of
parallel and adjacent p orbitals, and they can thus merge into two sets of π orbitals.

Fig. 21: Triple bond in acetylene.

The two pi bonds (occupying orthogonal planes) are depicted by the colored horizontal
lines on the right.

The triple bond in acetylene is seen to consist of one σ bond joining the line-of-centers
between the two carbon atoms, and two π bonds whose lobes of electron density are in
mutually-perpendicular planes. The acetylene molecule is of course linear, since the
angle between the two sp hybrid orbitals that produce the σ skeleton of the molecule is

Page 46 Chem1 Chemical Bonding

180°.

Multiple bonds between unlike atoms

Multiple bonds can also occur between dissimilar atoms. For
example, in carbon dioxide each carbon atom has two unhy-
bridized atomic p orbitals, and each oxygen atom still has one
p orbital available. When the two O-atoms are brought up to
opposite sides of the carbon atom, one of the p orbitals on
each oxygen forms a π bond with one of the carbon p-orbitals. In this case, sp-hybrid-
ization is seen to lead to two double bonds. Notice that the two C-O π bonds are mutu-
ally perpendicular.
Similarly, in hydrogen cyanide, HCN, we assume that
the carbon is sp-hybridized, since it is combined with only
two other atoms, and is hence in a divalent state. One of
the sp-hybrid orbitals overlaps with the hydrogen 1s
orbital, while the other overlaps end-to-end with one of
the three unhybridized p orbitals of the nitrogen atom.
This leaves us with two nitrogen p-orbitals, which form
two mutually perpendicular π bonds to the two unhybrid-
ized p orbitals on the carbon. Hydrogen cyanide thus contains one single and one triple
bond, the latter consisting of a σ bond from the overlap of the two sp hybrid orbitals,
and two mutually perpendicular π] bonds deriving from unhybridized atomic p orbitals
on the carbon and nitrogen atoms.

7.6 Conjugated double bonds

We have seen that the π bonding orbital is distinctly different in shape and symmetry
from the σ bond. There is another important feature of the π bond that is of far-reach-
ing consequence, particularly in organic and coordination chemistry.
Consider, for example, an extended hydrocarbon molecule in which alternate pairs of
carbon atoms are connected by double and single bonds. Each non-terminal carbon
atom forms two s bonds to two other carbons and to a hydrogen (not shown.) This mol-
ecule can be viewed as a series of ethylene molecules joined together end-to-end. Each
carbon, being sp hybridized, still has a half-filled atomic p orbital. Since these p orbit-
als on adjacent carbons are all parallel, we can expect them to interact with each other
to form p bonds between alternate pairs of carbon atoms as shown below.

Fig. 22:
A conjugated carbon
chain
When single and double
bonds alternate in a chain of
carbon atoms, the two π
bonding arrangements
shown at the top are equiva-
lent. The resulting structure
is said to be conjugated and
tend to be especially stable.
The C–C bond order is 1.5.

Page 47 Hybrid orbitals: the valence bond model

Each carbon, being sp hybridized, has a half-filled atomic p orbital that can form a p
bond with either of its neighbors (but not with both at the same time.) You will recall
that when there are two equivalent choices for the arrangements single and double
bonds in a molecule, we generally consider the structure to be a resonance hybrid. In
keeping with this idea, we would expect the electron density in the π system of this
kind to be extended or shared out evenly along the entire molecular framework (bottom
structure above.)
A system of alternating single and double bonds, as we have here, is called a conju-
gated system. Chemists say that the π bonds in a conjugated system are delocalized;
they are, in effect, “smeared out” over the entire length of the conjugated part of the
molecule.

An even higher degree of con-

jugation exists in compounds
containing extended (C=C)n
chains. These compounds,
known as cumulenes, form
polymers that exhibit inter-
esting electrical properties
and whose derivatives can act
as “organic wires”.

Benzene
The classic example of π bond delocalization is found in the cyclic molecule benzene
(C6H6), which consists of six carbon atoms bound together in a hexagonal arrange-
ment. Each carbon has a single hydrogen atom attached to it. Earlier, you learned to
represent the benzene structure as a composite of two resonance forms.
Modern hybridization theory offers a somewhat less contrived view: each carbon atom
is sp hybridized, producing trigonal bond geometry about each node in the hexagon.
Each carbon atom also has a half-occupied unhybridized p orbital; two of these
together, on adjacent carbon atoms, can form two equivalent sets of π bonds as shown
in the center section. The net effect is depicted at the right. The π bonding system is
drawn out into two donut shaped electron clouds, one on either side of the molecular
plane. The “double bonds” in benzene are delocalized so that they extend around the
entire hexagonal skeleton.

7.7 The nitrate ion

Pi bond delocalization furnishes a means of expressing the structures of other mole-
cules that require more than one electron-dot or structural formula for their accurate

Page 48 Chem1 Chemical Bonding

Fig. 23: Bonding in benzene
Left: the unhybridized 2p atomic orbitals of the carbon atoms
are shown superimposed on the trigonally-hybridized sp2
sigma bonding system. Right: overlap of the sp2 orbitals leads
to a pi orbital system that extends around the molecule

representation. A good example is the nitrate ion.

Nitrogen has three half-occupied p orbitals available for bonding, all perpendicular to
one another. Since the nitrate ion is known to be planar, we are forced to assume that
the nitrogen outer electrons are sp2 hybridized. The addition of an extra electron fills
all three hybrid orbitals completely. Each of these filled sp2 orbitals forms a σ bond by
overlap with an empty oxygen 2pz orbital; this, you will recall, is an example of coordi-
nate covalent bonding, in which one of the atoms contributes both of the bonding elec-
trons.
The empty oxygen 2p orbital is made available when the oxygen electrons themselves
become sp hybridized; we get three filled sp hybrid orbitals, and an empty 2p atomic
orbital, just as in the case of nitrogen.

Fig. 24: Sigma and pi bonds in the nitrate ion

The π bonding system arises from the interaction of one of the occupied oxygen sp
orbitals with the unoccupied 2pz orbital of the nitrogen. Notice that this, again, is a
coordinate covalent sharing, except that in this instance, it is the oxygen atom that

Page 49 Hybrid orbitals: the valence bond model

donates both electrons.
Pi bonds can form in this way between the nitrogen atom and any of the three oxygens;
there are thus three equivalent π bonds possible, but since nitrogen can only form one
complete π bond at a time, the π bonding is divided up three ways, so that each N-O
bond has a bond order of 4/3.

7.8 Hybrids involving d orbitals

In atoms that are below those in the first complete row of the periodic table, the simple
octet rule begins to break down. For example, we have seen that PCl3 does conform to
the octet rule but PCl5 does not. We can describe the bonding in PCl3 very much as we
do NH3: four sp3 hybridized orbitals, three of which are shared with electrons from
other atoms and the fourth containing a nonbonding pair.
In order to understand the bonding in PCl5, we have to consider the d orbitals in addi-
tion to the s- and p types. When d orbitals are energetically close to the outmost s- and
p orbitals, additional hybrid types can be built. In the case of PCl 5 we need five hybrid
orbitals, and these can be constructed by adding two d-orbital functions to the mathe-
matical mixture of one s- and two p-orbitals, resulting in five sp3d2 hybrid orbitals
directed toward the corners of a trigonal bipyramid, as is predicted by VSEPR theory.

Fig. 25: Bonding in PCl5

Notice that in the trigonal bipyramid, the axial and equatorial (in-plane) atoms are not
equivalent as evidenced by the shorter lengths of the equatorial bonds.

