Professional Documents
Culture Documents
Student Solutions Manual To Accompany Atkins' Physical Chemistry
Student Solutions Manual To Accompany Atkins' Physical Chemistry
Student Solutions Manual To Accompany Atkins' Physical Chemistry
Manual to
Accompany Atkins’
Physical Chemistry
INTERNATIONAL EDITION
Peter Bolgar
Haydn Lloyd
Aimee North
Vladimiras Oleinikovas
Stephanie Smith
and
James Keeler
Department of Chemistry
University of Cambridge
UK
Table of contents
Preface vii
2 Internal energy 39
2A Internal energy 39
2B Enthalpy 45
2C Thermochemistry 48
2D State functions and exact differentials 57
2E Adiabatic changes 63
18 Solids 663
18A Crystal structure 663
18B Diffraction techniques 666
18C Bonding in solids 673
18D The mechanical properties of solids 678
18E The electrical properties of solids 679
18F The magnetic properties of solids 681
18G The optical properties of solids 684
�is manual provides detailed solutions to the (a) Exercises and the odd-numbered Discus-
sion questions and Problems from the international edition of Atkins’ Physical Chemistry.
We have included page-speci�c references to equations, sections, �gures and other features
of the main text. Equation references are denoted [��B.�b–���], meaning eqn ��B.�b located
on page ��� (the page number is given in italics). Other features are referred to by name,
with a page number also given.
Generally speaking, the values of physical constants (from the �rst page of the main text)
are used to � signi�cant �gures except in a few cases where higher precision is required.
In line with the practice in the main text, intermediate results are simply truncated (not
rounded) to three �gures, with such truncation indicated by an ellipsis, as in 0.123...; the
value is used in subsequent calculations to its full precision.
�e �nal results of calculations, generally to be found in a box , are given to the precision
warranted by the data provided. We have been rigorous in including units for all quantities
so that the units of the �nal result can be tracked carefully. �e relationships given on
the back of the front cover are useful in resolving the units of more complex expressions,
especially where electrical quantities are involved.
Some of the problems either require the use of mathematical so�ware or are much easier
with the aid of such a tool. In such cases we have used Mathematica (Wolfram Research,
Inc.) in preparing these solutions, but there are no doubt other options available. Some of
the Discussion questions relate directly to speci�c section of the main text in which case we
have simply given a reference rather than repeating the material from the text.
Acknowledgements
In preparing this manual we have drawn on the equivalent volume prepared for the ��th edi-
tion of Atkins’ Physical Chemistry by Charles Trapp, Marshall Cady, and Carmen Giunta. In
particular, the solutions which use quantum chemical calculations or molecular modelling
so�ware, and some of the solutions to the Discussion questions, have been quoted directly
from the solutions manual for the ��th edition, without signi�cant modi�cation. More
generally, we have bene�ted from the ability to refer to the earlier volume and acknowledge,
with thanks, the in�uence that its authors have had on the present work.
�is manual has been prepared by the authors using the LATEX typesetting system, in
the implementation provided by MiKTEX (miktex.org); the vast majority of the �gures
and graphs have been generated using PGFPlots. We are grateful to the community who
maintain and develop these outstanding resources.
Finally, we are grateful to the editorial team at OUP, Jonathan Crowe and Roseanna
Levermore, for their invaluable support in bringing this project to a conclusion.
viii PREFACE
Solutions to exercises
E�A.�(a) Consider � m3 of air: the mass of gas is therefore �.��� kg. If perfect gas be-
haviour is assumed, the amount in moles is given by n = pV �RT
m = n O2 × M O2 + n N2 × M N2
�ese two equations are solved simultaneously for n O2 to give the follow-
ing expression, which is then evaluated using the data given
)
m − M N2 n
n O2 =
M O2 − M N2
(1146 g) − (28.02 g mol−1 ) × (39.5... mol)
= = 9.50... mol
(32.00 g mol−1 ) − (28.02 g mol−1 )
a)
1.0
(1�ρ)�(g−1 dm3 )
0.5
0.0
−300 −200 −100 0 100
θ�○ C
Figure 1.1
n H2 2.0 mol
x H2 = = = 2
x N2 = 1 − x H2 = 1
n H2 + n N2 2.0 mol + 1.0 mol 3 3
Expressed in atmospheres these are �.� atm and �.� atm, respectively.
(iii) �e total pressure is
E�A.�(a) From the inside the front cover the conversion between pressure units is: � atm
≡ 101.325 kPa ≡ ��� Torr; � bar is 105 Pa exactly.
(i) A pressure of ��� kPa is converted to Torr as follows
1 atm 760 Torr
108 kPa × × = ��� Torr
101.325 kPa 1 atm
(ii) A pressure of �.��� bar is 0.975 × 105 Pa, which is converted to atm as
follows
1 atm
0.975 × 105 Pa × = �.��� atm
101.325 kPa
E�A.�(a) �e perfect gas law [�A.�–�], pV = nRT, is rearranged to give the pressure,
p = nRT�V . �e amount n is found by dividing the mass by the molar mass of
Xe, ���.�� g mol−1 .
��� � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � ��
n
So no , the sample would not exert a pressure of �� atm, but ��.� atm if it were
a perfect gas.
p f Vf = p i Vi hence p i = p f Vf �Vi
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 5
�e initial volume is �.�� dm3 greater than the �nal volume so Vi = 4.65+2.20 =
6.85 dm3 .
Vf 4.65 dm3
pi = × pf = × (5.04 bar) = 3.42 bar
Vi 6.85 dm3
(i) �e initial pressure is �.�� bar
(ii) Because a pressure of � atm is equivalent to �.����� bar, the initial pressure
expressed in atm is
1 atm
× 3.40 bar = �.�� atm
1.01325 bar
E�A.�(a) If the gas is assumed to be perfect, the equation of state is [�A.�–�], pV = nRT.
In this case the volume and amount (in moles) of the gas are constant, so it
follows that the pressure is proportional to the temperature: p ∝ T. �e ratio
of the �nal and initial pressures is therefore equal to the ratio of the temper-
atures: p f �p i = Tf �Ti . �e pressure indicated on the gauge is that in excess
of atmospheric pressure, thus the initial pressure is 24 + 14.7 = 38.7 lb in−2 .
Solving for the �nal pressure p f (remember to use absolute temperatures) gives
Tf
pf = × pi
Ti
(35 + 273.15) K
= × (38.7 lb in−2 ) = 44.4... lb in−2
(−5 + 273.15) K
�e pressure indicated on the gauge is this �nal pressure, minus atmospheric
pressure: 44.4... − 14.7 = 30 lb in−2 . �is assumes that (i) the gas is behaving
perfectly and (ii) that the tyre is rigid.
where 1 dm3 = 10−3 m3 has been used along with 1 J = 1 kg m2 s−2 and 1 Pa =
1 kg m−1 s−2 .
6 1 THE PROPERTIES OF GASES
E�A.��(a) �e vapour is assumed to be a perfect gas, so the gas law pV = nRT applies. �e
task is to use this expression to relate the measured mass density to the molar
mass.
First, the amount n is expressed as the mass m divided by the molar mass M to
give pV = (m�M)RT; division of both sides by V gives p = (m�V )(RT�M).
�e quantity (m�V ) is the mass density ρ, so p = ρRT�M, which rearranges
to M = ρRT�p; this is the required relationship between M and the density.
where 1 J = 1 kg m2 s−2 and 1 Pa = 1 kg m−1 s−2 have been used. �e molar mass
of S is 32.06 g mol−1 , so the number of S atoms in the molecules comprising
the vapour is (0.255... × 103 g mol−1 )�(32.06 g mol−1 ) = 7.98. �e result is
expected to be an integer, so the formula is likely to be S8 .
E�A.��(a) �e vapour is assumed to be a perfect gas, so the gas law pV = nRT applies; the
task is to use this expression to relate the measured data to the mass m. �is
is done by expressing the amount n as m�M, where M is the the molar mass.
With this substitution it follows that m = MPV �RT.
�e partial pressure of water vapour is 0.60 times the saturated vapour pressure
M pV
m=
RT
(18.0158 g mol−1 ) × (0.60 × 0.0356 × 105 Pa) × (400 m3 )
=
(8.3145 J K−1 mol−1 ) × ([27 + 273.15] K)
= 6.2 × 103 g = �.� kg
Solutions to problems
P�A.� (a) �e expression ρgh gives the pressure in Pa if all the quantities are in
SI units, so it is helpful to work in Pa throughout. From the front cover,
��� Torr is exactly � atm, which is 1.01325×105 Pa. �e density of ��.�� g cm−3
is equivalent to 13.55 × 103 kg m−3 .
p = p ex + ρgh
= 1.01325 × 105 Pa + (13.55 × 103 kg m−3 ) × (9.806 m s−2 )
× (10.0 × 10−2 m) = 1.15 × 105 Pa
P�A.� �e perfect gas law pV = nRT implies that pVm = RT, where Vm is the molar
volume (the volume when n = 1). It follows that p = RT�Vm , so a plot of p
against T�Vm should be a straight line with slope R.
However, real gases only become ideal in the limit of zero pressure, so what is
needed is a method of extrapolating the data to zero pressure. One approach is
to rearrange the perfect gas law into the form pVm �T = R and then to realise
that this implies that for a real gas the quantity pVm �T will tend to R in the limit
of zero pressure. �erefore, the intercept at p = 0 of a plot of pVm �T against p
is an estimate of R. For the extrapolation of the line back to p = 0 to be reliable,
the data points must fall on a reasonable straight line. �e plot is shown in
Fig �.�.
0.08206
0.08204
0.08202
0.08200
�e estimate for R is therefore the intercept, 0.082062 atm dm3 mol−1 K−1 .
�e data are given to � �gures, but they do not fall on a very good straight line
so the value for R has been quoted to one fewer signi�cant �gure.
8 1 THE PROPERTIES OF GASES
P�A.� For a perfect gas pV = nRT which can be rearranged to give p = nRT�V . �e
amount in moles is n = m�M, where M is the molar mass and m is the mass of
the gas. �erefore p = (m�M)(RT�V ). �e quantity m�V is the mass density
ρ, and hence
p = ρRT�M
It follows that for a perfect gas p�ρ should be a constant at a given temperature.
Real gases are expected to approach this as the pressure goes to zero, so a
suitable plot is of p�ρ against p; the intercept when p = 0 gives the best estimate
of RT�M. �e plot is shown in Fig. �.�.
58
(p�ρ)�(kPa kg−1 m3 )
56
54
0 20 40 60 80 100
p�kPa
Figure 1.3
P�A.� (a) For a perfect gas pV = nRT so it follows that for a sample at constant
volume and temperature, p 1 �T1 = p 2 �T2 . If the pressure increases by
∆p for an increase in temperature of ∆T, then with p 2 = p 1 + ∆p and
T2 = T1 + ∆T is follows that
p 1 p 1 + ∆p p 1 ∆T
= hence ∆p =
T1 T1 + ∆T T1
For an increase by �.�� K, ∆T = 1.00 K and hence
∆p p 1 6.69 × 103 Pa
= = = 24.5... Pa K−1
∆T T1 273.16 K
∆p 6.69 × 103 Pa
∆p′ = ∆T ′ × � � = (99.99 K) × = 2.44... × 103 Pa
∆T 273.16 K
P�A.� �e molar mass of SO� is 32.06+2×16.00 = 64.06 g mol−1 . If the gas is assumed
to be perfect the volume is calculated from pV = nRT
��� � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � ��
n
Note the conversion of the mass in t to mass in g; repeating the calculation for
��� t gives a volume of 4.1 × 105 m3 .
�e volume of gas is therefore between �.�� km3 and �.�� km3 .
P�A.�� Imagine a column of the atmosphere with cross sectional area A. �e pressure
at any height is equal to the force acting down on that area; this force arises
from the gravitational attraction on the gas in the column above this height –
that is, the ‘weight’ of the gas.
Suppose that the height h is increased by dh. �e force on the area A is reduced
because less of the atmosphere is now bearing down on this area. Speci�cally,
the force is reduced by that due to the gravitational attraction on the gas con-
tained in a cylinder of cross-sectional area A and height dh. If the density of
10 1 THE PROPERTIES OF GASES
the gas is ρ, the mass of the gas in the cylinder is ρ × A dh and the force due to
gravity on this mass is ρgA dh, where g is the acceleration due to free fall. �e
change in pressure dp on increasing the height by dh is this force divided by
the area, so it follows that
dp = −ρgdh
�e minus sign is needed because the pressure decreases as the height increases.
�e density is related to the pressure by starting from the perfect gas equation,
pV = nRT. If the mass of gas is m and the molar mass is M, it follows that
n = m�M and hence pV = (m�M)RT. Taking the volume to the right gives
p = (m�MV )RT. �e quantity m�V is the mass density ρ, so p = (ρ�M)RT;
this is rearranged to give an expression for the density: ρ = M p�RT.
�is expression for ρ is substituted into dp = −ρgdh to give dp = −(M p�RT)gdh.
Division by p results in separation of the variables (1�p) dp = −(M�RT)gdh.
�e le�-hand side is integrated between p 0 , the pressure at h = 0 and p, the
pressure at h. �e right-hand side is integrated between h = 0 and h
p 1 h Mg
� dp = � − dh
p0 p 0 RT
Mg
[ln p] p 0 = − [h]0
p h
RT
p M gh
ln =−
p0 RT
RT
p = p 0 e−h�H with H=
Mg
(a) �e pressure decrease across such a small distance will be very small be-
cause h�H � 1. It is therefore admissible to expand the exponential and
retain just the �rst two terms: ex ≈ 1 + x
p = p 0 (1 − h�H)
p − p 0 = −p 0 h�H
p tot (X × 10−12 × V )
n trace =
RT
Taking the volume V to the le� gives the molar concentration, c trace
�e quantity in the square root is related to the acceleration due to free fall,
g, in the following way. A mass m at the surface of the Earth experiences
a gravitational force given GMm�R 2 (note that the force goes as R−2 ). �is
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 13
force accelerates the mass towards the Earth, and can be written mg. �e two
expressions for the force are equated to give
GMm GM
= mg hence = gR
R2 R
�is expression for GM�R is substituted into the above expression for υ e to give
�
2GM �
υe = = 2Rg
R
�e escape velocity is therefore a function of the radius of the Earth and the
acceleration due to free fall.
�e radius of the Earth is 6.37×106 m and g = 9.81 m s−2 so the escape velocity
is 1.11×104 m s−1 . For comparison, the mean speed of He at ��� K is ���� m s−1
and for N� the mean speed is ��� m s−1 . For He, only atoms with a speed in
excess of eight times the mean speed will be able to escape, whereas for N� the
speed will need to be more than twenty times the mean speed. �e fraction of
molecules with speeds many times the mean speed is small, and because this
fraction goes as e−υ it falls o� rapidly as the multiple increases. A tiny fraction
2
of He atoms will be able to escape, but the fraction of heavier molecules with
su�cient speed to escape will be utterly negligible.
Solutions to exercises
E�B.�(a) �e most probable speed is given by [�B.��–��], υ mp = (2RT�M)1�2 , the mean
speed is given by [�B.�–��], υ mean = (8RT�πM)1�2 , and the mean relative
√
speed between two molecules of the same mass is given by [�B.��a–��], υ rel =
2υ mean .
M CO2 = 12.01 + 2 × 16.00 = 44.01 g mol−1 .
−1
2 × (8.314501 / ) × (293.15/
J K−1 mol
1�2
2RT 1�2 K)
υ mp =� � =� −1 � = 333 m s−1
M 44.01 × 10 kg
−3
Omol
/
J1kg
MISFkg.ms#kjmts
=
14 1 THE PROPERTIES OF GASES
√ √
υ rel = 2υ mean = 2 × (376 m s−1 ) = 531 m s−1
where 1 J = 1 kg m2 s−2 and 1 Pa = 1 kg m−1 s−2 have been used. Note the
conversion of the collision cross-section σ to m2 : � nm2 = (1 × 10−9 )2 m2 =
1 × 10−18 m2 .
6 × 100 cm3
1�3
6V 1�3
d=� � =� � = 5.75... cm
π π
�e mean free path is given by [�B.��–��], λ = kT�σ p. �is is rearranged to
give the pressure p with λ equal to the diameter of the vessel
kT (1.3806 × 10−23 J K−1 ) × (298.15 K)
p= = = �.�� Pa
σ d (0.36 × 10−18 m2 ) × (5.75... × 10−2 m)
Note the conversion of the diameter from cm to m.
E�B.�(a) �e mean free path is given by [�B.��–��], λ = kT�σ p.
kT (1.3806 × 10−23 J K−1 ) × (217 K)
λ= =
σ p (0.43 × 10−18 m2 ) × (0.05 × 1.01325 × 105 Pa)
= 1.4 × 10−6 m = �.� µm
200.59 g mol−1
1�2 1�2
υ mean,H2 M Hg
=� � =� � = �.���
υ mean,Hg M H2 2 × 1.0079 g mol−1
where 1 J = 1 kg m2 s−2 has been used. Note that the molar mass is in kg mol−1 .
where 1 J = 1 kg m2 s−2 has been used. �us, �.���% of molecules have veloci-
ties in this range.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 17
Solutions to problems
P�B.� A rotating slotted-disc apparatus consists of a series of disks all mounted on a
common axle (sha�). Each disc has a narrow radial slot cut into it, and the slots
on successive discs are displaced from one another by a certain angle. �e discs
are then spun at a constant angular speed.
Detector
Source
Selector
Imagine a molecule moving along the direction of the axle with a certain veloc-
ity such that it passes through the slot in the �rst disc. By the time the molecule
reaches the second disc the slot in that disc will have moved around, and the
molecule will only pass through the slot if the speed of the molecule is such
that it arrives at the second disc at just the time at which the slot appears in
the path of the molecule. In this way, only molecules with a speci�c velocity
(or, because the slot has a �nite width, a small range of velocities) will pass
through the second slpt. �e velocity of the molecules which will pass through
the second disc is set by the angular speed at which the discs are rotated and
the angular displacement of the slots on successive discs.
�e angular velocity of the discs is 2πv rad s−1 so in time t the discs move
through an angle θ = 2πvt. If the spacing of the discs is d, a molecule with
velocity υ x will take time t = d�υ x to pass from one disc to the next. If the
second slit is set at an angle α relative to the �rst, such a molecule will only pass
through the second slit if
d 2πvd
2πv � �=α hence υx =
υx α
18 1 THE PROPERTIES OF GASES
m 1�2 mυ 2x
ln I = ln[Af (υ x )] = ln A + ln � � −
2πkT 2kT
A plot of ln I against υ 2x is expected to be a straight line with slope −m�2kT;
such a plot is shown in Fig. �.�.
At both temperatures the data fall on reasonable straight lines, with slope −1.33
at �� K and −0.516 at ��� K.
If the Maxwell–Boltzmann distribution applies the expected slope at �� K is
computed as
Again, the expected slope −0.504 compares reasonably well with that found
experimentally.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 19
0.0
T = 40 K
T = 100 K
−2.0
ln I
−4.0
−6.0
0 1 2 3 4 5
υ 2x �(104 m s )
2 −2
Figure 1.4
0 2a
With a = m�2kT the mean speed is
m 1�2 1 kT 1�2
υmean = �υ x � = � � � �=� �
kT 2(m�2kT) 2πm
A�er the beam emerges from the velocity selector, f (υ x ) is zero for υ x > υmean .
�e probability distribution is therefore changed and so needs to be re-normalized
such that υ mean
Kx � e−mυ x �2k T dυ x = 1
2
0
�is integral is best evaluated using mathematical so�ware which gives
υ mean πkT 1�2 1
� e−mυ x �2k T dυ = � � erf( √ )
2
0 2m 2 π
where erf(x) is the error function. �e normalized distribution is therefore
2m 1�2 1
f new (υ x ) = � � e−mυ x �2k T
2
�e new mean speed is computed using this distribution; again this intergral is
best evaluated using mathematical so�ware. Note that the integral extends up
to υmean
2m 1�2 1 υ mean
υmean, new = � � � υ x e−mυ x �2k T dυ x
2
��� � � � � � � � � ��� � � � � � � � � � �
υ mean
1�2
2kT 1 kT 1�2 1
= (1 − e1�4π ) � � = (1 − e 1�4π
)2 � �
πm erf( 2√1 π ) 2πm erf( 2√1 π )
1
= (1 − e1�4π )2υmean
erf( 2√1 π )
= = n 2 e−Mυmp (n −1)�2RT
2 2
In taking the ratio, with the exception of the term υ 2 , all of the terms in f (υ)
which multiply the exponential cancel. In this expression the term υmp is re-
placed by (2RT�M)1�2 to give
f (n × υmp )∆υ
= n 2 e−M υmp (n −1)�2RT = n 2 e−M(2RT�M)(n −1)�2RT = n 2 e(1−n )
2 2 2 2
f (υmp )∆υ
P�B.� �e key idea here is that for an object to escape the gravitational �eld of the
Earth it must acquire kinetic energy equal in magnitude to the gravitational
potential energy the object experiences at the surface of the Earth. �e grav-
itational potential energy between two objects with masses m 1 and m 2 when
separated by a distance r is
Gm 1 m 2
V =−
r
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 21
�e quantity in the square root is related to the acceleration due to free fall,
g, in the following way. A mass m at the surface of the Earth experiences
a gravitational force given GMm�R 2 (note that the force goes as R −2 ). �is
force accelerates the mass towards the Earth, and can be written mg. �e two
expressions for the force are equated to give
GMm GM
= mg hence = gR (�.�)
R2 R
�is expression for GM�R is substituted into the above expression for υ e to give
2GM 1�2
υe = � � = (2Rg)1�2
R
�e escape velocity is therefore a function of the radius of the Earth and the
acceleration due to free fall.
�e quoted values for the Earth give
� �
υ e = 2Rg = 2 × (6.37 × 106 m) × (9.81 m s−2 ) = 1.12 × 104 m s−1
For Mars, data is not given on the acceleration due to free fall. However, it
follows from eqn �.� that g = GM�R 2 , and hence
MMars REarth 2
gMars = gEarth � �
MEarth RMars
6.37 × 106 m
2
= (9.81 m s−2 ) × (0.108) × � � = 3.76... m s−2
3.38 × 106 m
22 1 THE PROPERTIES OF GASES
�ese results indicate that the lighter molecules have the greater chance of
escaping (because they are moving faster on average) and that increasing the
temperature increases the probability of escaping (again becuase this increases
the mean speed). Escape from Mars is easier than from the Earth because of
the lower escape velocity, and heavier molecules are seemingly very unlikely to
escape from the Earth.
At ��� K and with M = 2 × 16.00 g mol−1 the fraction is �.���� and at ���� K
the fraction is �.���� .
P�B.�� Two hard spheres will collide if their line of centres approach within 2r of one
another, where r is the radius of the sphere. �is distance de�nes the collision
diameter, d = 2r, and the collision cross-section is the area of a circle with this
radius, σ = πd 2 = π(2r)2 . �e pressure is computed from the other parameters
using the perfect gas law: p = nRT�V .
�e collision frequency is given by [�B.��b–��], z = σ υ rel p�kT, with the
√ relative
speed for two molecules of the same type given by [�B.��a–��], υ rel = 2υ mean .
�e mean speed is given by [�B.�–��], υ mean = (8RT�πM)1�2 .
Putting this all together gives
1C Real gases
Answer to discussion questions
D�C.� �e critical constants represent the state of a system at which the distinction
between the liquid and vapour phases disappears. �is situation is usually
described by saying that above the critical temperature the liquid phase cannot
be produced by the application of pressure alone. �e liquid and vapour phases
can no longer coexist, though supercritical �uids have both liquid and vapour
characteristics.
D�C.� �e van der Waals equation is a cubic equation in the volume V . Every cubic
equation has some values of the coe�cients for which the number of real roots
passes from three to one. In fact, any equation of state of odd degree n > 1 can
in principle account for critical behavior because for equations of odd degree
24 1 THE PROPERTIES OF GASES
in V there are necessarily some values of temperature and pressure for which
the number of real roots of V passes from n to �. �at is, the multiple values of
V converge from n to � as the temperature approaches the critical temperature.
�is mathematical result is consistent with passing from a two phase region
(more than one volume for a given T and p) to a one phase region (only one V
for a given T and p), and this corresponds to the observed experimental result
as the critical point is reached.
Solutions to exercises
E�C.�(a) �e relation between the critical constants and the van der Waals parameters
is given by [�C.�–��]
a 8a
Vc = 3b pc = Tc =
27b 2 27Rb
All three critical constants are given, so the problem is over-determined: any
pair of the these expressions is su�cient to �nd values of a and b. It is con-
venient to use R = 8.2057 × 10−2 dm3 atm K−1 mol−1 and volumes in units of
dm3 .
If the expressions for Vc and p c are used, a and b are found in the following
way
Vc = 3b hence b = Vc �3 = (0.0987 dm3 mol−1 )�3 = 0.0329 dm3 mol−1
a a
pc = = hence a = 27(Vc �3)2 p c
27b 2 27(Vc �3)2
a = 27(Vc �3)2 p c = 27([0.0987 dm3 mol−1 ]�3)2 × (45.6 atm)
= 1.33 atm dm6 mol−2
�ere are three possible ways of choosing two of the expressions with which to
�nd a and b, and each choice gives a di�erent value. For a the values are �.��,
�.��, and �.��, giving an average of �.�� atm dm6 mol−2 . For b the values are
�.����, �.����, and �.����, giving an average of �.���� dm3 mol−1 .
In Section �C.�(a) on page �� it is argued that b = 4Vmolec N A , where Vmolec is
the volume occupied by one molecule. �is volume is written in terms of the
radius r as 4πr 3 �3 so it follows that r = (3b�16πN A )1�3 .
E�C.�(a) (i) In Section �C.�(b) on page �� it is explained that at the Boyle temperature
Z = 1 and dZ�dp = 0; this latter condition corresponds to the second
virial coe�cient, B or B′ , being zero. �e task is to �nd the relationship
between the van der Waals parameters and the virial coe�cients, and the
starting point for this is the expressions for the product pVm is each case
([�C.�b–��] and [�C.�b–��])
RT a RT Vm a
van der Waals: p = − 2 hence pVm = −
(Vm − b) Vm (Vm − b) Vm
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 25
B
virial: pVm = RT �1 + �
Vm
�e van der Waals expression for pVm is rewritten by dividing the denom-
inator and numerator of the �rst fraction by Vm to give
RT a
pVm = −
(1 − b�Vm ) Vm
= 1.41 × 103 K
If the reduced pressure is the same for two gases (�) and (�) it follows that
(i) From the tables in the Resource section, for H� p c = 12.8 atm, Tc = 33.23 K,
and for NH� p c = 111.3 atm, Tc = 405.5 K. Taking gas (�) as H� and
gas (�) as NH� , the pressure and temperature of NH� corresponding to
p(H2 ) = 1.0 atm and T (H2 ) = 298.15 K is calculated as
T (H2 ) 298.15 K
T (NH3 ) = (H )
× Tc(NH3 ) = × (405.5 K) = 3.6 × 103 K
Tc 2 33.23 K
T (H2 ) 298.15 K
T (Xe) = (H )
× Tc(Xe) = × (289.75 K) = 2.6 × 103 K
Tc 2 33.23 K
T (H2 ) 298.15 K
T (He) = (H )
× Tc(He) = × (5.2 K) = �� K
Tc 2 33.23 K
E�C.�(a) �e van der Waals equation of state in terms of the molar volume is given by
[�C.�b–��], p = RT�(Vm − b) − a�Vm2 . �is relationship is rearranged to �nd b
RT a a RT
p= − hence p + 2 =
Vm − b Vm2 Vm Vm − b
pVm2 + a RT Vm2 Vm − b
hence = hence =
Vm 2 Vm − b pVm + a
2 RT
RT Vm2
hence b = Vm −
pVm2 + a
RT Vm2
b = Vm − = (5.00 × 10−4 m3 mol−1 )
pVm2 + a
(8.3145 J K−1 mol−1 ) × (273 K) × (5.00 × 10−4 m3 mol−1 )2
−
(3.0 × 106 Pa) × (5.00 × 10−4 m3 mol−1 )2 + (0.50 m6 Pa mol−2 )
= 4.6 × 10−5 m3 mol−1
E�C.�(a) �e van der Waals equation of state in terms of the volume is given by [�C.�a–
��], p = nRT�(V − b) − an 2 �V 2 . �e parameters a and b for ethane are
given in the Resource section as a = 5.507 atm dm6 mol−2 and b = 6.51 ×
10−2 dm3 mol−1 .
With these units it is convenient to use R = 8.2057 × 10−2 dm3 atm K−1 mol−1 .
(i) T = 273.15 K, V = 22.414 dm3 , n = 1.0 mol
nRT an 2
p= − 2
V − nb V
(1.0 mol) × (8.2057 × 10−2 dm3 atm K−1 mol−1 ) × (273.15 K)
=
(22.414 dm3 ) − (1.0 mol) × (6.51 × 10−2 dm3 mol−1 )
(5.507 atm dm6 mol−2 ) × (1.0 mol)2
− = 0.99 atm
(22.414 dm3 )2
E�C.�(a) Recall that 1 atm = 1.01325 × 105 Pa, 1 dm6 = 10−6 m6 , and 1 Pa = 1 kg m−1 s−2
1.01325 × 105 Pa 10−6 m6
a = (0.751 atm dm6 mol−2 ) × × = 0.0761 Pa m6 mol−2
1 atm 1 dm6
= 0.0760 kg m−1 s−2 m6 mol−2 = 0.0761 kg m5 s−2 mol−2
10−3 m3
b = (0.0226 dm3 mol−1 ) × = 2.26 × 10−5 m3 mol−1
1 dm3
(i) If Vm is ��% smaller than the molar volume of a perfect gas, it follows that
Vm = Vm○ (1 − 0.12) = 0.88Vm○ . �e compression factor is then computed
directly as
Vm 0.88 × Vm○
Z= ○ = = �.��
Vm Vm○
(ii) From [�C.�–��] it follows that Vm = ZRT�p
Because Z < 1, implying that Vm < Vm○ , attractive forces are dominant.
E�C.�(a) �e van der Waals equation of state in terms of the volume is given by [�C.�a–
��], p = nRT�(V −b)−an 2 �V 2 . �e molar mass of N� is M = 2×14.01 g mol−1 =
28.02 g mol−1 , so it follows that the amount in moles is
nRT an 2
p= − 2
V − nb V
(3.29... × 103 mol) × (8.2057 × 10−2 dm3 atm K−1 mol−1 ) × (500 K)
=
(1000 dm3 ) − (3.29... × 103 mol) × (0.0387 dm3 mol−1 )
(1.352 atm dm6 mol−2 ) × (3.29... × 103 mol)2
− = 140 atm
(1000 dm3 )2
E�C.�(a) (i) �e pressure is computed from the equation of state for a perfect gas,
[�A.�–�], as p = nRT�V
nRT (10.0) × (8.2057 × 10−2 dm3 atm K−1 mol−1 ) × ([27 + 273.15] K)
p= =
V 4.860 dm3
= ��.� atm
(ii) �e van der Waals equation of state in terms of the volume is given by
[�C.�a–��], p = nRT�(V − b) − an 2 �V 2 . �is is used to calculate the
pressure
nRT an 2
p= − 2
V − nb V
(10.0 mol) × (8.2057 × 10−2 dm3 atm K−1 mol−1 ) × ([27 + 273.15] K)
=
(4.860 dm3 ) − (10.0 mol) × (0.0651 dm3 mol−1 )
(5.507 atm dm6 mol−2 ) × (10.0 mol)2
− = 35.2... = 35.2 atm
(4.860 dm3 )2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 29
Solutions to problems
P�C.� �e virial equation is given by [�C.�b–��], pVm = RT(1 + B�Vm + . . .), and
from the Resource section the second virial coe�cient B for N� at ��� K is
−1
−10.5 cm3 mol . �e molar mass of N� is 2 × 14.01 = 28.02 g mol−1 , hence the
molar volume is
V V 2.25 dm3
Vm = = = = 13.8... dm3 mol−1
n m�M (4.56 g)�(28.02 g mol−1 )
�is is used to calculate the pressure using the virial equation. It is convenient
to use R = 8.2057 × 10−2 dm3 atm K−1 mol−1 and express all the volumes in
dm3
RT B
p= �1 + �
Vm Vm
(8.2057 × 10−2 dm3 atm K−1 mol−1 ) × (273 K) −1.05 × 10−2 dm3 mol−1
= �1 + �
13.8... dm3 mol−1 13.8... dm3 mol−1
= �.�� atm
P�C.� �e virial equation is [�C.�b–��], pVm = RT(1 + B�Vm + C�Vm2 + . . .). �e com-
pression factor is de�ned in [�C.�–��] as Z = Vm �Vm○ , and the molar volume of
a perfect gas, Vm○ is given by Vm○ = RT�p.
It follows that
Vm = (RT�p)(1 + B�Vm + C�Vm2 ) = Vm○ (1 + B�Vm + C�Vm2 )
Vm B C
hence Z = ○ = 1 + +
Vm Vm Vm2
To evaluate this expression, the molar volume is approximated by the molar
volume of a perfect gas under the prevailing conditions
−c −1131 K2
1�2 1�2
T =� � =� �
ln(−a�b) ln[−(−0.1993 bar−1 )�(0.2002 bar−1 )]
= ���.� K
(c) �e virial equation (up to the second term) in terms of the molar volume
is given by [�C.�b–��]
B
pVm = RT �1 + �
Vm
Division of each side by p gives
RT B
Vm = �1 + �
p Vm
�e quantity RT�p is recognised as the molar volume of a perfect gas, Vm○ ,
so it follows that
B Vm B
Vm = Vm○ �1 + � hence ○ = Z = �1 + �
Vm Vm Vm
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 31
In Problem P�C.� it is shown that B is related to the van der Waals con-
stants by B = b − a�RT; using this, it is then possible to compute the
compression factor
a
B=b− = (0.03049 dm3 mol−1 )
RT
(5.464 atm dm6 mol−2 )
−
(8.2057 × 10−2 dm3 atm K−1 mol−1 ) × (776.4 K)
= −0.552... dm3 mol−1
P�C.� According to Table �C.� on page ��, for the Dieterici equation of state the
critical constants are given by
a a
pc = Vc = 2b Tc =
4e2 b 2 4bR
From the Resource section the values for Xe are Tc = 289.75 K, p c = 58.0 atm,
−1
Vc = 118.8 cm3 mol . �e coe�cient b is computed directly from Vc
a = 2Tc Vc R
= 2 × (289.75 K) × (118.8 × 10−3 dm3 mol−1 )
× (8.2057 × 10−2 dm3 atm K−1 mol−1 ) = 5.649 atm dm6 mol−2
�e two values of a are not the same; their average is �.��� atm dm6 mol−2 .
From Table �C.� on page �� the expression for the pressure exerted by a Di-
eterici gas is
nRT exp(−a�[RT V �n])
p=
V − nb
32 1 THE PROPERTIES OF GASES
P�C.�� �e van der Waals equation in terms of the molar volume is given by [�C.�b–
��], p = RT�(Vm − b) − a�Vm2 . Multiplication of both sides by Vm gives
RT Vm a
pVm = −
(Vm − b) Vm
and then division of the numerator and denominator of the �rst fraction by Vm
gives
RT a
pVm = −
(1 − b�Vm ) Vm
�e approximation (1−x)−1 ≈ 1+x+x 2 is the used to approximate 1�(1−b�Vm )
to give
b b2 a
pVm = RT �1 + + 2�−
Vm Vm Vm
�e terms in 1�Vm and 1�Vm2 are gathered together to give
1 a b2
pVm = RT �1 + �b − �+ 2�
Vm RT Vm
�is result is then compared with the virial equation in terms of the molar
volume, [�C.�b–��]
B C
pVm = RT �1 + + �
Vm Vm2
�is comparison identi�es the virial coe�cients as
a
B=b− C = b2
RT
√
From the given value C = 1200 cm6 mol−2 it follows that b = C = 34.64 cm3 mol−1 .
Expressed in the usual units this is b = �.����� dm3 mol−1 . �e value of a is
found by rearranging B = b − a�RT to
dp RT 2B 3C d2 p 2RT 6B 12C
=− 2 + 3 − 4 =0 = 3 − 4 + 5 =0
dVm Vm Vm Vm dVm2 Vm Vm Vm
�e �rst of these equations is multiplied through by Vm4 and the second by Vm5
to give
�is expression for Vm is then substituted into −RT Vm2 + 2BVm − 3C = 0 to give
(3C)2 3C
−RT + 2B − 3C = 0
B2 B
A term 3C is cancelled and the equation is multiplied through by B 2 to give
P�C.�� �e virial equation in terms of the pressure, [�C.�a–��], is (up to the second
term)
pVm = RT (1 + B′ p)
�e mass density ρ is given by m�V , and the mass m can be written as nM,
where n is the amount in moles and M is the molar mass. It follows that
34 1 THE PROPERTIES OF GASES
58
(p�ρ)�(kPa kg−1 m3 )
56
54
0 20 40 60 80 100
p�kPa
Figure 1.5
�e slope is B′ RT�M
B′ RT
= 0.04610 kg−1 m3
M
For methoxymethane, CH� OCH� , M = 2 × 12.01 + 6 × 1.0079 + 16.00 =
46.0674 g mol−1 .
B = B′ RT
= (0.0868 atm−1 ) × (8.2057 × 10−2 dm3 atm K−1 mol−1 ) × (298.15 K)
= �.�� dm3 mol−1
P�C.�� A gas can only be lique�ed by the application of pressure if the temperature is
below the critical temperature, which for N� is ���.� K.
RT
p(V − nb) = nRT becomes p(Vm − b) = RT hence Vm = +b
p
Vm p (RT�p + b)p bp
Z= = = 1+
RT RT RT
Vm p 10bp bp RT
= =1+ hence b=
RT RT RT 9p
P�C.�� �e virial equation in terms of the molar volume, [�C.�b–��], is (up to the third
term)
B C
pVm = RT �1 + + �
Vm Vm2
For part (a) only the �rst two terms are considered, and it then follows that a
plot of pVm against 1�Vm is expected to be a straight line with slope BRT; such
a plot is shown in Fig. �.�.
36 1 THE PROPERTIES OF GASES
p�MPa Vm �(dm3 mol−1 ) (pVm )�(MPa dm3 mol−1 ) (1�Vm )�(dm−3 mol)
0.400 0 6.220 8 2.488 3 0.160 75
0.500 0 4.973 6 2.486 8 0.201 06
0.600 0 4.142 3 2.485 4 0.241 41
0.800 0 3.103 1 2.482 5 0.322 26
1.000 2.479 5 2.479 5 0.403 31
1.500 1.648 3 2.472 5 0.606 69
2.000 1.232 8 2.465 6 0.811 16
2.500 0.983 57 2.458 9 1.016 7
3.000 0.817 46 2.452 4 1.223 3
4.000 0.609 98 2.439 9 1.639 4
linear
(pVm )�(MPa dm3 mol−1 )
2.48 quadratic
2.46
2.44
�e slope is BRT
For part (b) the data points are �tted to polynomial of order � in 1�Vm using
mathematical so�ware; the data are a slightly better �t to such a function (see
the dashed line in the graph above) which is
P�C.�� �e van der Waals equation of state in terms of the molar volume is given by
[�C.�b–��], p = RT�(Vm − b) − a�Vm2 . �is equation is a cubic in Vm , as is seen
by multiplying both sides by (Vm −b)Vm2 and then gathering the terms together
From the Resource section the van der Waals parameters for Cl� are
and to use R = 8.2057 × 10−2 dm3 atm K−1 mol−1 ; inserting all of these values
and the temperature gives the polynomial
�e molar volume of the van der Waals gas is about �% smaller than that of
the perfect gas.
38 1 THE PROPERTIES OF GASES
b = 4Vmolec N A
4 16πN A σ 3�2
= 4 � πr 3 � N A = � �
3 3 4π
16π(6.0221 × 1023 mol−1 ) 0.46 × 10−18 m2
3�2
= � �
3 4π
= 7.1 × 10−5 m3 mol−1 = �.��� dm3 mol−1
I�.� According to the equipartition theorem (�e chemist’s toolkit � in Topic �A),
each quadratic contribution to the energy of a molecule contributes 12 kT to
the average energy per molecule. Translational kinetic energy is a quadratic
term, and because translation is possible in three dimensions, the quoted en-
ergy density of 0.15 J cm−3 is the result of three such contributions.
�e rotation of a molecule about an axis is also a quadratic contribution to the
energy, and in general three such contributions are expected corresponding to
rotation about three mutually perpendicular axes. However, linear molecules,
such as diatomics, do not show rotation about their long axes, so there are
only two contributions. �e contribution of rotation to the energy density will
therefore be 23 of that due to translation
��� � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
trans. rot.
2A Internal energy
Answers to discussion questions
D�A.� Table �A.� on page �� lists four varieties of work: expansion, surface expan-
sion, extension, and electrical. �ere is also work associated with processes in
magnetic and gravitational �elds which we will not describe in detail.
D�A.� An isothermal expansion of a gas may be achieved by making sure that the gas
and its container are in thermal contact with a large ‘bath’ which is held at a
constant temperature – that is, a thermostat.
D�A.� Work is done when a body is moves against an opposing force. For an in�nites-
imal displacement in the x-direction, dx, against a force F along that direction
the work done by the body is F dx.
When the energy of a system changes as a result of a temperature di�erence
between the system and its surroundings, the resulting energy transfer from the
hotter to the cooler body is described as heat. In thermodynamic terms, both
heat and work cause the internal energy of an object to change: if heat ‘�ows in’
the internal energy of the body rises, if the body ‘does work’, its internal energy
decreases.
If the internal energy of an object increases, this is interpreted in molecular
terms as the molecules moving up to higher energy levels. If the molecules
drop down to lower levels the resulting energy is available as heat or work.
Solutions to exercises
E�A.�(a) �e system is expanding against a constant external pressure, hence the expan-
sion work is given by [�A.�–��], w = −pex ∆V . �e change in volume is the
cross-sectional area times the linear displacement
�e external pressure is 1.0 atm = 1.01325 × 105 Pa, therefore the expansion
work is
w = −(1.01325 × 105 Pa) × (7.5 × 10−4 m3 ) = −76 J
E�A.�(a) For all cases ∆U = 0, because the internal energy of a perfect gas depends on
the temperature alone.
Vf
w = −nRT ln � �
Vi
30.0 dm3
= −(1.00 mol) × (8.3145 J K−1 mol−1 ) × (293.15 K) × ln � �
10.0 dm3
= −2.68 × 103 J = −�.�� kJ
(ii) �e �nal pressure of the expanding gas is found using the perfect gas law,
[�A.�–�]
�is pressure equals the constant external pressure against which the gas
is expanding, therefore the work of expansion is
Tf 400 K
pf = pi × = (1.00 atm) × � � = 1.33 atm
Ti 300 K
3
∆U = nC V ,m ∆T = (1.00 mol) × � × 8.3145 J K−1 mol−1 � × (400 K − 300 K)
2
= +1.25 × 103 J = +1.25 kJ
E�A.�(a) (i) �e work of expansion against constant external pressure is given by [�A.�–
��]
w = −pex ∆V
133.3 Pa
= −(200 Torr) × � � × (3.3 × 10−3 m3 ) = −88 J
1 Torr
Note that the pressure is expressed in Pa and the change in volume in m3 ,
to give the work in J.
(ii) �e work done in a reversible isothermal expansion is given by [�A.�–��],
w = −nRT ln(Vf �Vi ). �e amount in moles of methane is
m (4.50 g)
n= = = 0.280... mol
M (16.0416 g mol−1 )
w = −(0.280... mol) × (8.3145 J K−1 mol−1 ) × (310 K)
[12.7 + 3.3] dm3
× ln � � = −1.7 × 102 J
12.7 dm3
Note that the modulus of the work done in a reversible expansion is greater
than the work for expansion against constant external pressure because
the latter is an irreversible process.
E�A.�(a) �e chemist’s toolkit � in Topic �A gives an explanation of the equipartition
theorem. �e molar internal energy is given by
Um = 1
2
× (νt + νr + 2νv ) × RT
where νt is the number of translational degrees of freedom, νr is the number
of rotational degrees of freedom and νv is the number of vibrational degrees of
freedom. As each gas molecule can move independently along the x, y and z
axis, the number of translational degrees of freedom is three.
(i) Molecular iodine is a diatomic molecule, therefore it has two degrees of
rotational freedom. On account of its heavy atoms, molecular iodine is
likely to have one degree of vibrational freedom at room temperature.
�erefore, the molar internal energy of molecular iodine gas at room
temperature is
Um = 1
2
× (3 + 2 + 2) × RT = 7
2
× (8.3145 J K−1 mol−1 ) × (298.15 K)
= 8.7 kJ mol−1
(ii) and (iii) Both methane (tetrahedral) and benzene (planar) have three degrees of
rotational freedom. At room temperature it is unlikely that any of their
vibrational modes would be excited, therefore both are expected to have
approximately the same internal energy at room temperature:
Um = 1
2
× (3 + 3 + 0) × RT = 3 × (8.3145 J K−1 mol−1 ) × (298.15 K)
= 7.4 kJ mol−1
E�A.�(a) A state function is a property with a value that depends only on the current state
of the system and is independent of how the state has been prepared. Pressure,
temperature and enthalpy are all state functions.
42 2 INTERNAL ENERGY
Solutions to problems
P�A.� From the equipartition theorem (�e chemist’s toolkit � in Topic �A), the molar
internal energy is given by
Um = 1
2
× (νt + νr + 2νv ) × RT
U= 1
2
× (3 + 2 + 0) × RT = 5
2
× (8.3145 J K−1 mol−1 ) × (298.15 K)
= 6.2 kJ mol−1
0 0
1 2 5 l 1 2 5
= � ax 2 − bx 2 � = al 2 − bl 2
2 5 0 2 5
Note that the second term arises from the non Hooke’s Law behaviour of the
elastomer, reducing the overall work done.
P�A.� (a) �e natural logarithm can be expanded using the Taylor series as ln(1 +
ν) ≈ ν + ν 2 �2! + ν 3 �3! + ..., which, for ν << 1, can be approximated as
ln(1 + ν) ≈ ν, and similarly, ln(1 − ν) ≈ −ν. �erefore,
kT kT νkT
F= [ln(1 + ν) − ln(1 − ν)] ≈ [ν − (−ν)] =
2l 2l l
Because ν = n�N, it follows that
νkT nkT
F= =
l Nl
(b) Hooke’s law predicts F = const × x, that is the restoring force is directly
proportional to the displacement. Using n = x�l, the expression for the
force obtained in part (a) is rewritten as
nkT kTx
= ≡ const × x
Nl Nl2
�erefore Hooke’s law applies and kT�N l 2 is the force constant.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 43
30
perfect
repulsion only
25 attraction only
p�atm 20
15
10
0.5 1.0 1.5 2.0 2.5
V �dm 3
Figure 2.1
P�A.� (a) �e virial equation of state is given by [�C.�b–��]. �e �rst three terms
are
1 B C
p = RT � + + �
Vm Vm2 Vm3
Using Vm = V �n gives
1 nB n 2 C
p = nRT � + + 3 �
V V2 V
1 3
2
n RTC = 1
2
× n 2 × (nRT) × C
−2
= 1
2
× (1.0 mol)2 × (2.26... × 103 J) × (1.2 × 103 cm6 mol )
= 1.36... × 106 J cm6
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 45
1000 cm3
w = −(2.26... × 103 J) × ln � �
500 cm3
1 1
+ (−4.92... × 104 J cm3 ) × � − �
1000 cm3 500 cm3
1 1
+ (1.36... × 106 J cm6 ) × � − �
(1000 cm )
3 2 (500 cm3 )2
= −1.5 kJ
(b) For a perfect gas, the work of reversible isothermal expansion is given by
[�A.�–��]
Vf
w = −nRT ln � �
Vi
1000 cm3
= −(1.0 mol) × (8.3145 J K−1 mol−1 ) × (273 K) × ln � �
500 cm3
= −1.6 × 103 J = −1.6 kJ
2B Enthalpy
D�B.� If a substance is heated at constant volume all of the energy as heat is trans-
formed into internal energy of the substance. If the same process is carried out
under conditions of constant pressure some of the energy as heat will be used
to expand the substance against the external pressure and so less of the energy
as heat is transformed into internal energy. �is e�ect is largest for gases whose
volumes change much more rapidly with temperature than do solids or liquids.
Solutions to exercises
sides gives
T2 T2 T2
dH = � C p dT = � (a + bT)dT = �aT + 12 bT 2 �T
T2
�
T1 T1 T1 1
i
Hence
nR
w = −p × (Tf − Ti ) = −nR∆T
p
= −(1.00 mol) × (8.3145 J K−1 mol−1 ) × (373.15 K − 298.15 K)
= −6.23... × 102 J = −624 J
∆U = q + w = (+10.7... kJ) + (−0.623... kJ) = +10.1 kJ
(ii) �e energy and enthalpy of a perfect gas depends on the temperature
alone, hence ∆H and ∆U is the same as above, that is ∆H = +10.7 kJ and
∆U = +10.1 kJ . Under constant volume conditions there is no expansion
work, w = 0 , therefore the heat is equal to the change in internal energy,
qV = +10.1 kJ .
E�B.�(a) Under constant pressure qp = ∆H, therefore
qp = ∆H = nC p,m ∆T = (3.0 mol) × (29.4 J K mol−1 ) × (285 K − 260 K)
= 2.20... kJ = +2.2 kJ
�e de�nition of enthalpy is given by [�B.�–��], H = U + pV . For a change
at constant pressure, it follows that ∆H = ∆U + p∆V , where ∆V = Vf − Vi .
If the gas is assumed to be perfect, then Vf = nRTf �p and Vi = nRTi �p, so
∆V = (Tf − Ti )nR�p. Hence p∆V = nR(Tf − Ti ) = nR∆T.
∆U = ∆H − nR∆T
= (2.20... kJ) − (3.0 mol) × (8.3145 J K−1 mol−1 ) × (285 K − 260 K)
= +1.6 kJ
E�B.�(a) �e heat transferred under constant pressure equals the change in enthalpy
of the system, [�B.�b–��], qp = ∆H. �e relationship between the change in
enthalpy, change in temperature and the heat capacity is given by [�B.�b–��]
∆H (229 J)
C p,m = = = 29.9... J K mol−1 = 30 J K−1 mol−1
n∆T (3.0 mol) × (2.55 K)
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 47
E�B.�(a) In the reaction � moles of gas are formed from � moles of gas. �e relationship
between ∆H and ∆U is given by [�B.�–��]
Solutions to problems
P�B.� �� g of benzene is n = m�M = (10 g)�(78.1074 g mol−1 ) = 0.128... mol. �e
heat needed to vaporize �� g of benzene under constant pressure is qp = n ×
∆ vap Hm = (0.128... mol) × (30.8 kJ mol−1 ) = 3.94... kJ. A current I �owing
through a potential ∆� corresponds to a power of I∆�. If the current �ows for
a time interval ∆t the energy is power × time , that is q = I∆�∆t. �erefore
∆t = q�(I∆�).
qp (3.94... × 103 J)
∆t = = = 657 s = 11 min
I∆� (0.50 A) × (12 V)
P�B.� Fitting the data set using a computer program to an expression in the form of
C −p,m
○
(T) = a+bT+cT −2 yields a = 48.0 J K−1 mol−1 , b = 6.49×10−3 J K−2 mol−1
and c = −9.33 × 105 J K mol−1 . �e change in enthalpy in response to a change
in temperature at constant pressure is given by [�B.�a–��], dHm = C p,m dT.
Substituting in the expression for C p,m and integrating both sides gives
dHm = (a + bT + cT −2 ) dT
(T2 ) T2
� dHm = � (a + bT + cT −2 ) dT
(T1 ) T1
By using Integral A.� in the Resource section for each term, it follows that
1 1
Hm (T2 ) − Hm (T1 ) = a(T2 − T1 ) + 12 b(T22 − T12 ) − c � − �
T2 T1
= (48.0 J K−1 mol−1 ) × (1500 K − 298.15 K)
+ 12 × (6.49 × 10−3 J K−2 mol−1 )
× [(1500 K)2 − (298.15 K)2 ]
1 1
− (−9.33 × 105 J K mol−1 ) × � − �
1500 K 298.15 K
= 62.2 kJ
48 2 INTERNAL ENERGY
P�B.� Since the volume is �xed no expansion work is done, hence w = 0 and
∆U = qV = +2.35 kJ . From H = U + pV it follows that ∆H = ∆U + V ∆p at
constant volume. Using the van der Waals equation of state, [�C.�b–��],
RT a R∆T
p= − 2 it follows that ∆p =
Vm − b Vm Vm − b
Note that the term in a cancels because the volume is constant. �erefore
RV ∆T
∆H = ∆U + v∆p = ∆U +
Vm − b
From the given data Vm = V �n = (15.0 dm3 )�(2.0 mol) = 7.5 dm3 mol−1 ,
∆T = 341 K − 300 K = 41 K and from Table �C.� b = 4.29 × 10−2 dm3 mol−1 .
�erefore
(8.3145 J K−1 mol−1 ) × (15.0 dm3 ) × (41 K)
∆H = (+2.35 × 103 J) +
(7.5 dm3 mol−1 ) − (4.29 × 10−2 dm3 mol−1 )
= 3.03... × 103 J = 3.0 kJ
2C Thermochemistry
Answers to discussion questions
D�C.� When a system is subjected to constant pressure conditions, and only expan-
sion work can occur, the energy supplied as heat is the change in enthalpy of the
system. �us enthalpy changes in the system can be determined by measuring
the amount of heat supplied under constant-pressure conditions.
A very simple example o�en encountered in elementary laboratory classes is a
thermally insulated vessel (for example, a foam plastic co�ee cup) le� open to
the atmosphere: the heat released in the reaction is determined by measuring
the change in temperature of the contents.
For a combustion reaction a constant-pressure �ame calorimeter (Section �B.�(b)
on page ��) may be used. In this apparatus a certain amount of substance
burns in a supply of oxygen and the rise in temperature is monitored. More
sophisticated methods include isothermal titration calorimetry and di�erential
scanning calorimetry, both described in Section �C.� on page ��.
D�C.� �e main objection to the use of the term ‘heat’ to describe the energy change
associated with a physical or chemical process is that heat is not a state function.
�e value of the heat therefore depends on the path chosen.
If, in fact, the processes being described takes place at constant pressure, the
heat is equal to the enthalpy change. Because enthalpy is a state function, the
heat measured under these circumstance is a property of the physical or chem-
ical change itself, and not a�ected by the path taken, and so is a meaningful and
useful quantity to discuss.
It is more appropriate to talk about the enthalpy change, the change in a state
function, rather to talk about the heat which, because of an unstated restriction,
just so happens to have the same value.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 49
Solutions to exercises
∆ r H −○ = ∆ r U −○ + ∆ng RT
= (−1373 × 103 J mol−1 ) + (+2.0 mol) × (8.3145 J K−1 mol−1 ) × (298 K)
= −1368 kJ mol−1
E�C.�(a) (i) �e standard reaction enthalpy at 298 K is calculated from the standard
enthalpies of formation at 298 K using [�C.�a–��]
∆ r H −○ (298 K) = � ν∆ f H −
−
○
� ν∆ f H −○
products reactants
∆ r C −p○ = � νC p,m −
−
○
� νC −p,m
○
products reactants
= C −p,m
○
(CO, g) + C −p,m
○
(H2 , g) − C −p,m
○
(H2 O, g) − C −p,m
○
(graphite, s)
= (29.14 J K−1 mol−1 ) + (28.824 J K−1 mol−1 ) − (33.58 J K−1 mol−1 )
− (8.527 J K−1 mol−1 ) = +15.86 J K−1 mol−1
It is assumed that all heat capacities are constant over the temperature
range of interest, therefore the integrated form of Kirchho� ’s Law is ap-
50 2 INTERNAL ENERGY
plicable, [�C.�d–��]
E�C.�(a) �e reaction equation C(graphite, s) + O� (g) ��→ CO� (g) represents the for-
mation of CO� (g) from its elements in their reference states, therefore ∆ r H −○ (298 K) =
∆ f H −○ (CO2 , g) = −393.51 kJ mol−1 . �e variation of standard reaction en-
thalpy with temperature is given by Kirchho� ’s Law, [�C.�a–��]
T2
∆ r H −○ (T2 ) = ∆ r H −○ (T1 ) + � ∆ r C −p○ dT
T1
∆ r C −p○ = � νC p,m −
−
○
� νC −p,m
○
products reactants
= C −p,m
○
(CO2 , g) − C −p,m
○
(O2 , g) − C −p,m
○
(graphite, s)
E�C.�(a) �e equation for combustion of ethylbenzene is C� H�� (l) + ��.� O� (g) ��→
� CO� (g) + � H� O(l). �e standard enthalpy of combustion is calculated us-
ing [�C.�a–��] and using values for the standard enthalpies of formation from
Table �C.�.
∆ c H −○ = � ν∆ f H −
−
○
� ν∆ f H −○
products reactants
= 8∆ f H −○ (CO2 , g) + 5∆ f H −○ (H2 O, l) − 21
∆ H −○ (O2 , g) −
2 f
∆ f H −○ (C8 H10 , l)
= 8 × (−393.51 kJ mol−1 ) + 5 × (−285.83 kJ mol ) − 0 − (−12.5 kJ mol−1 )
−1
E�C.�(a) �e equation for the combustion of naphthalene is C�� H� (s) + �� O� (g) ��→
�� CO� (g) + � H� O(l). ∆ c H −○ is computed using the thermochemical data from
Tables �C.� and �C.� and equation [�C.�a–��]
∆ c H −○ = � ν∆ f H −
−
○
� ν∆ f H −○
products reactants
∆ c U −○ = ∆ c H −○ − ∆ng RT
= (−5.15... × 106 J) − (−2.0 mol) × (8.3145 J K−1 mol−1 ) × (298 K)
= −5.15... × 103 kJ mol−1
0.120 g
qv = n∆ c U −○ = � � × (−5.15... × 103 kJ mol−1 ) = −4.82 ...kJ
128.1632 g mol−1
0.150 g
qv = n∆ c U −○ = � � × (−3.05... × 103 kJ mol−1 ) = −4.86... kJ
94.1074 g mol−1
(ii) Reaction(�) represents the formation of � moles of HCl(g) from its ele-
ments in their reference states, therefore the standard enthalpy of forma-
tion of HCl(g) is
Solutions to problems
P�C.� At constant pressure the temperature rise is given by [�B.�–��], qp = C p ∆T.
�e heat capacity is approximated as nC −p,m
○
(H2 O, l) where n is the amount in
moles. �erefore
qp (1.0 × 107 J)
∆T = = = 36.8... K = 37 K
nC −p,m
○
� 18.0158 g mol−1 � × (75.29 J K−1 mol−1 )
65000 g
P�C.� (a) �e combustion equation for cyclopropane is C� H� (g) + �.� O� (g) ��→
� CO� (g) + � H� O(l). Using data from Table �C.� and applying [�C.�a–��]
for this case
∆ c H −○ = 3∆ f H −○ (CO2 , g) + 3∆ f H −○ (H2 O, l) − ∆ f H −○ (C3 H6 , g)
54 2 INTERNAL ENERGY
�erefore
∆ r H −○ = ∆ f H −○ (propene, g) − ∆ f H −○ (cyclopropane, g)
= (+20.42 kJ mol−1 ) − (+52.9... kJ mol−1 ) = −32.56 kJ mol−1
It follows that the heat released during the combustion of ribose is �qv � = C∆T =
(11.3... kJ K−1 ) × (0.910 K) = 10.3... kJ, and so the standard internal energy of
combustion is
−�qv � −(10.3... kJ)
∆ c U −○ = = = −2.12... × 103 kJ mol−1
n (0.727 g)�(150.129 g mol−1 )
�e combustion reaction is C� H�� O� (s) + � O� (g) ��→ � CO� (g) + � H� O(l).
�ere is no change in the number of gaseous species when going from reactants
to products, therefore ∆ c U −○ = ∆ c H −○ = −2.12... × 103 kJ mol−1 . Applying
[�C.�a–��] for this reaction and using the values for the standard enthalpies
of formation from Table �C.� gives
P�C.� �e chemical reaction for the combustion is C�� (s) + �� O� (g) ��→ �� CO� (g).
�e standard internal energy of combustion is
�ere is no change in the number of gaseous species when going from re-
actants to products, therefore ∆ c U −○ = ∆ c H −○ = −2.59... × 104 kJ mol−1 =
−25966 kJ mol−1 . Applying [�C.�a–��] for this case gives
�erefore
P�C.� �e reaction equation for combustion of methane is CH� (g) + � O� (g) ��→
CO� (g) + � H� O(g). �e standard enthalpy of combustion can be calculated
using [�C.�a–��] and using the values for the standard enthalpies of formation
from Tables �C.� and �C.�
∆ c H −○ = � ν∆ f H −
−
○
� ν∆ f H −○
products reactants
∆ r C −p○ = � νC p,m −
−
○
� νC −p,m
○
products reactants
= C −p,m
○
(CO2 , g) + 2C −p,m
○
(H2 O, g) − 2C −p,m
○
(O2 , g) − C −p,m
○
(CH4 , g)
Solutions to exercises
E�D.�(a) �e volume of the liquid can be written as
V = V ′ (a + bT + cT 2 )
where a = 0.75, b = 3.9 × 10−4 K−1 and c = 1.48 × 10−6 K−2 . �e expansion
coe�cient is de�ned in [�D.�–��] as α = (1�V )(∂V �∂T) p . �e derivative with
respect to T is
∂V
� � = V ′ (b + 2cT)
∂T p
�erefore
1 b + 2cT
α= × [V ′ (b + 2cT)] =
V ′ (a + bT + cT 2 ) a + bT + cT 2
Evaluating this expression at 320 K gives
(3.9 × 10−4 K−1 ) + 2 × (1.48 × 10−6 K−2 ) × (320 K)
α 320 =
(0.75) + (3.9 × 10−4 K−1 ) × (320 K) + (1.48 × 10−6 K−2 ) × (320 K)2
= +1.3 × 10−3 K−1
α 2 T Vm α 2 TM
C p,m − C V ,m = =
κT κT ρ
(12.4 × 10−4 K−1 )2 × (298 K) × (78.1074 g mol−1 )
=
[(92.1 × 10−6 bar−1 ) × (1 bar�105 Pa)] × (0.879 × 106 g m−3 )
= +44.2 J K−1 mol−1
�e van der Waals parameter a of water vapour is found in Table �C.�, and
needs to be converted to SI units
hence
Vm,f
a 1 1
∆U m = − � = −a � − �
Vm2 Vm,i Vm,f Vm,i
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 59
�e van der Waals parameter a for nitrogen gas is found in Table �C.�, and
needs to be converted to SI units
w = − � pdVm
Substituting in the expression for the pressure of a van der Waals gas, [�C.�b–
��]
RT a RT a
w = −� − dVm = − � dVm + � dVm
Vm − b Vm2 Vm − b Vm2
RT
= −� dVm + ∆Um
Vm − b
Vm,f RT Vm,f − b
q=� dVm = RT ln(Vm − b)�Vm,f = RT ln � �
V
Vm,i Vm − b m,i
Vm,i − b
20.00 dm3 − 3.87 × 10−2 dm3
= (8.3145 J K−1 mol−1 ) × (298 K) × ln � �
1.00 dm3 − 3.87 × 10−2 dm3
= +7.51... × 103 J mol−1 = +7.52 kJ mol−1
where the value for b is taken from the Resource section. From the First Law
the work done is w = −q + ∆Um , hence
Solutions to problems
P�D.� �e expansion coe�cient is de�ned in [�D.�–��] as α = (1�V )(∂V �∂T) p . For
su�ciently small changes the derivatives are approximated by �nite changes to
give
1 δV
α= � �
V δT
60 2 INTERNAL ENERGY
1 1 ∂V 1 ∂V
dV = � � dp + � � dT
V V ∂p T V ∂T p
Noting the de�nition of κ T , [�D.�–��], κ T = −(1�V )(∂V �∂p)T and that
of α, [�D.�–��], α = (1�V )(∂V �∂T) p , and rewriting (1�V )dV as d ln V
gives
d ln V = −κ T dp + αdT
Dividing the expression for dp by p gives
1 1 ∂p 1 ∂p
dp = � � dV + � � dT
p p ∂V T p ∂T V
1 ∂p 1 ∂p ∂V
d ln p = � � dV − � � � � dT
p ∂V T p ∂V T ∂T p
1 ∂p ∂V
= � � �dV − � � dT�
p ∂V T ∂T p
1 1� 1 � ∂V
= �− � � � �dV − � � dT�
V p �− � � �
1 ∂V ∂T p
V ∂p T
1 dV 1 ∂V
= �− + � � dT�
pκ T V V ∂T p
1 dV
= �− + αdT�
pκ T V
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 61
It follows that
∂V (V − nb)2 V 3
� � =
∂p T −nRT V 3 + 2n 2 a(V − nb)2
and hence
1 ∂V V 2 (V − nb)2
κT = − � � =
V ∂p T nRT V 3 − 2n 2 a(V − nb)2
�e expansion coe�cient is de�ned in [�D.�–��] as α T = (1�V )(∂V �∂T) p ,
which is rewritten using the reciprocal identity as
1 1
α=
V (∂T�∂V ) p
��� � � � � � � � � � ��� � � � � � � � � � � �
A
∂T p na 2na T 2na
� � = + − (V − nb) = − (V − nb)
∂V p nR RV 2 RV 3 V − nb RV 3
RT V 3 − 2na(V − nb)2
=
(V − nb)RV 3
where the term A is identi�ed from the previous equation as being equal to
T�(V − nb).Using the reciprocal identity
∂V (V − nb)RV 3
� � =
∂T p RT V 3 − 2na(V − nb)2
therefore
1 ∂V (V − nb)RV 2
α= � � =
V ∂T p RT V 3 − 2na(V − nb)2
62 2 INTERNAL ENERGY
It follows that
κT V 2 (V − nb)2 (V − nb)RV 2
= ×
α nRT V 3 − 2n 2 a(V − nb)2 RT V 3 − 2na(V − nb)2
(V − nb)
=
nR
For a molar quantity V → Vm and n → 1 to give κ T �α = (Vm − b)�R or
κ T R = α(Vm − b) .
1 ∂V
µ= �T � � − V�
Cp ∂T p
1 nR nRT nb
= �T � � − � + nb�� = −
Cp p p Cp
�e heat capacity and the van der Waals parameter b are always positive, there-
fore the Joule–�omson coe�cient µ is negative for this gas. �is means that
the temperature of the gas will increase when expanding in an isenthalpic pro-
cess.
P�D.� From [�D.��–��], dH = −µC p dp+C p dT, it follows that at constant temperature
1 ∂H
µ=− � �
C p ∂p T
426.6
426.4
H�kJ kg−1
426.2
426.0
425.8
425.6
0.07 0.08 0.09 0.10 0.11 0.12 0.13
p�MPa
Figure 2.2
462
460
−1
H�kJ kg
458
456
0.8 0.9 1.0 1.1 1.2
p�MPa
Figure 2.3
2E Adiabatic changes
Answers to discussion questions
D�E.� In an adiabatic expansion the system does work but as no energy as heat is
permitted to enter the system, the internal energy of the system falls and so
consequently does the temperature. From the First Law, ∆U = w because q =
0. However, the change in internal energy is also related to the temperature
change and the heat capacity: ∆U = C V ∆T. Equating these two expressions
for ∆U gives w = C V ∆T, or dw = C V dT for an in�nitesimal change.
For a reversible expansion, the work is dw = −pdV . Equating these two ex-
pressions for the work gives −pdV = C V dT. �is equation is the point from
which the relationships between pressure, volume and temperature for a re-
versible adiabatic expansion are found: the heat capacity comes into the �nal
expressions via this route.
In words, the key thing here is that in an adiabatic process there is a change
64 2 INTERNAL ENERGY
Solutions to exercises
E�E.�(a) For a reversible adiabatic expansion the initial and �nal states are related by
[�E.�–��], p i Vi = p f Vf , where γ is the ratio of heat capacities, γ = C p,m �C V ,m .
γ γ
�e initial and �nal states are also related by (Tf �Ti ) = (Vi �Vf )1�c where c =
C V ,m �R. For this gas c = (20.8 J K−1 mol−1 )�(8.3145 J K−1 mol−1 ) = 2.50...
and hence
1�2.50...
Vi 1�c 5.79... dm3
Tf = Ti � � = (300 K) × � � = 2.57... × 102 K = 258 K
Vf 8.46 dm3
Vi 1�c Vi 1�c
∆T = Tf − Ti = Ti � � − Ti = Ti �� � − 1�
Vf Vf
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 65
Vi 1�c
w ad = C V ∆T = nC V ,m ∆T = nC V ,m Ti �� � − 1�
Vf
Vi 1�c
= n(C p,m − R)Ti �� � − 1�
Vf
2.45 g
=� � × [(37.11 J K−1 mol−1 )
44.01 g mol−1
− (8.3145 J K−1 mol−1 )] × (300.15 K)
� 1�3.46... �
� �
×� − 1�
500 cm3
�
� 3.00 × 103 cm3 � � = −194 J
� �
� �
E�E.�(a) �e initial and �nal states in a reversible adiabatic expansion are related by
[�E.�–��], p i Vi = p f Vf , therefore
γ γ
1.4
Vi γ 0.50 dm3
p f = p i � � = (67.4 kPa) × � � = 9.7 kPa
Vf 2.00 dm3
E�E.�(a) Ammonia and methane are nonlinear polyatomic molecules which have three
degrees of translational and three degrees of rotational freedom. From the
equipartition theorem (�e chemist’s toolkit � in Topic �A)
C V ,m = 1
2
× (νt + νr + 2νv ) × R
(i) Without any vibrational contribution the calculation is the same for both
molecules. C V ,m = 12 × (3 + 3 + 0) × R = 3R. For a perfect gas, [�B.�–��],
C p,m = C V ,m + R, therefore
C V ,m + R 3R + R 4
γ= = =
C V ,m 3R 3
�e experimental values of γ for ammonia and methane are the same, γ = 1.31,
which is closer to the value calculated without taking the vibrational degrees of
freedom into account.
66 2 INTERNAL ENERGY
E�E.�(a) For a reversible adiabatic expansion the initial and �nal states are related by
[�E.�a–��], (Tf �Ti ) = (Vi �Vf )1�c , where c = C V ,m �R. For a monoatomic ideal
gas C V ,m = 32 R, so c = 32 . �erefore
1 2
Vi c 1.0 dm3 3
Tf = Ti � � = (273.15 K) × � � = 1.3 × 102 K
Vf 3.0 dm3
Solutions to problems
P�E.� �e work done in a reversible adiabatic expansion is given by [�E.�–��], w ad =
C V ∆T. �e initial and �nal states in a reversible adiabatic expansion are related
by [�E.�b–��], Vi Tic = Vf Tfc , where c = C V ,m �R. If the gas is assumed to be
perfect, C p,m − C V ,m = R [�B.�–��], and so
Integrated activities
I�.� �e change in reaction enthalpy with respect to temperature is described by
Kirchho� ’s Law, [�C.�a–��]. Assuming that the heat capacities are indepen-
dent of temperature, the integrated form of Kircho� ’s Law is applicable and is
given by [�C.�d–��]
∆ r H −○ (T2 ) = ∆ r H −○ (T1 ) + (T2 − T1 )∆ r C −p○
If ∆ r C −p○ is negative, then the reaction enthalpy will decrease with increasing
temperature, whereas if ∆ r C −p○ is positive, then the reaction enthalpy will in-
crease with increasing temperature.
(a)
∆ r C −p○ = � νC p,m −
−
○
� νC −p,m
○
products reactants
= 2C −p,m
○
(H2 O, l) − C −p,m
○
(O2 , g) − 2C −p,m
○
(H2 , g)
= 2 × (9R) − � 72 R� − 2 × � 72 R� = +7 12 R
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 67
If ∆ r C −p○ is negative, then the reaction enthalpy will decrease with increasing
temperature, whereas if ∆ r C −p○ is positive, then the reaction enthalpy will in-
crease with increasing temperature.
(a)
∆ r C −p○ = � νC p,m −
−
○
� νC −p,m
○
products reactants
= 2C −p,m
○
(H2 O, g) − C −p,m
○
(O2 , g) − 2C −p,m
○
(H2 , g)
= 2 × (4R) − � 72 R� − 2 × � 72 R� = −2 12 R
∆ r C −p○ = C −p,m
○
(CO2 , g) + 2C −p,m
○
(H2 O, g) − 2C −p,m
○
(O2 , g) − C −p,m
○
(CH4 , g)
= � 72 R� + 2 × (4R) − 2 × � 72 R� − (4R) = + 12 R
∆ r C −p○ = 2C −p,m
○
(NH3 , g) − C −p,m
○
(N2 , g) − 3C −p,m
○
(H2 , g)
= 2 × (4R) − � 72 R� − 3 × � 72 R� = −6R
3A Entropy
Answers to discussion questions
D�A.� Everyday experience indicates that the direction of spontaneous change in an
isolated system is accompanied by the dispersal of the total energy of the sys-
tem. For example, for a gas expanding freely and spontaneously into a vacuum,
the process is accompanied by a dispersal of energy and matter. For a perfect
gas this entropy change of such an expansion is derived in Section �A.�(a) on
page �� as ∆S = nR ln(Vf �Vi ). �e entropy change is clearly positive if Vf is
greater than Vi .
�e molecular interpretation of this thermodynamic result is based on the iden-
ti�cation of entropy with molecular disorder. An increase in disorder results
from the chaotic dispersal of matter and energy and the only changes that can
take place within an isolated system (the universe) are those in which this kind
of dispersal occurs. �is interpretation of entropy in terms of dispersal and
disorder allows for a direct connection of the thermodynamic entropy to the
statistical entropy through the Boltzmann formula S = k ln W, where W is
the number of microstates, the number of ways in which the molecules of the
system can be arranged while keeping the total energy constant.
�e concept of the number of microstates makes quantitative the more ill-
de�ned qualitative concepts of ‘disorder’ and ‘the dispersal of matter and en-
ergy’ used above to give a physical feel for the concept of entropy. A more
‘disorderly’ distribution of energy and matter corresponds to a greater number
of microstates associated with the same total energy.
D�A.� �e Second Law of thermodynamics states only that the total entropy of both
the system (here, the molecules organizing themselves into cells) and the sur-
roundings (here, the medium) must increase in a naturally occurring process.
It does not state that entropy must increase in a portion of the universe that
interacts with its surroundings. In this case, the cells grow by using chemical
energy from their surroundings (the medium) and in the process the increase
in the entropy of the medium outweighs the decrease in entropy of the system.
Hence, the Second Law is not violated.
Solutions to exercises
E�A.�(a) �e e�ciency is de�ned in [�A.�–��], η = �w���q h �, and for a Carnot cycle
e�ciency is given by [�A.�–��], η = 1 − (Tc �Th ). �ese two are combined
70 3 THE SECOND AND THIRD LAWS
and rearranged into an expression for the temperature of the cold sink
E�A.�(a) For the process to be spontaneous it must be irreversible and obey the Clausius
inequality [�A.��–��] implying that ∆S tot = ∆S + ∆S sur > 0. In this case,
∆S tot = 125 J K−1 + (−125 J K−1 ) = 0, thus the process is not spontaneous in
either direction and is at equilibrium.
14 g −1
∆S = � −1 � × (8.3145 J K mol ) × ln (2) = +2.9 J K
−1 −1
.
28.02 g mol
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 71
Solutions to problems
P�A.� (a) Isothermal reversible expansion
�e work of a reversible isothermal expansion of an ideal gas is given
by [�A.�–��], w = −nRT ln (Vf �Vi ). Because at �xed temperature p ∝
(1�V ) as given by Boyle’s law, an equivalent expression is
pi
w = −nRT ln � �
pf
= −(1.00 mol) × (8.3145 J K−1 mol−1 )
3.00 atm
× (273.15 K + 27 K) × ln � �
1.00 atm
= −2.74 × 103 ... J = −2.74 kJ .
∆S tot = ∆S + ∆S sur
= (+9.13... J K−1 ) + (−5.53... J K−1 ) = +3.59 J K−1 .
P�A.� (a) A�er Stage � the volume doubles, thus VB = 2 × VA = 2 × (1.00 dm3 ) =
2.00 dm3 . Assuming V T 3�2 = constant for the adiabatic stages, the
volume a�er Stage � is
(b) Again assuming V T 3�2 = constant for the adiabatic stage, the volume
a�er Stage � can be related to the initial volume
VB
q 1 = q h = nRTh ln � �
VA
= (0.100 mol) × (8.3145 J K−1 mol−1 ) × (373 K) × ln (2)
= +2.14... × 102 J = +215 J .
VD
q 3 = q c = nRTc ln � �
VC
1.59... dm3
= (0.100 mol) × (8.3145 J K−1 mol−1 ) × (273 K) × ln � �
3.19... dm3
= −1.57... × 102 J = −157 J .
Tc 273 K
η =1− =1− = 0.268 = 26.8% .
Th 373 K
the result is the same as the above (the di�erence is due to the use of fewer
signi�cant �gures in the previous calculation).
Using the values of the heat transfer calculated above in equation [�A.�–
��] gives
q c q c 214... J −157... J
+ = +
Th Tc 373 K 273 K
= 0.0 J .
P�A.� (a) Consider a process in which heat dq c is extracted from the cold source at
temperature Tc , and heat dq h is discarded into the hot sink at temperature
Th . �e overall entropy change of such process is
dq c dq h
dS = +
Tc Th
Assume that dq c = −dq and dq h = +dq, where dq is a positive quantity.
It follows that
+dq −dq 1 1
dS = + = dq × � − �
Th Tc Th Tc
Because Th > Tc , the term in parentheses is negative, therefore dS is
negative. �e process is therefore not spontaneous and not allowed by
the Second Law. If work is done on the engine, �dq h � will become greater
than �dq c � and eventually dS will be greater than zero.
(b) Assuming q c = −�q� and q h = �q� + �w� the overall change in entropy is
−�q� �q� + �w�
∆S = +
Tc Th
For the process to be permissible by the Second Law the Clausius inequal-
ity de�ned in [�A.��–��], dS ≥ 0, must be satis�ed. �erefore
−�q� �q� + �w�
+ ≥0
Tc Th
which implies
Th Th
�w� ≥ �q� × � − 1� = �q� × � − 1� .
Tc Tc
P�A.� Suppose two adiabatic paths intersect at point A as shown in the �gure. Two
remote points corresponding to the same temperature on each adiabat, A and
B, are then connected by an isothermal path forming a cycle.
Consider energy changes for each Stage of the cycle. Stage � (A → B) is adiabatic
and, thus, no heat exchange takes place q 1 = 0. �erefore, the total change in
internal energy is ∆U 1 = w 1 + q 1 = w 1 . Stage � (B → C) is an isothermal change
and assuming that the system energy is a function of temperature only (e.g.
ideal gas): ∆U 2 = w 2 + q 2 = 0. Stage � (C → A) is again adiabatic, q 3 = 0, with
∆U 3 = w 3 + q 3 = w 3 . Because the system energy is a function of temperature
only, U B = U C and, thus
∆U 3 = U A − U C = U A − U B = −∆U 1
Pressure, p
Stage �
Stage �
C
B
Stage �
Volume, V
Figure 3.1
Finally, analyse the net work done, w cycle = w 1 + w 2 + w 3 = w 2 , and the net heat
absorbed, q cycle = q 1 + q 2 + q 3 = q 2 , over the cycle. It is apparent that the sole
result of the process is the absorption of heat q 2 and its convertion to work w 2 ,
which directly contradicts the statement of the Second Law by Kelvin, unless
the q 2 = w 2 = 0, i.e. points B and C are the same and correspond to the same
path. �erefore, no two such adiabatic paths exist.
Mercury has quite strong interactions between the atoms, as evidenced by its
cohesiveness, and so its entropy of vaporization is expected to be greater than
that predicted by Trouton’s rule.
76 3 THE SECOND AND THIRD LAWS
Solutions to exercises
E�B.�(a) Two identical blocks must come to their average temperature. �erefore the
�nal temperature is
Tf = 12 (T1 + T2 ) = 1
2
× (50 ○ C + 0 ○ C) = 25 ○ C = 298 K .