Some of the most important and commonly encountered compounds which involve the
d orbitals in bonding are the transition metal complex. The term “complex” in this con-
text means that the molecule is composed of two or more kinds of species, each of
which can have an independent existence. For example, the ions Pt 2+ and Cl– can form
the ion [PtCl4]2– which is found experimentally to be square planar. To understand the
hybridization scheme, it helps to start with the neutral Pt atom, then imagine it losing
two electrons to become an ion, followed by grouping of the two unpaired 5d electrons
into a single d orbital, leaving one vacant.This vacant orbital, along with the 6s and two
of the 6p orbitals, can then accept an electron pair from four chlorines.

Octahedral coordination

Page 50 Chem1 Chemical Bonding

The molecule sulfur hexafluoride SF6 exemplifies one of the most common types of d-
orbital hybridization. The six bonds in this octahedrally-coordinated molecule are
derived from mixing six atomic orbitals into a hybrid set. The easiest way to under -
stand how these come about is to imagine that the molecule is made by combining an
imaginary S6+ ion (which we refer to as the S(VI) valence state) with six F– ions to form
the neutral molecule. These now-empty 3s and 3p orbitals then mix with two 3d orbit-
als to form the sp3d2 hybrids.

Fig. 26: Bonding in sulfur hexafluoride

“S(VI)” represents the six-coordinated valence state of sulfur, which is considered to have lost its
3s and 3p electrons, making six equivalent sp3d2-hybridized orbitals available for accepting lone
pairs contributed by the fluorine atoms.

This same type of hybridization can be invoked to explain the structures of many tran-
sition metal cations; the hexamminezinc(II) cation depicted below is typical.

7.9 Transition metal complexes

Some of the most important and commonly encountered compounds which involve the
d orbitals in bonding are the transition metal complexes. The term “complex” in this
context means that the molecule is composed of two or more kinds of species, each of
which can have an independent existence.

Square-planar molecules: dsp2 hybridization

For example, the ions Pt2+ and Cl– can form the ion [PtCl4]2–. To understand the
hybridization scheme, it helps to start with the neutral Pt atom, then imagine it losing
two electrons to become an ion, followed by grouping of the two unpaired 5d electrons
into a single d orbital, leaving one vacant.This vacant orbital, along with the 6s and two
of the 6p orbitals, can then accept an electron pair from four chlorines.

Fig. 27: bonding in [PtCl4]2–

All four Pt-Cl bonds are equivalent and point to the corners of a square centered on the Pt

All of the four-coordinated molecules we have discussed so far have tetrahedral geome-
try around the central atom. Methane, CH4, is the most well known example. It may
come as something as a surprise, then, to discover that the tetrachlorplatinum (II)
ion [PtCl4]2– has an essentially two-dimensional square-planar configuration. This

Page 51 Hybrid orbitals: the valence bond model

type of bonding pattern is quite common when the parent central ion (Pt2+ in this case)
contains only eight electrons in its outmost d-subshell.

Octahedral complexes: sp3d2 and d2sp3

Many of the most commonly encountered transition metal ions accept electron pairs
from donors such as CN– and NH3 (or lacking these, even from H2O) to form octahedral
coordination complexes. The hexamminezinc(II) cation depicted below is typical.

Fig. 28: bonding in

Zn(NH3)62+
Again, the bonding electron
pairs are contributed by the
ligands, in this case ammonia,
and they fill hybrids made by
mixing the lowest-energy
unoccupied orbitals of the
central atom.

In sp3d2 hybridization the bonding orbitals are derived by mixing atomic orbitals hav-
ing the same principal quantum number (n = 4 in the preceding example). A slightly
different arrangement, known as d2sp3 hybridization, involves d orbitals of lower prin-
cipal quantum number. This is possible because of the rather small energy differences
between the d orbitals in one “shell” with the s and p orbitals of the next higher one—
hence the term “inner orbital” complex which is sometimes used to describe ions such
as hexaminecobalt(III), shown below. Both arrangements produce octahedral coordi-
nation geometries.

Fig. 29: bonding in

Co(NH3)62+
Here the two d orbitals
needed for mixing are made
available by consolidating
the 3d electrons of Co.

In some cases, the same central atom can form either inner or outer complexes
depending on the particular ligand and the manner in which its electrostatic field
affects the relative energies of the different orbitals.Thus the hexacyanoiron(II) ion uti-
lizes the iron 3d orbitals, whereas hexaaquoiron(II) achieves a lower energy by accept-
ing two H2O molecules in its 4d orbitals.

Page 52 Chem1 Chemical Bonding

Fig. 30: “Inner” and “outer” orbital octahedral coordination

Although this “inner-outer” hybrid model was a instrumental in

explaining the properties of transition metal complexes in the first
half of the 20th century, it has now been replaced with a more com-
prehensive model known as ligand field theory which is introduced

7.10 Limitations of the hybrid model

As is the case with any scientific model, the hybridization model of bonding is useful
only to the degree to which it can predict phenomena that are actually observed. Most
models contain weaknesses that place limits on their general applicability. The need for
caution in accepting this particular model is made more apparent when we examine
the shapes of the molecules below the first row of the periodic table. For example, we
would expect the bonding in hydrogen sulfide to be similar to that in water, with tetra-
hedral geometry around the sulfur atom. Experiments, however, reveal that the bond
H-S-H bond angle is only 92°. Hydrogen sulfide thus deviates much more from tetrahe-
dral geometry than does water, and there is no apparent and clear reason why it
should. It is certainly difficult to argue that electron-repulsion between the two non-
bonding orbitals is pushing the H-S bonds closer together (as is supposed to happen to
the H-O bonds in water); many would argue that this repulsion would be less in hydro-
gen sulfide than in water, since sulfur is a larger atom and is hence less electronega-
tive.
It might be, then, that our model of orbital hybridization into four equivalent sp3 orbit-
als does not apply to H2S. It looks like the “simple” explanation that bonding occurs
through two half occupied atomic- p orbitals 90° apart comes closer to the mark. Per -
haps hybridization is not an all-or -nothing phenomenon; perhaps the two 3p orbitals
are substantially intact in hydrogen sulfide, or are only hybridized very slightly. In gen-
eral, the hybridization model does not work very well with nonmetallic elements below
the first row of the periodic table, and there is as yet no clear explanation why. We
must simply admit that we have reached one of the many points in chemistry where
our theory is not sufficiently developed to give a clear and unequivocal answer. This

Page 53 Hybrid orbitals: the valence bond model

does not detract, however, from the wide usefulness of the hybridization model in eluci-
dating the bond character and bond shapes in the millions of molecules based on first-
row elements, particularly of carbon.

Are hybrid orbitals real?

The justification we gave for invoking hybridization in molecules such as BeH 2, BF3
and CH4 was that the bonds in each are geometrically and chemically equivalent,
whereas the atomic s- and p-orbitals on the central atoms are not. By combining these
into new orbitals of sp, sp2 and sp3 types we obtain the required number of completely
equivalent orbitals. This seemed easy enough to do on paper; we just drew little boxes
and wrote “sp2” or whatever below them. But what is really going on here?
The full answer is beyond the scope of this course, so we can only of fer the following
very general explanation. First, recall what we mean by “orbital”: a mathematical func-
tion ψ having the character of a standing wave whose square ψ2 is proportional to the
probability of finding the electron at any particular location in space. The latter, the
electron density distribution, can be observed (by X-ray scattering, for example), and in
this sense is the only thing that is “real”.
A given standing wave (ψ function) can be synthesized by combining all kinds of funda-
mental wave patterns (that is, atomic orbitals) in much the same way that a color we
observe can be reproduced by combining different sets of primary colors in various pro-
portions. In neither case does it follow that these original orbitals or colors are actually
present in the final product.

An alternative to hybrids: the Bent-Bond model.

It turns out, in fact, that the electron distribution and bonding
in ethylene can be equally well described by assuming no
hybridization at all. The “bent bond” model depicted at the right
requires only that the directions of some of the atomic-p orbit-
als be distorted sufficiently to provide the overlap needed for
bonding. So one could well argue that hybrid orbitals are not
“real”; they do turn out to be convenient for understanding the
bonding of simple molecules at the elementary level, and this is why we use them.

8. The molecular orbital Model

In the models of chemical bonding we have discussed up to now, we have assumed that
the electrons that interpose themselves between adjacent nuclei (the “bonding elec-
trons”) are in orbitals associated with one or the other of the parent atoms. In the sim-
ple Lewis and VSEPR models, these were just the ordinary s, p, and d orbitals. The
more sophisticated hybridization model recognized that these orbitals will be modified
by the interaction with other atoms, and the concept of mixed (hybrid) orbitals was
introduced.
These models in which the bonding electrons are regarded as occupying atomic-type

Page 54 Chem1 Chemical Bonding

orbitals are known generally as valence-bond models. Although they are quite useful
for understanding many of the features of chemical bonding (such as molecular
shapes), valence-bond models are generally unable to predict more detailed character-
istics such as the numerical values of bond energies.
The main difficulty with the valence-bond approach is that the atomic orbitals, whether
hybridized or not, result from the interaction of electrons with a single central force
field-- that of the atomic nucleus. An electron that spends most of its time between two
nuclei will find itself in a very different, two-center force field, and this will give rise to
new types of orbitals that are better characterized as molecular, rather than as atomic
orbitals.

8.1 Molecular orbitals

OnPage 12 it was pointed out that an electron that occupies the region of space
between two nuclei exerts a mutual attraction on the two positive centers, leading to a
net binding effect. Conversely, if the electron is off to one side, in an anti-binding
region, it actually adds to the repulsion between the two nuclei.
The easiest way of visualizing a molecular orbital is to start by picturing two isolated
atoms and the electron orbitals that each would have separately. These are just the
orbitals of the separate atoms, by themselves, which we already understand. We will
then try to predict the manner in which these atomic orbitals interact as we gradually
move the two atoms closer together. Finally, we will reach some point where the inter-
nuclear distance corresponds to that of the molecule we are studying. The correspond-
ing orbitals will then be the molecular orbitals of our new molecule.

8.2 The hydrogen molecule ion

To see how this works, we will consider the simplest possible molecule, H 2+. This is the
hydrogen molecule ion, which consists of two nuclei of charge +1, and a single electron
shared between them.

As two H nuclei move toward each other, the 1s atomic orbitals of the isolated atoms
gradually merge into a new molecular orbital in which the greatest electron density
falls between the two nuclei. Since this is just the location in which electrons can exert
the most attractive force on the two nuclei simultaneously, this arrangement consti-
tutes a bonding molecular orbital. Regarding it as a three- dimensional region of space,
we see that it is symmetrical about the line of centers between the nuclei; in accord
with our usual nomenclature, we refer to this as a (sigma) orbital.

8.3 Bonding and antibonding molecular orbitals

Conserving orbitals
There is one minor difficulty: we started with two orbitals (the 1s atomic orbitals), and
ended up with only one orbital. Now according to the rules of quantum mechanics,

Page 55 The molecular orbital Model

orbitals cannot simply appear and disappear at our convenience. For one thing, this
would raise the question of at just what internuclear distance do we suddenly change
from having two orbitals, to having only one? It turns out that when orbitals interact,
they are free to change their forms, but there must always be the same number. This is
the same as saying that there must always be the same number of possible allowed
sets of electron quantum numbers.

In-phase and out-of-phase wave combinations

How can we find the missing orbital? To answer this question, we must go back to the
wave-like character of orbitals that we developed in our earlier treatment of the hydro-
gen atom. You are probably aware that wave phenomena such as sound waves, light
waves, or even ocean waves can combine or interact with one another in two ways: they
can either reinforce each other, resulting in a stronger wave, or they can interfere with
and partially destroy each other. A roughly similar thing occurs when the “matter
waves” corresponding to the two separate hydrogen 1s orbitals interact; both in-phase
and out-of-phase combinations are possible, and both occur. One of the resultants is
the bonding orbital that we just considered. The other, corresponding to out-of-phase
combination of the two orbitals, gives rise to a molecular orbital that has its greatest
electron probability in what is clearly the antibonding region of space. This second
orbital is therefore called an antibonding orbital.

Fig. 31: Bonding and antibonding molecular orbitals formed by in- and out-
of-phase combinations of atomic orbitals

When the two 1s wave functions combine out-of-phase, the regions of high electron
probability do not merge. In fact, the orbitals act as if they actually repel each other.
Notice particularly that there is a region of space exactly equidistant between the nuclei
at which the probability of finding the electron is zero. This region is called a nodal sur-
face, and is characteristic of antibonding orbitals. It should be clear that any electrons
that find themselves in an antibonding orbital cannot possibly contribute to bond for -
mation; in fact, they will actively oppose it.
We see, then, that whenever two orbitals, originally on separate atoms, begin to interact
as we push the two nuclei toward each other, these two atomic orbitals will gradually
merge into a pair of molecular orbitals, one of which will have bonding character, while
the other will be antibonding. In a more advanced treatment, it would be fairly easy to
show that this result follows quite naturally from the wave-like nature of the combining
orbitals.
What is the difference between these two kinds of orbitals, as far as their potential
energies are concerned? More precisely, which kind of orbital would enable an electron
to be at a lower potential energy? Clearly, the potential energy decreases as the electron
moves into a region that enables it to “see” the maximum amount of positive charge. In

Page 56 The molecular orbital Model

a simple diatomic molecule, this will be in the internuclear region-- where the electron
can be simultaneously close to two nuclei. The bonding orbital will therefore have the
lower potential energy.

8.4 Molecular orbital diagrams

This scheme of bonding and antibonding orbitals is usually depicted by a molecular
orbital diagram such as those shown below.
Electrons fill the lower-energy molecular orbitals before the
higher ones, just as is the case for atomic orbitals. Thus in
H2+, the single electron goes into the bonding orbital, leaving
the antibonding orbital empty. Since any orbital can hold a
maximum of two electrons, the bonding orbital in H2+ is only
half-full. This single electron is nevertheless enough to lower
the potential energy of one mole of hydrogen nuclei pairs by
270 kJ-- quite enough to make them stick together and
behave like a distinct molecular species. Although H2+ is sta-
ble in this energetic sense, it happens to be an extremely reactive molecule-- so much so
that it even reacts with itself, so these ions are not commonly encountered in everyday
chemistry.

Dihydrogen
If one electron in the bonding orbital is conducive to bond for-
mation, might two electrons be even better? We can arrange
this by combining two hydrogen atoms-- two nuclei, and two
electrons. Both electrons will enter the bonding orbital, as
depicted in the Figure. We recall that one electron lowered the
potential energy of the two nuclei by 270 kJ/mole, so we
might expect two electrons to produce twice this much stabili-
zation, or 540 kJ/mole.
Experimentally, one finds that it takes only 452 kJ to break apart a mole of hydrogen
molecules. The reason the potential energy was not lowered by the full amount is that
the presence of two electrons in the same orbital gives rise to a repulsion that acts
against the stabilization. This is exactly the same effect we saw in comparing the ion-
ization energies of the hydrogen and helium atoms.