Although the above result may seem self-evident, the more detailed explaina-
tion is as follows. �e heat capacity at constant volume is de�ned in [�A.��–
��], C V = (∂U�∂T)V . As shown in Section �A.�(b) on page ��, if the heat
capacity is constant, the internal energy changes linearly with the change in
temperature. �at is ∆U = C V ∆T = C V (Tf − Ti ). For the two blocks at
the initial temperatures of T1 and T2 , the change in internal energy to reach
the �nal temperature Tf is ∆U 1 = C V ,1 (Tf − T1 ) and ∆U 2 = C V ,2 (Tf − T2 ),
respectively. �e blocks of metal are made of the same substance and are of
the same size, therefore C V ,1 = C V ,2 = C V . Because the system is isolated
the total change in internal energy is ∆U = ∆U 1 + ∆U 2 = 0. �is means that
∆U = C V ((Tf − T1 ) − (Tf − T2 )) = C V × (2Tf − (T1 + T2 )) = 0, which implies
that the �nal temperature is Tf = 12 (T1 + T2 ), as stated above.
�e temperature variation of the entropy at constant volume is given by [�B.�–
��], ∆S = C V ln (Tf �Ti ), with C p replaced by C V . Expressed with the speci�c
heat C V ,s = C V �m it becomes
Tf
∆S = mC V ,s ln � �.
Ti
Note that for a solid the internal energy does not change signi�cantly with the
volume or pressure, thus it can be assumed that C V = C p = C. �e entropy
change for each block is found using this expression
Tf
∆S 1 = mC V ,s ln � �
T1
298 K
= (1.00 × 103 g) × (0.385 J K−1 g−1 ) × ln � �
50 K + 273.15 K
= −31.0... J K−1 = −31.0 J K−1 .
Tf
∆S 2 = mC V ,s ln � �
T2
298 K
= (1.00 × 103 g) × (0.385 J K−1 g−1 ) × ln � �
273.15 K
= 33.7... J K−1 = +33.7 J K−1 .
E�B.�(a) Because entropy is a state function, ∆S between the initial and �nal states is
the same irrespective of the path taken. �us the overall process can be broken
down into steps that are easier to evaluate. First consider heating the initial
system at constant pressure to the �nal temperature. �e variation of entropy
with temperature at constant pressure is given by [�B.�–��], S(Tf ) = S(Ti ) +
C p ln (Tf �Ti ). �us the change in entropy, ∆S = S(Tf ) − S(Ti ), of this step is
Tf Tf
∆S 1 = C p ln � � = nC p,m ln � �
Ti Ti
pi
∆S 2 = nR ln � �
pf
Tf pi
∆S = ∆S 1 + ∆S 2 = nC p,m ln � � + nR ln � �
Ti pf
273.15 K + 125 K
= (3.00 mol) × � 52 × 8.3145 J K−1 mol−1 � × ln � �
273.15 K + 25 K
1.00 atm
+ (3.00 mol) × (8.3145 J K−1 mol−1 ) × ln � �
5.00 atm
= (+18.0... J K−1 ) + (−40.1... J K−1 ) = −22.1 J K−1 .
E�B.�(a) Because entropy is a state function, ∆S between the initial and �nal states is
the same irrespective of the path taken. �us the overall process can be broken
down into steps that are easier to evaluate. First consider heating the ice at
constant pressure from the initial temperature to the melting point, Tm . �e
variation of entropy with temperature at constant pressure is given by [�B.�–
��], S(Tf ) = S(Ti ) + C p ln (Tf �Ti ). �us the change in entropy, ∆S = S(Tf ) −
S(Ti ), for this step is
Tm Tm
∆S 1 = C p ln � � = nC p,m (H2 O(s)) ln � �
Ti Ti
Next consider the phase transition from solid to liquid at the melting temper-
ature. �e entropy change of a phase transition is given by [�B.�–��], ∆ trs S =
∆ trs H�Ttrs , thus
−
○
∆ fus H m
∆S 2 = n
Tm
�en the liquid is heated to the boiling temperature, Tb . In analogy to the �rst
step
Tb
∆S 3 = nC p,m (H2 O(l)) ln � �
Tm
78 3 THE SECOND AND THIRD LAWS
Tf
∆S 5 = nC p,m (H2 O(g)) ln � �
Tb
∆S�n = ∆S 1 + ∆S 2 + ∆S 3 + ∆S 4 + ∆S 5
−
○
Tm ∆ fus H m Tb
= C p,m (H2 O(s)) ln � �+ + C p,m (H2 O(l)) ln � �
Ti Tm Tm
−
○
∆ vap H m Tf
+ + C p,m (H2 O(g)) ln � �
Tb Tb
273.15 K
= (37.6 J K−1 mol−1 ) × ln � �
273.15 K − 10.0 K
6.01 × 103 J mol−1
+
273.15 K
273.15 K + 100.0 K
+ (75.3 J K−1 mol−1 ) × ln � �
273.15 K
40.7 × 103 J mol−1
+
273.15 K + 100.0 K
273.15 K + 115.0 K
+ (33.6 J K−1 mol−1 ) × ln � �
273.15 K + 100.0 K
= (+1.40... J K−1 mol−1 ) + (+22.0... J K−1 mol−1 )
+ (+23.4... J K−1 mol−1 ) + (+1.09... × 102 J K−1 mol−1 )
+ (+1.32... J K−1 mol−1 )
= +1.57... × 102 J K−1 mol−1
Hence
10.0 g
∆S = × (+1.57... × 102 J K−1 ) = +87.3 J K−1 .
18.02 g mol−1
E�B.�(a) �e entropy change of a phase transition is given by [�B.�–��], ∆ trs S = ∆ trs H�Ttrs .
As discussed in Section �B.� on page �� because there is no hydrogen bonding
in liquid benzene it is safe to apply Trouton’s rule. �at is ∆ vap S −○ = +85 J K−1 mol−1 .
It follows that
∆ vap H −○ = Tb × ∆ vap S −○
= (273.15 K + 80.1 K) × (+85 J K−1 mol−1 )
= 3.00... × 104 J mol−1 = +30 kJ mol−1 .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 79
348 K
∆S m = S m (348 K) − S m (298 K) = (29.355 J K−1 mol−1 ) × ln � �
298 K
= +4.55 J K−1 mol−1 .
E�B.�(a) As explained in Section �B.� on page �� the temperature variation of the en-
tropy at constant volume is given by
Tf dT
∆S = S(Tf ) − S(Ti ) = � CV
Ti T
Assuming that C V = 32 R, the ideal gas limit, for the temperature range of
interest, the molar entropy at 500 K is given by
3 dT
500 K
S m (500 K) = S m (298 K) + � 2
R
298 K T
500 K
= S m (298 K) + 32 R × ln � �
298 K
= (146.22 J K−1 mol−1 )
500 K
+ ( 32 × 8.3145 J K−1 mol−1 ) × ln � �
298 K
= 153 J K−1 mol−1 .
Solutions to problems
P�B.� Because entropy is a state function, ∆S between the initial and �nal states is
the same irrespective of the path taken. �us the overall process can be broken
down into steps that are easier to evaluate.
80 3 THE SECOND AND THIRD LAWS
First consider heating the water at constant pressure from the initial tempera-
ture T to the melting point. �e variation of the entropy with temperature at
constant pressure is given by [�B.�–��], S(Tf ) = S(Ti ) + C p ln (Tf �Ti ). �us
the change in entropy for this step is
Tm Tm
∆S 1 = C p ln � � = nC p,m (H2 O(l)) ln � �
T T
Next consider the phase transition from liquid to solid at the melting temper-
ature; note that freezing is just the opposite of fusion, thus ∆H 2 = n(−∆ fus H m
−
○
).
�e entropy change of a phase transition is given by [�B.�–��], ∆ trs S = ∆ trs H�Ttrs ,
thus
∆H 2 −∆ fus H m
−
○
∆S 2 = =n
Tm Tm
�e ice is then cooled to the �nal temperature, T. Similarly to the �rst step
T
∆S 3 = nC p,m (H2 O(s)) ln � �
Tm
�erefore the overall entropy change for the system is
∆S = ∆S 1 + ∆S 2 + ∆S 3
Tm −∆ fus H −m
○
T
= nC p,m (H2 O(l)) ln � �+n + nC p,m (H2 O(s)) ln � �
T Tm Tm
−1 273.15 K
= (1.00 mol) × (75.3 J K mol ) × ln �
−1
�
273.15 K − 5.00 K
−6.01 × 103 J mol−1
+ (1.00 mol) ×
273.15 K
273.15 K − 5.00 K
+ (1.00 mol) × (37.6 J K−1 mol−1 ) × ln � �
273.15 K
= (+1.39... J K−1 ) + (−22.0... J K−1 ) + (−0.694... J K−1 )
= −21.3... J K−1 = −21.3 J K−1 .
Consider enthalphy change for the same path. �e variation of enthalpy with
temperature at constant pressure is given by [�B.�b–��], ∆H = C p ∆T. �us for
the �rst and third steps, respectively
∆H 1 = nC p,m (H2 O(l))(Tm − T) and ∆H 3 = nC p,m (H2 O(s))(T − Tm )
At constant pressure the heat released by the system is the enthalpy change of
the system, q = ∆H. Because q sur = −q, the entropy change of the surroundings
is
−q −(−5.82... × 103 J)
∆S sur = =
T 273.15 K − 5.00 K
= +21.7... J K−1 = +21.7 J K−1 .
−∆H 1
∆S sur = = − × (∆H 1 + ∆H 2 + ∆H 3 )
T2 T2
Tb − T2 −∆ vap H m
−
○
T2 − Tb
= − �nC p,m (H2 O(l)) +n + nC p,m (H2 O(g)) �
T2 T2 T2
5.00 K
= −(1.00 mol) × (75.3 J K−1 mol−1 )
273.15 K + 95.0 K
4.07 × 104 J mol−1
− (1.00 mol) ×
273.15 K + 95.0 K
−5.00 K
− (1.00 mol) × (33.6 J K−1 mol−1 ) ×
273.15 K + 95.0 K
= −(+1.02... J K−1 ) − (+1.10... × 102 J K−1 ) − (−0.456... J K−1 )
= −1.11... × 102 J K−1 = −111 J K−1 .
82 3 THE SECOND AND THIRD LAWS
P�B.� Two identical blocks must come to their average temperature. �erefore the
�nal temperature is
T = 12 (Tc + Th )
Although the above result may seem self-evident, the more detailed explaina-
tion is as follows. �e heat capacity at constant volume is de�ned in [�A.��–
��], C V = (∂U�∂T)V . As shown in Section �A.�(b) on page ��, if the heat
capacity is constant, the internal energy changes linearly with the change in
temperature. �at is ∆U = C V ∆T = C V (Tf − Ti ). For the two blocks at
the initial temperatures of Tc and Th , the change in internal energy to reach
the �nal temperature T is ∆U c = C V ,c (T − Tc ) and ∆U h = C V ,h (T − Th ),
respectively. �e blocks of metal are made of the same substance and are of
the same size, therefore C V ,c = C V ,h = C V . Note that for a given solid the
internal energy does not change signi�cantly on the volume or pressure, thus
it can be assumed that C V = C p . Assuming the system is isolated the total
change in internal energy is ∆U = ∆U c + ∆U h = 0. �is means that ∆U =
C p ((T − Tc ) − (T − Th )) = C p × (2T − (T1 + T2 )) = 0, which implies that the
�nal temperature is T = 12 (Tc + Th ), as stated above.
At constant pressure the temperature dependence of the entropy is given by
[�B.�–��],
Tf
∆S = nC p,m ln � �
Ti
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 83
Tf m Tf
∆S = S(Tf ) − S(Ti ) = C p ln � � = � � C p,m ln � �
Ti M Ti
500 g −1 3.71... × 102 K
=� −1 � × (24.4 J K mol ) × ln �
−1
�
63.55 g mol 293 K
= +45.4 J K−1 .
For the second experiment, the initial and �nal states of the metal block is the
same, therefore ∆S = 0 . All the heat is released into surroundings, that is water
bath, which can be assumed to be large enough to retain constant temperature.
�us
q I 2 R∆t
∆S sur = =
Tsur Tsur
(1.00 A)2 × (1.00 × 103 Ω) × (15.0 s)
= = +51.2 J K−1 .
293 K
P�B.� As suggested in the hint, �rst consider heating the folded protein at constant
pressure to from the initial temperature T to that of the transition, Ttrs . �e
variation of entropy with temperature at constant pressure is given by [�B.�–
��], S(Tf ) = S(Ti ) + C p ln (Tf �Ti ). �us the change in molar entropy, ∆S m =
S m (Tf ) − S m (Ti ), of this step is
Ttrs
∆S 1,m = C p,m (folded) ln � �
T
∆ trs H −m
○
∆S 2,m =
Ttrs
T Ttrs
∆S 3,m = C p,m (unfolded) ln � � = −C p,m (unfolded) ln � �.
Ttrs T
Given that C p,m (unfolded) − C p,m (folded) = 6.28 × 103 J K−1 mol−1 , the molar
entropy of unfolding at 25.0 ○ C is thus
P�B.�� (a) Consider a process in which heat �dq� is extracted from the cold source at
temperature Tc , and heat q h = �dq� + �dw� is discarded into the hot sink at
temperature Th . �e overall entropy change of such process is
For the process to be permissible by the Second Law, the Clausius inequal-
ity de�ned in [�A.��–��], dS ≥ 0, must be satis�ed. �erefore
the equality implies the minimum amount of work for which the process
is permissible. Hence it follows that
(b) �e expression in (a) is rearranged to �nd �dw� and the given relation,
dq = CdTc , is used to give
�dq�
�dw� = Th − �dq�
Tc
dTc
�dw� = CTh � � − C�dTc �
Tc
w Tf dTc Tf
� �dw ′ � = CTh � � � − C � �dTc �
0 Ti Tc Ti
which evaluates to
Tf
�w� = CTh �ln � �� − C�Tf − Ti � .
Ti
86 3 THE SECOND AND THIRD LAWS
(d) Assuming constant temperature, for �nite amounts of heat and work, the
expression derrived in (a) becomes
�q� �q� + �w�
=
Tc Th
�is is rearranged to give the work as
Th
�w� = � − 1� × �q�
Tc
�e heat transferred during freezing is equal to the enthalpy of the tran-
sition, which is the opposite of fusion, q = ∆ trs H = (m�M)(−∆ fus H −○ ).
�erefore the work needed is
293 K 250 g
�w� = � − 1� × � × (−6.01 × 103 J K−1 mol−1 )�
273 K 18.02 g mol−1
= 6.10... × 103 J = 6.11 × 103 J .
(e) �e total work is the sum of the two steps described in (c) and (d). �ere-
fore
(f) Assuming no energy losses, power is the total work divided by the time
interval over which the work is done, P = w tot �∆t, hence
Because the entropies of ions in water are values relative to the hydrogen ion in
water, they may be either positive or negative. A positive entropy means that
an ion has a higher molar entropy than H+ in water, and a negative entropy
means that the ion has a lower molar entropy than H+ in water. An ion with
zero entropy in fact has that same entropy as H+ .
Solutions to exercises
E�C.�(a) Consider the chemical equation
1
N (g) + 32 H2 (g)
2 2
�→ NH3 (g)
�e standard reaction entropy is given by [�C.�b–��], ∆ r S −○ = ∑J ν J S m
−
○
(J),
where ν J are singed stoichiometric coe�cients for a given reaction equation.
�erefore, using data from the Resource section
∆ r S −○ = nS m
−
○
(NH3 , (g)) − 32 nS m
−
○
(H2 , (g)) − 12 nS m
−
○
(N2 , (g))
= (1.00 mol) × (192.45 J K−1 mol−1 )
− � 32 × 1.00 mol� × (130.684 J K−1 mol−1 )
− � 12 × 1.00 mol� × (191.61 J K−1 mol−1 )
= −99.38 J K−1 .
E�C.�(a) Assuming that the Debye extrapolation is valid, the constant-pressure molar
heat capacity is C p,m (T) = aT 3 . �e temperature dependence of the entropy
is given by [�C.�a–��], S(T2 ) = S(T1 ) = ∫T12 (C p,m �T)dT. For a given temper-
T
(ii)
∆ r S −○ = 2S m
−
○
(AgBr, (s)) + S m
−
○
(Cl2 , (g)) − 2S m
−
○
(AgCl, (s)) − S m
−
○
(Br2 , (l))
= 2 × (107.1 J K−1 mol−1 ) + (223.07 J K−1 mol−1 )
− 2 × (96.2 J K−1 mol−1 ) − (152.23 J K−1 mol−1 )
= +92.6 J K−1 mol−1 .
(iii)
∆ r S −○ = S m
−
○
(HgCl2 , (s)) − S m
−
○
(Hg, (l)) − S m
−
○
(Cl2 , (g))
= (146.0 J K−1 mol−1 ) − (76.02 J K−1 mol−1 )
− (223.07 J K−1 mol−1 )
= −153.1 J K−1 mol−1 .
Solutions to problems
P�C.� Consider the process of determining the calorimetric entropy from zero to
the temperature of interest. Assuming that the Debye extrapolation is valid,
the constant-pressure molar heat capacity at the lowest temperatures is of a
form C p,m (T) = aT 3 . �e temperature dependence of the entropy is given
by [�C.�a–��], S(T2 ) = S(T1 ) = ∫T12 (C p,m �T)dT. �us for a given (low)
T
T C p,m T aT ′ 3
S m (T) − S m (0) = � dT ′
= � dT ′
0 T′ 0 T′
T a
= a � T ′ dT ′ = T 3 = 13 C p,m (T)
2
0 3
Hence
−
○
Sm (10 K) − S m
−
○
(0) = 1
3
× (4.64 J K−1 mol−1 ) = 1.54... J K−1 mol−1
−
○
�e increase in entropy on raising the temperature to the melting point is S m (234.4 K)−
−1
S m (10 K) = 57.74 J K mol . �e entropy change of a phase transition is
−
○ −1
2322 J mol−1
−
○
∆ fus S m (234.4 K) = = 9.90... J K−1 mol−1
234.4 K
�e �ird-Law standard molar entropy at 298 K is the sum of the above con-
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 89
tributions.
−
○
Sm (298 K) − S m
−
○
(0) = (S m
−
○
(10 K) − S m
−
○
(0)) + (S m
−
○
(234.4 K) − S m
−
○
(10 K))
+ ∆ fus S m
−
○
(234.4 K) + (S m
−
○
(298 K) − S m
−
○
(234.4 K))
= (1.54... J K−1 mol−1 ) + (57.74 J K−1 mol−1 )
+ (9.90... J K−1 mol−1 ) + (6.85 J K−1 mol−1 )
= 76.04 J K−1 mol−1 .
P�C.� (a) Assuming that the Debye extrapolation is valid, the constant-pressure
molar heat capacity is of a form C p,m (T) = aT 3 . �e temperature depen-
dence of the entropy is given by [�C.�a–��], S(T2 ) = S(T1 ) = ∫T12 (C p,m �T)dT.
T
�us for a given (low) temperature T the change in the molar entropy
from zero is
T C p,m T aT ′ 3
S m (T) − S m (0) = � dT ′
= � dT ′
0 T′ 0 T′
T a
= a � T ′ dT ′ = T 3 = 13 C p,m (T)
2
0 3
Hence
−
○
Sm (10 K) − S m
−
○
(0) = 1
3
× (2.8 J K−1 mol−1 ) = 0.933... J K−1 mol−1
= �.�� J K−1 mol−1 .
y = c3 x 3 + c2 x 2 + c1 x + c0
where the best �tted coe�cients c i for the respective temperature ranges
are
ci 10 K to 30 K 30 K to 298 K
c3 +5.0222 × 10−5 −5.2881 × 10−8
c2 −4.3010 × 10−3 +3.5425 × 10−5
c1 +1.2025 × 10−1 −8.1107 × 10−3
c0 −5.4187 × 10−1 +7.5533 × 10−1
�e integral of the �tted functions over the range x i to x f is
xf
I=� c 3 x 3 + c 2 x 2 + c 1 x + c 0 dx
xi
c3 c2 c1
= �x f 4 − x i 4 � + �x f 3 − x i 3 � + �x f 2 − x i 2 � + c 0 (x f − x i )
4 3 2
90 3 THE SECOND AND THIRD LAWS
0.6
(C p,m �T)�(J K−2 mol−1 )
0.4
0.2
0.0
0 50 100 150 200 250 300
T�K
Figure 3.2
Using the appropriate coe�cients and limits the integrals are evaluated to
give the respective changes in entropy
−
○
Sm (30 K) − S m
−
○
(10 K) = 10.0... J K−1 mol−1
−
○
Sm (298 K) − S m
−
○
(30 K) = 53.8... J K−1 mol−1
−
○
Sm (298 K) − S m
−
○
(10 K) = (10.0... J K−1 mol−1 ) + (53.8... J K−1 mol−1 )
= 63.9... J K−1 mol−1 = ��.� J K−1 mol−1 .
(c) �e standard �ird-Law entropy at 298 K is the sum of the above calcu-
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 91
∑J ν J C −p,m
○
(J). For the reaction at 298 K
∆ r C −p○ = C −p,m
○
(CO, (g)) + C −p,m
○
(H2 O, (g))
− C −p,m
○
(CO2 , (g)) − C −p,m
○
(H2 , (g))
= (29.14 J K−1 mol−1 ) + (33.58 J K−1 mol−1 )
− (37.11 J K−1 mol−1 ) − (28.824 J K−1 mol−1 )
= −3.21... J K−1 mol−1
92 3 THE SECOND AND THIRD LAWS
Assuming that ∆ r C −p○ is constant over the temperature range involved, the stan-
dard entropy and enthalpy changes of the reaction is given by, respectively,
[�C.�b–��], ∆ r S −○ (T2 ) = ∆ r S −○ (T1 )+∆ r C −p○ ln(T2 �T1 ), and [�C.�d–��], ∆ r H −○ (T2 ) =
∆ r H −○ (T1 ) + ∆ r C −p○ (T2 − T1 ).
398 K
∆ r S −○ (398 K) = ∆ r S −○ (298 K) + ∆ r C −p○ × ln � �
298 K
= (+42.0... J K−1 mol−1 )
398
+ (−3.21... J K−1 mol−1 ) × ln � �
298
= +41.15 J K−1 mol−1 .
P�C.� Assuming that the Debye extrapolation is valid, the constant-pressure molar
heat capacity is of a form C p,m (T) = aT 3 . �e temperature dependence of the
entropy is given by [�C.�a–��], S(T2 ) = S(T1 ) = ∫T12 (C p,m �T)dT. �us for a
T
given (low) temperature T the change in the molar entropy from zero is
C p,m
T T aT ′ 3
S m (T) − S m (0) = � dT ′
= � dT ′
0 T′ 0 T′
T a
= a � T ′ dT ′ = T 3 = 13 C p,m (T)
2
0 3
Hence
−
○
Sm (14.14 K) − S m
−
○
(0) = 1
3
× (9.492 J K−1 mol−1 ) = 3.16... J K−1 mol−1
1.2
(C p,m �T)�(J K−2 mol−1 )
1.0
0.8
0.6
where the best �tted coe�cients c i for the respective temperature ranges are
−
○
Sm (300 K) − S m
−
○
(0) = (3.16... J K−1 mol−1 ) + (29.6... J K−1 mol−1 )
+ (2.11... × 102 J K−1 mol−1 )
= ���.� J K−1 mol−1 .
P�C.� (a) Given the expression for the constant-pressure molar heat capacity, C p,m (T) =
aT 3 + bT, consider C p,m �T.
C p,m aT 3 + bT
= = aT 2 + b
T T
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 95
T�K C p,m �(J K−1 mol−1 ) T 2 �(K2 ) (C p,m �T)�(J K−2 mol−1 )
0.20 0.437 0.040 2.185 0
0.25 0.560 0.063 2.240 0
0.30 0.693 0.090 2.310 0
0.35 0.838 0.123 2.394 3
0.40 0.996 0.160 2.490 0
0.45 1.170 0.203 2.600 0
0.50 1.361 0.250 2.722 0
0.55 1.572 0.303 2.858 2
3.0
(C p,m �T)�(J K−2 mol−1 )
2.8
2.6
2.4
2.2
2.0
0.05 0.10 0.15 0.20 0.25 0.30 0.35
T�K
Figure 3.3
T C p,m T
S m (T) = S m (0) + � ′
= (0) + � (aT ′ + b) dT ′
2
dT S
T′
m
0 0
a
= S m (0) + T 3 + bT .
3
96 3 THE SECOND AND THIRD LAWS
(d) Assuming that the expression derived above can be extrapolated to 2.0 K
Both the Helmholtz and Gibbs energies refer to properties of the system alone.
However, because of the way they are de�ned these quantities e�ectively allow
the entropy change of the system plus surroundings to be evaluated. For exam-
ple, at constant volume and temperature the change in the Helmholtz energy
is expressed in terms of the internal energy change and the entropy change
of the system: dA = dU − TdS. If this expression is divided by −T to give
−dA�T = −dU�T + dS the two terms on the right can bothe be identi�ed as
entropy changes.
It is also possible to express the criterion for spontaneity in terms of the change
in H, U or S for the system. For example, as shown in Topic �D, dS U ,V ≥ 0.
However, the variables which are being held constant (here U and V ) do not
correspond to such easily realizable conditions such as constant temperature
and volume (or pressure) so such criteria are less applicable to chemical sys-
tems.
Solutions to exercises
E�D.�(a) �e maximum non-expansion work is equal to the Gibbs free energy as ex-
plained in Section �D.�(e) on page ��. �e standard reaction Gibbs energy is
given by [�D.��b–��], ∆ r G −○ = ∑J ν J ∆ f G −○ (J), where ν J are the signed stoichio-
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 97
metric numbers. For the reaction CH4 (g) + 3O2 (g) �→ CO2 (g) + 2H2 O(l)
∆ r G −○ = ∆ f G −○ (CO2 ,(g)) + 2∆ f G −○ (H2 O ,(l)) − ∆ f H −○ (CH4 ,(g))
− 3∆ f H −○ (O2 ,(g))
= (−394.36 kJ mol−1 ) + 2 × (−237.13 kJ mol−1 )
− (−50.72 kJ mol−1 ) − 3 × 0
= −817.90 kJ mol−1 .
(i)
∆ r G −○ = 2∆ f G −○ (CH3 COOH ,(l)) − 2∆ f G −○ (CH3 CHO ,(g)) − ∆ f G −○ (O2 ,(g))
= 2 × (−389.9 kJ mol−1 ) − 2 × (−128.86 kJ mol−1 ) − 0
= −522.1 kJ mol−1 .
(ii)
∆ r G −○ = 2∆ f G −○ (AgBr ,(s)) + ∆ f G −○ (Cl2 ,(g)) − 2∆ f G −○ (AgCl ,(s))
− ∆ f G −○ (Br2 ,(l))
= 2 × (−96.90 kJ mol−1 ) + 0 − 2 × (−109.79 kJ mol−1 ) − 0
= +25.78 kJ mol−1 .
(iii)
∆ r G −○ = ∆ f G −○ (HgCl2 ,(s)) − ∆ f G −○ (Hg ,(l)) − ∆ f G −○ (Cl2 ,(g))
= (−178.6 kJ mol−1 ) − 0 − 0 = −178.6 kJ mol−1 .
(i)
Given the result for the previous execise, ∆ r S −○ = −386.1 J K−1 mol−1
(ii)
Given the result for the previous execise, ∆ r S −○ = +92.6 J K−1 mol−1
(iii)
Given the result for the previous execise, ∆ r S −○ = −153.1 J K−1 mol−1 .
∆ r S −○ = 2S m
−
○
(I2 ,(s)) + 2S m
−
○
(H2 O ,(l)) − 4S m
−
○
(HI ,(g)) − S m
−
○
(O2 ,(g))
= 2 × (116.135 J K−1 mol−1 ) + 2 × (69.91 J K−1 mol−1 )
− 4 × (206.59 J K−1 mol−1 ) − (205.138 J K−1 mol−1 )
= −659.40... J K−1 mol−1
Solutions to problems
P�D.� (a) From the perfect gas law, pV = nRT, the �nal pressure is
n B RTB (2.00 mol) × (8.3145 × 10−2 dm3 bar K−1 mol−1 ) × (300 K)
p= =
VB,f 1.00 dm3
= 49.8... bar = 49.9 bar .
(b) Taking the hint, �rst consider the heating at constant volume. �e entropy
dependence on temperature at constant volume is given by [�B.�–��],
∆S = nC V ,m ln (Tf �Ti ), with C p replaced by nC V ,m . �e volume is then
allowed to expand to the �nal. �e entropy change for the isothermal
expansion of a perfect gas is given by [�B.�–��], ∆S = nR ln(Vf �Vi ).
�erefore the total change in the entropy for the gas in Section A is
TA,f VA,f
∆S A = nC V ,m ln � � + nR ln � �
TA VA
900 K
= (2.00 mol) × (20.0 J K−1 mol−1 ) × ln � �
300 K
3.00 dm3
+ (2.00 mol) × (8.3145 J K−1 mol−1 ) × ln � �
2.00 dm3
= +50.6... J K−1 = +50.7 J K−1 .
(c) Section B is kept at the constant temperature throughout the process, thus
only the change in the volume needs to be considered
VB,f
∆S B = nR ln � �
VB
1.00 dm3
= (2.00 mol) × (8.3145 J K−1 mol−1 ) × ln � �
2.00 dm3
= −11.5... J K−1 = −11.5 J K−1 .
∆A B = ∆U B − TB ∆S B
= 0 − (300 K) × (−11.5... J K−1 ) = +3.46 × 103 J .
Because ∆T is not zero and S is not given for the Section A, the equivalent
expression cannot be evaluated.
(f) Because the process is reversible, the total ∆A = ∆A A + ∆A B = 0 . �is
implies ∆A A = −∆A B .
∆ ec G −○ (I)
= −E a (I)
∆ ion G −○ (H) = I(H)
H+ (g) + I – (g)
H (g) + I (g)
∆ f G −○ (H (g))
+∆ f G −○ (I (g)) ∆ solv G −○ (H+ )
+∆ solv G −○ (I− )
1
H
2 �
(g) + 12 I� (g)
−∆ f G −○ (H+ (aq))
−∆ f G −○ (I− (aq))
H+ (aq) + I – (aq)
Figure 3.4
�e sum of the Gibbs energies for all the steps around a closed cycle is zero.
∆ r S 1−○ = S m
−
○
(Li+ ,(g)) + S m
−
○
(F− ,(g)) − S m
−
○
(LiF ,(s))
= (133 J K−1 mol−1 ) + (145 J K−1 mol−1 ) − (35.6 J K−1 mol−1 )
= 242.4 J K−1 mol−1
102 3 THE SECOND AND THIRD LAWS
And so
For the second step ∆ r G 2−○ = ∆ solv G −○ (Li+ ) + ∆ solv G −○ (F− ). �e Gibbs en-
ergy of solvation in water is given by Born equation [�D.��b–��], ∆ solv G −○ =
−(z i 2 �[r i �pm]) × 6.86 × 104 kJ mol−1 , thus
(+1)2 (−1)2
∆ r G 2−○ = − � + � × 6.86 × 104 kJ mol−1
[r(Li )�pm] [r(F− )�pm]
+
1 1
= −� + � × 6.86 × 104 kJ mol−1 = −9.61... × 102 kJ mol−1
127 163
= −961 kJ mol−1
Solutions to exercises
E�E.�(a) As explained in Section �E.�(c) on page ���, the change in Gibbs energy of a
phase transition varies with pressure as ∆ trs G m (p f ) = ∆ trs G m (p i ) ∫ p i f ∆ trs Vm dp.
p
Assuming that ∆ trs Vm changes little over the range of pressures considered
E�E.�(a) �e Gibbs energy dependence on pressure for a perfect gas is given by [�E.��–
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 103
pf
∆G m = RT ln � �
pi
100.0 atm
= (8.3145 J K−1 mol−1 ) × (298 K) × ln � �
1.0 atm
= +11.4... × 103 J mol−1 = +11 kJ mol−1 .
Vi
∆G = nRT ln � �
Vf
42 cm3
= (2.5 × 10−3 mol) × (8.3145 J K−1 mol−1 ) × (300 K) × ln � �
600 cm3
= −17 J .
E�E.�(a) �e variation of the Gibbs energy with pressure is given by [�E.�–���], (∂G�∂T) p =
−S. �e change in entropy is thus
∂G f ∂G i ∂(G f − G i )
∆S = S f − S i = − � � +� � = −� �
∂T p ∂T p ∂T p
E�E.�(a) �e Gibbs-Helmholtz relation for the change in Gibbs energy is given by [�E.��–
���], (∂[∆G�T]�∂T) p = −∆H�T 2 . Expressing for the change in enthalpy gives
∆G = n[G m (p f ) − G m (p i )] = (p f − p i )nVm = (p f − p i )V
1.01325 × 105 Pa
= (100 atm − 1.0 atm) × × (1.0 × 10−3 m3 )
1 atm
= +1.00... × 104 J = +10 kJ .
104 3 THE SECOND AND THIRD LAWS
Solutions to problems
P�E.� (a) �e Gibbs-Helmholtz relation for the change in Gibbs energy is given by
[�E.��–���], (∂[∆G�T]�∂T) p = −∆H�T 2 . Integrating the equation be-
tween the temperatures T1 and T2 and assuming that ∆H is temperature
independent gives
T2 ∂ ∆G(T) T2 1
� � � dT = ∆H � − 2 dT
T1 ∂T T p
T1 T
∆G(T2 ) ∆G(T1 ) 1 1
− = ∆H � − �
T2 T1 T2 T1
�erefore
∆G(T2 ) ∆G(T1 ) 1 1
= + ∆H � − �
T2 T1 T2 T1
(b) �e standard reaction entropy is given by [�C.�b–��], ∆ r G −○ = ∑J ν J ∆ f G −○ (J),
where ν J are the signed stoichiometric numbers.
∆ r G −○ (298 K) = 2∆ f G −○ (CO2 (g)) − 2∆ f G −○ (CO (g)) − ∆ f G −○ (O2 (g))
= 2 × (−394.36 kJ mol−1 ) − 2 × (−137.17 kJ mol−1 ) − 0
= −514.38 kJ mol−1 .
Similarly, the standard reaction enthalpy is given by [�C.�b–��], ∆ r H −○ =
∑J ν J ∆ f H −○ (J).
∆ r H −○ (298 K) = 2∆ f H −○ (CO2 (g)) − 2∆ f H −○ (CO (g)) − ∆ f H −○ (O2 (g))
= 2 × (−393.51 kJ mol−1 ) − 2 × (−110.53 kJ mol−1 ) − 0
= −565.96 kJ mol−1 .
P�E.� �e given expression for the reaction Gibbs energy dependence on temperature
is rearranged for ∆G(T2 ) and becomes
T2 T2
∆ r G −○ (T2 ) = ∆ r G −○ (T1 ) + ∆ r H −○ �1 − �
T1 T1
Hence at 37 ○ C = 310 K
273.15 K + 37 K
∆ r G −○ (310 K) = (−6333 kJ mol−1 )
298 K
−1 273.15 K + 37 K
+ (−5797 kJ mol ) �1 − �
298 K
= −6355 kJ mol−1
�erefore the result is an extra 22 kJ mol−1 of energy that is available for non-
expansion work.
dH = dU + d(pV ) = dU + V dp + pdV
∂H ∂H
� � =T and � � =V
∂S p ∂p S
∂ ∂H ∂ ∂H
� � � � =� � � �
∂p ∂S p S ∂S ∂p S p
�erefore
∂T ∂V
� � =� �
∂p S ∂S p
It follows that
∂p ∂S
� � =� �
∂T V ∂V T
For the Gibbs energy, the above derived result for dH is used
It follows that
∂V ∂S
� � = −� � .
∂T p ∂p T
P�E.� (a) Assuming that a = 0 and b ≠ 0, the van der Waals equation becomes
p = RT�(Vm − b). �e molar volume is thus
RT
Vm = +b
p
Consider the exact di�erential of the molar Gibbs energy at constant tem-
perature, dG m = (∂G m �∂p)T dp. Integrating this gives
G m,f pf ∂G m pf p f RT
� dG m = � � � dp = � Vm dp = � � + b� dp
G m,i pi ∂p T pi pi p
�erefore
pf
G m (p f ) = G m (p i ) + RT ln � � + b(p f − p i )
pi
pVm2 − RT Vm + a = 0
Because the van der Waals equation is a correction to the ideal gas, the
result should be approximately similar. Considering 4pa�(RT)2 � 1, it
is obvious that only a positive root reproduces the perfect gas and hence
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 107
�erefore
pf a pf
G m (p f ) = G m (p i ) + RT ln � �− ln � �
pi RT pi
150.0
∆G m (p)�atm dm3 mol−1
100.0
100.0
90.0
perfect gas
85.0
vdW, repulsive
vdW, attractive
80.0
25 30 35 40 45 50
−
○
p�p
Figure 3.6
where the value of b is taken from the tables in the Resource section; note
the conversion of the molar volumes to dm3 mol−1 so as to match the
units of b.
(b) �e variation of entropy with temperature at constant volume and pres-
sure are given by
20
atoms
linear rotors
15 non-linear rotors
∆S m �R
10
0
0 1 2 3 4 5
ln(T2 �T1 )
Figure 3.7
20
atoms
linear rotors
15 non-linear rotors
∆S m �R
10
0
0 1 2 3 4 5
ln(T2 �T1 )
Figure 3.8
this is integrated for the particular form of the heat capacity suggested
CTf Tf a c
∆S = � dT = � � + b + 3 � dT
Ti T Ti T T
Tf 1 1
= a ln + b(Tf − Tf ) − 12 c � 2 − 2 �
Ti ��� � � � � � � � � �� � � � � � � � � � �� Tf Ti
��� � � � � � � � term � ��� � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � ��
term � term �
perature range; from the plots it is clear that the �rst term makes by far
the greatest contribution. Terms � and � both result in an increase in the
entropy with temperature, but term � will make a negative contribution
to the entropy change if c < 0, which is commonly the case.
(term �)�(J K−1 mol−1 )
20 20
0 0
20 40 60 80 0.00 0.02 0.04
−1
a�(J K −1
mol ) b�(J K −2
mol )
−1
(term �)�(J K−1 mol−1 )
−2
−4
−10 −5 0
c�(10 J K mol )
5 −1
Figure 3.9
where the critical values of p, V , and T are given in terms of the van der
Waals parameters by
p c = a�27b 2 Vc = 3b Tc = 8a�27bR
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 111
2.0
1.5
∆G�nRT
1.0
0.5
0.0
1 2 3 4 5
p f �p i
Figure 3.10
1.2
Tr = 1
Tr = 2
Tr = 3
1.0
�
0.8
0.6
1.0 1.5 2.0 2.5 3.0
Vr
Figure 3.11
D�A.� Chemical potential is the single function that governs phase stability. �e phase
whose chemical potential is least under a set of given conditions is the most
stable. Conditions under which two or more phases have equal chemical poten-
tials are conditions under which those phases are in equilibrium. Understand-
ing how chemical potential varies with physical conditions such as temperature,
pressure, and composition makes it possible to compute chemical potentials for
various phases and to map out the conditions for stability of those phases and
for equilibrium between them.
Solutions to exercises
E�A.�(a) Use the phase rule [�A.�–���], F = C − P + 2, with C = 1 (one component).
Inserting P = 1 gives F = 1 − 1 + 2 = 2. �e condition P = 1 therefore
represents an area . An area has F = 2 because it is possible to vary pressure
and temperature independently (within limits) and stay within the area. P = 1
indicates that a single phase is present, so this result con�rms that a single phase
is represented by an area in a phase diagram.
E�A.�(a) (i) ��� K and �.� atm lies on the boundary between solid and gas phases.
Two phases , solid and gas, would therefore be present in equilibrium
under these conditions.
114 4 PHYSICAL TRANSFORMATIONS OF PURE SUBSTANCES
(ii) ��� K and � atm lies in the vapour region, so only one phase , vapour,
will be present.
(iii) ��� K is greater than the critical temperature, which means that there
is no distinction between gas and liquid. �erefore only one phase (a
supercritical �uid) will be present at all pressures.
E�A.�(a) In a phase diagram, a single phase is represented by an area, while a line repre-
sents a phase boundary where two phases coexist in equilibrium. Point a lies
within an area and therefore only one phase is present. Points b and d each
lie on the boundary between two areas, and therefore in each case two phases
are present. Point c lies at the intersection of three phase boundaries, where
three phases are present in equilibrium.
dG = (µ 2 − µ 1 )dn
E�A.�(a) Use the phase rule [�A.�–���], F = C − P + 2, with C = 2 (for two components).
Rearranging for the number of phases gives
Solutions to problems
P�A.� (a) ��� K and � atm lies in the solid region of the phase diagram, so initially
only solid carbon dioxide (dry ice) will be present. When the temper-
ature reaches ���.� K, the sublimation point of CO2 at � atm, solid and
gas phases will be present in equilibrium. Above this temperature only
gaseous CO2 is present.
(b) ��� K and �� atm lies in the solid region of the phase diagram, so again
CO2 will initially be a solid. On heating, a point is reached at which the
solid melts; at this temperature solid and liquid phases are both present in
equilibrium. Above this temperature only a liquid phase is present until
the boiling temperature is reached, at which point liquid and gas will be in
equilibrium. Above this temperature, only the gas phase will be present.
P�A.� A schematic phase diagram is shown in Fig �.�. Note that in reality the phase
boundaries may be curved rather than straight. �ere are two triple points
which are marked with dots.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 115
⇧
s Triple points
⇧ δ
s(
β ⇧
p ⇧ (((
s
⌅ α
γ ⌅
⌅
T
Figure 4.1
Solutions to exercises
E�B.�(a) �e relationship between pressure and temperature along the solid–liquid bound-
ary is given by [�B.�–���], p = p∗ + (∆ fus H�T ∗ ∆ fus V )(T − T ∗ ). In this case
p∗ = 1 atm (corresponding to the normal melting point, T ∗ = 273.15 K) and
p = 1 bar (corresponding to the standard melting point). Rearranging for
(T − T ∗ ), the di�erence in melting points, gives
T ∗ ∆ fus V
(T − T ∗ ) = (p − p∗ )
∆ fus H
1.01325 × 105 Pa
= �1 × 105 Pa − 1 atm × �
1 atm
(273.15 K) × (−1.6 × 10−6 m3 mol−1 )
× = 9.6 × 10−5 K
6.008 × 103 J mol−1
�is result shows that the standard melting point of ice is slightly higher than
the normal melting point, but the di�erence is negligibly small for most pur-
poses.
116 4 PHYSICAL TRANSFORMATIONS OF PURE SUBSTANCES
E�B.�(a) Since 1 W = 1 J s−1 , the rate at which energy is absorbed is (1.2 kW m−2 ) ×
(50 m2 ) = 60 kJ s−1 . �e rate of vaporization is then
rate of energy absorption 60 kJ s−1
= = 1.36... mol s−1
∆ vap H 44 kJ mol−1
Multiplication by the molar mass of water gives the rate of loss of water as
(1.36... mol s−1 ) × (18.0158 g mol−1 ) = �� g s−1 .
E�B.�(a) �e perfect gas equation [�A.�–�], pV = nRT, is used to calculate the amount
as n = pV �RT. V is the volume of the laboratory (75 m3 ) and p is the vapour
pressure. �e mass is found from m = nM, where M is the molar mass; hence
m = pV M�RT
pV M (3.2 × 103 Pa) × (75 m3 ) × (18.0158 g mol−1 )
Water: m= = = �.� kg
RT (8.3145 J K−1 mol−1 ) × ([25 + 273.15] K)
pV M (13.1 × 103 Pa)×(75 m3 )×(78.1074 g mol−1 )
Benzene: m= = = �� kg
RT (8.3145 J K−1 mol−1 ) × ([25 + 273.15] K)
pV M (0.23 Pa) × (75 m3 ) × (200.59 g mol−1 )
Mercury: m= = = �.� g
RT (8.3145 J K−1 mol−1 ) × ([25 + 273.15] K)
1 Pa = 1 kg m−1 s−2 and 1 J = 1 kg m2 s−2 have been used. Note that an typically
sized bottle of benzene (containing less than �� kg of benzene) would evaporate
completely before saturating the air of the laboratory with benzene vapour.
E�B.�(a) (i) �e integrated form of the Clausius–Clapeyron equation [�B.��–���] is
p ∆ vap H 1 1
ln ∗
=− � − ∗�
p R T T
Rearranging for ∆ vap H and substituting in the numbers, taking p∗ , T ∗ at
��.� ○ C and p,T at ���.� ○ C, gives
1 1 −1 p
∆ vap H = −R � − ∗ � ln ∗
T T p
−1
−1 1 1
= −(8.3145 J K −1
mol ) × � − �
[119.3 + 273.15] K [85.5 + 273.15] K
5.3 kPa
× ln � � = 4.86... × 104 J mol−1 = �� kJ mol−1
1.3 kPa
(ii) �e integrated form of the Clausius–Clapeyron equation is now rearranged
for T. Substituting in p = 1 atm, or 1.01325 × 105 Pa, corresponding to
the normal boiling point, together with the value of ∆ vap H from above
and the same values for p∗ , T ∗ as before, gives
−1
1 R p
T =� ∗
− ln ∗ �
T ∆ vap H p
−1
1 8.3145 J K−1 mol−1 1.01325 × 105 Pa
=� − × ln �
[85.5 + 273.15] K 4.86... × 104 1.3 × 103 Pa
= 4.89... × 102 K = 4.9 × 102 K or 2.2 × 102 ○ C
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 117
E�B.��(a) �e relationship between pressure and temperature along the solid–liquid bound-
ary is given by [�B.�–���], p = p∗ + (∆ fus H�T ∗ ∆ fus V )(T − T ∗ ), which is
rearranged to give ∆ fus H. In this case p∗ = 1.00 atm, T ∗ = 350.75 K, p =
100 atm and T = 351.26 K.
p − p∗ ∗
∆ fus H = T ∆ fus V
T − T∗
([100 − 1] atm) × (1.01325 × 105 Pa�1 atm)
= × (350.75 K)
([351.26 − 350.75] K)
× ([163.3 − 161.0] × 10−6 m3 mol−1 )
= 1.58... × 104 J mol−1 = ��.� kJ mol−1
does not matter that the pressure is given in units of Torr because only
the slope of ln p is required.
d ln p d log p d 1780 K
∆ vap H = RT 2 = RT 2 ln 10 = RT 2 ln 10 �7.960 − �
dT dT dT T
1780 K
= RT 2 ln 10 � � = (1780 K)R ln 10
T2
= (1780 K) × (8.3145 J K−1 mol−1 ) × ln 10 = ��.�� kJ mol−1
(ii) �e normal boiling point refers to the temperature at which the vapour
pressure is � atm which is ��� Torr. �e given expression, log(p�Torr) =
7.960 − (1780K)�T, is rearranged for T and a pressure of ��� Torr is
substituted into it to give
1780 K 1780 K
T= = = ���.� K or ��.�� ○ C
7.960 − log(p�Torr) 7.960 − log 760
Note that this temperature lies outside the range 10 ○ C to 30 ○ C for which
the expression for log(p�Torr) is known to be valid, and is therefore an
estimate.
E�B.��(a) �e relationship between pressure and temperature along the solid–liquid bound-
ary is given by [�B.�–���], p = p∗ + (∆ fus H�T ∗ ∆ fus V )(T − T ∗ ). �e value of
∆ fus V is found by using Vm = M�ρ where M is the molar mass and ρ is the
mass density:
M M
∆ fus V = Vm (l) − Vm (s) = −
ρ(l) ρ(s)
78.1074 g mol−1 78.1074 g mol−1
= − = 1.19... × 10−6 m3 mol−1
0.879 × 106 g m−3 0.891 × 106 g m−3
T ∗ ∆ fus V
T = T ∗ + (p − p∗ )
∆ fus H
1.01325 × 105 Pa
= ([5.5 + 273.15] K) + �([1000 − 1] atm) × �
1 atm
([5.5 + 273.15] K) × (1.19... × 10−6 m3 )
× = 2.8 × 102 K or �.� ○ C
10.59 × 103 J mol−1
Solutions to problems
P�B.� �e work done in expanding against a constant external pressure is given by
equation [�A.�–��], w = −p ex ∆V . Because the molar volume of a gas is so
much greater the molar volume of a liquid, ∆ vap V ≈ Vm (g). In addition, if
the gas behaves perfectly, Vm = RT�p (from the perfect gas law, [�A.�–�]) with
p = p ex as the gas expands against constant external pressure. �e work of
120 4 PHYSICAL TRANSFORMATIONS OF PURE SUBSTANCES
expansion is therefore
RT
w = −p ex × = −RT = −(8.3145 J K−1 mol−1 ) × ([100 + 273.15] K)
p ex
= −3.10... × 103 J mol−1 = −3.10 kJ mol−1
�e negative sign indicates that the system has done work on the surround-
ings, so the internal energy of the system falls. �e fraction of the enthalpy of
vaporization spent on expanding the vapour is
3.10... kJ mol−1
× 100 % = �.�� %
40.7 kJ mol−1
∆ vap H 1 1
p = p∗ exp �− � − ∗ ��
R T T
= (1 atm)
20.25 × 103 J mol−1 1 1
× exp �− −1 × � − ��
−1
8.3145 J K mol (40 + 273.15) K (−29.2 + 273.15) K
= �.�� atm or ��� kPa
P�B.� (a) From the variation of chemical potential with temperature (at constant
pressure) [�B.�a–���], (∂µ�∂T) p = −S m , the slope of the chemical po-
tential against temperature is equal to the negative of the molar entropy.
�e di�erence in slope on either side of the normal freezing point of water
is therefore
∂µ(l) ∂µ(s)
� � −� � = −S m (l) − (−S m (s))
∂T p ∂T p
(b) In a similar way, the di�erence in slope on either side of the normal boil-
ing point of water is
∂µ(g) ∂µ(l)
� � −� � = −S m (g) − (−S m (l))
∂T p ∂T p
p = ρgd + 1 atm
10−3 kg 106 cm3
= �13.6 g cm−3 × × � × (9.81 m s−2 ) × (10 m)
1g 1 m3
+ (1.01325 × 105 Pa) = 1.44... × 106 Pa
To �nd the freezing point, use [�B.�–���], p = p∗ +(∆ fus H�T ∗ ∆ fus V )(T −T ∗ ).
Rearranging for T gives
T ∗ ∆ fus V
T = T∗ + (p − p∗ )
∆ fus H
(234.3 K) × (0.517 × 10−6 m3 mol−1 )
= (234.3 K) +
2.292 × 103 J mol−1
× ([1.44... × 10 − 1.01325 × 105 ] Pa) = ���.� K
6
Note that this is not a very large di�erence from the normal freezing point,
re�ecting the fact that the slope of the solid-liquid boundary is generally very
steep compared to the liquid-vapour boundary. Large changes in pressure are
therefore needed to bring about signi�cant changes in freezing point.
p ∆ vap H 1 ∆ vap H
ln ∗
=− +
p R T RT ∗
�is implies that a plot of ln(p�p∗ ) against 1�T should be a straight line of slope
−∆ vap H�R and intercept ∆ vap H�RT ∗ ; such a plot is shown in Fig. �.�. If p∗ is
taken to be 1 atm, or 101.325 kPa, then T ∗ corresponds to the normal boiling
point which can then be obtained from the intercept.
122 4 PHYSICAL TRANSFORMATIONS OF PURE SUBSTANCES
0
ln(p�p∗ )
−2
−4
0.0025 0.0030 0.0035
T −1 �K−1
Figure 4.2
�e values of ∆ vap H and T ∗ are obtained from the slope and intercept respec-
tively:
P�B.�� (a) �e Clapeyron equation [�B.�a–���] is dp�dT = ∆ trs S�∆ trs V . For subli-
mation, and with ∆ trs S = ∆ trs H�T this becomes
dp ∆ sub H
=
dT T∆ sub V
Since the molar volume of a gas is much greater than that of a solid, ∆ sub V
can be approximated as ∆ sub V = Vm (g) − Vm (s) ≈ Vm (g), and if the gas
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 123
∆ sub H 1
ln p = − + constant
R T
�is implies that a plot of ln p against 1�T should be a straight line of
slope −∆ sub H�R; such a plot is shown in Fig. �.�.
4.5
4.0
ln(p�Pa)
3.5
3.0
2.5
P�B.�� (a) If the mass of the liquid decreases by m, then the amount in moles of
vapour formed is n vap = m�M. �e amount in moles of the input gas
is given by n gas = PV �RT (from the perfect gas equation) so the mole
fraction of the vapour is
n vap n vap m�M mRT
x vap = = = =
n tot n vap + n gas PV �RT + m�M MPV + mRT
(b) If the total pressure remains at P, the partial pressure of the vapour is
mRT mRTP
p = x vap × P = ×P =
MPV + mRT MPV + mRT
(c) Dividing top and bottom of this expression by MPV gives
AmP RT
p= where A=
1 + Am MPV
(d) For geraniol, noting that P = 760 Torr = 1.01325 × 105 Pa,
50
40
30
p�bar
Solid Liquid
20
10
Vapour
0
0 100 200 300 400 500 600 700
T�K
Figure 4.4
(d) Use the Clapeyron equation for the liquid-vapour boundary [�B.�–���]:
dp ∆ vap H dp
= which rearranges to ∆ vap H = T∆ vap V ×
dT T∆ vap V dT
126 4 PHYSICAL TRANSFORMATIONS OF PURE SUBSTANCES
− 8.04678(1 − y)0.70 �
105 Pa 10.418
= �
593.95 K (0.645...)2
− 15.996 + 28.030(0.645...) − 15.0360(0.645...)2
�en
dp
∆ vap H = T∆ vap V ×
dT
= (383.54 K) × ([30.3 − 0.12] × 10−3 m3 mol−1 ) × (2.84... × 103 Pa K−1 )
= 33.0 kJ mol−1
4.0
p�MPa
2.0
0.0
100 120 140 160 180 200
T�K
Figure 4.5
d(p�MPa)
= (14.967 × 10−3 K−3 )T 2 − (2.904 × 10−3 K−2 )T + (0.1461 K−1 )
dT
= (14.967 × 10−3 K−3 ) × (1.11... × 102 K)2
− (2.904 × 10−3 K−2 ) × (1.11... × 102 K) + (0.1461 K−1 )
= 8.57... × 10−3 K−1
so that
dp
= 8.57... × 10−3 MPa K−1 = 8.57... × 103 Pa K−1
dT
�erefore
dp
∆ vap H = T∆ vap V ×
dT
10−3 m3
= (1.11... × 102 K) × ([8.89 − 3.80 × 10−2 ] dm3 mol−1 ) ×
1 dm3
× (8.57 × 103 Pa K−1 ) = 8.49 kJ mol−1
�is is used in equation [�B.�–���] together with the value of ∆ fus H quoted.
Taking p∗ and T ∗ as corresponding to the triple point, p∗ = 36 Torr = 4.80 kPa
and T ∗ = 5.50 ○ C = 278.65 K gives the equation of the solid-liquid boundary
as
�is takes the form of a steep straight line with a positive gradient extending
upwards from the triple point. �is is plotted in Fig. �.�. �e line is only drawn
for T ≥ T ∗ (p ≥ p∗ ) because the liquid does not exist below the triple point.
�e relationship between p and T along the liquid-vapour boundary is given by
equation [�B.��–���]; p∗ and T ∗ are again taken as corresponding to the triple
point.
∆ vap H 1 1
p = p∗ exp �− � − ∗ ��
R T T
30.8 × 103 J mol−1 1 1
= (4.80 × 103 Pa) × exp �− � − ��
8.3145 J K−1 mol−1 T 278.65 K
1 1
or (p�kPa) = 4.80 × exp �−3.70 × 103 � − ��
T�K 278.65
�is equation is also plotted in Fig. �.�, again only for values of T in the range
T ≥ 278.65 K since the liquid does not exist below this temperature.
�e relationship between p and T along the solid-vapour boundary is given by
an equation that is analogous to the liquid-vapour one except that ∆ vap H is
replaced by ∆ sub H. ∆ sub H = ∆ fus H + ∆ vap H so the required equation is
∆ sub H 1 1
p = p∗ exp �− � − ∗ ��
R T T
[10.6 + 30.8] × 103 J mol−1 1 1
= (4.80 × 103 Pa) × exp �− � − ��
8.3145 J K−1 mol−1 T 278.65 K
1 1
or (p�kPa) = 4.80 × exp �−4.98 × 103 � − ��
T�K 278.65
�is equation is plotted Fig. �.� for values in the range T ≤ 278.65 K since the
solid and vapour phases are only in equilibrium at the triple point and below.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 129
25
20
Liquid
15 Solid
p�kPa
10
5 Vapour
0
250 260 270 280 290 300 310 320
T�K
Figure 4.6
5 Simple mixtures
D�A.� Raoult’s law, [�A.��–���] de�nes the behaviour of ideal solutions. Like perfect
gases, what makes the behaviour ideal can be expressed in terms of intermolec-
ular interactions. Unlike perfect gases, however, the interactions in an ideal
solution cannot be neglected. Instead, ideal behaviour amounts to having the
same interactions between molecules of the di�erent components of the mix-
ture as there are between molecules of the same type.
In short, ideal behaviour consists of A–B interactions being the same as A–A
and B–B interactions. If that is the case, then the cohesive forces that would
keep a molecule in the liquid phase would be the same in the solution as in a
pure liquid, and the vapour pressure of a component will di�er from that of a
pure liquid only in proportion to its abundance (mole fraction). �us, Raoult’s
132 5 SIMPLE MIXTURES
law is expected to be valid for mixtures of components that have very simi-
lar chemical structures. Similar structures imply both similar intermolecular
interactions and similar sizes.
In an ideal dilute solution, on the other hand, Raoult’s law holds for the solvent
in the limit as x A approaches �, not because A–B interactions are like A–A
interactions, but because there are so many more A–A interactions than A–
B interactions that A–A interactions dominate the behaviour of the solvent.
For the solute, on the other hand, there are many more A–B interactions than
B–B interactions in the limit as x B approaches zero. �us, only one kind of
interaction (A–B) is important in determining the a�nity of the solute for the
solution.
dw add,max = µ A dn A + µ B dn B . . .
and so non-expansion work can arise from the changing composition of a sys-
tem.
Solutions to exercises
E�A.�(a) �e partial pressure of gas A, p A above a liquid mixture is given by Raoult’s
Law, [�A.��–���], p A = x A p∗A , where x A is the mole fraction of A in the liquid
and p∗A is the vapour pressure over pure A. �e total pressure over a mixture
of A and B is p A + p B .
�ese mole fractions are used with Raoult’s law to give the total pressure
1�M A 1�M B
p = x A p∗A + x B p∗B = p∗ + p∗
1�M A + 1�M B A 1�M A + 1�M B B
92.1332 g mol−1 .
1�M A 1�M B
p= p∗ + p∗
1�M A + 1�M B A 1�M A + 1�M B B
1�(78.1074 g mol−1 )
= × (10 kPa)
1�(78.1074 g mol−1 ) + 1�(92.1332 g mol−1 )
1�(92.1332 g mol−1 )
+ × (2.8 kPa)
1�(78.1074 g mol−1 ) + 1�(92.1332 g mol−1 )
= 5.41... kPa + 1.28... kPa = �.� kPa
E�A.�(a) �e total volume is calculated from the partial molar volumes of the two com-
ponents using [�A.�–���], V = n A VA + n B VB . �e task is therefore to �nd the
amount in moles, n A and n B , of A and B in a given mass m of solution. If the
molar masses of A and B are M A and M B then it follows that
m = nA MA + nB MB
�is latter expression for the total amount in moles, (n A + n B ), is used with
n A = x A (n A + n B ) to give
mx A
n A = x A (n A + n B ) =
xA MA + xB MB
and likewise
mx B
nB =
xA MA + xB MB
With these expressions for n A and n B the total volume is computed from the
partial molar volumes
mx A VA mx B VB
V = n A VA + n B VB = +
xA MA + xB MB xA MA + xB MB
m
= [x A VA + x B VB ]
xA MA + xB MB
m
= [x A VA + (1 − x A )VB ]
x A M A + (1 − x A )M B
�.��� kg evaluates as
1000 g
V=
0.4693 × (119.3679 g mol ) + (1 − 0.4693) × (58.0774 g mol−1 )
−1
−1 −1
× �0.4693 × (80.235 cm3 mol ) + (1 − 0.4693) × (74.166 cm3 mol )�
= ���.� cm3
E�A.�(a) Consider a solution of A and B in which the fraction (by mass) of A is α (here
α = 12 ). �e total volume of a solution of A and B is calculated from the partial
molar volumes of the two components using [�A.�–���], V = n A VA + n B VB . In
this exercise V and VA are known, so the task is therefore to �nd the amount
in moles, n A and n B , of A and B in the solution of known mass density ρ.
�e mass of a volume V of the solution is ρV , so the mass of A is αρV . If
the molar mass of A is M A , then the amount in moles of A is n A = αρV �M A .
Similarly, n B = (1 − α)ρV �M B . �e volume is expressed using these quantities
as
αρV VA (1 − α)ρV VB
V = n A VA + n B VB = +
MA MB
�e term V cancels between the �rst and third terms to give
αρVA (1 − α)ρVB
1= +
MA MB
MA (1 − α)VB ρ
VA = �1 − �
αρ MB
MA (1 − α)VB ρ
VA = �1 − �
αρ MB
(46.0674 g mol−1 )
=
0.5 × (0.914 g cm−3 )
−1
� (1 − 0.5) × (17.4 cm3 mol ) × (0.914 g cm−3 ) �
× 1−
� 18.0158 g mol−1 �
−1
= ��.� cm3 mol
100
p HCl �(kPa)
50
0
0.000 0.005 0.010 0.015 0.020
x HCl
Figure 5.1
nB nB nB
cB = = =ρ = ρb B
V m�ρ m
Using Henry’s law the concentration is therefore given by
�
bB
pB ρx B p
c B = ρb B = ρ =
KB KB
where the partial pressure p B is expressed in terms of the mole fraction and the
total pressure p, p B = x B p.
�e mole fraction of N� in air is �.���, the Henry’s law constant for N� in
benzene is 1.87 × 104 kPa kg mol−1 and the density of benzene is �.��� g cm−3 .
If it is assumed that the total pressure is � atm then
ρx N2 p (0.879 × 103 kg m−3 )×(0.780)×(101.325 kPa)
c N2 = = = 3.71... mol m−3
K N2 1.87 × 104 kPa kg mol−1
136 5 SIMPLE MIXTURES
When the pressure is ten times greater at �.�� atm the solubility is increased by
the same factor to 3.37 × 10−2 mol kg−1
E�A.�(a) As explained in Exercise E�A.�(a) the concentration of a solute is estimated
as c B = ρp B �K B where ρ is the mass density of the solvent. �e Henry’s law
constant for CO� in water is 3.01 × 103 kPa kg mol−1 and the density of water
is �.��� g cm−3 or ��� kg m−3 .
E�A.�(a) �e partial molar volume of solute B (here NaCl) is de�ned from [�A.�–���] as
∂V
VB = � �
∂n B p,T ,n ′
�e total volume is given as a function of the molality, but this volume is de-
scribed as that arising from adding the solute to � kg of solvent. �e molality of
a solute is de�ned as (amount in moles of solute)/(mass of solvent in kg), there-
fore because in this case the mass of solvent is � kg, the molality is numerically
equal to the amount in moles, n B .
�e polynomial given relates υ to x, and so from this it is possible to compute
the derivative dυ�dx. �is required derivative is dV �dn B (where the partials
are dropped for simplicity), which is related to dυ�dx in the following way
dV dV dυ dx
� �=� �� �� �
dn B dυ dx dn B
�e quantity x is de�ned as b�b −○ , but it has already been argued that the molal-
ity can be expressed as n B �(1 kg), hence x = n B �(mol) and therefore dx�dn B =
mol−1 . Because υ = V �cm3 , dυ�dV = cm−3 and so dV �dυ = cm3 . Hence
dV dV dυ dx dυ
� �=� �� �� � = � � cm3 mol−1
dn B dυ dx dn B dx
�e required derivative is
dυ
� � = 16.62 + 2.655 x 1�2 + 0.24 x
dx
Hence the expression for the partial molar volume of B (NaCl) is
VB = �16.62 + 2.655 x 1�2 + 0.24 x� cm3 mol−1
�e partial molar volume when b�b −○ = 0.1 is given by
VB �(cm3 mol−1 ) = (16.62 + 2.655 x 1�2 + 0.24 x)
= (16.62 + 2.655(0.100)1�2 + 0.24 × 0.100) = 17.4...
E�A.��(a) For a binary mixture the Gibbs–Duhem equation, [�A.��b–���], relates changes
in the chemical potentials of A and B
n A dµ A + n B dµ B = 0
E�A.��(a) Because the gases are assumed to be perfect and are at the same temperature
and pressure when they are separated, the pressure and temperature will not
change upon mixing. �erefore [�A.��–���], ∆ mix S = −nR(x A ln x A +x B ln x B ),
applies. �e amount in moles is computed from the total volume, pressure
and temperature using the perfect gas equation: n = pV �RT. Because the
separate volumes are equal, and at the same pressure and temperature, each
compartment contains the same amount of gas, so the mole fractions of each
gas in the mixture are equal at 0.5.
Note that the pressure in expressed in Pa and the volume in m3 ; the units of the
result are therefore (N m−2 ) × (m3 ) × (K−1 ) = N m K−1 = J K−1 .
Under these conditions the Gibbs energy of mixing is given by [�A.��–���],
∆ mix G = nRT(x A ln x A + x B ln x B ); as before n = pV �RT.
Solutions to problems
P�A.� �is problem is similar to the Example given in Section �A.�(d) on page ���.
�e Gibbs–Duhem equation [�A.��b–���], expressed in terms of partial molar
volumes is n A dVA + n B dVB = 0 which is rearranged to
nB
dVA = − dVB
nA
If the variation of the solute partial molar volume VB with concentration is
described by a known function, then integration of this equation gives an ex-
pression for how the solvent partial molar volume VA varies.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 139
�e integral is then
υA x x
� dυ A = −M A b −○ � x(9.5605 x −1�2 ) dx = −M A b −○ � 9.5605 x 1�2 dx
υ ∗A 0 0
P�A.� �e required molar masses are: N� ��.�� g mol−1 ; O� ��.�� g mol−1 ; Ar ��.�� g mol−1 ;
CO� ��.�� g mol−1 .
140 5 SIMPLE MIXTURES
Consider ��� g of the mixture. Of this ��.� g is N� so the amount in moles of this
gas is n N2 = (75.5 g)�(28.02 g mol−1 ) = 2.69... mol. Similar calculations are
made for the other cases to give the results shown below in the table. �e total
amount in moles n is found by summing these individual contributions and
this is then used to compute the mole fractions from x J = n J �n: the resulting
values are also shown in the table.
∆ mix S = −nR � x J ln x J
J
(∆ mix S)�n = −R � x J ln x J
J
�is expression is used togther with the values given in the table to compute
the entropy of mixing for the �rst set of data as +4.70 J K−1 mol−1 and for the
second set of data as +4.711 J K−1 mol−1 . �e di�erence is of the order of
0.01 J K−1 mol−1 .
��� � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � �
multiples of ��� g
m A �(g) + m B �(g)
×m�(g) = m B �(g)
100
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 141
m A �(g) × m�(g)
m B �(g) =
100 − m�(g)
1 015
1 010
V �cm3
1 005
1 000
0.4 0.6 0.8 1.0 1.2 1.4 1.6
n B �mol
Figure 5.2
�e data �t well to the polynomial (shown as the smooth curve on the plot)
�e partial molar volume is the slope of this curve which is the derivative with
respect to n B
−1
VB �(cm3 mol ) = 14.450(n B �mol) − 1.8512
�e following table gives values of VB for each of the data points. �ese are
plotted in Fig. �.�; the line is the function above.
142 5 SIMPLE MIXTURES
−1
m(CuSO4 )�g ρ�g cm−3 n B �mol VB �cm3 mol
5 1.051 0.330 2.91
10 1.107 0.696 8.21
15 1.167 1.106 14.13
20 1.230 1.566 20.78
20
15
−1
VB �cm3 mol
10
P�A.� In Example �A.� on page ��� the partial molar volume of ethanol is found to be
given by
υ = 54.6664 − 0.72788 z + 0.084768 z 2
−1
where υ = VE �(cm3 mol ) and z = n E �mol. �e value of z at which υ is a
minimum or maximum is found by setting the derivative dυ�dz = 0
dυ 0.72788
= −0.72788 + 0.169536 z = 0 hence z= = 4.2934
dz 0.169536
�is value of z corresponds to 4.2934 mol in �.��� kg of solvent water (speci�ed
in the Example). �e molality is the amount in moles divided by the mass of the
solvent in kg, thus the corresponding molality is 4.2934 mol kg−1 . �e plot in
the text con�rms that this is indeed the position of the minimum in the partial
molar volume.
D�B.� A regular solution has excess entropy S E of zero, but an excess enthalpy H E that
is non-zero. A regular solution of A and B can be thought of as one in which the
di�erent molecules of A and B are distributed randomly, as in an ideal solution,
but where the energy of A–A, B–B, and A–B interactions are di�erent.
In real solutions both S E and H E are non-zero, and in general both are likely
to vary with composition. �e non-zero value for S E is interpreted as arising
from the non-random distribution of molecules. �is is exempli�ed by ionic
solutions, in which ions of one charge are more likely to be surrounded by ions
of the opposite charge than of the same charge (Topic �F).
D�B.� All of the colligative properties result from the lowering of the chemical poten-
tial of the solvent due to the presence of the solute. For an ideal solution, this
reduction is predicted by µ A = µ ∗A + RT ln x A . �e relationship shows that as
the amount of solute increases, the mole fraction of the solvent x A decreases
and hence the chemical potential the solvent A decreases.
If the chemical potential of the solvent is lowered, then the chemical potential
of the vapour in equilibrium with the solvent is also lowered because at equi-
librium these two chemical potentials must be equal. �e chemical potential
of a perfect gas is given by µ A = µ −A○ + RT ln p A , so a lowering of the chemical
potential results in a reduction in the pressure.
�e overall result is that addition of a solute reduces the vapour pressure of
the solvent, and therefore the temperature at which the solvent boils is raised
because a greater increase in temperature is needed to make the vapour pres-
sure equal to the external pressure. Similarly, the freezing point of the solvent
is decreased because the chemical potential of the solid will equal that of the
solvent at a lower temperature.
At a molecular level the decrease in vapour pressure can be thought of as being
due to the solute molecules getting in the way of the solvent molecules, thus
reducing their tendency to escape. Another way of looking at this is that the
presence of a solute increases the ‘randomness’, and hence the entropy, of the
solution, thus reducing the tendency for the formation of the (pure) vapour or
solid.
144 5 SIMPLE MIXTURES
Solutions to exercises
E�B.�(a) In Exercise E�A.�(a) it is found that the vapour pressure obeys
�e task is to work out the mole fraction that corresponds to the given molality.
�e molality of HCl is de�ned as b HCl = n HCl �m GeCl4 , where n HCl is the amount
in moles of HCl and m GeCl4 is the mass in kg of solvent GeCl� . �e mole
fraction of HCl is n HCl �(n HCl + n GeCl4 ), where n GeCl4 is the amount in moles of
GeCl� , which is given by n GeCl4 = m GeCl4 �M GeCl4 , where M GeCl4 is the molar
mass of GeCl� . �ese relationships allow the mole fraction to be rewritten as
follows
n HCl n HCl
x HCl = =
n HCl + n GeCl4 n HCl + m GeCl4 �M GeCl4
�e amount in moles of HCl is written is n HCl = b HCl m GeCl4 ; using this the
above expression for the mole fraction becomes
�e molar mass of GeCl� is ���.�� g mol−1 , therefore the mole fraction corre-
sponding to b = 0.10 mol kg−1 is
E�B.�(a) Raoult’s law, [�A.��–���], p A = x A p∗A relates the vapour pressure to the mole
fraction of A, therefore from the given data is it possible to compute x A . �e
task is to relate the mole fraction of A to the masses of A (the solvent) and B
(the solute), and to do this the molar masses M A and M B are introduced. With
these n A = m A �M A , where m A is the mass of A, and similarly for n B . It follows
that
nA m A �M A MB mA
xA = = =
n A + n B m A �M A + m B �M B M B m A + M A m B
xA MA mB (p A �p∗A )M A m B
MB = =
m A (1 − x A ) m A [1 − (p A �p∗A )]
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 145
E�B.�(a) �e freezing point depression ∆Tf is related to the molality of the solute B, b B ,
by [�B.��–���], ∆Tf = K f b B , where K f is the freezing-point constant. From the
data and the known value of K f it is possible to calculate b B . �e task is then to
relate this to the given masses and the desired molar mass of the solute, M B .
�e molality of B is de�ned as b B = n B �m A , where m A is the mass of the solvent
A in kg. It follows that
nB m B �M B
bB = =
mA mA
where m B is the mass of solute B. From the freezing point data b B = ∆Tf �K f ,
therefore
∆Tf m B �M B mB Kf
= hence M B =
Kf mA m A ∆Tf
With the data given and the value of the freezing-point constant from the Re-
source section
(100 g) × (30 K kg mol−1 )
MB = = ��� g mol−1
(0.750 kg) × (10.5 K)
Note that because molality is de�ned as (amount in moles)/(mass of solvent in
kg), the mass of solvent m A is used as �.��� kg.
E�B.�(a) �e freezing point depression ∆Tf is related to the molality of the solute B,
b B , by [�B.��–���], ∆Tf = K f b B , where K f is the freezing-point constant. �e
molality of the solute B is de�ned as b B = n B �m A , where n B is the amount in
moles of B and m A is the mass in kg of solvent A. �e amount is related to the
mass of B, m B , using the molar mass M B : n B = m B �M B . It therefore follows
that
Kf mB
∆Tf = K f b B =
MB mA
�e molar mass of sucrose C�� H�� O�� is ���.���� g mol−1 . A volume ��� cm3 of
water has mass ��� g to a good approximation. Using these values with the data
given and the value of the freezing-point constant from the Resource section
gives the freezing point depression as
nB nB nB
[B] = = = ρ = bB ρ
V m A �ρ m A
With this the osmotic pressure is related to the molality
Π Π Π
[B] = hence bB ρ = and so bB =
RT RT ρRT
In this expression all of the quantities are in SI units therefore the temperature
is expected to be in K, which is veri�ed as follows
(K kg mol−1 ) × (Pa) Pa
=
−3 −1
(kg m ) × (J K mol ) × (K)
−1 J × m −3 × K−1
kg m−1 s−2
= =K
(kg m2 s−2 ) × m−3 × K−1
�e total amount in moles is 0.50 mol+2.00 mol = 2.50 mol. With A as hexane
and B as heptane the thermodynamic quantities are calculated as
∆ mix G = nRT(x A ln x A + x B ln x B )
0.50 0.50 2.00 2.00
= (2.50 mol) × (8.3145 J K−1 mol−1 ) × (298 K) × � ln + ln �
2.50 2.50 2.50 2.50
= −3.10 × 103 J
∆ mix S = −nR(x A ln x A + x B ln x B )
0.50 0.50 2.00 2.00
= −(2.50 mol) × (8.3145 J K−1 mol−1 ) × � ln + ln �
2.50 2.50 2.50 2.50
= +10.4 J K−1
nA m A �M A MB mA
xA = = =
n A + n B m A �M A + m B �M B M B m A + M A m B
MB mA MB mB MB 1
xA = = hence = � − 1�
M B m A + M A m B M B + M A (m B �m A ) mA MA xA
mB MB 1 86.1706 g mol−1 1
= � − 1� = � − 1� = �.����
mA MA xA 100.1964 g mol−1 1�2
More simply, if equal amounts in moles of A and B are required, the ratio of
the corresponding masses of A and B must be equal to the ratio of their molar
masses: m B �m A = M B �M A .
∆ fus H 1 1
ln x B = � − �
R Tf T
28.8 × 103 J mol−1 1 1
= −1 � − � = −4.55...
−1
8.3145 J K mol (217 + 273.15) K (25 + 273.15) K
hence x B = 0.0105....
148 5 SIMPLE MIXTURES
E�B.��(a) �e vapour pressure of component J in the solution obeys Raoult’s law, [�A.��–
���], p J = p∗J x J , where x J is the mole fraction in the solution. In the gas the
partial pressure is p J = y J p tot , where y J is the mole fraction in the vapour.
�ese relationships give rise to four equations
where x A + x B = 1 is used and likewise for the gas. In these equations x A and
p tot are the unknowns to be found. �e expressions for p A are set equal, as are
those for p B , to give
p∗A x A
p∗A x A = p tot y A hence p tot =
yA
p∗B (1 − x A )
p∗B (1 − x A ) = p tot (1 − y A ) hence p tot =
1 − yA
�ese two expressions for p tot are set equal and the resulting equation rear-
ranged to �nd x A
E�B.��(a) If the solution is ideal, the vapour pressure of component J in the solution
obeys Raoult’s law, [�A.��–���], p J = p∗J x J , where x J is the mole fraction in
the solution. In the gas the partial pressure is p J = y J p tot , where y J is the mole
fraction in the vapour.
150 5 SIMPLE MIXTURES
�e normal boiling point is when the total pressure is � atm, and this is exactly
the pressure found by assuming Raoult’s law applies. �e solution is therefore
ideal .
�e composition of the vapour is computed from p A = p tot y A and p A = p∗A x A
hence
pA p∗ x A (127.6 kPa) × (0.6589)
yA = = A = = 0.829...
p tot p tot 101.325 kPa
It follows that y B = 1 − 0.829... = 0.170.... �e composition of the vapour is
therefore y A = 0.830 and y B = 0.170 .
Solutions to problems
P�B.� �e freezing point depression in terms of mole fraction is predicted by [�B.��–
���]
RT ∗2
∆T = x B K ′ K′ =
∆ fus H
With the data given
nB nB n B �m A bB
xB = = = =
n A + n B m A �M A + n B 1�M A + n B �m A 1�M A + b B
�
xB
∆T = b B M A K ′ hence b B = ∆T�M A K ′
�e table below gives values of b B calculated from the given ∆T and this ex-
pression; to distinguish these values for the experimental values of b B , the cal-
culated values are termed apparent molalities, b B,app
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 151
b B �(mol kg−1 ) ∆T�K b B,app �(mol kg−1 ) b B,app �b B M B,app �(g mol−1 )
0.015 0.115 0.031 2.081 27.9
0.037 0.295 0.080 2.165 26.8
0.077 0.470 0.128 1.657 35.1
0.295 1.381 0.375 1.271 45.7
0.602 2.67 0.725 1.204 48.3
M B,app bB
=
MB b B,app
with M B = 58.01 g mol−1 , the molar mass of KF. �e argument that leads
to this is that the greater the apparent molality the smaller the molar mass:
M B,app ∝ 1�b B,app .
�e data in the table show that the molality predicted from the experimental
freezing point depression using the value of the freezing-point constant deter-
mined by [�B.��–���] is always greater than the molality know from the way
the solution was prepared. Presumably this latter molality is based on adding
a know mass of KF to a known mass of solvent, and assuming that the molar
mass of KF is ��.� g mol−1 . �e fact that the apparent molality is higher than
the molality of the prepared solution implies that the number of solute species
is greater than expected.
�e freezing point depression depends on the mole fraction of the solute, re-
gardless of its identity. �erefore if the added KF were to dissociate completely
on dissolution in ethanoic acid the mole fraction of the solute would be twice as
large as expected on the basis of the amount of added KF, and in turn this would
mean that the apparent molality (based on the freezing-point depression) is
twice as large as expected.
�e data in the table can be interpreted as indicating that there is dissociation
of the KF, and that this dissociation is greater at lower molalities. However, this
only part of the picture as it does no explain why b B,app �b B is greater than � at
some molalities.
P�B.� Let the two components of the mixture be labelled � (propionic acid) and �
(THP). �e de�nition of the partial molar volume of �, V1 , is
∂V
V1 = � �
∂n 1 n 2
n1 M1 n2 M2 n1 M1 n2 M2
(n 1 + n 2 )V E = V − − hence V = (n 1 + n 2 )V E + +
ρ1 ρ2 ρ1 ρ2
n1 M1 n2 M2
V = (n 1 + n 2 )V E + +
ρ1 ρ2
∂V ∂V E M1
hence V1 = � � = (n 1 + n 2 ) � � + VE + (�.�)
∂n 1 n 2 ∂n 1 n 2 ρ1
�e �rst step is to compute (∂V E �∂n 1 )n 2 , and this requires rewriting the mole
fractions in terms of the n i
V E = x 1 x 2 a 0 + x 12 x 2 a 1 − x 1 x 22 a 1
n1 n2 a0 n 12 n 2 a 1 n 1 n 22 a 1
= + −
(n 1 + n 2 )2 (n 1 + n 2 )3 (n 1 + n 2 )3
∂V E n2 a0 2n 1 n 2 a 0 2n 1 n 2 a 1
� � = − +
∂n 1 n 2 (n 1 + n 2 ) 2 (n 1 + n 2 ) 3 (n 1 + n 2 )3
3n 12 n 2 a 1 n 22 a 1 3n 1 n 22 a 1
− − +
(n 1 + n 2 )4 (n 1 + n 2 )3 (n 1 + n 2 )4
∂V E M1
V1 = (n 1 + n 2 ) � � + VE +
∂n 1 n 2 ρ1
= (x 2 a 0 − 2x 1 x 2 a 0 + 2x 1 x 2 a 1 − 3x 12 x 2 a 1 − x 22 a 1 + 3x 1 x 22 a 1 )
+ (x 1 x 2 a 0 + x 12 x 2 a 1 − x 1 x 22 a 1 ) + M 1 �ρ 1
= a 0 x 2 (1 − x 1 ) + a 1 x 2 (2x 1 − 2x 12 − x 2 + 2x 1 x 2 ) + M 1 �ρ 1
= a 0 x 2 (x 2 ) + a 1 x 2 (2x 1 [1 − x 1 ] − x 2 + 2x 1 x 2 ) + M 1 �ρ 1
= a 0 x 22 + a 1 x 2 (2x 1 x 2 − x 2 + 2x 1 x 2 ) + M 1 �ρ 1
= a 0 x 22 + a 1 x 2 (4x 1 x 2 − x 2 ) + M 1 �ρ 1
= a 0 x 22 + a 1 x 22 (4x 1 − 1) + M 1 �ρ 1
V E = x 2 x 1 a 0 + x 22 x 1 a 1 − x 2 x 1 a 1
which, when compared with the original expression, shows that the sign of the
term in a 1 is reversed: this change needs to be carried through to the end. In
summary
V1 = a 0 x 22 + a 1 x 22 (4x 1 − 1) + M 1 �ρ 1
V2 = a 0 x 12 − a 1 x 12 (4x 2 − 1) + M 2 �ρ 2
V1 = a 0 x 22 + a 1 x 22 (4x 1 − 1) + M 1 �ρ 1 = 14 a 0 + 14 a 1 + M 1 �ρ 1
−1 −1
= 0.25 × (−2.4697 cm3 mol ) + 0.25 × (0.0608 cm3 mol )
−1
+ (74.0774 g mol−1 )�(0.97174 g cm−3 ) = ��.� cm3 mol
V2 = a 0 x 12 − a 1 x 12 (4x 2 − 1) + M 2 �ρ 2 = 14 a 0 − 14 a 1 + M 2 �ρ 2
−1 −1
= 0.25 × (−2.4697 cm3 mol ) − 0.25 × (0.0608 cm3 mol )
−1
+ (86.129 g mol−1 )�(0.86398 g cm−3 ) = ��.� cm3 mol
�
cJ
n J m J �M J m J 1 cJ
[J] = = = =
V V V MJ MJ
where c J is the mass concentration. With this the virial equation is rewritten
cJ cJ
Π = [J]RT(1 + B[J]) = RT �1 + B �
MJ MJ
A plot of Π�c J against c J will have intercept RT�M J when c J = 0, and from this
it is possible to determine the molar mass.
�e pressure is given by Π = hρg; for the pressure to be in Pa the height needs
to be in m and ρ in kg m−3 ; for the present case ρ = 1 g cm−3 = 1000 kg m−3 .
�e data are plotted in Fig. �.�.
175.02
174.98
0 1 2 3 4 5 6 7
c�(mg cm )
−3
Figure 5.4
1 m3 1 mg
(175.03 Pa mg−1 cm3 ) × × −6 = 175.03 Pa kg−1 m3
10 cm
6 3 10 kg
40
30
25
20
0 2 4 6 8 10
c�(mg cm )
−3
Figure 5.5
0.4 ξ = −2
ξ = −1
0.2 ξ=0
ξ = +1
H E �(nRT)0.0 ξ = +2
−0.2
−0.4
200 K
80
250 K
300 K
(H E �(nRξ))�K
60 350 K
40
20
0
0.0 0.2 0.4 0.6 0.8 1.0
xA
Figure 5.7
Π RT BRT
= + cJ
cJ MJ M J2
A plot of Π�c J against c J will have intercept RT�M J when c J = 0: from this it is
possible to determine the molar mass. �e second virial coe�cient is obtained
from the slope. Such a plot is shown in Fig. �.�.
158 5 SIMPLE MIXTURES
80
(Π�c)�(Pa g−1 dm3 )
60
40
20
0 2 4 6 8 10
c�(g dm )
−3
Figure 5.8
D�C.� A low-boiling azeotrope has a boiling temperature lower than that of either
component, so it is easier for the molecules to move into the vapour phase
than in a ‘normal’ (non-azeotropic) mixture. �erefore, the liquid phase has
less favorable intermolecular interactions than in a ‘normal’ mixture, a sign
that the components are less attracted to each other in the liquid phase than to
molecules of their own kind. �ese intermolecular interactions are determined
by factors such as dipole moment (polarity) and hydrogen bonding.
Conversely, a high-boiling azeotrope has a boiling temperature higher than that
of either component, so it is more di�cult for the molecules to move into the
vapour phase. �is re�ects the relatively unusual situation of components that
have more favorable intermolecular interactions with each other in the liquid
phase than with molecules of their own kind.
Solutions to exercises
E�C.�(a) �e molar masses of phenol and water are ��.���� g mol−1 and ��.���� g mol−1 ,
respectively. �e mole fraction of phenol (P) is
n β l α 0.149... − 0.042
= = = �.��
n α l β 0.161 − 0.149...
E�C.�(a) An approximate phase diagram is shown in Fig. �.�; the given data points are
shown with dots and the curve is a quadratic which is a modest �t to these
points. �e shape conforms to the expected phase diagram for such a system.
23.0
22.5
θ�○ C
22.0
21.5
21.0
0.0 0.2 0.4 0.6 0.8 1.0
x C6 F14
Figure 5.9
θ�○ C xM yM θ�○ C xM yM
110.6 1 1 117.3 0.408 0.527
110.9 0.908 0.923 119.0 0.300 0.410
112.0 0.795 0.836 121.1 0.203 0.297
114.0 0.615 0.698 123.0 0.097 0.164
115.8 0.527 0.624 125.6 0 0
125 liquid
vapour
120
θ�○ C 115
y M = 0.354
x M = 0.250
110
0.0 0.2 0.4 0.6 0.8 1.0
zM
Figure 5.10
E�C.�(a) At the lowest temperature shown in the diagram the mixture is in the two-
phase region, and the two phases have composition of approximately x B =
0.88 and x B = 0.05. �e level rule shows that there is about � times more of
the B-rich than of the B-poor phase. As the temperature is raised the B-rich
phase becomes slightly less rich in B, and the other phase becomes richer in B.
�e lever rule implies that the proportion of the B-rich phase increases as the
temperature rises.
At temperature T1 the vertical line intersects the phase boundary. At this point
the B-poor phase disappears and only one phase, with x B = 0.8, is present.
Solutions to problems
P�C.� If it is assumed that Raoult’s law applies, [�A.��–���], the partial vapour pres-
sures of benzene (B) and methylbenzene (M) are
p B = x B p∗B p M = x M p∗M
where x J are the mole fractions and p∗J are the vapour pressures over the pure
liquids. �e total pressure is taken to be p tot = p B + p M .
�e mole fraction in the vapour, y J , is related to the total pressure by p J = y J p tot ,
so it follows that
pJ x J p∗J
yJ = =
p tot p tot
�erefore
x B p∗B 0.75 × (75 Torr)
yB = = = �.��
p tot 0.75 × (75 Torr) + 0.25 × (21 Torr)
x M p∗M 0.25 × (21 Torr)
yM = = = �.���
p tot 0.75 × (75 Torr) + 0.25 × (21 Torr)
162 5 SIMPLE MIXTURES
where x J are the mole fractions and p∗J are the vapour pressures over the pure
liquids. �is equation is used to construct the liquid line on the graph shown
in Fig. �.��, where z B is interpreted as x B .
�e mole fraction in the vapour, y J , is related to the total pressure by p J = y J p tot .
Using this it can be shown that the total pressure in terms of the mole fraction
in the vapour in given by [�C.�–���],
p∗B p∗M
p tot =
p∗B + (p∗M − p∗B )y B
�is equation is used to construct the vapour line on the phase diagram, where
z B is interpreted as y B .
10.0 liquid
vapour
8.0
a ′1
p tot �kPa
a1
6.0
a ′′2 ′
a2 a2
a′′3
4.0 a3
2.0
0.0 0.2 0.4 0.6 0.8 1.0
zB
Figure 5.11
�is is the pressure at which boiling �rst occurs, point a 1 in the diagram.
(b) �e composition of the vapour is given by
(c) As the pressure is reduced further, say to point a 2 , the tie line indicates
that the liquid will have composition a ′′2 and the vapour will have com-
position a′2 , the latter being richer in the more volatile component B.
�e level rule indicates that the proportion of the vapour phase is now
signi�cant.
�e process continues until point a 3 is reached. At this pressure the com-
position of the liquid is given by point a ′′3 , and the level rule indicates that
the proportion of the liquid phase is very small. It is also evident from the
diagram that at point a 3 the vapour composition is y B = y M = 12 , therefore
p B = 12 p tot and p M = 12 p tot . Raoult’s law gives the partial vapour pressures
of B and M are p B = p∗B x B and p M = p∗M x M . It follows that
1
p
2 tot
= x B p∗B and 1
p
2 tot
= x M p∗M = (1 − x B )p∗M
P�C.� �e annotated phase diagrams are shown in Fig. �.��. Given that the normal
boiling point of hexane is certainly lower than that of heptane the horizontal
scale should presumably be mole fraction of heptane; however, the solution
provided follows the labelling of the diagram in the text.
900
liquid line
70°C
700
liquid
Pressure, p/Torr
d a
625
500 c 500
b
90
f e g
v l
80
liquid
70
1.0
0.8
0.6
yA 0.4
(p∗A �p∗B ) = 1
(p∗A �p∗B ) = 4
0.2
(p∗A �p∗B ) = 50
0.0
0.0 0.2 0.4 0.6 0.8 1.0
xA
Figure 5.13
As before, the derivative is zero at x A = 0.5 for all values of ξ; this corresponds
either to a minimum when ξ is small, or to a maximum when ξ is su�ciently
large. Qualitatively the behaviour is similar to that shown in Fig. �B.�.
Apart from this solution at x A = 0.5, there are no analytical solutions for when
this derivative is zero. However, solutions can be found by graphically by look-
ing for the intersection between ln (x A �[1 − x A ]) and −2ξx A (1− x A )(1−2x A ).
�is is done with the aid of Fig. �.�b. From the graph it is evident that for ξ = 1
there are no values of x A at which the curves intersect, and so no minima, but
at su�ciently high values of ξ (such as ξ = 6) such intersections do occur and
lead to two minima.
ξ=0
ξ=3
0.0 ξ=6
ξ = 15
∆ mix G�nRT
−0.5
liquid
A(s) + B(l)
+ A(l) AB2(s),B(l)
B(l) + A(l)
A(l)
Temperature
+ B(s)
AB2(s),B(l)
A(l)
A(s) + AB2(s)
AB2(s) + B(s)
Solutions to exercises
E�D.�(a) �e feature that indicates incongruent melting is the intersection of the two
liquid curves at around x B = 0.6. �e incongruent melting point is marked as
T1 ≈ 350 ○ C. �e composition of the eutectic is x B ≈ 0.25 and its melting point
is labelled T2 ≈ 190 ○ C .
E�D.�(a) �e cooling curves are shown in Fig �.��; the break points are shown by the
short horizontal lines. For isopleth a the �rst break point is at ��� ○ C where
the isopleth crosses the liquid curve, there is a second break point where the
isopleth crosses the boundary at T1 ; there is then a eutectic halt at ��� ○ C. For
isopleth b the �rst break point is at ��� ○ C where the isopleth crosses the liquid
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 167
curve, there is a second break point where the isopleth crosses the boundary at
T1 , and then a eutectic halt at ��� ○ C.
500
400
T�K
300
a b
200
time
Figure 5.16
E�D.�(a) Figure �.�� shows the relevant phase diagram to which dotted horizontal lines
have been added at the relevant temperatures.
1000
c1
800
Temperature, θ/°C
Liquid
600
c2
Ag 3Sn
400
c3
200
0
0 20 40 60 80 100
Figure 5.17 Mass percentage of Ag/%
mass fraction, namely the mass of silver (from the compound and from
atomic silver) over the total sample mass
m Ag (3n c + n a )M Ag
c Ag = =
m Ag + m Sn (3n c + n a )M Ag + n c M Sn
n c �3M Ag (c Ag − 1) + M Sn c Ag � = n a M Ag (1 − c Ag )
nc M Ag (1 − c Ag )
=
n a 3M Ag (c Ag − 1) + M Sn c Ag
E�D.�(a) �e schematic phase diagram is shown in Fig �.��. �e solid points are the data
given in the Exercise, and the lines are simply plausible connections between
these points; it is assumed that the compound in �:�. Note that to the right of
x B = 0.5 the solids are AB and B, whereas to the le� of this composition the
solids are AB and A.
AB(s)+AB(l)
+B(l)
110 A(s)+AB(s)
B(s)+AB(s)
100
0.0 0.2 0.4 0.6 0.8 1.0
xB
Figure 5.18
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 169
92
90
two-phase liquid
T�K
88
86
82
0.0 0.2 0.4 0.6 0.8 1.0
x CF4
Figure 5.19
E�D.�(a) �e schematic phase diagram is shown in Fig �.��. �e solid points are the data
given in the Exercise, and lines are simply plausible connections between these
points. (�e dash-dotted lines are referred in to Exercise E�D.�(a).)
E�D.�(a) �e compositions at which the cooling curves are plotted are indicated by the
vertical dash-dotted lines on the phase diagram for Exercise E�D.�(a), Fig. �.��.
�e cooling curves are shown in Fig �.��. �e break points, where solid phases
start to form are shown by the short horizontal lines, and the dotted lines in-
dicate the temperatures of the two eutectics (�� K and �� K). �e horizontal
segments correspond to solidi�cation of a eutectic.
94
92
90 (i) (ii) (iii) (iv) (v)
T�K
88
86
84
82
time
Figure 5.20
Solutions to problems
P�D.� �e schematic phase diagram is shown in Fig �.��. �e solid points are the
data given in the Exercise, and in the absence of any further information and
170 5 SIMPLE MIXTURES
because there are so few points, these have just been joined by straight lines.
1,000 solid
liquid
θ�○ C
900
800
On cooling a liquid with composition x ZrF4 = 0.4 solid �rst starts to appear
when the isopleth intersects the liquid line (at about ��� ○ C). �e composition
of the small amount of solid that forms is given by the le�-hand open circle in
the diagram (about x = 0.29; containing less ZrF� than the liquid). As the
temperature drops further more solid is formed and its composition moves
along the solid line to the right becoming richer in ZrF� until it reaches the
point where the isopleth crosses the solid line. At this point what remains of
the liquid has composition given by the right-hand open circle (about x = 0.48).
A further drop in temperature results in complete solidi�cation.
P�D.� �e phase diagram is shown in Fig. �.��, along with the relevant cooling curves.
�e fact that there is a phase boundary indicated by the vertical line at x B = 0.67
is taken to indicate the formation of compound AB� which has x B = 23 . By
analogy with the phase diagram shown in Fig. �D.� on page ���, the form of
the given phase diagram indicates that AB� does not exist in the liquid phase.
�e number of distinct chemical species (as opposed to components) and phases
present at the indicated points are, respectively
b(3, 2), d(2, 2), e(4, 3), f (4, 3), g(4, 3), k(2, 2)
P�D.� (a) Note that, as indicated on the diagram, Ca� Si, CaSi, CaSi� appear at mole
fractions of Si 13 , 12 and 23 , as expected.
eutectics: x Si = 0.056 at approximately ��� ○ C, x Si = 0.402 at ���� ○ C,
x Si = 0.694 at ���� ○ C
congruent melting compounds: Ca� Si Tf = 1314 ○ C, CaSi Tf = 1324 ○ C
incongruent melting compound: CaSi� Tf = 1040 ○ C (melts into CaSi(s)
and Si-rich liquid with x Si around �.��)
(b) For an isopleth at x Si = 0.2 and at ���� ○ C the phases in equilibrium are
CaSi� and a Ca-rich liquid (x Si = 0.11). �e lever rule, [�C.�–���], gives
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 171
A(l)+B(l)+B(s)
Temperature, T
f
A(l)+B(l)
+A(s) g
c A(l)+B(l)+AB2(s)
k
e
d AB2(s)+A(s) AB2(s)+B(s)
0.23
0.57
0.67
0.84
0.16
0 0.2 0.4 0.6 0.8 1
Mole fraction of B, xB
xB = 0.16 xB = 0.57 xB = 0.84
xB = 0.23
xB = 0.67
Figure 5.22
(c) For an isopleth at x Si = 0.8 Si(s) begins to appear at about ���� ○ C. Fur-
ther cooling causes more Si(s) to freeze out of the melt so that the melt
becomes more concentrated in Ca. �ere is a eutectic at x Si = 0.694 and
���� ○ C.
At a temperature just above the eutectic point the liquid has composition
x Si = 0.694 and the lever rule gives that the relative amounts of the Si(s)
and liquid phases as:
At the eutectic temperature a third phase appears, CaSi� (s). As the melt
cools at this temperature, both Si(s) and CaSi� (s) freeze out of the melt
while the composition of the melt remains constant. At a temperature
slightly below the eutectic point all the melt will have frozen to Si(s) and
CaSi� (s) with the relative amounts:
n Si l CaSi2 0.80 − 0.667
= = = �.��
n CaSi2 l Si 1.0 − 0.80
P�D.� �e data are plotted as the phase diagram shown in Fig �.��; the �lled and open
circles are the data points and the solid/dashed line is a best-�t cubic function.
A mixture of �.��� mol of N,N-dimethylacetamide with �.��� mol of heptane
has mole fraction of the former of x 1 = 0.750�(0.750 + 0.250) = 0.750. �e
tie line at ���.� ○ C is shown on the diagram, and this intersects with the two
curves at x 1 = 0.167 and x 2 = 0.805 (determined from the best-�t polynomial -
172 5 SIMPLE MIXTURES
310 x1
x2
305
T�K
300
295
an alternative would be to use the data points given for this temperature). �e
lever rule, [�C.�–���] gives the proportion of the two phases as
n x=0.805 0.750 − 0.167
= = ��.�
n x=0.167 0.805 − 0.750
�e N,N-dimethylacetamide-rich phase is therefore more than ten times more
abundant than the other phase.
A mixture of this composition will become a single phase at the temperature at
which the x 1 = 0.750 isopleth intersects the right-hand phase boundary. Using
the �tted function, this intersection is at ���.� ○ C .
D�E.� �e lever rule, [�C.�–���], applies in a ternary system, but with an important
caveat. For binary systems the tie lines to which the rule appplies are always
horizontal and so can be added to the phase diagram at will. In contrast, for
a ternary system the tie lines have no such simple orientation and have to be
determined experimentally. �us the lever rule applies, but in order to use
it additional information is needed about the tie lines at the composition of
interest.
Solutions to exercises
E�E.�(a) �e points corresponding to the given compositions are marked with letters on
the phase diagram shown in Fig. �.��. Composition (i) is in a two-phase region,
(ii) is in a three-phase region, (iii) is in a region where there is only one phase.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 173
Composition (iv) corresponds to the point at which the phase boundaries meet
and all of the phases are in equilibrium
0 1 H2O
S1
0.2 P = 1 0.8
iii S2
0.4 0.6
P=2 P=2
0.6 i 0.4
iv
NH4Cl 1 0 (NH4)2SO4
0 0.2 0.4 0.6 0.8 1
Figure 5.24
E�E.�(a) (i) �e phase equilibrium between NH� Cl and H� O is indicated by the le�-
hand edge of the phase diagram shown in Fig. �.��. At the point S 1 the
system moves from two phases to one – in other words from a system of
solid NH� Cl in equilibrium with a solution of NH� Cl in H� O, to a system
in which there is just one phase, a solution of NH� Cl in H� O. �is point
therefore marks the solubility of NH� Cl, and from the diagram is occurs
at x NH4 Cl = 0.19.
�e task is to convert this mole fraction to a molar concentration. Imagine
a solution made from a mass m of H� O: the amount in moles of H� O is
m�M, where M is the molar mass of H� O. �e mole fraction of NH� Cl
is therefore
n NH4 Cl n NH4 Cl
x NH4 Cl = =
n NH4 Cl + n H2 O n NH4 Cl + m�M
�is rearranges to n NH4 Cl = x NH4 Cl (m�M)�(1 − x NH4 Cl ). If it is assumed
that the mass density of the solution is approximately the same as the mass
density of water, ρ, the volume of this solution is given by V = m�ρ. With
this, the molar concentration of NH� Cl is computed as
n NH4 Cl 1 x NH4 Cl (m�M) ρ x NH4 Cl (m�M)
[NH4 Cl] = = =
V V 1 − x NH4 Cl m 1 − x NH4 Cl
x NH4 Cl ρ
=
M(1 − x NH4 Cl )
With the data from the diagram and assuming ρ = 1000 g dm−3
x NH4 Cl ρ 0.19 × (1000 g dm−3 )
[NH4 Cl] = =
M(1 − x NH4 Cl ) (18.0158 g mol−1 ) × (1 − 0.19)
= �� mol dm−3
�is high concentration rather casts doubt on assuming the density of the
solution is the same as that of water.
174 5 SIMPLE MIXTURES
A
0.0
1.0
0.2
0.8
xC 0.4 xA
0.6
0.6
0.4
(iii)
0.8
(i) 0.2
1.0 (ii)
C 0.0 B
0.0 0.2 0.4 0.6 0.8 1.0
xB
Figure 5.25
= 0.13
Likewise, x Na2 SO4 ⋅ 10 H2 O = 0.024 and x H2 O = 0.85; this point is plotted in the
ternary phase diagram shown in Fig �.��.
�e line with varying amounts of water but the same relative amounts of the
two salts (in this case, equal by mass), passes through this point and the vertex
corresponding to x H2 O = 1. �is line intersects the NaCl axis at a mole fraction
corresponding to a ��:�� mixture (by mass) of the two salts
NaCl
0.0
1.0
0.2
0.8
x H2 O 0.4 x NaCl
0.6
0.6
0.4
0.8
0.2
1.0
H� O 0.0 Na� SO�
0.0 0.2 0.4 0.6 0.8 1.0
x Na2 SO4 ⋅ 10 H2 O
Figure 5.26
m CHCl3 �M CHCl3
x CHCl3 =
m CHCl3 �M CHCl3 + m CH3 COOH �M CH3 COOH + m H2 O �M H2 O
9.2 g
119.37 g mol−1
= = �.��
+ +
9.2 g 3.1 g 2.3 g
119.37 g mol−1 60.052 g mol−1 18.016 g mol−1
Likewise, x CH3 COOH = 0.20 and x H2 O = 0.50 . �is point in marked with the
open circle on the phase diagram shown in Fig. �.��; it falls clearly in the two-
phase region. �e point almost lies on the tie line a ′2 –a ′′2 , and using this as a
guide the lever rule indicates that the phase with composition a′2 , (x W = 0.57,
x T = 0.20, x E = 0.23), is approximately � times more abundant than the phase
with composition a′′2 , (x W = 0.06, x T = 0.82, x E = 0.12).
(i) When water is added to the mixture the composition moves along the
dashed line to the lower-le� corner. �e system will pass from the two-
phase to the one-phase region when the line crosses the phase boundary,
which is at approximately (x W = 0.75, x T = 0.14, x E = 0.10).
(ii) When ethanoic acid is added to the mixture the composition moves along
the dashed line to the vertex. �e system will pass from the two-phase to
the one-phase region when the line crosses the phase boundary, which is
at approximately (x W = 0.44, x T = 0.26, x E = 0.30).
Solutions to problems
P�E.� �e given points are shown by �lled dots in the phase diagram shown in Fig. �.��,
and they are connected by a straight line, as indicated in the problem. Beneath
176 5 SIMPLE MIXTURES
CH� COOH
0.0
1.0
0.2
0.8
xW 0.4 xE
0.6
0.6
0.4
0.8 a′2
0.2
1.0 a ′′2
H� O 0.0 CHCl�
0.0 0.2 0.4 0.6 0.8 1.0
xT
Figure 5.27
this line lies a one-phase region because CO� and nitrobenzene are miscible
in all proportions. Addition of I� eventually causes phase separation into a
two-phase region for all compositions about the line.
0 1 I2
0.2 0.8
0.4 0.6
0.8 0.2
P=1
CO2 1 0 nitrobenzene
0 0.2 0.4 0.6 0.8 1
Figure 5.28
P�E.� Consider the construction shown in Fig. �.��. �e line is interest is AW, where
as indicated the position of W is determined by the mole fractions of B and
C in the binary mixture; to avoid confusing these particular mole fractions
with others, they are denoted y B and y C . �e point P lies on this line and its
perpendicular distance from each of the edges of the triangle gives the mole
fractions of B and C, x B and x C .
Construct the line UV passing through P and parallel to the base of the triangle.
It follows that AWB and APV are similar triangles, therefore there exists the
following relationship between the ratios of the sides.
WB PV
= (�.�)
AW AP
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 177
N
M xC
xB
U V
P
C B
yB W yC
Figure 5.29
Alternatively (and avoiding the use of the angle explicitly) PNV and PMU are
also recognised as similar triangles, so that it follows directly that x C �PV =
x B �PU.
5F Activities
Answers to discussion questions
D�F.� �e Debye–Hückel theory of electrolyte solutions formulates deviations from
ideal behaviour (essentially, deviations due to electrostatic interactions between
the ions) in terms of the work of charging the ions. �e assumption is that
the solute particles would behave ideally if they were not charged, and the
di�erence in chemical potential between real and ideal behaviour amounts to
the work of putting electrical charges onto the ions.
To �nd the work of charging, the distribution of ions must be found, and that is
done using the shielded Coulomb potential which takes into account the ionic
178 5 SIMPLE MIXTURES
strength of the solution and the dielectric constant of the solvent. �e Debye–
Hückel limiting law, [�F.��–���], relates the mean ionic activity coe�cient to
the charges of the ions involved, the ionic strength of the solution, and depends
on a constant that takes into account solvent properties and temperature.
D�F.� If a solvent or solute has a certain chemical potential, then this is related to
the activity of the solvent or solute through [�F.�–���] and [�F.�–���]. At �rst
sight it seems odd to think of the chemical potential determining the activity,
but this approach is in fact logical as the chemical potential is the experimen-
tally measurable quantity. For example, measurements of cell potentials or the
vapour pressure of liquids provide (slightly indirect, it must be admitted) ways
of measuring the chemical potential.
For ideal systems expressions are available which relate the chemical potential
to the concentration (in the form of its various measures, such as mole fraction
or molality). It therefore follows that the di�erence between the measured
chemical potential and that predicted from idealised models is attributable to
factors not taken into account in the ideal systems. Such factors are collectively
described as non-ideal interactions; the di�erence between activity and con-
centration is therefore ascribed to the presence of such interactions
Non-ideal interactions are no di�erent from the usual interactions between
molecular species. For example, they may include interactions between charged
or polar species, hydrogen bonding or more speci�c interactions.
D�F.� �e main way of measuring activities described in this Topic is from measure-
ments of partial vapour pressures, as given in [�F.�–���] for the solvent and in
[�F.��–���] for the solute activity. Other measurements from which the value
of the chemical potential can be inferred (for example, cell potentials) are used
to determine activities via the general relationship µ J = µ −J○ + RT ln a J .
Solutions to exercises
E�F.�(a) Ionic strength is de�ned in [�F.��–���]
I = 12 � z 2i (b i �b −○ )
i
where the sum runs over all the ions in the solution, z i is the charge number
on ion i, and b i is its molality. For KCl the molality of K+ and Cl – are both
�.�� mol kg−1 ; z K+ = +1 and z Cl− = −1. For CuSO� the molality of Cu�+ and
SO� � – are both �.�� mol kg−1 ; z Cu2+ = +2 and z SO4 2− = −2. �e ionic strength
is therefore
I = 12 [1�(1 mol kg−1 )] �(+1)2 × (0.10 mol kg−1 ) + (−1)2 × (0.10 mol kg−1 )
+(+2)2 × (0.20 mol kg−1 ) + (−2)2 × (0.20 mol kg−1 )� = �.�
I = 12 � z 2i (b i �b −○ )
i
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 179
where the sum runs over all the ions in the solution, z i is the charge number
on ion i, and b i is its molality.
(i) �e aim here is to increase the ionic strength by 0.250−0.150 = 0.100; the
task is therefore to compute the mass m of Ca(NO� )� which, when added
to a mass m w of water, gives this increase in the ionic strength.
A solution of Ca(NO� )� of molality b contributes Ca�+ at molality b and
NO� – at molality 2b. �e contribution to the ionic strength is therefore
1
2
[(+2)2 × b + (−1)2 × 2b]�b −○ = 3b�b−○ .
�e molality arising from dissolving mass m of Ca(NO� )� in a mass m w
of solvent is (m�M)�m w , where M is the molar mass. It therefore follows
that to achieve the desired increase in ionic strength
m 1
3× × = 0.100 hence m= 1
× 0.100 × Mm w b −○
Mm w b−○ 3
m= 1
3
× 0.100 × (164.10 × 10−3 kg mol−1 ) × (0.500 kg) × (1 mol kg−1 )
= 2.73... × 10−3 kg
Hence the �.�� g of NaCl needs to be added to achieve the desired ionic
strength.
E�F.�(a) �e Debye–Hückel limiting law, [�F.��–���], is used to estimate the mean ac-
tivity coe�cient, γ± , at �� ○ C in water
where z± are the charge numbers on the ions from the salt of interest and I is
the ionic strength, de�ned in [�F.��–���]. In the de�nition of I the sum runs
over all the ions in the solution, z i is the charge number on ion i, and b i is its
molality.
180 5 SIMPLE MIXTURES
−A �z+ z− � I 1�2
log γ± = + CI
1 + BI 1�2
Because the electrolyte is �:� with univalent ions, the ionic strength is simply
I = b HBr �b −○ . �ere is no obvious straight-line plot using which the data can
be tested against the Davies equation, therefore a non-linear �t is made us-
ing mathematical so�ware and assuming that A = 0.509; remember that the
molalities must be expressed in mol kg−1 . �e best-�t values are B = 1.96 and
C = 0.0183; With these values the predicted activity coe�cients are �.���, �.���
and �.���, which is very good agreement.
E�F.�(a) �e activity in terms of the vapour pressure p is given by [�F.�–���], a = p�p∗ ,
where p∗ is the vapour pressure of the pure solvent. With the data given a =
p�p∗ = (1.381 kPa)�(2.3393 kPa) = �.���� .
E�F.�(a) On the basis of Raoult’s law, the activity in terms of the vapour pressure p A
is given by [�F.�–���], a A = p A �p∗A , where p∗A is the vapour pressure of the
pure solvent. With the data given a A = p A �p∗A = (250 Torr)�(300 Torr) =
0.833... = �.��� . �e activity coe�cient is de�ned through [�F.�–���], a A =
γ A x A , therefore γ A = a A �x A = 0.833...�0.900 = �.��� .
For the solute, Henry’s law is used as the basis and the activity is given by [�F.��–
���], a B = p B �K B , where K B is the Henry’s law constant expressed in terms of
mole fraction. In this case a B = p B �K B = (25 Torr�200 Torr) = �.��� .
E�F.�(a) On the basis of Raoult’s law, the activity in terms of the vapour pressure p J is
given by [�F.�–���], a J = p J �p∗J , where p∗J is the vapour pressure of the pure
solvent. �e partial vapour pressure of component J in the gas is given by p J =
y J p tot . In this case
p P y P p tot 0.516 × (1.00 atm) × [(101.325 kPa)�(1 atm)]
aP = = = = 0.497...
p∗P p∗P 105 kPa
�e activity of propanone is therefore a P = �.��� . �e activity coe�cient is
de�ned through [�F.�–���], a J = γ J x J , therefore γ P = a P �x P = 0.497...�0.400 =
�.�� .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 181
For the other component the mole fractions are y M = 1 − y P = 0.484 and
x M = 1 − x P = 0.600. �e rest of the calculation follows as before
E�F.�(a) For this model of non-ideal solutions the vapour pressures are given by [�F.��–
���], p A = p∗A x A exp(ξ[1 − x A ]2 ) and likewise for p B ; the total pressure is given
by p tot = p A + p B . �e vapour pressures are plotted in Fig. �.��.
20
pA
pB
15 p tot
p�kPa
10
0
0.0 0.2 0.4 0.6 0.8 1.0
xA
Figure 5.30
Solutions to problems
P�F.� In the Raoult’s law basis the activity is given by [�F.�–���], a J = p J �p∗J , and the
activity coe�cient by [�F.�–���], a J = γ J x J . �e partial pressure of the vapour
is given by p J = y J p, where y J is the mole fraction of J in the vapour and p the
total pressure.
It follows that γ J = a J �x J = y J p�x J p∗J . In this binary system the activity coe�-
cient for �,�-epoxybutane (E) is found using the given data for trichloromethane
(T) by using and x E = 1 − x T , and likewise for y E . It follows that γ E = (1 −
y T )p�(1 − x T )p∗E . �e resulting activity coe�cients are shown in the table.
182 5 SIMPLE MIXTURES
p�kPa xT yT γT γE
23.40 0 0 1
21.75 0.129 0.065 0.417 0.998
20.25 0.228 0.145 0.490 0.958
18.75 0.353 0.285 0.576 0.885
18.15 0.511 0.535 0.723 0.738
20.25 0.700 0.805 0.885 0.563
22.50 0.810 0.915 0.966 0.430
26.30 1 1 1
where z± are the charge numbers on the ions from the salt of interest and I is
the ionic strength, de�ned in [�F.��–���]. For a �:� electrolyte of univalent ions
at molality b, I = b�b −○ (recall that b −○ = 1 mol kg−1 so b must also be in units
of mol kg−1 ). A test of this equation is to plot log γ± against I 1�2 , such a plot is
shown in Fig. �.��.
−A �z+ z− � I 1�2
log γ± = + CI
1 + BI 1�2
�e data can be �tted to this equation using mathematical so�ware to imple-
ment a non-linear �t; the value of A is �xed as 0.509 for the �t. �e best-�t
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 183
0.00
−0.02
log γ±
−0.04
−0.06
�e table shows that the values predicted by the limiting law are increasingly in
error as the ionic strength increases, whereas the Davies equation reproduces
most of the experimental data to within the stated precision. However, it must
be kept in mind that the B and C parameters in the Davies equation have been
adjusted speci�cally to �t these data.
�e given data and the derived values of ν are shown in the following table
(1 µM = 1 µmol dm−3 ), and plotted in Fig. �.��.
6.0
(ν�[EB]out )�(µM−1 )
4.0
2.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
ν
Figure 5.32
K s = γ±v+1 b P b Xv
I�.� In Section �B.�(e) on page ��� the derivation of the expression for the osmostic
pressure starts by equating the chemical potential of A as a pure liquid subject
to pressure p with that of A in a solution of mole fraction x A containing solute B
and subject to pressure p+Π: µ ∗A (p) = µ A (x A , p+Π). �e chemical potential of
A in the solution is then expressed as µ A (x A , p + Π) = µ∗A (p + Π) + RT ln x A ,
which assumes ideality. If the solution is not ideal, then the mole fraction is
replaced by the activity a A to give
µ ∗A (p) = µ ∗A (p + Π) + RT ln a A
I�.� On the basis of Raoult’s law, the activity in terms of the vapour pressure p J
is given by [�F.�–���], a J = p J �p∗J , where p∗J is the vapour pressure of the
pure solvent. �e activity coe�cient is de�ned through [�F.�–���], a J = γ J x J ,
therefore γ J = p J �p∗J x J .
�e data as given do not include values for the vapour pressure over the pure
liquids, so the �rst task is to plot p J against x J and extrapolate to x J = 1 to
�nd p∗J . �e vapour pressures are plotted in this way Fig. �.��, and the lin-
ear extrapolations to �nd the vapour pressures of the pure substances are also
shown. �ese give the values p∗E = 7.45 kPa and p∗B = 35.41 kPa; using these
values the activity coe�cients are computed as shown in the table.
pE
pB
30
p J �kPa
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0
xJ
Figure 5.33
On the basis of Henry’s law, the activity in terms of the vapour pressure p J is
given by [�F.��–���], a J = p J �K J , where K J is the Henry’s law constant for J as
188 5 SIMPLE MIXTURES
To �nd the Henry’s law constant for E, the limiting slope of a plot of p E against
x E is taken. �e three data points given at the lowest values of x E do not ex-
trapolate back to the origin, which is not in accord with Henry’s law. Arguably
there are several equally valid ways of proceeding here, but one is to force the
best-�t line to pass through the origin and then use the �rst three data points;
this leads to the slope is shown by the dashed line in Fig. �.��. �e limiting
slope, taken in this way is 17.77 and so K E = 17.77 kPa. �is value is used to
compute the activity coe�cients for E based on Henry’s law, and the results are
shown in column headed γ E (Henry) in the table above. �e outcome is not
satisfactory because the expecting limiting behaviour γ E → 1 as x E → 0 is not
evidenced.
Using the �nal expression G E �n is computed from the given data and using
the activity coe�cients (based on Raoult’s law) already derived. �e computed
values are given in the table.
I�.� On the basis of Raoult’s law, the activity in terms of the vapour pressure p J is
given by [�F.�–���], a J = p J �p∗J , where p∗J is the vapour pressure of the pure
substance. �e activity coe�cient is de�ned through [�F.�–���], a J = γ J x J ,
therefore γ J = p J �p∗J x J . �e partial pressure in the gas phase is determined
from the mole fraction in the gas phase, y J , p J = y J p tot , so the �nal calculation
is γ J = y J p tot �p∗J x J .
�e total pressure is given in kPa, whereas the vapour pressure over pure oxy-
gen is given in Torr. �e conversion is
95
vapour
liquid
90
T�K
85
80
75
0.0 0.2 0.4 0.6 0.8 1.0
x O2 or y O2
Figure 5.34
I�.�� In Section �B.�(a) on page ��� the derivation of the expression for the freezing
point depression starts by equating the chemical potential of A as a pure solid
with that of A in a solution of mole fraction x A containing solute B: µ ∗A (s) =
µ A (l, x A ). �e latter chemical potential is written, for an ideal solution, as
µ A (l, x A ) = µ∗A (l)+ RT ln x A . If the solution is not ideal, then the mole fraction
is replaced by the activity a A to give
µ∗A (s) = µ ∗A (l) + RT ln a A
�e derivation then proceeds as before. First, µ ∗A (s) − µ ∗A (l) is identi�ed as
−∆ fus G to give
µ ∗ (s) − µ∗A (l) −∆ fus G
ln a A = A =
RT RT
Next, both sides are di�erentiated with respect to T and the Gibbs–Helmholtz
equation, d(G�T)�dT = −H�T 2 , is applied to the right-hand side
d ln a A −1 d ∆ fus G ∆ fus H
= � �=
dT R dT T RT 2
�e freezing point depression ∆T is de�ned as ∆T = T ∗ − T, where T is the
freezing point of the solution and T ∗ is the freezing point of the pure solvent.
190 5 SIMPLE MIXTURES
It follows that d∆T = −dT. Finally, because the freezing point depression is
small, T on the right-hand side of the previous equation can be replaced by T ∗
to give
d ln a A ∆ fus H
=− (�.�)
d∆T RT ∗ 2
�e empirical freezing-point constant K f is introduced in [�B.��–���], ∆T =
K f b B , where b B is the molality of the solvent. �is expression is developed
by writing b B = n B �m A , where m A is the mass of solvent A (in kg) and then
m A = n A M, where M is the molar mass of the solvent. It follows that ∆T =
K f n B �(n A M). For dilute solutions x B ≈ n B �n A so ∆T = K f x B �M.
�e expression for the freezing point depression given in [�B.��–���] is
RT ∗ 2
∆T = xB
∆ fus H
Comparison of this with the expression just derived, ∆T = K f x B �M, gives
Kf RT ∗ 2
=
M ∆ fus H
Using this expression for the right-hand side in eqn �.� gives the required form
d ln a A M
=− (�.�)
d∆T Kf
�is expression for d ln a B is used in eqn �.� and the integral is then evaluated
1
�= � b B d ln a B
bB
1 1 1 1�2
= � b � − 1 ′
A � � � db B
b B b −○
B
bB bB 2
� b B 1�2 �
1 � �
= � �
�
1 − 12 A′ � −○ � � db B
�
bB � b �
bB
1 1 1�2 3�2
= �b B − 12 × 23 × A′ � −○ � b B �
bB b b B =0
b B 1�2
= 1 − 13 A′ � �
b −○
6 Chemical equilibrium
Solutions to exercises
�e equilibrium constant is de�ned by [�A.��–���], K = �∏J a J J �equilibrium .
ν
E�A.�(a)
�e ‘equilibrium’ subscript indicates that the activities are those at equilibrium
rather than at an arbitrary stage in the reaction; however this subscript is not
usually written explicitly. In this case
4
a PH 3 (g)
K = a−1
P 4 (s) a H 2 (g) a PH 3 (g) =
−6 4
6
a P4 (s) a H 2 (g)
�en
∆ r G −○ = 2∆ f G −○ (AgBr, s) + ∆ f G −○ (Cl2 , g)
− {2∆ f G −○ (AgCl, s) + ∆ f G −○ (Br2 , l)}
= 2∆ f G −○ (AgBr, s) − 2∆ f G −○ (AgCl, s)
= 2(−96.90 kJ mol−1 ) − 2(−109.79 kJ mol−1 ) = +25.7... kJ mol−1
�en
∆ r G 1−○ − ∆ r G 2−○
−
○
K 1 e−∆ r G 1 �RT
= −∆ G −○ �RT = exp �− �
K2 e r 2 RT
(−320 × 103 J mol−1 ) − (−55 × 103 J mol−1 )
= exp �− � = 1.4 × 1046
(8.3145 J K−1 mol−1 ) × (300 K)
(i) At Q = 0.010
(ii) At Q = 1.0
(iii) At Q = 10
(iv) At Q = 105
(v) At Q = 106
E�A.�(a) For the reaction 2H2 O(g) � 2H2 (g) + O2 (g) the following table is drawn up
by supposing that there are n moles of H2 O initially and that at equilibrium a
fraction α has dissociated.
2H2 O � 2H2 + O2
Initial amount n � �
Change to reach equilibrium −αn +αn + 12 αn
Amount at equilibrium (1 − α)n αn 1
2
αn
1−α
1
α 2
α
Mole fraction, x J
1 + 12 α 1 + 12 α 1 + 12 α
(1 − α)p
1
αp 2
αp
Partial pressure, p J
1 + 12 α 1 + 12 α 1 + 12 α
used. Treating all species as perfect gases so that a J = (p J �p−○ ), the equilibrium
constant is
α p (1 + 12 α)2
1 3 3
α3 p
= 2
=
(1 − α)2 p2 (1 + 12 α)3 p−○ (1 − α)2 (2 + α) p−○
In this case α = 1.77% (= 0.0177) and p = 1.00 bar; recall that p−○ = 1 bar.
0.01773 1.00 bar
K= × = 2.85 × 10−6
(1 − 0.0177) × (2 + 0.0177)
2 1 bar
∆ν = ν CO + ν H2 O − ν H2 CO = 1 + 1 − 1 = +1 hence K = K c × (c −○ RT�p−○ )
2A + B � 3C + 2D
Initial amount, n J,0 �mol �.�� �.�� � �.��
Change, ∆n J �mol −0.60 −0.30 +0.90 +0.60
Equilibrium amount,
�.�� �.�� �.�� �.��
n J �mol
Mole fraction, x J 0.0869... 0.369... 0.195... 0.347...
Partial pressure, p J (0.0869...)p (0.369...)p (0.195...)p (0.347...)p
To go to the second line, the fact that 0.90 mol of C has been produced is used to
deduce the changes in the other species given the stoichiometry of the reaction.
For example, � mol of A is consumed for every � mol of C produced so ∆ν A =
− 23 ∆ν C = 23 ×+0.90 mol = −0.60 mol. �e total amount in moles is (0.40 mol)+
(1.70 mol) + (0.90 mol) + (1.60 mol) = 4.6 mol. �is value has been used to
�nd the mole fractions. In the last line, p J = x J p has been used.
E�A.�(a) �e reaction Gibbs energy for an arbitrary reaction quotient is given by [�A.��–
���], ∆ r G = ∆ r G −○ + RT ln Q. Treating borneol and isoborneol as perfect gases
so that a J = p J �p−○ , the reaction quotient Q is
Hence
∆ r G = ∆ r G −○ + RT ln Q
= (+9.4 × 103 J mol−1 ) + (8.3145 J K−1 mol−1 ) × (503 K) × ln 2.0
= +12 kJ mol−1
Hence
∆ r G = ∆ r G −○ + RT ln Q
= (−16.5 × 103 J mol−1 ) + (8.3145 J K−1 mol−1 ) × (298 K) × ln(2.30...)
= −14 kJ mol−1
Because ∆ r G < 0 the spontaneous direction of the reaction under these condi-
tions is from le� to right.
198 6 CHEMICAL EQUILIBRIUM
E�A.��(a) �e standard Gibbs energy change for the reaction is given in terms of the
standard Gibbs energies of formation by [�A.��a–���]:
E�A.��(a) In general if the extent of a reaction changes by an amount ∆ξ then the amount
of a component J changes by ν J ∆ξ where ν J is the stoichiometric number for
species J (positive for products, negative for reactants). In this case ν A = −1
and ν B = +2.
E�A.��(a) A reaction is exergonic if ∆ r G < 0 and endergonic if ∆ r G > 0. From the Re-
source section the standard Gibbs energy change for the formation of methane
from its elements in their reference states at 298 K is ∆ f G −○ = −50.72 kJ mol−1 .
�is is negative so the reaction is exergonic .
a C3
Q = a−1
A aB aC =
−2 3
a A a B2
Solutions to problems
P�A.� (a) �e relationship between the equilibrium constant and ∆ r G −○ is [�A.��–
���], ∆ r G −○ = −RT ln K.
(b) �e following table is drawn up. Iodine is not included in the calculations
as it is a solid.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 199
by straightforward algebra.
K p−○
1
α=� �
2
4p + K p−○
0.164 × (1 bar)
1
=� �
2
�� 1+α � (p − p I2 )� p − p I2
2α 2
4α 2
K= = � �
� 1−α
1+α
� (p − p I2 )p−○ (1 + α)(1 − α) p−○
Given the partial pressure of I2 this equation can be solved for α, and p IBr
calculated as before.
200 6 CHEMICAL EQUILIBRIUM
P�A.� �e following table is drawn up for the reaction, assuming that to reach equi-
librium the reaction proceeds by an amount z in the direction of the products.
Rearranging gives
Substituting in the values for n J and K, dividing through by mol2 and writing
x = z�mol yields the quadratic
P�A.� If the extent of reaction at equilibrium is ξ, then from the stoichiometry of the
reaction the amounts of A and B that have reacted are ξ and 3ξ respectively and
the amount of C that has been formed is 2ξ. If the initial amounts of A, B and
C are n, 3n and 0, the following table is drawn up.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 201
A + 3B � 2C
Initial amount n 3n �
Change to reach equilibrium −ξ −3ξ +2ξ
Amount at equilibrium n−ξ 3(n − ξ) 2ξ
n−ξ 3(n − ξ) 2ξ
Mole fraction, x J
4n − 2ξ 4n − 2ξ 4n − 2ξ
(n − ξ)p 3(n − ξ)p 2ξp
Partial pressure, p J
4n − 2ξ 4n − 2ξ 4n − 2ξ
� 4n−2ξ � p−○
2
(p C �p−○ )2
2ξ p 2
p2C p−○
2
a2
K = C3 = = =
a A a B (p A �p−○ )(p B �p−○ )3 p A p3B (n−ξ)p
� 4n−2ξ � � 4n−2ξ �
3(n−ξ)p 3
� �
1
2
1
(ξ�n) = 1 − √
� 1 + 4 27K(p�p−○ ) �
1
1.0
K = 100
0.8
K=1
0.6
ξ�n
0.4
K = 0.01
0.2
0.0
0 10 20 30 40 50
−
○
p�p
Figure 6.1
Solutions to exercises
E�B.�(a) �e relationship between ∆ r G −○ and K is given by [�A.��–���], ∆ r G −○ = −RT ln K.
Hence if K = 1, ∆ r G −○ = −RT ln 1 = 0. Furthermore ∆ r G −○ is related to ∆ r H −○
and ∆ r S −○ by [�D.�–��], ∆ r G −○ = ∆ r H −○ − T∆ r S −○ , so if K = 1
∆ r H −○
∆ r H −○ − T∆ r S −○ = 0 hence T=
∆ r S −○
Values of ∆ r H −○ and ∆ r S −○ at ��� K are calculated using data from the Resource
section.
∆ r H −○ = ∆ f H −○ (CaO, s) + ∆ f H −○ (CO2 , g) − ∆ f H −○ (CaCO3 , s, calcite)
= (−635.09 kJ mol−1 ) + (−393.51 kJ mol−1 ) − (−1206.9 kJ mol−1 )
= +178.3 kJ mol−1
∆ r S −○ = S m
−
○
(CaO, s) + S m
−
○
(CO2 , g) − S m
−
○
(CaCO3 , s, calcite)
= (39.75 J K−1 mol−1 ) + (213.74 J K−1 mol−1 ) − (92.9 J K−1 mol−1 )
= 160.59 J K−1 mol−1
Substituting these values into the equation found above, assuming that ∆ r H −○
and ∆ r S −○ do not vary signi�cantly with temperature over the range of interest,
gives:
∆ r H −○ 178.13 × 103 J mol−1
T= = = 1109 K
∆ r S −○ 160.59 J K−1 mol−1
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 203
E�B.�(a) Treating the vapour as a perfect gas, so that a J = p J �p−○ , and noting that a A2 B = 1
because it is a pure solid, the equilibrium constant for the dissociation A2 B(s) �
A2 (g) + B(g) is
( 12 p)( 12 p) p2
K= =
p−○ 2 4p−○ 2
∆ r H −○ 1 1 R ln(K 2 �K 1 )
ln K 2 − ln K 1 = − � − � hence ∆ r H −○ = −
R T2 T1 (1�T2 ) − (1�T1 )
Noting that the above expression for K implies that ln(K 2 �K 1 ) = ln(p22 �p21 ),
∆ r H −○ is calculated as
∆ r H −○ − ∆ r G −○ ∆ r H −○ + RT ln K ∆ r H −○ p2
∆ r S −○ = = = + R ln � −○ �
T T T 4p
Using the data for 367 ○ C (both temperatures give the same result) gives:
E�B.�(a) For the reaction N2 O4 (g) � 2NO2 (g) the following table is drawn up by
supposing that there are n moles of N2 O4 initially and that at equilibrium a
fraction α has dissociated.
N2 O4 � �NO2
Initial amount n �
Change to reach equilibrium −αn +2αn
Amount at equilibrium (1 − α)n 2αn
1−α 2α
Mole fraction, x J
1+α 1+α
(1 − α)p 2α p
Partial pressure, p J
1+α 1+α
In this case α = 0.1846 and p = 1.00 bar; recall that p−○ = 1 bar.
4 × 0.18462 1.00 bar
K= × = 0.141... = �.���
(1 − 0.1846) × (1 + 0.1846) 1 bar
�at is, K 2 = 1.27 × 109 , a smaller value than at 298 K, as expected for this
exothermic reaction.
E�B.�(a) Assuming that ∆ r H −○ is constant over the temperature range of interest, the
temperature dependence of K is given by [�B.�–���],
∆ r H −○ 1 1
ln K 2 − ln K 1 = − � − �
R T2 T1
Using ∆ r G −○ = −RT ln K to substitute for K 1 and setting ln K 2 = ln 1 = 0 (the
crossover point) gives
∆ r G −○ (T1 ) ∆ r H −○ 1 1
=− � − �
RT1 R T2 T1
Rearranging for T2 gives
Note that this temperature is outside the range over which ∆ r H −○ is known to
be constant and is therefore an estimate.
206 6 CHEMICAL EQUILIBRIUM
� �
∆ r H −○ − ∆ r G −○ 1 � �
∆r S =
−
○
= �−R �B + 2C
� + R �AT + B + ��
C C
= R �A − 2 �
� T �
T T � T � T
���� � � � � � � � � � � � � � � �−○� � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � �� � −�○� � � � � � � � � � � � � � � � � � ���
∆r H −∆ r G
1.51 × 105 K2
= (8.3145 J K−1 mol−1 ) × �−1.04 − � = −16.5 J K−1 mol−1
(400 K)2
� �
d∆ r G −○ � �
∆r S = −
−
○
=−
d �−R �AT + B + �� = R �A − C � C
� T �
dT dT � � T2
���� � � � � � � � � � � � � � � � � � � � � � � �−
○
� � � � � � � � � � � � � � � � � � � � � � � ��
∆r G
E�B.�(a) �e following table is drawn up for the borneol � isoborneol reaction, denoting
the initial amounts of borneol and isoborneol by n b,0 and n iso,0 and supposing
that in order to reach equilibrium an amount z of borneol has converted to
isoborneol.
Borneol � Isoborneol
Initial amount n b,0 n iso,0
Change to reach equilibrium −z +z
Amount at equilibrium n b,0 − z n iso,0 + z
n b,0 − z n iso,0 + z
Mole fraction, x J
n b,0 + n iso,0 n b,0 + n iso,0
Rearranging for z gives z = (Kn b,0 − n iso,0 )�(1 + K). Noting that n = m�M
where M = 154.2422 g mol−1 is the molar mass of borneol and isoborneol,
gives
∆ r H −○ 1 1 R ln(K 2 �K 1 )
ln K 2 − ln K 1 = − � − � hence ∆ r H −○ = −
R T2 T1 (1�T2 ) − (1�T1 )
208 6 CHEMICAL EQUILIBRIUM
Solutions to problems
P�B.� Assuming ∆ r H −○ to be constant over the temperature range of interest, the tem-
perature dependence of K is given by [�B.�–���]
∆ r H −○ 1 1 R ln(K 2 �K 1 )
ln K 2 − ln K 1 = − � − � hence ∆ r H −○ = −
R T2 T1 (1�T2 ) − (1�T1 )
�erefore
P�B.� �e reaction for which ∆ r H −○ is the standard enthalpy of formation of UH3 is:
Treating H2 as a perfect gas (so that a H2 = p H2 �p−○ ) and noting that pure solids
have a J = 1, the equilibrium constant for this reaction is written
−3�2
p−○
3�2
a UH3 1 p
K = 3�2 = = � � = � −○ �
a H2 a U (p�p ) × 1
−
○ 3�2 p p
−3�2
d ln K d p d p
∆ f H −○ (UH3 , s) = RT 2 = RT 2 ln � −○ � = − 32 RT 2 ln � −○ �
dT dT p dT p
d d
= − 32 RT 2 � ln p − ln p−○ � = − 32 RT 2 ln p
dT dT
d B B C
= − 32 RT 2 �A + + C ln T� = − 32 RT 2 �− 2 + �
dT T T T
= − 32 R(CT − B)
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 209
� 12 α 3 �
1�2 1�2
p
= � −○ �
(1 − α)(1 + 12 α)1�2 p
Using this expression, with p = 1 bar, K is calculated at each temperature
and − ln K is plotted against 1�(T�K) as described above; the plot is shown
in Fig. �.�.
14.0
13.0
− ln K
12.0
Figure 6.2
�e equilibrium constant K has already been calculated; from the table above
the value of K at 1443 K is 2.79 × 10−6 .
�e standard reaction Gibbs energy is then calculated using ∆ r G −○ = −RT ln K,
[�A.��–���], and the standard reaction entropy from ∆ r G −○ = ∆ r H −○ − T∆ r S −○
[�D.�–��].
P�B.� �e equilibrium is 2CH3 COOH(g) � (CH3 COOH)2 (g), the dimer being
held together by hydrogen bonds. �e following table is drawn up, assuming
that the initial amount in moles of ethanoic acid is n and that at equilibrium a
fraction α of the ethanoic acid has dimerised.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 211
2pV 2pV
pV = (1 − 12 α)nRT hence α =2− =2−
nRT (m�M)RT
1
� 1 � p−○
αp
2
a(CH COOH)2 (p(CH3 COOH)2 �p−○ ) p(CH3 COOH)2 p−○
K= 2 3 = = =
1− α
2
a CH3 COOH (p CH3 COOH �p−○ )2 p2CH3 COOH (1−α)p
2
� 1 �
1− α
2
1
α(1 − 12 α)p−○
= 2
(1 − α)2 p
�e values of α and K at the two temperatures are then calculated using these
formulae as
At 437 K
At 471 K
2 × (101.9 × 103 Pa) × (21.45 × 10−6 m3 )
α =2− = 0.235...
(0.0380 g�60.0516 g mol−1 ) × (8.3145 J K−1 mol−1 ) × (471 K)
1
α(1 − 12 α)p−○ 1
× (0.235...) × (1 − 12 × 0.235...) × (100 kPa)
K= 2
= 2
= 0.174...
(1 − α)2 p (1 − 0.235...)2 × (101.9 kPa)
= �.���
�e ratio of the values of K that would be obtained using the lowest and highest
values of ∆ r H −○ is
K lowH
−
○ −
○
e−∆ r H low �RT e∆ r S �R ∆ r H low − ∆ r H high
−
○ −
○
= −∆ H −○ �RT = exp �− �
K highH e r high e∆ r S −○ �R RT
For the given data, the value of this factor is
At ��� K:
K lowH (243 − 289) × 103 J mol−1
= exp �− � = 1.2 × 108
K highH (8.3145 J K−1 mol−1 ) × (298 K)
At ��� K:
K lowH (243 − 289) × 103 J mol−1
= exp �− � = 2.7 × 103
K highH (8.3145 J K−1 mol−1 ) × (700 K)
P�B.�� �e standard reaction Gibbs energy is related to the standard reaction enthalpy
and entropy according to [�D.�–��], ∆ r G −○ = ∆ r H −○ − T∆ r S −○ . However, ∆ r H −○
and ∆ r S −○ themselves vary with temperature according to Kirchho� ’s law [�C.�a–
��] for ∆ r H −○ and the analogous equation [�C.�a–��] for ∆ r S −○
T2
∆ r H −○ (T2 ) = ∆ r H −○ (T1 ) + � ∆ r C −p○ dT
T1
T2 ∆ r C −p○
∆ r S −○ (T2 ) = ∆ r S −○ (T1 ) + � dT
T1 T
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 213
A + BT + C�T 2 T2
∆ r S −○ (T2 ) = ∆ r S −○ (T1 ) + � dT
T1 T
T2 A C
= ∆ r S −○ (T1 ) + � + B + 3 dT
T1 T T
T 1 1
= ∆ r S −○ (T1 ) + A ln + B(T2 − T1 ) − 12 C � 2 − 2 �
2
T1 T2 T1
T2
∆ r G −○ (T2 ) = ∆ r G −○ (T1 ) − (T2 − T1 )∆ r S −○ (T1 ) + A �T2 − T1 − T2 ln � ��
T1
+ B � 12 �T22 − T12 � − T2 (T2 − T1 )�
T2 1 1 1 1
+C� � − � − � − ��
2 T22 T12 T2 T1
�e standard Gibbs energy for the formation of H2 O(l) is ∆ r H −○ for the reaction
H2 (g) + 12 O2 (g) → H2 O(l). From the data in the Resource section, at 298 K,
214 6 CHEMICAL EQUILIBRIUM
∆ f S −○ = S m
−
○
(H2 O, l) − S m
−
○
(H2 , g) − 12 S m
−
○
(O2 , g)
= (69.91 − 130.684 − 12 × 205.138) J K−1 mol−1 = −163.343 J K−1 mol−1
�e quantities A, B, and C are also calculated using the data from the Resource
section:
A = a H2 O,l − a H2 ,g − 12 a O2 ,g
= �75.29 − 27.28 − 12 × 29.96� J K−1 mol−1 = 33.03 J K−1 mol−1
B = b H2 O,l − b H2 ,g − 12 b O2 ,g
= �0 − 3.26 − 12 × 4.18� × 10−3 J K−2 mol−1 = −5.35 × 10−3 J K−2 mol−1
C = c H2 O,l − c H2 ,g − 12 c O2 ,g
= �0 − 0.50 − 12 × (−1.67)� × 105 J K mol−1 = 0.335 × 105 J K mol−1
Hence, using the expressions derived above with T1 = 298 K and T2 = 372 K:
∆ f H −○ (298 K)
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
∆ f H −○ (372 K) = (−285.83 × 103 J mol−1 )
+ (33.03 J K−1 mol−1 ) × ([372 − 298] K)
+ 12 (−5.35 × 10−3 J K−2 mol−1 ) × [(372 K)2 − (298 K)2 ]
1 1
− (0.335 × 105 J K mol−1 ) × � − �
372 K 298 K
= −283.49... kJ mol−1
∆ f S −○ (298 K)
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 372 K
∆ f S −○ (372 K) = (−163.343 J K−1 mol−1 ) +(33.03 J K−1 mol−1 ) × ln � �
298 K
+ (−5.35 × 10−3 J K−2 mol−1 ) × [(372 K) − (298 K)]
1 1
− 12 (0.335 × 105 J K mol−1 ) × � − �
(372 K)2 (298 K)2
= −156.34... J K−1 mol−1
�is compares to −237.13 kJ mol−1 at 298 K (from the Resource section). Note
that ∆ f H −○ and ∆ r S −○ do not change very much between 298 K and 372 K in
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 215
this case. In fact, assuming that they are constant gives almost the same value
of ∆ f G −○ (372 K), as is seen by calculating ∆ f G −○ at ��� K using the values of
∆ f H −○ and ∆ f S −○ at ��� K:
which di�ers from the value obtained above by less than 0.3 kJ mol−1 .
6C Electrochemical cells
Answers to discussion questions
D�C.� �e role of a salt bridge is to minimise the liquid junction potential which
would otherwise occur as a result of the contact between the electrolytes in
the two half cells. For a cell to generate a potential these solutions must be
in electrical contact: the salt bridge achieves this without involving a physical
contact between the two solutions.
D�C.� When a current is being drawn from an electrochemical cell, the cell potential
is altered by the formation of charge double layers at the surface of electrodes
and by the formation of solution chemical potential gradients (concentration
gradients). Resistive heating of the cell circuits may occur and junction poten-
tials between dissimilar materials both external and external to the cell may
change.
Solutions to exercises
E�C.�(a) �e reduction half-reactions for the cell in question are:
E cell = E cell
−
○
− (RT�νF) ln Q
216 6 CHEMICAL EQUILIBRIUM
� ���2� � � �� � � � � � �
1
1
a Cd(s) a AgBr(s) 1
Q= 2 2
= 2
a Cd2+ (aq) a Ag(s) a Br − (aq) a Cd2+ (s) a Br − (aq)
�
1
where a J = 1 for pure solids has been used. For ions in solution the activity is
written as a = γ± b�b −○ , where γ± is the mean activity coe�cient, as established
in Section �F.� on page ���. �e Debye–Hückel limiting law [�F.��–���], which
applies at low molalities, is
log γ± = −A�z+ z− �I 1�2
where A = 0.502 for an aqueous solution at 298 K, z+ and z− are the charges
on the ions, and I is the dimensionless ionic strength of the solution which
for a solution containing two types of ion at molality b+ and b− is given by
[�F.��–���], I = 12 (b+ z+2 + b− z−2 )�b −○ .
For the cell in question, the right-hand electrode contains a solution of Cd(NO3 )2
of molality b R = 0.010 mol kg−1 . In this case z+ = +2 (for Cd2+ ), z− = −2 (for
NO−3 ), b+ = b Cd2+ = b R and b− = b NO−3 = 2b R . �e ionic strength is
Putting these activities into the Nernst equation for the cell with ν = 2 and the
expression for Q obtained above gives
RT RT 1
E cell = E cell
−
○
− ln Q = [E −○ (R) − E −○ (L)] − ln � �
νF ��� � � � � � � � � � � � � � � � � � � � � � ��
−
○
����������������������� 2F a a2
Cd 2+ Br−
E cell
which reveal that ν = 2 for the given cell reaction. �e relationship between
∆ r G −○ and E cell
−
○
is given by [�C.�–���], ∆ r G −○ = −νFE cell
−
○
where 1 C V = 1 J is used.
RT RT RT Q2
E cell,1 − E cell,2 = �E cell
−
○
− ln Q 2 � − �E cell
−
○
− ln Q 1 � = − ln � �
νF νF νF Q1
E�C.�(a) (i) �e reduction half-reactions for the cell Zn(s)|ZnSO4 (aq)||AgNO3 |Ag(s),
together with their standard electrode potentials from the Resource sec-
tion, are
R: Ag+ (aq) + e− → Ag(s) E −○ (R) = +0.80 V
L: Zn2+ (aq) + 2e− → Zn(s) E −○ (L) = −0.76 V
218 6 CHEMICAL EQUILIBRIUM
Pt(s)�H2 (g)�HCl(aq)�AgCl(s)�Ag(s)
�e overall cell reaction (R − L) is the same and so, therefore, is the stan-
dard cell potential:
−
○
E cell = (+0.40 V) − (−0.83V) = +1.23 V
Solutions to problems
P�C.� (a) �e reaction of hydrogen and oxygen, 2H2 (g) + O2 (g) → 2H2 O(l), can
be broken down into the reduction half-reactions
R: O2 (g) + 4H+ (aq) + 4e− → 2H2 O(l) E −○ (R) = +1.23 V
L: 4H+ (aq) + 4e− → 2H2 (g) E −○ (L) = 0 (by de�nition)
220 6 CHEMICAL EQUILIBRIUM
(b) �e balanced chemical equation for the combustion of butane, C4 H10 (g)+
13
2
O2 (g) → 4CO2 (g) + 5H2 O(g) can be broken down into the reduction
half-reactions
R: 13 O2 (g) + 26H+ (aq) + 26e− → 13H2 O(l)
L: 4CO2 (g) + 26H+ (aq) + 26e− → C4 H10 (g) + 8H2 O(l)
2
�rst using standard Gibbs energies of formation and then using E cell −
○
=
−∆ r G �νF [�C.�–���] to calculate E cell . Note from the above half-reactions
−
○ −○
that ν = 26.
Hence
for which ν = 2. �e value of E −○ (L) is taken from the Resource section. �e cell
reaction (R − L) is
Q(aq) + 2H+ (aq) + 2Hg(l) + 2Cl− (aq) → QH2 (aq) + Hg2 Cl2 (s)
−
○
E cell = E −○ (R) − E −○ (L) = (+0.6994 V) − (+0.27 V) = +0.4294 V
Noting that ν = 2 and that a J = 1 for pure solids and liquids, the Nernst
equation is
RT a QH2
E cell = E cell
−
○
− ln � 2 a2
�
2F aQ aH + Cl−
Taking a QH2 = a Q , because Q and QH2 are present at the same concentration,
and a H+ = a Cl− gives:
RT 1 2RT 2RT
E cell = E cell
−
○
− ln � 4 � = E cell
−
○
+ ln a H+ = E cell
−
○
+ ln 10 × log a H+
2F a H+ F F
2RT ln 10
= E cell
−
○
− × pH
F
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 221
6D Electrode potentials
Answer to discussion questions
D�D.� �is is discussed in Impact ��.
Solutions to exercises
E�D.�(a) �e reduction half-reactions for the given cell are
νF −○ 1 × (96485 C mol−1 )
ln K = E cell = × (0.9509 V) = 37.0...
RT (8.3145 J K−1 mol−1 ) × (298.15 K)
E�D.�(a) (i) �e reduction half-reactions for the speci�ed cell and their corresponding
electrode potentials from the Resource section are
R: Cu2+ (aq) + 2e− → Cu(s) E −○ (R) = +0.34 V
L: 2Ag+ (aq) + 2e− → 2Ag(s) E −○ (L) = +0.80 V
�e overall cell reaction is
∆ r G −○ = −νFE cell
−
○
= −2 × (96485 C mol−1 ) × (−0.46 V) = 88.7... kJ mol−1
= +89 kJ mol−1
E�D.�(a) Assuming that the mercury forms Hg2 SO4 (s) in the reaction, the required
reduction half-equations and the corresponding standard electrode potentials
are
νF −○ 2 × (96485 C mol−1 )
ln K = E cell = × (+0.29 V) = 22.5...
RT (8.3145 J K−1 mol−1 ) × (298 K)
Solutions to problems
P�D.� �e given reaction can be broken down into the following reduction half equa-
tions
R: 2Fe3+ (aq) + 2e− → 2Fe2+ (aq)
L: Ag2 CrO4 (s) + 2e− → 2Ag(s) + CrO2−
4 (s)
where the K+ and Cl− spectator ions have been ignored. �ese half-equations
show that ν = 2 for the given reaction.
E −○ (L) = E cell
−
○
− E −○ (R) = (+0.77 V) − (+0.323... V) = +0.45 V
224 6 CHEMICAL EQUILIBRIUM
P�D.� (a) �e equilibrium HCO−3 (aq) � CO2− 3 (aq) + H (aq) is broken down into
+
∆ r G −○ = ∆ f G −○ (CO2−
3 , aq) − ∆ f G (HCO3 , aq)
−
○ −
Hence
RT � a HCO−3 a OH− �
E cell = E cell
−
○
− ln
F � a CO2−
3
�
(d) �e standard cell potential corresponds to all species involved in the cell
reaction, which includes OH− , being present at unit activity. �is means
that the pH will need to be approximately ��, in order to give a OH− = 1. At
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 225
pH �.�, the concentration of OH− will be lower than at pH ��, which will
mean that the cell reaction as written above will have a greater tendency
to move in the forward direction. As a result E cell is predicted to be larger
at pH �.� than when a OH− = 1.
Assuming that the activities of all other species remain the same, the
change in cell potential on going from a OH− = 1 to pH � is
� � � − × 1 ��
�
� −○ RT � a HCO−3 a OH− �� �
� − �E −○ − RT ln � HCO3
a
∆E cell = �
� cell
E − ln � � cell
�
�
� F � a CO2− � � � F � a � �
� � � �
2−
CO
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
3 3
Noting that a J = 1 for pure solids and that in this cell a H2 = 1 because the
hydrogen is at standard pressure, the Nernst equation is
RT
E cell = E cell
−
○
− ln (a Cl− a H+ )
F
In addition, the base B and its conjugate acid are in equilibrium:
a B a H+
BH+ (aq) � B(aq) + H+ (aq) Ka =
a BH+
�e expression for K a is rearranged to give a H+ = K a a BH+ �a B and this is sub-
stituted into the Nernst equation to give
RT RT a Cl− a BH+ K a
E cell = E cell
−
○
− ln (a Cl− a H+ ) = E cell
−
○
− ln � �
F F aB
Replacing activities by a J = γ J (b J �b −○ ) [�F.��–���] gives
RT γ Cl− γ BH+ bK a RT γ 2 bK a
E cell = E cell
−
○
− ln � × −○ � = E cell
−
○
− ln � ± −○ �
F γB b F b
where the mean activity coe�cient of the BH+ and Cl – ions is given by [�F.��–
���], γ± = (γ Cl− γ BH+ )1�2 and the neutral base B is assumed to be an ideal solute
so that γ B = 1. Noting from inside the front cover that ln x = ln 10 log x, the
Nernst equation becomes
RT ln 10 γ 2 bK a
E cell = E cell
−
○
− log � ± −○ �
F b
RT ln 10 b
= E cell
−
○
− �2 log γ± + log � −○ � − pK a �
F b
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 227
where pK a = − log K a has been used. Next the Davies equation [�F.��b–���],
log γ± = −A�z+ z− �I 1�2 �(1+BI 1�2 )+CI, is used to substitute for log γ± . �e ionic
strength I is given by [�F.��–���], I = 12 ∑ i z 2i (b i �b −○ ), where z i is the charge
on ion species i and the sum extends over all the ions present in the solution.
In this case, b BH+ = b Cl− = b, and b H+ is neglected because it will be much
smaller on account of the equilibrium involving the base B. �erefore the ionic
strength is
I = 12 �z BH
2
+ b + z Cl− b� �b
2 −
○
= 12 �12 × b + (−1)2 × b� �b −○ = b�b−○
and therefore
Substitution of this expression into the Nernst equation derived above gives
RT ln 10 A(b�b−○ )1�2 b b
E cell = E cell
−
○
− �2 �− + C � −○ �� + log � −○ � − pK a �
F 1 + B(b�b )
−○ 1�2 b b
which rearranges to
De�ning (b�b −○ )1�2 as x and the le�-hand side as y, and introducing A = 0.5091
gives
1.0182x
y= − 2Cx 2 − 2 log x + pK a
1 + Bx
which is �tted to the data using mathematical so�ware to give the following val-
ues for the parameters: B = 2.54 , C = −0.204 , and pK a = 6.74 . �ese values
have been used to draw the line on the graph shown in Fig. �.�.
8.8
)�RT ln 10
8.6
8.2
8.0
0.05 0.10 0.15 0.20 0.25 0.30 0.35
(b�b )
−
○ 1�2
Figure 6.3
I�.� From Impact � the reaction for the hydrolysis of ATP to ADP and inorganic
phosphate P−i is
In an environment in which pH = 7.0 and the ATP, ADP and P−i concentrations
are all �.� mmol dm−3 , the reaction Gibbs energy is given by [�A.��–���],
∆ r G = ∆ r G ⊕ + RT ln Q ⊕
a ADP a Pi a H3 O+
∆ r G = ∆ r G −○ + RT ln � �
a ATP a H2 O
setting all the activities to � except for that for H� O+ which is set to 10−7 gives
∆r G ⊕
∆ r G ⊕ = ∆ r G −○ + RT ln 10−7
hence
∆ r G −○ = ∆ r G ⊕ − RT ln 10−7
= (−31 × 103 J mol−1 )−(8.3145 J K−1 mol−1 )×([37 + 273.15] K)×ln 10−7
= +11 kJ mol−1
�e right-hand half reaction is multiplied by three and the le�-hand half re-
action by two in order that both involve the same number of electrons. �e
overall reaction is
2CH3 CH2 OH(aq) + 3NO−3 (aq) + 6H+ (aq) → 4CO2 (g) + 3NH+4 (aq) + 3H2 O(l)
230 6 CHEMICAL EQUILIBRIUM
inside the front cover that ln x = ln 10 log x, and also that pH = − log a H+ , and
assuming T = 298 K, gives
1
∆ r G = ∆ r G −○ + RT ln Q = ∆ r G −○ + RT ln � 6
�
aH +
T�K K 1�(T�K) − ln K
233 4.13 × 108 0.004 29 −19.8
248 5.00 × 107 0.004 03 −17.7
258 1.45 × 107 0.003 88 −16.5
268 5.37 × 106 0.003 73 −15.5
273 3.20 × 106 0.003 66 −15.0
280 9.62 × 105 0.003 57 −13.8
288 4.28 × 105 0.003 47 −13.0
295 1.67 × 105 0.003 39 −12.0
303 6.02 × 104 0.003 30 −11.0
−10
−15
− ln K
−20
0.0032 0.0036 0.0040 0.0044
1�(T�K)
Figure 6.4
(b) �e standard reaction enthalpy for the reaction 2ClO(g) → (ClO)2 (g) is
expressed in terms of standard enthalpies of formation
∆ r H −○ = ∆ f H −○ [(ClO)2 , g] − 2∆ f H −○ (ClO, g)
232 6 CHEMICAL EQUILIBRIUM
Hence
∆ f H −○ [(ClO)2 , g] = ∆ r H −○ + 2∆ f H −○ (ClO, g)
= (−73.0... kJ mol−1 ) + 2 × (+101.8 kJ mol−1 )
= +131 kJ mol−1
Similarly
∆ r S −○ = S m
−
○
[(ClO)2 , g] − 2S m
−
○
(ClO, g)
Hence
−
○
Sm [(ClO)2 , g] = ∆ r S −○ + 2S m
−
○
(ClO, g)
= (−1.46... × 102 J K−1 mol−1 ) + 2 × (226.6 J K−1 mol−1 )
= 307 J K−1 mol−1
I�.� (a) �e ionic strength is given by [�F.��–���], I = 12 �b+ z+2 + b− z−2 � �b −○ , where
z+ and z− are the charges on the ions. For the CuSO4 compartment,
z+ = 2, z− = −2, and b+ = b− = b CuSO4 :
Because the charges are the same for ZnSO� it follows that I = 4(b ZnSO4 �b−○ )
= 1.20 × 10−2 .
(b) According to the Debye–Hückel limiting law (Section �F.�(b) on page
���), the mean activity coe�cient is given by [�F.��–���], log γ± = −A�z+ z− �I 1�2 ,
where A = 0.509 for aqueous solutions at �� ○ C. For the CuSO4 solution
Hence γ±,CuSO4 = 10−0.128 ... = 0.743... = 0.743 . For the ZnSO4 solution
is given by
RT
E cell = E cell
−
○
− ln Q
νF
(8.3145 J K−1 mol−1 ) × ([25 + 273.15] K)
= (+1.102 V) − × ln(2.41...)
2 × (96485 C mol−1 )
= +1.09 V
Because the standard electrode potential for the le�-hand half-reaction is zero
by de�nition, the standard cell potential for this cell is equal to the standard
electrode potential of the H2 O/H2 ,OH− electrode. �e equilibrium constant
−
○
K w for the cell reaction is then given by [�C.�–���], E cell = (RT�νF) ln K.