Page 57 The molecular orbital Model

Dihelium
With two electrons we are still ahead, anyway, so let’s try for
three. The dihelium positive ion is a three-electron molecule.
We can think of it as being from two helium nuclei and three
electrons. This molecule is stable, but not as stable as dihy-
drogen; the energy required to break He2+ is 301 kJ/mole.
The reason for this should be obvious; two electrons were
accommodated in the bonding orbital, but the third electron
must go into the next higher slot-- which turns out to be the
sigma antibonding orbital. The presence of an electron in
this orbital, as we have seen, gives rise to a repulsive com-
ponent which acts against, and partially cancels out, the
attractive effect of the filled bonding orbital.
Taking our building-up process one step further, we can
look at the possibilities of combining to helium atoms to
form dihelium. You should now be able to predict that He2 cannot be a stable molecule;
the reason, of course, is that we now have four electrons-- two in the bonding orbital,
and two in the antibonding orbital. The one orbital almost exactly cancels out the effect
of the other. Experimentally, the bond energy of dihelium is only.084 kJ/mol; this is
not enough to hold the molecule together in the presence of random thermal motion at
ordinary temperatures, so the molecule is unstable.

8.5 Diatomic molecules containing second-row atoms

The four simplest molecules we have examined so far involve molecular orbitals that
derived from two 1s atomic orbitals. If we wish to extend our model to larger atoms, we
will have to contend with higher atomic orbitals as well. One greatly simplifying princi-
ple here is that only the valence-shell orbitals need to be considered. Inner atomic
orbitals such as 1s are deep within the atom and well-shielded from the electric field of
a neighboring nucleus, so that these orbitals largely retain their atomic character when
bonds are formed.

Dilithium
For example, when lithium, whose configuration is
1s22s1,bonds with itself to form Li2, we can forget about the
1s atomic orbitals and consider only the σ bonding and anti-
bonding orbitals shown at the right. Since there are not
enough electrons to populate the antibonding orbital, the
attractive forces win out and we have a stable molecule. The
bond energy of dilithium is 104.6 kJ/mole; notice that this
value is far less than the 270 kJ bond energy in dihydrogen,
which also has two electrons in a bonding orbital.
The reason, of course, is that the 2s orbital of Li is much farther from its nucleus than
is the 1s orbital of H, and this is equally true for the corresponding molecular orbitals.
It is a general rule, then, that the larger the parent atom, the less stable will be the cor-
responding diatomic molecule.

Page 58 Chem1 Chemical Bonding

Lithium hydride
All the molecules we have considered thus far are homonuclear; they are made up of
one kind of atom. As an example of a heteronuclear molecule, let’s take a look at a very
simple example-- lithium hydride. Lithium hydride is a stable, though highly reactive
molecule. The diagram below shows how the molecular orbitals in lithium hydride can
be related to the atomic orbitals of the parent atoms. One thing that makes this dia-
gram look different from the ones we have seen previously is that the parent atomic
orbitals have widely differing energies; the s orbital of lithium has a much lower energy
than the 1s hydrogen orbital, owing to the greater nuclear charge of lithium. On this
diagram, the 2p orbitals of both parent atoms, and the 2s orbital of hydrogen, are
shown. These orbitals are not occupied in the ground states of the parent atoms, so we
need not concern ourselves with them, although in a more thorough treatment their
existence would need to be taken into account.
There are two occupied atomic orbitals on the lithium
atom, and only one on the hydrogen. With which of
the lithium orbitals does the hydrogen 1s orbital
interact? The lithium 1s orbital is the lowest-energy
atomic orbital on the diagram. Because this orbital is
so small and retains its electrons so tightly, we can
forget about it as far as bonding goes; we need con-
sider only 2s orbital of lithium which combines with
the 1s orbital of hydrogen to form the usual pair of
sigma bonding and antibonding orbitals. Of the four
electrons in lithium and hydrogen, two are retained in the lithium 1s orbital, and the
two remaining ones go into the 2s bonding orbital.
The resulting molecule is 243 kJ/mole more stable than the parent atoms. As we
might expect, the bond energy of the heteronuclear molecule is very close to the aver-
age of the energies of the corresponding homonuclear molecules. Actually, it turns
out that the correct way to make this comparison is to take the geometric mean,
rather than the arithmetic mean, of the two bond energies. The geometric mean is
simply the square root of the product of the two energies.
The geometric mean of the H2 and Li2 bond energies is 213 kJ/mole, so it appears
that the lithium hydride molecule is 30 kJ/mole more stable than it “is supposed” to
be. This is attributed to the fact that the electrons in the 2 bonding orbital are not
equally shared between the two nuclei; the orbital is skewed slightly so that the elec-
trons are attracted somewhat more to the hydrogen atom. This bond polarity, which
we considered in some detail near the beginning of our study of covalent bonding,
arises from the greater electron-attracting power of hydrogen-- a consequence of the
very small size of this atom. The electrons can be at a lower potential energy if they
are slightly closer to the hydrogen end of the lithium hydride molecule. It is worth
pointing out, however, that the electrons are, on the average, also closer to the lith-
ium nucleus, compared to where they would be in the 2s orbital of the isolated lith-
ium atom. So it appears that everyone gains and no one loses here!

8.6 Sigma and pi orbitals

The molecules we have considered thus far are composed of atoms that have no more
than four electrons each; our molecular orbitals have therefore been derived from s-
type atomic orbitals only. If we wish to apply our model to molecules involving larger
Page 59 The molecular orbital Model
atoms, we must take a close look at the way in which p-type orbitals interact as well.
Although two atomic p orbitals will be expected to split into bonding and antibonding
orbitals just as before, it turns out that the extent of this splitting, and thus the relative
energies of the resulting molecular orbitals, depend very much on the nature of the
particular p orbital that is involved.

The importance of direction

You will recall that there are three possible p orbitals for any value of the principal
quantum number. You should also recall that p orbitals are not spherical, like s orbit-
als, but are elongated, and thus possess definite directional properties. The three p
orbitals correspond to the three directions of Cartesian space, and are frequently desig-
nated px, py, and pz, to indicate the axis along which the orbital is aligned. Of course,
in the free atom, where no coordinate system is defined, all direction are equivalent,
and so are the p orbitals. But when the atom is near another atom, the electric field
due to that other atom acts as a point of reference that defines a set of directions. The
line of centers between the two nuclei is conventionally taken as the x axis. If this
direction is represented horizontally on a sheet of paper, then the y axis is in the verti-
cal direction and the z axis would be normal to the page.

Fig. 32: Formation of sigma bonding- and antibonding molecular orbitals from end-to-end
atomic p orbitals.

These directional differences lead to the formation of two different classes of molecular
orbitals. The above figure shows how two px atomic orbitals interact. In many ways the
resulting molecular orbitals are similar to what we got when s atomic orbitals com-
bined; the bonding orbital has a large electron density in the region between the two
nuclei, and thus corresponds to the lower potential energy. In the out-of-phase combi-
nation, most of the electron density is away from the internuclear region, and as
before, there is a surface exactly halfway between the nuclei that corresponds to zero
electron density. This is clearly an antibonding orbital-- again, in general shape, very
much like the kind we saw in hydrogen and similar molecules. Like the ones derived
from s-atomic orbitals, these molecular orbitals are orbitals.
Sigma orbitals are cylindrically symmetric with respect to the line of centers of the
nuclei; this means that if you could look down this line of centers, the electron density

Page 60 Chem1 Chemical Bonding

would be the same in all directions.

Fig. 33: Formation of pi bonding and antibonding orbitals from two parallel
atomic-p orbitals.

When we examine the results of the in- and out-of-phase combination of py and pz
orbitals, we get the bonding and antibonding pairs that we would expect, but the
resulting molecular orbitals have a different symmetry: rather than being rotationally
symmetric about the line of centers, these orbitals extend in both perpendicular direc-
tions from this line of centers.Orbitals having this more complicated symmetry are
called π (pi) orbitals. There are two of them, py and pz differing only in orientation, but
otherwise completely equivalent.
The different geometric properties of the π
σ* and orbitals causes the sigma orbitals to
split more than the pi orbitals, so that the
π* antibonding orbitals sigma antibonding orbital always has the
highest energy. The sigma bonding orbital
can be either higher or lower than the pi
bonding orbitals, depending on the partic-
px py pz px py pz ular atom.
atomic orbitals π atomic orbitals
If we combine the splitting schemes for
the 2s and 2p orbitals, we can predict
σ bonding orbitals bond order in all of the diatomic molecules
and ions composed of elements in the first
complete row of the periodic table.
Fig. 34: Splitting scheme for p orbitals
Remember that only the valence orbitals
of the atoms need be considered; as we
saw in the cases of lithium hydride and dilithium, the inner orbitals remain tightly
bound and retain their localized atomic character.