Rearranging for ln K and using ν = 1, K = K w , and E cell = E −○ (H2 O�H2 , OH− )
−
○
gives
F
ln K w = × E −○ (H2 O�H2 , OH− )
RT
Noting from inside the front cover that ln x = ln 10 × log x, and also that pK w =
− log K w allows the above equation to be rewritten as
ln K w F
pK w = − log K w = − =− × E −○ (H2 O�H2 , OH− )
ln 10 RT ln 10
�e task is therefore to �nd E −○ (H2 O�H2 , OH− ) from the given data. To do
this, the speci�ed cell is written in terms of its reduction half-reactions
Noting that a J = 1 for pure solids, and that in this cell a H2 = 1 because the
hydrogen is at standard pressure, the Nernst equation for the cell is
RT RT a Cl−
E cell = E cell
−
○
− ln Q = E cell
−
○
− ln � �
νF F a OH−
RT γ± (b Cl− �b −○ ) RT b Cl−
E cell = E cell
−
○
− ln � � = E cell
−
○
− ln � �
F γ± (b OH− �b )−○ F b OH−
�e standard cell potential is split into contributions from the two electrodes
−
○
using [�D.�–���], E cell = E −○ (R) − E −○ (L)
RT b Cl−
E cell = E −○ (AgCl�Ag, Cl− ) − E −○ (H2 O�H2 , OH− ) − ln � �
F b OH−
Hence
RT b Cl−
E −○ (H2 O�H2 , OH− ) = E −○ (AgCl�Ag, Cl− ) − E cell − ln � �
F b OH−
�is equation is used with b Cl− = 0.01125 mol kg−1 , b OH− = 0.0100 mol kg−1 ,
and the values of E cell and E −○ (AgCl�Ag, Cl− ) to calculate E −○ (H2 O�H2 , OH− )
at each temperature. �e relation derived earlier
θ�○ C T�K E cell �V E −○ (AgCl�Ag, Cl− )�V E −○ (H2 O�H2 , OH− )�V pK w
20.0 293.15 1.047 74 0.225 02 −0.825 70 14.20
25.0 298.15 1.048 64 0.222 30 −0.829 37 14.02
30.0 303.15 1.049 42 0.219 59 −0.832 91 13.85
−0.825
−0.835
292 294 296 298 300 302 304
T�K
Figure 6.5
∆ r S −○ = νF × slope × V K−1
= 1 × (96485 C mol−1 ) × (−7.229 × 10−4 V K−1 )
= −68.5... J K−1 mol−1 = −68.6 J K−1 mol−1
Solutions to exercises
E�A.�(a) If the power, P, is constant, the total energy emitted in time ∆t is P∆t. �e
energy of each emitted photon is E photon = hν = hc�λ. �e total number of
photons emitted in this time period is therefore the total energy emitted divided
238 7 QUANTUM THEORY
υ = P∆t�cmglow-worm
(0.10 W) × (10 y) × (3.1536 × 107 s y−1 )
= = �� m s−1
(2.9979 × 108 m s−1 ) × (0.0050 kg)
h 6.6261 × 10−34 J s
υ= = = 7.27 × 106 m s−1
m e λ (9.1094 × 10−31 kg) × (100 × 10−12 m)
h 6.6261 × 10−34 J s
υ= = = 2.4 × 10−2 m s−1
m e λ (9.1094 × 10−31 kg) × (3 × 10−2 m)
(iii)
6.6261 × 10−34 J s
λ=
((4.00 × 10−3 kg mol−1 )�(6.0221 × 1023 mol−1 )) × (1.00 × 103 m s−1 )
= ��.� pm
E�A.�(a) Wien’s law [�A.�–���], λ max T = 2.9 × 10−3 m K, is rearranged to give the
wavelength at which intensity is maximised
E�A.�(a) Assuming that the object is a black body is equivalent to assuming that Wien’s
law [�A.�–���], λ max T = 2.9 × 10−3 m K, holds. Using ν̃ = λ−1 , Wien’s law is
expressed in terms of the wavenumber of maximum intensity (ν̃ max )
E�A.�(a) Molar heat capacities of monatomic non-metallic solids obey the Einstein re-
lation [�A.�a–���],
θ E 2 eθ E �2T
2
C V ,m (T) = 3R f E (T), f E (T) = � � � θ �T �
T e E −1
where the solid is at temperature T and is characterized by an Einstein temper-
ature θ E . �us, for a solid at 298 K with an Einstein temperature of 2000 K
E�A.�(a) �e energy of the quantum is given by the Bohr frequency condition [�A.�–
���], ∆E = hν, and the frequency is ν = 1�T. �e energy per mole is ∆E m =
N A ∆E.
(i) For T = 1.0 fs
(ii) For T = 10 fs
E�A.��(a) When a photon is absorbed by a free hydrogen atom, the law of conservation of
energy requires that the kinetic energy acquired by the atom is E k , the energy
of the absorbed photon. Assuming relativistic corrections are negligible the
kinetic energy is E k = E photon = 12 m H υ 2 . �e atom is accelerated to the speed
=� �
(1.0079 × 10−3 kg mol−1 )
= (3.45... × 1013 m s−1 ) × �E photon �J�
1�2
�e photon energies have been calculated in Exercise E�A.��(a), and thus the
following table is drawn up
(i) P = 1 W
(1 W) × (1 s)(550 × 10−9 m)
N= = 2.77 × 1018
(6.6261 × 10−34 J s) × (2.9979 × 108 m s−1 )
(ii) P = 100 W
E�A.��(a) As described in Section �A.� on page ���, photoejection can only occur if the
energy of the incident photon is greater than or equal to the work function of
the metal �. If this condition is ful�lled, the energy of the emitted photon is
given by [�A.��–���], E k = hν − Φ = hc�λ − Φ. To convert the work function to
Joules, multiply through by the elementary charge, as described in Section �A.�
on page ���,
�is is less than the threshold energy, hence no electron ejection occurs.
(ii) For λ = 300 nm
�is is greater than the the threshold frequency, and so photoejection can
occur. �e kinetic energy of the electron is,
�
υ= 2E photon �m e
�
= 2 × (3.19 × 10−19 J)�(9.109 × 10−31 kg) = 837 km s−1
Solutions to problems
P�A.� A cavity approximates an ideal black body, hence the Planck distribution [�A.�a–
���], applies
8πhc
ρ(λ, T) =
λ 5 �e hc�λk T − 1�
242 7 QUANTUM THEORY
Because the wavelength range is small (5 nm), the energy density is approxi-
mated by
∆E(T) = ρ(λ, T)∆λ
Taking λ = 652.2 nm gives
8πhc
ρ=
λ 5 �1 + x + 1 2
2!
+
x 1 3
3!
x ... − 1�
When x << 1 the second and higher order terms in x become negligibly small
compared to x. Consequently
P�A.� Each data point is used to �nd a value for λ max T, and then the mean of these is
used with λ max T = hc�(4.965k) to �nd a value of h, h = (4.965k�c) (λ max T)mean .
�e following table is drawn up
�is is the Stefan–Boltzmannn law, that the total energy density is proportional
to T 4 , with a constant of proportionality of 8π 5 k 4 �15(hc)5 .
P�A.� Assuming the Sun approximates an ideal black body, Wien’s law, λ max T = 2.9 ×
10−3 m K, applies. Hence, λ max = (2.9×10−3 m K)�(5800 K) = ��� nm which
corresponds to blue-green .
7B Wavefunctions
Answers to discussion questions
D�B.� �is is discussed in Section �B.�(b) on page ���.
Solutions to exercises
E�B.�(a) (i) �e function ψ = x 2 cannot be normalized as the area under ψ ∗ ψ = x 4
is in�nite when the integral is evaluated over all space. However, if the
function were restricted to a �nite region of space, it could be normalized.
(ii) �e function ψ = 1�x cannot be normalized as it goes to ∞ as x goes to
0 so the integral of ψ ∗ ψ over all space is in�nite. However, if the function
were restricted to a �nite region of space which did not include x = 0 it
could be normalized.
(iii) �e function ψ = exp(−x 2 ) can be normalized as ψ ∗ ψ = exp(−2x 2 )
goes to 0 at x = ±∞ so that the integral of ψ ∗ ψ over all space is �nite.
244 7 QUANTUM THEORY
E�B.�(a) In two dimensions �ψ(x, y)�2 dxdy is the probability of �nding the particle in
the in�nitesimal area dxdy at the position (x, y). �e probability is a dimen-
sionless quantity, and the area dxdy has units of (length)2 . Hence, �ψ(x, y)�2
must have units of (length)−2 , implying that ψ has units of length−1 .
Solutions to problems
P�B.� (a) �e task is to �nd N such that ψ = Nei� satis�es the normalization condi-
tion [�B.�c–���], ∫ ψ ∗ ψ dτ = 1. In this case the integration is over � and
the range is 0 to 2π. �e function is complex so ψ ∗ = N exp(−i�); note
that the normalization factor N is always real. �e integrand is therefore
ψ ∗ ψ = e−i� ei� = e−i�+i� = e0 = 1
�e integral to evaluate is therefore
2π
N2 � 1 d� = N 2 ��0 = 2πN 2
2π
0
and similarly the integral over y is equal to L y �2. Hence the integral is
�
evaluated as N 2 (L x �2)(L y �2). Setting this equal to 1 gives N = 2� L x L y .
(b) In the case when L x = L y = L this reduces to N = 2�L .
P�B.� �e task is to �nd N such that ψ(x) = Ne−ax , satis�es the normalization
condition [�B.�c–���], ∫ ψ ∗ ψ dτ = 1. �e integration is over x ranging from 0
to ∞.
∞ ∞
N2 � e−2ax dx = −(N 2 �2a) e−2ax �0 = −(N 2 �2a)(e−∞ − e0 ) = N 2 �2a
0
where e−∞ = 0 and e0 = 1 are used. Setting this equal to 1 gives N = (2a)1�2 .
Hence, the total probability of �nding the particle at a distance x ≥ x 0 is
∞ ∞ ∞
� �ψ(x)�2 dx = 2a � e−2ax dx = −(2a�2a) e−2ax �x = e−2ax 0
x0 x0 0
where a 0 = 53 pm is the Bohr radius. If the sphere has radius 1.0 pm then the
expression evaluates to
4 × (1.0 pm)3
× e−2r�a 0 = 8.95... × 10−6 × e−2r�a 0
3 × (53 pm)3
P�B.�� �e probability �nding the atom between x and x +dx, where dx is an in�nites-
imal displacement, is given by
� d x2 d e−x �a �
2 2
dP(x) d
= �N 2 x 2 e−x �a � = N 2 × e−x �a + x 2 ×
2 2 2 2
dx dx � dx dx �
2x −x 2 �a 2 x 2
= N 2 �2xe−x �a 2
− x2 × � = 2 −x 2 �a 2
�1 − � � �
2
e 2xN e
a2 a
Solutions to exercises
E�C.�(a) Two wavefunctions ψ i and ψ j are orthogonal if ∫ ψ ∗i ψ j dτ = 0, [�C.�–���].
In this case the integration is from � = 0 to � = 2π. Let ψ i = exp(i�), the
wavefunction with m l = +1, and ψ j = exp(2i�), the wavefunction with m l =
+2. Note that the functions are complex, so ψ ∗i = exp(−i�). �e integrand is
therefore
ψ ∗i ψ j = exp(−i�) exp(2i�) = exp(i�)
and the integral evaluates as
=1 =1
2π ��� � � � � � � � � � � �� � � � � � � � � � � �� ��� � � � � � � � ��� � � � � � � � � �
� exp(i�) d� = (1�i) exp(i�)�0 = (1�i)[exp(i × 2π) − exp(i × 0)] = 0
2π
0
E�C.�(a) For the case when m l = +1 the normalized wavefunction is ψ+1 (�) = (2π)−1�2 ei� .
�is is complex, and so ψ+1 ∗
= (2π)−1�2 e−i� . �e expectation value of the
position, speci�ed by the operator �, is
2π 2π
��� = � ∗
ψ+1 �ψ+1 d� = (1�2π) � e−i� � ei� d�
0 0
2π
= (1�2π) � � d� = (1�2π) × (� 2 �2)�0 = π
2π
0
which evaluates to give the same result, ���m l = π. �is result is interpreted
by noting that the probability density around the ring is (1�2π)e−im l � eim l � =
1�2π, which is constant. �erefore the average position is half way round the
ring at � = π.
1.0546 × 10−34 J s
= 7.01 × 10−10 m
2 × (7.52... × 10−26 kg m s−1 )
E�C.�(a) To construct the potential energy operator, replace the position x in the clas-
sical expression by the operator for position x̂ to give V̂ = 12 k f x̂ 2 . However,
because x̂ = x× the potential energy operator is V̂ = 12 k f x 2 .
operator d�dx.
where sin(nπ) = 0 for integer n is used. �us, the two wavefunctions are
orthogonal.
L�2
� ψ 1 ψ 2 dτ = �
∗
cos(πx�L) cos(3πx�L) dx
−L�2
where sin(nπ) = 0 for integer n is used. �us, the two wavefunctions are
orthogonal.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 251
Solutions to problems
P�C.� Operate on each function f with i,ˆ the inversion operator, which has the e�ect
of making the replacement x → −x. If the result of the operation is f multiplied
by a constant, f is an eigenfunction of iˆ and the constant is the eigenvalue.
ˆ 3 − kx) = (−x)3 − k(−x) = −(x 3 − kx). Yes , f is an eigenfunction,
(a) i(x
eigenvalue −1 .
ˆ
(b) i(cos(kx)) = cos(k(−x)) = cos(kx). Yes , f is an eigenfunction, eigen-
value +1 .
ˆ 2 + 3x − 1) = (−x)2 + 3(−x) − 1 = x 2 − 3x − 1. No , f is not an
(c) i(x
eigenfunction.
∗
P�C.� An operator Ω̂ is hermitian if ∫ ψ ∗i Ω̂ψ j dτ = �∫ ψ ∗j Ω̂ψ i dτ� , [�C.�–���]. For
the kinetic energy operator, it is necessary to integrate by parts (�e chemist’s
toolkit �� in Topic �C on page ���) twice
ħ 2 d2 ħ2 ∞ 2
∗ d ψj
� ψ i �− � = −
∗
ψ j dx � ψ i dx
2m dx 2 2m −∞ dx 2
� �
ħ2 � � ħ2
∞
∞ dψ ∗ dψ ∞ dψ ∗ dψ
� ∗ j dψ �
=− � ψi � −� dx � =
� 2m �−∞ dx dx
i j i j
2m �
dx
� dx −∞
−∞ dx dx �
���� � � � � � � � � � ��
A
������������ �
� �
∞ dψ ∗ dψ 2 � ∗ ∞ ∞ d2 ψ ∗ �
ħ2
�
j
dx =
ħ � i ψ j� − �
dψ
ψ j dx �
2m � �
i i
2m −∞ dx dx � dx −∞ −∞ dx 2
�
���� � � � � � � � � � �� � � � � � � � � � � ��
B
�
ħ 2 d2 ∞ ħ 2 d2
� ψ i �− � = �− � ψ ∗i dx
∗
ψ j dx � ψ j
2m dx 2 −∞ 2m dx 2
Note that because the complex conjugate of the whole term is taken, to com-
pensate for this the complex conjugate of the terms inside the bracket need to
be taken too ψ = [ψ ∗ ]∗ . �is �nal equation is consistent with [�C.�–���] and
so demonstrates that the kinetic energy operator is hermitian.
252 7 QUANTUM THEORY
P�C.� (a) Consider the sum of two arbitrary hermitian operators  and B̂,
∗
As  is hermitian, ∫ ψ ∗i Âψ j dτ = �∫ ψ ∗j Âψ i dτ� and similarly for B̂.
∗ ∗
� ψ i ( + B̂)ψ j dτ = �� ψ j Âψ i dτ� + �� ψ j B̂ψ i dτ�
∗ ∗ ∗
∗
= �� ψ ∗j ( + B̂)ψ i dτ�
which demonstrates that the sum of two hermitian operators is also her-
mitian.
(b) As Ω̂ψ j = ψ k is also a function, the integral can be rewritten
� ψ i Ω̂ Ω̂ψ j dτ = � ψ i Ω̂ψ k dτ
∗ ∗
= � ψ ∗l ψ k dτ
= � ψ ∗l Ω̂ψ j dτ
To go to the second line the two functions have been re-ordered, which
is always permitted, and to go to the third line the de�nition Ω̂ψ j = ψ k
has been used. Because Ω̂ is hermitian the �nal term can be rewritten as
[∫ ψ ∗j Ω̂ψ l dτ]∗ and so
∗
� ψ i Ω̂ψ k dτ = �� ψ j Ω̂ψ l dτ�
∗ ∗
In the term on the right the de�nition Ω̂ψ i = ψ l is used, and in the term
on the le� Ω̂ψ j = ψ k is used to give
∗
� ψ i Ω̂ Ω̂ψ j dτ = �� ψ j Ω̂ Ω̂ψ i dτ�
∗ ∗
P�C.� (a) �e expectation value is given by [�C.��–���], �Ω� = ∫ ψ ∗ Ω̂ψ dτ, where
ψ is normalized. A hermitian operator has the property ∫ ψ ∗i Ω̂ψ j dτ =
[∫ ψ ∗j Ω̂ψ i dτ]∗ , which for the case ψ i = ψ j = ψ becomes
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � � � � � � � �∗
A B
� �
� ψ Ω̂ψ dτ = � �
∗ ∗
� � ψ Ω̂ψ dτ �
� �
� �
�e quantities A and B and the same, so it follows that A = B∗ = A∗ ,
which is only true if A is real. �e quantity A is the expectation value, so
it follows that the expectation value of a hermitian operator is real.
(b) �e expectation value of the square of the hermitian operator Ω̂ Ω̂ is
�ΩΩ� = � ψ ∗ Ω̂ Ω̂ψ dτ
−∞
as the integrand is odd and the integration is over a symmetric range. For
�x 2 � the required integral is
∞ ∞
�x 2 � = (2a�π)1�2 � x 2 e−2ax dx = 2(2a�π)1�2 � x 2 e−2ax dx
2 2
−∞ 0
254 7 QUANTUM THEORY
ħ 2a 1�2 ∞
�p x � = � ψ ∗ p̂ x ψ dτ = � � � −2axe−2ax dx = 0
2
i π −∞
ħ d
p̂2x ψ = p̂ x ( p̂ x )ψ = �(ħ�i)(2a�π)1�4 × [−2axe−ax ]�
2
i dx
�is is equal to
which gives
�p2x � = � ψ ∗ p̂2x ψ dτ
+∞
= −ħ 2 (2a�π) � (4a 2 x 2 e−2ax − 2ae−2ax )
1�2 2 2
−∞
� �
2a 1�2 � 2 π 1�2 �
1�2
= −ħ 2 � � × � �
π
� 2a � � − 2a � � �
π � (2a)3 2a �
� �
= ħ a
2
P�C.�� To evaluate the commutator of the operators consider the e�ect of the commu-
tator on an arbitrary function ψ; use the product rule to evaluate the derivatives
of products as necessary.
(a)
d 1 d ψ 1 dψ 1 dψ dx −1 1 dψ
� , �ψ = � �− = +ψ −
dx x dx x x dx x dx dx x dx
1
=− 2 ψ
x
�e commutator is therefore identi�ed as the term multiplying ψ, −1�x 2 .
(b)
d 2 d(x 2 ψ) dψ dψ dx 2 dψ
� , x �ψ = − x2 = x2 +ψ − x2
dx dx dx dx dx dx
= 2x ψ
�e commutator is therefore identi�ed as the term multiplying ψ, 2x .
P�C.�� To complete this problem it is useful to establish some general properties of
commutators. First, an operator  commutes with powers of itself: for example,
[Â, Â2 ] = 0. �is is proved by simply multiplying out the commutator:
[Â, Â2 ] = Â(ÂÂ) − (ÂÂ)Â = 0
Note that multiplying either of the terms by a constant does not alter this prop-
erty. Second, any operator commutes with a constant: [Â, c] = 0, which follows
trivially on multiplying out the commutator. Finally, the useful property that
[Â + B̂, Ĉ] = [Â, Ĉ] + [B̂, Ĉ]. Again, this follows by multiplying out the com-
mutator
[Â + B̂, Ĉ] = (Â + B̂)Ĉ − Ĉ(Â + B̂) = ÂĈ − Ĉ Â + B̂Ĉ − Ĉ B̂ = [Â, Ĉ] + [B̂, Ĉ]
(a) For the case V̂ = V0 :
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � �
A B
[Ĥ, x̂] = [ p̂2x �2m + V0 , x̂] = [ p̂2x �2m, x̂] + [V0 , x̂]
Reintroducing the constants gives the result [Ĥ, x̂] = −(ħ 2 �m)(d�dx)
For the case V̂ = 12 k f x 2 :
7D Translational motion
Answers to discussion questions
D�D.� �e physical origin of tunnelling is related to the probability density of the
particle, which according to the Born interpretation is the square of the wave-
function that represents the particle. �is interpretation requires that the wave-
function of the system be everywhere continuous, even at barriers. �erefore,
if the wavefunction is non zero on one side of a barrier it must be non zero on
the other side of the barrier and this implies that the particle has tunnelled into
the barrier. �e transmission probability depends upon the mass of the par-
ticle: the greater the mass the smaller the probability of tunnelling. Electrons
and protons have small masses, molecular groups large masses; therefore, tun-
nelling e�ects are more observable in process involving electrons and protons.
Solutions to exercises
E�D.�(a) �e energy levels of a particle in a box with length L are given by [�D.�–���],
E n = h 2 n 2 �8mL 2 and when the length is increased to 1.1 × L the energies
become E n′ = h 2 n 2 �8m(1.1 × L)2 = h 2 n 2 �8m(1.21 × L 2 ) giving a fractional
change of
E n′ − E n [h 2 n 2 �8m(1.21 × L 2 )] − [h 2 n 2 �8mL 2 ] 1
= = − 1 = −0.174
En h n �8mL
2 2 2 1.21
h 2 (n + 1)2 h 2 n 2 h2
∆E(n) = E n+1 − E n = − = �(n + 1)2 − n 2 �
8mL 2 8mL 2 8mL 2
h2 h2
= (n 2 + 2n + 1 − n 2 ) = (2n + 1)
8mL 2 8mL 2
Setting ∆E(n) to the average thermal energy gives h 2 (2n + 1)�8mL 2 = kT�2,
and so (2n + 1) = 4mkTL 2 �h 2 , leading to
2mkTL 2
n= − 1
2
h2
For a helium atom, mass 4.00 m u in a box for length 1 cm,
sin (2πx�L) = ±1
2πx�L = π�2, 3π�2 hence x = L�4, 3L�4
and similarly for y. Hence the maxima in probability density occurs at (x, y) =
(L�4, L�4), (L�4, 3L�4), (3L�4, L�4), (3L�4, 3L�4)
258 7 QUANTUM THEORY
Nodes occur when the wavefunction passes through zero, which is when either
of the sin terms are zero, excluding the boundaries at x = 0, L y = 0, L because
at these points the wavefunction does not pass through zero. �ere is thus a
node when sin (2πx�L) = 0, corresponding to x = L�2 and any value of y. �is
node is therefore a line at x = L�2 and parallel to the y axis . Similarly there is
another nodal line at y = L�2 and parallel to the x axis .
E�D.�(a) �e energy levels for a �D rectangular box, side lengths L 1 , L 2 are
h 2 n 12 n 22
E n 1 ,n 2 = � + �
8m L 21 L 22
where n 1 and n 2 are integers greater than or equal to 1.
For the speci�c case where L 1 = L, L 2 = 2L,
h 2 n 12 n 22 h2 n 22
E n 1 ,n 2 = � 2+ � = �n 2
+ �
8m L (2L)2 8mL 2 1
4
�e energy of the state with n 1 = n 2 = 2 is then
h2 22 h2
E 2,2 = �22 + � = (5)
8mL 2 4 8mL 2
�is is degenerate with the state with n 1 = 1, n 2 = 4
h2 42 h2
E(1,4) = �1 2
+ � = (5)
8mL 2 4 8mL 2
1
= n 2 × (6.02... × 10−20 J) × = n 2 × (0.376... eV)
1.6022 × 10−19 C
(6.02... × 10−20 J)
= n2 × = n 2 × (3.03... × 103 cm−1 )
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
�e values in the other units are found by using the appropriate value of the
constant, computed above.
(i) ∆E(1, 2) = 1.8 × 10−19 J , 1.1 × 102 kJ mol−1 , �.� eV , or 9.1 × 103 cm−1
(ii) ∆E(5, 6) = 6.6 × 10−19 J , 4.0 × 102 kJ mol−1 , �.� eV , or 3.3 × 104 cm−1
ψ1
ψ 12
x�L
0.2 0.4 0.6 0.8 1.0
Figure 7.1
E�D.��(a) �e wavefunction � of the lowest energy state for a particle in a box has n = 1
and is ψ 1 (x) = 2�L sin(πx�L), which leads to a probability density P1 (x) =
�ψ 1 (x)�2 = (2�L) sin2 (πx�L). Graphs of these functions are shown in Fig �.�.
�e probability density is symmetric about x = L�2. �erefore, there is an equal
probability of observing the particle at an arbitrary position x ′ and at L − x ′ , so
it follows that the average position of the particle must be at L�2.
�
E�D.��(a) �e wavefunction for the state with n = 1 is ψ 1 (x) = 2�L sin(πx�L), which
leads to a probability density P1 (x) = �ψ 1 (x)�2 = (2�L) sin2 (πx�L); these are
plotted in Fig �.�.
�e plotted probability density gives the probability of the particle being found
in an interval at a particular value of x – that is, it is the probability density of
x. It is not the probability density of x 2 , so there is no simple way of inferring
from the plot the value of �x 2 �. �e fact that the probability density of x is
symmetric about L�2 does not imply that the probability density of x 2 has the
same property. �ere is therefore no reason to assume that �x 2 � = (L�2)2 .
E�D.��(a) For an electron in a square well of side L the energy of the state characterized
by quantum numbers n 1 and n 2 is given by [�D.��b–���], E n x ,n y = h 2 (n 2x +
n 2y )�8m e L 2 , where n x , n y are integers greater than or equal to one. Hence, the
minimum or zero-point energy has n x = n y = 1,
Setting this equal to the rest energy m e c 2 , and rearranging for the length, L
h 2 �4m e L 2 = m e c 2
L = h�2m e c = λ C �2
Solutions to problems
P�D.� �e energy levels of a particle in a one dimensional box are E n = h 2 n 2 �8mL 2 .
�e mass of an O� molecule is 2 × 16.00 amu × 1.6605 × 10−27 kg = 5.3136 ×
10−26 kg. �e separation in energy of the lowest two levels, being n = 1 and
n = 2, is given by
h 2 22 h 2 12 h2
∆E = E 2 − E 1 = − = (4 − 1)
8mL 2 8mL 2 8mL 2
(6.6261 × 10−34 J s)2
= ×3
8 × (5.3136 × 10−26 kg) × (5.0 × 10−2 m)
= 6.2 × 10−41 J
To �nd the value of n that makes the energy equal to the average thermal energy,
set E n = 12 kT and solve for n
√
2L mkT
n=
h
2×(5.0 × 10−2 m)×�(5.3136 × 10−26 kg)×(300 K)×(1.3806 × 10−23 J K−1 )�
1�2
=
6.6261 × 10−34 J s
= 2.23... × 109 = 2.2 × 109
h 2 n 2 h 2 (n − 1)2 h 2 (2n − 1)
E n − E n−1 = − =
8mL 2 8mL 2 8mL 2
(6.6261 × 10 J s) × (2.23... × 109 − 1)
−34 2
=
8 × (5.3136 × 10−26 kg) × (5.0 × 10−2 m)2
= 1.8 × 10−30 J
= L�2
L L
�x 2 � = � ψ ∗1 x 2 ψ 1 dx = (2�L) � x 2 sin2 (πx�L) dx
0 0
P�D.� (a) Assuming that the system can be modelled as a �D particle in a box, the
energy levels of the system are given by [�D.�–���], E n = h 2 n 2 �8mL 2 .
Hence, the separation between the levels n and n + 1 is
(c) �e terms in the energy expression that change with the number of con-
jugated atoms N are the quantum number of the highest energy occupied
state, n, and the length of the box, L. �e energy, and hence frequency, of
the transition is directly proportional to 2n +1 and inversely proportional
to L 2 . Because the there are N electrons which occupy the states in pairs,
the quantum number of the highest occupied state is n = N�2 (rounded
up if N is odd). �e length L is proportional to the number of bonds
in the molecule, which is N − 1. Hence, the frequency of the transition is
proportional to (2(N�2)+1)�(N −1)2 = (N +1)�(N −1)2 ≈ N −1 , and so
the absorption spectrum of a linear polyene shi�s to a lower frequency
as the number of conjugated atoms increases .
and for n = 2,
P�D.� (a) In �D the Schrödinger equation with a potential energy function V (x, y, z)
is given by
ħ2 ∂2 ∂2 ∂2
− � 2 + 2 + 2 � ψ + V (x, y, z)ψ = Eψ
2m ∂x ∂y ∂z
In this case V (x, y, z) = 0 inside the box and in�nite elsewhere. It fol-
lows that the wavefunction is zero outside the box. Inside the box the
Schrödinger equation is
ħ2 ∂2 ∂2 ∂2
− � 2 + 2 + 2 � ψ = Eψ
2m ∂x ∂y ∂z
ħ2 d2 X d2 Y d2 Z
− �Y Z 2 + XZ 2 + Y Z 2 � = EXY Z
2m dx dy dz
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 265
��
(�,�,�)
(�,�,�),(�,�,�)
(�,�,�)
(�,�,�)
(�,�,�)
(�,�,�)
��
(�,�,�)
E n x ,n y ,n z �(h 2 �8ml 2 ) (�,�,�)
(�,�,�)
(�,�,�)
(�,�,�)
(�,�,�)
��
(�,�,�)
(�,�,�)
(�,�,�)
�
Figure 7.2
(d) Comparing the �D and �D energy diagrams, the one dimensional case is
more sparse than the three dimensional case, as the ��eenth level is not
reached until E n �(h 2 �8ml 2 ) = 225. In addition, none of the levels in the
one-dimensional case are degenerate.
T1 16ε(1 − ε)e−2κ L 1
= = e−2κ(L 1 −L 2 )
T2 16ε(1 − ε)e−2κ L 2
T2
e2k 2 W (k 2 − ik 3 ) e2k 2 W (k 2 − ik 3 )
A 2 �1 + � = 2ik 1 − A 2 k 2 �1 − �
k 2 + ik 3 k 2 + ik 3
�is can be rearranged to give
−1
1 e2k 2 W (k 2 − ik 3 ) k 2 (k 2 − ik 3 ) k 2 e2k 2 W (k 2 − ik 3 )
A2 = � + + − �
2 2(k 2 + ik 3 ) 2ik 1 (k 2 + ik 3 ) 2ik 1 (k 2 + ik 3 )
Hence,
−1
A 2 e2k 2 W (k 2 − ik 3 ) k 2 + ik 3
B2 = = A 2 � 2k W �
k 2 + ik 3 e 2 (k 2 − ik 3 )
−1
k 2 + ik 3 1 k2 k2
=� + + − �
2e 2k 2 W (k 2 − ik 3 ) 2 2ik 1 e 2k 2 W 2ik 1
�is gives
4k 1 k 2 eik 2 W
A 3 = e−ik 3 W (A 2 e k 2 W + B 2 e−k 2 W ) =
(ia + b)e k 2 W − (ia − b)e−k 2 W
2k 1 k 2 eik 2 L
A3 =
(ia + b) sinh(k 2 W) + be−k 2 W
268 7 QUANTUM THEORY
a 2 + b 2 = (k 12 + k 22 )2 = k 24 [1 + (k 1 �k 2 )2 ]2
b 2 = 4k 12 k 22
Hence
−1
b2 a2 + b2
T= = �1 + sinh2 k 2 W�
b 2 + (a 2 + b 2 ) sinh (k 2 W)
2
b2
−1 −1
sinh2 (k 2 W) [ 1 (e k 2 W − e−k 2 W )]2
= �1 + � = �1 + 2 �
4ε(1 − ε) 4ε(1 − ε)
−1
(e k 2 W − e−k 2 W )2
= �1 + � (�.�)
16ε(1 − ε)
�
P�D.�� (a) �e particle in a box wavefunctions are given by ψ n (x) = 2�L sin(nπx�L),
giving a probability density of Pn (x) = �ψ n (x)�2 = (2�L) sin2 (nπx�L).
�is probability density is plotted in Fig. �.� for n = 1 . . . 5, and for n = 50
in Fig. �.�.
�e correspondence principle is that as the quantum number becomes
large the quantum result must converge with the classical result. For a
particle in a box, the classical result is an even probability. As n increases
the oscillations become more rapid leading to a more uniform distribu-
tion especially if it is averaged over short distance which is a small fraction
of the width of the box.
(b) �e transmission probability of a particle of mass m passing through a
barrier of height V0 , length W is given by, [�D.��a–���]
−1
(eκW − e−κW )2
T = �1 + �
16ε(1 − ε)
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 269
0.2
T
0.1
0.0
10 12 14 16 18 20
−1
V2 �kJ mol
Figure 7.3
1.0 n=1
n=2
n=3
n=4
n=5
ψ 2n �(2�L)
0.5
0.0
0.0 0.2 0.4 0.6 0.8 1.0
x�L
Figure 7.4
� �
where ε = E�V0 and κ = 2m(V0 − E)�ħ = 2mV0 (1 − ε)�ħ. A plot of
T versus ε is shown in Fig. �.� for the passage by a H� molecule, a proton,
and an electron.
�e curves in the �gure di�er in the value of W(mV0 )1�2 �ħ , a measure of
the obstruction that the barrier represents taking into account its height,
width and the type of particle. �e curves can be thought of as having the
same value of W and V0 , but di�ering only in m. �e values of W and V0
were chosen such that the proton and hydrogen molecule exhibit ‘typical’
tunnelling behaviour: if the incident energy is small enough, there is
practically no transmission, and if the incident energy is high enough,
transmission is virtually complete. A barrier through which a proton and
hydrogen molecule exhibit tunnelling behaviour is practically no barrier
for an electron on account of the much smaller mass of the latter. On this
plot, T for the electron is essentially �.
(c) �e wavefunction of a �D particle in a box with quantum numbers n 1 and
270 7 QUANTUM THEORY
1.0
ψ 50
2
�(2�L)
0.5
0.0
0.0 0.2 0.4 0.6 0.8 1.0
x�L
Figure 7.5
1.0 H�
p+
e–
T
0.5
0.0
0.0 1.0 2.0 3.0 4.0
ε
Figure 7.6
n 2 is given by [�D.��a–���]
As can be seen from the plots, the function sin(n 1 πx�L x ) has n 1 −1 nodes
where the function passes through zero; the boundaries do not count
as nodes because the wavefunction does not pass through zero at these
points. Similarly the function sin(n 2 πy�L y ) has n 2 − 1 nodes. In two
dimensions a node in the wavefunction along x or y leads to a nodal line,
and the total number of these is (n 1 − 1) + (n 2 − 1) = n 1 + n 2 − 2 .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 271
n 1 = 1, n 2 = 1
1
0
0 0.2 0.5
0.4 0.6 0.8 1 0 y�L y
x�L x
n 1 = 1, n 2 = 2
psi�(2�L x )1�2 (2�L y )1�2
1
−1
0 0.2 0.5
0.4 0.6 0.8 1 0 y�L y
x�L x
Figure 7.7
7E Vibrational motion
Answers to discussion questions
D�E.� If the harmonic oscillator were to have zero energy the particle would be at
rest and located at the bottom of the potential energy well (x = 0). Such
an arrangement has no uncertainty in either the position or the momentum,
which is not in accord with the uncertainty principle. By giving the oscillator
zero-point energy, there is some uncertainty in the position as the particle
oscillates back and forth, and then it is possible for there to be the appropriate
uncertainty in the momentum such that the uncertainty principle is satis�ed.
�e existence of zero-point energy is thus traced to the need to satisfy the
uncertainty principle.
D�E.� For the harmonic oscillator the spacing of the energy levels is constant. �ere-
fore, relative to the energy of the oscillator, the spacing becomes progressively
smaller as the quantum number increases. In the limit of very high quantum
numbers this spacing becomes negligible compared to the total energy, and
e�ectively the energy can take any value, as in the classical case.
272 7 QUANTUM THEORY
n 1 = 2, n 2 = 1
1
−1
0 0.2 0.5
0.4 0.6 0.8 1 0 y�L y
x�L x
n 1 = 2, n 2 = 2
psi�(2�L x )1�2 (2�L y )1�2
1
−1
0 0.2 0.5
0.4 0.6 0.8 1 0 y�L y
x�L x
Figure 7.8
As is shown in Fig. �E.� on page ���, as the quantum number becomes large the
probability density clusters more and more around the classical turning points
of the classical harmonic oscillator (that is, the points at which the kinetic
energy is zero). Because the classical oscillator is moving most slowly near
these points, they are the displacements at which it is most probable that the
oscillator will be found. Again, the quantum and classical results converge at
high quantum numbers
Solutions to exercises
E�E.�(a) �e wavefunctions are depicted in Fig. �E.� on page ���. �e general form of
the harmonic oscillator wavefunctions is ψ υ = N υ H υ (y)e−y �2 , where N υ is
2
ing the wavefunction, setting the result to 0 and solving for y. �e di�erential
is evaluated using the product rule
dψ 1 (y) � dy 2 −y 2 de−y �
2
= N 12 e + y2 = N 12 [2ye−y + y 2 × (−2ye−y )]
2 2
dy � dy dy �
= N 12 2y(1 − y 2 )e−y
2
= 4.30 × 10−21 J
k f = (1.33 × 10−25 kg) × �(4.82 × 10−21 J)�(1.0546 × 10−34 J s)� = 278 N m−1
2
given
= 2.64 × 10−6 m
E�E.�(a) �e wavefunctions are depicted in Fig. �E.� on page ���; they are real. Two
wavefunctions are orthogonal if ∫ ψ ∗i ψ j dτ = 0. In this case the wavefunctions
are ψ 0 (y) = N 0 e−y �2 and ψ 1 (y) = N 1 ye−y �2 , and the integration is from
2 2
meaning that its value at −y is the negative of its value at +y. �e integral of
an odd function over a symmetric range is zero, hence these wavefunctions are
orthogonal.
E0 = 1
2
× (1.0546 × 10−34 J s)
× �(329 N m−1 )�( 12 × 34.9688 × 1.6605 × 10−27 kg)�
1�2
= 5.61 × 10−21 J
E�E.�(a) For the state with υ = 0 the energy of a harmonic oscillator is given by [�E.�–
���], E 0 = 12 ħω, where the frequency ω is given by [�E.�–���], ω = (k f �m)1�2 .
(i) For the system with k f = 1000 N m−1 the energy of the state with υ = 0 is
E0 = 1
2
× (1.0546 × 10−34 J s) × [(1000 N m−1 )�(1 × 1.6605 × 10−27 kg)]1�2
= 4.09... × 10−20 J = 4.09 × 10−20 J
(ii) For the system with k f = 100 N m−1 the energy of the state with υ = 0 is
E0 = 1
2
× (1.0546 × 10−34 J s) × [(100 N m−1 )�(1 × 1.6605 × 10−27 kg)]1�2
= 1.29 × 10−20 J
Solutions to problems
P�E.� �e intermolecular potential is electrostatic in origin and arises from the in-
teraction between the protons and electrons, the number and distribution of
which remain the same on isotopic substitution and so therefore does the po-
tential. �e force constant, which re�ects the shape of the intermolecular po-
tential, is therefore una�ected by such a change.
(a) �e force constants for the molecules are the same therefore ω A′ B �ω AB =
(k f �µ A′ B )1�2 �(k f �µ AB )1�2 = (µ AB �µ A′ B )1�2 ; multiplying through by ω AB ,
gives ω A′ B = ω AB (µ AB �µ A′ B )1�2 . �e vibrational frequency in Hz, ν, is
related to ω by ν = ω�2π, hence ν A′ B = ν AB (µ AB �µ A′ B )1�2 .
(b) �e e�ective mass of �H ��Cl is µ 1 = (1 × 35)�(35 + 1) = 0.972... m u , that
of �H ��Cl is µ 2 = (2 × 35)�(35 + 2) = 1.89... m u , and that of �H ��Cl is
µ 3 = (1 × 37)�(1 + 37) = 0.973... m u .
�e vibrational frequency of �H ��Cl is
P�E.� Assuming that the force constant is the same � for all � the molecules,
� then the
ratio of frequencies is ω 2 �ω 1 = ν 2 �ν 1 = k f �µ 2 � k f �µ 1 = µ 1 �µ 2 . For a
homonuclear diatomic molecule A� the e�ective mass is µ A2 = m A2 �2m A =
m A �2. Hence, µ 1H2 = 1�2 = 0.5 m u , µ 2H2 = 2�2 = 1 m u , µ 3H2 = 3�2 = 1.5 m u .
� �
ν 2H2 = ν 1H2 µ 1H2 �µ 2H2 = 131.9 THz × 0.5 m u �1 m u = 93.27 THz
� �
ν 3H2 = ν 1H2 µ 1H2 �µ 3H2 = 131.9 THz × 0.5 m u �1.5 m u = 76.15 THz
P�E.� (a) �e wavenumber, ν̃ is given by ν̃ = ν�c, where c is the speed of light (see
�e chemist’s toolkit ��). �e frequency in Hz is related to the angular
frequency by ν = ω�2π, hence ν̃ = ω�2πc .
(b) �e vibrational frequency of �H ��Cl is 5.63×1014 s−1 , which gives a wave-
number of
ν̃ = (5.63 × 1014 s−1 )�[2π × (2.9979 × 1010 cm s−1 )] = 2.99 × 103 cm−1
Note that in order to express the wavenumber in cm−1 , the speed of light
is used in cm s−1 .
(c) �e frequency ω is given by [�E.�–���], ω = (k f �µ) , and so the vibra-
1�2
P�E.� If the C atom is immobilized, only the O will be moving, and the force constant
will be the same as that in the CO molecule. �e wavenumber of the CO
vibration is 2170 cm−1 , as given in Problem P�E.�, and as the force constant is
the�same this is linked to the wavenumber of the vibration when bound as ν̃ 2 =
ν̃ 1 µ 1 �µ 2 , where 1 refers to the free molecule and 2 that bound to the haem
group. �e e�ective mass of ��C ��O is µ 1 = (12×16)�(12+16) = 6.85... m u , and
in the bound case the e�ective mass is simply the mass of the O atom, 16 m u ,
as this is the only atom which is moving. Hence
�
ν̃ 2 = (2170 cm−1 ) × (6.85... m u )�(16 m u ) = 1420 cm−1
acts on e−gx the result is not a constant times that function. However, this
2
Solving this for g gives g = (mk f )1�2 �2ħ ; this is the value of g needed
for e−gx to be an eigenfunction of the hamiltonian. �e resulting wave-
2
[�E.�b–���].
(c) Returning to eqn �.�, if the terms in x 2 cancel the remaining terms are
−(ħ 2 �2m)2g(−1)e−gx
2
�e energy is indeed the same as that of the lowest level of the harmonic
oscillator as described in the text.
P�E.�� (a) �e variable y is de�ned as y = x�α where α = (ħ 2 �mk f )1�4 , and the ki-
netic energy operator in terms of x is T̂ = −(ħ 2 �2m)d2 �dx 2 . Substituting
in y leads to
ħ 2 d2 ħ 2 d2
T̂ = − =− 2
2m d(α y) 2 2α m dy 2
ħ 2 mk f 1�2 d2 k f 1�2 d2 d2
=− � 2 � = − 1
2
ħ � � = − 1
2
ħω
2m ħ dy 2 m dy 2 dy 2
where for the last line the de�nition of the frequency ω is used, ω =
(k f �m)1�2 .
(b) �e expectation value of the kinetic energy of the state with quantum
number υ is given by
∞ d2
�T�υ = − 12 ħωN υ2 � H υ e−y �2
H υ e−y �2 dy
2 2
−∞ dy 2
d e−y �2
2
d2 d d
−y 2 �2
= (H ′ −y 2 �2
+ )= (H ′ − yH υ )e−y �2
2
H υ e υ e H υ
dy 2 dy dy dy υ
�e product rule is applied once more
d(H ′υ − yH υ ) −y 2 �2 d e−y �2
2
d2
H υ e−y �2 = + (H ′υ − yH υ )
2
e
dy 2 dy dy
= (H ′′υ − yH ′υ − H υ )e−y �2
− y(H ′υ − yH υ )e−y �2
2 2
From Table �E.� on page ��� one of the properties of the Hermite polyno-
mials is H ′′υ − 2yH ′υ + 2υH υ = 0, so it follows that H ′′υ − 2yH ′υ = −2υH υ .
Using this in the last line above gives
d2
H υ e−y �2 = −(2υ + 1)H υ e−y �2 + y 2 H υ e−y �2
2 2 2
dy 2
278 7 QUANTUM THEORY
(c) A further property of the Hermite polynomials is H υ+1 −2yH υ +2υH υ−1 =
0, which rearranges to yH υ = υH υ−1 + 12 H υ+1 . Using this, the term y 2 H υ
is rewritten
�e same relationship is used to substitute for the two terms on the right,
but rewritten �rstly by making the substitution υ → υ − 1 to give yH υ−1 =
(υ − 1)H υ−2 + 12 H υ , and secondly with with the substitution υ → υ + 1 to
give yH υ+1 = (υ + 1)H υ + 12 H υ+2 . With these substitutions
d2
H υ e−y �2 = �−(2υ + 1)H υ + y 2 H υ � e−y �2
2 2
dy 2
= �−(2υ + 1)H υ + υ(υ − 1)H υ−2 + (υ + 12 )H υ + 14 H υ+2 � e−y �2
2
−∞
− (υ + 12 )H υ + 14 H υ+2 ]e−y �2
2
dy
∞ ∞
= − 12 ħωN υ2 �υ(υ − 1) � H υ H υ−2 e−y dy + 14 � H υ H υ+2 e−y dy
2 2
−∞ −∞
∞
− (υ + )
2 �
1
Hυ Hυ e −y 2
dy�
−∞
�e �rst two integrals are zero due to orthogonality of the Hermite poly-
nomials, leaving
∞
�T�υ = 12 (υ + 12 )ħωN υ2 � H υ2 e−y dy
2
(�.�)
−∞
−∞
A property of the Hermite polynomials is (Table �E.� on page ���)
∞
� H υ2 e−y dy = π 1�2 2υ υ!
2
−∞
�T�υ = 12 (υ + 12 )ħω = 12 E υ
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 279
P�E.�� As is shown in Example �E.� on page ���, in terms of the dimensionless variable
y the classical turning points are at y tp = ±(2υ + 1)1�2 , where υ is the quantum
√
number of the state. For the �rst excited state υ = 1, and so y tp = ± 3.
�e wavefunction is ψ 1 = N 1 ye−y �2
2
, which is normalized by solving
+∞ +∞
N 12 � y 2 e−y dy = 2N 12 � y 2 e−y dy = 1
2 2
−∞ 0
3 π 1�2 3
−∞
∞
= α 2 N υ N υ′ � H υ′ (yH υ )e−y dy
2
−∞
Using the properties of the Hermite polynomials given in Table �E.� on page
���, the term yH υ can be rewritten as yH υ = 12 H υ+1 +υH υ−1 , and so the integral
can be rewritten as
∞ ∞
I = α 2 N υ N υ′ � 12 � H υ′ H υ+1 e−y dy + υ � H υ′ H υ−1 e−y dy�
2 2
(�.�)
−∞ −∞
∞
Due to orthogonality of the Hermite polynomials, the integral ∫−∞ H υ′ H υ e−y dy
2
−∞
∞
It is a property of the Hermite polynomials that ∫−∞ H υ2 e−y dy = π 1�2 2υ υ! so
2
In the case that υ′ = υ − 1 only the second integral in eqn �.� is non-zero. Using
∞
the same procedure as before with ∫−∞ H υ−1 e−y dy = [π 1�2 2υ−1 (υ − 1)!]1�2 ,
2
2
gives
∞
I = α 2 N υ N υ−1 υ � e−y dy
2
2
H υ−1
−∞
1 1 α2 √
= υα 2 1�2 υ 1�2 1�2 υ−1 [π 1�2 υ−1
2 (υ − 1)!] = υ
[π 2 υ!] (π 2 (υ − 1)!)1�2 21�2
7F Rotational motion
Answers to discussion questions
D�F.� �e vector model of angular momentum is described in Section �F.�(c) on page
���. �e model captures the key idea that the magnitude and the z-component
of the angular momentum can be known simultaneously and precisely. How-
ever, if this is so there is no knowledge of the other components of the angular
momentum other than that the sum of their squares must be equal to a partic-
ular value.
D�F.� Rotational motion on a ring and on a sphere share the following features: (a)
quantization arising as a result of the need to satisfy a cyclic boundary con-
dition; (b) energy levels which go inversely with the moment of inertia; (c)
the lack of zero-point energy; (d) degeneracy; (e) quantization of the angular
momentum about one axis.
Solutions to exercises
E�F.�(a) �e diagrams shown in Fig. �.� are drawn by forming a vector of length [l(l +
1)]1�2 and√with a projection m l on the z-axis. For l = 1, m l = +1 the vector is
of length √2 and has projection +1 on the z-axis; for l = 2, m l = 0 the vector is
of length 6 and has projection 0 on the z-axis. Each vector may lie anywhere
on a cone described by rotating the vector about the z-axis.
z
+1 l = 1, m l = 1
0
l = 2, m l = 0
Figure 7.9
�e nodes occur when Y3,0 passes through zero, which � happens when either
cos θ = 0 or 5 cos2 θ − 3 = 0, that is when cos θ = ± 3�5; recall that θ is in the
range 0 to π. Hence, the nodes are θ = π�2, 0.684, 2.46 . �ere are 3 angular
nodes.
�
E�F.�(a) �e real part of the spherical harmonic Y1,+1 is − 12 3�π sin θ cos �. When �
varies, an angular node therefore occurs when cos � = 0, i.e. at � = π�2, 3π�2 .
�is plane corresponds to the plane x = 0, i.e. the yz plane.
�
�e imaginary part of the same spherical harmonic is − 12 3�π sin θ sin �. When
� varies, an angular node therefore occurs when sin � = 0, i.e. at � = 0, π. �is
plane corresponds to the plane y = 0, i.e. the xz plane.
282 7 QUANTUM THEORY
E�F.�(a) �e rotational energy depends only on the quantum number l [�F.��–���], but
there are distinct states for every allowed value of m l , which can range from −l
to +l in integer steps. �ere are 2l + 1 such states, as there are l of these with
m l > 0, l of these with m l < 0 and m l = 0. Hence l = 3 has a degeneracy of � .
E�F.�(a) �e magnitude of the angular momentum associated with a wavefunction with
angular momentum quantum number l is given by [�F.��–���], magnitude =
ħ[l(l + 1)]1�2 . Hence for l = 1 the magnitude is ħ[1(1 + 1)]1�2 = 21�2 ħ .
�e projection of the angular momentum onto the z-axis is given by [�F.�–���],
ħm l , where m l is a quantum number that takes values between −l and +l in
integer steps, m l = −l , −l + 1, . . . + l. Hence the possible projections onto the
z-axis are −ħ, 0, ħ .
E�F.�(a) �e wavefunction of a particle on a ring, with quantum number m l is ψ m l =
eim l � = cos(m l �) + i sin(m l �) in the range 0 ≤ � ≤ 2π. �e real and imaginary
parts of the wavefunction are therefore cos(m l �) and sin(m l �) respectively.
Nodes occur when the function passes through zero, which for trigonometric
functions are the same points at which the function is zero. Hence in the real
part, nodes occur when cos(m l �) = 0, and so when m l � = (2n + 1)π�2 for
integer n, which gives � = (2n + 1)π�2m l . In the imaginary part, nodes occur
when sin(m l �) = 0 and so when m l � = nπ for an integer n, which gives � =
nπ�m l .
(i) With m l = 0 the real part is a constant and has no nodes ; the imaginary
part is zero everywhere.
(ii) With m l = 3, nodes in the real part occur at π�6, π�2, 5π�6, 7π�6, 3π�2 ,
11π�6 . In the imaginary part nodes occur at 0, π�3, 2π�3, π, 4π�3, 5π�3 .
�ere are � nodes in each of the parts.
E�F.�(a) �e normalization condition is ∫ ψ ∗m l ψ m l dτ = 1. In this case the integral is
over � in the range 0 ≤ � ≤ 2π, and the wavefunction is ψ m l = Neim l � . Hence,
noting that the wavefunction is complex
2π 2π
N 2 � ψ ∗m l ψ m l dτ = N 2 � e−im l � eim l � d� = N 2 � d� = 2πN 2
0 0
is complex and so ψ ∗m ′ = e−im l � . Note that both m l and m′l are integers and
′
2π 2π
e−im l � eim l � d� = �
′ ′
� ei(m l −m l )� d�
0 0
E�F.��(a) �e energy levels are [�F.��–���], E l = ħ 2 l(l + 1)�2I, where I is the moment
of inertia. �e minimum energy to start it rotating is the minimum excitation
energy, the energy to take it from the motionless l = 0 to the rotating l = 1
state, ∆E = E 1 − E 0 = (ħ 2 �2I)(1(1 + 1) − 0(0 + 1)] = ħ 2 �I. Evaluating this gives
E�F.��(a) �e energy levels are [�F.��–���], E l = ħ 2 l(l + 1)�2I, where I is the moment
of inertia. So that the excitation energy is ∆E = E 2 − E 1 = (ħ 2 �2I)[2(2 + 1) −
1(1 + 1)] = 2ħ 2 �I. Evaluating this gives
E�F.��(a) �e diagrams shown in Fig. �.�� are drawn by forming a vector of length [l(l +
√
1)] 1�2
and with a projection m l on the z-axis. For l = 1 the vector is of length
√ 2 and has projection −1, 0, +1 on the z-axis. For l = 2 the vector is of length
6 and has projection −2, . . . + 2 in integer steps on the z-axis. Each vector
may lie anywhere on a cone described by rotating the vector about the z-axis.
z
+2 +2
+1 +1
z
+1 +1 0 0
0 0 −1 −1
−1 −1 −2 −2
Figure 7.10
284 7 QUANTUM THEORY
E�F.��(a) �e angle in question is that between the z-axis and the vector representing the
angular momentum. �e projection� of the vector onto the z-axis is m l ħ, and
the length of the vector is ħ l(l + 1). �erefore the angle θ that the vector
�
makes to the z-axis is given by cos θ = m l � l(l + 1).
� √
When m l = l, cos θ = l� l(l + 1), which for l = 1 gives cos θ = 1� 2, and so
√
θ = π�4 , and for l = 5 gives cos θ = 5� 30, and so θ = 0.420 .
Solutions to problems
P�F.� �e angular momentum states have quantum numbers m l = 0, ±1, ±2.... �e
energy levels for a particle on a ring are given by [�F.�–���], E m l = m 2l ħ 2 �2I,
and have angular momentum [�F.�–���], J z = m l ħ. �e moment of inertia for
an electron on this ring is I = mr 2 = (9.1094 × 10−31 kg) × (440 × 10−12 m)2 =
1.76... × 10−49 kg m2
(a) If there are 22 electrons in the system the highest occupied state will be
the degenerate levels m l = ±5. �ese states have an energy of E±5 =
(±5)2 (1.0546 × 10−34 J s)2 �2(1.76... × 10−49 kg m2 ) = 7.88 × 10−19 J , and
angular momenta of
(b) �e lowest unoccupied levels are those with m l = ±6, and so the dif-
ference in energy between the highest occupied and lowest unoccupied
levels is
3.46... × 10−19 J
ν = ∆E�h = = 5.23 × 1014 Hz
6.6261 × 10−34 J s
(a) Y0,0 = (1�2)π−1�2 , which is a constant, and so its derivatives with respect
to all θ and � are zero, so Λ̂ 2 Y0,0 = 0 implying that E = 0 and lˆz Y0,0 = 0,
so that J z = 0 .
(b) Y2,−1 = N sin θ cos θe−i� , thus
∂Y2,−1 �∂θ = Ne−i� (cos2 θ−sin2 θ) ∂Y2,−1 �∂� = −iN sin θ cos θe−i� = −iY2,−1
1 ∂ 2 Y2,−1 1 ∂ ∂Y2,−1
Λ̂ 2 Y2,−1 = 2 + sin θ
sin θ ∂� 2 sin θ ∂θ ∂θ
N cos θe−i� Ne−i� ∂
= + sin θ(cos2 θ − sin2 θ)
sin θ sin θ ∂θ
N cos θe−i�
=
sin θ
Ne−i�
+ �sin θ(−4 cos θ sin θ) + cos θ(cos2 θ − sin2 θ)�
sin θ
∂Y3,+3 �∂θ = 3Ne3i� sin2 θ cos θ ∂Y3,+3 �∂� = 3iNe3i� sin3 θ = 3iY3,+3
1 ∂ 2 Y3,+3 1 ∂ ∂Y3,+3
Λ̂ 2 Y3,+3 = + sin θ
sin2 θ ∂� 2 sin θ ∂θ ∂θ
−9Ne sin θ
3i� 3
1 ∂
= + 3Ne3i� sin3 θ cos θ
sin2 θ sin θ ∂�
286 7 QUANTUM THEORY
3Ne3i�
= −9Ne3i� sin θ + (3 sin2 θ cos2 θ − sin4 θ)
sin θ
= −9Ne3i� sin θ + 3Ne3i� (3 sin θ cos2 θ − sin3 θ)
= −12Ne3i� sin3 θ = −12Y3,+3
Hence
π 2π
� �
∗
Y1,0 Y1,+1 sin θ dθ d� = 0
θ=0 �=0
∂z ∂ f ∂2 f ∂2 f ∂2 f ∂2 f
= −ħ 2 �y + yz − x y 2 − z2 + zx �
∂z ∂x ∂z∂x ∂z ∂y∂x ∂y∂z
∂f ∂2 f ∂2 f ∂2 f ∂2 f
= −ħ 2 �y + yz − x y 2 − z2 + zx �
∂x ∂z∂x ∂z ∂y∂x ∂y∂z
(b) Similarly,
∂ ∂ ∂f ∂f
lˆy lˆx f = −ħ 2 �z − x � �y −z �
∂x ∂z ∂z ∂y
∂ ∂f ∂ ∂f ∂ ∂f ∂ ∂f
= −ħ 2 �z �y � − z �z � − x �y � + x �z ��
∂x ∂z ∂x ∂y ∂z ∂z ∂z ∂y
To evaluate the �nal term in this, the product rule must be used
∂2 f ∂2 f ∂2 f ∂z ∂ f ∂2 f
= −ħ 2 �yz − z2 − xy 2 + x + xz �
∂x∂z ∂x∂y ∂z ∂z ∂y ∂z∂y
∂2 f ∂2 f ∂2 f ∂f ∂2 f
= −ħ 2 �yz − z2 − xy 2 + x + xz �
∂x∂z ∂x∂y ∂z ∂y ∂z∂y
(c) Due to the symmetry of mixed partial derivatives, the only terms that are
not repeated in both of these terms are the �rst derivatives. Hence,
∂f ∂f ħ ∂f ∂f
lˆx lˆy f − lˆy lˆx f = ħ 2 �x − y � = iħ × �x − y � = iħ lˆz f
∂y ∂x i ∂y ∂x
where the de�nition of the lˆz operator given in [�F.��–���] is used. It
follow that [ lˆx , lˆy ] = lˆz .
(d) Applying cyclic permutation to lˆx gives (ħ�i)(z∂�∂x − x∂�∂z) = lˆy . Like-
wise lˆy gives (ħ�i)(x∂�∂y − y∂�∂x) = lˆz , and lˆz gives (ħ�i)(y∂�∂z −
z∂�∂y) = lˆx .
(e) Applying this permutation to the expression [ lˆx , lˆy ] = iħ lˆz gives [ lˆy , lˆz ] =
iħ lˆx . Permuting this expression gives [ lˆz , lˆx ] = iħ lˆy , and permuting that
expression gives [ lˆx , lˆy ] = iħ lˆz .
P�F.�� �e Cartesian coordinates expressed in terms of the spherical polar coordinates
are x = r sin θ cos �, y = r sin θ sin �, z = r cos θ, see �e chemist’s toolkit �� in
Topic �F on page ���. �e chain rule is therefore used to express ∂�∂� in terms
of derivatives of x, y and z.
∂ ∂x ∂ ∂y ∂ ∂z ∂
= + +
∂� ∂� ∂x ∂� ∂y ∂� ∂z
Evaluating the derivatives gives
∂ ∂ ∂ ∂ ∂
= −r sin θ sin � + r sin θ cos � + 0 = −y +x
∂x ∂y ∂z ∂x ∂y
where it has been noted that the factors multiplying the derivatives are Carte-
sian coordinates. Hence, lˆz = (ħ�i)(x∂�∂y − y∂�∂x) = (ħ�i)∂�∂�
288 7 QUANTUM THEORY
∂U ∂ NA h2 n2 −2 N A h 2 n 2 NA h2 n2
= � � = × = −
∂V ∂V 8mV 2�3 adia 3 8mV 5�3 12mV 5�3
��� � � � � � � ��� � � � � � � � �
A
∂U NA h2 n2
dU = � � dV = − dV (�.�)
∂V adia 12mV 5�3
�e work is equal to this change in internal energy. For a �nite change
the expression is integrated with respect to V between limits V1 and V2 ,
with ∆V = V2 − V1 .
(b) It is evident from eqn �.� that dU, and hence the work, goes as n 2 .
(c) �e work of expansion against an external pressure p ex is given by [�A.�a–
��], dw = −p ex dV . In eqn �.� the term A which multiplies dV refers to the
sample itself, and so must presumably in some way re�ect the pressure
of the sample, not the external pressure. However, if the expansion is
reversible, the external pressure is equal to the internal pressure and the
term A can then be identi�ed as the pressure. �erefore, if it is assumed
that the expansion is both adiabatic and reversible
NA h2 n2
p=
12mV 5�3
�e expression can be rewritten in terms of the average energy of each
particle which, because they all occupy the same level, is simply E av =
n 2 h 2 �8mL 2 = n 2 h 2 �8mV 2�3 , hence
��� � � � � �� � � � � � ��
E av
N A h 2 n 2 8N A n 2 h 2 2N A E av
p= = =
12mV 5�3 12V 8mV 2�3 3V
�is expression is reminiscent of the form of the pressure derived using
the kinetic theory of gases (Topic �B): pV = 13 nMυ 2rms , where n is the
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 289
��� � � � � �� � � � � �
Ek
pV = 1
3
nmN A υ 2rms = n 23 N A 12 mυ 2rms
I�.� (a) In Problem P�D.� and Problem P�D.� it is shown that for a particle in a
box in a state with quantum number n
hence
For n = 1
8A Hydrogenic Atoms
Answers to discussion questions
D�A.� (i) A boundary surface for a hydrogenic orbital is drawn so as to contain
most (say ��%) of the probability density of an electron in that orbital. Its
shape varies from orbital to orbital because the electron density distribu-
tion is di�erent for di�erent orbitals.
(ii) �e radial distribution function P(r) gives the probability density that the
electron will be found at a distance r from the nucleus. It is de�ned such
that P(r) dr is the probability of �nding the electron in a shell of radius r
and thickness dr. Because the radial distribution function gives the total
density, summed over all angles, it has no angular dependence and, as a
result, perhaps gives a clearer indication of how the electron density varies
with distance from the nucleus.
Solutions to exercises
E�A.�(a) �e radial wavefunction of a �s orbital is given in Table �A.� on page ��� as
R 3,0 = N(6 − 6ρ + ρ 2 )e−ρ�2 , where ρ = 2Zr�3a 0 . Radial nodes occur when the
wavefunction passes through 0, which is when 6 − 6ρ + ρ 2 = 0. �e roots of this
292 8 ATOMIC STRUCTURE AND SPECTRA
√
quadratic equation are at ρ = 3 ± 3 and hence the nodes are at
√
r = (3 ± 3)(3a 0 �2Z)
E�A.�(a) Angular nodes occur when cos θ sin θ cos � = 0, which occurs when any of
cos θ, sin θ, or cos � is equal to zero; recall that the range of θ is 0 → π and of
� is 0 → 2π.
Although the function is zero for θ = 0 this does not describe a plane, and so
is discounted. �e function is zero for θ = π�2 with any value of �: this is the
x y plane. �e function is also zero for � = π�2 with any value of θ: this is the
yz plane. �ere are two nodal planes, as expected for a d orbital.
E�A.�(a) �e radius at which the electron is most likely to be found is that at which the
radial distribution function is a maximum. �e radial distribution function is
de�ned in [�A.��b–���], P(r) = r 2 R(r)2 . For the �p orbital R(r) is given in
Table �A.� on page ��� as R 2,1 = N ρe−ρ�2 where ρ = 2Zr�na 0 , which for n = 2
is ρ = Zr�a 0 . With the substitution r 2 = ρ 2 (a 0 �Z)2 , the radial distribution
function is therefore P(ρ) = N 2 (a 0 �Z)2 ρ 4 e−ρ .
To �nd the maximum in this function the derivative is set to zero; the multi-
plying constants can be discarded for the purposes of this calculation
d 4 −ρ
ρ e = 4ρ 3 e−ρ − ρ 4 e−ρ
dr
Setting this derivative to zero gives the solutions ρ = 0 and ρ = 4. P(ρ) is zero
for ρ = 0 and as ρ → ∞, therefore ρ = 4 must be a maximum. �is occurs at
r = 4a 0 �Z .
E�A.�(a) All the �p orbitals have the same value of n and l, and hence have the same
radial function, which is given in Table �A.� on page ��� as R 2,1 = N ρe−ρ�2
where ρ = 2Zr�na 0 , which for n = 2 is ρ = Zr�a 0 . Radial nodes occur when
the wavefunction passes through zero. �e function goes to zero at ρ = 0 and as
ρ → ∞, but it does not pass through zero at these points so they are not nodes.
�e number of radial nodes is therefore � .
E�A.�(a) �e energy of the level of the H atom with quantum number n is given by
[�A.�–���], E n = −hc R̃ H �n 2 . As described in Section �A.�(d) on page ���,
the degeneracy of a state with quantum number n is n 2 .
�e state with E = −hc R̃ H has n = 1 and degeneracy (1)2 = � ; that with
E = −hc R̃ H �9 has n = 3 and degeneracy (3)2 = � ; and that with E = −hc R̃ H �25
has n = 5 and degeneracy (5)2 = �� .
E�A.�(a) �e task is to �nd the value of N such that the integral ∫ ψ ∗ ψ dτ = 1, where
ψ = Ne−r�a 0 . �e integration is over the range r = 0 to ∞, θ = 0 to π, and
� = 0 to 2π; the volume element is r 2 sin θ dr dθ d�. �e required integral is
therefore
∞ π 2π
N2 � � � r 2 e−2r�a 0 sin θ dr dθ d�
0 0 0
= N 2 [2!�(2�a 0 )3 ] × 2 × 2π = N 2 a 03 π
E�A.��(a) �e wavefunction is given by [�A.��–���], ψ n,l ,m l = Yl ,m l (θ, �)R n,l (r); for the
state with n = 2, l = 0, m l = 0 this is
ψ 2,0,0 = Y0,0 (θ, �)R 1,0 (r) = (4π)−1�2 (Z�2a 0 )3�2 (2 − ρ)e−ρ�2
294 8 ATOMIC STRUCTURE AND SPECTRA
where the radial wavefunction is taken from Table �A.� on page ���, the angular
wavefunction (the spherical harmonic) is taken from Table �F.� on page ���,
and ρ = 2Zr�na 0 . �e probability density is therefore
E�A.��(a) �e radial wavefunction of a �s orbital is taken from Table �A.� on page ���,
R 2,0 (r) = N(2−ρ)e−ρ�2 , where ρ = 2Zr�na 0 ; for n = 2, ρ = Zr�a 0 . �e extrema
are located by �nding the values of ρ for which dR 2,0 �dρ = 0; the product rule
is required
E�A.��(a) Assuming that the electron is in the ground state, the wavefunction is ψ =
Ne−r�a 0 , and so the probability density is P(r) = ψ 2 = N 2 e−2r�a 0 . P(r) is a
maximum at r = 0 and then simply falls o� as r increases; it falls to 12 its initial
value when P(r ′ )�P(0) = 12
Hence r ′ = − 12 ln 12 a 0 = 0.347a 0 .
Solutions to problems
P�A.� �e �p orbitals only di�er in the axes along which they are directed. �erefore,
the distance from the origin to the position of maximum probability density
will be the same for each.
�e radial function for the �p orbitals is R 2,1 = N ρe−ρ�2 , where ρ = Zr�a 0 . �e
2
probability density is the square of the radial function, R 2,1 = N 2 ρ 2 e−ρ , and the
maximum in this is found by setting dR �dρ = 0
2
2
dR 2,1
= 2N 2 ρe−ρ − N 2 ρ 2 e−ρ = 0
dρ
Turning points occur at ρ = 0, 2, and it is evident from a plot of R(ρ) that ρ = 2
is the maximum. �is corresponds to r = 2a 0 �Z. For the 2pz orbital, for which
the angular part goes as cos θ, the maximum will be at θ = 0, which corresponds
to x = y = 0. �e position of maximum probability density is therefore at
x = 0, y = 0, z = 2a 0 �Z . �e corresponding positions for the other �p orbitals
are found by permuting the x, y and z coordinates.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 295
P�A.� �e energy levels of a hydrogenic atom with atomic number Z are given by
[�A.��–���], E n = −hcZ 2 R̃ N �n 2 , where the Rydberg constant for the atom is
given by [�A.��–���], R̃ N = (µ�m e )R̃∞ ; µ is the reduced mass of the atom.
For the D atom, with nuclear mass m D , the reduced mass is
mD me
µ=
mD + me
and therefore the Rydberg constant for D is
µ mD me mD
R̃ D = R̃∞ = R̃∞ = R̃∞
me m e (m D + m e ) mD + me
2.01355 × (1.660 539 × 10−27 kg)
=
[2.01355 × (1.660 539 × 10−27 kg)]+(9.109 382 × 10−31 kg)
× (109 737 cm−1 ) = 109 707 cm−1
where the constants have been used to su�cient precision to match the data.
�e energy of the ground state is
E 1 = −hc R̃ D = −(6.626 070 × 10−34 J s) × (2.997 925 × 1010 cm s−1 )
× (109 707 cm−1 ) = −2.179 27 × 10−18 J
Expressed as a molar quantity this is −1312.39 kJ mol−1 .
P�A.� (a) By analogy with [�A.��–���], the wavefunction for a �px orbital is ψ 3p x =
R 3,1 (r) × [Y1,+1 (θ, �) − Y1,−1 (θ, �)](2)−1�2 . �e required integral to
verify normalization is
∞ π 2π
� r 2 dr � � 2−1 �Y1,+1 (θ, �) − Y1,−1 (θ, �)� sin θ dθ d�
2 2
R 3,1
0 0 0
Consider �rst the integral over r. From Table �A.� on page ��� R 3,1 (ρ) =
(486)−1�2 (Z�a 0 )3�2 (4 − ρ)ρe−ρ�2 , where ρ = 2Zr�na 0 which is this case
is 2Zr�3a 0 . It is convenient to calculate the integral over ρ, noting that
r = ρ(3a 0 �2Z) so that r 2 = ρ 2 (3a 0 �2Z)2 and dr = dρ (3a 0 �2Z). �e
integral becomes
1 Z3 ∞
3 �
(4 − ρ)2 ρ 2 (3a 0 �2Z)2 ρ 2 e−ρ (3a 0 �2Z) dρ
486 a 0 0
1 33 ∞ 1 ∞
= � (4 − ρ) ρ e dρ =
2 4 −ρ
� (16ρ − 8ρ + ρ )e dρ
4 5 6 −ρ
486 2 0
3 144 0
1
= �16(4!�15 ) − 8(5!�16 ) + (6!�17 )� = 1
144
where Integral E.� is used, with k = 1 and the appropriate value of n. �e
radial part of the function is therefore normalized.
�e angular function is found using the explicit form of the spherical
harmonics listed in Table �F.� on page ���
1 3 1�2 1
Yx = (Y1,+1 − Y1,−1 ) = − � � �sin θ ei� + sin θ e−i� �
2 1�2 8π 21�2
3 1�2 1 3 1�2
= −� � × 2 sin θ cos � = − � � sin θ cos �
8π 21�2 4π
296 8 ATOMIC STRUCTURE AND SPECTRA
Using Integral T.� with a = π and k = 1 gives the integral over θ as 4�3.
�e term cos2 � is written as 1 − sin2 �. �e integral ∫0 1 d� = 2π and,
2π
x y f (r) = (r sin θ cos �)(r sin θ sin �) f (r) = r 2 f (r) sin2 θ cos � sin �
= 12 r 2 f (r) sin2 θ sin 2�
�e spherical harmonics Y2,±2 have the form (again, omitting the normal-
ization factors) sin2 θ e±2i� . �e angular part of dx y is therefore obtained
by the combination
�is integral is zero because ∫0 eni� d� is zero for integer n. All of the
2π
terms in eqn �.� follow this pattern and therefore the overall integral is
zero; the orbitals are therefore orthogonal.
(b) �e radial nodes for the �s, �p and �d orbitals are found by examining
the radial wavefunctions, which are listed in Table �A.� on page ���, ex-
pressed as functions of ρ = 2Zr�3a 0 . �ese functions all go to zero as
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 297
where the integrals are evaluated using Integral E.� with k = 1 and the
appropriate value of n.
298 8 ATOMIC STRUCTURE AND SPECTRA
1 Z 3 3a 0 2
P3s = � � � � (6 − 6ρ + ρ 2 )2 ρ 2 e−ρ
243 a 0 2Z
1 Z 6
= �ρ − 12ρ 5 + 48ρ 4 − 72ρ 3 + 36ρ 2 � e−ρ
108 a 0
1 Z 3 3a 0 2
P3p = � � � � (4 − ρ)2 ρ 2 ρ 2 e−ρ
486 a 0 2Z
1 Z 6
= �ρ − 8ρ 5 + 16ρ 4 � e−ρ
216 a 0
1 Z 3 3a 0 2 2 2 2 −ρ
P3d = � � � � (ρ ) ρ e
2430 a 0 2Z
1 Z 6 −ρ
= ρ e
1080 a 0
Plots of these three functions are shown in Fig. �.�
0.10 �s
�p
�d
P�(Z�a 0 )
0.05
0.00
0 5 10 15 20
ρ
Figure 8.1
�e radial distribution function for the �s orbital has two subsidiary max-
ima which lie close in to the nucleus, and that for �p has one such max-
imum. In multi-electron atoms this density close to the nucleus results
in the energies of the �s, �p and �d orbitals no longer being equal: see
Section �B.� on page ���.
�e ground state of the H atom is the �s orbital for which the wavefunction is
ψ 1s = (πa 03 )−1�2 e−r�a 0 therefore P(r) = (πa 03 )−1 e−2r�a 0 . Because P(r) does not
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 299
depend on the angles, the integral over the angles can be evaluated separately
to give 4π. �e expression for P(σ) therefore becomes
σ
P(σ) = (4�a 03 ) � r 2 e−2r�a 0 dr
0
hence
To �nd the radius at which P(σ) = 0.9 needs the solution to the equation
0.9 = 1 − e−2σ�a 0 [2(σ�a 0 )2 + 2(σ�a 0 ) + 1], which is found numerically to be
σ = 2.66a 0 . Figure �.� is a plot of P(σ) as a function of σ; it is a sigmoid curve
and shows, as expected, that the the radius of the sphere increases as the total
enclose probability increases.
1.0
0.8
0.6
P(σ)
0.4
0.2
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
σ�a 0
Figure 8.2
8B Many-electron atoms
Answers to discussion questions
D�B.� See Section �B.� on page ���.
D�B.� �is is covered in any introductory or general chemistry text.
Solutions to exercises
E�B.�(a) All con�gurations have the [Ar] core.
Sc Ti V Cr Mn
2 1 2 2 2 3 1 5
4s 3d 4s 3d 4s 3d 4s 3d 4s2 3d5
Fe Co Ni Cu Zn
2 6 2 7 2 8 1 10
4s 3d 4s 3d 4s 3d 4s 3d 4s 3d10
2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 301
E�B.�(a) Across the period the energy of the orbitals generally decreases as a result of
the increasing nuclear charge. �erefore Li is expected to have the lowest
ionization energy as its outer electron has the highest orbital energy.
E�B.�(a) Hydrogenic orbitals are written in the form [�A.��–���], R n, l (r)Yl ,m l (θ, �),
where the appropriate radial function R n, l is selected from Table �A.� on page
��� and the appropriate angular function Yl ,m l is selected from Table �F.� on
page ���. Using Z = 2 for the �s and Z = 1 for the �s gives
Ψ(r 1 , r 2 ) = ψ 1s (r 1 )ψ 2s (r 2 )
E�B.�(a) For a subshell with angular momentum quantum number l there are 2l + 1
values of m l , each of which corresponds to a separate orbital. Each orbital can
accommodate two electrons, therefore the total number of electrons is 2×(2l +
1). �e subshell with l = 3 can therefore accommodate 2(6+1) = �� electrons.
Solutions to problems
P�B.� �e radial distribution function for a �s orbital is given by [�A.��–���], P(r) =
(4Z 3 �a 03 )r 2 e−2Zr�a 0 . �is gives the probability density of �nding the electron
in a shell of radius r. �e most probable radius is found by �nding the maxi-
mum in P(r), when dP(r)�dr = 0. In �nding this maximum the multiplying
constants are not relevant and can be discarded
d 2 −2Zr�a 0
r e = [2r − (2Z�a 0 )r 2 ] e−2Zr�a 0 = 0
dr
It follows that r max = a 0 �Z; that this is a maximum is most easily seen by
plotting P(r). For Z = 126 the most probable radius will be a 0 �126 .
P�B.� Toward the middle of the �rst transition series (Cr, Mn, and Fe) elements ex-
hibit the widest ranges of oxidation states. �is is due to the large number
of electrons in the �d and �s subshells that have similar energies, and as the
�d electrons that are generally removed provide very little shielding to the �s
orbitals, the e�ective nuclear charge does not increase signi�cantly between
adjacent oxidation states, meaning that the ionization energies of these levels
are close, and as these are the outermost electrons the ionization levels are
relatively small, meaning that large numbers of reactions will release enough
energy to lose many electrons.
However, it should be noted that the higher oxidation states of the middle
transition metals do not exist as cations, but only in compounds or compound
ions where there is a signi�cant stabilization of this ion by electron rich atoms,
302 8 ATOMIC STRUCTURE AND SPECTRA
for example the MnVII state only exists in the MnO� – ion where there is large
electron donation from bonding with four O� – ions, as here the e�ective nu-
clear charge has increased a lot over the neutral atom.
�is phenomenon is related to the availability of both electrons and orbitals
favourable for bonding. Elements to the le� (Sc and Ti) of the series have few
electrons and relatively low e�ective nuclear charge leaves d orbitals at high
energies that are relatively unsuitable for bonding. To the far right (Cu and
Zn) e�ective nuclear charge may be higher but there are few, if any, orbitals
available for bonding. Consequently, it is more di�cult to produce a range of
compounds that promote a wide range of oxidation states for elements at either
end of the series. At the middle and right of the series the +� oxidation state is
very commonly observed because normal reactions can provide the requisite
ionization energies for the removal of �s electrons.
P�B.� �e �rst, second and third ionization energies for the group �� elements are
plotted in Fig. �.�
40
�rst IE
second IE
Ionization energy/eV
30 third IE
20
10
0
0 20 40 60 80
Atomic number, Z
Figure 8.3
(a) In all cases, I 1 < I 2 < I 3 because of decreased nuclear shielding as each
successive electron is removed.
(b) �e ionization energies of boron are much larger than those of the re-
maining group elements because the valence shell of boron is very small
and compact with little nuclear shielding. �e boron atom is much smaller
than the aluminum atom.
(c) �e ionization energies of Al, Ga, In, and Tl are comparable even though
successive valence shells are further from the nucleus because the ioniza-
tion energy decrease expected from large atomic radii is balanced by an
increase in e�ective nuclear charge.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 303
8C Atomic spectra
Answers to discussion questions
D�C.� �is is discussed in Section �C.�(b) on page ���.
D�C.� �e selection rules are given in [�C.�–���]. In part these can be rationalised
by noting that a photon has one unit of (spin) angular momentum and that
in the spectroscopic transition this angular momentum must be conserved.
�e selection rule for l, ∆l = ±1, can be understood as a single electron in
the atom changing angular momentum by one unit in order to accommodate
the angular momentum from the photon. �is selection is derived in How
is that done? �C.� on page ��� by considering the relevant transition dipole
moment. �e selection rule for the total spin, ∆S = 0, stems from the fact that
the electromagnetic radiation does not a�ect the spin directly.
�e selection rules for multi-electron atoms are harder to rationalise not least
because the change in the overall angular momentum (L and J) is a�ected both
by changes in the angular momenta of individual electrons and by the way in
which these couple together.
Solutions to exercises
E�C.�(a) For a d electron l = 2 and s = 12 . Using the Clebsh–Gordon series, [�C.�–���],
the possible values of j are l + s, l + s − 1, . . . �l − s�, which in this case are
j = 52 , 32 .
E�C.�(a) �e symbol D implies that the total orbital angular momentum L = 2 , the su-
perscript � implies that the multiplicity 2S + 1 = 1, so that the total spin angular
momentum S = 0 . �e subscript � implies that the total angular momentum
J =2.
E�C.�(a) �e Clebsch–Gordan series, [�C.�–���], is used to combine two spin angular
momenta s 1 and s 2 to give S = s 1 + s 2 , s 1 + s 2 − 1 ..., �s 1 − s 2 �. (i) For two
electrons, each with s = 12 , S = 1, 0 with multiplicities, 2S = 1, of 3, 1 . (ii)
�ree electrons are treated by �rst combining the angular momenta of two of
them to give S ′ = 0, 1 and then combining each value of S ′ with s 3 = 12 for
the third spin. �erefore, for S ′ = 1, S = 1 + 12 , �1 − 12 � = 32 , 12 . Combining
S ′ = 0 with s = 12 simply results in S = 12 . �e overall result is S = 32 , 1
2
with
corresponding multiplicities �, � .
E�C.�(a) �e valence electron con�guration of the Ni�+ is [Ar] �d8 . In principle the
same process could be adopted as in Exercise E�C.�(a), in which the spin an-
gular momenta of all eight electrons are coupled together in successive steps to
304 8 ATOMIC STRUCTURE AND SPECTRA
�nd the overall spin angular momentum. Such an approach would be rather
tedious and would also run the risk of generating values of S which come from
arrangements of electrons which violate the Pauli principle. A quicker method,
and one which ensures that the Pauli principle is not violated, is to consider
combinations of the quantum number m s which gives the z-component of the
spin angular momentum and which takes values ± 12 . �e total z-component of
the spin angular momentum is found by simply adding together the m s values:
M S = m s 1 + m s 2 + . . ..
With � electrons in the � d orbitals, � of these electrons must doubly occupy
three of the orbitals, and the Pauli principle requires that the two electrons in
each orbital are spin paired: one has m s = + 12 and one has m s = − 12 . �ese six
electrons therefore make no net contribution to M S , in the sense that the sum
of the individual m s values is �. �e remaining two electrons can either occupy
the same orbital with spins paired, giving M S = + 12 − 12 = 0, or they can occupy
di�erent orbitals with either their spins paired, giving M S = 0 once more, or
with their spins parallel, giving M S = + 12 + 12 = +1 or M S = − 12 − 12 = −1. Recall
that a total spin S gives M S values of S, (S − 1) . . . − S. �erefore the �rst
arrangement with just M S = 0 is interpreted as arising from S = 0 , and the
second arrangement with M S = 0, ±1 is interpreted as arising from S = 1 .
E�C.�(a) �ese electrons are not equivalent, as they are in di�erent subshells, hence all
the terms that arise from the vector model and the Clebsch–Gordan series are
allowed. �e orbital angular momentum of the s and d electrons are l 1 = 0 and
l 2 = 2 respectively, and these are combined using L = l 1 +l 2 , l 1 +l 2 −1, ... �l 1 −l 2 �
which in this case gives L = 2 only. �e spin angular momenta of each electron
is s 1 = s 2 = 12 , and these combine in the same way to give S = 1, 0; these values
of S have spin multiplicities of 2S +1 = 3, 1. �e terms which arise are therefore
3
D and 1 D.
�e possible values of J are given by J = L+S, L+S −1, ..., �L−S�, and hence for
S = 1, L = 2 the values of J are �, �, and �. For S = 0, L = 2 only J = 2 is possible.
�e term symbols are therefore 3 D3 , 3 D2 , 3 D1 , and 1 D2 . From Hund’s rules,
described in Section �C.�(d) on page ���, the lowest energy state is the one with
the greatest spin and then, because the shell is less than half full, the smallest J.
�is is 3 D1 .
E�C.�(a) Closed shells have total spin and orbital angular momenta of zero, and so do
not contribute to the overall values of S and L. (i) For the con�guration �s1
there is just one electron to consider with l = 0 and s = 12 , so L = 0, S = 12 , and
J = 12 . �e term symbol is 2 S1�2 . (ii) For the con�guration �p1 there is just one
electron to consider with l = 1 and s = 12 , so L = 1, S = 12 , and J = 32 , 1
2
. �e
term symbols are therefore 2 P3�2 and 2 P1�2 .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 305
E�C.�(a) �e two terms arising from a d1 con�guration are 2 D3�2 , 2 D5�2 , which have
S = 12 , L = 2 and J = 32 , 52 . �e energy shi� due to spin-orbit coupling is given
by [�C.�–���], E L,S , J = 12 hc Ã[J(J + 1) − L(L + 1) − S(S + 1)], where à is the
spin-orbit coupling constant. Hence, E 2,1�2,3�2 = −(3�2)hc à , and E 2,1�2,5�2 =
+hc à .
E�C.��(a) �e energy levels of a hydrogenic atom are E n = −hcZ 2 R̃ N n−2 , where Z is the
atomic number; for all but the most precise work it is su�cient to approximate
R̃ N by R̃∞ . �e wavenumber of the transition between states with quantum
numbers n 1 and n 2 in the He+ ion is given by a modi�ed version of [�A.�–���],
ν̃ = Z 2 R̃∞ (n−2
1 − n 2 ). For the 2 → 1 transition and with Z = 2
−2
E�C.��(a) �e selection rules for a many-electron atom are given in [�C.�–���]. For a
single electron these reduce to ∆l = ±1; there is no restriction on changes in n.
(i) �s (n = 2, l = 0) → �s (n = 1, l = 0) has ∆l = 0, and so is forbidden .
(ii) �p (n = 2, l = 1) → �s (n = 1, l = 0) has ∆l = −1, and so is allowed .
(iii) �d (n = 3, l = 2) → �p (n = 2, l = 1) has ∆l = −1, and so is allowed .
Solutions to problems
P�C.� �e wavenumbers of the spectral lines of the H atom for the n 2 → n 1 transition
are given by [�A.�–���], ν̃ = R̃ H (n−2
1 − n 2 ), where R̃ H is the Rydberg constant
−2
λ = ν̃−1 = R̃−1
H (n 1 − n 2 ) .
−2 −2 −1
�e lowest energy, and therefore the longest wavelength transition (the one at
λ max = 12368 nm = 1.2368 × 10−3 cm) corresponds to the transition from
n 1 + 1 → n 1 , therefore
1 1 1 (n 1 + 1)2 − n 12 2n 1 + 1
= − = = 2
λ max R̃ H n 1 (n 1 + 1)
2 2 n 1 (n 1 + 1)
2 2 n 1 (n 1 + 1)2
From the given data (λ max R̃ H )−1 = [(1.2368 × 10−3 cm) × (109677 cm−1 )]−1 =
(135.6...)−1 . �e value of n 1 is found by seeking an integer value of n 1 for
which n 12 (n 1 + 1)2 �(2n 1 + 1) = 135.6.... For n 1 = 6 the fraction on the le� is
62 × 72 �13 = 135.6.... �erefore, the Humphreys series is that with n 1 = 6 .
�e wavelengths of the transitions in the Humphreys series are therefore given
by λ = (109677 cm−1 )−1 × (6−2 − n−22 )
−1
for n 2 = 7, 8, .... �e next few lines,
with n 2 = 8, 9, and 10 are at ����.� nm, ����.� nm, ����.� nm, respectively.
�e convergence limit, corresponding to n 2 = ∞ is ����.� nm, as given in the
data.
ν̃ = Z 2 R̃ N �n−2
1 − n2 �
−2
(�.�)
where Z is the nuclear charge and R̃ N is the Rydberg constant for the nucleus
in question. In turn this is given by [�A.��–���]
µ me mN
R̃ N = R̃∞ µ=
me me + mN
where m N is the mass of the nucleus. �e spectra of 4 He+ and 3 He+ di�er
because R̃ N is di�erent for the two atoms. However, this di�erence is very
small because the value of the reduced mass µ is dominated by the mass of
the electron (m e � m N ). It is therefore necessary to work at high precision.
�e �rst step is to compute R̃ N for each nucleus, using m 4 He = 4.002 602m u
and m3 He = 3.016 029m u .
µ mN
R̃4 He = R̃∞ = R̃∞
me me + mN
4.002 602 × (1.660 539 × 10−27 kg)
=
(9.109 383 × 10−31 kg) + 4.002 602 × (1.660 539 × 10−27 kg)
× (1.097 373 × 105 cm−1 ) = 1.097 223 × 105 cm−1
A similar calculation gives R̃ 3 He = 1.097 173 × 105 cm−1 . With these values of
the Rydberg constant the wavenumber of the relevant transitions is computed
using eqn �.�; the results are given in the table.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 307
If the spectrometer has su�cient resolution these di�erences are detectable; the
greatest di�erence is for the higher wavenumber transition.
P�C.� �e three transitions originate from the same level, the 2 P, with energy E 2 P ,
and if it is assumed that the nd 2 D states are hydrogenic their energies may be
written E n = −A�n 2 , where A is some constant. It follows that the wavenumber
of the transitions can be written
ν̃ n = E 2 P �hc − (A�hc)�n 2
A plot of ν̃ n against 1�n 2 is therefore expected to a straight line with y-intercept
(at 1�n 2 = 0) E 2 P �hc. �e data are tabulated below and the graph is given in
Fig. �.�.
30 000
ν̃∞
25 000
ν̃ n �cm−1
20 000
15 000
0.00 0.02 0.04 0.06 0.08 0.10 0.12
2
1�n
Figure 8.4
�e data fall on a good straight line which has y-intercept ν̃∞ = E 2 P �hc =
28 595 cm−1 . �e transition from the 2 S ground state to the 2 P state is at a
wavelength of ���.�� nm, which corresponds to a wavenumber of �� ��� cm−1 .
�erefore the transition from 2 S to the ionization limit of the 2 P–2 D series will
be at wavenumber
14 908 cm−1 + ν̃∞ = 14 908 cm−1 + 28 595 cm−1 = 43 503 cm−1
308 8 ATOMIC STRUCTURE AND SPECTRA
�is corresponds to the ionization energy of the ground state, which can be
expressed in eV as �.�� eV . Although the data are given to high precision,
quite a long extrapolation is needed to �nd the energy of the 2 P state and it also
has been assumed that the constant A is independent of n, which may not be
the case. As a result, the ionization energy is quoted to more modest precision.
∆E = E 1,1�2,3�2 − E 1,1�2,1�2
= 12 hc Ã[ 32 ( 32 + 1) − L(L + 1) − S(S + 1)]
− 12 hc Ã[ 12 ( 12 + 1) − L(L + 1) − S(S + 1)] = 32 hc Ã
�e wavenumber of the separation between the two lines is therefore 32 Ã, hence
�e Balmer series are those lines with n 1 = 2, and so the wavenumbers of these
are
ν̃ = R̃ Ps (2−2 − n−2
2 ) = (54868.5 cm ) × (2
−1 −2
− n−2
2 ) n 2 = 3, 4 . . .
�e �rst three lines have n 2 = 3, 4, 5 and are at 7 621 cm−1 , 10 288 cm−1 , and
11 522 cm−1 , respectively. �e ionization energy is simply the binding energy
of the ground state, which is hc R̃ Ps . Hence I = hc R̃ Ps �e = �.��� eV
P�C.�� �e derivation follows the method used in How is that done? �C.� on page
���. For a transition to be allowed the transition dipole moment µfi must
be non-zero. It is convenient to explore this condition by examining the x-,
y-, and z-components of the moment: if any of these are non-zero, the over-
all moment will also be non-zero. �e x- and y-components are given by
µ x ,fi = −e ∫ ψ ∗f xψ i dτ and µ y,fi = −e ∫ ψ ∗f yψ i dτ, respectively. �e limits of
integration are r = 0 to ∞, θ = 0 to π, and � = 0 to 2π, and the volume element
is dτ = r 2 sin θ dr dθ d�.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 309
Similarly
2
P3�2 and 2 P1�2 , the lowest of which is that with J = 12 . According to the
selection rules the transitions from 2 S1�2 to both 2 P states are allowed:
hence, the transitions are 2 S1�2 → 2 P1�2 and 2 S1�2 → 2 P3�2 .
(b) �e wavenumber the spectral line corresponding to the n 1 → n 2 transi-
tion is given by a modi�ed version of [�A.�–���] which takes into account
the nuclear charge Z: ν̃ = Z 2 R̃ H (n−2
1 − n 2 ), where R̃ H is the Rydberg
−2
Rydberg constant is di�erent for He, but the change is so small that it can
safely be ignored. Hence for a transition for n = 1 → 4, the wavenumber
is ν̃ = 4 × (109 677 cm−1 ) × (1−2 − 4−2 ) = 411 289 cm−1 .
�is corresponds to a wavelength of λ = ν̃−1 = (411 289 cm−1 )−1 =
2.43... × 10−6 cm = ��.��� � nm . �e corresponding frequency is
n2 a0 l(l + 1)
�r�n, l ,m l = �1 + 12 �1 − ��
Z n2
(12 )a 0 0(0 + 1) 3a 0
�r�1,0,0 = �1 + 12 �1 − �� =
2 1 2 4
Hence, the mean radius of the atom increases by 23a 0 �2−3a 0 �4 = 43a 0 �4 .
I�.� Because the beam splits into two, with de�ections ±(µ B L 2 �4E k )dB�dz, a split-
ting between the two beams of ∆x is achieved by satisfying the condition ∆x =
(µ B L 2 �2E k )dB�dz, which is rearranged to give an expression for the �eld gra-
dient dB�dz = 2E k ∆x�µ B L 2 . A reasonable estimate for the mean kinetic energy
is to take the equipartition value of 32 kT.
dB 2E k ∆x 3kT∆x
= =
dz µB L2 µB L2
3 × (1.3806 × 10−23 J K−1 ) × (1000 K) × (1.00 × 10−3 m)
=
(9.2740 × 10−24 J T−1 ) × (50 × 10−2 m)2
= 17.9 T m−1
9 Molecular Structure
9A Valence-bond theory
Answers to discussion questions
D�A.� See Section �A.�(b) on page ��� for details on hybridization applied to simple
carbon compounds. �e carbon atoms in alkanes are sp3 hybridized. �is
explains the nearly tetrahedral bond angles about the carbon atoms in such
molecules.
�e double-bonded carbon atoms in alkenes are sp2 hybridized. �is explains
the bond angles of approximately ���○ about these atoms. �e simultaneous
overlap of sp2 hybridized orbitals and unhybridized p orbitals in C=C double
bonds explains the resistance of such bonds to torsion and the co-planarity of
the atoms attached to those atoms.
�e triple-bonded carbon atoms in alkynes are sp hybridized, which explains
the ���○ bond angles about these atoms. �e central carbon atom in allene is
also sp hybridized. Each of its C=C double bonds involves one of its sp hybrids
and one unhybridized p orbital. �e two resulting π orbitals are oriented per-
pendicular to one another, which is why the two CH� groups are rotated by
��○ relative to one another. �is arrangement of orbitals also accounts for the
resistance to the two CH� groups being rotated relative to one another about
the long axis.
D�A.� Resonance refers to the superposition of the wave functions representing dif-
ferent electron distributions in the same nuclear framework. �e wavefunction
resulting from the superposition is called a resonance hybrid.
Resonance allows for a more re�ned description of the electron distribution,
and hence bonding, than is given by a single valence bond wavefunction. Dif-
ferent valence bond structures are allowed to contribute to di�erent extents,
meaning that the overall wavefunction is built up from contributions from
di�erent valence-bond wavefunctions.
�is approach makes it possible to describe polar bonds as a combination of a
purely covalent and a purely ionic structure, and delocalized bonding in terms
of combinations of valence-bond structures in which, for example, a double
bond is located in di�erent parts of a molecule. Resonance is a device for
calculating an improved wavefunction: it does not imply that wavefunction
�ickers between those for the di�erent structures.
D�A.� Promotion and hybridization are two modi�cations to the simplest version of
valence-bond (VB) theory, adopted to overcome obvious mismatches between
312 9 MOLECULAR STRUCTURE
Solutions to exercises
E�A.�(a) �e ammonium ion is iso-electronic with methane, therefore the two species
are expected to have the same description of bonding. Four sp3 hybrid atomic
orbitals are formed from the �s and the three �p orbitals of the nitrogen atom;
each hybrid then forms a σ bond by overlapping with a hydrogen �s orbital.
E�A.�(a) All the carbon atoms in �,�-butadiene are sp2 hybridized. �e σ framework of
the molecule consists of C–H and C–C σ bonds. Each C–H σ bond is formed
by the overlap of an sp2 hybrid atomic orbital on a carbon atom with a �s atomic
orbital on a neighbouring hydrogen atom. Similarly, C–C σ bonds are formed
by the overlap of sp2 hybrid atomic orbitals on neighbouring carbon atoms.
�e two π bonds are formed by the side-by-side overlap of unhybridized �p
orbitals on carbon atoms C� and C�, and likewise between C� and C�.
E�A.�(a) �e carbon and nitrogen atoms in methylamine are sp3 hybridized. �e C–N
bond is formed by the overlap of an sp3 orbital on carbon with an sp3 orbital
on nitrogen. �e C–H bonds are formed by the overlap of a carbon sp3 hybrid
atomic orbital with a hydrogen �s atomic orbital. Similarly, the N–H bonds are
formed by the overlap of a nitrogen sp3 hybrid atomic orbital with a hydrogen
�s atomic orbital. �e lone pair on nitrogen resides on an sp3 hybrid atomic
orbital.
becomes
� h 1 h 2 dτ = � (s + px + p y + pz )(s − px − p y + pz ) dτ
∗
= � s2 dτ − � spx dτ − � sp y dτ + � spz dτ
=1−1−1+1=0
All the integrals of the form ∫ sp i dτ are zero because the s and p orbitals are
orthogonal, and all the integrals of the form ∫ s2 dτ and ∫ p2i dτ are � because
the orbitals are normalized. �e condition for the orthogonality of h 1 and h 2
is satis�ed.
� h h dτ = N � (s + 2 p) dτ
∗ 2 1�2 2
� 1 �
���� � � � � ��� � � � � � ��� � � � � � � � � � � � � � ��� � � � � � ��� � � � � � ��
1 0
� �
� �
= N 2 �� s2 dτ +2 � p2 dτ +23�2� sp dτ � = 3N 2
� �
� �
� �
� �
�e integral ∫ sp dτ is zero because the s and p orbitals are orthogonal, and the
integrals ∫ s2 dτ and ∫ p2 dτ are � because the orbitals are normalized. From
the normalization condition it follows that 3N 2 = 1 and hence N = 1�31�2 .
E�A.�(a) Using [�A.�–���] and assuming that the valence-bond in HF is formed be-
tween the H�s and F�pz atomic orbitals, the spatial part of the valence-bond
wavefunction is written as Ψ(1, 2) = ψF�pz (1)ψH�s (2) + ψF�pz (2)ψH�s (1). �e
overall wavefunction must be antisymmetric to satisfy the Pauli principle, there-
fore the symmetric spatial part has to be combined with the antisymmetric two-
electron spin wavefunction given by [�B.�–���], σ− (1, 2). �e (unnormalized)
complete two-electron wavefunction is therefore
[ψF�pz (1)ψH�s (2) + ψF�pz (2)ψH�s (1)] × σ− (1, 2). �e wavefunction ΨH+ F− de-
scribes the electron distribution when both electrons reside on the F2pz or-
bital. �e spatial part of this wavefunction is given by ψF2pz (1) ψF2pz (2) , which
is symmetric, therefore it has to be combined with the antisymmetric spin
wavefunction resulting in ΨH+ F− = [ψF2pz (1) ψF2pz (2) ] × σ− (1, 2). Similarly, the
other ionic structure has ΨH− F+ = [ψH�s(1) ψH�s(2) ] × σ− (1, 2).
E�A.�(a) Both phosphorus and nitrogen are in Group ��, therefore the valence bond
description of the bonding in P2 is similar to that of N2 . �ere is a triple bond
between the two sp hybridized phosphorus atoms. A σ bond is formed by the
overlap of two sp hybrid atomic orbitals projecting towards each other along
the internuclear axis. �e two π bonds are the result of the side-by-side overlap
of 3px with 3px and 3p y with 3p y orbitals. �ere is one lone pair on each
phosphorus atom, contained in the sp orbital projecting outwards along the
internuclear axis.
Consider the equilibrium P4 ��� ⇀ 2 P2 . In the tetrahedral P4 there are six σ
bonds, whereas in two molecules of P2 there are two σ and four π bonds overall.
π bonds are generally weaker than σ bonds, therefore the equilibrium favors P4 .
Solutions to problems
P�A.� �e wavefunction in terms of the polar coordinates of each electron is given in
Brief illustration �A.� on page ��� as
1
Ψ(1, 2) = �e−(rA� +rB� )�a 0 + e−(rA� +rB� )�a 0 �
πa 03
Given that the internuclear separation along the z-axis is R, in Cartesian coor-
dinates rAi and rBi becomes
1
Ψ(1, 2) =
πa 03
× �e−[(x 1 +y 1 +z 1 ) +(x 22 +y 22 +(z 2 −R)2 )1�2 ]�a 0
+ e−[(x 2 +y 2 +z 2 ) +(x 12 +y 12 +(z 1 −R)2 )1�2 ]�a 0
�
2 2 2 1�2 2 2 2 1�2
P�A.� For the purposes of this problem, the px and p y orbitals are represented by
unit vectors along the x-and y-axes, respectively. �e given hybrid atomic
orbitals are created by the linear combination of the s, px and p y orbitals. �e s
orbital is spherically symmetric about the origin, therefore it does not modify
the directions in which the hybrids point. �e vector representations of the
hybrid atomic orbitals are
√ � � � �
h1 = 2j h2 = 3�2 i − 1�2 j h3 = − 3�2 i − 1�2 j
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 315
y
h1
�
3
2
α α �x
1
2
h3 h2
√
From the diagram it is evident that α = tan−1 �1� 3� = 30○ . It follows that the
angle between adjacent hybrids is ���○ .
Solutions to exercises
E�B.�(a) �e energy of the σ bonding orbital in H� + is given by [�B.�–���], E σ = EH�s +
j 0 �R − ( j + k)�(1 + S). Molecular potential energy curves are usually plotted
with respect to the energy of the separated atoms, therefore the energies to be
plotted are E σ − EH�s = j 0 �R − ( j + k)�(1 + S).
Using [�B.�d–���], j 0 �a 0 = 27.21 eV = 1 Eh the energy for R�a 0 = 1 is com-
puted as
(1 E h ) (0.729 E h ) + (0.736 E h )
E σ − EH�s = − = +0.211 E h
1 (1 + 0.858)
R�a 0 � � � �
(E σ − EH�s )�Eh +0.211 −5.32 × 10 −2
−5.88 × 10 −2
−3.76 × 10−2
0.2
(E σ − EH�s )�E h
0.1
R�a 0
0.0
1.0 1.5 2.0 2.5 3.0 3.5 4.0
−0.1
Figure 9.1
�ese data are plotted in Fig. �.�; with so few data points it is di�cult to locate
the minimum. �e data are �tted well by the following cubic
Note that this cubic equation has no physical meaning, it is only used to draw
the line on the plot above and to locate the minimum by setting the derivative
to zero; in particular the maximum close to the �nal data point has no physical
basis. �is minimum is found to be at R = 2.5 a 0 , which corresponds to the
predicted equilibrium bond length. �e depth of the potential energy well at
this distance is about −0.073 E h which is 2.0 eV .
E�B.�(a) A sketch of the bonding and the antibonding molecular orbitals resulting from
the side-by-side overlap of two p orbitals is shown in Fig. �C.� on page ���. �e
bonding molecular orbital is antisymmetric with respect to inversion, therefore
it is denoted as a πu orbital. �e antibonding molecular orbital is symmetric
with respect to inversion, therefore it is a πg orbital.
� ψ ψ dτ = N � (ψA + λψB ) dτ
∗ 2 2
� �
���� � � � � � � � � � � � � � � � ��� � � � � � � �� � � � � � � � ��� � � � � � � � � � � � ��� � � � � � � � � � � � ��
1 1 S
� �
� �
= N 2 �� ψA2 dτ +λ 2 � ψB2 dτ +2λ� ψA ψB dτ �
� �
� �
� �
� �
= N (1 + λ + 2λS)
2 2
Using S = 0.250 the value of the integral becomes 0.356 + 0.88025β. �is value
must be zero for the two wavefunctions to be orthogonal, therefore β = −0.404
and hence ψ j = A − 0.404B.
� ψ i ψ i dτ = � [N(0.145A + 0.844B)] dτ
∗ 2
= N 2 (0.733 + 0.245S)
√
Using S = 0.250 gives a value of 0.794N 2 for the integral, therefore N = 1� 0.794 =
1.12. �erefore the normalized wavefunction is
Solutions to problems
P�B.� Inspection of [�B.�–���] reveals that the repulsion energy between two hydro-
gen nuclei is given by e 2 �4πε 0 R, where R is the internuclear separation. In
molar quantities, the repulsion energy is N A e 2 �4πε 0 R, which, at an equilibrium
separation of R = 74.1 pm becomes
P�B.� Refer to the data presented in Exercise E�B.�(a). �e energy of the σ bonding
orbital in H� + is given by [�B.�–���], E σ = EH�s + j 0 �R − ( j + k)�(1 + S). �is
energy in usually measured with respect to the energy of the separated atoms,
therefore the energy is E σ − EH�s = j 0 �R − ( j + k)�(1 + S). Likewise for the
σ∗ antibonding orbital the energy is given by [�B.�–���], E σ∗ = EH�s + j 0 �R −
( j − k)�(1 − S). Relative to the separated atoms, the energy is E σ∗ − EH�s =
j 0 �R − ( j − k)�(1 − S). With the data given, and using j 0 �a 0 = 27.21 eV = 1 Eh ,
the energies of these molecular orbitals are
R�a 0 � � � �
(E σ − EH�s )�Eh +0.211 −5.32 × 10 −2
−5.88 × 10 −2
−3.76 × 10−2
(E σ∗ − EH�s )�Eh +1.05 +0.340 +0.132 +5.52 × 10−2
Figure �.� shows plots of (a) the bonding and antibonding wavefunctions, and
(b) the squares of these functions (the probability density) for the case R = 2a 0 .
�e quantity plotted in (a) is (a 0 )3�2 ψ± , and in (b) it is (a 0 )3 ψ±2 ; the same scale
is used for each orbital.
�e antibonding orbital has a node at the mid-point of the bond, whereas the
bonding orbital has signi�cant electron density at this point. From the dia-
grams it appears that the antibonding orbital has greater overall probability, but
this is not in fact the case – both orbitals are normalized. It is just that when
the functions are plotted along the z-axis there appears to be such a di�erence
due to the di�erent distribution of electron density elsewhere in the orbitals.
�e di�erence density is the di�erence in electron density between the molecu-
lar orbital and two non-interacting �s orbitals, one on each atom. It is a measure
of the way in which the electron density is changed when the molecular orbitals
are compared to non-interacting atomic orbitals. �e di�erence density is given
by
ψ±2 − 12 �ψ A2 + ψ B2 �
Also shown in Fig. �.� is this di�erence density for (c) the bonding and (d)
the antibonding molecular orbital; the same scale is used for each plot. If
the di�erence density is positive the implication is that the electron density
is greater than for two non-interacting atomic orbitals, whereas if the di�er-
ence density is negative the implication is that there is a reduction in electron
density. It is evident from the plots that in the bonding molecular orbital there
is an increase in the electron density in the internuclear region, whereas for
the antibonding orbital the density in this region is reduced, but the density is
increased further away. �ese observations account (partially, at least) for the
fact that occupying the bonding molecular orbital promotes bond formation.
�e apparent di�erence in size between the di�erence densities between (c) and
(d) is a result of simply plotting the function along the z-axis.
D�C.� �e bond strength is related to the extent to which the occupied bonding molec-
ular orbitals are lowered in energy compared to the constituent atomic orbitals.
320 9 MOLECULAR STRUCTURE
(a) (b)
antibonding
bonding
–4 –2 0 2 4 6
z/a0
–4 –2 0 2 4 6
z/a0
(c) (d)
antibonding
bonding
–4 –2 0 2 4 6 –4 –2 0 2 4 6
z/a0 z/a0
Figure 9.2
Solutions to exercises
E�C.�(a) �e molecular orbital energy level diagram for Li2 , Be2 , B2 , C2 and N2 is shown
in Fig. �C.�� on page ���, and for O2 , F2 and Ne2 in Fig. �C.�� on page ���.
Following the same logic as in Exercise E�C.�(a) and Exercise E�C.�(a) gives
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 321
Li2 1 + 1 = 2 VE 1σ 2g b = 12 (2 − 0) = 1
Be2 2 + 2 = 4 VE 1σ 2g 1σ∗2
u b = 12 (2 − 2) = 0
B2 3 + 3 = 6 VE 1σ 2g 1σ∗2 2
u 1π u b = 12 (4 − 2) = 1
C2 4 + 4 = 8 VE 1σ 2g 1σ∗2 4
u 1π u b = 12 (6 − 2) = 2
N2 5 + 5 = 10 VE 1σ 2g 1σ∗2 4 2
u 1π u 2σ g b = 12 (8 − 2) = 3
O2 6 + 6 = 12 VE 1σ 2g 1σ∗2 2 4 ∗2
u 2σ g 1π u 1π g b = 12 (8 − 4) = 2
F2 7 + 7 = 14 VE 1σ 2g 1σ∗2 2 4 ∗4
u 2σ g 1π u 1π g b = 12 (8 − 6) = 1
Ne2 8 + 8 = 16 VE 1σ 2g 1σ∗2 2 4 ∗4 ∗2
u 2σ g 1π u 1π g 2σ u b = 12 (8 − 8) = 0
E�C.�(a) �e molecular orbital energy level diagram for Li2 , Be2 , B2 , C2 , N2 and their
ions is shown in Fig. �C.�� on page ���, and for O2 , F2 , Ne2 and their ions in
Fig. �C.�� on page ���. �e highest occupied molecular orbital (HOMO) is the
molecular orbital which is the highest in energy and is at least singly occupied.
�e HOMO of each of the listed ions is indicated by a box around it.
Li2 + 1 + 1 − 1 = 1 VE 1σ 1g
Be2 + 2 + 2 − 1 = 3 VE 1σ 2g 1σ∗1
u
B2 + 3 + 3 − 1 = 5 VE 1σ 2g 1σ∗2 1
u 1π u
C2 + 4 + 4 − 1 = 7 VE 1σ 2g 1σ∗2 3
u 1π u
N2 + 5 + 5 − 1 = 9 VE 1σ 2g 1σ∗2 4 1
u 1π u 2σ g
O2 + 6 + 6 − 1 = 11 VE 1σ 2g 1σ∗2 2 4 ∗1
u 2σ g 1π u 1π g
F2 + 7 + 7 − 1 = 13 VE 1σ 2g 1σ∗2 2 4 ∗3
u 2σ g 1π u 1π g
Ne2 + 8 + 8 − 1 = 15 VE 1σ 2g 1σ∗2 2 4 ∗4 ∗1
u 2σ g 1π u 1π g 2σ u
Li2 − 1 + 1 + 1 = 3 VE 1σ 2g 1σ∗1
u
Be2 − 2 + 2 + 1 = 5 VE 1σ 2g 1σ∗2 1
u 1π u
B2 − 3 + 3 + 1 = 7 VE 1σ 2g 1σ∗2 3
u 1π u
C2 − 4 + 4 + 1 = 9 VE 1σ 2g 1σ∗2 4 1
u 1π u 2σ g
N2 − 5 + 5 + 1 = 11 VE 1σ 2g 1σ∗2 4 2 ∗1
u 1π u 2σ g 1π g
O2 − 6 + 6 + 1 = 13 VE 1σ 2g 1σ∗2 2 4 ∗3
u 2σ g 1π u 1π g
F2 − 7 + 7 + 1 = 15 VE 1σ 2g 1σ∗2 2 4 ∗4 ∗1
u 2σ g 1π u 1π g 2σ u
Ne2 − 8 + 8 + 1 = 17 VE 1σ 2g 1σ∗2 2 4 ∗4 ∗2 1
u 2σ g 1π u 1π g 2σ u 3σ g
E�C.�(a) �e energy of the incident photon must equal the sum of the ionization energy
of the orbital and the kinetic energy of the ejected photoelectron, [�C.�–���],
hν = I + 12 me υ 2 . �e energy of the incident photon is given by hν = hc�λ =
(6.6261 × 10−34 J s) × (2.9979 × 108 m s−1 )�(100 × 10−9 m) = 1.98... × 10−18 J.
Rearranging the equation to give the speed of the ejected electron gives
1�2
2 hc
υ=� � − I��
me λ
2
=� × [(1.98... × 10−18 J)
(9.1094 × 10−31 kg)
1�2
− (12.0 eV) × (1.6022 × 10−19 J eV−1 )]� = 3.70 × 105 m s−1
E�C.�(a) �e molecular orbital diagram for the homonuclear diatomic molecules Li2 ,
Be2 , and C2 is shown in Fig. �C.�� on page ���. According to the Pauli principle,
up to two valence electrons can be placed in each of the molecular orbitals. First
the lowest energy orbital is �lled up, then the next lowest and so on, until all
the valence electrons are used up.
(i) Li2 has 1+1 = 2 valence electrons (VE) overall, therefore the ground-state
electron con�guration is 1σ 2g . �e bond order is de�ned in [�C.�–���] as
b = 12 (N − N ∗ ), therefore b = 12 (2 − 0) = 1 .
(ii) Be2 : 2 + 2 = 4 VE; 1σ 2g 1σ∗2
u ; b = 2 (2 − 2) = 0 .
1
E�C.�(a) �e molecule with the greater bond order is expected to have the larger dis-
sociation energy. Qualitatively B2 and C2 share the same molecular orbital
energy level diagram, shown in Fig. �C.�� on page ���. B2 has 3 + 3 = 6
valence electrons overall, therefore its ground-state electron con�guration is
u 1π u . �e bond order is de�ned in [�C.�–���] as b = 2 (N − N ),
1σ 2g 1σ∗2 2 1 ∗
therefore b = 12 (4 − 2) = 1.
C2 has 4 + 4 = 8 valence electrons, its con�guration is 1σ 2g 1σ∗2 4
u 1π u , and the
bond order is b = 12 (6 − 2) = 2. C2 has greater bond order than B2 , therefore
C2 is expected to have the larger bond dissociation energy.
E�C.�(a) �e molecule with the greater bond order is expected to have the larger dissoci-
ation energy. �e molecular orbital energy level diagram of F2 and F2 + is shown
in Fig. �C.�� on page ���. F2 has 7 + 7 = 14 valence electrons overall, therefore
the ground-state electron con�guration is 1σ 2g 1σ∗2 2 4 ∗4
u 2σ g 1π u 1π g . �e bond
order is de�ned in [�C.�–���] as b = 2 (N − N ), therefore b = 2 (8 − 6) = 1.
1 ∗ 1
Removing one electron from F� gives F� + , which has one fewer electron in the
antibonding π∗g orbital, therefore the bond order is b = 12 (8 − 5) = 32 . F2 + has
greater bond order than F2 , therefore F2 + is expected to have the larger bond
dissociation energy.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 323
Solutions to problems
P�C.� (a) Figure �.� shows a plot of the overlap integral (Z = 1 is assumed)
1 R 1 R 2 1 R 4
S(2s, 2s) = �1 + � �+ � � + � � � e−R�2a 0
2 a0 12 a 0 240 a 0
1.0
S(2s,2s)
0.8 S(2p,2p)
0.6
S
0.4
0.2
0.0
0 5 10 15 20
R�a 0
Figure 9.3
(b) �e overlap integral S(2s, 2s) reaches a value of �.�� at R�a 0 = 8.03 ; this
value can be read o� a graph or found by using mathematical so�ware.
(c) Figure �.� shows a plot of the overlap integral (Z = 1 is assumed)
1 R 1 R 2 1 R 3
S(2p, 2p) = �1 + � �+ � � + � � � e−R�2a 0
2 a0 10 a 0 120 a 0
(d) �e value of the overlap integral at R�a 0 = 8.03 is
S(2p, 2p) = �1 + 12 ×8.03 + 1
10
×(8.03)2 + 1
120
×(8.03)3 �×e−8.03�2 = �.��
P�C.� Figure �.� shows contour plots of the bonding and antibonding �pσ and �pπ
molecular orbitals for a representative internuclear distance of R = 6a 0 ; nega-
tive amplitude is indicated by dashed contours, and the locations of the nuclei
are shown by the black dots. Density plots, in which the intensity of the shading
is proportional to the square of the wavefucntion, are shown in Fig. �.�.
�ese plots are useful as they identify aspects of the symmetry of the wavefunc-
tions and the positions of nodal planes. In addition, they illustrate that for the
bonding orbitals electron density is accumulating in the internuclear region.
2pσg 2pσu
x / a0 4 4
0 0
−4 −4
−12 −8 −4 0 4 8 12 −12 −8 −4 0 4 8 12
z / a0 z / a0
2pπu 2pπg
8 8
4 4
x / a0
0 0
−4 −4
−8 −8
−12 −8 −4 0 4 8 12 −12 −8 −4 0 4 8 12
z / a0 z / a0
Figure 9.4
D�D.� In forming a bond an atom must, to some extent, give up electron density to
be shared with other atoms in the molecule. �e energy needed to do this is
connected with the value of the ionization energy of the orbital. Equally, the
atom will to some extent acquire additional electron density to interact with,
and the energy gained from acquiring this density is connected in some way to
the electron a�nity. �us both electron gain and electron loss, in the loosest
sense, are involved in the process of bonding. It is for this reason that ionization
energy and electron a�nity are involved in the estimation of atomic orbital
energies for participation in bonding. See Section �D.�(a) on page ���.
Solutions to exercises
E�D.�(a) A suitable MO diagram in shown in the solution to Exercise E�D.�(a). �e ion
with the greater bond order is expected to have the shorter bond length. NO+
has 5 + 6 − 1 = 10 valence electrons, just enough to completely �ll up all the
bonding molecular orbitals, leading to a ground state electron con�guration
of 1σ 2 2σ 2 3σ 2 1π 4 . NO− has two more electrons, both accommodated in the
antibonding 2π orbital. It follows that NO+ has a greater bond order than NO− ,
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 325
2p g
2p u
4 4
x / a0
0 0
−4 −4
−12 −8 −4 0 4 8 12 −12 −8 −4 0 4 8 12
z / a0 z / a0
8 2p u 8 2p g
4 4
x / a0
0 0
−4 −4
−8 −8
−12 −8 −4 0 4 8 12 −12 −8 −4 0 4 8 12
z / a0 z / a0
Figure 9.5
ativities of Period � atoms against the square root of their Mulliken electroneg-
ativities is shown in Fig. �.�.
�e equation of the best �t line is χPauling = 3.18χMulliken − 2.57, which is very
1�2
E�D.�(a) �e orbital energy of an atomic orbital in a given atom is estimated using the
procedure outlined in Brief illustration �D.� on page ���, and using data from
the Resource section. �e orbital energy of hydrogen is
αH = − 12 [I + Eea ]
(1 eV)
= − 12 ×[(1312.0 kJ mol−1 )+(72.8 kJ mol−1 )]× = −7.18 eV
(96.485 kJ mol−1 )
�e conversion factor between kJ mol−1 and eV is taken from inside the front
326 9 MOLECULAR STRUCTURE
4.0
3.0
χPauling
2.0
1.0
1.0 1.2 1.4 1.6 1.8 2.0 2.2
1�2
χMulliken
Figure 9.6
αCl = − 12 [I + Eea ]
(1 eV)
= − 12 ×[(1251.1 kJ mol−1 )+(348.7 kJ mol−1 )]× = −8.29 eV
(96.485 kJ mol−1 )
E�D.�(a) �e orbital energies of hydrogen (αH = −7.18 eV) and chlorine (αCl = −8.29 eV)
are calculated in Exercise E�D.�(a). Taking β = −1.0 eV as a typical value and
setting S = 0 for simplicity, substitution into [�D.�c–���] gives
2 1�2
2β
E± = 12 (αH + αCl ) ± 12 (αH − αCl ) �1 + � � �
αH − αCl
= 12 [(−7.18 eV) + (−8.29 eV)]
� 2 �1�2
� (−2.0 eV) �
± [(−7.18 �
eV) − (−8.29 eV)] �1 + � � �
(−7.18 eV) − (−8.29 eV) �
1
2
� �
� �
= (−7.73... eV) ± (1.14... eV)
E�D.�(a) �e orbital energies of hydrogen (αH = −7.18 eV) and chlorine (αCl = −8.29 eV)
are calculated in Exercise E�D.�(a). Taking β = −1.0 eV as a typical value, and
setting S = 0.2, substitution into [�D.�a–���] gives
αH + αCl − 2βS ± [(2βS − (αH + αCl ))2 − 4(1 − S 2 )(αH αCl − β 2 )]1�2
E± =
2(1 − S 2 )
(−15.0... eV) ± (1.54... eV)
= = (−7.84... eV) ± (0.803... eV)
(1.92)
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 327
A Molecule B
4σ
2π
2p
2p
1π
3σ
2σ
2s
2s
1σ
Figure 9.7
E�D.�(a) �e molecular orbital energy level diagram of the heteronuclear diatomic molecule
XeF is similar to the one shown in the solution to Exercise E�D.�(a) except that
the orbitals on atom A are �s and �p. It is not possible to predict the precise
energy ordering of the orbitals from simple considerations, so this diagram is
simply a plausible suggestion.
XeF has 8 + 7 = 15 valence electrons, therefore the ground state electron con-
�guration is 1σ 2 2σ 2 3σ 2 1π 4 2π 4 4σ 1 . �e con�guration of XeF+ is the same
except that, as there is one fewer electrons, the antibonding 4σ orbital is not
occupied. �is means that the bond order in XeF+ (b = 1) is greater than the
bond order in XeF (b = 12 ), therefore XeF+ is likely to have a shorter bond
length than XeF.
328 9 MOLECULAR STRUCTURE
Solutions to problems
P�D.� (a) A normalized wavefunction satis�es the condition given by [�B.�c–���],
∫ ψψ dτ = 1. �e given wavefunction is real, therefore ψ = ψ .
∗ ∗
= cos2 θ + sin2 θ = 1
�e values of the integrals come from the fact that ψA and ψB are or-
thonormal.
(b) �e wavefunction which describes the bonding molecular orbital is formed
by the in-phase interference of the atomic orbitals ψA and ψB , therefore
the coe�cients of ψA and ψB must have the same sign. Similarly, the
antibonding orbital is the result of the out-of-phase interference of the
basis atomic orbitals, therefore the corresponding coe�cients must have
opposite signs. A plot of the coe�cients, cos θ and sin θ as a function of
θ is shown in Fig. �.�.
1.0 cos θ
sin θ
0.5
value of function
0.0
−0.5
−1.0
0.0 0.2 0.4 0.6 0.8 1.0
θ�π
Figure 9.8
�erefore ψ describes a bonding molecular orbital for 0 < θ < π�2, and
an antibonding molecular orbital for π�2 < θ < π.
P�D.� �e energy of ψA and ψC are kept constant in the following, but the energy of
ψB is progressively lowered.
(a) Taking the energy of the orbital ψB to be −12.0 eV, the secular determi-
nant becomes
��� (−7.2 eV) − E (−1.0 eV) (−0.8 eV) ����
��� ���
��� (−1.0 eV) (−12.0 eV) − E 0
��
��� (−0.8 eV) (−8.4 eV) − E ����
� 0
= −E 3 − (27.6 eV)E 2 − (246.04 eV2 )E − (709.68 eV3 )
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 329
Setting the polynominal to zero and solving the cubic gives the following
energies: E 1 = −12.2 eV , E 2 = −8.75 eV and E 3 = −6.65 eV . �e ma-
trix which diagonalizes the hamiltonian matrix is
�e entries in each column of the matrix above give the coe�cients of the
atomic orbitals for the corresponding molecular orbital. Note that E 1 is
close to the energy of ψB , and that in the �rst column the orbital with the
largest coe�cient by far is ψB .
(b) If the energy of ψB is lowered further to −15.0 eV, the secular determinant
becomes
��� (−7.2 eV) − E (−1.0 eV) (−0.8 eV) ����
��� ���
��� (−1.0 eV) (−15.0 eV) − E 0 ��
��� (−0.8 eV) (−8.4 eV) − E ����
� 0
= −E 3 − (30.6 eV)E 2 − (292.84 eV2 )E − (889.2 eV3 )
D�E.� In ab initio methods an attempt is made to evaluate all integrals that appear
in the secular determinant. Approximations are still employed, but these are
mainly associated with the construction of the wavefunctions involved in the
integrals. In semi-empirical methods, many of the integrals are expressed in
terms of spectroscopic data or physical properties. Semi-empirical methods
exist at several levels. At some levels, in order to simplify the calculations, many
of the integrals are set equal to zero.
330 9 MOLECULAR STRUCTURE
Density functional theory (DFT) is di�erent from the Hartree-Fock (HF) self-
consistent �eld (HF-SCF) methods, the ab initio methods, in that DFT focuses
on the electron density while HF-SCF methods focus on the wavefunction.
However, they both attempt to evaluate integrals from �rst principles, so DFT
methods are in that sense ab initio methods: both are iterative self-consistent
methods in that the calculations are repeated until the energy and wavefunc-
tions (HF-SCF) or energy and electron density (DFT) are unchanged to within
some acceptable tolerance.
D�E.� �ese are all terms originally associated with the Hückel approximation used
in the treatment of conjugated π electron molecules, in which the π electrons
are considered independent of the σ electrons. �e π electron binding energy is
the sum of the energies of each π electron in the molecule. �e delocalization
energy is the di�erence in energy of the π electrons between the conjugated
molecule with n π bonds and the energy of n ethene molecules, each of which
has one π bond. �e π bond formation energy is the energy released when a
π bond is formed. It is obtained from the total π electron binding energy by
subtracting the contribution from the Coulomb integrals, α.
Solutions to exercises
E�E.�(a) (i) Following the same logic as in Exercise E�E.�(a) and applying the Hückel
approximations as explained there the secular determinant for anthracene
is written as (the numbers in bold refer to the numbering of the carbon
atoms in the molecule)
� � � � � � � � � �� �� �� �� ��
� α−E β 0 0 0 0 0 0 0 0 0 0 0 β
� β α−E β 0 0 0 0 0 0 0 0 0 0 0
� 0 β α−E β 0 0 0 0 0 0 0 0 0 0
� 0 0 β α−E β 0 0 0 0 0 0 0 0 0
� 0 0 0 β α−E β 0 0 0 0 0 0 0 β
� 0 0 0 0 β α−E β 0 0 0 0 0 0 0
� 0 0 0 0 0 β α−E β 0 0 0 β 0 0
� 0 0 0 0 0 0 β α−E β 0 0 0 0 0
� 0 0 0 0 0 0 0 β α−E β 0 0 0 0
�� 0 0 0 0 0 0 0 0 β α−E β 0 0 0
�� 0 0 0 0 0 0 0 0 0 β α−E β 0 0
�� 0 0 0 0 0 0 β 0 0 0 β α −E β 0
�� 0 0 0 0 0 0 0 0 0 0 0 β α−E β
�� β 0 0 0 β 0 0 0 0 0 0 0 β α−E
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 331
� � � � � � � � � �� �� �� �� ��
� α−E β 0 0 0 0 0 0 0 0 0 0 0 β
� β α−E β 0 0 0 0 0 0 0 0 0 0 0
� 0 β α−E β 0 0 0 0 0 0 0 0 0 0
� 0 0 β α−E β 0 0 0 0 0 0 0 0 0
� 0 0 0 β α−E β 0 0 0 0 0 0 0 β
� 0 0 0 0 β α−E β 0 0 0 0 0 0 0
� 0 0 0 0 0 β α−E β 0 0 0 0 0 0
� 0 0 0 0 0 0 β α−E β 0 0 0 β 0
� 0 0 0 0 0 0 0 β α−E β 0 0 0 0
�� 0 0 0 0 0 0 0 0 β α−E β 0 0 0
�� 0 0 0 0 0 0 0 0 0 β α−E β 0 0
�� 0 0 0 0 0 0 0 0 0 0 β α−E β 0
�� 0 0 0 0 0 0 0 β 0 0 0 β α−E β
�� β 0 0 0 β 0 0 0 0 0 0 0 β α−E
E�E.�(a) To calculate the π-electron binding energy of the given systems, it is necessary
to calculate the energies of the occupied molecular orbitals. �is is done by di-
agonalising the hamiltonian matrix: the diagonal elements of the resulting ma-
trix are the energies of the molecular orbitals. �e hamiltonian matrix has the
same form as the secular matrix except that the diagonal elements are α instead
of α − E. Alternatively, the energies can be found my �nding the eigenvalues of
the hamiltonian matrix, or by multiplying out the secular determinant, setting
the resulting polynomial in E to zero and then �nding the roots. Mathematical
so�ware is needed for all of these approaches.
�e secular determinants are derived in Exercise E�E.�(a), and from these the
form of the hamiltonain matrix is easily found.
species HeH+ has two electrons, therefore the kinetic energy term is written as
ħ2 2 ħ2 2
T̂ = − ∇1 − ∇
2me 2me 2
�e energy of interaction between an electron and a nucleus with charge num-
ber Z at distance r is given by −Ze 2 �4πε 0 r. �e potential energy operator
consists of terms for each electron interacting with the H nuclues (Z = 1) and
the He nucleus (Z = 2)
e2 1 1 2 2 1
V̂ = − � + + + − �
4πε 0 r 1H r 2H r 1He r 2He r 12
�e �rst term represent the interaction between electron � and the H nucleus,
and the second is for electron � with the same nucleus. �e third and fourth
terms represent the interactions of the two electrons with the He nucleus. �e
last term accounts for the repulsion between the two electrons. �e complete
electronic hamiltonian is Ĥ elec = T̂ + V̂ . Because only the electronic hamilto-
nian is required, the repulsion between the two nuclei is not included.
E�E.�(a) (i) Without making the Hückel approximations, the secular determinant of
the H3 molecule is written as
��� α 1 − E β 12 − S 12 E β 13 − S 13 E ���
��� ���
��� β 21 − S 21 E α2 − E β 23 − S 23 E ���
��� β − S E β 32 − S 32 E α3 − E ���
� 31 31 �
where α n is the Coulomb integral of the orbital on atom n, β nm is the
resonance integral accounting for the interaction between the orbitals on
atoms n and m, E is the energy of the molecular orbital and S nm is the
overlap integral between the orbtials on atoms n and m.
Within the Hückel approximations the energy of the basis atomic orbitals
is taken to be independent of the position of the corresponding atoms in
the molecule, therefore all Coulomb integrals are set equal to α (given
that there is only one type of basis atomic orbital and only one type of
atom is involved in the problem). Interaction between orbitals on non-
neighbouring atoms is neglected, that is β nm = 0 if atoms n and m are
not neighbouring. All other resonance integrals are set equal to β. �e
overlap between atomic orbitals is also neglected, therefore all overlap
integrals S nm with n ≠ m are set to zero. Hence the secular determinant
for linear H� is
��� α − E 0 ����
��� β
��� β α−E β ����
��� 0 �
� β α − E ����
(ii) In this case hydrogen atoms � and � are neighbours, therefore β 13 = β,
and the secular determinant is
��� α − E ���
��� β β ���
��� β α−E β ���
��� β α−E ���
� β �
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 333
Solutions to problems
P�E.� Number the oxygen atoms from � to � and the carbon atom as number �. Taking
the Hückel approximations the secular determinant of the carbonate ion is
written as
���� αO −E 0 0 β ����
��� �
��� 0 αO −E 0 β ����
= (αO − E)2 [(αO − E)(αC − E) − 3β 2 ]
��� 0 0 αO −E β ����
��� �
�� β β β αC −E ����
Hence the energies of the molecular orbitals of the carbonate ion are the solu-
tions of the equation (αO − E)2 [(αO − E)(αC − E) − 3β 2 ] = 0. �e solution
E = αO occurs twice and represents a degenerate pair of orbitals. �e other two
roots are the solutions of the equation (αO − E)(αC − E) − 3β 2 = 0, which gives
the energies �
E± = 12 [αO + αC ± (αO − αC )2 + 12β 2 ]
where E− < E+ . Because oxygen is the more electronegative element it is likely
that the three lowest energy molecular orbitlas will be those with E = αO and
E = E− . �e carbonate ion has � π electrons which will occupy these orbitals
and hence give a π-electron binding energy of
� �
E π = (αO +αC − (αO − αC )2 + 12β 2 )+4αO = 5αO +αC − (αO − αC )2 + 12β 2
334 9 MOLECULAR STRUCTURE
Edeloc = E π − E loc
� �
= [5αO + αC + (αO − αC )2 + 12β 2 ] − [5αO + αC + (αO − αC )2 + 4β 2 ]
� �
= (αO − αC )2 + 12β 2 − (αO − αC )2 + 4β 2
In this case, orbitals B and C are on the same atom. It follows that the resonance
integral βBC and the overlap integral SBC are zero, as the atomic orbitals on one
atom are orthogonal to each other. �erefore the secular equations simplify to
� α β 0 β � � 0 1 0 1 �
� β α β 0 � � 1 0 1 0 �
H=� � H = α� + β � �
� 0 β α β � � 0 1 0 1 �
� β 0 β α � � 1 0 1 0 �
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 335
�e diagonal elements in the resulting diagonalized matrix are +2, +1, +1, −1,
−1, and −2, giving the energies α +2β, α + β (doubly degenerate), α − β (doubly
degenerate), and α − 2β.
�e secular determinant and hamiltonian matrix for cyclooctatetraene is
��� α − E β β ����
��� 0 0 0 0 0 �0 1 0 0 0 0 0 1�
���� β α − E β 0 ���� �1 0 1 0 0 0 0 0�
�
0 0 0 0
� �
��� 0 β α−E β 0 0 0 0 ���� �0 1 0 1 0 0 0 0�
� �
��� �
���� 0 β α−E β 0 ���� �0 0 1 0 1 0 0 0�
H = α�+β � �
0 0 0
�
��� 0 0 0 β α−E β 0 0 ���� �0 0 0 1 0 1 0 0�
� �
��� � �0 0 0 0 1 0 1 0�
��� 0 0 0 0 β α−E β 0 ���� � �
� �
��� 0 β α − E β ���� �0 0 0 0 0 1 0 1�
��� �
0 0 0 0
�� β 0 0 0 0 0 β α − E ���� �1 0 0 0 0 0 1 0�
P�E.� (a) Within the Hückel approximations, the secular determinant of the trian-
gular species H� is
��� α − E β ����
��� β
��� β α − E β ����
��� β �
� β α − E ����
�e hamiltonian matrix is of the same form, but with diagonal elements
α
� α β β � � 0 1 1 �
H=� β α β � H = α� + β � 1 0 1 �
� β β α � � 1 1 0 �
As explained in the text, in order to �nd the energies it is su�cient to
diagonalize the matrix on the right, and this is convenient because most
mathematical so�ware packages are only able to diagonalize numerical
336 9 MOLECULAR STRUCTURE
E =α−β
energy
E = α + 2β
H3 + 2 VE Etot = 2α + 4β
H3 3 VE Etot = 3α + 3β
H3 − 4 VE Etot = 4α + 2β
(c) �e change in the total electron binding energy directly gives the change
in the internal energy in the reaction H+ (g)+H2 (g) ��→ H3 + (g). �ere-
fore ∆ r U −○ is expressed in terms of the resonance and Coulomb integrals
as
P�E.� For H� the �-��G* basis set is equivalent to the �-��G basis set because the star
indicates that the basis set adds d-type polarization functions for each atom
other than hydrogen. Consequently, the basis sets (a) �-��G* and (b) �-���+G**
were chosen. Since the calculated energy is with respect to the energy of widely
separated stationary electrons and nuclei, the experimental ground electronic
energy of dihydrogen is calculated as D e + 2I.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 337
H�
�-��G* �-���+G** experimental
R�pm ��.� ��.� ��.�
ground-state energy, E 0 �eV −30.6626 −30.8167 −32.06
F�
R�pm ���.� ���.� ���.�
ground-state energy, E 0 �eV −5406.30 −5407.92
Both computational basis sets give satisfactory bond length agreement with the
experimental value for H� . However the �-��G* basis set is not as accurate as the
larger basis set as illustrated by consideration of both its higher ground-state
energy and the variation principle that the energy of a trial wavefunction is
never less than the true energy. �at is, the energy provided by the �-���+G**
basis set is closer to the true energy.
P�E.�� (a) Linear conjugated polyenes do not have degenerate energy levels, so each
value of k corresponds to a molecular orbital which can be occupied by
up to two electrons. �erefore for ethene, with � electrons, the HOMO
has k = 1, and the LUMO has k = 2. �e HOMO–LUMO energy gap is
2π π
∆E = �α + 2β cos � − �α + 2β cos � = (α − β) − (α + β) = −2β
3 3
�is energy gap corresponds to 61500 cm−1 , therefore β = −3.07... ×
104 cm−1 .
Butadiene has � electrons, the HOMO has k = 2 and the LUMO has k = 3;
the HOMO–LUMO energy gap is given by
3π 2π
∆E = �α + 2β cos � − �α + 2β cos � = −1.23β
5 5
∆E corresponds to 46080 cm−1 , therefore β = −3.72... × 104 cm−1 .
Hexatriene has � electrons, the HOMO has k = 3 and the LUMO has
k = 4; the HOMO–LUMO energy gap is given by
4π 3π
∆E = �α + 2β cos � − �α + 2β cos � = −0.890β
7 7
∆E corresponds to 39750 cm−1 , therefore β = −4.46... × 104 cm−1 .
Octatetraene has � electrons, the HOMO has k = 4 and the LUMO has
k = 5; the HOMO–LUMO energy gap is given by
5π 4π
∆E = �α + 2β cos � − �α + 2β cos � = −0.695β
9 9
∆E corresponds to 32900 cm−1 , therefore β = −4.73... × 104 cm−1 .
�e average value of β is −4.00... × 104 cm−1 , which in electronvolts is
(−4.00... × 104 cm−1 )×(6.6261 × 10−34 J s)×(2.9979 × 1010 cm s−1 )
β=
(1.6022 × 10−19 J eV−1 )
= −4.96 eV
338 9 MOLECULAR STRUCTURE
(b) Octatetraene has � π electrons, therefore the orbitals with quantum num-
bers k = 1, �, � and � are all fully occupied. Hence the π-electron binding
energy is
π 2π 3π
E π = 2 �α + 2β cos � + 2 �α + 2β cos � + 2 �α + 2β cos �
9 9 9
4π
+ 2 �α + 2β cos � = 8α + 9.52β
9
�erefore the delocalization energy is Edeloc = 8α + 9.52β − 8(α + β) =
1.52β .
(c) �e energies and the orbital coe�cients are calculated according to the
given formulae and presented in the following tabe; k is the index for the
molecular orbital.
Integrated activities
I�.� (a) �e calculated and the measured values of the standard enthalpy of for-
mation (∆ f H −○ �kJ mol−1 ) of ethene, butadiene, hexatriene and octatetraene
are shown in the table below, together with the relative error in the calcu-
lated values.
molecule computed experimental % error
C� H� ��.�� ��.����� ��.�
C� H� ���.� 108.8 ± 0.79 ��.�
C� H� ���.� 168 ± 3 ��.�
C� H�� ���.� ���.� ��.�
�e experimental values are taken from webbook.nist.gov/chemistry/ and
book �ermodynamic Data of Organic Compounds by Pedley, Naylor and
Kirby.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 339
(b) �e % errors shown in the table above are calculated using the expression
�∆ f H −○ (calc) − ∆ f H −○ (expt)�
% error =
∆ f H −○ (expt)
(c) For all of the molecules, the computed enthalpies of formation deviate
from the experimental values by much more than the uncertainty in the
experimental value. It is clear that molecular modeling so�ware is not a
substitute for experimentation when it comes to quantitative measures.
I�.� (a) �e energies of the LUMOs of the given molecules are calculated using the
DF/B�LYP/�-��G* method. �e results along with the standard reduction
potentials are listed in the table below.
−3.0
−3.2
ELUMO �eV
−3.4
−3.6
−0.3 −0.2 −0.1 0.0 0.1
E −○ �V
Figure 9.9
(b) �e energy of the LUMO of this molecule is calculated using the same
method as above as −2.99 eV. Hence the predicted reduction potential is
(c) �e energy of the LUMO of the given molecule is calculated as −3.11 eV.
Hence the predicted reduction potential is
1 ∂2 1
∇2 = r + 2 Λ2
r ∂r 2 r
but an entirely equivalent form, which is more convenient here, is
∂2 2 ∂ 1
∇2 = + + 2 Λ2
∂r 2 r ∂r r
�e legendrian operator Λ 2 contains derivatives with respect to angles only. As
the given trial wavefunction is independent of angles, Λ 2 Ψ = 0 and therefore
the laplacian operating on the trial wavefunction gives
∂2 2 ∂
∇2 Ψtrial = �Ne−αr � + �Ne−αr � + 0 = 4N α 2 r 2 e−αr − 6N αe−αr
2 2 2 2
∂r 2 r ∂r
∗
�e wavefunction Ψtrial is real, therefore Ψtrial = Ψtrial . �erefore the hamilto-
nian operating on the trial wavefunction gives
N αħ 2 Ne 2 −αr 2
ĤΨtrial = − �2αr 2 e−αr − 3e−αr � −
2 2
e
µ 4πε 0 r
E = � Ψtrial
∗
ĤΨtrial dτ
∞ N 2 αħ 2 N 2 e 2 −2αr 2
=� − �2αr 2 e−2αr − 3e−2αr � −
2 2
e dτ
0 µ 4πε 0 r
�e integral is best evaluated term by term. To evaluate the �rst term, Integral
G.� is used from the Resource section to give
∞ 2α 2 ħ 2 4 −2αr 2 3ħ 2 π 1�2
� − r e dr = − � �
0 µ 16µ 2α
Using Integral G.�, the second term gives
∞ 3αħ 2 2 −2αr 2 3ħ 2 π 1�2
� r e dr = � �
0 µ 8µ 2α
�e last term is evaluated using Integral G.�
∞ e 2 r −2αr 2 e2
� − e dr = −
0 4πε 0 16πεα
�erefore the energy is given by
3ħ 2 π π 1�2 e2
E = N2 � � � − �
4µ 2α 4ε 0 α
�e value of N 2 is found using the normalization condition given by [�B.�c–
∗
���], ∫ Ψtrial Ψtrial dτ = 1. Again, note that in polar coordinates the volume
element dτ is given by r 2 sin θ dθ d� dr, and because the wavefunction is spher-
ically symmetric, integration over all angles gives 4π. �erefore the normaliza-
tion condition becomes
∞
4πN 2 � r 2 e−2αr dr = 1
2
I�.� (a) All the coe�cients of the atomic orbitals for the molecular orbital ψ 1 are
positive, therefore it is a fully bonding molecular orbital. �e molecular
orbital ψ 3 has alternating positive, negative and then positive coe�cients,
therefore it is a fully antibonding molecular orbital. �e molecular orbital
ψ 2 is non-bonding, with no interaction between atomic orbitals on adja-
cent atoms.
(b) �e oxygen, carbon and nitrogen atoms all contribute one p orbital (the
�pz atomic orbital) to the π system, therefore they must be sp2 hybridized.
All the σ bonds to an sp2 hybridized atom must lie in the same plane,
therefore all the atoms connected to the oxygen, carbon and nitrogen
atoms must lie in the same plane together with the oxygen, carbon and
the nitrogen atoms.
(c) �e molecular orbital energy level diagram is shown in part (a). �ere
are four electrons in the π system of the peptide link. Two electrons come
from the formal π bond between the oxygen and the carbon atoms, and
two from the formal lone pair on the nitrogen atom. Hence the two low-
est energy molecular orbitals, that is the bonding and the non-bonding
molecular orbitals, are �lled.
(d) �is picture of bonding in the peptide group does not invoke delocal-
ization. �e lowest energy, bonding molecular orbital, ψ 4 , represents a
localized π bond between the oxygen and the carbon atoms. According
to this representation, the next highest energy molecular orbital, ψ 5 , is
the lone pair on nitrogen, a non-bonding molecular orbital. �e high-
est energy molecular orbital, ψ 6 is the antibonding π∗ molecular orbital
between the oxygen and the carbon atoms. �e four electrons in the π
system occupy the two lowest energy molecular orbitals.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 343
E O
C N ψ6
O ψ5
C N
O ψ4
C N
(e) Assume that the oxygen, carbon and nitrogen atoms are sp2 hybridized.
�e p orbital on nitrogen is in the plane in which the oxygen, carbon and
their neighbouring atoms are, therefore the two other atoms connected
to nitrogen must lie in a plane perpendicular to this one. �erefore the
oxygen, carbon, nitrogen and the neighbouring atoms cannot all lie in the
same plane.
(f) �e bonding molecular orbital ψ 1 must be lower in energy than ψ 4 , be-
cause ψ 1 extends over three atoms and is bonding between two pairs of
atoms, whereas ψ 4 extends over two atoms only and is bonding between
one pair of atoms only. Following the same logic, ψ 3 must be higher in
energy than ψ 6 . Disregarding the small di�erence in the Coulomb inte-
gral of nitrogen and oxygen, the nonbonding orbitals have approximately
the same energies.
(g) Only the two lowest energy molecular orbitals are occupied in either case.
Occupation of the non-bonding orbital does not result in any di�erence
in the π-electron binding energies. �e bonding molecular orbital in the
case of the planar peptide link is lower in energy, hence occupation of
this orbital results in greater π-electron binding energy when compared
with the occupation of the bonding molecular orbital in the non-planar
peptide link. �erefore planar geometry is the lower energy conformation
of the peptide link.
10 Molecular symmetry
D��A.�
Solutions to exercises
E��A.�(a) �e molecules to be assigned are shown in Fig. ��.�. For clarity not all symmetry
elements are shown.
C3,S3
C3
σv σv
F H C2',σv
C2,σv σh
C2 F σh
F P C2,i
σv F
N Cl Cl F F C2',σv
O O F Cl
Figure 10.1
In each case the point group is identi�ed using the �ow diagram in Fig. ��A.� on
page ��� having identi�ed the symmetry elements, or by drawing an analogy
with one of the shapes in the summary table in Fig. ��A.� on page ���.
346 10 MOLECULAR SYMMETRY
(i) Nitrogen dioxide, NO� , is a V-shaped molecule shape with a bond angle
of approximately 134○ . It possesses
• the identity, E
• a C 2 axis that lies in the plane of the molecule and passes through
the nitrogen atom.
• two di�erent vertical mirror planes σv , both containing the C 2 axis
but with one lying in the plane of the molecule and the other per-
pendicular to it.
�e point group is C 2v .
(ii) PF� has a trigonal bipyramidal shape with two axial and three equatorial
�uorine atoms. It possesses
• the identity, E
• a C 3 axis passing through the phosphorus and the two axial chlorine
atoms
• a horizontal mirror plane σh perpendicular to the C 3 axis and con-
taining the phosphorus and the three equatorial �uorine atoms
• three C 2′ axes perpendicular to the C 3 axis, each passing through
the phosphorus and one of the equatorial �uorine atoms. Only one
of the C 2′ axes is shown in Fig. ��.�.
• a S 3 axis coincident with the C 3 axis
• three vertical mirror planes σv , each containing the phosphorus, the
two axial �uorine atoms, and one of the three equatorial �uorine
atoms. Only one of the σv mirror planes is shown in Fig. ��.�.
�e point group is D 3h
(iii) CHCl� , chloroform, possesses
• the identity E
• a C 3 axis passing through the hydrogen and the carbon atom
• three vertical mirror planes σv , each containing the hydrogen, car-
bon, and one of the chlorine atoms; only one of these is shown in
Fig. ��.�.
�e point group is C 3v
(iv) �,�-di�uorobenzene possesses
• the identity, E
• a C 2 axis perpendicular to the plane of the molecule and passing
through the centre of the benzene ring
• two di�erent C 2′ axes, both perpendicular to the C 2 axis and lying
in the plane of the molecule. One C 2′ axis passes through the two
�uorine atoms, while the other perpendicular to this.
• a horizontal mirror plane σh lying in the plane of the molecule
• two di�erent vertical mirror planes σv , both containing the C 2 axis
and perpendicular to the plane of the molecule. One contains the two
�uorine atoms while the other is perpendicular to the line joining the
two �uorine atoms.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 347
C2,σv
σv σh
Cl
Cl Cl C2
Cl
Figure 10.2
E��A.�(a) �e molecules are shown in Fig. ��.� along with their point groups. For clarity
not all symmetry elements are shown.
C2,σv C2
σv
H
σ O H H
C2 B
H i B H
N N H
i C∞ C2
O
H H
BeH� , D∞h (Be not
Pyridine, C 2v Nitroethane, C s B� H� , D 2h
shown for clarity)
Figure 10.3
(i) Pyridine has point group C 2v , so it may be polar . �e dipole must lie
along the C 2 axis, which passes through the nitrogen and the carbon atom
opposite it in the ring.
348 10 MOLECULAR SYMMETRY
(ii) Nitroethane, CH� CH� NO� , belongs to point group C s , so it may be polar .
(iii) Linear BeH� belongs to point group D∞h , so it may not be polar.
(iv) Diborane, B� H� , belongs to point group D 2h , so it may not be polar.
E��A.�(a) �ere are �� distinct isomers of dichloronaphthalene, shown in Fig. ��.� to-
gether with their point groups. All isomers have a mirror plane in the plane of
the paper; additional symmetry elements are marked.
Cl Cl Cl
8 Cl
Cl 2 1
7 C2, σv C2
3 6
4 5 Cl Cl Cl
Cl Cl
Cl
C2
Cl C2, σv
Figure 10.4
E��A.�(a) As explained in Section ��A.�(b) on page ���, to be chiral a molecule must not
possess an axis of improper rotation, S n . �is includes mirror planes, which
are the same as S 1 , and centres of inversion, which are the same as S 2 . �e
table in Section ��A.�(c) on page ��� shows that a molecule belonging to D 2h
possesses three mirror planes, so such a molecule may not be chiral . Similarly
the table in Section ��A.�(b) on page ��� shows that a molecule belonging to
C 3h possesses a σh mirror plane, so such a molecule may not be chiral .
E��A.�(a) From the table in Section ��A.�(b) on page ��� a molecule belonging to the
point group C 3v , such as chloromethane, possesses
• the identity E
• a C 3 axis passing through the chlorine and carbon atoms
• three vertical mirror planes σv , each containing the chlorine, carbon and
one of the hydrogen atoms
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 349
C3
σv
Cl
H H
H
Figure 10.5
E��A.�(a) �e point group of the naphthalene molecule is identi�ed by following the �ow
diagram in Fig. ��A.� on page ���. Firstly, naphthalene is not linear, which
leads to the question “Two or more C n , n > 2?” Naphthalene has three two-fold
axes of rotation but no higher order axes, so the answer to this question is ‘No’.
However the fact that it does have three C 2 axes means that the answer to the
next question “C n ?” is ‘Yes’.
�is leads to the question “Select C n with the highest n; then, are there nC 2 per-
pendicular to C n ?” As already identi�ed there are three mutually perpendicular
C 2 axes and therefore no one axis with highest n, that is, no principal axis. As
explained in Section ��A.� on page ��� in the case of a planar molecule such as
naphthalene with more than one axis competing for the title of principal axis,
it is common to choose the one perpendicular to the plane of the molecule.
However, because all three axes are mutually perpendicular, whichever of the
three is selected it still has two other C 2 axes perpendicular to it, so the answer
to this question is necessarily ‘Yes’.
�is leads to the question “σh ?” to which the answer is ‘Yes’ because, whichever
C 2 axis is selected, there is a mirror plane perpendicular to it. In the case of the
C 2 axis perpendicular to the plane of the molecule, the σh mirror plane lies in
the plane of the molecule. Answering ‘Yes’ to this question leads to the result
D nh , and because the C n axis with highest n in this molecule is C 2 , it follows
that the point group is D 2h .
Alternatively, the point group may be identi�ed from the table in Fig. ��A.� on
page ��� by drawing an analogy between the planar naphthalene molecule and
the rectangle which belongs to D 2h .
From the table in Section ��A.�(c) on page ��� a molecule belonging to the
point group D 2h possesses
• the identity E
• a C 2 axis , which in the case of a planar molecule such as naphthalene is
commonly taken to be the axis perpendicular to the plane of the molecule.
�is axis passes through the mid-point of the central bond joining the two
rings.
• two C 2′ axes , perpendicular to the C 2 axis and lying in the plane of the
molecule. One of these passes along the line of the central bond joining
350 10 MOLECULAR SYMMETRY
C2',σv
σh
C2,i C2',σv
Figure 10.6
E��A.�(a) �e objects to be assigned are shown in Fig. ��.�. For clarity not all symmetry
elements are shown.
C2,σv
C2,σv
σv C3,S3
C2,σv C2,σv i C∞
Figure 10.7
(iv) Modelling the unsharpened cylindrical pencil as a cylinder (Fig. ��.�) and
assuming no lettering or other pattern on it, its point group is determined
using the �ow diagram in Fig. ��A.� on page ���. It is linear, in the sense
that rotation by any angle around the long axis of the pencil is a symmetry
operation so that the pencil possesses a C∞ axis. Because the pencil has
not been sharpened, both ends are the same and this means that the pencil
possesses a centre of inversion i. Using the �ow diagram this establishes
the point group as D∞h .
Solutions to problems
P��A.� �e molecules concerned are shown in Fig. ��.�. For clarity not every symmetry
element is shown in each diagram.
(a) C2 σd
H H H H
H H H H
i C3 C3 H
C2
C3,S6 H H H H
σd H H H σd H σd
H
C2
(b) σd C2
C2
C3 C3
i C2
σd σv
σd σd σv
C2 C2
C3,S6
C2
C2 (d)
(c) H2N
H2
H N NH2
C2 H H Co C3
B i B C2
H H N NH2
H H2 H N
C2 2
C2
(e)
C2 C2 σd
σd σd
C2
C2
S S C4 C4 σd
S
S S
S
σd S S
C4,S8
Figure 10.8
elements
• �e identity E
• A C 3 axis along the the C–C bond
• �ree C 2 axes perpendicular to the C 3 axis. �ese are best seen in
the Newman projection also shown in Fig. ��.�(a), that is, the view
along the C 3 axis
• �ree dihedral mirror planes σd , each containing the C 3 principal
axis and two hydrogen atoms. �ese mirror planes are ‘dihedral’
rather than ‘vertical’ because they bisect the angles between the C 2
axes. �e three σd mirror planes are seen most easily in the Newman
projection; for clarity only one σd plane is shown
• An S 6 axis coincident with the C 3 axis
• A centre of inversion i at the midpoint of the C–C bond.
Using the �ow diagram in Fig. ��A.� on page ��� the point group is D 3d .
(b) �e chair conformation of cyclohexane possesses the following symmetry
elements
• �e identity E
• A C 3 axis perpendicular to the approximate plane of the molecule
• �ree C 2 axis perpendicular to the C 3 axis, each passing through
the midpoint of two opposite C–C bonds
• �ree dihedral mirror planes σd , each containing the C 3 axis as well
as two opposite carbon atoms
• An S 6 axis coincident with the C 3 axis
• A centre of inversion i at the intersection of the C 3 and C 2 axes.
Using the �ow diagram in Fig. ��A.� on page ��� the point group is D 3d .
�e boat conformation of cyclohexane possesses
• �e identity E
• A C 2 axis passing vertically through the centre of the boat where the
mast would be
• Two vertical mirror planes σv , both containing the C 2 axis but one
also containing the bow and stern of the boat and the other perpen-
dicular to this
Using the �ow diagram in Fig. ��A.� on page ��� the point group is C 2v .
(c) Diborane, B� H� , possesses
• �e identity E
• �ree di�erent C 2 axes , all perpendicular to each other. One passes
through the two bridging hydrogen atoms, one passes through the
two boron atoms, and the third is perpendicular to the plane de�ned
by the boron and bridging hydrogen atoms.
• �ree di�erent mirror planes σ , one perpendicular to each of the
three C 2 axes. One mirror plane contains the boron atoms and the
four terminal hydrogen atoms. �e second contains the two bridging
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 353
P��A.� �e molecules are shown in Fig. ��.� along with some of their key symmetry
elements.
(a) In addition to the identity element, ethene possesses three C 2 axes, three
mirror planes, and a centre of inversion i. As explained in Section ��A.�
on page ���, in the case of a planar molecule with several C n axes com-
peting for the title of principal axis it is common to choose the axis per-
pendicular to plane of the molecule to be the principal axis. �is axis is
labelled C 2 in Fig. ��.�, while the other C 2 axes are labelled C 2′ . �e mirror
plane that lies in the plane of the molecule is denoted σh while the other
354 10 MOLECULAR SYMMETRY
(a) Ethene
C2',σv
Allene σd
σh C2'
H σd H
C2,i H H
H C2
C C C H H σ
C2',σv C2,S4 H d
H H H H
C2'
C2'
(iii) 45o (iv) 60o
C2' C2'
C2 45o C2 60o
C2' C2'
Figure 10.9
two are denoted σv . Using the �ow diagram in Fig. ��A.� on page ��� the
point group is established as D 2h .
In the case of allene there is a C 2 axis, coincident with an S 4 axis, that
passes through all three carbon atoms. �ere are also two other C 2 axes,
denoted C 2′ , that are perpendicular to C 2 and which pass through the
central carbon; these are most clearly seen in the Newman projection
shown in Fig. ��.�(a). Finally there are two dihedral mirror planes σd
which bisect the angle between the two C 2′ axes and which each pass
through all three carbon atoms and two of the hydrogen atoms. Using the
�ow diagram in Fig. ��A.� on page ��� these symmetry elements establish
the point group as D 2d .
(b) (i) �e biphenyl molecule with a dihedral angle of 0○ , that is, with both
phenyl groups in the same plane, has a shape analogous to that of the
ethene molecule from part (a). Like ethene it belongs to the point
group D 2h and possesses the symmetry elements E, C 2 , 2C 2′ , 2σv ,
σh and i.
(ii) �e biphenyl molecule with a dihedral angle of 90○ , that is, with the
two rings perpendicular, has a shape analogous to that of allene. Like
allene it belongs to the point group D 2d and possesses the symmetry
elements E, C 2 , 2C 2′ , 2σd , and S 4 .
(iii) If the dihedral angle is changed from 90○ to 45○ the C 2 and C 2′ axes
are retained but the σd mirror planes and the S 4 axis are no longer
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 355
present. Using the �ow diagram in Fig. ��A.� on page ��� the sym-
metry elements establish the point group as D 2 .
(iv) �e biphenyl molecule with a dihedral angle of 60○ possesses the
same set of symmetry elements as when the dihedral angle is 45○
and therefore belongs to the same point group of D 2 .
P��A.� �e ion is shown in Fig. ��.��. In each diagram only the mirror plane referred
to in the question is shown; any other mirror planes are omitted for clarity.
C2 C2
σ σv
σh F F
F F CN
i CN F
F3C CF3 i CN
NC
C2 F F
C2 F F F
F C2 F
Figure 10.10
(a) If the CF� groups are treated as structureless ligands then in addition to
the identity E the ion possesses three C 2 axes, a centre of inversion, and
three mirror planes, only one of which is shown in Fig. ��.��. �e point
group is D 2h .
(b) Fig. ��.�� shows the staggered and eclipsed conformations of the CF� groups.
In the staggered arrangement the centre of inversion i is retained, as is the
C 2 axis that lies along the CN–Ag–CN bond and the mirror plane perpen-
dicular to this, but the other C 2 axes and mirror planes are lost. �e point
group is C 2h . In the eclipsed arrangement, the C 2 axis perpendicular to
the plane of the Ag and the four ligands is retained, as are the two mirror
planes containing this axis, only one of which is shown in Fig. ��.��. �e
other C 2 axes, the other mirror plane, and the centre of inversion are lost.
�e point group is C 2v .
sentations are labelled E and T, respectively, and have characters � and � under
the operation E.
On the right of the table various cartesian functions are listed on the row cor-
responding to the irreducible representation as which they transform. Where
a pair (or more) of functions transform as a two (or higher) dimensional rep-
resentation, the functions are bracketed together.
�e order of the group, h, is equal to the number of operations. Alternatively,
as the operations are grouped into classes, the order is the sum of the number
of operations belonging to each class.
�e number of irreducible representations equals the number of classes. Also,
apart from the groups C n with n > 2, the sum of the squares of the dimensions
of each irreducible representation is equal to the order.
Solutions to exercises
E��B.�(a) BF� belongs to the point group D 3h . �e σh operation corresponds to a re-
�ection in the plane of the molecule, while the C 3 operation corresponds to
rotation by 120○ around the C 3 axis which passes through the boron atom and
is perpendicular to the plane of the molecule (Fig. ��.��).