Page 61 The molecular orbital Model

One minor complication that
you should be aware of is
that the relative energies of
the σ and π bonding molecu-
lar orbitals are reversed in
some of the second-row
diatomics. However, the
order in which these two
orbitals are filled has no
effect on the predicted bond
orders, so there is ordinarily
no need to know which mole-
cules follow which scheme.

Dicarbon
Carbon has four outer-shell electrons, two 2s
and two 2p. For two carbon atoms, we therefore
have a total of eight electrons, which can be
accommodated in the first four molecular orbit-
als. The lowest two are the 2s -derived bonding
and antibonding pair, so the “first” four electrons
make no net contribution to bonding. The other
four electrons go into the pair of pi bonding orbit-
als, and there are no more electrons for the anti-
bonding orbitals— so we would expect the
dicarbon molecule to be stable, and it is.
You will recall that one pair of electrons shared
between two atoms constitutes a “single” chemi-
cal bond; this is Lewis’ original definition of the
covalent bond. In C2 there are two paris of elec-
tron in the pi bonding orbitals, so we have what
amounts to a double bond here. Actually, the
preferred nomenclature is “bond order”; the bond order in dicarbon is two.

Page 62 Chem1 Chemical Bonding

Dioxygen
The electron configuration of oxygen is 1s22s22p4. In
O2, therefore, we need to accommodate twelve val-
ance electrons in molecular orbitals. As you can see
from the diagram, this places two electrons in anti-
bonding orbitals. Each of these electrons occupies a
separate π* orbital because this leads to less elec-
tron-electron repulsion. Owing to these two
unpaired electrons, O2 is paramagnetic. The para-
magnetism of oxygen can readily be demonstrated
by pouring liquid O2 past a pole of a strong perma-
nent magnet; the liquid stream will be deflected
toward the magnet.
The potential energy of molecular oxygen is 494 kJ/
mole below that of the parent atoms. This is a lower
bond energy than nitrogen has— not surprising,
considering that oxygen has two electrons in an
antibonding orbital, compared to nitrogen’s one.
Although this ground-state form of O2 (also known as triplet oxygen) is the energetically
favored, and therefor the common form, the other variety, in which the two electrons
are paired up in a single pi antibonding orbital, is also well known. This “singlet” oxy-
gen, as it is called, has a bond energy of only 402 kJ/mole. The lower value reflects the
action of electrostatic repulsion between the two electrons in the same orbital.
Although it does not exist under normal conditions, singlet oxygen can be formed by
the action of light and in certain chemical reactions, and it has an interesting and
unique chemistry of its own.
Since molecular oxygen contains two electrons in an antibonding orbital, it might be
possible to make the molecule more stable by removing one of these electrons, thus
increasing the ratio of bonding to antibonding electrons in the molecule. Just as we
would expect, and in accord with our model, O2+ has a bond energy higher than that of
neutral dioxygen; removing the one electron actually gives us a more stable molecule.
This constitutes a very good test of our model of bonding and antibonding orbitals. In
the same way, adding an electron to O2 results in a weakening of the bond, as evi-
denced by the lower bond energy of O2-minus. The bond energy in this ion is not
known, but the length of the bond is greater, and this is indicative of a lower bond
energy.

molecule H2 He2+ He2 Li Be2 B2 C2 N2 O2 F2

bond order 1 0.5 0 1 0 1 2 3 2 1
bond energy, kJ/mol 458 299 0 99 0 290 569 941 490 154
bond length, pm 74 108 267 131 131 109 121 144

Table 2: Bonding in the second-row diatomic molecules

Page 63 The molecular orbital Model

9. Bonding in coordination complexes
What is a complex?
If you have taken a lab course in chemistry, you have very likely admired the deep blue
color of copper sulfate crystals, CuSO4·5H2O. The proper name of this substance is
copper(II) sulfate pentahydrate, and it is typical of many salts that incorporate waters
of hydration into their crystal structures. It is also a complex, a term used by chemists
to describe a substance composed of two other substances (in this case, CuSO 4 and
H2O) each of which is capable of an independent existence. The binding between the
components of a complex is usually weaker than a regular chemical bond; thus most
solid hydrates can be decomposed by heating, driving off the water and yielding the
anhydrous salt:
CuSO4·5H2O → CuSO4 (s) + 5 H2O
Driving off the water in this way also destroys the color, turning it from a beautiful
deep blue to a nondescript pale yellow. If the anhydrous salt is now dissolved in water,
the blue color now pervades the entire solution. It is apparent that the presence of
water is somehow necessary for the copper(II) ion to take on a blue color, but why
should this be?
A very common lab experiment that most students carry out is to add some dilute
ammonia to a copper sulfate solution. At first, the solutions turns milky as the alkaline
ammonia causes the precipitation of copper hydroxide: Cu2+ + 2 OH– → Cu(OH)2(s).
But if more ammonia is added, the cloudiness disappears and the solution assumes an
intense deep blue color that makes the original solution seem pale by comparison. The
equation for this reaction is usually given as
Cu2+ + 6NH3 → Cu(NH3)62+
The new product is commonly known as the copper-ammonia complex ion, or more offi-
cially, hexamminecopper(II). This equation is somewhat misleading, however, in that it
implies the formation of a new complex where none existed before. In fact, since about
1895 it has been known that the ions of most transition metals dissolve in water to
form complexes with water itself, so a better representation of the reaction of dissolved
copper with ammonia would be
Cu(H2O)62+ + 6 NH3 → Cu(NH3)62+ + 6 H2O
In effect, the ammonia binds more tightly to the copper ion than does water, and it thus
displaces the latter when it comes into contact with the hexaaquocopper(II) ion, as the
dissolved form of Cu2+ is properly known.

9.1 The basics of coordination complexes

Although our primary focus in this unit is on bonding, the topic of coordination com-
plexes is to important in chemistry and biochemistry that some of their basic features
are worth knowing about, even if their detailed chemistry is beyond the scope of this
course. These complexes play an especially crucial role in physiology and biochemistry.
Thus heme, the oxygen-carrying component of red blood cells (and the source of the
red color) is basically a complex of iron, and the part of chlorophyll that converts sun-
light into chemical energy within green plants is a magnesium complex.

Page 64 Chem1 Chemical Bonding

Some definitions
We have already defined a complex as a sub-
stance composed of two or more components
capable of an independent existence. A coordi-
nation complex is one in which a central atom
or ion is joined to one or more ligands (Latin
ligare, to tie) through what is formally a coordi-
nate covalent bond in which both of the bond-
ing electrons are supplied by the ligand. In
such a complex the central atom acts as an
electron-pair acceptor (Lewis base) and the
ligand as an electron-pair donor (Lewis acid).
The central atom and the ligands coordinated
to it constitute the coordination sphere. Thus the salt [Co(NH3)5CuCl2 can be consid-
ered to be composed of the complex ion [Co(NH3)5Cl]2+ and two Cl– ions; components
within the square brackets are inside the coordination sphere, whereas the two chlo-
ride ions are situated outside the coordination sphere. These latter two ions could be
replaced by other ions such as NO3– without otherwise materially changing the nature
of the salt.
The central atoms of coordination complexes are most often positive ions, but may in
some cases be neutral atoms, as in nickel carbonyl Ni(CO) 4.
Ligands composed of ions such as F– or small molecules such as H2O or CN– possess
more than one set of lone pair electrons, but only one of these pairs can coordinate
with a central ion. Such ligands are said to be monodentate (“one tooth”.) Larger
ligands may contain more than atom capable of coordinating with a single central ion,
and are described as polydentate. Thus ethylenediamine (shown below) is a bidentate
ligand. Polydentate ligands whose geometry enables them to occupy more than one
coordinating position of a central ion act as chelating agents (Greek χελοσ chelos, claw)
and tend to form extremely stable complexes known as chelates.