Using the orbital numbering shown in Fig. ��.��, the matrix representatives
for the σh and C 3 operations were found in Exercise E��B.�(a) and Exercise
E��B.�(b) to be
� −1 0 0 0 � �1 0 0 0�
� 0 −1 0 0 � �0 0 0 1�
D(σh ) = � � D(C 3 ) = � �
� 0 0 � �0 0�
and
0 −1 1 0
� 0 0 0 −1 � �0 0 1 0�
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 357
C3
p1
p3 p2
σh
p4
Figure 10.11
� −1 0 0 0 �� 1 0 0 0� � −1 0 0 0 �
� 0 −1 0 0 �� 0 0 0 1� � 0 0 0 −1 �
D(σh )D(C 3 ) = � �� �= � �
� 0 0 −1 0 �� 0 1 0 0� � 0 −1 0 0 �
� 0 0 0 −1 � � 0 0 1 0� � 0 0 −1 0 �
� −1 0 0 0 �
� 0 0 0 −1 �
( p1 p2 p3 p4 ) � � = ( −p1 −p3 −p4 −p2 )
� 0 −1 0 0 �
� 0 0 −1 0 �
�e operation σh C 3 therefore changes the sign of p1 , and converts p2 , p3 and
p4 into −p3 , −p4 , and −p2 respectively. �is is precisely the same outcome as
achieved by the S 3 operation , that is, a ���○ rotation around the C 3 axis fol-
lowed by a re�ection in the σh plane. �us, D(σh )D(C 3 ) = D(S 3 ), as expected
from the fact that by de�nition the S 3 operation corresponds to a C 3 rotation
followed by a re�ection in the plane perpendicular to the C 3 axis.
E��B.�(a) Fig. ��.�� shows BF� , an example of a molecule with D 3h symmetry. �e three
C 2 axes are labelled C 2 , C 2′ and C 2′′ , and likewise for the σv mirror planes.
C2',σv'
C2',σv'
C2',σv'
4 F 1
2 F 3 F
2
1 3 4
F F F F F F
C2'',σv'' 3 C2,σv C2'',σv'' 4 C2,σv C2'',σv'' C2,σv
2 1
Figure 10.12
1 0 for i ≠ j
� N(C)χ (C)χ (C) = �
Γ (i) Γ ( j)
h C 1 for i = j
where the sum is over all classes of the group, N(C) is the number of operations
in class C, and h is the number of operations in the group (its order). χ Γ (C)
(i)
(i)
is the character of class C in irreducible representation Γ , and similarly for
χ Γ (C).
( j)
�e character table for the point group C 2h is available in the Online resource
centre. �e operations of the group are {E, C 2 , i, σh }, so h = 4. �ere are four
classes and in this group each class has just one member, so all N(C) = 1. �e
irreducible representations have the following characters
Ag and Bg 1
4
{1×1×1 + 1×1×(−1) + 1×1×1 + 1×1×(−1)} = 0
Ag and Au 1
4
{1×1×1 + 1×1×1 + 1×1×(−1) + 1×1×(−1)} = 0
Ag and Bu 1
4
{1×1×1 + 1×1×(−1) + 1×1×(−1) + 1×1×1} = 0
Bg and Au 1
4
{1×1×1 + 1×(−1)×1 + 1×1×(−1) + 1×(−1)×(−1)} = 0
Bg and Bu 1
4
{1×1×1 + 1×(−1)×(−1) + 1×1×(−1) + 1×(−1)×1} = 0
Au and Bu 1
4
{1×1×1 + 1×1×(−1) + 1×(−1)×(−1) + 1×(−1)×1} = 0
Ag and Ag 1
4
{1×1×1 + 1×1×1 + 1×1×1 + 1×1×1} = 1
Bg and Bg 1
4
{1×1×1 + 1×(−1)×(−1) + 1×1×1 + 1×(−1)×(−1)} = 1
Au and Au 1
4
{1×1×1 + 1×1×1 + 1×(−1)×(−1) + 1×(−1)×(−1)} = 1
Bu and Bu 1
4
{1×1×1 + 1×(−1)×(−1) + 1×(−1)×(−1) + 1×1×1} = 1
z z
C3 C3
p1
p3 p2 σh
σh σh
p4
Figure 10.13
As shown in Fig. ��.�� the e�ect of this operation is to change the sign of all
four p orbitals. �is e�ect is written as ( −p1 −p2 −p3 −p4 ) ← ( p1 p2 p3 p4 )
360 10 MOLECULAR SYMMETRY
� −1 0 0 0 �
� 0 −1 0 0 �
D(σh ) = � �
� 0 0 −1 0 �
� 0 0 0 −1 �
Solutions to problems
P��B.� Fig. ��.�� shows trans-di�uoroethene, an example of a molecule with C 2h sym-
metry.
C2
1
2
H i F
σh F H
Figure 10.14
R ↓ R′ → E C2 σh i
E E C2 σh i
C2 C2 E i σh
σh σh i E C2
i i σh C2 E
P��B.� Fig. ��.�� shows that a water molecule together with the speci�ed axis system
and the six basis orbitals, which are labelled sA , sB , sO , px , p y , and pz . �e σv
plane is the xz plane, and the σv′ plane is the yz plane.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 361
z z z
C2 pz
sO
O y y py y
x HA HB x x px
sA sB
Figure 10.15
D(C 2 )
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
� 0 1 0 0 0 0 �
� 1 0 0 0 0 0 �
� �
� 0 0 1 0 0 0 �
( sB sA sO −px −p y pz ) = ( sA sB sO px p y pz )�
� 0 0 0 −1 0 0 �
�
� �
� 0 0 0 0 −1 0 �
� �
� 0 0 0 0 0 1 �
D(σ v )
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��
� 0 1 0 0 0 0 �
� 1 0 0 0 0 0 �
� �
� 0 0 1 0 0 0 �
( sB sA sO px −p y pz ) = ( sA sB sO px p y pz )�
� 0 0 0 1 0 0 �
�
� �
� 0 0 0 0 −1 0 �
� �
� 0 0 0 0 0 1 �
�e σv′ operation leaves all orbitals unchanged except px , which changes sign.
D(σ v′ )
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��
� 1 0 0 0 0 0 �
� 0 1 0 0 0 0 �
� �
� 0 0 1 0 0 0 �
( sA sB sO −px p y pz ) = ( sA sB sO px p y pz )�
� 0 0 0 −1 0 0 �
�
� �
� 0 0 0 0 1 0 �
� �
� 0 0 0 0 0 1 �
362 10 MOLECULAR SYMMETRY
D(E)
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
� 1 0 0 0 0 0 �
� 0 1 0 0 0 0 �
� �
� 0 0 1 0 0 0 �
( sA sB sO px p y pz ) = ( sA sB sO px p y pz )�
� 0 0 0 1 0 0 �
�
� �
� 0 0 0 0 1 0 �
� �
� 0 0 0 0 0 1 �
� 0 0 0 0 �
� �
� 0 0 0 0 �
� �
D=� �
0 0 0 0 0
� �
� 0 0 0 0 0 �
� �
� 0 0 0 0 0 �
� 0 0 0 0 0 �
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 363
�1 0� �1 0 � �1 0 � �1 0�
D(E) = D(C 2 ) = D(σv ) = D(σv′ ) =
�0 1� � 0 −1 � � 0 −1 � �0 1�
�e new representatives are all in block diagonal format, in this case in
0
the form � �, which means that the two combinations are not mixed
0
with each other by any operation of the group. �e two dimensional basis
( s1 s2 ) can therefore be cut into two one-dimensional bases correspond-
ing to s1 and s2 . �e representations for these bases are
For s1 : D(E) = 1 D(C 2 ) = 1 D(σv ) = 1 D(σv′ ) = 1
For s2 : D(E) = 1 D(C 2 ) = −1 D(σv ) = −1 D(σv′ ) = 1
Because the characters of one-dimensional representatives are just the
representatives themselves, inspection of the C 2v character table imme-
diately indicates that these one-dimensional representations correspond
364 10 MOLECULAR SYMMETRY
P��B.� �e ethene molecule and axis system are shown in Fig. ��.��.
y
y
C2,σyz
σxy
sB H H s
C2z A
x
x
sC sD C2,σzx
H H
Figure 10.16
( sD sC sB sA ) ← ( sA sB sC sD )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 1 0 0 �
� 1 0 0 0 �
For C 2 : ( sB sA sD sC ) = ( sA sB sC sD )� �
y
� 0 0 0 1 �
� 0 0 1 0 �
D(C 2z )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 0 1 0 �
� 0 0 0 1 �
For C 2z : ( sC sD sA sB ) = ( sA sB sC sD )� �
� 1 0 0 0 �
� 0 1 0 0 �
D(i)
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 0 1 0 �
� 0 0 0 1 �
( sC sD sA sB ) = ( sA sB sC sD )� �
� 1 0 0 0 �
For i:
� 0 1 0 0 �
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 365
D(σ x y )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 1 0 0 0 �
� 0 1 0 0 �
For σ x y : ( sA sB sC sD ) = ( sA sB sC sD )� �
� 0 0 1 0 �
� 0 0 0 1 �
D(σ yz )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 1 0 0 �
� 1 0 0 0 �
For σ yz : ( sB sA sD sC ) = ( sA sB sC sD )� �
� 0 0 0 1 �
� 0 0 1 0 �
D(σ zx )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 0 0 1 �
� 0 0 1 0 �
For σ zx : ( sD sC sB sA ) = ( sA sB sC sD )� �
� 0 1 0 0 �
� 1 0 0 0 �
D(E)
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 0 1 0 �
� 0 0 0 1 �
( sC sD sA sB ) = ( sA sB sC sD )� �
� 1 0 0 0 �
For E:
� 0 1 0 0 �
Solutions to exercises
E��C.�(a) �e D 2h character table shows that x y spans B1g in D 2h . To show this explicitly,
the direct product is formed between the irreducible representations spanned
by x and y individually. �e character table shows that x spans B3u and y spans
B2u ; the direct product B3u × B2u is calculated using the method described in
Section ��C.�(a) on page ���
y
E C 2x C 2 C 2z i σ x y σ yz σ zx
B3u � −1 −1 � −1 � −1 �
B2u � −1 � −1 −1 � � −1
product � � −1 −1 � � −1 −1
�e characters in the product row are those of symmetry species B1g , thus
con�rming that the function x y has symmetry species B1g in D 2h .
E��C.�(a) As explained in Section ��C.�(a) on page ���, only orbitals of the same symme-
try species may have a nonzero overlap. Inspection of the C 2v character table
shows that the oxygen �s and �pz orbitals both have A1 symmetry, so they
can interact with the A1 combination of �uorine orbitals. �e oxygen �p y
orbital has B2 symmetry, so it can interact with the B2 combination of �uorine
orbitals. �e oxygen �px orbital has B1 symmetry, so cannot interact with either
combination of �uorine orbitals and therefore remains nonbonding.
In SF� the same interactions with the sulfur s and p orbitals are possible but
there is now the possibility of additional interactions involving the d orbitals.
�e C 2v character table shows that dz 2 and dx 2 −y 2 have A1 symmetry, so they
can interact with the A1 combination of �uorine orbitals. �e d yz orbital
has B2 symmetry so can interact with the B2 combination. �e dx y and dzx
orbitals have A2 and B1 symmetry respectively, so they cannot interact with
either combination of �uorine orbitals and therefore remain nonbonding.
E��C.�(a) As explained in Section ��C.�(a) on page ���, only orbitals of the same symme-
try species may have a nonzero overlap. Inspection of the C 2v character table
shows that the nitrogen �s, px , p y , and pz orbitals span A1 , B1 , B2 , and A1
respectively. Because none of these orbitals have A2 symmetry, none of them
can interact with the A2 combination of oxygen orbitals.
�e character table also shows that the dz 2 , dx 2 −y 2 , dx y , d yz , and dzx orbitals
of the sulfur in SO� transform as A1 , A1 , A2 , B2 and B1 respectively. �erefore
the dx y orbital has the correct symmetry to interact with the A2 combination
of oxygen p orbitals.
E��C.�(a) As explained in Section ��C.� on page ���, a transition from a state with sym-
metry Γ(i) to one with symmetry Γ(f) is only allowed if the direct product
368 10 MOLECULAR SYMMETRY
Similarly
n(A1 ) = 1
16
�N(E)× χ(A1g ) (E)× χ(E) + N(C 2′ )× χ(A1g ) (C 2′ )× χ(C 2′ )
+ N(σh )× χ(A1g ) (σh )× χ(σh ) + N(σv )× χ(A1g ) (σv )× χ(σv )�
= 1
16
(1×1×4 + 2×1×2 + 1×1×4 + 2×1×2) = 1
Similarly
n(A2g ) = 1
16
(1×1×4 + 2×(−1)×2 + 1×1×4 + 2×(−1)×2) = 0
n(B1g ) = 16 (1×1×4 + 2×1×2 + 1×1×4 + 2×1×2) = 1
1
n(B2g ) = 161
(1×1×4 + 2×(−1)×2 + 1×1×4 + 2×(−1)×2) = 0
n(Eg ) = 16 (1×2×4 + 2×0×2 + 1×(−2)×4 + 2×0×2) = 0
1
n(A1u ) = 161
(1×1×4 + 2×1×2 + 1×(−1)×4 + 2×(−1)×2) = 0
n(A2u ) = 16 (1×1×4 + 2×(−1)×2 + 1×(−1)×4 + 2×1×2) = 0
1
n(B1u ) = 161
(1×1×4 + 2×1×2 + 1×(−1)×4 + 2×(−1)×2) = 0
n(B2u ) = 16 (1×1×4 + 2×(−1)×2 + 1×(−1)×4 + 2×1×2) = 0
1
n(Eu ) = 16
1
(1×2×4 + 2×0×2 + 1×0×4 + 2×1×2) = 1
E��C.�(a) As explained in Section ��C.� on page ���, a transition from a state with sym-
metry Γ(i) to one with symmetry Γ(f) is only allowed if the direct product Γ(f) ×
Γ(q) × Γ(i) contains the totally symmetric irreducible representation, which
for both molecules is A1g . �e ground state is totally symmetric, implying
that it transforms as A1g . �erefore the direct product becomes Γ(f) × Γ(q) ×
A1g . �is is simply Γ(f) × Γ(q) because, from the �rst simplifying feature of
direct products listed in Section ��C.�(a) on page ���, the direct product of the
totally symmetric representation A1g with any other representation is the latter
representation itself.
If Γ(f) × Γ(q) is to be A1g , then Γ(f) must equal Γ(q) because, from the second
simplifying feature of direct products listed in Section ��C.�(a) on page ���,
the direct product of two irreducible representations only contains the totally
symmetric irreducible representation if the two irreducible representations are
identical.
E��C.�(a) As explained in Section ��C.� on page ��� an integral can only be non-zero if
the integrand spans the totally symmetric irreducible representation, which in
C 2v is A1 . From Section ��C.�(a) on page ��� the symmetry species spanned by
the integrand px zpz is found by the forming the direct product of the symmetry
species spanned by px , z, and pz separately. �ese are read o� the C 2v character
table by looking for the appropriate Cartesian functions listed on the right of
the table: x and hence px spans B1 , while z and pz both span A1 . �e direct
product required is therefore B1 × A1 × A1 .
E��C.�(a) As explained in Section ��C.� on page ���, an electric dipole transition is for-
bidden if the electric transition dipole moment µ q,fi is zero. �e transition
dipole moment is given by [��C.�–���], µ q,fi = −e ∫ ψ ∗f qψ i dτ where q is x,
y, or z. �e integral is only non-zero if the integrand contains the totally sym-
metric representation, which from the C 3v character table is A1 .
Solutions to problems
P��C.� Methane belongs to point group Td . �e methane molecule and its H�s orbitals
are shown in Fig. ��.��, along with one operation of each class. It is su�cient to
consider just one operation in each class because, by de�nition, all operations
the same class have the same character.
C3
C2,S4 σd
sA
H
sB
sD
H H
H
sC
Figure 10.17
D(C 3 )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 1 0 0 0 �
� 0 0 0 1 �
( sA sC sD sB ) = ( sA sB sC sD )� �
� 0 1 0 0 �
� 0 0 1 0 �
D(C 2 )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 1 0 0 �
� 1 0 0 0 �
( sB sA sD sC ) = ( sA sB sC sD )� �
� 0 0 0 1 �
� 0 0 1 0 �
D(σ d )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 1 0 0 0 �
� 0 1 0 0 �
( sA sB sD sc ) = ( sA sB sC sD )� �
� 0 0 0 1 �
� 0 0 1 0 �
D(S 4 )
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 0 0 1 0 �
� 0 0 0 1 �
( sD sC sA sB ) = ( sA sB sC sD )� �
� 0 1 0 0 �
� 1 0 0 0 �
372 10 MOLECULAR SYMMETRY
Finally, the E operation leaves all orbitals unchanged, meaning that its repre-
sentative is simply the identity matrix
D(E)
��� � � � � � � � � � � � � � � � � � � � � � ��� � � � � � � � � � � � � � � � � � � � � � �
� 1 0 0 0 �
� 0 1 0 0 �
( sA sB sC sD ) = ( sA sB sC sD )� �
� 0 0 1 0 �
� 0 0 0 1 �
�is result can be arrived at much more quickly by noting that: (�) only the
diagonal elements of the representative matrix contribute to the trace; (�) or-
bitals which are unmoved by an operation will result in a � on the diagonal; (�)
orbitals which are moved to other positions by an operation will result in a � on
the diagonal. �e character is found simply by counting the number of orbitals
which do not move. In the present case � are unmoved by E, � is unmoved by
C 3 , none are unmoved by C 2 , � are unmoved by σd , and none are unmoved by
S 4 . �e characters are thus {4, 1, 0, 2, 0}.
�is representation is decomposed using the method described in Section ��C.�(b)
on page ���. �e number of times n(Γ) that a given irreducible representation
Γ occurs in a representation is given by [��C.�a–���],
1
n(Γ) = � N(C)χ (C)χ(C)
(Γ)
h C
where h is the order of the group, N(C) is the number of operations in class C,
χ(Γ) is the character of class C in the irreducible representation Γ, and χ(C) is
the character of class C in the representation being reduced. In the case of the
group Td , h = 24. �e number of times that the irreducible representation A1
occurs is
n(A1 ) = 1
24
�N(E)× χ(A1 ) (E)× χ(E) + N(C 3 )× χ(A1 ) (C 3 )× χ(C 3 )
+ N(C 2 )× χ(A1 ) (C 2 )× χ(C 2 ) + N(σd )× χ(A1 ) (σd )× χ(σd )
+ N(S 4 )× χ(A1 ) (S 4 )× χ(S 4 )�
= 1
24
(1×1×4 + 8×1×1 + 3×1×0 + 6×1×2 + 6×1×0) = 1
Similarly
n(A2 ) = 1
24
(1×1×4 + 8×1×1 + 3×1×0 + 6×(−1)×2 + 6×(−1)×0) = 0
n(E) = 24
1
(1×2×4 + 8×(−1)×1 + 3×2×0 + 6×0×2 + 6×0×0) = 0
n(T1 ) = 24 (1×3×4 + 8×0×1 +
1
3×(−1)×0 + 6×(−1)×2 + 6×1×0) = 0
n(T2 ) = 24
1
(1×3×4 + 8×0×1 + 3×(−1)×0 + 6×1×2 + 6×(−1)×0) = 1
As explained in Section ��C.�(a) on page ���, only orbitals of the same sym-
metry species may have a nonzero overlap. �e carbon �s orbital spans the
totally symmetric irreducible representation A1 as it is unchanged under all
symmetry operations in the group. It can therefore form molecular orbitals
with the A1 combination of hydrogen orbitals. Inspection of the Td character
table shows that the carbon px , p y , and pz orbitals jointly span T2 , so they can
form molecular orbitals with the T2 combinations of hydrogen orbitals.
�e character table also shows that the dz 2 and dx 2 −y 2 orbitals on the silicon in
SiH� jointly span E; note that the dz 2 orbital appears as (3z 2 − r 2 ) in the char-
acter table. �ese d orbitals therefore cannot form molecular orbitals with the
A1 or T2 combinations of hydrogen orbitals. However, the dx y , d yz , and dzx
orbitals on the silicon span T2 and can therefore form molecular orbitals with
the T2 combinations of hydrogen orbitals.
P��C.� As explained in Section ��C.� on page ���, an integral over a region will neces-
sarily vanish unless the integrand, or part thereof, spans the totally symmetric
irreducible representation of the point group of the region. In the case of the
function 3x 2 − 1, the −1 part will always span the totally symmetric represen-
tation because it is invariant to all symmetry operations. �erefore the integral
will not necessarily vanish in any of the ranges.
P��C.� �e p 1 symmetry adapted linear combination is shown in Fig. ��.��, along with
some of the symmetry elements.
C2',σv C2'',σd
σh
A
B
x
D
C4,C2,i,S4
C
Figure 10.18
class C E C 4 C 2 C 2′ C 2′′ i S 4 σh σv σd
D(C) � −1 � −1 � −1 � −1 � −1
y
y
C2,σyz
σxy
sB H H s
C2z A
x
x
sC sD C2,σzx
H H
Figure 10.19
y
Row E C 2z C2 C 2x i σxy σ yz σ zx
� e�ect on sA sA sC sB sD sC sA sB sD
� characters for Ag � � � � � � � �
� product of rows � and � sA sC sB sD sC sA sB sD
� characters for B2u � −1 � −1 −1 � � −1
� product of rows � and � sA −sC sB −sD −sC sA sB −sD
� characters for B3u � −1 −1 � −1 � −1 �
� product of rows � and � sA −sC −sB sD −sC sA −sB sD
� characters for B1g � � −1 −1 � � −1 −1
� product of rows � and � sA sC −sB −sD sC sA −sB −sD
�� characters for B1u � � −1 −1 −1 −1 � �
�� product of rows � and �� sA sC −sB −sD −sC −sA sB sD
�e results from row �� show that attempting to project out a SALC with sym-
metry B1u gives zero . �is is because the four hydrogen �s orbitals do not span
B1u .
11 Molecular Spectroscopy
D��A.� Doppler broadening. �is contribution to the linewidth is due to the Doppler
e�ect which shi�s the frequency of the radiation emitted or absorbed when the
molecules involved are moving towards or away from the detecting device. In
a gas, molecules have a wide range of speeds in all directions and the detected
spectral line is the absorption or emission pro�le arising from the resulting
Doppler shi�s. �e shape of a Doppler-broadened spectral line re�ects the
Maxwell distribution of speeds in the sample.
Lifetime broadening. �is kind of broadening is a quantum mechanical e�ect
which predicts that for a state with a lifetime τ there is an energy uncertainty
δE given by δE ≈ ħ�τ. �is uncertainty in the energy translates to absorption
(or emission) over a range of frequencies and hence a linewidth. �e lifetime
of a state may be limited by the rate of spontaneous emission from the state, in
which case the resulting broadening is called natural line broadening.
Collisions between molecules are e�cient at changing their rotational and vi-
brational energies, and therefore the lifetimes of such states are limited by the
the collision rate. �e resulting line broadening is called collisional or pressure
line broadening.
�e rate of spontaneous emission cannot be changed; hence its contribution is
the same regardless of phase. Doppler broadening is expected to contribute in a
similar way for both gases and liquids. �e higher density of liquids compared
to gases implies that collisions will be more frequent and hence the collisional
line broadening will be greater for a liquid.
Solutions to exercises
E��A.�(a) If a light source of frequency ν 0 is approached at a speed s, the Doppler shi�ed
frequency ν a is [��A.��a–���],
1 + s�c
1�2
νa = ν0 � �
1 − s�c
378 11 MOLECULAR SPECTROSCOPY
Writing the frequencies in terms of the wavelength as ν = c�λ and then invert-
ing before sides gives
1 − s�c
1�2
λa = λ0 � �
1 + s�c
At nonrelativistic speeds, s � c, this simpli�es to λ a = λ 0 (1 − s�c)
1�2
. Hence
Within the precision of the data given, the Doppler shi� is insigni�cant.
δ ν̃ = [2π × (2.9979 × 1010 cm s−1 ) × 100�(1.0 × 1013 s−1 )]−1 = �.�� cm−1
I�I 0 = 10−ε[J]L = 10−(723 dm mol−1 cm−1 )×(4.25×10−3 mol dm−3 )×(0.250 cm)
3
= 0.171...
Note the use of L in cm and the conversion of [J] to mol dm−3 in order to give
the usual units of ε.
E��A.�(a) �e transmittance T is the ratio I�I 0 , hence the Beer–Lambert law [��A.�–���]
can be written T = I�I 0 = 10−ε[J]L . It follows that log T = −ε[J]L and hence
ε = −(log T)�[J]L. With this, the following table is drawn up, with L = 0.20 cm
380 11 MOLECULAR SPECTROSCOPY
E��A.�(a) �e transmittance T is the ratio I�I 0 , hence the Beer–Lambert law [��A.�–���]
can be written T = I�I 0 = 10−ε[J]L . It follows that log T = −ε[J]L and hence
ε = −(log T)�[J]L. With the given data, T = 0.48 and L = 0.20 cm, the molar
absorption coe�cient is calculated as
ε = −(log 0.48)�[(0.010 mol dm−3 )×(0.20 cm)] = 1.59...×102 dm3 mol−1 cm−1
Hence T = 23% .
E��A.��(a) �e ratio of the incident to the transmitted intensities of light a�er passing
through a sample of length L, molar concentration [H2 O], and molar absorp-
tion coe�cient ε is given by [��A.�–���], T = I�I 0 = 10−ε[H2 O]L . It follows that
log T = −ε[H2 O]L, which rearranges to give L = −(log T)�ε[H2 O].
�e molar concentration of H� O is calculated by noting that its mass density
is ρ = 1000 kg m−3 and its molar mass is M = 18.016 g mol−1 . �e con-
centration is therefore ρ�M = (1000 kg m−3 )�(18.016 × 10−3 kg mol−1 ) =
55.5... × 103 mol m−3 = 55.5... mol dm−3 .
�e light intensity is half that at the surface when T = 0.5, hence the depth is
calculated as
A= 1
2
× [(4.54... − 3.33...) × 104 cm−1 ] × (2.21 × 104 dm3 mol−1 cm−1 )
= 1.33... × 108 dm3 mol−1 cm−2
Solutions to problems
P��A.� �e fraction of the incident photons that reach the retina is (1 − 0.30) × (1 −
0.25)×(1−0.09)×(1−0.43) = 0.272.... Hence the number of photons reaching
the retina in �.� s is
(4.0 × 103 mm−2 s−1 ) × (40 mm2 ) × (0.1 s) × 0.272... = 4.4 × 103
ε A2 A 1 = ε A2 ε A1 [A]L + ε A2 ε B1 [B]L
ε A1 A 2 = ε A1 ε A2 [A]L + ε A1 ε B2 [B]L
Subtracting the two equations eliminates [A], and rearrangement gives the re-
quired expression for [B]
ε A2 A 1 − ε A1 A 2
ε A2 A 1 − ε A1 A 2 = ε A2 ε B1 [B]L − ε A1 ε B2 [B]L [B] =
(ε A2 ε B1 − ε A1 ε B2 )L
Simply exchanging the labels A and B gives the corresponding expression for
[A]
ε B2 A 1 − ε B1 A 2
[A] =
(ε B2 ε A1 − ε B1 ε A2 )L
382 11 MOLECULAR SPECTROSCOPY
P��A.� Following the hint, a plot is made of ln ε against ν̃; the data are shown in the
following table and the plot is shown in Fig. ��.�.
7
ln[ε�(dm3 mol−1 cm−1 )]
= e (1�a) �e
b
−e �
a ν̃ max a ν̃ min
1
= e−35.793 �e(1.2609×10 )×(3.425×10 ) − e(1.2609×10 )×(3.125×10 ) �
−3 4 −3 4
1.2609 × 10−3
A= 1
2
× [(34483 − 31250) cm−1 ] × (150 dm3 mol−1 cm−1 )
= 2.42 × 105 dm3 mol−1 cm−2
A = ε[M]L
= (150 dm3 mol−1 cm−1 ) × (1.02... × 10−4 mol dm−3 ) × (12.0 cm)
= �.��
(c) With the data at ��� Torr and ��% dimers, the concentration of the monomer
is
(100 Torr)×[(1 atm)�(760 Torr)]×[(1.01325 × 105 Pa)�(1 atm)]
[M] =
(8.3145 J K−1 mol−1 ) × (373 K)
× (1 − 0.18) = 3.52... mol m−3 = 3.52... × 10−3 mol dm−3
384 11 MOLECULAR SPECTROSCOPY
A = ε[M]L
= (150 dm3 mol−1 cm−1 ) × (3.52... × 10−3 mol dm−3 ) × (12.0 cm)
= 6.34...
Hence
P��A.� �e line from the star is at longer wavelength, and hence lower frequency, than
for the Earth-bound observation, therefore the object is receding. �e Doppler
shi� is given by [��A.��a–���]
1 − (s�c)
1�2
f = ν r �ν 0 = � �
1 + (s�c)
1 − x 1�2 1− f2
f =� � hence f 2 (1 + x) = (1 − x) hence x=
1+x 1+ f2
δλ 2 2kT ln 2 1�2 δλ 2 c 2 m
= � � hence T =� �
λ0 c m λ 0 8k ln 2
With the data given
P��A.�� If each collision is e�ective at changing the energy of a state, the lifetime is
simply the inverse of the collision rate: τ = 1�z .
�e uncertainty in the energy of a state with lifetime τ is δE ≈ ħ�τ. �erefore a
spectroscopic transition involving this state has an uncertainty in its frequency,
and hence a linewidth, of the order of δν = δE�h ≈ (2πτ)−1 . Using τ = 1�z
and the given expression for z gives the linewidth as
1�2
4σ kT 1�2 p 4σ 2
δν = 1�2πτ = z�2π = � � =� 3 � p
2π πm kT π mkT
4 × (0.30 × 10−18 m2 )2
1�2
δν = � 3 �
π ×(36)×(1.6605 × 10−27 kg)×(1.3806 × 10−23 J K−1 ) × (298 K)
× (1.01325 × 105 Pa) = �.�� GHz
��� � � � �� � � � ��
apodization
where a j and ν̃ j are the intensity and wavenumber, respectively, of the jth peak
in the spectrum; i runs from 0 to N, the number of data points. �e value of
N is a matter of choice, but a sensible starting value might be ���; the reason
386 11 MOLECULAR SPECTROSCOPY
for this apparently odd choice is that some numerical implementations of the
Fourier transform require that the number of points be a power of � (256 = 28 ).
N = 256 ν̃ max = 100 cm−1 δ = 1�(2 × 100 cm−1 ) = 0.005 cm α = 2.5 cm−2
a 1 = 0.25 ν̃ 1 = 5.0 cm−1 a 2 = 1.00 ν̃ 2 = 15 cm−1 a 3 = 0.75 ν̃ 3 = 50 cm−1
2
I
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
p�cm
Figure 11.2
To �nd the spectrum it is necessary to compute the Fourier transform of the in-
terferogram. �ere are many variants of the way this transform is implemented
as a numerical procedure, and the one needed here is usually referred to as the
discrete cosine Fourier transform. As can be seen from Fig. ��.�, the interfer-
ogram is always positive and decays to zero; this will give a large peak in the
spectrum at a wavenumber of zero, in addition to the peaks corresponding to
the wavenumbers of the oscillating terms that have been introduced. Figure ��.�
shows the spectrum obtained by Fourier transformation of the interferogram;
the peak at zero wavenumber has been truncated.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 387
0 20 40 60 80 100
ν̃�cm
Figure 11.3
D��B.� Symmetric rotor: �e energy depends on J and K 2 , hence each level except the
K = 0 level is doubly degenerate. In addition, states of a given J have (2J + 1)
values of the component of their angular momentum along an external axis,
characterized by the quantum number M J . �e energy is not a�ected by M J ,
388 11 MOLECULAR SPECTROSCOPY
D��B.� A molecule has three principal moments of inertia about perpendicular axes:
these moments are labelled I a , I b , and I c , with I c ≥ I b ≥ I a . A prolate symmetric
rotor has I a ≠ I b = I c ; examples include a thin rod, any linear molecule, CH� F
and CH� CN. An oblate symmetric rotor has I a = I b ≠ I c ; examples include a
�at disc, benzene and BF� . In terms of I�� and I� , prolate rotors have I�� < I� and
oblate tops have I�� > I� .
Solutions to exercises
E��B.�(a) �e wavenumbers of the lines in the rotational spectrum are given by [��B.��a–
���], ν̃(J) = 2B̃(J + 1); the J = 3 ← 2 transition is therefore at ν̃(2) = 2B̃(2 +
1) = 6B̃. �e rotational constant is given by [��B.�–���], B̃ = ħ�4πcI, and the
moment of inertia is given by m eff R 2 , where m eff = m 1 m 2 �(m 1 + m 2 ).
E��B.�(a) �e wavenumbers of the lines in the rotational spectrum are given by [��B.��a–
���], ν̃(J) = 2B̃(J + 1). �e J = 3 ← 2 transition is therefore at ν̃(2) = 2B̃(2 +
1) = 6B̃, hence B̃ = (63.56�6) cm−1 . �e rotational constant is given by [��B.�–
���], B̃ = ħ�4πcI, and the moment of inertia is given by m eff R 2 , where m eff =
m 1 m 2 �(m 1 + m 2 ). It follows that R = (ħ�4πcm eff B̃)1�2 .
1.0546 × 10−34 J s
1�2
R=� �
4π×(2.9979 × 1010 cm s−1 )×(1.62... × 10−27 kg)×[(65.36�6)cm−1 ]
= 125.7 pm
E��B.�(a) �e wavenumbers of the lines in the rotational spectrum are given by [��B.��a–
���], ν̃(J) = 2B̃(J + 1); the lines are therefore spaced by 2B̃, it therefore follows
that B̃ = (12.604�2) cm−1 . �e rotational constant is given by [��B.�–���],
B̃ = ħ�4πcI, and the moment of inertia is given by m eff R 2 , where m eff =
m 1 m 2 �(m 1 + m 2 ). It follows that I = ħ�4πc B̃ and R = (I�m eff )1�2 .
I = ħ�4πc B̃
1.0546 × 10−34 J s
= = 4.4420 × 10−47 kg m2
4π × (2.9979 × 1010 cm s−1 ) × [(12.604�2] cm−1 )
J max =
(1.3806 × 10−23 J K−1 )×(298 K)
1�2
� � − 1
2×(6.6261 × 10−34 J s)×(2.9979 × 1010 cm s−1 )×(0.244 cm−1 ) 2
= ��
J max =
(1.3806 × 10−23 J K−1 )×(373 K)
1�2
� � − 1
2×(6.6261 × 10−34 J s)×(2.9979 × 1010 cm s−1 )×(0.244 cm−1 ) 2
= ��
E��B.�(a) For a molecule to show a pure rotational Raman spectrum it must have an
anisotropic polarizability. With the exception of spherical rotors, all molecules
satisfy this requirement. �erefore H� , HCl, CH� Cl all give rotational Raman
spectra.
1.14... × 10−45 kg m2
1�2
R = (I�m eff )1�2 = � 1 �
2
× 34.9688 × (1.6605 × 10−27 kg)
= 198.9 pm
E��B.�(a) �e ratio of the weights for (odd J)/(even J) is given by [��B.��–���]. For ��Cl,
I = 32 and the nucleus is therefore a fermion. �e ratio is (odd J)/(even J)
= (I + 1)�I = ( 32 + 1)�( 32 ) = 5
3
.
R θ�2
E��B.��(a) �e required expressions are the �rst listed under symmetric rotors in Table ��B.�
on page ���
m A (m B + m A )
I� = m A f 1 (θ)R 2 + f 2 (θ)R 2
m
mC
+ �(3m A + m B )R ′ + 6m A R[ 13 f 2 (θ)]1�2 � R′
m
I�� = 2m A f 1 (θ)R 2
Note that the molecule described by these relationships is BA� C, which be-
comes BA� by letting C=A; the question refers to a molecule AB� , but for con-
sistency with the main text the exercise will be continued with BA� . Let m C =
m A and R ′ = R to give
m A (m B + m A )
I� = m A f 1 (θ)R 2 + f 2 (θ)R 2
m
mA
+ �(3m A + m B ) + 6m A [ 13 f 2 (θ)]1�2 � R 2
m
with m = m B + 4m A . To simplify the expression somewhat let m B = αm A . �is
gives m = αm A + 4m A = (4 + α)m A
m A2 (1 + α)
I� = m A f 1 (θ)R 2 + f 2 (θ)R 2
(4 + α)m A
mA
+ �m A (3 + α) + 6m A [ 13 f 2 (θ)]1�2 � R 2
(4 + α)m A
(1 + α) 1
I� �(m A R 2 ) = f 1 (θ) + f 2 (θ) + �(3 + α) + 6[ 13 f 2 (θ)]1�2 �
(4 + α) (4 + α)
I�� �(m A R 2 ) = 2 f 1 (θ)
�e variation of the moments of inertia with θ are shown in Fig. ��.�; I� is shown
for three representative values of α. Not surprisingly, I� and I�� converge onto
the same value when θ is the tetrahedral angle (shown by the vertical dotted
line). �is is because in this limit the molecule becomes tetrahedral and is then
a spherical rotor, for which all the moments of inertia are the same.
At the tetrahedral angle cos θ tet = − 13 ; hence f 1 (θ tet ) = 4
3
and f 2 (θ tet ) = 1
3
(1 + α) 1 1
I� �(m A R 2 ) = 4
+ ×3+ �(3 + α) + 6[ 13 × 13 ]1�2 �
3
(4 + α) (4 + α)
(1 + α) 1 1
= 43 + ×3+ {(3 + α) + 2}
(4 + α) (4 + α)
4(4 + α) + (1 + α) + 3(5 + α) 32 + 8α
= = =8
3(4 + α) 3(4 + α) 3
�e moment of inertia for a tetrahedral molecule is, from the table, I�(m A R 2 ) =
8
3
, in agreement with the result just derived. In this limit the moment of inertia
does not depend on the mass of B (the central atom), as the axes pass through
this atom.
392 11 MOLECULAR SPECTROSCOPY
3.0
I��
I� α = 1
2.8
I� α = .2
I� α = 5
2.6
I�m A R 2
2.4
2.2
2.0
90.0 95.0 100.0 105.0 110.0
θ�○
Figure 11.4
E��B.��(a) To be a symmetric rotor a molecule most possess an n-fold axis with n > 2.
(i) O� is bent (like H� O), it has a two-fold axis and so is an asymmetric rotor.
(ii) CH� CH� has a three-fold axis and so is a symmetric rotor. (iii) XeO� is
tetrahedral, and so is a spherical rotor. (iv) Ferrocene has a �ve-fold axis and
so is a symmetric rotor.
E��B.��(a) In order to determine two unknowns, data from two independent experiments
are needed. In this exercise two values of B for two isotopologues of HCN are
given; these are used to �nd two moments of inertia. �e moment of inertia of
a linear triatomic is given in Table ��B.� on page ���, and if it is assumed that
the bond lengths are una�ected by isotopic substitution, the expressions for
the moment of inertia of the two isotopologues can be solved simultaneously
to obtain the two bond lengths.
�e rotational constant in wavenumber is given by [��B.�–���], B̃ = ħ�4πcI;
multiplication by the speed of light gives the rotational constant in frequency
units B = ħ�4πI, which rearranges to I = ħ�4πB
Using the expressions from Table ��B.� on page ���, the moments of inertia
are expressed in terms of the masses and bond lengths, where the former are
expressed as multiples on m u . In this case A = �H or �H, B = ��C and C = ��N.
(m H R − m N R ′ )2
I HCN = m H R 2 + m N R′2 −
mH + mC + mN
(1.0078R − 14.0031R ′ )2
= 1.0078R 2 + 14.0031R′2 −
1.0078 + 12.0000 + 14.0031
(1.0078R − 14.0031R′ )2
= 1.0078R 2 + 14.0031R′2 −
27.0109
(m D R − m N R′ )2
I DCN = mD R + mN R −
2 ′2
mD + mC + mN
(2.0141R − 14.0031R ′ )2
= 2.0141R 2 + 14.0031R′2 −
2.0141 + 12.0000 + 14.0031
(2.0141R − 14.0031R′ )2
= 2.0141R 2 + 14.0031R′2 −
28.0172
�ese two equations need to be solved simultaneously for R and R ′ , but be-
cause they are quadratics this is a very laborious process by hand: it is best
achieved using mathematical so�ware. �is gives the resulting bond lengths as
R = R CH = 0.1062 nm and R′ = R CN = 0.1157 nm .
1 × 127 2 + 127 2
D̃ 2HI �D̃ 1HI = (m eff 1HI �m eff 2HI ) = � × � = �.��
2
1 + 127 2 × 127
Solutions to problems
P��B.� Suppose that the bond length is R and that the centre of mass is at a distance x
from mass m 1 and therefore (R − x) from mass m 2 . Balancing moments gives
m 1 x = m 2 (R − x), hence x = m 2 R�(m 1 + m 2 ). Using this result it follows that
(R − x) = R − m 2 R�(m 1 + m 2 ) = m 1 R�(m 1 + m 2 ). �e moment of inertia is
394 11 MOLECULAR SPECTROSCOPY
therefore
m 1 m 22 R 2 m 2 m 12 R 2
I = m 1 x 2 + m 2 (R − x)2 = +
(m 1 + m 2 )2 (m 1 + m 2 )2
m 1 m 2 (m 2 + m 1 )R 2
m1 m2 R2
= = = m eff R 2
(m 1 + m 2 )2 (m 1 + m 2 )
P��B.� �e rotational terms for a symmetric rotor are given by [��B.��a–���], F̃(J, K) =
B̃J(J+1)+(Ã− B̃)K 2 . �e selection rules are ∆J = ±1 and ∆K = 0, and therefore
the term in K does not a�ect the wavenumber of the lines in the spectrum; the
result is that the lines are at exactly the same wavenumbers as for a linear rotor,
[��B.��a–���], ν̃(J) = 2B̃(J + 1). �e separation of the lines is 2B̃.
In frequency units the spacing is 2B = 2 × (298 GHz) = 596 GHz . Expressed
as a wavenumber this spacing is (596 × 109 Hz)�(2.9979 × 1010 cm s−1 ) =
��.� cm−1 .
�e rotational constant is given by [��B.�–���], B̃ = ħ�4πcI. Expressed in
frequency units this is B = ħ�4πI. It follows that I = ħ�4πB
1.0546 × 10−34 J s
I = ħ�4πB = = 2.82 × 10−47 kg m2
4π × (298 × 109 Hz)
Expressions for the moment of inertia are given in Table ��B.� on page ���;
NH� is a symmetric rotor and the second entry under symmetric rotors is the
required one. �e moment of inertia corresponding to the rotational constant
B is I� . With the data given
mH mN
I� = m H (1 − cos θ)R 2 + (1 + 2 cos θ)R 2
m N + 3m H
�ese two quantities di�er by less than �.�% so the hypothesis that the bond
length is invariant to isotopic substitution is con�rmed to quite a high level of
precision; with the accuracy of the data given there is, however, some percep-
tible change.
34
P��B.� Note: there is an error in the problem; for the S isotopologue the line at
��.��� �� GHz is for J = 3.
�e wavenumbers of the lines expected for a linear rotor are given by [��B.��a–
���], ν̃(J) = 2B̃(J + 1); the separation of the lines is 2B̃. For OC32 S the average
spacing of the lines is ��.����� GHz, so the best estimate for the rotational
constant is B OCS = 12 × (12.16272 GHz) = 6.08136 GHz. For OC34 S there
are just two lines, one for J = 1 and one for J = 3; these are separated by
��.����� GHz, which is 4B. �e best estimate for the rotational constant is
B OCS′ = 14 × (23.73007 GHz) = 5.93252 GHz.
�e rotational constant in wavenumber is given by [��B.�–���], B̃ = ħ�4πcI;
multiplication by the speed of light gives the rotational constant in frequency
units B = ħ�4πI, hence I = ħ�4πB
where for short S implies ��S and S′ implies ��S. It is somewhat more convenient
for the subsequent manipulations to express the moments of inertia in units of
the atomic mass constant m u and nm.
2
109 nm 1 mu
I OCS = (1.37... × 10−45 kg m2 ) × � � ×
1m 1.6605 × 10−27 kg
= 0.831... m u nm2
2
109 nm 1 mu
I OCS′ = (1.41... × 10−45 kg m2 ) × � � ×
1m 1.6605 × 10−27 kg
= 0.851... m u nm2
396 11 MOLECULAR SPECTROSCOPY
Using the expressions from Table ��B.� on page ���, the moments of inertia
are expressed in terms of the masses and bond lengths, where the former are
expressed as multiples on m u . In this case A = ��O, B = ��C and C = ��S or ��S.
(m O R − m S R′ )2
I OCS = m O R 2 + m S R′2 −
mO + mC + mS
(15.9949R − 31.9721R′ )2
= 15.9949R 2 + 31.9721R ′2 −
15.9949 + 12.0000 + 31.9721
(15.9949R − 31.9721R ′ )2
= 15.9949R 2 + 31.9721R ′2 −
59.967
(m R − m ′R )
′ 2
= m O R 2 + m S′ R ′2 −
O S
I OCS′
m O + m C + m S′
(15.9949R − 33.9679R′ )2
= 15.9949R 2 + 33.9679R ′2 −
15.9949 + 12.0000 + 33.9679
(15.9949R − 33.9679R ′ )2
= 15.9949R 2 + 33.9679R ′2 −
61.9628
�ese two equations need to be solved simultaneously for R and R ′ , but be-
cause they are quadratics this is a very laborious process by hand: it is best
achieved using mathematical so�ware. �is gives the resulting bond lengths as
R = R OC = 0.1167 nm and R′ = R CS = 0.1565 nm .
P��B.� �e wavenumbers of the lines expected for a linear rotor are given by [��B.��a–
���], ν̃(J) = 2B̃(J +1); the separation of the lines is 2B̃. However, the separation
between adjacent lines in the given data is not constant, but increases along the
series. To account for this, the e�ects of centrifugal distortion are included,
and in this case the frequencies of the lines are given by [��B.��b–���], ν(J) =
2B(J+1)−4D J (J+1)3 (written with the constants in frequency units). Division
of both side of this expression by 2(J +1) indicates that a plot of [ν(J)]�2(J +1)
against (J + 1)2 should be a straight line with slope −2D J and intercept B. �e
data are tabulated below; δ is the di�erence between successive lines. �e plot
is shown in Fig. ��.�.
4 296.5
[ν(J)�2(J + 1)]�MHz
4 296.0
4 295.5
600 650 700 750 800 850 900
(J + 1) 2
Figure 11.5
about 0.03 MHz in the intercept, so the best estimate for the rotational constant
is B = 4293.28 ± 0.03 MHz or, expressed as a wavenumber, B̃ = 0.1432 cm−1 .
It is somewhat unusual that the centrifugal distortion constant appears to be
negative.
�e most occupied J state is given by [��B.��–���], J max = (kT�2hc B̃)1�2 − 12 .
At ��� K
J max =
(1.3806 × 10−23 J K−1 )×(298 K)
1�2
� � − 1
2×(6.6261 × 10 J s)×(2.9979 × 1010 cm s−1 )×(0.1432 cm−1 )
−34 2
= ��
d
A(2J + 1)e−hc B̃ J(J+1)�k T
dJ
= A×2×e−hc B̃ J(J+1)�k T − A(2J + 1)×(2J + 1)×(hc B̃�kT)e−hc B̃ J(J+1)�k T
= ��
For a spherical rotor the degeneracy of each level is (2J + 1)2 . Finding the most
populated level proceeds as before
d
A(2J + 1)2 e−hc B̃ J(J+1)�k T
dJ
= A × 4(2J + 1) × e−hc B̃ J(J+1)�k T
− A(2J + 1)2 × (2J + 1) × (hc B̃�kT)e−hc B̃ J(J+1)�k T
As before the maximum occurs when the term in square brackets is zero
= �
P��B.�� Temperature e�ects. At extremely low temperatures (�� K) only the lowest rota-
tional states are populated. No emission spectrum is expected for the CO in the
cloud and star-light microwave absorptions by the CO in the cloud are from the
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 399
Density e�ects. �e density of an interstellar cloud may range from one particle
to a billion particles per cm3 . �is is still very much a vacuum compared
to the laboratory high vacuum of a trillion particles per cm3 . Under such
extreme vacuum conditions the half-life of any quantum state is expected to
be extremely long and absorption lines should be very narrow. At the higher
densities the vast size of nebulae obscures distant stars. High densities and high
temperatures may create conditions in which emissions stimulate emissions of
the same wavelength by molecules. A cascade of stimulated emissions greatly
ampli�es normally weak lines – the maser phenomena of microwave ampli�-
cation by stimulated emission of radiation.
Particle velocity e�ects. Particle velocity can cause Doppler broadening of spec-
tral lines. �e e�ect is extremely small for interstellar clouds at �� K but is ap-
preciable for clouds near high temperature stars. Out�ows of gas from pulsing
stars exhibit a red Doppler shi� when moving away at high speed and a blue
shi� when moving toward us.
Solutions to exercises
E��C.�(a) �e wavenumber of the fundamental vibrational transition is simply equal to
the vibrational frequency expressed as a wavenumber. �is is given by [��C.�b–
���], ν̃ = (1�2πc)(k f �m eff )1�2 , where m eff is the e�ective mass, given by m eff =
m 1 m 2 �(m 1 + m 2 ). It follows that k f = m eff (2πc ν̃)2 . With the data given the
following table is drawn up.
�
H ��F �
H ��Cl �
H ��Br �
H ���I
ν̃�cm−1 4141.3 2988.9 2649.7 2309.5
m eff �m u 0.9570 0.9796 0.9954 0.9999
k f �N m −1
967.0 515.6 411.7 314.2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 401
As expected, the relative population of the upper level increases with tempera-
ture.
E��C.�(a) Taking y e = 0 is equivalent to using the terms for the Morse oscillator, which
are given in [��C.�–���], G̃(υ) = (υ + 12 )ν̃ − (υ + 12 )2 ν̃x e . �e transition υ ← 0
has wavenumber
Data on three transitions are provided, but only two are needed to obtain values
for ν̃ and x e . �e ∆G̃(υ) values for the �rst two transitions are
�is value for ν̃ is used in the �rst equation, which is then solved for x e to give
x e = 12 − (1556.22 cm−1 )�[2 × (1580.4 cm−1 )] = 7.65 × 10−3 .
E��C.�(a) Following the discussion in Section ��C.�(b) on page ���, D̃ 0 is given by the area
under a plot of ∆G̃ υ+1�2 against (υ + 12 ), where ∆G̃ υ+1�2 = G̃(υ + 1) − G̃(υ).
�e data are shown in the table and the plot in Fig. ��.�.
402 11 MOLECULAR SPECTROSCOPY
3 000
∆G̃ υ+1�2 �cm−1
2 000
1 000
0
0 5 10 15 20 25 30
υ+ 1
2
Figure 11.6
�e area under the line is simply the area of a triangle, 12 × base × height, which
in this case is 12 × (28.22) × (2 937.7) = 4.14 × 104 . �e dissociation energy
is therefore D̃ 0 = 4.14 × 104 cm−1 ; only modest precision is quoted because a
long extrapolation is made on the basis of few data points.
�e fact that the data fall on a good straight line indicates that the Morse levels
apply, in which case, according to [��C.�b–���], ∆G̃ υ+1�2 = ν̃ − 2(υ + 1)x e ν̃.
�is expression is rewritten
which implies that a plot of ∆G̃ υ+1�2 against (υ + 12 ) will have slope −2x e ν̃ and
intercept (ν̃ − x e ν̃). Hence, using the slope of the plot already made
ν̃ 2
D̃ 0 = D̃ e − G̃(0) = − G̃(0)
4ν̃x e
(2989.8 cm−1 )2
= − (1481.86 cm−1 ) = 4.14 × 104 cm−1
4 × (52.06 cm−1 )
To within the precision quoted, both methods give the same result.
To convert to eV, the conversion 1 eV = 8065.5 cm−1 from inside the front cover
is used to give D 0 = 5.14 eV .
E��C.�(a) �e wavenumber of the transition arising from the rotational state J in the R
branch (∆J = +1) of the fundamental transition (υ = 1 ← υ = 0) is given
by [��C.��c–���], ν̃ R (J) = ν̃ + 2B̃(J + 1). In this case ν̃ = 2308.09 cm−1 and
B̃ = 6.511 cm−1 hence
ν̃ Na 35Cl − ν̃ Na 37Cl
1�2
ν̃ 37 m
= 1 − Na Cl = 1 − � eff Na Cl �
35
Solutions to problems
P��C.� (a) Figure ��.� shows plot of the total electronic energy (with respect to the
free atoms) as a function of the bond length for each of the hydrogen
halides. Calculations are performed with Spartan �� using the MP� method
with the �-���++G** basis set.
10 HI
HBr
8
6
HCl
4
2
V / eV
0 HF
–2
–4
–6
–8
50 100 150 200 250 300 350
Figure 11.7 R / pm
�e plot clearly shows that in going down the halogen group from HF to
HI the equilibrium bond length increases and the depth of the potential
well decreases. �e equilibrium properties of each molecule are sum-
marized in the following table. �e force constants are computed in the
harmonic approximation using [��C.�b–���], ν̃ = (1�2πc)(k f �m eff )1�2 ,
with m eff = m 1 m 2 �(m 1 + m 2 ). It follows that k f = m eff (2πc ν̃)2 . �e
calculated bond lengths are in good agreement with the experimental
values, but the vibrational frequencies do not agree very well at all.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 405
P��C.� Figure ��.� shows a plot of V (x)�V0 as a function of x�a; the minimum at x = 0
is clearly rather ‘�at’.
0.0
−0.5
V (x)�V0
−1.0
−3 −2 −1 0 1 2 3
x�a
Figure 11.8
d 2a 2 2 2
V0 (e−a �x − 1) = V0 3 e−a �x
2 2
dx x
d2 d 2a 2 −a 2 �x 2 2V0 a 2 3 2a 2 −a 2 �x 2
V (x) = V 0 e = �− + 3 �e
dx 2 dx x3 x3 x x
P��C.� (a) �e dissociation energy is D̃ 0 = D̃ e − G̃(0) (Fig. ��C.� on page ���), where
G̃(0) is the energy of the lowest vibrational term. For the Morse energy
levels given by [��C.�–���], G̃(υ) = (υ + 12 )ν̃ − (υ + 12 )2 x e ν̃, it follows
that G̃(0) = 12 ν̃ − 14 x e ν̃. �e conversion between cm−1 and eV is achieved
using 1 eV = 8065.5 cm−1 from inside the front cover. For �H ��Cl
hc D̃ 0 = hc D̃ e − G̃(0) = hc D̃ e − ( 12 ν̃ − 14 x e ν̃)
= (5.33 eV) − ( 12 × 2989.7 − 14 × 52.05) × [(1 eV)�(8065.5 cm−1 )]
= �.�� eV
(b) �e task is to calculate the values of ν̃ and x e ν̃ for the isotopologue �H ��Cl.
�e potential energy curve, and hence the value of the depth of the well
D̃ e , is the same for the two isotopologues.
In the harmonic limit the vibrational frequency is given by [��C.�b–���],
ν̃ = (1�2πc)(k f �m eff )1�2 , with m eff = m 1 m 2 �(m 1 + m 2 ). Assuming that
−1�2
the force constants of the two isotopologues are the same, ν̃ ∝ m eff .
From [��C.�–���] it is seen that x e = ν̃�4 D̃ e which rearranges to D̃ e =
−1�2 −1�2
ν̃�4x e . Because ν̃ ∝ m eff it follows that x e ∝ m eff also in order for D̃ e
to be una�ected by isotopic substitution. �us x e ν̃ ∝ m−1 eff .
1�2 1�2
ν̃ 2H 35Cl m 1 m eff , 1H 35Cl
= � eff , HX � hence ν̃ 2H 35Cl = ν̃ 1H 35Cl × � �
ν̃ 1H 35Cl m eff , 2H 35Cl m eff , 2H 35Cl
1.0078 × 34.9688 2.0140 + 34.9688 1�2
ν̃ 2H 35Cl = (2989.7 cm−1 ) � × �
1.0078 + 34.9688 2.02140 × 34.9688
= 2144.25 cm−1
Similarly
m eff , 1H 35Cl
x e ν̃ 2H 35Cl = x e ν̃ 1H 35Cl × � �
m eff , 2H 35Cl
1.0078 × 34.9688 2.0140 + 34.9688
= (52.05 cm−1 ) � × �
1.0078 + 34.9688 2.02140 × 34.9688
= 26.77 cm−1
hc D̃ 0 = hc D̃ e − G̃(0) = hc D̃ e − ( 12 ν̃ − 14 x e ν̃)
= (5.33 eV) − ( 12 × 2144.25 − 14 × 26.77) × [(1 eV)�(8065.5 cm−1 )]
= �.�� eV
P��C.� (a) �e dissociation energy D̃ 0 and the well depth D̃ e are related by D̃ e =
D̃ 0 + G̃(0) (Fig. ��C.� on page ���), where G̃(0) is the vibrational term of
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 407
kf = 1
2
× (4.0026) × (1.6605 × 10−27 kg)
× [2π × (2.9979 × 1010 cm s−1 ) × (1.5 cm−1 )]2
= 2.7 × 10−4 N m−1
I= 1
2
× (4.0026) × (1.6605 × 10−27 kg) × (297 × 10−12 m)2
= 2.93... × 10−46 kg m2 = 2.93 × 10−46 kg m2
hence
ħ
B̃ =
4πcI
(1.0546 × 10−34 J s)
= = �.��� cm−1
4π × (2.9979 × 1010 cm s−1 ) × (2.93... × 10−46 kg m2 )
D̃ e = D̃ 0 + 12 ν̃ − 14 x e ν̃ = D̃ 0 + 12 ν̃ − ν̃ 2 �16 D̃ e
hence
2 150
∆G̃ υ+1�2 �cm−1
2 100
2 050
1 2 3 4 5
υ+1
Figure 11.9
P��C.�� �e data provided allow the calculation of two independent moments of inertia.
If it is assumed that the bond lengths are una�ected by isotopic substitution,
then it is possible to set up two equations and solve them simultaneously for
the CC and CH bond lengths.
Expressions for the wavenumbers of the lines in the P and R branches are given
by [��C.��a–���] and [��C.��c–���]; from these it follows that the spacing be-
tween the lines is 2B̃. �e rotational constant is given by [��B.�–���], B̃ =
ħ�4πcI; it follows that I = ħ�4πc B̃.
I H = ħ�4πc B̃ H
1.0546 × 10−34 J s
= = 2.38... × 10−46 kg m2
4π × (2.9979 × 1010 cm s−1 ) × [(2.352�2) cm−1 ]
I D = ħ�4πc B̃ D
1.0546 × 10−34 J s
= = 3.30... × 10−46 kg m2
4π × (2.9979 × 1010 cm s−1 ) × [(1.696�2) cm−1 ]
I H = 2m C (r CC �2)2 + 2m H (r CH + r CC �2)2
I D = 2m C (r CC �2)2 + 2m D (r CH + r CC �2)2
�ese are the two equations which need to be solved simultaneously. Finding
the solution is much simpli�ed by letting p = (r CC �2)2 and q = (r CH + r CC �2)2
to give
I H = 2m C p + 2m H q I D = 2m C p + 2m D q
It follows that
IH − ID
q=
2(m H − m D )
(2.38... × 10−46 kg m2 ) − (3.30... × 10−46 kg m2 )
= = 2.75... × 10−20 m2
2(1.0078 − 2.0140) × (1.6605 × 10−27 kg)
mD IH − mH ID
p=
2m C (m D − m H )
(2.0140) × (2.38... × 10−46 kg m2 ) − (1.0078) × (3.30... × 10−46 kg m2 )
=
2 × (12.0000) × (2.0140 − 1.0078) × (1.6605 × 10−27 kg)
= 3.65... × 10−20 m2
On the other hand �x 2 � is de�nitely non-zero, as was seen for the case of the
harmonic oscillator in Topic �E.
It follows straightforwardly that 1��R�2 = 1�R e2 .
�R 2 � is found in the following way
� ��� � � �� � � �
A B
= �R e2 � + �2xR e � +�x 2 �
Term A is simply the average of a constant term, which is equal to the term itself,
in this case R e2 . Term B is rewritten 2R e �x� by taking constant terms outside
the averaging; this term is zero because �x� = 0. �erefore �R 2 � = R e2 + �x 2 �.
Using this 1��R 2 � is found in the following way
1 1 1 1
= 2 = 2×
�R � R e + �x � R e 1 + �x 2 ��R e2
2 2
1 �x 2 �
≈ �1 − �
R e2 R e2
where to go to the last line the expansion (1+ y)−1 ≈ 1− y is used. �e resulting
expression includes the lowest power of �x 2 ��R e2 , as required
�1�R 2 � is found in the following way
1 1 1 1
� �=� �= 2 � �
R2 (R e + x)2 R e (1 + x�R e )2
1 1
≈ 2 �1 − 2x�R e + 3x 2 �R e2 � = 2 ��1� − (2�R e )�x� + (3�R e2 )�x 2 ��
Re Re
1 3�x 2 �
= �1 + �
R e2 R e2
ν̃ 2
D̃ 0 = − 1 ν̃ + 14 ν̃x e
4ν̃x e 2
(610.258 cm−1 )2 1
= − (610.258 cm−1 ) + 14 (3.141 cm−1 )
4 × (3.141 cm−1 ) 2
= 2.93 × 104 cm−1
To convert to eV, the conversion 1 eV = 8065.5 cm−1 from inside the front cover
is used to give D 0 = 3.64 eV .
P��C.�� �e features centred about ����.�� cm−1 are the P and R branches. From [��C.��a–
���] and [��C.��c–���] the �rst line in the R branch occurs at ν̃ + 2B̃, and the
�rst line in the P branch is ν̃ − 2B̃. �e separation of these two, �.��� cm−1 , is
therefore 4B̃.
(d) As noted at the start of the answer, 4B̃ = 7.655 cm−1 , hence B̃ = 1.914 cm−1 .
(e) �e rotational constant is given by [��B.�–���], B̃ = ħ�4πcI, and the mo-
ment of inertia is given by m eff R 2 . It follows that R = (ħ�4πcm eff B̃)1�2 .
12.0000 × 15.9949
m eff = × (1.6605 × 10−27 kg) = 1.13... × 10−26 kg
12.0000 + 15.9949
R = (ħ�4πcm eff B̃)1�2
1.0546 × 10−34 J s
1�2
=� �
4π(2.9979 × 1010 cm s−1 )×(1.13... × 10−26 kg)×(1.914 cm−1 )
= 113.3 pm
Although the data are given to quite high precision the assumption that the
harmonic oscillator/rigid rotor models apply means that the derived values of
the bond length and so on are likely to have systematic errors which are higher
than the apparent precision of the data.
P��C.�� �e method of combination di�erences, described in Section ��C.�(b) on page
���, involves taking the di�erence between two transitions which share a com-
mon lower rotational level or a common upper rotational level. In the case of
O and S branches, which correspond to ∆J = −2 and ∆J = +2, respectively,
the two transitions which share a common lower level are ν̃ O (J) and ν̃ S (J):
these are the transitions from J to J − 2, and from J to J + 2. As is evident from
Fig. ��.��, the di�erence in wavenumber between these two transitions is the
interval indicated by the dashed arrow which is simply G̃(J + 2) − G̃(J − 2) for
the upper vibrational state (assumed to be υ = 1)
G̃(J + 2) − G̃(J − 2) = B̃ 1 (J + 2)(J + 3) − B̃ 1 (J − 2)(J − 1) = B̃ 1 (8J + 4)
hence ν̃ S (J) − ν̃ O (J) = 8B̃ 1 (J + 12 )
J+2
J υ=1
J−2
ν̃ O (J +2)
ν̃ S (J −2)
ν̃ O (J)
ν̃ S (J)
J+2
J υ=0
J−2
Figure 11.10
Solutions to exercises
E��D.�(a) According to [��D.�–���], a linear molecule has 3N − 5 normal modes, where
N is the number of atoms in the molecule. �ere are �� atoms in this linear
molecule, and so there are 3(44) − 5 = ��� normal modes.
E��D.�(a) According to [��D.�–���], a non-linear molecule has 3N −6 vibrational normal
modes, where N is the number of atoms in the molecule; therefore H� O has
� normal modes. �e terms (energies expressed as wavenumbers) for normal
mode q are given by [��D.�–���], G̃ q (υ) = (υ q + 12 )ν̃ q , where υ q is the quantum
number for that mode and ν̃ q is the wavenumber of the vibration of that mode.
�ese terms are additive, so the ground state term corresponds to each mode
having υ q = 0
E��D.�(a) A mode is infrared active if the vibration results in a dipole which changes
as the molecule oscillates back and forth about the equilibrium geometry. A
414 11 MOLECULAR SPECTROSCOPY
E��D.�(a) �e benzene molecule has a centre of symmetry, so the rule of mutual exclusion
applies. �e molecule has no permanent dipole moment and if the ring expands
uniformly this situation does not change: such a vibration does not lead to a
changing dipole and so the mode is infrared inactive .
�is kind of ‘breathing’ vibration does lead to a change in the polarizability, so
the mode is Raman active .
E��D.�(a) With the exception of homonuclear diatomics, all molecules have at least one
infrared active normal mode. Of the molecules listed, HCl, CO� , and H� O
have infrared active modes.
Solutions to problems
P��D.� Figure ��.�� shows a plot of V (h)�V0 as a function of hb 1�4 .
For a given potential the force constant is de�ned in [��C.�b–���] in terms of
the second derivative of the potential with respect to the displacement from
equilibrium; in this case h is the varaible which describes the displacement.
�us k f = (d2 V �dh 2 )0 .
d
V0 (1 − e−bh ) = V0 4bh 3 e−bh
4 4
dh
d2 d
V (h) = V0 4bh 3 e−bh = 4V0 b �3h 2 − 4bh 6 � e−bh
4 4
dh 2 dh
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 415
1.0
0.8
0.6
V (h)�V0
0.4
0.2
0.0
−2 −1 0 1 2
hb 1�4
Figure 11.11
�e second derivative, and hence the force constant, goes to zero at h = 0. �us,
for small displacements there is no restoring force.
�e potential is con�ning so it is expected that there will be quantized vibra-
tional energy levels. By loose analogy with the harmonic case the ground state
wavefunction is expected to have a maximum at h = 0 and then decay away to
zero as h → ±∞.
P��D.� (a) Calculations on SO� are performed with Spartan �� using the MP� method
with the �-���++G** basis set. �e calculated equilibrium structure is
shown in Fig. ��.�� where it is seen that the two S–O bonds have equal
length, as expected. �e calculated bond length and angle agrees quite
well with the experimental values of ���.�� pm and ���.��○ .
(b) �e calculated values of the fundamental vibrational wavenumbers, and
illustrations of the displacements involved in the normal modes, are also
shown in Fig. ��.��. �ey correlate well with experimental values but
are about ��–�� cm−1 lower. SCF calculations o�en yield systematically
lower or higher values than experiment while approximately paralleling
the experimental to within an additive constant.
C2v, 119.26°
146.9 pm 146.9 pm
Figure 11.12
416 11 MOLECULAR SPECTROSCOPY
Solutions to exercises
E��E.�(a) A mode is infrared active if it has the same symmetry species as one of the
functions x, y, and z; in this point group these span B1 + B2 + A1 . A mode is
Raman active if it has the same symmetry as a quadratic form; in this group
such forms span A1 + A2 + B1 + B2 . �erefore all of the normal modes are both
Raman and infrared active.
E��E.�(a) (i) H� O belongs to the point group C 2v . Rather than considering all � dis-
placement vectors together it is convenient to consider them in sub-sets
of displacement vectors which are mapped onto one another by the op-
erations of the group. �e x, y, and z vectors on the oxygen are not
mapped onto the displacements of the H atoms and so can be considered
separately. In fact, because these displacement vectors are attached to the
principal axis, they transform as the cartesian functions x, y, and z as
listed in the character table: that is as B1 + B2 + A1 .
Assuming the same axis system as in Fig. ��E.� on page ���, the two x
displacements on the H atoms map onto one another, as do the two y
displacements, as do the two z displacements: however, the x, y, and z
displacements are not mixed with one another. For the two z displace-
ments the operation E leaves both una�ected so the character is �; the C 2
operation swaps the two displacements so the character is �; the σv (xz)
operation swaps the two displacements so the character is �; the σv′ (yz)
operation leaves the two displacements una�ected so the character is �.
�e representation is therefore (2, 0, 0, 2), which is easily reduced by in-
spection to A1 + B2 . For the two y displacements the argument is essen-
tially the same, resulting in the representation (2, 0, 0, 2), which reduces
to A1 + B2 .
For the two x displacements the operation E leaves both una�ected so
the character is �; the C 2 operation swaps the two displacements so the
character is �; the σv (xz) operation swaps the two displacements so the
character is �; the σv′ (yz) operation leaves the two displacements in the
same position by changes their direction, so the character is −2. �e rep-
resentation is therefore (2, 0, 0, −2), which is easily reduced by inspection
to A2 + B1 .
�e � displacements therefore transform as 3A1 + A2 + 2B1 + 3B2 .
�e displacements include translations and rotations. For the point group
C 2v , x, y, and z transform as B1 , B2 , and A1 , respectively. �e rotations
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 417
E��E.�(a) �e displacements include translations and rotations. For the point group C 2v ,
x, y, and z transform as B1 , B2 , and A1 , respectively. �e rotations R x , R y , and
R z transform as B2 , B1 , and A2 , respectively. Taking these symmetry species
away leaves just the normal modes as 4A1 + A2 + 2B1 + 2B2 . �ese correspond
to � normal modes, which is the number expected for CH� Cl� .
Solutions to problems
P��E.� (a) CH� Cl has a C 3 axis (the principal axis) along the C–Cl bond, and three
σv planes, one passing along each C–H bond and containing the principal
axis. �e point group is therefore C 3v .
(b) �e molecule is non-linear and has N = 5, there are thus 3 × 5 − 6 = �
normal modes.
(c) �e task is to �nd the symmetry species spanned by the set of (x, y, z)
displacement vectors on each atom. �e (x, y, z) displacement vectors on
the Cl can be considered separately and, as these vectors are connected to
the principal axis, their symmetry species is simply read from the char-
acter table as E + A1 . �e (x, y, z) displacement vectors on the C behave
in the same way and so transform as E + A1 .
Consider the set of � (x, y, z) displacement vectors on the H atoms. Un-
der the operation E these are all una�ected so the character is �; under
the C 3 operation they are all moved to new positions so the character is �.
Next consider one of the σv planes: the (x, y, z) vectors on the H atoms
which do not lie in this plane are all moved and so contribute � to the
character. Now consider the H atom which lies in the plane, and arrange
a local axis system such that z points along the C–H bond, y lies in the
plane and x lies perpendicular to the plane. �e e�ect of σv on (x, y, z)
is to transform it to (−x, +y, +z); two of the basis functions remain the
same and one changes sign, so the character is −1 + 1 + 1 = +1. �e
representation formed from the � (x, y, z) displacement vectors on the
H atoms is thus (9, 0, 1). �is is reduced using the reduction formula,
[��C.�a–���], to give 2A1 + A2 + 3E.
418 11 MOLECULAR SPECTROSCOPY
where µ̂ x is the operator for the dipole moment along x, and the normalized
wavefunctions ψ n = (2�L)1�2 sin(nπx�L) are used. �e dipole moment oper-
ator is proportional to x, so setting aside the constants of proportion and using
(n + 1) ≈ n for large n, the required integral is
L
I = (2�L) � x sin2 (nπx�L) dx
0
�e integral is of the form of Integral T.�� with k = π�d and a = nd; it evaluates
to n 2 d 2 �4. Taking into account the normalization factor gives the �nal result
nd�2. �e conclusion is therefore that the transition dipole moment increases
with n, the number of conjugated double bonds.
In summary, the expectation is that increasing the number of conjugated dou-
ble bonds will increase the wavelength of the absorption and also the intensity
of the absorption.
D��F.� �e wavenumbers of the lines in the P, Q and R branches are given in [��F.�–
���]. Consider �rst the P branch ν̃ P (J) = ν̃ − (B̃′ + B̃)J + (B̃′ − B̃)J 2 , and recall
that (B̃′ + B̃) � �(B̃′ − B̃)�. As J increases the lines move to lower wavenumber
on account of the term −(B̃′ + B̃)J. However, as J becomes larger still the term in
J 2 becomes proportionately more and more important. If (B̃′ − B̃) > 0 this term
contributes to an increase in the wavenumber of the lines, and for su�ciently
large J it will overcome the −(B̃′ + B̃)J term and cause the lines to start to move
to higher wavenumber as J increases further. �ere will therefore be a lowest
wavenumber at which any line appears: this is the band head. If (B̃′ − B̃) < 0
the term in J 2 simply causes the lines to move to lower wavenumber and no
band head is formed.
Next consider the R branch ν̃ R (J) = ν̃ + (B̃′ + B̃)(J + 1) + (B̃′ − B̃)(J + 1)2 .
As J increases the lines move to higher wavenumber on account of the term
+(B̃′ + B̃)(J +1). However, as J becomes larger still the term in (J +1)2 becomes
proportionately more and more important. If (B̃′ − B̃) < 0 this term contributes
to an decrease in the wavenumber of the lines, and for su�ciently large J it
will overcome the +(B̃′ + B̃)(J + 1) term and cause the lines to start to move
to lower wavenumber as J increases further. �ere will therefore be a highest
wavenumber at which any line appears: this is the band head. If (B̃′ − B̃) > 0
the term in (J + 1)2 simply causes the lines to move to higher wavenumber and
no band head is formed.
�e lines in the Q branch, ν̃ Q (J) = ν̃ + (B̃′ − B̃)J(J + 1), all appear at higher or
lower wavenumber than ν̃ depending on the sign of (B̃′ − B̃); no band heads
are formed.
If a parameter α is de�ned as α = B̃′ �B̃, so that B̃′ = α B̃, the wavenumbers of
the lines in the P and R branches can be written
�ese functions are plotted in Fig. ��.�� for representative values of α. If α < 1,
meaning that (B̃′ − B̃) < 0, the band head occurs in the R branch, but if α > 1,
meaning that (B̃′ − B̃) > 0, the band head occurs in the P branch.
Solutions to exercises
�e Gaussian functions are written e−α x �2 and e−α(x−a�2) �2 , where the param-
2 2
E��F.�(a)
eter α determines the width. To evaluate the normalizing factor for the function
420 11 MOLECULAR SPECTROSCOPY
20
P branch, α = 1.2
P branch, α = 0.8
10 R branch, α = 1.2
[ν̃(J)P or R − ν̃]�B̃
R branch, α = 0.8
0
−10
−20
0 2 4 6 8 10 12
J or (J + 1)
Figure 11.13
e−α x �2
2
requires the integral
+∞
� e−α x dx
2
−∞
�e �nal equality above is veri�ed by expanding out the square and recombin-
ing the terms. Taking out the constant factor and then writing x as (x − a�4) +
a�4 gives
+∞
I = (α�π)1�2 e−α a �16
� �(x − a�4)e−α(x−a�4) + (a�4)e−α(x−a�4) � dx
2 2 2
−∞
= e−α(w�2) �2
α = (8 ln 2)�w 2
2
1
2
hence
(a�4)e−α a �16
= (a�4)e−(8 ln 2)a �16a 2
= (a�4)e−(ln 2)�2
2 2
= (a�4)e−(ln 2 )
= (a�4)(1�21�2 ) = a�(4 × 21�2 )
1�2
E��F.�(a) A simple model for the wavelength of the HOMO–LUMO transition in a poly-
ene is discussed in Example �D.� on page ���. In this model the energy levels
of the π electrons in a polyene are modelled by those of a particle in a one-
dimensional box of length L. If the polyene consists of n conjugated double
bonds, the length L may be written as L = nd, where d is the length of a single
conjugated bond. A molecule with n conjugated double bonds will have 2n π
electrons which will occupy the energy levels pairwise. �erefore the HOMO
is the level with quantum number n and the LUMO has quantum number n+1.
�e energy of the HOMO–LUMO transition is therefore
h2 (2n + 1)h 2
E n+1 − E n = [(n + 1)2 − n 2 ] =
8mL 2 8mL 2
For large n, E n+1 − E n goes as n. However, L = nd, therefore overall E n+1 −
E n goes as 1�n, and hence the wavelength of the transition goes as n. �us,
increasing the number of conjugated double bonds will increase the wavelength
of the absorption; that is, shi� it to the red. �e transition at ��� nm is therefore
likely to be from �, and that at ��� nm is likely to be from �.
−∞
−∞ −∞
the �nal equality above is veri�ed by expanding out the square and recombining
the terms. Taking out the constant factors gives
+∞
I = (a�π)1�2 e−ax 0 �4 � e−a(x−x 0 �2) dx
2 2
−∞
as before, the integral is form of Integral G.� with k = a and evaluates to (π�a)1�2
a maximum of � when x 0 = 0, that is when the two functions are aligned, and
falls o� towards zero as x 0 increases.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 423
L π 2πx π 2πx π
= (1�L) � cos � � − �cos � � cos � � + sin � � sin � �� dx
L�4 4 L 4 L 4
Recognising that cos π�4 = sin π�4 = (2)−1�2 allows this factor to be taken
L 2πx 2πx
= (1�L)2−1�2 � 1 − cos � � − sin � � dx
L�4 L L
L 2πx L 2πx L
= (1�L)2−1�2 �x − sin � �+ cos � ��
2π L 2π L L�4
L 2πL 2π(L�4)
= (1�L)2−1�2 �L − L�4 + � − sin � � + sin � �
2π L L
2πL 2πL�4
+ cos � � − cos � � ��
L L
L
= (1�L)2−1�2 �3L�4 + � − sin(2π) + sin(π�2) + cos(2π) − cos(π�2)��
2π
3 + 4�π
= (1�L)2−1�2 (3L�4 + L�π) = 2−1�2 (3�4 + 1�π) =
4 × 21�2
E��F.�(a) �e wavenumbers of the lines in the P branch are given in [��F.�–���], ν̃ P (J) =
ν̃ − (B̃′ + B̃)J + (B̃′ − B̃)J 2 . �e band head is located by �nding the value
of J which gives the smallest wavenumber, which can be inferred by solving
dν̃ P (J)�dJ = 0.
d
�ν̃ − (B̃′ + B̃)J + (B̃′ − B̃)J 2 � = −(B̃′ + B̃) + 2J(B̃′ − B̃)
dJ
424 11 MOLECULAR SPECTROSCOPY
B̃′ + B̃
J head =
2(B̃′ − B̃)
1�2 1�2
1 3a 2 3 a3
=� � = � �
a(b − a�2) 8 8 b − a�2
Solutions to problems
P��F.� �e �rst transition is not allowed because it violates the spin selection rule,
∆S = 0; this transition has ∆S = 1. �e second transition is not allowed because
it violates the selection rule for Λ, ∆Λ = 0, ±1: this transition has ∆Λ = 2.
P��F.� Figure ��.�� is helpful in understanding the various quantities involved in this
problem. �e pure electronic energy of the ground electronic state, that is the
energy at the equilibrium separation, is (as a wavenumber) T̃e (X). Likewise,
the pure electronic energy of the excited state is T̃e (B). �e vibrational terms
of the ground state (in the harmonic approximation) are G̃ X (υ X ) = (υ X + 12 )ν̃ X ,
with these terms measured from T̃e (X). Likewise, those of the excited state are
G̃ B (υ B ) = (υ B + 12 )ν̃ B measured from T̃e (B).
�e wavenumber of the �–� transition, ν̃ 00 , is therefore
�e di�erence T̃e (B) − T̃e (X) is the value quoted as �.��� eV; this is converted
to a wavenumber using the factor from inside the front cover
8065.5 cm−1 1
ν̃ 00 = (6.175 eV) × + 2 × (700 cm−1 ) − 12 × (1580 cm−1 )
1 eV
= 4.936 × 104 cm−1
P��F.� (a) �e photoelectron spectrum involves a transition from the ground state
of the molecule to an electronic state of the molecular ion. �e energy
needed for the transition is measured indirectly by measuring the energy
of the ejected electron, but in all other respects the spectrum is interpreted
in the same way as electronic absorption spectra.
426 11 MOLECULAR SPECTROSCOPY
T̃e (B) υ′ = 0
ν̃ 00
Figure 11.14
In HBr only the ground vibrational state of the ground electronic state
will be signi�cantly populated, so only transitions from this level need
be considered. In principle there can be transitions from this vibrational
level to a range of di�erent vibrational levels of the upper electronic state
(a υ = 0 progression) and the intensities of these transitions will be gov-
erned by the Franck–Condon factors. As described in Section ��F.�(c)
on page ���, if the two electronic states have similar equilibrium bond
lengths the υ = 0 → υ′ = 0 transition will be the strongest, and then the
intensity will drop of quickly for higher values of υ′ . On the other hand, if
the upper state is displaced to the le� or right, several transitions will have
signi�cant Franck–Condon factors and several lines in the progression
will be observed.
In the photoelectron spectrum of HBr the band centred at about �� eV
shows extensive structure which is interpreted as being due to several
lines of a vibrational progression. �is is consistent with this band being
due to the removal of a bonding electron, resulting in the upper state (of
the ion) having a signi�cantly longer bond length than the ground state.
(b) �e band at around ��.� eV shows two peaks: these are not due to vi-
brational �ne structure, but to spin-orbit coupling in the molecular ion,
speci�cally such coupling associated with the Br atom. In the ion there
are unpaired electrons, so spin-orbit coupling in manifested in the spec-
trum. Removal of a nonbonding electron from a lone pair on the Br is
not expected to give a change in equilibrium bond length, so only the
υ = 0 → υ′ = 0 transition has signi�cant intensity: there is no vibrational
progression.
�
�
� L2 L2
= −e(1�L)� [cos(n − 1)π − 1] + sin(n − 1)π
�
� (n − 1) π (n − 1)π
�
2 2
�
�
L2 L2 �
− [cos(n + 1)π − 1] − sin(n + 1)π �
(n + 1) π (n + 1)π �
�
�
2 2
L2 L2 16Le
I 1,2 = −e(1�L) � [−1 − 1] − 2 [−1 − 1]� =
π 2 9π 9π 2
L2 L2
I 1,3 = −e(1�L) � [1 − 1] + − [1 − 1]� = 0
4π 2 16π 2
�us, as was to be shown, the transition dipole is non-zero for the transition
1 → 2, but zero for 1 → 3.
�is result is obtained much more simply by rewriting the integral using x =
(x − a�2) + a�2
L
I = −e(1�L) � sin(nπx�L) [(x − a�2) + a�2] sin(πx�L) dx
0
L
= −e(1�L) � sin(nπx�L) (x − a�2) sin(πx�L) dx
0
L
− e(1�L) � sin(nπx�L) (a�2) sin(πx�L) dx
0
R 1 R 2 −R�a 0
µ = −eR �1 + + � � �e
a0 3 a0
hence
R R 1 R 2 −R�a 0
−µ�ea 0 = �1 + + � � �e
a0 a0 3 a0
1.0
−µ�ea 0
0.5
0.0
0 2 4 6 8 10
R�a 0
Figure 11.15
selection rule is broken. Spin-orbit coupling is one of the e�ects that leads to
this rule being broken, and it is observed that the rates of these spin forbidden
transitions is enhanced by the presence of heavy atoms which have signi�cant
spin orbit coupling.
In the present case, if the added iodide ion is able to interact with the chro-
mophore for a signi�cant period of time (a time comparable with ISC or phos-
phorescence) then the spin-orbit coupling in the transient species may increase
the rate of ISC and/or of the phosphorescence, leading to an increase in the
intensity of the latter.
D��G.� �e characteristics of �uorescence which are consistent with the accepted mech-
anism are: (�) it ceases as soon as the source of illumination is removed; (�)
the time scale of �uorescence, 10−9 s, is typical of a process in which the rate
determining step is a spontaneous radiative transition between states of the
same multiplicity; (�) it occurs at longer wavelength (lower frequency) than
the exciting radiation; (�) its vibrational structure is characteristic of that of a
transition from the ground vibrational level of the excited electronic state to the
vibrational levels of the ground electronic state; and (�) the observed shi�ing,
and in some instances quenching, of the �uorescence spectrum by interactions
with the solvent.
Solutions to exercises
E��G.�(a) �is observed increase in the linewidth is a result of predissociation, as il-
lustrated in Fig. ��G.� on page ���. Where the dissociative 5 Рu state crosses
the bound upper electronic state the possibility exists that molecules in the
upper electronic state will undergo radiationless transitions to the dissociative
state leading to subsequent dissociation. �is process reduces the lifetime of
the excited states and so increases the linewidth of the associated transitions
(lifetime broadening, see Section ��A.�(b) on page ���).
Solutions to problems
P��G.� �e anthracene �uorescence spectrum re�ects the vibrational levels of the lower
electronic state. �e given wavelengths correspond to wavenumbers 22 730 cm−1 ,
24 390 cm−1 , 25 640 cm−1 , and 27 030 cm−1 ; these indicate spacings of the
vibrational levels of (24 390 cm−1 ) − (22 730 cm−1 ) = 1660 cm−1 , and, by
similar calculations, 1250 cm−1 , and 1390 cm−1 .
430 11 MOLECULAR SPECTROSCOPY
P��G.� (a) �e resonant modes satisfy nλ�2 = L therefore λ = 2L�n and ν = nc�2L.
With the data given
Integrated activities
I��.� �e electronic ground state of a closed-shell molecule transforms as the totally
symmetric irreducible representation. �is is because all of the orbitals are
doubly occupied by spin paired electrons. �e overall symmetry of a �lled
orbital is found from the direct product Γ(i) × Γ(i) , where Γ(i) is the irreducible
representation as which the orbital transforms. When multiplied out, such a
product always includes the totally symmetric irreducible representation and it
is this symmetry species which needs to be combined with the anti-symmetric
spin function for two spin-paired electrons.
(a) Ethene belongs to the point group D 2h : assume that the molecule lies
in the x y-plane, with the C–C bond along x. �e π bonding molecular
orbital transforms in the same way as the cartesian function z, that is as
B1u . �e π∗ anti-bonding molecular orbital transforms in the same way
as the cartesian function xz, that is as B2g . �e excited state (π)1 (π∗ )1
therefore has symmetry B1u × B2g = B3u (the direct product is found by
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 431
� � �
ψf µ̂ x ψi
B3u × B3u × Ag = Ag
� � �
ψf µ̂ ψi
A2 × Γx , y,z × A1 = A2 × Γx , y,z
I��.� �e energy of the HOMO, E HOMO , is reported in the table below, based on
calculations performed with Spartan �� using the DFT/B�LYP/�-��G* method.
�e experimentally determined energy of the I� –aromatic hydrocarbon charge
transfer bands is also given.
Figure ��.�� shows a plot of E HOMO against hν max ; the best-�t straight line is
also shown. �ere is a modest correlation between the two quantities.
−5.5
E HOMO �eV
−6.0
−6.5
R�2
d
R
R�2
Figure 11.17
I� = 2m H (R�2)2 = 1
m R2
2 H
= ��.�� pm
1�2
ħ
R=� �
4πm H c C̃
1.0546 × 10−34 J s
1�2
=� �
4π(1.67... × 10 kg) × (2.9979 × 1010 cm s−1 ) × (20.71 cm−1 )
−27
= ��.�� pm
(c) With the given value of R, the rotational constant is computed from B̃ =
ħ�4πcI� = ħ�2πm H cR 2
1.0546 × 10−34 J s
B̃ =
2π(1.67... × 10−27 kg) × (2.9979 × 1010 cm s−1 ) × (87.32 × 10−12 m)2
= ��.�� cm−1
m H 1�2
ν̃ D = ν̃ H � �
mD
1.0079 1�2
= (2521.6 cm−1 ) × � � = ����.� cm−1
2.0140
(a) Graphs of A(ν̃) and ε(ν̃) are shown in Fig. ��.��. �is band is likely to be
due to the fundamental transition of the antisymmetric stretching normal
mode of CO� , which has ν̃ 2 = 2349 cm−1 . �e broad features are the
unresolved P and R branches; no Q branch is expected for this mode. �e
principal contribution to the linewidth of an infrared transition is likely
to be pressure broadening.
1.0
100
0.5
0.0 0
2 300 2 350 2 400 2 300 2 350 2 400
−1 −1
ν̃�cm ν̃�cm
Figure 11.18
(b) Expressions for the wavenumbers of the lines in the P and R branches are
given in [��C.��a–���] and [��C.��c–���]
Using these expressions the positions and intensities of the lines in the P
and R branches are computed and a synthetic spectrum constructed by
assuming a linewidth and lineshape (here arbitrarily taken as a Gaussian).
Two such spectra are shown in Fig. ��.��: in (a) the linewidth has been
chosen so that the lines are well resolved; in (b) a much wider line is used
so that the lines in the P and R branches merge into a broad contour. �e
spectrum in (b) is roughly similar to that in Fig. ��.��, but the asymmetry
between the two branches is not reproduced by the calculation.
436 11 MOLECULAR SPECTROSCOPY
(a) (b)
2300 2320 2340 2360 2380 2400 2300 2320 2340 2360 2380 2400
~ ~
v / cm−1 v / cm−1
Figure 11.19
where h is in m.
�is integral cannot be solved by hand, but some work with mathematical
so�ware and a typical value of ε shows that the absorbance exceeds � by
the time h 0 = 30 m. Over such a small height it is safe to assume that the
pressure and temperature are constant, in which case the calculation of
the absorbance is much simpler
x CO2 p
A = ε[CO2 ]h 0 = ε h0
RT
= [ε�(mol−1 dm3 cm−1 )]
(3.3 × 10−4 ) × (1 bar)
× × (h 0 �cm)
(8.3145 × 10−2 dm3 bar K−1 mol−1 ) × (288 K)
= (1.378 × 10−5 ) × [ε�(mol−1 dm3 cm−1 )] × (h 0 �cm)
�e transmittance is T = 10−A .
Figure ��.�� shows plots of the transmittance as a function of height for
some representative values of the molar absorption coe�cient. For the
maximum value of ε seen in this absorption band (refer to Fig. ��.��) the
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 437
transmittance drops to �.� a�er less than � m. For a value of ε more typical
of the average (say ε = 100 mol−1 dm3 cm−1 ), the transmission drops to
�.� a�er about �.� m. Even for values of ε typical of the extremities of
the band, the transmission has fallen to �.� within �� m or so. A surface
plot of the transmission as a function of both wavenumber and height is
shown in Fig. ��.��.
1.0
ε = 200 mol−1 dm3 cm−1
0.8 ε = 100
ε = 50
0.6 ε = 25
T
0.4
0.2
0.0
0 5 10 15 20 25 30
h 0 �m
Figure 11.20
~
v /cm−1
h0 /m
Figure 11.21
I��.�� In Problem P��A.� it is shown that the integrated absorption coe�cient for a
Gaussian line shape is A = 1.0645 ε max ∆ν̃ 1�2 , where ∆ν̃ 1�2 is the width at half
height. Interpolating, by eye, a smooth curve across the band centred at about
��� nm, gives ε max = 250 mol−1 dm3 cm−1 . �e molar absorbance drops to half
438 11 MOLECULAR SPECTROSCOPY
this value at about ��� nm, which is 3.70×104 cm−1 and at about ��� nm, which
is 3.23 × 104 cm−1 , giving a width of ���� cm−1 . �e integrated absorption
coe�cient is therefore
� � �
ψf µ̂ x ψi
Γf × Γx , y,z × A1 = Γf × Γx , y,z
where that fact that the direct product with the totally symmetric irreducible
representation has no e�ect is used. �e only way this product can contain
the totally symmetric irreducible representation is if the irreducible represen-
tation of the �nal state, Γf , is equal to the irreducible representation of Γx , y,z .
�us, transitions from the A1 ground state to A1 , B1 , or B2 excited states are
allowed.
12 Statistical
thermodynamics
D��A.� In terms of molecular energy levels the thermodynamic temperature is the one
quantity that determines the most probable populations of the states of the
system at thermal equilibrium, as discussed in Section ��A.�(b) on page ���.
Solutions to exercises
E��A.�(a) �e Boltzmann population ratio for degenerate energy levels is given by [��A.��b–
���], N i �N j = (g i �g j )e−β(ε i −ε j ) . �e rotational term of a linear rotor is given
by [��B.��–���], F̃(J) = B̃J(J + 1) and, as explained in Section ��B.�(c) on page
���, its degeneracy is given as g J = 2J + 1. �e rotational energy is related to
the rotational term as ε J = hc F̃(J). �erefore
N5 2 × 5 + 1
= × e−hc B̃[5×(5+1)−0×(0+1)]�k T = 11 × e−30 B̃hc�k T
N0 2 × 0 + 1
using kT�hc = 207.224 cm−1 at 298.15 K (from inside the front cover)
N5
= 11 × e−30×(2.71 cm )�(207.224 cm ) = �.��
−1 −1
N0
Solutions to problems
P��A.� (a) �ere is no con�guration in which the molecules are distributed evenly
over the states and which, at the same time, satis�es the constraint that
the energy is 5ε.
(b) �e energy of a con�guration is E�ε = N 1 + 2N 2 + 3N 3 . . ., and the weight
of a con�guration is given by [��A.�–���], W = N!�(N 0 !N 1 !N 2 !...). �e
con�gurations satisfying the total energy constraint E = 5ε are
N0 N1 N2 N3 N4 N5 W
� � � � � � �
� � � � � � ��
� � � � � � ��
� � � � � � ��
� � � � � � ��
� � � � � � ��
� � � � � � �
N 0 N 1 N 2 N 3 N 4 N 5 N 6 N 7 N 8 N 9 N 10 W
�� � � � � � � � � � � ��
�� � � � � � � � � � � ���
�� � � � � � � � � � � � ���
�� � � � � � � � � � � ��� ���
�� � � � � � � � � � � ��� ���
�� � � � � � � � � � � � ��� ���
�� � � � � � � � � � � � ��� ���
�� � � � � � � � � � � � ��� ���
�� � � � � � � � � � � � ��� ���
�� �� � � � � � � � � � ��� ���
Of the ones listed in the table the con�guration with the greatest weight is
{12, 6, 2, 0, 0, 0, 0, 0, 0, 0, 0} .
�e Boltzmann population ratio is given by [��A.��a–���], N i �N 0 = e−βε i ,
where it is assumed that ε 0 = 0. It follows that ln N i �N 0 = −βε i = −iβε, and
hence βε = (ln N i �N 0 )�(−i). For i = 1, βε = (ln 6�12)�(−1) = 0.693; for
i = 2, βε = (ln 2�12)�(−2) = 0.896. Taking an average of these two values gives
βε = 0.795 and hence T = ε�(0.795k) .
442 12 STATISTICAL THERMODYNAMICS
With this value for the temperature the populations predicted by the Boltz-
mann distribution are
P��A.� �e Boltzmann population ratio for degenerate energy levels is given by [��A.��b–
���], N i �N j = (g i �g j )e−β(ε i −ε j ) . Taking logarithms gives
∆ε hc ν̃
T =− =−
k ln[(N 1 �N 0 )(g 0 �g 1 )] k ln [(N 1 �N 0 )(g 0 �g 1 )]
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (450 cm−1 )
=−
(1.3806 × 10−23 J K−1 ) × ln [(30%�70%) × (2�4)]
= 420 K
From the perfect gas law, pV = nRT, and because N ∝ n, it follows that p ∝ N,
and therefore p(h)�p 0 = N(h)�N(0). Because β = 1�(kT), m = M�N A , and
k = R�N A it follows that
mgh (M�N A )gh M gh h
βmgh = = = =
kT (R�N A )T RT H
where H = RT�M g is used. Hence
and for H� O
M gh
= exp �− �
N(h)
N0 RT
(18.02 × 10−3 kg mol−1 ) × (9.807 m s−2 ) × (8.0 × 103 m)
= exp �− �
(8.3145 J K−1 mol−1 ) × (298 K)
= 0.57
D��B.� It is possible for there to be di�erent wavefunctions which have the same en-
ergy: such wavefunctions are said to be degenerate. If this is the case, for a
given ‘energy level’, that is a given value of the energy, there are several ‘states’
each of which is distinct but has the same energy.
�e partition function is computed as a sum over the states. However, because
degenerate states have the same energy, the sum may be computed as a sum over
energy levels, as long as the degeneracy g i of each level is taken into account.
q = � e−βε i = � g i e−βε i
states i levels i
Solutions to exercises
E��B.�(a) �e rotational partition function of a heteronuclear diatomic is given by [��B.��–
���], q R = ∑ J (2J + 1)e−β hc B̃ J(J+1) . �is is evaluated explicitly by summing
successive terms until they become too small to a�ect the result to a given level
of precision. �e partition function in the high-temperature limit is given by
444 12 STATISTICAL THERMODYNAMICS
= (0.359... K−1 ) × T
�e values of q R computed in these two di�erent ways are compared in Fig. ��.�.
�e high temperature limit becomes accurate to within � % of the exact solution
at around 18 K .
10
qR
exact
high T limit
0
0 5 10 15 20 25 30
T�K
Figure 12.1
k × T 3�2
q R = π 1�2 � �
hc B̃
(1.3806 × 10−23 J K−1 )
3�2
= π 1�2 � � × T 3�2
(6.6261 × 10 J s)×(2.9979 × 1010 cm s−1 )×(5.241 cm−1 )
−34
�e values of q R computed in these two di�erent ways are compared in Fig. ��.�.
�e high temperature limit becomes accurate to within � % of the exact solution
at around 37 K .
30
20
qR 10
exact
high T limit
0
0 10 20 30 40 50
T�K
Figure 12.2
1 kT 3�2 π 1�2
qR = � � � �
4 hc ÃB̃C̃
(1.3806 × 10−23 J K−1 ) × (298.15 K)
3�2
1
= � �
4 (6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
1�2
π
� �
(4.828 cm−1 ) × (1.0012 cm−1 ) × (0.8282 cm−1 )
= ���.�
qV 10
exact
high T limit
0
0 1 000 2 000 3 000 4 000 5 000
T�K
Figure 12.3
−1
q V1 = �1 − e−hc β ν̃ 1 �
−1
= �1 − e−(2.87 ...×10 � = 1.17...
−3
cm)×(658 cm−1 )
Similarly
−1
q V2 = �1 − e−(2.87 ...×10 � = 1.46...
−3
cm)×(397 cm−1 )
−1
q V3 = �1 − e−(2.87 ...×10 � = 1.01...
−3
cm)×(1535 cm−1 )
−1
q V1 = �1 − e−hc β ν̃ 1 �
−1
= �1 − e−(2.87 ...×10 � = 1.36...
−3
cm)×(459 cm−1 )
Similarly
−1
q V2 = �1 − e−(2.87 ...×10 � = 2.15...
−3
cm)×(217 cm−1 )
−1
q V3 = �1 − e−(2.87 ...×10 � = 1.12...
−3 −1
cm)×(776 cm )
−1
q V4 = �1 − e−(2.87 ...×10 � = 1.68...
−3
cm)×(314 cm−1 )
q V = q V1 × (q V2 )2 × (q V3 )3 × (q V4 )3
= (1.36...) × (2.15...)2 × (1.12...)3 × (1.68...)3 = 42.1 .
q E = g 0 + g 1 × e−βε 1 + g 2 × e−βε 2
= 4 + 1×e−(7.57 ...×10 cm)×(2500 cm ) + 2×e−(7.57 ...×10
−4 −1 −4
cm)×(3500 cm−1 )
qHT 2 V �Λ H2 3
3 3
Λ He h�(2πm He kT)1�2 m H2 3�2
= = � � = � � = � �
T
qHe V �Λ He 3 Λ H2 h�(2πm H2 kT)1�2 m He
2 × 1.0079 g mol−1
3�2
qHT 2 M H2 3�2
= � � = � � = �.���
T
qHe M He 4.00 g mol−1
kT 3�2 π 1�2
qR = � � � �
hc ÃB̃C̃
(1.3806 × 10−23 J K−1 ) × (298.15 K)
3�2
=� �
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
1�2
π
� �
(3.1752 cm ) × (0.3951 cm−1 ) × (0.3505 cm−1 )
−1
= 7.97 × 103
= 1.12 × 104
Solutions to problems
P��B.� �e partition function is given by [��B.�b–���], q = ∑ i g i e−βε i , where g i is de-
generacy and the corresponding energy is ε i . �erefore the partition function
is
�is is plotted in Fig ��.�. As expected, the partition function rises from a
value of � at low temperatures, where only the ground state is occupied, and
approaches a value of � at high temperatures, when all three states are nearly
equally populated.
P��B.� As discussed in Section ��C.�(a) on page ���, the Morse oscillator has a �nite
number of bound levels between the ground level and the dissociation limit.
�e number of these is found by noting that E υ reaches a maximum value at
the dissociation limit, and therefore this limit is found by solving dE υ �dυ = 0
d
�(υ + 12 )hc ν̃ − (υ + 12 )2 hcx e ν̃� = hc ν̃ − 2(υ + 12 )hcx e ν̃
dυ
solving 0 = hc ν̃ − 2(υ max + 12 )hcx e ν̃ gives υ max = 1�2x e − 1
2
2
q
0 5 10 15 20 25 30
kT�ε
Figure 12.4
υ max
q VM = � e−(υhc ν̃−(υ +υ)hc x e ν̃)�k T
2
υ=0
υ max
q VM = � e−(υ−(υ +υ)x e )θ V �T
2
υ=0
For the harmonic oscillator the partition is given by the exact expression [��B.��–
���], q VHO = (1 − e−θ �T )−1 .
V
Figure ��.� compares the partition functions for various values of x e with that
for the harmonic case. For the smallest value of x e the partition function is
initially larger than that for the harmonic oscillator. �is can be attributed
to more energy levels contributing at these temperatures as they are closer in
energy for the Morse oscillator than for the harmonic case. However, at higher
temperatures the partition function for the Morse oscillator starts to level o�
because there are a �nite number of levels, whereas for the harmonic case the
partition function continues without limit as there are an in�nite number of
levels.
�is behaviour is even more pronounced for x e = 0.05 and x e = 0.10. In these
two cases υ max is �� and �, respectively, and these values set the limiting high-
temperature value of the partition function.
harmonic
x e = 0.02
15
x e = 0.05
x e = 0.10
10
qV
0
0 2 4 6 8 10 12 14 16
V
T�θ
Figure 12.5
+ 5 × e−(4.82 ...×10
−3
cm)×(10559 cm−1 )
(ii) At T = 5000 K
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
hcβ =
(1.3806 × 10−23 J K−1 ) × (5000 K)
= 2.87... × 10−4 cm
+ 5 × e−(2.87 ...×10
−4
cm)×(10559 cm−1 )
N 0 �N = 5�5.00... = �.��
N 2 �N = (3.26... × 10−10 )�5.00... = 6.54 × 10−11
452 12 STATISTICAL THERMODYNAMICS
(ii) T = 5000 K
N0 5 N 2 0.764...
= = �.��� = = �.���
N 6.262... N 6.262...
P��B.� �e partition function is given by [��B.�b–���], q = ∑ i g i e−βε i , where g i is
degeneracy and the corresponding energy is given as ε i = hc ν̃ i , and β = 1�kT.
Here g i = 2J +1, where J is the right subscript in the term symbol. At T = 298 K
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
hcβ = = 4.82... × 10−3 cm
(1.3806 × 10−23 J K−1 ) × (298 K)
�erefore the electronic partition function is
q E = 1 + 3 × e−(4.82 ...×10 cm)×(557.1 cm )
−3 −1
+ 1 × e−(4.82 ...×10
−3
cm)×(16367.3 cm−1 )
+ 1 × e−(1.43 ...×10
−3
cm)×(16367.3 cm−1 )
30
20
qR 10
exact
high T limit
0
0 2 4 6 8 10
T�K
Figure 12.6
Solutions to exercises
E��C.�(a) �e mean vibrational energy is given by [��C.�–���], �ε V � = hc ν̃�(e β hc ν̃ − 1);
this result is exact. �e equipartition value is �ε V � = kT, because there are two
quadratic terms for a harmonic oscillator. �ese two expressions for the energy
are plotted as a function of T in Fig. ��.�. �e value from the equipartition
theorem comes within � % of the exact value at 4.80 × 103 K .
6
�ε V � × 1020 �J
2
exact
equipartition
0
0 1 000 2 000 3 000 4 000 5 000
T�K
Figure 12.7
�ε V � = �ε V1 � + 2 × �ε V2 � + �ε V3 �
60
�ε V � × 1020 �J
40
20
exact
equipartition
0
0 2 000 4 000 6 000 8 000 10 000 12 000
T�K
Figure 12.8
�ε V � = �ε V1 � + 2 × �ε V2 � + 3 × �ε V3 � + 3 × �ε V4 �
100
�ε V � × 1020 �J
50
exact
equipartition
0
0 2 000 4 000 6 000 8 000
T�K
Figure 12.9
1 ∂q E hc
�ε E � = − � � = E �g 1 ν̃ 1 e−β hc ν̃ 1 + g 2 ν̃ 2 e−β hc ν̃ 2 �
q E ∂β V q
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
=
4.29...
× �1 × (2500 cm ) × e−(7.57 ...×10 cm)×(2500 cm )
−4 −1
−1
= 4.03 × 10−21 J
0 + εe−βε ε hc ν̃
�ε� = = βε = hc ν̃�k T
1+e −βε e +1 e +1
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (500 cm−1 )
= (6.6261×10−34 J s)×(2.9979×10 10 cm s−1 )×(500 cm−1 )
e (1.3806×10−23 J K−1 )×(298 K) +1
= 8.15 × 10−22 J
1 ∂q R 1
�ε R � = − � � = R � hc B̃J(J + 1)(2J + 1)e−β hc B̃ J(J+1)
q R ∂β V q J
456 12 STATISTICAL THERMODYNAMICS
�e terms of the sum above, and also of the sum needed to compute q R , are
evaluated and summed until the result has converged to the required precision.
�e equipartition value is �ε R � = kT. �ese two expressions for the energy
are plotted as a function of T in Fig. ��.��. �e value from the equipartition
theorem comes within � % of the exact value at 19.6 K .
2
�ε R � × 1022 �J
exact
equipartition
0
0 5 10 15 20
T�K
Figure 12.10
1 ∂q R 1
�ε R � = − � � = � hc B̃J(J + 1)(2J + 1) e
2 −β hc B̃ J(J+1)
q R ∂β V q R J
�e terms in the sum needed to compute q R and �ε R � are evaluated and summed
until the result has converged to the required precision. �e equipartition value
is �ε R � = 32 kT, because for this non-linear molecule there are three rotational
degrees of freedom. �ese two expressions for the energy are plotted as a func-
tion of T in Fig. ��.��. �e value from the equipartition theorem comes within
� % of the exact value at 26.4 K .
Solutions to problems
P��C.� �e mean molecular energy is given by [��C.�a–���], �ε� = −(1�q)(∂q�∂β)V ,
where β = 1�kT, and q is the partition function given by [��B.�b–���], q =
∑ i g i e−βε i . For a symmetric rotor the rotational terms are given in [��B.��a–
���], F̃(J, K) = B̃J(J + 1) + (à − B̃)K 2 , with J = 0, 1, 2, ... and K = 0, ±1, ..., ±J;
the corresponding energies are hc F̃(J, K) and the degeneracy is (2J + 1). �e
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 457
�ε R � × 1022 �J
4
2
exact
equipartition
0
0 5 10 15 20 25 30
T�K
Figure 12.11
1 ∂q R
�ε R � = − � �
q R ∂β V
1 �∞ � +J �
−β hc B̃ J(J+1) � −β hc( Ã− B̃)K 2 �
= � + + � � �
�K=−J �
hc B̃J(J 1)(2J 1)e e
q R � J=0 � �
∞ � +J �
2 ��
�
+ �(2J + 1)e−β hc B̃ J(J+1) � � hc(Ã − B̃)K 2 e−β hc( Ã−B̃)K �
�K=−J ��
J=0 � �
1 ∞
= �(2J + 1)e
−β hc B̃ J(J+1)
q R J=0
� +J �
� 2�
× �hc � �B̃J(J + 1) + (Ã − B̃)K 2 � e−β hc( Ã−B̃)K �
� K=−J �
� �
�e terms in the sum in this expression, and the terms in q R , are evaluated and
summed until the value converges to the required precision. �e equipartition
value of the energy is �ε R � = 32 kT because there are three rotational degrees
of freedom. �e two expressions for the energy are plotted as a function of
temperature in Fig. ��.��. �e equipartition value is within �% of the exact
solution at 4.59 K .
P��C.� H� O has three translational and three rotational degrees of freedom. It there-
fore follows from the equipartition principle that the molar internal energy is
U m = (3 + 3) × 12 RT = 3RT. �e constant-volume molar heat capacity is
therefore C V ,m = (∂U m �∂T)V = 3R.
458 12 STATISTICAL THERMODYNAMICS
1.0
�ε R � × 1022 �J
0.5
exact
equipartition
0.0
0 1 2 3 4 5 6
T�K
Figure 12.12
P��C.� �e energy levels for a spin in a magnetic �eld are given by [��A.�d–���], E m I =
−g I µ N B0 m I , where m I = +1, 0, −1. �ese energies are conveniently written as
E m I = −m I δ, with δ = g I µ N B0 . If the energy of the lowest state is de�ned as the
energy zero, then the three levels have energies E 0 = 0, E 1 = δ, and E 2 = 2δ.
�e partition function is
q = 1 + e−βδ + e−2βδ
1 ∂q
�ε� = ε gs − � �
q ∂β V
1
= ε gs − �−δe−βδ − 2δe−2βδ �
1 + e−βδ + e−2βδ
δe−βδ + 2δe−2βδ δe−βδ + 2δe−2βδ
= ε gs + = −δ +
1 + e−βδ + e−2βδ 1 + e−βδ + e−2βδ
In the last step the fact that the energy of ground state (the one with m I = +1)
is −δ is used. With the data given
�e separation of the energy levels is very much smaller than kT, therefore the
three levels have almost equal populations giving a mean energy of very close
to zero.
4.0
3.5
3.0
qE
2.5
2.0
0 200 400 600 800 1 000
T�K
Figure 12.13
N i �N j = (g i �g j )e−β(ε i −ε j )
1 dq E e−hc ν̃�k T hc ν̃
�ε E � = − = hc ν̃ × =
q dβ
E 1 + e−hc ν̃�k T e hc ν̃�k T + 1
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (121.1 cm−1 )
= (6.6261×10−34 J s)×(2.9979×10 10 cm s−1 )×(121.1 cm−1 )
e (1.3806×10−23 J K−1 )×(300 K) +1
= 8.63 × 10−22 J
P��C.� Mean values of any observable are given by a sum of the observed value over
all the possible states weighted by the probability of each state. �us, the mean
of the square of energy is given by �ε 2 � = (1�q) ∑ j ε j 2 e−βε j , where q is the
partition function given by [��A.��–���], q = ∑ i e−βε i .
It is useful to consider the �rst and second derivatives of the term e−βε j with
respect to β. �e �rst derivative is
d −βε j
e = −ε j e−βε j
dβ
d2 d d −βε j
2e
−βε j
= �−ε j e−βε j � = (−ε j ) × e
dβ dβ dβ
= (−ε j ) × (−ε j ) × e−βε j = ε j 2 × e−βε j
1
�ε 2 � = 2 −βε
�ε e j
q j j
� �
� �
�� e−βε j � = 1 d q
2
1 d2 −βε 1 d2
= � 2e j = � � q 2
q dβ 2 � j �
� �
q j dβ dβ
�erefore
1�2
1 d2 q
�ε �
2 1�2
= � �
q dβ 2
�ese results are also used to �nd the root mean square of the deviation from
the mean as
� 1 d2 q 1 dq � 1 � d2 q dq �
2 1�2 2 1�2
��ε � − �ε� � = 2 −� � = q 2 −� �
2 2 1�2
� q dβ q dβ � q � dβ dβ �
dq −hc ν̃ e−β hc ν̃
= = −hc ν̃ e−β hc ν̃ q 2
dβ (1 − e−β hc ν̃ )2
d2 q −hc ν̃ e−β hc ν̃ −2hc ν̃ e−β hc ν̃ e−β hc ν̃
= −hc ν̃ � + �
dβ 2 (1 − e−β hc ν̃ )2 (1 − e−β hc ν̃ )3
1 2e−β hc ν̃
= (hc ν̃)2 e−β hc ν̃ � + �
(1 − e−β hc ν̃ )2 (1 − e−β hc ν̃ )3
= (hc ν̃)2 e−β hc ν̃ �q 2 + 2e−β hc ν̃ q 3 �
= q 2 (hc ν̃)2 e−β hc ν̃ �1 + 2e−β hc ν̃ q�
x 2 e−βx x 2 e−2βx
= q x 2 e−βx + q 2 x 2 e−2βx = +
1−e −βx (1 − e−βx )2
x 2 e−βx (1 − e−βx ) + x 2 e−2βx x 2 e−βx
= =
(1 − e−βx )2 (1 − e−βx )2
hence
hc ν̃ e−β hc ν̃�2
��ε 2 � − �ε� � =
2 1�2
1 − e−β hc ν̃
kθ R e−θ �2T
R
1 − e−θ R �T
particles, the same volume, and the same temperature, but they need not have
the same energy.
D��D.� In the context of ensembles, the thermodynamic limit is achieved as the num-
ber of replications Ñ approaches in�nity. In this limit, the dominating con�g-
uration is overwhelmingly the most probable con�guration, and its properties
are essentially the same as those of the system.
Solutions to exercises
E��D.�(a) It is essential to include the factor 1�N! when considering indistinguishable
particles which are free to move. �us, such a factor is always needed for gases.
In the solid state, particles are distinguished by their positions in the lattice
and therefore the particles are regarded as distinguishable on the basis that
their locations are distinguishable. For the cases mentioned, the factor 1�N!
is needed for all but solid CO.
Solutions to problems
P��D.�
∂ ln(V �Λ 3 ) ∂(ln V − ln Λ 3 )
p = N kT � � = N kT � �
∂V T
∂V T
∂ ln V N kT nRT
= N kT � � = =
∂V T V V
Vm e5�2
S m = R ln � � Λ = h�(2πmkT)1�2 Vm = RT�p
NA Λ3
Because the molar volume appears in the numerator, the molar entropy in-
creases with the molar volume. In terms of the Boltzmann distribution, this
relationship is expected: large containers have more closely spaced energy lev-
els than do small ones, so more states are thermally accessible. Temperature
appears in the numerator of the expression (through the denominator of Λ),
so the molar entropy increases with the temperature. Again, this is consistent
with the Boltzmann distribution, because more states are accessible at higher
temperatures than at lower ones.
�e fact that diatomic and polyatomic gases have rotational and vibrational
modes of motion as well does not change the above arguments. �e parti-
tion functions of those modes are independent of volume, so the volume de-
pendence of the entropy is as described above. At most temperatures, rota-
tional modes of motion are active and contribute to the entropy, as expressed
in [��E.��a–���]; the contribution increases with temperature. Finally, most
vibrational modes contribute little if at all to the entropy, but as with rotation
the contribution increases with temperature.
464 12 STATISTICAL THERMODYNAMICS
D��E.� �e temperature is always high enough for the mean translational energy to
be 32 kT, the equipartition value. �erefore, the molar constant-volume heat
capacity for translation is C VT ,m = 32 R.
When the temperature is high enough for the rotations of the molecules to
be highly excited (when T � θ R ) the equipartition value kT for the mean
rotational energy (for a linear rotor) can be used to obtain C VR ,m = R. For non-
linear molecules, the mean rotational energy is 32 kT, so the molar rotational
heat capacity rises to 32 R when T � θ R . At intermediate temperatures the total
heat capacity takes a value between that due to translation, 32 R, and 52 R (for a
linear molecule) when both translation and rotation contribute fully.
Molecular vibrations contribute to the heat capacity, but only when the temper-
ature is high enough for them to be signi�cantly excited. For each vibrational
mode, the equipartition mean energy is kT, so the maximum contribution to
the molar heat capacity is R. However, it is unusual for the vibrations to be so
highly excited that equipartition is valid, and in general the contribution to the
heat capacity has to be calculated using [��E.�–���].
Solutions to exercises
E��E.�(a) For atoms with �lled shells the only contribution to the entropy is translational.
�e standard molar entropy of a monatomic perfect gas is given by the Sackur–
Tetrode equation [��E.�b–���]
kTe5�2
−
○
Sm = R ln � � Λ = h�(2πmkT)1�2
p−○ Λ 3
6.6261 × 10−34 J s
Λ=
�2π(4.00×1.6605 × 10−27 kg)×(1.3806 × 10−23 J K−1 )×(298 K)�
1�2
= 5.05... × 10−11 m
(1.3806 × 10−23 J K−1 ) × (298 K) × e5�2
−
○
Sm = (8.3145 J K−1 mol−1 ) × ln � �
(105 N m−2 ) × (5.05 . . . × 10−11 m)3
= ��� J K−1 mol−1
6.6261 × 10−34 J s
Λ=
�2π(131.29×1.6605 × 10−27 kg)×(1.3806 × 10−23 J K−1 )×(298 K)�
1�2
= 8.82... × 10−12 m
(1.3806 × 10−23 J K−1 ) × (298 K) × e5�2
−
○
Sm = (8.3145 J K−1 mol−1 ) × ln � �
(105 N m−2 ) × (8.82... × 10−12 m)3
= ���.� J K−1 mol−1
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 465
E��E.�(a) For atoms with �lled shells the only contribution to the entropy is translational.
�e standard molar entropy of a monatomic perfect gas is given by the Sackur–
Tetrode equation [��E.�b–���]
kTe5�2
−
○
Sm = R ln � � Λ = h�(2πmkT)1�2
p−○ Λ 3
−
○
It follows that S m = A ln(T 5�2 m 3�2 ), where A is a constant. �erefore
−
○
(He, T1 ) − S m
−
○
(Xe, T2 ) = A ln(T1 m He ) − A ln(T2 m Xe )
5�2 3�2 5�2 3�2
Sm
−
○
If S m (He, T1 ) = S m
−
○
(Xe, 298)
S m = [U m (T) − U m (0)]�T + R ln q
�is is the appropriate form for the rotational contribution; for the translational
contribution the ln term is ln qe�N. At ��� K kT�hc = 207 cm−1 which is signif-
icantly greater than any of the rotational constants, therefore the equipartition
theorem can be used to �nd U m (T): there are three rotational modes, therefore
U m (T) − U m (0) = 32 RT.
R
Sm = ( 32 RT)�T + R ln q R = R( 32 + ln q R )
= (8.3145 J K−1 mol−1 ) × [ 32 + ln(43.1...)]
= ��.�� J K−1 mol−1
466 12 STATISTICAL THERMODYNAMICS
E��E.�(a) Only the ground electronic state contributes to the electronic partition func-
tion, which is therefore simply the degeneracy of the ground state q E = g 0 .
For a given term the degeneracy is given by the value of J, which is the right
subscript: g 0 = (2J + 1) = (2 × 92 + 1) = 10. �e entropy is given in terms of the
partition function by [��E.�a–���]
S m = [U m (T) − U m (0)]�T + R ln q
�is is the appropriate form for the electronic contribution; for the translational
contribution the ln term is ln qe�N. In this case U m (T) − U m (0) = 0 as only
the ground state is considered
θ V �T
= R� − ln(1 − e−θ �T )� θ V = hc ν̃�k
V
V
Sm
eθ V �T − 1
��� K ��� K
ν̃�cm −1
θ �K
V
θ �T
V V
Sm �R θ �TV V
Sm �R
��� ��� �.�� �.��� �.�� �.���
��� ��� �.�� �.��� �.�� �.���
���� ���� �.��� �.����� �.��� �.����
���� ���� �.��� �.����� �.��� �.����
���� ���� �.��� �.����� �.��� �.����
���� ���� �.��� 9.515 × 10−3 �.��� �.�����
���� ���� �.��� 1.855 × 10−3 �.��� �.�����
���� ���� ��.�� 1.026 × 10−5 �.��� 1.988 × 10−3
���� ���� ��.�� 5.957 × 10−7 ��.�� 3.895 × 10−4
(ii) CH� : three translational modes, three rotational modes (non-linear), and
no active vibrational modes: C V ,m �R = 3 × 12 + 3 × 12 = 3 .
(iii) C� H� : three translational modes, three rotational modes (non-linear),
and no active vibrational modes: C V ,m �R = 3 × 12 + 3 × 12 = 3 . �ere
are four low-frequency normal modes which, if active, will contribute a
further 4R.
�e experimental value for γ is �.�� for both gases: evidently the vibrational
modes are not active.
q = g 0 + g 1 e−β hc ν̃
where g 0 and g 1 are the degeneracies of the ground and �rst excited state,
respectively. �e mean energy is given by [��C.�a–���], �ε� = −(1�q)(∂q�∂β)V
g 1 hc ν̃ e−β hc ν̃ g 1 hc ν̃
�ε� = =
g 0 + g 1 e−β hc ν̃ g 0 e β hc ν̃ + g 1
N A g 1 hc ν̃
hence U m = N A �ε� =
g 0 e β hc ν̃ + g 1
∂U m ∂U m dβ ∂U m −1
C V ,m = � � =� � =� � ×
∂T V ∂β V dT ∂β V kT 2
1 g 0 hc ν̃e β hc ν̃
= × N A g 1 hc ν̃
kT 2 (g 0 e β hc ν̃ + g 1 )2
N A (hc ν̃)2 g 0 g 1 e β hc ν̃
=
kT 2 (g 0 e β hc ν̃ + g 1 )2
N A (hc ν̃)2 8 e β hc ν̃
C V ,m (500) =
kT 2 (4 e β hc ν̃ + 2)2
(6.0221 × 1023 mol−1 )×(1.75... × 10−20 J)2 8 e2.53 ...
=
(1.3806 × 10 J K ) × (500 K)
−23 −1 2 (4 e2.53 ... + 2)2
= �.�� J K−1 mol−1
(6.0221 × 1023 mol−1 )×(1.75... × 10−20 J)2 8 e1.40 ...
C V ,m (900) =
(1.3806 × 10 J K ) × (900 K)
−23 −1 2 (4 e1.40 ... + 2)2
= �.�� J K−1 mol−1
θ V � e−θ �2T �
2 V 2
C V ,m = R� � θ V = hc ν̃�k
T � 1 − e−θ �T �
V
��� K ��� K
ν̃�cm−1 θ V �K T�θ V C V ,m �R T�θ V C V ,m �R
��� ��� �.��� �.��� �.��� �.���
��� ���� �.��� �.��� �.��� �.���
���� ���� �.���� 6.593 × 10−3 �.���� �.����
���� ���� �.����� 3.226 × 10−5 �.���� 6.980 × 10−3
���� ���� �.����� 2.233 × 10−5 �.���� 5.725 × 10−3
1.0
0.8
0.6
C V ,m �R
0.4
0.2
0.0
0.0 0.2 0.4 0.6 0.8 1.0
V
T�θ
Figure 12.14
Solutions to problems
P��E.� �e electronic levels of NO form a two-level system, an expression for the heat
capacity of which is derived in the solution to Exercise E��E.�(a).
N A (hc ν̃)2 g 0 g 1 e β hc ν̃ g 0 g 1 e β hc ν̃
C V ,m = = N A k(βhc ν̃)2
kT 2 (g 0 e β hc ν̃ + g1 ) 2 (g 0 e β hc ν̃ + g 1 )2
where g 0 and g 1 are the degeneracies of the ground and �rst excited state,
respectively. For NO g 0 = 2 and g 1 = 2. With the data given
βhc ν̃ = hc ν̃�k × T −1
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (121.1 cm−1 )
= × T −1
1.3806 × 10−23 J K−1
= (174.2... K) × T −1
A plot of C V ,m as a function of T is shown in Fig. ��.��.
P��E.� �e energy levels for a particle on a ring are given by [�F.�–���], E m = m 2 ħ 2 �2I
where I is the moment of inertia and m = 0, ±1, ±2, . . .. In the high-temperature
limit the partition function is well-approximated by an intergral
+∞
q R = � e−βE m = � e−βm ħ �2I
2 2
dm
m −∞
�e integral is of the form of Integral G.� with k ′ = βħ 2 �2I; the value needed is
twice that given for G.� as that integral is from 0 to +∞.
1�2
2πI
q = � 2�
R
βħ
�ese energy levels contribute one quadratic contribution to the energy, that is
there is one rotational mode. In the high-temperature limit the equipartition
theorem applies and hence the internal energy is U m = 12 RT and C VR ,m = 12 R .
470 12 STATISTICAL THERMODYNAMICS
0.4
C V ,m �R
0.2
0.0
0 100 200 300 400 500
T�K
Figure 12.15
q = 1 + e−βε + e−2βε
ε e−βε + 2ε e−2βε ε e βε + 2ε
E mean = =
1 + e−βε + e−2βε e2βε + e βε + 1
e βε + 2
hence U m = N A E mean = N A ε 2βε
e + e βε + 1
�e entropy is given in terms of the partition function by [��E.�a–���]
S m = [U m (T) − U m (0)]�T + R ln q
NA ε e βε + 2
= + R ln(1 + e−βε + e−2βε )
T e + e βε + 1
2βε
βε(e βε + 2)
= R � 2βε + ln(1 + e−βε + e−2βε )�
e + e βε + 1
P��E.� �e data in Problem P��B.� �t very well to the terms F̃(J) = B̃J(J + 1) with B̃ =
10.593 cm−1 . �e rotational contribution to the entropy is given by [��E.�a–
���]
R
Sm = [U m (T) − U m (0)]�T + R ln q R
where the internal energy is given by [��E.�a–���]
N A ∂q R
U m (T) − U m (0) = − � �
q R ∂β V
∂q R
� � = −hc B̃ �(2J + 1)[J(J + 1)]e−hc β B̃ J(J+1)
∂β V J
472 12 STATISTICAL THERMODYNAMICS
∂q R
q R = �(2J +1)e−θ � � = −kθ R �(2J +1)[J(J +1)]e−θ J(J+1)�T
R R
J(J+1)�T
J ∂β V J
Rθ R
U m (T) − U m (0) = �(2J + 1)[J(J + 1)]e
−θ R J(J+1)�T
qR J
�e sums are best evaluated using mathematical so�ware and the results are
expressed in terms of the dimensionless parameter T�θ R . �e result of such a
calculation is shown in Fig. ��.��
2
Sm
R
�R
0
0 1 2 3 4 5
R
T�θ
Figure 12.16
P��E.�� Contributions to the entropy from translation, rotation and vibration are ex-
pected. �e molecule has a doubly-degenerate ground electronic state, so this
will also contribute to the entropy. However, the excited electronic states are at
energies very much greater than kT at ��� K (kT at ��� K is �.��� eV), so their
contribution is negligible.
kTe5�2
T
Sm = R ln � � Λ = h�(2πmkT)1�2
p−○ Λ 3
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 473
= 1.64... × 10−11 m
(1.3806 × 10−23 J K−1 ) × (298 K) × e5�2
T
Sm = (8.3145 J K−1 mol−1 ) × ln � �
(105 N m−2 ) × (1.64... × 10−11 m)3
= 1.54... × 102 J K−1 mol−1
P��E.�� It is convenient to calculate the entropy using the approach set out in Problem
P��E.� in which the two quantities q (the partition function) and q̇ are de�ned
q = � e−βε j q̇ = � βε j e−βε j
j j
It is shown in the solution to that problem that the entropy can be written in
terms of these as
q̇
S m = R � + ln q� (��.�)
q
As discussed in Section ��C.�(a) on page ���, the Morse oscillator has a �nite
number of bound levels between the ground level and the dissociation limit.
�e number of these is found by noting that E υ reaches a maximum value at
the dissociation limit, and therefore this limit is found by solving dE υ �dυ = 0
d
�(υ + 12 )hc ν̃ − (υ + 12 )2 hcx e ν̃� = hc ν̃ − 2(υ + 12 )hcx e ν̃
dυ
solving 0 = hc ν̃ − 2(υ max + 12 )hcx e ν̃ gives υ max = 1�2x e − 1
2
υ=0
υ=0
�ese results are used with eqn ��.� to compute the entropy. For comparison,
for a harmonic oscillator the entropy is given by [��E.��b–���] (note that there
is an error in the expression in the text: the argument of the exponential term
in the ln should be negative)
θ V �T
S m �R = − ln(1 − e−θ �T )
V
eθ V �T −1
Figure ��.�� compares the entropy for various values of x e with that for the
harmonic case. For the smallest value of x e the entropy is initially larger than
that for the harmonic oscillator. �is can be attributed to fact that the energy
levels are more closely spaced for the Morse oscillator than for the harmonic
oscillator. However, at higher temperatures the entropy for the Morse oscillator
starts to level o� because there are a �nite number of levels, whereas for the
harmonic case the entropy continues to increase without limit as there are an
in�nite number of levels. �is behaviour is even more pronounced for x e = 0.05
and x e = 0.10, with the plateau at high temperatures being evident.
�ese plots are, however, somewhat unrealistic. For a typical molecule θ V ≈
1000 K, so at ��� K T�θ V ≈ 0.1, and x e is around �.���. With these parameters
the contribution to the entropy determined using the Morse levels is 5.0 ×
10−4 × R; the result obtained using the harmonic levels is the same. �is is
because there is very little contribution from excited vibrational states, so the
small di�erence between these low-lying states for the Morse and harmonic
oscillators has no signi�cant e�ect on the partition function.
harmonic
3.0 x e = 0.02
x e = 0.05
x e = 0.10
2.0
S m �R
1.0
0.0
0 2 4 6 8 10
V
T�θ
Figure 12.17
P��E.�� �e partition function for a particle con�ned to a box of length X in one di-
mension is given by [��B.�–���]
q X = X�Λ Λ = h�(2πmkT)1�2
�erefore the partition function for a particle con�ned to a two-dimensional
box of dimensions X and Y is
q XY = q X qY = XY�Λ 2 = A�Λ 2
476 12 STATISTICAL THERMODYNAMICS
where A is the area. Because there are two translational modes, the internal
energy of such a system is given by the equipartition theorem as U = nRT,
where n is the amount in moles. �e entropy of n moles is given by [��E.�b–
���]
S = U�T + N k ln qe�N
Ae Ae
= nR + N k ln 2 = nR + nN A k ln 2
Λ N Λ nN A
Am e
= nR + nR ln 2
Λ NA
where in the last step the molar area, A m = A�n, is introduced. �e molar
entropy is therefore
Am e Am e
2D
Sm = R + R ln = R ln e + R ln 2
Λ2 NA Λ NA
A m e2
= R ln
Λ2 NA
Vm e5�2
3D
Sm = R ln
Λ3 NA
∆S cond. = S m
2D
− Sm
3D
A m e2 Vm e5�2
= R ln − R ln
Λ2 NA Λ3 NA
Am Λ
= R ln
Vm e1�2
Solutions to exercises
E��F.�(a) �e equilibrium constant for this dissociation reaction is computed using [��F.��–
���]
g I2 kTΛ 3I2
K= e−hc D̃ 0 �k T
g I2 p−○ qIR2 qIV2 Λ 6I
Λ = h�(2πmkT)1�2
6.6261 × 10−34 J s
Λ I2 =
[2π(253.8×1.6605 × 10−27 kg)×(1.3806 × 10−23 J K−1 )×(1000 K)]1�2
= 3.46... × 10−12 m
Λ I = 4.90... × 10−12 m
kT
qR =
2hc B̃
(1.3806 × 10−23 J K−1 ) × (1000 K)
= = 9.31... × 103
2(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (0.0373 cm−1 )
�e dissociation energy is computed from the well depth (the conversion factor
from eV to cm−1 from inside the front cover is used), using the energy of the
ground state of the harmonic oscillator ε̃ 0 = 12 ν̃
D̃ 0 = D̃ e − ε̃ 0
8065.5 cm−1 1
= (1.5422 eV) × − 2 × (214.36 cm−1 )
1 eV
= 1.23... × 104 cm−1
hc D̃ 0 (6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (1.23... × 104 cm−1 )
=
kT (1.3806 × 10−23 J K−1 ) × (1000 K)
= 17.7...
e−hc D̃ 0 �k T
= e−17.7 ... = 1.96... × 10−8
478 12 STATISTICAL THERMODYNAMICS
E��F.�(a) �e Gibbs energy is computed from the partition function using [��F.�–���],
G(T) = G(0) − nRT ln q�N. As usual, the partition function is factored into
separate contributions from translation, rotation and so on. �e factor of 1�N
is usually taken with the translational contribution, so that, for example, the
rotational contribution to the Gibbs energy is −nRT ln q R , or −RT ln q R for the
molar quantity.
For a centro-symmetric linear molecule the rotational partition function in the
high-temperature limit is given by [��B.��a–���]
kT
qR =
2hc B̃
(1.3806 × 10−23 J K−1 ) × (298 K)
=
2(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (0.3902 cm−1 )
= 265.3...
R
Gm = −RT ln q R
= −(8.3145 J K−1 mol−1 ) × (298 K) × ln(265.3...) = −13.83 kJ mol−1
−1
q V1 = �1 − e−hc β ν̃ 1 �
−1
= �1 − e−(4.82 ...×10 � = 1.00...
−3
cm)×(1388.2 cm−1 )
−1
q V2 = �1 − e−(4.82 ...×10 � = 1.00...
−3
cm)×(2349.2 cm−1 )
−1
q V3 = �1 − e−(4.82 ...×10 � = 1.04...
−3
cm)×(667.4 cm−1 )
Hence
V
Gm = −RT ln q V
= −(8.3145 J K−1 mol−1 ) × (298 K) × ln(1.08...) = −0.204 kJ mol−1
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 479
E��F.�(a) �e Gibbs energy is computed from the partition function using [��F.�–���],
G(T) = G(0) − nRT ln q�N. As usual, the partition function is factored into
separate contributions from translation, rotation and so on. �e factor of 1�N
is usually taken with the translational contribution, therefore the electronic
contribution to the Gibbs energy is −nRT ln q E , or −RT ln q E for the molar
quantity.
�e electronic partition function of this two-level system is
q E = g 0 + g 1 e−β hc ν̃
where g 0 and g 1 are the degeneracies of the ground and �rst excited state,
respectively. For an electronic term the degeneracy is 2J + 1, hence g 0 = 2 × 32 +
1 = 4 and g 1 = 2 × 12 + 1 = 2. With the data given
βhc ν̃ = hc ν̃�k × T −1
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 ) × (881 cm−1 )
= × T −1
1.3806 × 10−23 J K−1
= (1.26... × 103 K) × T −1
at ��� K
E
Gm = −RT ln q E
at ��� K
E
Gm = −(8.3145 J K −1
mol ) × (900 K) × ln(4.48...) = −11.2 kJ mol−1
−1
Solutions to problems
P��F.� �e equilibrium constant for this reaction is given by [��F.��b–���]
−
○
qCHD q −○
K= −
○
3 ,m DCl,m −∆ r E 0 �RT
e
qCD q −○
4 ,m HCl,m
12 � �
1�2
qCHD3 qDCl B̃ 3CD4 B̃ HCl
� � =
qCD4 qHCl rot 3 � Ã CHD3 B̃ CHD3 �
2 B̃ DCl
(2.63)3
1�2
12 10.59
= � � = 6.23...
3 (2.63) × (3.28)2 5.445
�e term ∆ r E 0 is computed as
−∆ r E 0 −N A hc(−1053 cm−1 )
=
RT RT
= −(6.0221 × 1023 mol−1 ) × (6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
(−1053 cm−1 )
× = (1515 K)�T
(8.3145 J K−1 mol−1 ) × T
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 481
800
600
K
400
200
P��F.� In the absence of a magnetic �eld the ground state of the I atom, with term
symbol 2 P3�2 , has a degeneracy given by (2J + 1) = (2 × 32 + 1) = 4. When a
magnetic �eld is applied this level splits into four states characterised by M J =
+ 32 , + 12 , − 12 , − 32 . �e energy of these states is given, by analogy with [��A.��c–
���], by E M J = g µ B BM J , where the g-value is given as 43 , µ B is the Bohr mag-
neton, and B is the applied magnetic �eld. �e energies are therefore E±3�2 =
±2µ B B, E±1�2 = ± 23 µ B B, giving the partition function as
q E = [1 + 2x + 2x 2 ] + [1 + (2�3)x + (2�9)x 2 ]
+ [1 − 2x + 2x 2 ] + [1 − (2�3)x + (2�9)x 2 ]
= 4 + (40�9)x 2 = 4 �1 + (10�9)(µ B B�kT)2 �
=
K(0) � 4 �
�is is a very strong magnetic �eld which at present can only be generated by
special techniques and only then for a very short times.
P��F.� �e standard molar Gibbs energy is computed from the partition function us-
ing [��F.�b–���], G m−
○
(T) = G m−
○
(0) − RT ln qm−○ �N A . As usual, the partition
function is factored into separate contributions from translation, rotation and
so on. �e factor of 1�N A is usually taken with the translational contribution.
�e standard molar translational partition function is given by qm−○ = Vm−○ �Λ 3 =
RT�p−○ Λ 3 . Taking the mass of Cl� O� as 2(35.45+16.00) = 102.9 m u , Λ is given
by [��B.�–���]
Λ = h�(2πmkT)1�2
6.6261 × 10−34 J s
=
[2π(102.9×1.6605 × 10−27 kg)×(1.3806 × 10−23 J K−1 )×(200 K)]1�2
= 1.21... × 10−11 m
RT
qm−○ �N A = −○ 3
p Λ NA
(8.3145 J K−1 mol−1 ) × (200 K)
=
(105 N m−2 ) × (1.21... × 10−11 m)3 × (6.0221 × 1023 mol−1 )
= 1.53... × 107
q R = (1�σ)(kT�h)3�2 (π�ABC)1�2
(1.3806 × 10−23 J K−1 ) × (200 K)
3�2
= 12 � �
6.6261 × 10−34 J s
1�2
π
� �
(1.31094 × 1010 Hz) × (2.4098 × 109 Hz) × (2.1397 × 109 Hz)
= 2.89... × 104
�e partition functions for the other normal modes evaluate to 1.02..., 1.12...,
1.66..., 1.00..., 1.05... in order of the given modes. �e overall vibrational par-
tition function is the product of these individual contributions: q V = 2.03....
�e overall partition function is the product of these contributions from the
di�erent modes, therefore
−
○
Gm (200)−G m
−
○
(0) = − RT ln qm−○ �N A
= −(8.3145 J K−1 mol−1 ) × (200 K)
×ln �(1.53... × 107 )×(2.89... × 104 )×(2.03...)�
= −45.8 kJ mol−1
1 dq 1 d
U = −N = −N −βε
� gJ e J
q dβ q dβ J
1
= N � g J ε J e−βε J
q J
dU dU
CV = = −kβ 2
dT dβ
�
d �1 �
�
= −N kβ 2 � � g J ε J e−βε J �
dβ �
� J
q �
�
� −1 dq �
� 1 �
= −N kβ 2 � 2 � g J ε J e−βε J − � g J ε 2J e−βε J �
� q dβ J �
� q J �
� �
�
2� 1 � −βε J ′ � � −βε J � 1 �
2 −βε J �
= −N kβ � 2 � g J ′ ε J ′ e � gJ εJ e − � gJ εJ e �
� q � J′ �� J � q J �
� �
484 12 STATISTICAL THERMODYNAMICS
�e numerator and denominator of the �nal term in the bracket are both mul-
tiplied by q, and then a factor of 1�q 2 is taken outside the bracket to give
� �
�� �
CV =
−N kβ 2 � � g J ′ ε J ′ e−βε J′ � �� g J ε J e−βε J � − q � g J ε 2 e−βε J �
�� �� J � �
� J′ �
J
q2
� J �
�
2��
−N kβ � −βε � � −βε �
= � g J ′ ε J ′ e J′ � g J ε J e J
q2 � �
� J′ � � �
� J
�
� �� ��
− � g J ′ e−βε J′ � g J ε 2J e−βε J �
� J′ �� J ���
�
−N kβ 2 � �
� �
CV = �� g J g J ′ ε J ε J ′ e−β(ε J +ε J′ ) − � g J g J ′ ε 2J e−β(ε J +ε J′ ) �
� J, J ′ �
q2 � J, J ′
�
Taking a hint from the �nal result, consider the double sum
�(ε J − ε J ′ ) g J g J ′ e
2 −β(ε J +ε J ′ )
J, J ′
�e �rst two sums only di�er by swapping the indices J and J ′ , so they are in
fact identical. Hence the last line may be written
= 2 � ε 2J g J g J ′ e−β(ε J +ε J′ ) − 2 � ε J ε J ′ g J g J ′ e−β(ε J +ε J′ )
J, J ′ J, J ′
Apart from an overall sign and a factor of 12 , these two terms are the same as
those in the bracket in the expression for C V above, hence
−N kβ 2 � �
� �
CV = �� g J g J ′ ε J ε J ′ e−β(ε J +ε J′ ) − � g J g J ′ ε 2J e−β(ε J +ε J′ ) �
� J, J ′ �
q2 � J, J ′ �
N kβ 2
= �(ε J − ε J ′ ) g J g J ′ e
2 −β(ε J +ε J ′ )
2q 2 J, J ′
1 1
C V ,m �R = (hc B̃)2
k2 T 2 2q 2
× �[J(J + 1) − J ′ (J ′ + 1)]2 (2J + 1)(2J ′ + 1) e−θ [J(J+1)+J ′ (J ′ +1)]�T
R
J, J ′
2
θR 1
=� �
T 2q 2
× �[J(J + 1) − J ′ (J ′ + 1)]2 (2J + 1)(2J ′ + 1) e−θ [J(J+1)+J ′ (J ′ +1)]�T
R
J, J ′
�is expression is used to generate the curves in Fig. ��.�� for particular pairs of
values of J and J ′ , that is just one term from the double sum. However, the term
for J = 0, J ′ = 1 is identical to that for J = 1, J ′ = 0, so the curves plotted in the
�gure are twice the value for the particular combination of J and J ′ indicated.
�is double sum is not a particularly e�cient method for computing the heat
capacity, but it can be evaluated using mathematical so�ware to give the curve
also shown in Fig. ��.��. For a plot up to T�θ R = 5 if is su�cient to consider
contributions from levels with J ≤ 10; this makes the calculation more tractable.
J = 0, J ′ = 1
1.0 J = 0, J ′ = 2
J = 1, J ′ = 2
J = 1, J ′ = 3
C V ,m �R
J = 0, J ′ = 3
0.5 total
0.0
0 1 2 3 4 5
T�θ R
Figure 12.19
I��.� (a) In the high-temperature limit, the rotational partition function of an asym-
metric rotor is given by [��B.��–���], q R = (1�σ)(kT�hc)3�2 (π�ÃB̃C̃)1�2 ,
where σ is the symmetry number. �e point group for ethene is D 2h
486 12 STATISTICAL THERMODYNAMICS
1 kT 3�2 π 1�2
qR = � � � �
4 hc ÃB̃C̃
(1.3806 × 10−23 J K−1 ) × (298.15 K)
3�2
1
= � �
4 (6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
1�2
π
� �
(4.828 cm−1 ) × (1.0012 cm−1 ) × (0.8282 cm−1 )
= ���.�
(b) Pyridine belongs to the point group C 2v which contains the rotational
operations (E, C 2 ); therefore σ = 2.
1 kT 3�2 π 1�2
qR = � � � �
2 hc ÃB̃C̃
(1.3806 × 10−23 J K−1 ) × (298.15 K)
3�2
1
= � �
2 (6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
1�2
π
� �
(0.2014 cm ) × (0.1936 cm−1 ) × (0.0987 cm−1 )
−1
= 4.26 × 104
13 Molecules in motion
Solutions to exercises
E��A.�(a) �e rate of e�usion, r is given by [��A.��–���], r = pA 0 N A �(2πMRT)1�2 ; this
rate is the number of molecules escaping through the hole in a particular period
of time, divided by that time. �e mass loss ∆m in period ∆t is therefore ∆m =
∆t pA 0 N A �(2πMRT)1�2 × m, where m is the mass of a molecule. �is mass is
written m = M�N A and so it follows ∆m = ∆t pA 0 M 1�2 �(2πRT)1�2 . �is is
rearranged to give an expression for p
�erefore
42 s M N2 1�2 52 2
=� � hence M A = (28.02 g mol−1 ) � � = ��.� g mol−1
52 s MA 42
dp kT pA 0 N A RTA 0 A 0 RT 1�2
=− = − × p = − � � ×p
dt V (2πMRT)1�2 V (2πMRT)1�2 V 2πM
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
α
�e minus sign is needed because the pressure falls with time. �is di�eren-
tial equation is separable and can be integrated between p = p i and p = p f ,
corresponding to t = 0 and t = t.
pf t
� (1�p) dp = � −α dt hence ln(p f �p i ) = −αt
pi 0
V 2πM 1�2
t = − ln(p f �p i )�α = ln(p i �p f ) � �
A 0 RT
8.0 × 104 Pa (3.0 m3 ) 2π × (3.200 × 10−2 kg mol−1 )
1�2
= ln � � � �
7.0 × 10 Pa [π(10 m) ] (8.3145 J K−1 mol−1 ) × (298 K)
4 −4 2
m = M�N A .
1�2
p pN A
N = Z w At = At = At
(2πmkT) 1�2 (2πMkT)1�2
(90 Pa) × (6.0221 × 1023 mol−1 )1�2 × [(2.5 × 3.0) × 10−6 m2 ] × (15 s)
=
[2π × (0.03995 kg mol−1 ) × (1.3806 × 10−23 J K−1 ) × (500 K)]1�2
= 1.9 × 1020 collisions
1 kT 8RT 1�2
D= � �
3 σ p πM
(1.3806 × 10−23 J K−1 )×(293.15 K)
=
3 × (3.6 × 10−19 m2 ) × (p�Pa)
8×(8.3145 J K−1 mol−1 )×(293.15 K)
1�2
� �
π×(0.03995 kg mol−1 )
1
= (1.477... m2 s−1 ) ×
p�Pa
C V ,m 8RT 1�2
σ= � �
3κN A πM
�e value of C p,m is given in the Resource section; C V ,m is found using C p,m −
C V ,m = R for a perfect gas.
∆T
r = J z A = −κA
∆z
[(−15) − (28)] K
= −(2.4 × 10−2 J K−1 m−1 s−1 ) × (1.0 m2 ) ×
0.010 m
= 103 J s−1 = 103 W
Hence a heater of power 103 W is required to make good the loss of heat.
where 1 J = 1 kg m2 s−2 has been used to arrive at the units on the �nal line. Us-
ing this expression the following table is drawn up (recall that 10−7 kg m−1 s−1 =
1 µP)
Recalling that 10−7 kg m−1 s−1 = 1 µP, the cross section is computed as
1
σ=
3 × (2.98 × 10−5 kg m−1 s−1 ) × (6.0221 × 1023 mol−1 )
8 × (8.3145 J K−1 mol−1 ) × (0.02018 kg mol−1 ) × (273 K)
1�2
� �
π
= �.��� nm2
∆t pA 0 M 1�2
∆m =
(2πRT)1�2
(7200 s) × (0.835 Pa) × π × ( 12 × 2.50 × 10−3 m)2 × (0.260 kg mol−1 )1�2
=
[2 × π × (8.3145 J K−1 mol−1 ) × (400 K)]1�2
= 1.04 × 10−4 kg = ��� mg
Solutions to problems
P��A.� In the solution to Exercise E��A.��(a) it is shown that
P��A.� In the solution to Exercise E��A.�(a) it is shown that the di�usion constant is
given by
1 kT 8RT 1�2
D= � �
3 σ p πM
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 493
For a gas at ambient temperature and pressure a typical value for the di�usion
coe�cient is D = 1.5 × 10−5 m2 s−1 , and a typical value for the thermal con-
ductivity is κ = 0.025 J K−1 m−1 s−1 . �e di�usion constant is much higher in
interstellar space when compared to ambient conditions because in interstellar
space the much higher temperature results in a higher mean speed, and the
much lower pressure results in a longer mean free path. Molecules move more
quickly and experience fewer collisions, resulting in more rapid di�usion.
Because κ ∝ N D and D ∝ 1�N , the value of the thermal conductivity is
una�ected by the change in number density in going from ambient pressure to
interstellar conditions. �e higher thermal conductivity in the latter is therefore
attributable to the higher mean speed.
�e kinetic theory of gases assumes that the rate of atomic collisions is very
high such that thermal equilibrium is established quickly. However, at such
a dilute concentration, the timescales on which particles exchange energy by
collision make this assumption questionable. In fact, atoms are more likely to
interact with photons from stellar radiation than with other atoms.
Solutions to exercises
E��B.�(a) �e ion molar conductivity λ is given in terms of the mobility u by [��B.��–���],
λ = zuF, where z is the charge number of the ion (unsigned) and F is Faraday’s
constant; it follows that u = λ�zF. Note that 1 S = 1 C V−1 s−1 .
3.87 mS m2 mol−1
u Li+ = = 4.01×10−5 mS m2 C−1 = 4.01 × 10−8 m2 V−1 s−1
(1)(96485 C mol−1 )
5.01 mS m2 mol−1
u Na+ = = 5.19×10−5 mS m2 C−1 = 5.19 × 10−8 m2 V−1 s−1
(1)(96485 C mol−1 )
7.35 mS m2 mol−1
u K+ = = 7.62×10−5 mS m2 C−1 = 7.62 × 10−8 m2 V−1 s−1
(1)(96485 C mol−1 )
∆� 25.0 V
s = uE = u = (7.92 × 10−8 m2 V−1 s−1 ) ×
l 7.00 × 10−3 m
= 2.83 × 10 m s = ��� µm s
−4 −1 −1
cient D.
Ea 1 1
ln η 1 − ln η 2 = (ln η 0 + E a �RT1 ) − (ln η 0 + E a �RT2 ) = � − �
R T1 T2
ln (η 1 �η 2 )
Ea = R
(T1−1 − T2−1 )
ln [(1.002 cP)�(0.7975 cP)]
= (8.3145 J K−1 mol−1 )
[(293.15 K)−1 − (303.15 K)−1 ]
= ��.� kJ mol−1
E��B.�(a) According to the law of independent migration of ions, the limiting molar
conductivity Λ○m of an electrolyte is given by the sum of the limiting molar
conductivities λ i of the ions present, [��B.�–���], Λ○m = ν+ λ+ + ν− λ− ; in this
expression ν+ and ν− are the numbers of cations and anions provided by each
formula unit of electrolyte. For each of the given electrolytes it follows that
Λ○AgI = λ Ag+ + λ I−
= (Λ○AgNO3 − λ NO3 − ) + (Λ○NaI − λ Na+ ) = Λ○AgNO3 + Λ○NaI − Λ○NaNO3
= (13.34 + 12.69 − 12.16) mS m2 mol−1 = ��.�� mS m2 mol−1
Solutions to problems
P��B.� �e temperature dependence of the viscosity η is given by [��B.�–���], η =
η 0 eE a �RT , where E a is the activation energy . Taking the natural logarithm gives
ln η = ln η 0 +E a �RT. A plot of ln η against (1�T) therefore has slope E a �R; such
a plot is shown in Fig. ��.�.
496 13 MOLECULES IN MOTION
0.0
−0.5
ln (η�cP)
−1.0
−1.5
0.0028 0.0030 0.0032 0.0034 0.0036
1�(T�K)
Figure 13.1
E a = R × (slope)
= (8.3145 J K−1 mol−1 )(1.2207 × 103 K) = ��.�� kJ mol−1
κ�S m−1 c�mol dm−3 Λm �mS m2 mol−1 c 1�2 �(mol dm−3 )1�2
13.1 1.334 9.82 1.155
13.9 1.432 9.71 1.197
14.7 1.529 9.61 1.237
15.6 1.672 9.33 1.293
16.4 1.725 9.51 1.313
9.8
Λm �mS m2 mol−1
9.6
9.4
12.5
Λm �mS m2 mol−1
12.0
11.5
11.0
0.00 0.05 0.10 0.15 0.20 0.25
−3 1�2
c 1�2
�(mol dm )
Figure 13.3
�is is substantially larger than the �.� nm van der Waals radius of a Buck-
minsterfullerene (C�� ) molecule because the anion attracts a considerable hy-
dration shell through the London dispersion attraction to the nonpolar solvent
molecules and through the ion-induced dipole interaction. �e Stokes radius
re�ects the larger e�ective radius of the combined anion and its solvation shell.
(c A �c −○ ) (c B �c −○ ) c A c B (αc AB )2 α 2 c AB
K= = = =
c�c −○ cc −○ (1 − α)c AB c −○ (1 − α)c −○
(b) As the solution becomes more dilute, the degree of dissociation increases
and, in the limit of in�nite dilution, α = 1. �is is a consequence of the
form of K derived in part (a): because K is constant, a decrease in c AB
requires an increase in α towards �.
�e concentration of ions in the solution scales directly with α, therefore
the conductivity, and hence the molar conductivity, will be proportional
to α: Λ m ∝ α. At in�nite dilution the molar conductivity takes the value
Λ m,l and α = 1, therefore α = Λm �Λm,l .
(c) Substitution of this expression for α into the equilibrium expression gives
α 2 c AB Λm2
c AB
K= =
(1 − α)c −
○ Λm,l [1 − (Λm �Λm,l )]c −○
2
α2 Λm2 α 2 Λm,1
2
Λm
= 2 hence =
(1 − α) Λm,1 (1 − Λm
Λ m,l
) (1 − α)Λm (1 − ΛΛm )
m,l
13C Diffusion
Answers to discussion questions
D��C.� See the text following [��C.�–���].
500 13 MOLECULES IN MOTION
Solutions to exercises
E��C.�(a) �e Einstein–Smoluchowski equation [��C.��–���], D = d 2 �2τ, relates the dif-
fusion coe�cient D to the jump distance d and time τ required for a jump.
Approximating the jump length as the molecular diameter, then d ≈ 2a where
a is the e�ective molecular radius. �is is estimated using the Stokes–Einstein
equation [��C.�b–���], D = kT�6πηa, to give 2a = 2kT�6πηD.
Combining these expressions and using the value for viscosity of benzene from
the Resource section gives
2 2
d2 1 2kT 1 kT
τ= = � � = � �
2D 2D 6πηD 18D 3 πη
(1.3806 × 10−23 J K−1 ) × (298 K)
2
1
= � �
18 × (2.13 × 10 m s )
−9 2 −1 3
π × (0.601 × 10−3 kg m−1 s−1 )
= 2.73 × 10−11 s = 27.3 ps
A(πDt)1�2
where c(x, t) is the concentration at time t and distance x from the layer, and
n 0 is the amount in moles in the layer of area A placed at x = 0. If the mass of
sucrose is m, then n 0 = m�M, where M is the molar mass (���.�� g mol−1 ).
m
c(x, t) = e−x �4D t
2
MA(πDt) 1�2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 501
c(10 cm, 10 s) = (9.12... × 102 mol dm−3 ) × (10)−1�2 × e−4.79 ...×10 �(10)
5
RT ∂c
F =− � �
c ∂x T , p
RT αRT
F =− (−αc 0 ) =
c 0 − αc 0 x 1 − αx
RT ∂c
F =− � �
c ∂x T , p
RT
F =− (−2αc 0 xe−α x ) = 2αxRT
2
c 0 e−α x 2
Solutions to problems
P��C.� �ermodynamic force, F, is given by [��C.�b–���].
RT ∂c
F =− � �
c ∂x T , p
where c is the concentration.For a linear gradation of intensity, that is concen-
tration, down the tube
dc�dx = ∆c�∆x = [(0.050 − 0.100) × 103 mol m−3 ]�(0.10 m) = 500 mol m−4
2axRTe−ax
2
RT 2axRT
F =− (2ac xe−ax 2
) = − =
c 0 (1 − e−ax ) (1 − e−ax 2 ) (1 − e ax 2 )
2 0
2
�e �nal step involves multiplying top and bottom of the fraction by e ax . For
thermodynamic force for a = 0.10 cm−2 = 1000 m−2 and T = 298 K is
(5.0 MN mol−1 ) × (x�m)
F=
(1 − e1000×(x�m)2 )
(8.2 × 10−18 N molecule−1 ) × (x�m)
=
(1 − e1000×(x�m)2 )
2 000
F�kN mol−1
−2 000
A plot of the thermodynamic force per mole against x is shown in Fig. ��.�. It
demonstrates that the force is directed such that mass is pushed by the ther-
modynamic force toward the centre of the tube to where the concentration is
lowest. A negative force pushes mass toward the le� (x > 0) and a positive
force pushes mass toward the right (x < 0).
At x = 0 the gradient of the concentration is zero, so the thermodynamic force
is also zero. However, as x approaches zero the modulus of the thermodynamic
force increases without limit on account of the concentration becoming smaller
and smaller.
P��C.� �e generalised di�usion equation is [��C.�–���], where c is concentration, t
is time, D is the di�usion coe�cient and x is displacement.
∂c ∂2 c
=D 2
∂t ∂x
504 13 MOLECULES IN MOTION
A(πDt) 1�2
LHS
∂c n 0 e−x �4D t
2
x2 1
= � − �
∂t A(πDt) 1�2 4Dt 2 2t
RHS
� n e−x 2 �4D t −x � n 0 e−x �4D t
� 0 �
2
∂2 c ∂ x 2 1
D 2 =D � � �� = �� � − �
� A(πDt)1�2 2Dt �
D
∂x ∂x � � A(πDt) 2Dt 2Dt
1�2
n 0 e−x �4D t
2
x2 1
= � − �
A(πDt)1�2 4Dt 2 2t
As required the LHS = the RHS, hence the proposed form of c(x, t) is indeed
a solution to the di�usion equation.
As t → 0 the exponential term e−x �4D t falls o� more and more rapidly, imply-
2
P��C.� As discussed in Section ��C.�(c) on page ��� the probability of �nding a molecule
in an interval dx at distance x from the origin at time t is P(x, t)dx, where
P(x, t) is given by
1
P(x, t) = e−x �4D t
2
(πDt) 1�2
�e mean value of x 4 is found by integrating P(x, t)x 4 dx over the full range
of x, which in this case is � to ∞
∞ 1 ∞
�x 4 � = � x 4 P(x) dx = � x 4 e−x �4D t dx
2
0 (πDt) 1�2 0
1
= × 3 (4Dt)2 × (4πDt)1�2 = 12D 2 t 2
(πDt)1�2 8
where to go to the �nal line Integral G.� is used with k = 1�4Dt. Hence,
�x 4 �1�4 = (12D 2 t 2 )1�4 .
A similar calculation is used to �nd �x 2 �
∞ 1 ∞
�x 2 � = � x 2 P(x) dx = � x 2 e−x �4D t dx
2
0 (πDt) 1�2 0
1
= × 1 π 1�2 (4Dt)3�2 = 2Dt
(πDt)1�2 4
where to go to the �nal line Integral G.� is used with k = 1�4Dt. Hence,
�x 2 �1�2 = (2Dt)1�2 .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 505
P��C.� �e probability of being n steps from the origin is P(nd) = N!�(N−N R )!N R !2 N
where N R is the number of steps taken to the right and N is the total number
of steps. Note that n = N R − N L and N = N R + N L , where N L is the number of
steps taken to the le�.
N−n
NR = N − NL = NL + n hence NL =
2
N+n
NL = N − NR = NR − n hence NR =
2
N! N!
therefore P(nd) = = N+n
[N − ( N−n
2
)]! � N−n
2
�! 2 N � 2
�! � N−n
2
�! 2 N
N!
Pexact (6d) =
� N+6
2
�! � N−6
2
�! 2 N
�is is the ‘exact’ value of the probability according to the random walk model.
In the limit of large N the probability density of being at distance x and time t
is given by [��C.��–���]
2τ 1�2 −x 2 τ�2td 2
P(x, t) = � � e
πt
For the present case the value of x is taken as nd, and t�τ is taken as N because
the time to take N steps is N τ. With these substitutions
2 1�2 −n 2 �2N
Plim (nd, N) = � � e
πN
�e following table compares the exact values of the probability with those
predicted for large N. �e discrepancy between the two values falls to less than
�.�% when N is greater than about ��.
506 13 MOLECULES IN MOTION
∂c ∂2 c
=D 2
∂t ∂x
�e proposed solution is
where ξ(x, t) = x�(4Dt)1�2 and erf(ξ) = 2π−1�2 ∫0 e−y dy. Note that, by
ξ 2
=− =− � e �
(4πDt)1�2 ∂x (4πDt)1�2 2Dt
x(c s − c 0 )
= ex �4D t
2
(4πD)1�2 (t)3�2
�erefore the LHS is equal to the RHS and the proposed form of c(x, t)
does indeed satisfy the di�usion equation, as well as the initial and bound-
ary conditions.
(b) Di�usion through aveoli sites (about � cell thick) of oxygen and carbon
dioxide between lungs and blood capillaries (also about � cell thick) oc-
curs through about 75 µm (the diameter of a red blood cell). �us, the
range 0 ≤ x ≤ 100 µm is reasonable for concentration pro�les for the
di�usion of oxygen into water. Given the maximum distance, the longest
time is estimated using [��C.��–���].