9.2 Structures and bonding in transition metal complexes

Complexes such as Cu(NH3)62+ have been known and studied since the mid-nineteenth
century. Why they should form, or what their structures might be, were complete mys-
teries. At that time all inorganic compounds were thought to be held together by ionic
charges, but ammonia, of course, is electrically neutral. A variety of theories such as

Page 65 Bonding in coordination complexes

the existence of “secondary valences” were concocted, and various chain-like struc-
tures such as CuNH3-NH3-NH3-NH3-NH3-NH3 were proposed. Finally, in the mid-
1890s, after a series of painstaking experiments, the chemist ALFRED WERNER (Swiss,
1866-1919) presented the first workable theory of complex ion structures. (Werner
claimed that his theory first came to him in a flash after a night of fitful sleep; he had
written his landmark paper by the end of the next day. He was awarded the 1913 Nobel
Prize in Chemistry for his work.)
Werner was able to show, in spite of considerable opposition, that transition metal
complexes consist of a central ion surrounded by ligands in a square-planar, tetrahe-
dral, or octahedral arrangement. This an especially impressive accomplishment at a
time long before X-ray diffraction and other methods had become available to observe
structures directly. His basic method was to make inferences of the structures from a
careful examination of the chemistry of these complexes and particularly the existence
of structural isomers. For example, the existence of two different compounds AX4 hav-
ing the same composition shows that its structure must be square-planar rather than
tetrahedral.

What holds them together?

An understanding of the nature of the bond between the central ion and its ligands
would have to await the development of Lewis’ shared-electron pair theory. and Paul-
ing’s valence-bond picture. We have already shown (Page 50) how hybridization of the d
orbitals of the central ion creates vacancies able to accommodate one or more pairs of
unshared electrons on the ligands. Although these models correctly predict the struc-
tures of transition metals, they are by themselves unable to account for several of their
special properties:
• The metal-to-ligand bonds are generally much weaker than ordinary covalent bonds;
• Some complexes utilize “inner” d orbitals of the central ion, while others are “outer-
orbital” complexes;
• Transition metal ions tend to be intensely colored.

Paramagnetism of coordination complexes

Unpaired electrons act as tiny magnets; if a substance that contains unpaired elec-
trons is placed near an external magnet, it will undergo an attraction that tends to
draw it into the field. Such substances are said to be paramagnetic, and the degree of
Page 66 Chem1 Chemical Bonding
paramagnetism is directly proportional to the number of unpaired electrons in the mol-
ecule. Magnetic studies have played an especially prominent role in determining how
electrons are distributed among the various orbitals in transition metal complexes.

Studies of this kind are carried out by placing a sample consisting of a solution of the
complex between the poles of an electromagnet. The sample is suspended from the
arm of a sensitive balance, and the change in weight is measured with the magnet on
and off. An increase in the weight when the magnet is turned on indicates that the
sample is attracted to the magnet (paramagnetism) and must therefore possess one
or more unpaired electrons. The precise number can be determined by calibrating the
system with a substance whose electron configuration is known.

9.3 Ligand field theory

The current model of bonding in coordination complexes developed gradually between
1930-1950, and has largely superseded the hybridization model discussed previously.
It is essentially a simplified adaptation of molecular orbital theory which focuses on the
manner in which the electric field due to the unpaired electrons on the ligands interact
with the five different d orbitals of the central ion.
In an isolated transition metal atom the five outermost d orbitals all have the same
energy which depends solely on the spherically symmetric electric field due to the
nuclear charge and the other electrons of the atom. Suppose now that this atom is
made into a cation and is placed in solution, where it forms a hydrated species in
which six H2O molecules are coordinated to the central ion in an octahedral arrange-
ment. An example of such an ion might be hexaaquotitanium(III), T i(H2O)63+.
The ligands (H2O in this example) are bound to the central ion by electron pairs con-
tributed by each ligand. Because the six ligands are located at the corners of an octa-
hedron centered around the metal ion, these electron pairs are equivalent to clouds of
negative charge that are directed from near the central ion out toward the corners of
the octahedron. We will call this an octahedral electric field, or the ligand field.

d-orbital splitting
Although the five d orbitals of the central atom all have the same energy in a spheri-
cally symmetric field, their energies will not all be the same in the octahedral field
imposed by the presence of the ligands. The reason for this is apparent when we con-
sider the different geometrical properties of the five d orbitals. Two of the d orbitals,
designated dx2 and dx2-y2, have their electron clouds pointing directly toward ligand
atoms. We would expect that any electrons that occupy these orbitals would be subject
to repulsion by the electron pairs that bind the ligands that are situated at correspond-
ing corners of the octahedron. As a consequence, the energies of these two d orbitals
will be raised in relation to the three other d orbitals whose lobes are not directed
toward the octahedral positions.
The number of electrons in the d
orbital of the central atom is easily
determined from the location of the
element in the periodic table, taking
in account, of course, of the number
of electrons removed in order to form
the positive ion.

Page 67 Bonding in coordination complexes

Fig. 35: d orbitals in an
octahedral field
The green circles represent
the coordinating electron-pairs
of the ligands located at the six
corners of the octahedron
around the central atom. The
two d orbitals at the bottom
have regions of high electron
density pointing directly toward
the ligand orbitals; the result-
ing electron-electron repulsion
raises the energy of these d
orbitals. (Diagram from a Per-
due U. chemistry site.]

.
The effect of the octahedral ligand field due to the
ligand electron pairs is to split the d orbitals into
two sets whose energies differ by a quantity denoted
by ∆ which is known as the d orbital splitting. Note
that both sets of central-ion d orbitals are repelled
by the ligands and are both raised in energy; the
upper set is simply raised by a greater amount.
Both the total energy shift and ∆ are strongly depen-
dent on the particular ligands.

Why are transition metal complexes often highly colored?

Returning to our example of Ti(H2O)63+, we note that Ti has an outer configuration of
4s23d2, so that Ti3+ will be a d1 ion. This means that in its ground state, one electron
will occupy the lower group of d orbitals, and the upper group will be empty. The d-
orbital splitting in this case is 240 kJ per mole which corresponds to light of blue-green
color; absorption of this light promotes the electron to the upper set of d orbitals, which
represents the exited state of the complex. If we illuminate a solution of Ti(H2O)63+ with
white light, the blue-green light is absorbed and the solution appears violet in color.

Page 68 Chem1 Chemical Bonding

Fig. 36: Origin of color in a coordination complex
Absorption of light which is energetically equal to ∆ promotes the electron to the upper d-orbital level,
producing an excited state atom. Removal of this color from white life causes Ti3+ solutions to appear
purple in color.

High- and low spin complexes

The magnitude of the d orbital splitting depends strongly on the nature of the ligand
and in particular on how strong an electrostatic field is produced by its electron pair
bond to the central ion. If ∆ is not too large then the electrons that occupy the d orbitals
do so with their spins unpaired until a d5 configuration is reached, just as occurs in
the normal Aufbau sequence for atomic electron configurations. Thus a weak-field
ligand such as H2O leads to a “high spin” complex with Fe(II).

Fig. 37: Effect of ligand type on

magnetic moment of complex

In contrast to this, the cyanide ion acts as a strong-field ligand; the d orbital splitting is
so great that it is energetically more favorable for the electrons to pair up in the lower
group of d orbitals rather than to enter the upper group with unpaired spins. Thus
hexacyanoiron(II) is a “low spin” complex— actually zero spin, in this particular case.
Different d orbital splitting occur in square planar and tetrahedral coordination geome-
tries, so a very large number of arrangements are possible. In most complexes the
value of ∆ corresponds to the absorption of visible light, accounting for the colored
nature of many such compounds in solution and in solids such as CuSO 4·5H2O.

9.4 Coordination complexes in biochemistry

Approximately one-third of the chemical elements are present in living organisms.

Page 69 Bonding in coordination complexes

Many of these are metallic ions whose passage through cell walls and function within
the cell depends on the formation of coordination complexes. The ions Na+, K+, Ca2+
and Mg2+ are particularly important and serve a wide range of functions. These ions,
being in the s-block of the periodic table, bind to suitable polydentate ligands through
ordinary polarization forces rather then through d-orbital complexing. Many of the
micronutrient metals such as Fe, Cu, Mo, Zn and Mn are essential components of
enzymes in which the metal ion is bound through d-orbitals.

Chlorophyll
Chlorophyll is the light-harvesting pigment present in green plants. Its name comes
from the Greek word chloros, meaning “green”— the same root from which chlorine gets
its name. Chlorophyll consists of a ring-shaped tetradentate ligand known as a porphin
coordinated to a central magnesium ion. A histidine residue from one of several types
of associated proteins forms a fifth coordinate bond to the Mg atom.

Fig. 38: Chlorophyll

Left: detail of Mg coordination; notice
that the metal is slightly out of the
plane of the porphin ring. Right: plan
view of Mg within the porphin ligand.
A histidine residue from an associated
protein forms the fifth coordination
point to the Mg atom.

The light energy trapped by chlorophyll is utilized to drive a sequence of reactions

whose net effect is to bring about the reduction of CO2 to glucose (C6H12O6) which
serves as the fuel for all life processes in both plants and animals.

Hemoglobin
Hemoglobin performs the essential task of transporting dioxygen molecules from the
lungs to the tissues in which it is used to oxidize glucose, this oxidation serving as the
source of free energy required for cellular metabolic processes. Like chlorophyll, hemo-
globin consists of a tetradentate ligand, heme, combined with a polypeptide (mini-pro-
tein) chain which together coordinate with an iron atom in five positions. The sixth
position in the octahedrally-coordinated iron is taken up either by an oxygen molecule
or by a water molecule, depending on whether the hemoglobin is in its oxygenated state
(in arteries) or deoxygenated state (in veins.)

Page 70 Chem1 Chemical Bonding

Fig. 39: Hemoglobin
Left: The heme molecules (purple) are enfolded within a peptide chain. There are for of these subunits in
each hemoglobin molecule.
Right: the heme molecule.

Other ligands, notably cyanide ion and carbon monoxide, are able to bind to hemoglo-
bin much more strongly than does iron, thereby displacing it and rendering hemoglo-
bin unable to transport oxygen. Air containing as little as 1 percent CO will convert
hemoglobin to carboxyhemoglobin in a few hours, leading to loss of consciousness and
death. Even small amounts of carbon monoxide can lead to substantial reductions in
the availability of oxygen. The 400-ppm concentration of CO in cigarette smoke will tie
up about 6% of the hemoglobin in heavy smokers; the increased stress this places on
the heart as it works harder to compensate for the oxygen deficit is believed to be one
reason why smokers are at higher risk for heart attacks.

10. Bonding in metals and semiconductors

Most of the known chemical elements are metals, and many of these combine with
each other to form a large number of intermetallic compounds. The special properties
of metals-- their bright, lustrous appearance, their high electrical and thermal conduc-
tivities, and their malleability-- suggest that these substances are bound together in a
very special way.
The fact that the metallic elements are found on the left side of the periodic table gives
us two important preliminary clues to the nature of their bonding.
• These elements all possess low electronegativities, and readily form positive ions Mn+.
They show no tendency to form negative ions, so the kind of bonding present in ionic sol-
ids can immediately be ruled out.

Page 71 Bonding in metals and semiconductors

• The metallic elements have empty or nearly-empty outer p orbitals, so there are never
enough outer-shell electrons to place a complete octet around an atom.
These points lead us to the simplest picture of metals, which regards them as a lattice
Fig. 40: Electron-fluid
model of metallic bonding

of positive ions immersed in a “sea of electrons” which can freely migrate throughout
the solid. In effect the electropositive nature of the metallic atoms allows their valence
electrons to exist as a mobile fluid which can be displaced by an applied electric field,
hence giving rise to their high electrical conductivities. Because each ion is surrounded
by the electron fluid in all directions, the bonding has no directional properties; this
accounts for the high malleability and ductility of metals.
This view is really an oversimplification that fails to explain metals in a quantitative
way, nor can it account for the differences in the properties of individual metals. A
more detailed treatment, known as the bond theory of metals, applies the idea of reso-
nance hybrids to metallic lattices. In the case of an alkali metal, for example, this
would involve a large number of hybrid structures in which a given Na atom shares its
electron with its various neighbors.

10.1 Molecular orbitals in metals

The most useful treatment of metallic solids is based on the molecular orbital
approach. It is best understood by considering first a succession of molecules based on
lithium (or any other alkali metal having a single s electron in its valence shell). The fig-
ure on the next page shows how the m.o. wave functions for Li 2, Li3 and Li4 will look.
These are all constructed by combining the individual atom s functions just as is done
in simple m.o. theory. The only thing new here is that the new molecular orbitals
extend over all the atoms involved, and that the orbitals of intermediate energy have
both bonding and antibonding character in different regions. Every time we add
another atom, we get two new m.o.’s, but since each atom contributes only a single
valence electron, the m.o.’s are never more than half filled. If we extrapolate this to a
giant “molecule” LiN containing a very large number of atoms, we get 2N m.o.’s that are
so closely spaced in energy that they form what is known as a band of allowed energies.
In metallic lithium only the lower half of this band is occupied.

Page 72 Chem1 Chemical Bonding

Take two atoms, add
a third, then a fourth,
and keep on going
until you have about
a mole! Here we
start with two Li
atoms, each of which
has a single valence
electron in its 2s
orbital. (The purple
dots indicate loca-
tions of the Li nuclei.)
In Li2 we get the
usual pair of bonding
and antibonding
MOs, but only the
former are occu-
pied, so this is a sta-
ble molecule. In Li3
and beyond, only the
highest- and lowest-
energy MOs are
“pure” bonding and
antibonding in char-
acter.

In Li4 we begin to
see a trend in which
the lower (mostly
bonding) MO’s are
filled and the upper
(mostly antibonding)
ones are empty.

Now jump to a one-

mole chunk of Li.
The 1023 MO’s now
coalesce into a band
of closely-spaced
levels, with only the
bottom half occu-
pied, implying strong
bonding.

Fig. 41: Recipe for a metal

10.2 Why metals are different

Metallic solids possess special properties that set them apart from other classes of sol-

Page 73 Bonding in metals and semiconductors

ids and make them easy to identify and familiar to everyone. All of these properties
derive from the liberation of the valence electrons from the control of individual atoms,
allowing them to behave as a highly mobile fluid that fills the entire crystal lattice.
What were previously valence-shell orbitals of individual atoms become split into huge
numbers of closely-spaced levels known as bands that extend throughout the crystal.
Melting point and strength The strength of a metal derives from the
electrostatic attraction between the lattice of positive ions and the
fluid of valence electrons in which they are immersed. The larger the
nuclear charge (atomic number) of the atomic kernel and the smaller
its size, the greater this attraction. As with many other periodic prop-
erties, these work in opposite ways, as is seen by comparing the melt-
ing points of some of the Group 1-3 metals (right). Other factors,
particularly the lattice geometry are also important, so exceptions Fig. 42:
such as is seen in Mg are not surprising. In general, the transition Melting points
metals with their valence-level d electrons are stronger and have of some metals
(temperatures
higher melting points: Fe, 1535°C; Os 3700°C; W 3370°C (this is tung-
are in °C.)
sten, the highest-melting of all the metals; do you know what principal
use of tungsten derives from this very high melting point?)
Malleability and ductility These terms refer respectively to how readily a solid can be
shaped by pressure (forging, hammering, rolling into a sheet) and by being drawn out
into a wire. Metallic solids are known and valued for these qualities, which derive from
the non-directional nature of the attractions between the kernel atoms and the electron
fluid. The bonding within ionic or covalent solids may be stronger, but it is also direc-
tional, making these solids subject to fracture (brittle) when struck with a hammer, for
example. A metal, by contrast, is more likely to be simply deformed or dented.
Electrical conductivity In order for a substance to conduct electricity, it must contain
charged particles (charge carriers) that are sufficiently mobile to move in response to an
applied electric field. In the case of ionic solutions and melts, the ions themselves serve
this function. (Ionic solids contain the same charge carriers, but because they are fixed
in place, these solids are insulators.) In metals the charge carriers are the electrons,
and because they move freely through the lattice, metals are highly conductive. The
very low mass and inertia of the electrons allows them to conduct high-frequency alter-
nating currents, something that electrolytic solutions are incapable of. In terms of the
band structure, application of an external field simply raises some of the electrons to
previously unoccupied levels which possess greater momentum.
The conductivity of an electrolytic solution decreases as the temperature falls due to
the decrease in viscosity which inhibits ionic mobility. The mobility of the electron
fluid in metals is practically unaffected by temperature, but metals do suffer a slight
conductivity decrease (opposite to ionic solutions) as the temperature rises; this hap-
pens because the more vigorous thermal motions of the kernel ions disrupts the uni-
form lattice structure that is required for free motion of the electrons within the
crystal. Silver is the most conductive metal, followed by copper, gold, and aluminum.

Thermal conductivity Everyone knows that touching a metallic surface at room temper-
ature produces a colder sensation than touching a piece of wood or plastic at the same
temperature. The very high thermal conductivity of metals allows them to draw heat
out of our bodies very efficiently if they are below body temperature. In the same way, a
metallic surface that is above body temperature will feel much warmer than one made
of some other material. The high thermal conductivity of metals is attributed to vibra-

Page 74 Chem1 Chemical Bonding

tional excitations of the fluid-like electrons; this excitation spreads through the crystal
far more rapidly than it does in non-metallic solids which depend on vibrational
motions of atoms which are much heavier and possess greater inertia.
Appearance We usually recognize a metal by its “metallic lustre”, which refers to its
ability of reflect light. When light falls on a metal, its rapidly changing electromagnetic
field induces similar motions in the more loosely-bound electrons near the surface. A
vibrating charge is itself an emitter of electromagnetic radiation, so the effect is to
cause the metal to re-emit, or reflect, the incident light, producing the shiny appear-
ance. What color is a metal? With the two exceptions of copper and gold, the closely-
spaced levels in the bands allow metals to absorb all wavelengths equally well, so most
metals are basically black, but this is ordinarily evident only when the metallic parti-
cles are so small that the band structure is not established.
Thermionic effect The electrons within the electron fluid have a distribution of velocities
very much like that of molecules in a gas. When a metal is heated suf ficiently, a frac-
tion of these electrons will acquire sufficient kinetic energy to escape the metal alto-
gether; some of the electrons are essentially “boiled out” of the metal. This thermionic
effect, which was first observed by Thomas Edison, was utilized in vacuum tubes which
served as the basis of electronics from its beginning around 1910 until semiconductors
became dominant in the 1960’s.

10.3 Band structure in metals

Most metals are made of atoms that have an outer configuration of s2, which we would
expect to completely fill the band of m.o.’s we have described. With the band completely
filled and no empty levels above, we would not expect elements such as beryllium to be
metallic. What happens is that the empty p orbitals also split into a band. Although the
energy of the 2p orbital of an isolated Be atom is about 160 kJ greater than that of the
2s orbital, the bottom part of the 2p band overlaps the upper part of the 2s band, yield-
ing a continuous conduction band that has plenty of unoccupied orbitals. It is only
when these bands become filled with 2p electrons that the elements lose their metallic
character.

Page 75 Bonding in metals and semiconductors

Fig. 43: Band structure in
3rd-row metals
This diagram illustrates the
band structure in a 3rd-row
metal such as Na or Mg, and
how it arises from MO splitting
in very small aggregates M2 -
M6. The conduction bands for
the “infinite” molecule MN are
shaded.

In most metals there will be bands derived from the outermost s-, p-, and d atomic lev-
els, leading to a system of bands, some of which will overlap as described above. Where
overlap does not occur, the almost continuous energy levels of the bands are separated
by a forbidden zone, or band gap. Only the outermost atomic orbitals form bands; the
inner orbitals remain localized on the individual atoms and are not involved in bond-
ing.

10.4 Metals, insulators and semiconductors

The band theory of solids provides a clear set of criteria for distinguishing between con-
ductors (metals), insulators and semiconductors. As we have seen, a conductor must
posses an upper range of allowed levels that are only partially filled with valence elec-
trons. These levels can be within a single band, or they can be the combination of two
overlapping bands. A band structure of this type is known as a conduction band.

Fig. 44: Possible band organizations in conducting solids

An insulator is characterized by a large band gap between the highest filled band and
an even higher empty band. The band gap is sufficiently great to prevent any signifi-
cant population of the upper band by thermal excitation of electrons from the lower
one. The presence of a very intense electric field may be able to supply the required
energy, in which case the insulator undergoes dielectric breakdown. Most molecular
crystals are insulators, as are covalent crystals such as diamond.
If the band gap is sufficiently small to allow electrons in the filled band below it to jump

Page 76 Chem1 Chemical Bonding

Fig. 45: Band organizations in insulators and intrinsic semiconductors

into the upper empty band by thermal excitation, the solid is known as a semiconduc-
tor. In contrast to metals, whose electrical conductivity decreases with temperature
(the more intense lattice vibrations interfere with the transfer of momentum by the
electron fluid), the conductivity of semiconductors increases with temperature. In
many cases the excitation energy can be provided by absorption of light, so most semi-
conductors are also photoconductors. Examples of semiconducting elements are Se,
Te, Bi, Ge, Si, and graphite.

Fig. 46: Doped semiconductors

The presence of an impurity in a semiconductor can introduce a new band into the sys-
tem. If this new band is situated within the forbidden region, it creates a new and
smaller band gap that will increase the conductivity. The huge semiconductor industry
is based on the ability to tailor the band gap to fit the desired application by introduc-
ing an appropriate impurity atom (dopant) into the semiconductor lattice. The dopant
elements are normally atoms whose valance shells contain one electron more or less
than the atoms of the host crystal. For example, a phosphorus atom introduced as an
impurity into a silicon lattice possesses one more valence electron than Si. This elec-
tron is delocalized within the impurity band and serves as the charge carrier in what is
known as an N-type semiconductor. In a semiconductor of the P-type, the dopant might
be arsenic, which has only three valence electrons. This creates what amounts to an
electron deficiency, or hole in the electron fabric of the crystal, although the solid
remains electrically neutral overall. As this vacancy is filled by the electrons from sili-
con atoms the vacancy hops to another location, so the charge carrier is effectively a
positively charged hole, hence the P-type designation.

Page 77 Bonding in metals and semiconductors