2
πxmax π(0.1 × 10−3 m)2
tmax ≈ = = 3.74 s
4D 4(2.10 × 10−9 m2 s−1 )
�e plots shown in Fig. ��.� are with c 0 = 0, c s = 2.9 × 10−4 mol dm−3 ,
and D = 2.10 × 10−9 m2 s−1 .
508 13 MOLECULES IN MOTION
3.0
t = 0.05 s
t = 0.25 s
t = 0.75 s
104 c�(mol dm−3 )
2.0 t=2s
1.0
0.0
0.00 0.02 0.04 0.06 0.08 0.10
x�mm
Figure 13.5
14 Chemical kinetics
Solutions to exercises
E��A.�(a) For a homogeneous reaction in a constant volume system the rate of reaction
is given by [��A.�b–���], υ = (1�ν J )d[J]�dt, which is rearranged to d[J]�dt =
ν J υ. In these expressions ν J is the stoichiometric number of species J, which
is negative for reactants and positive for products. For this reaction ν A = −1,
ν B = −2, ν C = +3 and ν D = +1.
For A d[A]�dt = ν A υ = (−1) × (2.7 mol dm−3 s−1 ) = −2.7 mol dm−3 s−1
For B d[B]�dt = ν B υ = (−2) × (2.7 mol dm−3 s−1 ) = −5.4 mol dm−3 s−1
For C d[C]�dt = ν C υ = (+3) × (2.7 mol dm−3 s−1 ) = +8.1 mol dm−3 s−1
For D d[D]�dt = ν D υ = (+1) × (2.7 mol dm−3 s−1 ) = +2.7 mol dm−3 s−1
E��A.�(a) For a homogeneous reaction in a constant volume system the rate of reaction
is given by [��A.�b–���], υ = (1�ν J )d[J]�dt, where ν J is the stoichiometric
number of species J which is negative for reactants and positive for products.
For species C, which has ν C = +2, this gives
1 d[C] 1
υ= = × (2.7 mol dm−3 s−1 ) = 1.35... mol dm−3 s−1
ν C dt +2
= �.� mol dm−3 s−1
510 14 CHEMICAL KINETICS
For A d[A]�dt = ν A υ = (−2) × (1.35... mol dm−3 s−1 ) = −2.70... mol dm−3 s−1
For B d[B]�dt = ν B υ = (−1) × (1.35... mol dm−3 s−1 ) = −1.35... mol dm−3 s−1
For D d[D]�dt = ν D υ = (+3) × (1.35... mol dm−3 s−1 ) = +4.05... mol dm−3 s−1
E��A.�(a) As explained in Section ��A.�(b) on page ��� the units of k r are always such as
to convert the product of concentrations, each raised to the appropriate power,
into a rate expressed as a change in concentration divided by time. In this
case the rate is given in mol dm−3 s−1 , so if the concentrations are expressed
in mol dm−3 the units of k r will be dm3 mol−1 s−1 because
[A] [B]
��� � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � �� ��� � � � � � � � � � � � �� � � � � � � � � � � � � ��� � � � � � � � � � � � �� � � � � � � � � � � � �
kr
(dm3 mol−1 s−1 ) × (mol dm−3 ) × (mol dm−3 ) = mol dm−3 s−1
d[C] d[A]
= ν C υ = 3k r [A][B] = ν A υ = −k r [A][B]
dt dt
1 d[C] 1
υ= = k r [A][B][C] = 1
k [A][B][C]
ν J dt +2 2 r
As explained in Section ��A.�(b) on page ��� the units of k r are always such as
to convert the product of concentrations, each raised to the appropriate power,
into a rate expressed as a change in concentration divided by time. In this
case the rate is given in mol dm−3 s−1 , so if the concentrations are expressed
in mol dm−3 the units of k r will be dm6 mol−2 s−1 because
(dm6 mol−2 s−1 ) × (mol dm−3 ) × (mol dm−3 ) × (mol dm−3 ) = mol dm−3 s−1
E��A.�(a) As explained in Section ��A.�(b) on page ��� the units of k r are always such as
to convert the product of concentrations, each raised to the appropriate power,
into a rate expressed as a change in concentration divided by time.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 511
(dm3 mol−1 s−1 ) × (mol dm−3 ) × (mol dm−3 ) = mol dm−3 s−1
A third-order reaction expressed with concentrations in moles per cubic
decimetre is one with a rate law such as v = k r [A][B][C]. �e units of k r
will then be dm6 mol−2 s−1 because
[A] [B] [C]
��� � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � �� ��� � � � � � � � � � � � �� � � � � � � � � � � � � ��� � � � � � � � � � � � �� � � � � � � � � � � � � ��� � � � � � � � � � � � �� � � � � � � � � � � � �
kr
(dm6 mol−2 s−1 )×(mol dm−3 )×(mol dm−3 )×(mol dm−3 ) = mol dm−3 s−1
(ii) If the rate laws are expressed with pressures in kilopascals then a second-
order reaction is one with a rate law such as υ = k r p A p B and a third-order
reaction is one with a rate law such as υ = k r p A p B p C . If the rate is given in
kPa s−1 then the units of k r will be kPa−1 s−1 and kPa−2 s−1 respectively.
��� � � � � � � � � � � �� � � � � � � � � � � �� � �
kr pA pB
��� � � � � � � � � � � �� � � � � � � � � � � �� � � �
kr pA pB pC
(iii) An overall order can be assigned only if all of the individual orders can
be assigned. Consequently the reaction can only be assigned an overall
order if k r2 � k r3 [B]1�2 or k r2 � k r3 [B]1�2 . �e overall order in these
two cases is +2 and + 32 .
E��A.�(a) �e gaseous species is denoted A and the order with respect to A as a. �e rate
law expressed in terms of partial pressure is then υ = k r p Aa . Taking (common)
logarithms gives
where the properties of logarithms log(x y) = log x + log y and log x a = a log x
are used.
�is expression implies that a graph of log υ against log p A will be a straight line
of slope a, from which the order can be determined. However, because there
are only two data points a graph is not necessary so an alternative approach is
used.
If the initial partial pressure of the compound is p A,0 then the partial pressure
when a fraction f has reacted, so that a fraction 1 − f remains, is (1 − f )p A,0 .
Data are given for two points, f 1 = 0.100 and f 2 = 0.200. Denoting the rates
at these points by υ 1 and υ 2 and using the expression log υ = log k r + a log p A
from above gives the equations
log υ 1 = log k r + a log [(1 − f 1 )p A,0 ] log υ 2 = log k r + a log [(1 − f 2 )p A,0 ]
Hence
υ1 (1 − f 1 )p A,0 1 − f1
log � � = a log � � = a log � �
υ2 (1 − f 2 )p A,0 1 − f2
where the property of logarithms log x −log y = log(x�y) is used and the factor
of p A,0 is cancelled.
Rearranging for a gives
E��A.�(a) Assuming perfect gas behaviour the total pressure is proportional to the total
amount in moles of gas present, provided that the temperature is constant and
the volume of the container is �xed. �e reaction � ICl(g) + H� (g) → I� (g) +
� HCl(g) involves the same number of gas molecules on both sides of the re-
action arrow and therefore the total amount in moles of gas present does not
change as the reaction proceeds. Consequently there is no change in the total
pressure during the reaction. �is means that the composition of the reaction
mixture cannot be determined by measuring the total pressure in this case.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 513
E��A.�(a) �e stoichiometry of the reaction shows that one mole of Br� is formed for
every two moles of NO formed. �erefore the rate of formation of Br� is half
the rate of formation of NO.
d[Br2 ] d[NO]
= 1
2
× = 1
2
× (0.24 mmol dm−3 s−1 ) = �.�� mmol dm−3 s−1
dt dt
Solutions to problems
P��A.� �e rate law is assumed to take the form υ 0 = k r [C6 H12 O6 ] a where υ 0 is
the initial rate and a is the order with respect to glucose. Taking (common)
logarithms gives
where the properties of logarithms log(x y) = log x + log y and log x a = a log x
are used.
�is expression implies that a graph of log υ 0 against log[C6 H12 O6 ] will be a
straight line of slope a and intercept log k r . �e data are plotted in Fig. ��.�.
1.4
log(υ 0 �mol dm−3 s−1 )
1.2
1.0
0.8
0.6
−3.0 −2.8 −2.6 −2.4
−3
log([C6 H12 O6 ]�mol dm )
Figure 14.1
log(υ 0 �mol dm−3 s−1 ) = 1.00 × log([CH6 H12 O6 ]�mol dm−3 ) + 3.692
514 14 CHEMICAL KINETICS
(a) Identifying the order a with the slope gives a = 1.00; that is, the reaction
is �rst order in glucose.
(b) �e intercept at log([C6 H12 O6 ]�mol dm−3 ) = 0 is log(υ 0 �mol dm−3 s−1 ) =
3.692, which corresponds to υ 0 = 4.92×103 mol dm−3 s−1 when [C6 H12 O6 ]
= 1 mol dm−3 . Because a = 1, the rate law is υ 0 = k r [C6 H12 O6 ]1 , which
is rearranged to give
P��A.� (a) Experiments � and � both have the same initial H� concentration, but
experiment � has an ICl concentration twice that of experiment �. Because
the rate of experiment � is also twice that of experiment �, it follows that
the rate is proportional to [ICl] and hence that the reaction is �rst order
in ICl.
Experiments � and � both have the same initial ICl concentration, but
experiment � has an H� concentration three times that of experiment
�. Because the rate of experiment � is approximately three times that of
experiment �, it follows that the rate is proportional to [H2 ] and hence
that the reaction is �rst order in H� .
�erefore the rate law is υ = k r [ICl][H2 ] .
(b) �e rate law υ = k r [ICl][H2 ] implies that a graph of υ 0 against [ICl][H2 ]
should be a straight line of slope k r and intercept zero. �e data are plotted
in Fig. ��.�.
20
10
0
0 2 4 6 8 10 12 14
−6
[ICl]0 [H2 ]0 �(10
−6
mol dm )
2
Figure 14.2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 515
υ 0 �mol dm−3 s−1 = 0.163 × {[ICl]0 [H2 ]0 �(mol2 dm−6 )} + 6.19 × 10−9
υ 0 = k r [ICl][H2 ]
= (0.163... dm3 mol−1 s−1 ) × (4.7 × 10−3 mol dm−3 )
× (2.7 × 10−3 mol dm−3 ) = 2.1 × 10−6 mol dm−3 s−1
which has the form of a �rst-order rate law. Because the true rate law has been
forced into �rst-order form by assuming that the concentration of B is constant,
it is called a pseudo�rst-order rate law. �e dependence of the rate on the
concentration of each of the reactants may be found by isolating them in turn
(by having all the other substances present in large excess), and so constructing
the overall rate law.
In the method of initial rates, which is o�en used in conjunction with the
isolation method, the rate is measured at the beginning of the reaction for
several di�erent initial concentrations of reactants. Suppose that the rate law
for a reaction with A isolated is υ = k r [A] a ; then its initial rate, υ 0 , is given by
the initial values of the concentration of A and is written υ 0 = k r [A]0a . Taking
logarithms gives
log υ 0 = log k r + a log[A]0
516 14 CHEMICAL KINETICS
For a series of initial concentrations, a plot of the logarithms of the initial rates
against the logarithms of the initial concentrations of A should be a straight
lime with slope a.
�e method of initial rates might not reveal the full rate law, for the prod-
ucts may participate in the reaction and a�ect the rate. For example, products
participate in the synthesis of HBr, where the full rate law depends on the
concentration of HBr. To avoid this di�culty, the rate law should be �tted to the
data throughout the reaction. �e �tting may be done, in simple cases at least,
by using a proposed rate law to predict the concentration of any component at
any time, and comparing it with the data.
Because rate laws are di�erential equations, they must be integrated in order to
�nd the concentrations as a function of time. Even the most complex rate laws
may be integrated numerically. However, in a number of simple cases analytical
solutions are easily obtained and prove to be very useful. �ese are summarized
in Table ��B.� on page ���. Experimental data can be tested against an assumed
rate law by manipulating the integrated rate law into a form which will give a
straight line plot. If the data do indeed fall on a good straight line, then the data
are consistent with the assumed rate law.
Solutions to exercises
E��B.�(a) Using [��A.�b–���], υ = (1�ν J )(d[J]dt), the rate of the reaction 2A → P is
υ = − 12 (d[A]�dt). Combining this with the rate law υ = k r [A]2 gives
1 d[A] d[A]
− = k r [A]2 hence = −2k r [A]2
2 dt dt
�is is essentially the same as [��B.�a–���], d[A]�dt = −k r [A]2 except with
k r replaced by 2k r . �e integrated rate law is therefore essentially the same as
that for [��B.�a–���], that is, [��B.�b–���] 1�[A] − 1�[A]0 = k r t, except with
k r replaced by 2k r . Hence for the reaction in question
1 1
− = 2k r t
[A] [A]0
Rearranging for t gives
1 1 1
t= � − �
2k r [A] [A]0
1 1 1
= −1 −1 × � −3 − �
2 × (4.30 × 10 dm mol s )
−4 3
0.010 mol dm 0.210 mol dm−3
= 1.1 × 105 s or �.� days
[B]�[B]0
ln = ([B]0 − [A]0 ) k r t
[A]�[A]0
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 517
(i) In � hour the concentration of B falls from 0.060 mol dm−3 to 0.030 mol dm−3 ,
so the change in the concentration of B in this time period is −0.030 mol dm−3 .
It follows from the reaction stoichiometry that the concentration of A
must fall by the same amount, so the concentration of A a�er � hour is
[A]0
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
[A] = (0.080 mol dm−3 ) −(0.030 mol dm−3 ) = 0.050 mol dm−3
1 [B]�[B]0
kr = ln
([B]0 − [A]0 ) t [A]�[A]0
1
=
��0.060 mol dm � − �0.080 mol dm−3 �� × (3600 s)
−3
= 3.09... × 10−3 dm3 mol−1 s−1 = 3.1 × 10−3 dm3 mol−1 s−1
(ii) �e half-life of a particular reactant is the time taken for the concentration
of that reactant to fall to half its initial value. �e half-life of B is � hour
because it is given in the question that a�er � hour the concentration of
B had fallen from 0.060 mol dm−3 to 0.030 mol dm−3 , half the original
value.
�e initial concentration of A is 0.080 mol dm−3 so the half-life is the
time at which the concentration of A has dropped by �.��� mol dm−3 to
�.��� mol dm−3 . It follows from the stoichiometry of the reaction that the
concentration of B must also fall by �.��� mol dm−3 during this period,
so the concentration of B will be
[B] = 0.060 mol dm−3 − 0.040 mol dm−3 = 0.020 mol dm−3
1 1 [B]�[B]0
t= ln � �
k r ([B]0 − [A]0 ) [A]�[A]0
1 1
= ×
3.09... × 10−3 dm3 mol−1 s−1 (0.060 mol dm−3 ) − (0.080 mol dm−3 )
� �0.020 mol dm � � �0.060 mol dm � �
−3 −3
× ln
� �0.040 mol dm−3 � � �0.080 mol dm−3 � �
= 6.5 × 103 s or �.� hours
E��B.�(a) (i) �e integrated rate law for a zeroth-order reaction is given by [��B.�–���],
[A] = [A]0 − k r t where in this case A is NH� . If concentrations are
518 14 CHEMICAL KINETICS
(ii) When all the ammonia has been consumed, p NH3 = 0. Rearranging the
rate law for t gives
E��B.�(a) �e fact that the two half-lives are not the same establishes that the reaction is
not �rst-order because, as explained in Section ��B.� on page ���, a �rst-order
reaction has a constant half-life. For orders n ≠ 1 the half-life is given by [��B.�–
���], t 1�2 = (2n−1 − 1)�[(n − 1)k r [A]0n−1 ]. Denoting the two measurements by
t 1�2,i and t 1�2,ii and expressing concentration in terms of partial pressure gives
the two equations
2n−1 − 1 2n−1 − 1
t 1�2,i = t 1�2,ii =
(n − 1)k r p A,i
n−1
(n − 1)k r p A,ii
n−1
E��B.�(a) For the reaction 2N2 O5 (g) → 4NO2 (g) + O2 (g) the rate, as given by [��A.�b–
���], υ = (1�ν J )(d[J]�dt), is
1 dp N2 O5
υ=
−2 dt
where concentrations are expressed in terms of partial pressures. It is given that
the reaction is �rst-order in N� O� , so υ = k r p N2 O5 . Combining this with the
above expression for υ gives
1 dp N2 O5 dp N2 O5
= k r p N2 O5 hence = −2k r p N2 O5
−2 dt dt
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 519
�is has the same form, except with 2k r instead of k r , as [��B.�a–���], (d[A]�dt) =
−k r [A], for which it is shown in Section ��B.� on page ��� that the half-life and
the integrated rate law are
ln 2
t 1�2 = [A] = [A]0 e−k r t
kr
�e expressions for the reaction in question are analogous, but with k r replaced
by 2k r .
ln 2
t 1�2 = p N2 O5 = (p N2 O5 ,0 )e−2k r t
2k r
�e half-life is
ln 2 ln 2
t 1�2 = = = 1.03 × 104 s
2k r 2 × 3.38 × 10−5 s−1
�e partial pressures at the speci�ed times are calculated from the above inte-
grated form of the rate law. Hence
E��B.�(a) �e reaction is of the form A + B → products. Assuming that it has rate law
υ = k r [A][B], the integrated rate law is given by [��B.�b–���]
[B]�[B]0
ln = ([B]0 − [A]0 )k r t
[A]�[A]0
([B]0 − x)�[B]0
ln = ([B]0 − [A]0 )k r t
([A]0 − x)�[A]0
Hence
([B]0 − x)[A]0
= e([B]0 −[A]0 )k r t
([A]0 − x)[B]0
Rearranging gives
[B]0 [A]0 − x[A]0 = [B]0 [A]0 e([B]0 −[A]0 )k r t − x[B]0 e([B]0 −[A]0 )k r t
Hence
[B]0 [A]0 �e([B]0 −[A]0 )k r t − 1� [B]0 [A]0 (e λ − 1)
x= =
[B]0 e([B]0 −[A]0 )k r t − [A]0 [B]0 e λ − [A]0
where λ = ([B]0 − [A]0 )k r t. Taking A and B as OH− and CH� COOC� H�
respectively, the concentrations at the speci�ed times are
520 14 CHEMICAL KINETICS
For t = 20 s
λ = ([B]0 − [A]0 )k r t
= �(0.110 mol dm−3 ) − (0.060 mol dm−3 )� × (0.11 dm3 mol−1 s−1 ) × (20 s)
= 0.11
[B]0 [A]0 (e λ − 1) (0.110 mol dm−3 ) × (0.060 mol dm−3 ) × (e0.11 − 1)
x= =
[B]0 e λ − [A]0 (0.110 mol dm−3 ) × e0.11 − (0.060 mol dm−3 )
= 0.0122... mol dm−3
Hence the concentration of ester is
[B] = [B]0 − x
= (0.110 mol dm−3 ) − (0.0122... mol dm−3 ) = �.���� mol dm−3
For t = 15 min
λ = ([B]0 − [A]0 )k r t
= �(0.110 mol dm−3 ) − (0.060 mol dm−3 )� × (0.11 dm3 mol−1 s−1 )
× ([15 × 60] s) = 4.95...
[B]0 [A]0 (e λ − 1) (0.110 mol dm−3 ) × (0.060 mol dm−3 ) × (e4.95 ... − 1)
x= =
[B]0 e λ − [A]0 (0.110 mol dm−3 ) × e4.95 ... ) − (0.060 mol dm−3 )
= 0.0598... mol dm−3
Hence the concentration of ester is
[B] = [B]0 − x
= (0.110 mol dm−3 ) − (0.0598... mol dm−3 ) = �.���� mol dm−3
Solutions to problems
P��B.� �e concentration of B is given in the question as
[B] = n[A]0 (1 − e−k r t ) hence [B]�[A]0 = n(1 − e−k r t )
�e concentration of A for a �rst-order reaction is given by [��B.�b–���],
[A] = [A]0 e−k r t hence [A]�[A]0 = e−k r t
�ese expressions are plotted in Fig. ��.�
P��B.� �e �rst task is to convert the masses of urea into concentrations of ammonium
cyanate A. Because the only fate of the ammonium cyanate is to be converted
into urea, the mass of ammonium cyanate m A remaining at any given time
is equal to the original mass of ammonium cyanate minus the mass of urea,
m A = m A,0 − m urea . In this case m urea = 22.9 g. Dividing by the molar mass
of the ammonium cyanate, M A = 60.0616 g mol−1 , gives the amount of A in
moles, and division by the volume of the solution then gives the concentration
in mol dm−3 .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 521
2.0 B, n = 2
[A]�[A]0 or [B]�[A]0
1.0 B, n = 1
B, n = 1
2
A
0.0
0 1 2 3 4 5
kr t
Figure 14.3
1 1 1 1 1
kr t = � − � hence = (n−1)k r t+
n − 1 ([A]0 − [P]) n−1 [A]0
n−1
[A] n−1 [A]0n−1
where to obtain the second expression the relation [P] = [A]0 − [A] is substi-
tuted and the equation rearranged. �is expression implies that if the reaction
has order n ≥ 2 a plot of 1�[A]n−1 against t will be a straight line of slope
(n − 1)k r .
�e data are plotted assuming zeroth-, �rst-, second-, and third-order in Fig. ��.�
and using the data in the table below. �e second-order plot shows a good
straight line, while the other three plots show the data lying on distinct curves.
It is therefore concluded that the reaction is second-order .
522 14 CHEMICAL KINETICS
ln([A]�[A]0 )
−0.5
0.3
0.2
−1.0
0.1
−1.5
0.0
0 4 000 8 000 0 4 000 8 000
t�s t�s
150
second-order third-order
[A]−1 �dm3 mol−1
10 100
5 50
0
0
0 4 000 8 000 0 4 000 8 000
t�s t�s
Figure 14.4
Identifying the slope with (n − 1)k r as discussed above, and noting that n = 2
for a second-order reaction, gives k r = 9.95 × 10−4 dm3 mol−1 s−1 .
�e concentration of ammonium cyanate le� a�er 300 min, which is (300min)×
(60s�1 min) = 18000 s, is calculated using the integrated rate law for a second-
order reaction [��B.�b–���], [A] = [A]0 �(1 + k r t[A]0 )
(0.381... mol dm−3 )
[A] =
1 + (9.95... × 10−4 dm3 mol−1 s−1
)
× (18000 s) × (0.381... mol dm−3 )
= 0.0487... mol dm−3
Multiplication by the volume gives the amount in moles, and multiplication of
this by the molar mass gives the mass of A in g.
m A = MV [A] = (60.0616 g mol−1 )×(1.00 dm)×(0.0487... mol dm−3 ) = �.� g
P��B.� �e order is determined by testing the �t of the data to integrated rate law
expressions. A zeroth-order reaction of the form A → P has an integrated rate
law given by [��B.�–���], [A] = [A]0 −k r t, so if the reaction is zeroth-order then
a plot of [A] against t will be a straight line of slope −k r . In this case, A is the
organic nitrile. On the other hand, a �rst-order reaction has an integrated rate
law given by [��B.�b–���], ln([A]�[A]0 ) = −k r t, so if the reaction is �rst-order
then a plot of ln [A]�[A]0 against t will be a straight line of slope −k r . Finally,
if the order is n ≥ 2 the integrated rate law is given in Table ��B.� on page ���
as
1 1 1 1 1
kr t = � − � hence = (n−1)k r t+
n − 1 ([A]0 − [P]) n−1 [A]0
n−1
[A] n−1 [A]0n−1
where to obtain the second expression the relation [P] = [A]0 − [A] is substi-
tuted and the equation rearranged. �is expression implies that if the reaction
has order n ≥ 2 a plot of 1�[A]n−1 against t will be a straight line of slope
(n − 1)k r .
�e data are plotted assuming zeroth-, �rst-, second-, and third-order in Fig. ��.�.
�e second-order plot shows the best �t to a straight line, so it is concluded
that the reaction is likely to be second-order . However, the �rst-order and
third-order plots also give a reasonable �t to a straight line, so experimental
data over a wider range of concentrations would be needed to establish the
order with greater con�dence.
ln([A]�[A]0 )
1.0 −0.5
0.5 −1.0
0 5 10 0 5 10
3 3
t�10 s t�10 s
1.0
1
0.5
0
0 5 10 0 5 10
3 3
t�10 s t�10 s
Figure 14.5
Identifying the slope with (n − 1)k r as discussed above and noting that n = 2
for a second-order reaction gives k r = 7.33 × 10−5 dm3 mol−1 s−1 .
P��B.� �e order is determined by testing the �t of the data to integrated rate law
expressions. A �rst-order reaction has an integrated rate law given by [��B.�b–
���], ln([A]�[A]0 ) = −k r t, or ln[A] − ln[A]0 = −k r t, so if the reaction is
�rst-order then a plot of ln [A] against t will be a straight line of slope −k r .
On the other hand, a second-order reaction has an integrated rate law given
by [��B.�b–���], 1�[A] − 1�[A]0 = k r t, which implies that if the reaction is
second-order then a plot of 1�[A] against t will be a straight line of slope k r .
�e data are plotted in Fig. ��.�. �e �rst-order plot shows a good straight line
while in the second-order plot the data lie on a curve. It is therefore concluded
that the reaction is �rst-order .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 525
7
�rst-order second-order
6 0.04
1�(c�ng cm−3 )
ln(c�ng cm−3 )
5
0.02
4
3
0.00
0 200 400 0 200 400
t�min t�min
Figure 14.6
Identifying the slope with −k r as discussed above gives the �rst-order rate con-
stant as k r = 7.65 × 10−3 min−1 .
�e half-life of a �rst-order reaction is given by [��B.�–���], t 1�2 = ln 2�k r .
ln 2 ln 2
t 1�2 = = = 91 min
kr 7.65 × 10−3 min−1
P��B.� �e units of the rate constants show that both reactions are �rst-order, so their
rate equations are assumed to be
d[CH4 ] d[CH2 CO]
υ1 = = k 1 [CH3 COOH] and υ2 = = k 2 [CH3 COOH]
dt dt
where [��A.�b–���], υ = (1�ν J )(d[J]�dt), is used to express the rates in terms
of rate of formation of CH� and CH� CO. �e ratio of the rate of ketene forma-
tion to the total rate of product formation is therefore
P��B.�� �e �rst task is to calculate the concentrations of the reactant A at each time.
�e stoichiometry of the reaction 2A → B means that the initial concentration
of A is twice the �nal concentration of B, [A]0 = 2[B]∞ . In addition, the
amount of A that has reacted at any given time is equal to twice the amount
of B that has been formed. It follows that
��� � � � � � � � � � � � � � � � � � � � � �
A that has reacted
Substituting [A]0 = 2[B]∞ from above gives [A] = 2([B]0 − [B]); this expres-
sion is used to calculate the concentration of [A] at each of the times.
�e order is determined by testing the �t of the data to integrated rate law
expressions. If the rate law is υ = k r [A]n , where n is the order to be deter-
mined, expressing υ in terms of the rate of change of concentration of [A] using
[��A.�b–���], υ = (1�ν J )(d[J]�dt) gives
1 d[A] d[A]
υ= = k r [A]n hence = −2k r [A]n
−2 dt dt
Integrated rate laws are given in Table ��B.� on page ���, but care is needed
because these are for reactions of the form A → P but here the reaction is 2A →
B.
For n = 0, Table ��B.� on page ��� shows that a reaction A → P with rate law
υ = d[P]�dt = k r has integrated rate law A = A0 − k r t. To adapt this expression
for the reaction in question, the rate law for the reaction in the table is �rst
written as d[A]�dt = −k r using d[P]�dt = −d[A]�dt for a reaction of the form
A → P. �is rate law matches that found above, d[A]�dt = −2k r [A]n , for n = 0
except that k r is replaced by 2k r . �e integrated rate law will therefore be the
same except with k r replaced by 2k r , that is, [A] = [A]0 − 2k r t. �is expression
implies that if the reaction is zeroth-order a plot of [A] against t will give a
straight line of slope −2k r .
Similarly Table ��B.� on page ��� gives the integrated rate law for a �rst-order
reaction A → P with rate law υ = d[P]�dt = k r [A] as ln([A]0 �[A]) = k r t,
equivalent to [��B.�b–���], ln([A]�[A]0 ) = −k r t. By the same reasoning as
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 527
above the integrated rate law for the reaction will therefore be ln([A]�[A]0 ) =
−2k r t, implying that a plot of ln([A]�[A]0 ) against t will give a straight line of
slope −2k r .
Finally, if the order is n ≥ 2 the integrated rate law for a reaction A → P with
rate law υ = d[P]�dt = k r [A]n is given in Table ��B.� on page ��� as
1 1 1 1 1
kr t = � − � hence = (n−1)k r t+
n − 1 ([A]0 − [P])n−1 [A]0n−1 [A]n−1 [A]0n−1
where to obtain the second expression the relation [P] = [A]0 − [A] is substi-
tuted and the equation rearranged. Adapting this expression for the reaction
in question gives 1�[A]n−1 = 2(n − 1)k r t + 1�[A]0n−1 . �is expression implies
that if the reaction has order n ≥ 2 a plot of 1�[A]n−1 against t will be a straight
line of slope 2(n − 1)k r .
�e data are plotted assuming zeroth-, �rst-, second-, and third-order in Fig. ��.�.
�e �rst-order plot shows a good �t to a straight line, while the other plots are
curved, so it is concluded that the reaction is �rst-order .
P��B.�� �e order is determined by �tting the data to integrated rate laws. Table ��B.�
on page ��� gives integrated rate laws for the reaction A → P with rate law
υ = d[P]�dt = k r [A]n as
n=0 [A] = [A]0 − k r t hence k r = ([A]0 − [A]) �t
[A]0 1 [A]0
n=1 k r t = ln hence kr = ln
[A] t [A]
1 1 1
n≥2 kr t = � − �
n − 1 ([A]0 − [P]) n−1 [A]0n−1
1 1 1
hence kr = � − �
(n − 1)t [A]n−1 [A]0n−1
528 14 CHEMICAL KINETICS
ln([A]�[A]0 )
−0.5
0.4
−1.0
0.2
−1.5
0.0
0 20 40 0 20 40
t�min t�min
40
6 second-order third-order
[A]−1 �dm3 mol−1
0 0
0 20 40 0 20 40
t�min t�min
Figure 14.7
P��B.�� A reaction of the form A → P that is nth order in A has rate law υ = k r [A]n .
Combining this with [��A.�b–���], υ = (1�ν J )(d[J]�dt) gives
1 d[A]
= k r [A]n hence − [A]−n d[A] = k r dt
−1 dt
Initially, at t = 0, the concentration of A is [A]0 , and at a later time t it is [A].
�ese are used as the limits of the integration to give
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
Integral A.�
[A] [A] t
t 1
� −[A]−n d[A] = � k r dt hence [A]−(n−1) � = k r t�
[A]0 0 n−1 [A]0 0
1 1 1
hence � − � = kr t (for n ≠ 1)
n − 1 [A]n−1 [A]0n−1
�is integrated rate law is equivalent to that given in Table ��B.� on page ���.
1 � 1 1 � 1 2n−1 1
t 1�2 = − = � − �
(n − 1)k r � � [A]0 �
1 n−1
[A]0 � (n − 1)k r [A]0
n−1 n−1
[A]0n−1
2
2n−1 − 1
=
(n − 1)k r [A]0n−1
(b) �e time taken for the concentration of a substance to fall to one-third its
initial value is denoted t 1�3 . �us, at t = t 1�3 , [A] = 13 [A]0
1 � 1 1 � 1 3n−1 1
t 1�3 = − = � − �
(n − 1)k r � � 1 [A]0 �n−1 [A]0n−1 � (n − 1)k r [A]0n−1 [A]0n−1
3
3n−1 − 1
=
(n − 1)k r [A]0n−1
d[P] dx
= k r ([A]0 − 2x)2 ([B]0 − x) hence = k r ([A]0 − 2x)2 ([B] − x)
dt dt
530 14 CHEMICAL KINETICS
where to go to the second expression [P] = x is used, which implies that d[P]dt =
dx�dt. �e expression is rearranged and the initial condition x = 0 when t = 0
is applied. �is gives the integrations required as
x 1 t
� dx = � k r dt
0 ([A]0 − 2x) ([B]0 − x)
2 0
x 1 x 1
� dx = � dx
0 ([A]0 − 2x)2 ( [A]0 − x) 0 ([A]0 − 2x)2 × ([A]0 − 2x)
1 1
2 2
x
x 1 1 1
= � 2([A]0 − 2x) dx = ([A]0 − 2x) � =
−3 −2
−
0 2 0 2([A]0 − 2x)2 2[A]20
Combining this with the right-hand side from above gives the integrated
rate law as
1 1
− = kr t
2([A]0 − 2x)2 2[A]20
(b) If [B]0 = [A]0 the le�-hand side is integrated using the method of partial
fractions described in �e chemist’s toolkit �� in Topic ��B. �e integrand
is �rst written as
1 A B C
= + +
([A]0 − 2x) ([A]0 − x) ([A]0 − 2x)
2 2 [A]0 − 2x [A]0 − x
�is expression must be true for all x, so the values of A, B and C are most
conveniently found by substituting particular values of x.
Combining this with the right-hand side integral found above gives the
integrated rate law as
1 1 [A]0 − 2x 1
+ ln − = kr t
[A]0 ([A]0 − 2x) [A]20 [A]0 − x [A]20
Solutions to exercises
E��C.�(a) �e relaxation time in a jump experiment is given by [��C.�a–���], τ = 1�(k r +
k r′ ). �is equation is rearranged for k r′ . It is convenient to convert τ to ms.
1 1
k r′ = − kr = − (12.4 ms−1 ) = 23.8 ms−1
τ 27.6 × 10−3 ms
Solutions to problems
P��C.� �e expression for [A] in [��C.�–���] is di�erentiated
k r′ + k r e−(k r +k r )t
′
d[A]
[A] = [A]0 = −k r [A]0 e−(k r +k r )t
′
hence
k r + k r′ dt
According to [��C.�–���], d[A]�dt = −(k r + k r′ )[A]+ k r′ [A]0 . To verify that the
two expressions for d[A]�dt are the same, the expression for [A] from [��C.�–
���] is substituted into [��C.�–���]
[A]
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
k ′ + k r e−(k r +k r )t
′
d[A]
= −(k r + k r′ ) � r [A]0 � +k r′ [A]0
dt k r + k r′
= −k r′ [A]0 − k r e−(k r +k r )t [A]0 + k r′ [A]0 = −k r [A]0 e−(k r +k r )t
′ ′
�erefore the two expressions for d[A]�dt are the same and so the equation is
satis�ed.
P��C.� (a) �e forward and backward reactions are
d[A] d[A]
A→B = −k r [A] B→A = +k r′ [B]
dt dt
�e overall rate of change of [A] is therefore
d[A]
= −k r [A] + k r′ [B]
dt
�e stoichiometry of the reaction A � B means that the amount of B
present at any time is equal to the initial amount plus the amount of A
that has reacted. Hence [B] = [B]0 + ([A]0 − [A]). �is is substituted
into the above expression to give
d[A]
= −k r [A] + k r′ ([B]0 + [A]0 − [A])
dt
Rearranging and integrating with the initial condition that [A] = [A]0
when t = 0 gives
[A] d[A] t
� = � dt
[A]0 k r′ ([A]0 + [B]0 ) − (k r + k r )[A]
′ 0
[A] =
k r + k r′
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 533
k r′ ([A]0 + [B]0 )
[A]eq =
k r + k r′
k r′ ([A]0 + [B]0 )
[B]eq = [B]0 + [A]0 − [A]eq = [B]0 + [A]0 −
k r + k r′
(k r + k r′ )([B]0 + [A]0 ) − k r′ ([A]0 + [B]0 ) k r′ ([A]0 + [B]0 )
= =
k r + k r′ k r + k r′
1 d[A] d[A]
Forward 2A → A2 υ = k a [A]2 = hence = −2k a [A]2
−2 dt dt
1 d[A] d[A]
Backward A2 → 2A υ = k a′ [A2 ] = hence = 2k a′ [A2 ]
2 dt dt
�e overall rate of change of A is therefore
d[A]
= −2k a [A]2 + 2k a′ [A2 ]
dt
If the deviation of [A] from its new equilibrium value is denoted 2x, so
that [A] = [A]eq + 2x, the stoichiometry of the reaction implies that
[A2 ] = [A2 ]eq − x. �ese are substituted into the above expression to
give
d[A]
= −2k a �[A]eq + 2x� + 2k a′ �[A2 ]eq − x�
2
dt
= −2k a �[A]2eq + 4x[A]eq + 4x 2 � + 2k a′ �[A2 ]eq − x�
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
0 Neglect
In the third line the �rst two terms cancel because at equilibrium the rates
of the forward reaction k a [A]2eq and the backward reaction k a′ [A2 ]eq are
equal. �e last term is neglected because x is assumed to be small. Next,
because [A] = [A]eq + 2x it follows that d[A]�dt = 2 dx�dt. �is is
substituted into the above expression to give
dx dx
2 = − �8k a [A]eq + 2k a′ � x hence = − �4k a [A]eq + k a′ � x
dt dt
Rearranging, and integrating with the condition that x = x 0 when t = 0
gives
dx
x t x
� = � − �4k a [A]eq + k a′ � dt hence ln = − �4k a [A]eq + k a′ � t
x0 x 0 x0
1
hence x = x 0 e−(4k a [A]eq +k a )t = x 0 e−t�τ where = 4k a [A]eq + k a′
′
τ
Squaring both sides of the expression for 1�τ gives
1
= �4k a [A]eq + k a′ � = 16k a2 [A]2eq + 8k a k a′ [A]eq + k a′2
2
τ2
k a′ [A 2 ]eq
��� � � � � � � � � �� � � � � � � � � �
= 16k a �k a [A]2eq � +8k a k a′ [A]eq + k a′2 = 16k a k a′ [A2 ]eq + 8k a k a′ [A]eq + k a′2
[A]tot
��� � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � ��
= 8k a k a′ �2[A2 ]eq + [A]eq � +k a′2 = 8k a k a′ [A]tot + k a′2
In the second line, k a [A]2eq = k a′ [A2 ]eq is used; these quantities are equal
because as explained above the rates of the forward and backward reac-
tions are equal at equilibrium. In the third line, the relationship [A]tot =
[A] + 2[A2 ] is used; this expression is valid at all stages of the reaction
including at equilibrium.
(b) �e result 1�τ 2 = 8k a k a′ [A]tot + k a′2 implies that a plot of 1�τ 2 against
[A]tot should give a straight line of intercept k a′2 and slope 8k a k a′ ; from
these quantities k a′ and k a are determined.
(c) �e data are plotted in Fig. ��.�.
0.20
0.15
τ −2 �ns−2
0.10
0.05
0.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6
−3
[P]�mol dm
Figure 14.8
Identifying the intercept with k a′2 gives k a′2 = 2.87 × 10−4 ns−2 , hence
√
k a′ = 2.87 × 10−4 ns−2 = 0.0169... ns−1 = 1.7 × 107 s−1
Solutions to exercises
E��D.�(a) �e relationship between the values of a rate constant at two di�erent temper-
atures is given by [��D.�–���], ln(k r,2 �k r,1 ) = (E a �R)(1�T1 − 1�T2 ). Hence,
taking T1 = 37 ○ C and T2 = 15 ○ C,
k r,2 Ea 1 1
= exp � � − ��
k r,1 R T1 T2
87 × 103 J mol−1 1 1
= exp � ×� − ��
8.3145 J K−1 mol−1 [37+273.15] K [15+273.15] K
= �.���
�e rate constant therefore drops to about 7.6 % of its original value when the
temperature is lowered for 37 ○ C to 15 ○ C.
E��D.�(a) As explained in Section ��D.�(a) on page ��� the fraction f of collisions that are
su�ciently energetic to be successful is given by the exponential factor e−E a �RT .
Rearranging f = e−E a �RT for T and setting f = 0.10 gives
Ea 50 × 103 J mol−1
T =− =− = 2.6 × 103 K
R ln f (8.3145 J K−1 mol−1 ) × ln 0.10
23 × 103 J mol−1
k r = (8.1 × 10−10 dm3 mol−1 s−1 ) × exp �− �
(8.3145 J K−1 mol−1 ) × (500 K)
= 3.2 × 10−12 dm3 mol−1 s−1
E��D.�(a) �e relationship between the values of a rate constant at two di�erent tempera-
tures is given by [��D.�–���], ln(k r,2 �k r,1 ) = (E a �R)(1�T1 − 1�T2 ). Rearrang-
ing for E a gives
At T1 A = k r eE a �RT1
1.07...×105 J mol−1
= (3.80×10−3 dm3 mol−1 s−1 )×exp
(8.3145 J K−1 mol−1 )×([35+273.15] K)
= 6.62 × 1015 dm3 mol−1 s−1
At T2 A = k r eE a �RT2
1.07...×105 J mol−1
= (2.67×10−2 dm3 mol−1 s−1 )×exp
(8.3145 J K−1 mol−1 )×([50+273.15] K)
= 6.62 × 1015 dm3 mol−1 s−1
E��D.�(a) �e relationship between the values of a rate constant at two di�erent tempera-
tures is given by [��D.�–���], ln(k r,2 �k r,1 ) = (E a �R)(1�T1 − 1�T2 ). Rearrang-
ing for E a , and using k r,2 �k r,1 = 3 because the rate constant triples between the
two temperatures, gives
Solutions to problems
P��D.� �e de�nition of E a in [��D.�–���], E a = RT 2 (d ln k r �dT), is rearranged and
integrated.
Ea d ln k r Ea
= hence � d ln k r = � dT
RT 2 dT RT 2
�e le�-hand side integral is simply ln k r . If E a does not vary with temperature
then the integral on the right is evaluated by taking E a �R outside the integral
to give
Ea 1 Ea
ln k r = � dT = − +c
R T 2 RT
�is is [��D.�–���], ln k r = ln A − E a �RT, once the constant of integration c is
identi�ed with ln A.
−20
15
13
12
0.0035 0.0040 0.0045 0.0050
1�(T�K)
Figure 14.10
D��E.� Suppose that a reactant R can give alternative products P and Q by di�erent
reactions. If the rate constant for the formation of P is greater than that for
forming Q, then to start with more P will be formed. However, as time proceeds
it may be that the reverse reactions from P and Q back to R start to become
signi�cant, and eventually the reactions reach equilibrium. It may be that at
equilibrium the amount of Q exceeds that of P, even though initially the amount
of P exceeded that of Q.
If the relative proportions of the products are determined by the rate at which
they are formed, the reaction is said to be under kinetic control. If the amounts
are determined by the relevant equilibrium constants, the reaction is said to
be under thermodynamic control. �e latter will only occur if the reverse
reactions are signi�cant.
D��E.� �e overall reaction order is the sum of the powers of the concentrations of
all of the substances appearing in the experimental rate law for the reaction;
hence, it is the sum of the individual orders (exponents) associated with a each
reactant. Reaction order is an experimentally determined quantity.
Molecularity is the number of reactant molecules participating in an elemen-
tary reaction. Molecularity has meaning only for an elementary reaction, but
reaction order applies to any reaction. In general, reaction order bears no
necessary relation to the stoichiometry of the reaction, with the exception of
elementary reactions, where the order of the reaction corresponds to the num-
ber of molecules participating in the reaction; that is, to its molecularity. �us
for an elementary reaction, overall order and molecularity are the same and are
determined by the stoichiometry.
Solutions to exercises
E��E.�(a) �e steady-state approximation is applied to the intermediate species O.
d[O]
= k a [O3 ] − k a′ [O2 ][O] − k b [O][O3 ] = 0
dt
k a [O3 ]
(k a′ [O2 ] + k b [O3 ]) [O] = k a [O3 ] hence [O] =
k a′ [O2 ] + k b [O3 ]
�e rate of decomposition of O� is
d[O3 ]
= −k a [O3 ]+k a′ [O2 ][O]−k b [O][O3 ] = −k a [O3 ]+[O]{k a′ [O2 ]−k b [O3 ]}
dt
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 541
If step � is rate limiting, such that k a′ [O2 ][O] � k b [O][O3 ], and hence k a′ [O2 ] �
k b [O3 ], the denominator simpli�es to k a′ [O2 ] and hence
1�2 1�2
d[P] ka ka
= k b [A][B] = k b � ′ � [A2 ][B] = k b � ′ � [A2 ]1�2 [B]
dt ka ka
d[A]
= 2k a [A2 ] − 2k a′ [A]2 − k b [A][B]
dt
In the steady-state approximation this is assumed to be zero
hence [A] =
4k a′
where the positive square root is chosen in order to avoid obtaining a
negative value for [A]. �e rate of formation of P is given by d[P]�dt =
k b [A][B]; substituting the above expression for [A] into this gives
Under certain circumstances this rate law simpli�es. If 16k a′ k a [A2 ]�k b2 [B]2 �
1 then
� 1�2 �
d[P] k b2 [B]2 � �
�−1 + � 16k a k a [A2 ] � �
′
≈ � �
dt 4k a′ � k b [B]
2 2
�
� �
k b2 [B]2 16k a′ k a [A2 ]
1�2 1�2
ka
≈ ×� � = kb � � [A2 ]1�2 [B]
4k a′ k b2 [B]2 k a′
which is the same as the rate law derived in part (i) assuming a pre-
equilibrium. �e condition 16k a′ k a [A2 ]�k b2 [B]2 � 1 corresponds to the
A2 ⇄ A + A steps being much faster than the step involving B and k b ;
this is precisely the situation corresponding to a pre-equilibrium because
the removal of A in the reaction with B is then too slow to a�ect the
maintenance of the equilibrium.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 543
On the other hand, if 16k a′ k a [A2 ]�k b2 [B]2 � 1 then the square root is
approximated by the expansion (1 + x)1�2 ≈ 1 + 12 x to give
Solutions to problems
0.8 ka = 10 s−1
ka = 2 s−1
ka = 0.5 s−1
0.6 ka = 0.2 s−1
[I]�mol dm−3
ka = 0.1 s−1
0.4 ka = 0.02 s−1
0.2
0.0
0 1 2 3 4 5
t�s
Figure 14.11
If k b � k a , the concentration of I remains low and, apart from the initial induc-
tion period, approximately constant during the reaction. �us the steady-state
544 14 CHEMICAL KINETICS
P��E.� It is shown in Example ��E.� on page ��� that for the case of two consecutive
unimolecular reactions the concentration of the intermediate is greatest at a
time given by t max = (ln[k a �k b ])�(k a − k b ). �e half-life of a �rst-order reac-
tion is related to the rate constant according to [��B.�–���], t 1�2 = ln 2�k r . �is
is rearranged to k r = ln 2�t 1�2 and used to substitute for the rate constants in
the expression for t max .
1 ka 1 ln 2�t 1�2,a
t max = ln = ln
k a − k b k b (ln 2�t 1�2,a ) − (ln 2�t 1�2,b ) ln 2�t 1�2,b
1 t 1�2,b
= ln
ln 2 �1�t 1�2,a − 1�t 1�2,b � t 1�2,a
1 33.0 d
Hence t 1�2 = × ln = ��.� d
ln 2 × [1�(22.5 d) − 1�(33.0 d)] 22.5 d
(k b′ + k c )[C]
k b [B] − k b′ [C] − k c [C] = 0 hence [B] =
kb
�e expression for d[B]�dt becomes
where on the second line the expression for [B] derived above is substituted in.
Finally, the rate of formation of D is
d[D] k a k b k c [A]
= k c [C] − k c′ [D] = k c [C] = ′ ′
dt k a k b + k a′ k c + k b k c
where [D] = 0 is used and the expression for [C] is substituted in.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 545
remain constant throughout the reaction. It follows, then, that the degree of
polymerization is given by [��F.��b–���], �N� = 1 + k r t[A]0 . �erefore, the
average molar mass can be controlled by adjusting the initial concentration of
monomer and the length of time that the polymerization is allowed to proceed.
Solutions to exercises
E��F.�(a) �e Michaelis–Menten equation for the rate of an enzyme-catalysed reaction
is given by [��F.��a–���], υ = υ max �(1 + K M �[S]0 ). Rearranging for υ max gives
E��F.�(a) Example ��F.� on page ��� gives the values K M = 10.0 mmol dm−3 and υ max =
0.250 mmol dm−3 s−1 for an enzyme concentration of [E]0 = 2.3 nmol dm−3 .
�e catalytic e�ciency is de�ned in the exercise as k b �K M , and υ max is related to
k b according to [��F.��b–���], υ max = k b [E]0 , hence k b = υ max �[E]0 . �erefore,
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 547
Solutions to problems
P��F.� �e e�ective rate constant k r in the Lindemann–Hinshelwood mechanism is
given by [��F.�–���], 1�k r = k a′ �k a k b + 1�k a [A]. �is expression implies that
a plot of 1�k r against 1�[A] should be a straight line. �e data are plotted in
Fig. ��.��, using pressure as a measure of concentration.
548 14 CHEMICAL KINETICS
3
1�(104 k r �s−1 )
0
0 5 10 15
1�(p�Torr)
Figure 14.12
�e data lie on a curve rather than on a straight line, so it is concluded that the
Lindemann–Hinshelwood mechanism does not �t these data.
P��F.� Each molecule of hydroxyacid has one OH group and one COOH group (A),
so [OH] = [A]. Hence the given rate expression, d[A]�dt = −k r [A]2 [OH],
becomes
d[A] 1
= −k r [A]3 hence − d[A] = k r dt
dt [A]3
Integration of this expression, with the limits that the concentration is [A]0 at
time t = 0 and [A] at some later time t, gives
[A]
[A] 1 t 1
� − = � � = k r t�0
t
d[A] k r dt hence
[A]0 [A]3 0 2[A]2 [A]0
1 1
hence − = 2k r t
[A] 2 [A]20
Rearranging gives
1 1 1 [A]20
= 2k r t + hence [A]2 = =
[A]2 [A]20 2k r t + 1�[A]20 2k r t[A]20 + 1
To go to the �nal expression the top and bottom of the fraction are multiplied by
[A]20 . Taking the square root gives [A] = [A]0 �(2k r t[A]20 + 1)1�2 . As explained
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 549
in Section ��F.�(a) on page ���, the degree of polymerisation �N� is the ratio of
the initial concentration of A, [A]0 , to the concentration of end groups, [A], at
the time of interest. Hence
[A]0 [A]0
�N� = = = (2k r t[A]20 + 1)1�2
[A] [A]0 �(2k r t[A]20 + 1)1�2
�is equation is plotted for �xed υ max with a range of K M values in Fig. ��.��,
and for �xed K M with a range of υ max values in Fig. ��.��.
1.0
K M = 0.01 mol dm−3
0.8
K M = 0.1 mol dm−3
υ�mmol dm−3 s−1
0.6
0.2
υ max = 1 mmol dm−3 s−1
0.0
0.0 0.2 0.4 0.6 0.8 1.0
−3
[S]0 �mol dm
Figure 14.13
[ATP]� υ� 1 1
µmol dm−3 µmol dm−3 s−1 [ATP]�(µmol dm−3 ) υ�(µmol dm−3 s−1 )
0.6 0.81 1.67 1.23
0.8 0.97 1.25 1.03
1.4 1.30 0.71 0.77
2.0 1.47 0.50 0.68
3.0 1.69 0.33 0.59
550 14 CHEMICAL KINETICS
1.0
K M = 0.1 mol dm−3
0.8
υ max = 1 mmol dm−3 s−1
υ�mmol dm−3 s−1
0.6
0.4
υ max = 0.5 mmol dm−3 s−1
0.2
υ max = 0.1 mmol dm−3 s−1
0.0
0.0 0.2 0.4 0.6 0.8 1.0
−3
[S]0 �mol dm
Figure 14.14
1�(υ�µmol dm−3 s−1 )
1.0
0.5
1
υ max = = 2.32... µmol dm−3 s−1 = �.� µmol dm−3 s−1
0.43 µmol dm−3 s−1
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
υ max
�
slope
14G Photochemistry
Answer to discussion question
D��G.� �e time scales of atomic processes are rapid indeed. Note that the times given
here are in some way typical values for times that may vary over two or three
orders of magnitude. For example, vibrational wavenumbers can range from
about ���� cm−1 (for H� ) to ��� cm−1 (for I� ) and even lower, with a corre-
sponding range of associated times. Radiative decay rates of electronic states
can vary even more widely: times associated with phosphorescence can be in
the millisecond and even second range. A large number of time scales for phys-
ical, chemical, and biological processes on the atomic and molecular scale are
reported in Figure � of A. H. Zewail, ‘Femtochemistry: Atomic-Scale Dynamics
of the Chemical Bond’, Journal of Physical Chemistry A, ���, ���� (����).
Radiative decay of excited electronic states can range from about 10−9 s to
10−4 s, and even longer for phosphorescence involving ‘forbidden’ decay paths.
Molecular rotational motion takes place on a scale of 10−12 s to 10−9 s. Molecu-
lar vibrations are faster still, about 10−14 s to 10−12 s. Proton transfer reactions
occur on a timescale of about 10−10 s to 10−9 s, although protons can hop from
molecule to molecule in water even more rapidly (1.5 × 10−12 s).
Harvesting of light during plant photosynthesis involves very fast time scales
of several energy-transfer and electron-transfer steps in photosynthesis. Initial
energy transfer (to a nearby pigment) has a time scale of around 10−13 s to
5 × 10−12 s, with longer-range transfer (to the reaction centre) taking about
10−10 s. Immediate electron transfer is also very fast (about � ps), with ultimate
transfer (leading to oxidation of water and reduction of plastoquinone) taking
from 10−10 s to 10−3 s. �e mean time between collisions in liquids is similar
to vibrational periods, around 10−13 s.
Solutions to exercises
E��G.�(a) �e e�ciency of resonance energy transfer η T is de�ned by [��G.�–���], η T =
1 − � F �� F,0 , and the distance-dependence of the e�ciency is given by [��G.�–
���], η T = R 06 �(R 06 + R 6 ), where R is the donor–acceptor distance and R 0 is a
constant characteristic of the particular donor–acceptor pair.
In this case a decrease of the �uorescence quantum yield by ��% implies that
� F = 0.9� F,0 . Hence the e�ciency is η T = 1 − � F �� F,0 = 1 − 0.9 = 0.1.
Rearranging [��G.�–���] for R, and taking R 0 = 4.9 nm from Table ��G.� on
page ���, gives
1 − ηT 1 − 0.1 1�6
1�6
R = R0 � � = (4.9 nm) × � � = �.� nm
ηT 0.1
gives
N decomposed n decomposed N A (n formed �2)N A
N abs = = =
� � �
In the �nal expression, n formed is the amount in moles of B that is formed. �e
stoichiometry of the reaction A → 2B + C implies that the amount of A that
decomposes is half the amount of B that is formed, n decomposed = n formed �2.
�e quantum yield is 210 mmol einstein−1 , or 0.210 mol mol−1 = 0.210, hence
� F,0 �� F − 1 2−1
[Q] = =
τ0 kQ (6.0 × 10 s)×(3.0 × 108 dm3 mol−1 s−1 )
−9
Solutions to problems
P��G.� �e quantum yield is given by [��G.�a–���], � = N events �N abs where N abs is
the number of photons absorbed and N events is, in this case, the number of
molecules of the absorbing substance that decomposed. �e latter is equal
to n decomposed N A , where n decomposed is the amount in moles of substance that
decomposed.
�e number of photons absorbed is found by noting that the energy transferred
by each photon is given by [�A.�–���], ∆E = hν = hc�λ. �erefore the total
energy absorbed is E abs = N abs hc�λ. �is energy is also given by E abs = f Pt,
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 553
where P is the incident power, t is the time of exposure, and f is the fraction
of incident radiation that is absorbed. In this case f = 1 − 0.257 = 0.743.
Combining these expressions gives
N abs hc f Ptλ
f Pt = hence N abs =
λ hc
�is is substituted into � = N events �N abs , together with N events = n decomposed N A ,
to give
Note that 1 W = 1 J s−1 . �e fact that the quantum yield is greater than �
indicates that each absorbed photon can lead to the decomposition of more
than one molecule of absorbing material.
P��G.� (a) �e concentration of the excited dansyl chloride decays with time accord-
ing to [��G.�a–���], [S∗ ] = [S∗ ]0 e−t�τ 0 , or [S∗ ]�[S∗ ]0 = e−t�τ 0 . �e rate
of �uorescence is given by υ = k F [S∗ ] so the rate of �uorescence, and
hence the �uorescence intensity I F , is proportional to [S∗ ]. �erefore
I F �I 0 = [S∗ ]�[S∗ ]0 , and hence
IF [S∗ ] IF t
= ∗ = e−t�τ 0 hence ln � �=−
I 0 [S ]0 I0 τ0
t�ns I F �I 0 ln(I F �I 0 )
5.0 0.45 −0.799
10.0 0.21 −1.561
15.0 0.11 −2.207
20.0 0.05 −2.996
�e data lie on a good straight line that passes close to the origin. �e
equation of the line is
1 1
− = −0.145 ns−1 hence τ0 = = 6.89... ns = �.� ns
τ0 0.145 ns−1
554 14 CHEMICAL KINETICS
−1
ln(I F �I 0 )
−2
−3
0 5 10 15 20 25
t�ns
Figure 14.16
� F,0 0.70
kF = = = 1.0 × 108 s−1
τ0 6.89... × 10−9 s
[Hg∗ ] [Hg∗ ] t
= e−t�τ hence ln � �=−
[Hg∗ ]0 [Hg∗ ]0 τ
ln(I�I 0 )
−1 −1
−2 −2
0 10 20 0 10 20
t�µs t�µs
Figure 14.17
�F τ
ηT = 1 − =1−
� F,0 τ0
where the second expression comes from the fact that, according to [��G.�–
���], � F,0 = k F τ 0 , the lifetime is proportional to quantum yield. �e e�ciency
of resonance energy transfer in terms of donor–acceptor distance is given by
[��G.�–���], η T = R 06 �(R 06 + R 6 ). Equating the two expressions for η T gives
1�6
τ R6 1 R6 1
1− = 6 0 6 hence = 1 + 6 hence R = R 0 � − 1�
τ0 R0 + R 1 − τ�τ 0 R0 1 − τ�τ 0
d[B] d[B]
A→B = IA B→A = −k r [B]2
dt dt
I a 1�2
k r [B]2 = I a hence [B] = � �
kr
I��.� (a) �e expressions [A] = [A]0 − x and [P] = [P]0 + x are substituted into
the rate law to give
d[A]
υ=− = k r [A][P] = k r ([A]0 − x)([P]0 + x)
dt
�e expression [A] = [A]0 − x implies that d[A]�dt = −dx�dt so the
expression becomes
dx dx
= k r ([A]0 − x)([P]0 + x) hence = k r dt
dt ([A]0 − x)([P0 ] + x)
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 557
1 (−[P]0 − x)[A]0
=− ln � �
(−[P]0 ) − [A]0 ([A]0 − x)(−[P]0 )
1 [A]0 ([P]0 + x)
= ln � �
[A]0 + [P]0 [P]0 ([A]0 − x)
�e right-hand side is k r t, hence the integrated rate law is
1 [A]0 ([P]0 + x)
ln � � = kr t
[A]0 + [P]0 [P]0 ([A]0 − x)
1 [A]0 [P]
ln � � = kr t
[A]0 + [P]0 [P]0 ([A]0 − [P] + [P]0 )
��� � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � �
a
[A]0 [P]
hence ln � � = ([A]0 + [P]0 )k r t
[P]0 ([A]0 − [P] + [P]0 )
��� � � � � � � � ��� � � � � � � � � �
b
(b) �e quantity [P]�[P]0 is plotted against at in Fig. ��.�� for various values
of b.
�e plots are sigmoid in shape because the reaction is initially slow be-
cause only a small amount of P is present. As more product is formed, the
rate of the reaction υ = k r [A][P] increases and the curve becomes steeper,
until the reaction slows down towards the end due to the reactant A being
used up.
�e curves level o� at di�erent values because [P]�[P]0 is being plotted.
In each case the �nal concentration of P is given by the initial concentra-
tion of A, because all the A is eventually converted to P, plus the concen-
tration of P that was present at the start, that is, [P]∞ = [A]0 + [P]0 . �e
558 14 CHEMICAL KINETICS
100
b = 0.01
[P]�[P]0
b = 0.02
50
b = 0.1
0
0 2 4 6 8 10
at
Figure 14.18
�is expression is plotted against t in Fig. ��.�� for various values of b, tak-
ing a = 1 s−1 in each case. �e quantity [P]�[A]0 = 1−e−k r t is also plotted,
taking k r = 1 s−1 . As already noted, the autocatalytic curves are sigmoid,
in contrast to the �rst-order curve which is not. �e autocatalytic curves
with larger b, that is a greater initial amount of P relative to the initial
amount of A, reach their maximum value faster than those with smaller
b. �is is because, if less P is present to begin with, the autocatalytic step
is initially slower and the amount of P present builds up more slowly.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 559
1.0 �rst-order
b = 0.1
b = 0.03
0.0
0 2 4 6 8 10
t�s
Figure 14.19
(c) �e rate law found in part (a), [P]�[P]0 = (1 + b)e at �(1 + be at ), is rear-
ranged to [P] = (1 + b)e at [P]0 �(1 + be at ) and di�erentiated to give an
expression for the rate.
d[P] d (1 + b)e at [P]0
υ= = � �
dt dt 1 + be at
(1 + be at ) × a(1 + b)e at [P]0 − (1 + b)e at [P]0 × abe at
=
(1 + be at )2
a(1 + b)e at [P]0
=
(1 + be at )2
�e maximum rate is found by di�erentiating υ with respect to t and
setting the derivative equal to zero
dυ (1 + be at )2 × (a 2 (1 + b)e at [P]0 ) − a(1 + b)e at [P]0 × 2abe at (1 + be at )
=
dt (1 + be at )4
At the maximum, when dυ�dt = 0, the numerator of this expression is
zero
a 2 (1 + b)(1 + be at )2 e at [P]0 − 2a 2 b(1 + b)(1 + be at )e2at [P]0 = 0
Cancelling of terms followed by rearrangement gives
1 + be at − 2be at = 0 hence be at = 1 hence t = −(1�a) ln b
(d) As in part (a), [A] and [P] are written as [A]0 −x and [P]0 +x respectively.
�e rate law is
d[P]
υ= = k r [A]2 [P] = k r ([A]0 − x)2 ([P]0 + x)
dt
�e expression [P] = [P]0 + x implies that d[P]�dt = dx�dt so the rate
law becomes
dx dx
= k r ([A]0 − x)2 ([P]0 + x) hence = k r dt
dt ([A]0 − x)2 ([P]0 + x)
560 14 CHEMICAL KINETICS
1 A B C
= + +
([A]0 − x) ([P] + x) ([A]0 − x)
2 2 [A] − x [P]0 + x
C−B=0
A + B[A]0 − B[P]0 − 2C[A]0 = 0
A[P]0 + B[A]0 [P]0 + C[A]20 = 1
A − B([A]0 + [P]0 ) = 0
A[P]0 + B([A]0 [P]0 + [A]20 ) = 1
Rearranging gives
1
B([A]20 + 2[A]0 [P]0 + [P]20 ) = 1 hence B=C=
([A]0 + [P]0 )2
1 1
A− × ([A]0 + [P]0 ) = 0 hence A=
([A]0 + [P]0 )2 [A]0 + [P]0
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 561
Hence
x = [A]0 or [A]0 − x = 2([P]0 + x)
Because x = [A]0 − [A], the solution x = [A]0 corresponds to [A] = 0.
From the rate law υ = k r [A]2 [P] this corresponds to v = 0 and therefore to
a minimum rate rather than a maximum. �e maximum rate is therefore
given by the second expression, which is rearranged to yield x = 13 ([A]0 −
2[P]0 ).
�is is then substituted into the integrated rate law from above
1 1 1 1 [A]0 ([P]0 + x)
kr t = � − �+ ln
[A]0 + [P]0 [A]0 − x [A]0 ([A]0 + [P]0 )2 [P]0 ([A]0 − x)
1 1 1
= �2 − �
[A]0 + [P]0 3 [A]0 + 3 [P]0 [A]0
2
1 1 [P]0 [A]0
t= � + + ln �
k r ([A]0 + [P]0 ) 2 [A]0
2 2[P]0
(e) �e rate law is integrated as in part (d). Writing [A] = [A] − x and [P] =
[P]0 + x the rate law is
d[P]
υ= = k r [A][P]2 = k r ([A]0 − x)([P]0 + x)2
dt
�e expression [P] = [P]0 + x implies that d[P]�dt = dx�dt. �erefore
dx x dx t
= k r ([A]0 −x)([P]0 +x)2 hence � = � k dt
dt 0 ([A]0 − x)([P]0 + x)2 0
1 A B C
= + +
([A]0 − x)([P]0 + x)2 [A]0 − x [P]0 + x ([P]0 + x)2
A−B = 0
2A[P]0 + B[A]0 − B[P]0 − C) = 0
A[P]20 + B[A]0 [P]0 + C[A]0 = 1
A([A]0 + [P]0 ) − C = 0
A([P]20 + [A]0 [P]0 ) + C[A]0 = 1
1
A([A]20 + 2[A]0 [P]0 + [P]20 ) = 1 hence A=B=
([A]0 + [P]0 )2
1 1
× ([A]0 + [P]0 ) − C = 0 hence C=
([A]0 + [P]0 )2 ([A]0 + [P]0 )
x 1 1
kr t = � � +
0 ([A]0 + [P]0 ) ([A]0 − x) ([A]0 + [P]0 )2 )([P]0 + x)
2
1
+ � dx
([A]0 + [P]0 )([P]0 + x)2
− ln([A]0 − x) ln([P]0 + x)
x
1
=� + − �
([A]0 + [P]0 ) 2 ([A]0 + [P]0 ) 2 ([A]0 + [P]0 )([P]0 + x) 0
1 [P]0 + x 1
=� ln − �
([A]0 + [P]0 )2 [A]0 − x ([A]0 + [P]0 )([P]0 + x)
1 [P]0 1
−� ln − �
([A]0 + [P]0 ) 2 [A]0 ([A]0 + [P]0 )[P]0
1 [A]0 ([P]0 + x) 1 1 1
= ln + � − �
([A]0 + [P]0 ) 2 [P]0 ([A]0 − x) [A]0 + [P]0 [P]0 [P]0 + x
1 [A]0 [P] 1 1 1
kr t = ln + � − �
([A]0 +[P]0 ) 2 [P]0 ([A]0 +[P]0 −[P]) [A]0 +[P]0 [P]0 [P]
I��.� (a) Because the second step is rate-determining, the �rst step and its reverse
are treated as a pre-equilibrium because the rate of reaction of A− with AH
to form product is assumed to be too slow to a�ect the maintenance of the
pre-equilibrium. As explained in Section ��E.� on page ��� it follows that
k a [BH+ ][A− ] k a [AH][B]
K= = hence [A− ] =
k a′ [AH][B] k a′ [BH+ ]
�e rate formation of product is υ = d[P]�dt = k b [A− ][AH]. �e expres-
sion for [A− ] is substituted into this to give
k a [AH][B]
[A− ] =
k a [BH+ ] + k b [AH]
′
k a [HA][H+ ]
k a [HA][H+ ] = k a′ [HAH+ ] hence [HAH+ ] =
k a′
k a [HA][H+ ] ka kb
v = k b [HAH+ ][B] = k b ′
[B] = [HA][H+ ][B]
ka k a′
�e sum ∑∞ 2 N−1
N=1 N p is evaluated by noting that N p N−1 is the derivative of
N
p
∞ ∞ ∞
d N d ∞
�N p
2 N−1
= � N × N p N−1 = � N × p = � Np
N
�e sum ∑∞
N=1 N p
N−1
was already evaluated above; its value is 1�(1 − p)2 .
Hence
∞
d 1 1+ p
�N p
2 N−1
= �p × �=
N=1 dp (1 − p) 2 (1 − p)3
N=1 dp dp N=1
(1+p)�(1−p)3
��� � � � � � � � � � � � ��� � � � � � � � � � � � �
d ∞
d 1+ p
= (1 − p) �p � N 2 p N−1 � = (1 − p) �p × �
dp N=1 dp (1 − p)3
p2 + 4p + 1 p2 + 4p + 1
= (1 − p) × =
(1 − p)4 (1 − p)4
(b) �e average number of monomers per polymer, that is the chain length,
is given by [��F.��a–���], �N� = 1�(1 − p). �is expression is rearranged
to p = 1 − 1��N�. �is is then substituted into the expression derived in
(b) to give the ratio in terms of chain length.
M 1 (6�N�2 − 6�N� + 1)
=
2�N� − 1
15 Reaction dynamics
D��A.� Reactions between complex molecules might be expected to have strong steric
requirements (small steric factors) as a result of the reaction requiring a par-
ticular orientation and approach of the reacting parts of the molecule: the
more complex the molecules, the smaller the fraction of collisions which are
potentially reactive.
E ∗ s−1
k b (E) = �1 − � kb
E
D��A.� To the extent that real gases deviate from perfect gas behaviour, they do so
because of intermolecular interactions. Interactions tend to be more important
at high pressures, when the size of the molecules themselves is not negligible
compared to the average intermolecular distance (mean free path). Attractive
interactions might enhance a reaction rate compared to the predictions of col-
lision theory, particularly if the parts of the molecules that are attracted to each
other are the reactive sites. Similarly, repulsive interactions might reduce the
frequency of collisions compared to what would be predicted for perfect gases.
570 15 REACTION DYNAMICS
Solutions to exercises
E��A.�(a) �e collision theory expression for the rate constant is given in [��A.�–���].
1�2
8kT
kr = σ NA � � e−E a �RT
πµ
= (0.36 × 10−18 m2 ) × (6.0221 × 1023 mol−1 )
8 × (1.3806 × 10−23 J K−1 ) × (650 K)
1�2
� �
π(3.32 × 10−27 kg)
× e−(171×10 J mol−1 )�[(8.3145 J K−1 mol−1 )×(650 K)]
3
�e units are best resolved by realising that (8kT�πµ)1�2 is a speed, with units
m s−1 . Note that �.�� nm2 is 0.36 × 10−18 m2 .
E��A.�(a) As described in Section ��A.�(b) on page ���, the reactive cross section may be
estimated from the (non-reactive) collision cross sections of A and B: σest =
(σA + σB )2 . �e steric factor is given by the ratio of the experimental
1 1�2 1�2
4
reactive cross section, σexp , to the estimated cross section
E��A.�(a) In the RRK theory the rate constant for the unimolecular decay of an energized
molecule A* is given by [��A.��–���],
E ∗ s−1
k b (E) = �1 − � k b = (1 − x)s−1 k b
E
where x = E ∗ �E. For a non-linear molecule with � atoms there are 3N − 6 =
3 × 5 − 6 = 9 normal modes, so s = 9. �is expression is rearranged for x to give
x = 1 − [k b (E)�k b ]1�(s−1)
= 1 − [3.0 × 10−5 ]1�(9−1) = �.��
E��A.�(a) In the RRK theory the rate constant for the unimolecular decay of an energized
molecule A* is given by [��A.��–���],
k b (E) E ∗ s−1
= �1 − �
kb E
where E ∗ is the minimum energy needed to break the bond, and E is the energy
available from the collision. With the data given
4kT 1�2 2
Z AA = σ � � N A [A]2
πm
where [A] is the molar concentration of the gas. In turn, this is expressed in
terms of the pressure using the perfect gas equation to give [A] = n A �V =
p A �RT.
3131�2 − 3031�2
= 0.0163... = �.�%
3031�2
Similarly the �nal expression for the collision density shows Z AA ∝ p2 T −3�2
which, with p ∝ T, gives Z AA ∝ T 2 T −3�2 ∝ T 1�2 . �is is the same dependence
as z, so the same percentage increase will result.
E��A.�(a) �e collision theory expression for the rate constant is given in [��A.�–���].
In this expression, the factor e−E a �RT is identi�ed as the fraction of collisions f
572 15 REACTION DYNAMICS
having at least kinetic energy E a along the �ight path. For example with E a =
20 kJ mol−1 and T = 350 K
Ea 20 × 103 J mol−1
= = 6.87... f = e−6.87 ... = 1.04 × 10−3
RT (8.3145 J K−1 mol−1 ) × (350 K)
E��A.�(a) �e method for calculating the fractions is shown in the solution to Exercise
E��A.�(a). For E a = 20 kJ mol−1 and T = 350 K it is found that f = 1.03...×10−3
and increasing the temperature to ��� K gives f = 1.25...×10−3 . �e percentage
increase is
(1.25... × 10−3 ) − (1.03... × 10−3 )
100 × = ��%
1.03... × 10−3
Solutions to problems
P��A.� �e collision theory expression for the rate constant is given in [��A.�–���]
1�2
8kT
kr = σ NA �
∗
� e−E a �RT
πµ
A πµ 1�2
σ∗ = � �
N A 8kT
2.4 × 107 m3 mol−1 s−1 π(1.24... × 10−26 kg)
1�2
= � �
6.0221 × 1023 mol−1 8 × (1.3806 × 10−23 J K−1 ) × (298 K)
= 4.34... × 10−20 m2 = 0.043 nm2
�e units are best resolved by realising that (8kT�πµ)1�2 is a speed, with units
m s−1 .
To estimate the collision cross-section assume that d is twice the C–H bond
length and compute σ = πd 2 = π(2 × 154 × 10−12 m)2 = 2.98... × 10−19 m2 . �e
steric factor is P = σ ∗ �σ = (4.34... × 10−20 m2 )�(2.98... × 10−19 ) = �.�� .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 573
P��A.� �e collision theory expression for the rate constant is given in [��A.�–���]
1�2
8kT
kr = σ NA � � e−E a �RT
πµ
�e maximum value for the rate constant is when E a = 0. �e collision cross
section is taken as σ = πd 2 = π(308 × 10−12 m)2 = 2.98... × 10−19 m2 . �e
mass of a CH� radical is 15.03 m u , therefore the reduced mass of the collision
is µ = 12 × 15.03 m u = 1.24... × 10−26 kg
1�2
8kT
kr = σ NA � � = (2.98... × 10−19 m2 ) × (6.0221 × 1023 mol−1 )
πµ
8 × (1.3806 × 10−23 J K−1 ) × (298 K)
1�2
� �
π(1.24... × 10−26 kg)
= 1.64 × 108 mol−1 m3 s−1
�e units are best resolved by realising that (8kT�πµ)1�2 is a speed, with units
m s−1 .
For a second-order reaction the integrated rate law is [��B.�b–���], 1�[CH3 ] −
1�[CH3 ]0 = k r t. Suppose that initially an amount in moles n 0 of C� H� is in-
troduced into the vessel, and that a fraction α dissociates. �e amount of C� H�
remaining is n 0 (1 − α) and the amount of CH� produced is 2n 0 α. �e total
amount of gas is n 0 (1+α), therefore the mole fraction of CH� is 2α�(1+α) and
hence the partial pressure of CH� is 2α p tot �(1 + α). �e molar concentration
corresponding to this pressure is found using the perfect gas law as
n CH3 p CH3 2α p tot
[CH3 ] = = =
V RT RT(1 + α)
With the data given this evaluates as
2 × 0.1 × (100 × 103 Pa)
[CH3 ] = = 7.33... mol m−3
(8.3145 J K−1 mol−1 )×(298 K)×(1 + 0.1)
1
If recombination proceeds to ��%, the amount of CH� remaining is 10
of the
initial. �e time for this to take place is found by solving
10 1
− = kr t
[CH3 ]0 [CH3 ]0
Hence
9 9
t= =
[CH3 ]0 k r (7.33... mol m−3 ) × (1.64 × 108 mol−1 m3 s−1 )
= �.� ns
P��A.� �e collision theory expression for the rate constant, including the steric factor
P, is given in [��A.��–���]
1�2
8kT
k r = Pσ N A � � e−E a �RT
πµ
574 15 REACTION DYNAMICS
�.�� nm2 . No values are given for the ethyl and cyclohexyl radicals, so these will
be approximated by the values for ethene (�.�� nm2 ) and benzene (�.�� nm2 ),
respectively. �e reactive cross sections are therefore
m O2 m C2 H5 32.00 × 29.06
µ= = × (1.6605 × 10−27 kg) = 2.52... × 10−26 kg
m O2 + m C2 H5 32.00 + 29.06
For the O� –C� H�� collision the reduced mass is 3.83... × 10−26 kg.
Taking the activation energy as E a = 0, the steric factor is given by
kr πµ 1�2
P= � �
σ N A 8kT
Solutions to exercises
E��B.�(a) For a di�usion-controlled reaction the rate constant is approximated by [��B.�–
���], k d = 8RT�3η, where η is the viscosity. Recall that 1 P = 10−1 kg m−1 s−1 ,
so that 1 cP = 10−3 kg m−1 s−1 . �erefore the rate constant is
�e initial concentrations are [A] = 0.170 mol dm−3 = 0.170 × 103 mol m−3
and [B] = 0.350 mol dm−3 = 0.350 × 103 mol m−3 . �e initial rate is therefore
d[P]
= k d [A][B]
dt
= (3.04... × 106 m3 mol−1 s−1 )
× (0.170 × 103 mol m−3 ) × (0.350 × 103 mol m−3 )
= 1.81 × 1011 mol m−3 s−1
Solutions to problems
∗
P��B.� To simplify the notation the dependence of [J] and [J] on x and t will not
∗
be written explicitly. �e proposed solution, [��B.�–���], [J] = [J]e−k r t , is
substituted into the right-hand side of [��B.�–���]
∗
∂ 2 [J] ∗ ∂2
D − k r [J] = D 2 [J]e−k r t − k r [J]e−k r t
∂x 2 ∂x
∂ 2 [J] −k r t
=D e − k r [J]e−k r t
∂x 2
�e solution is now substituted into the le�-hand side of [��B.�–���]
∗
∂[J] ∂ ∂[J] −k r t
= [J]e−k r t = e − k r e−k r t [J]
∂t ∂t ∂t
�e le�-and right-hand sides are now set equal
∂ 2 [J] −k r t ∂[J] −k r t
D e − k r [J]e−k r t = e − k r e−k r t [J]
∂x 2 ∂t
�e term k r e−k r t [J] cancels to give
Solutions to exercises
E��C.�(a) �e variation of the rate constant with ionic strength is given by [��C.��–���],
lg k r = lg k r○ + 2Az A z B I 1�2 ; at ��� K and for aqueous solutions A = 0.509. In
the absence of further information assume z A = +1 and z A = −1. Rearranging
for lg k r○ gives
lg k r○ = lg k r − 2Az A z B I 1�2
= lg(12.2 dm6 mol−2 min−1 ) − 2 × (0.509) × (+1) × (−1) × (0.0525)1�2
= 1.31...
� 1�2 �
k r (C–D) ħω(C–H) �
� µ CH �
�
= e−ζ ζ= �1 − � � �
k r (C–H) �
� �
�
� �
2kT µ CD
mC mT 12 × 3.016
µ CT = = m u = 2.41... m u
m C + m T 12 + 3.016
450 N m−1
1�2
ω(C–H) = � � = 5.39... × 1014 s−1
1.54... × 10−27 kg
At ��� K
Raising the temperature will decrease ζ which will have the e�ect of increasing
the ratio k r (C–T)�k r (C–H) and thus moving it closer to �. �at is, the isotope
e�ect will be reduced.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 579
kT RT ∆‡ S�R
A=e e
h p−○
Ap−○ h
hence ∆‡ S = R ln
ekRT 2
= (8.3145 J K−1 mol−1 )
(2.05 × 1010 m3 mol−1 s−1 ) × (105 Pa) × (6.6261 × 10−34 J s)
× ln
e(1.3806 × 10−23 J K−1 ) × (8.3145 J K−1 mol−1 ) × (303 K)2
= −25.3 J K−1 mol−1
kT RT ∆‡ S�R
A=e e
h p−○
Ap−○ h
hence ∆‡ S = R ln
ekRT 2
= (8.3145 J K−1 mol−1 )
(7.78 × 1011 m3 mol−1 s−1 ) × (105 Pa) × (6.6261 × 10−34 J s)
× ln
e(1.3806 × 10−23 J K−1 ) × (8.3145 J K−1 mol−1 ) × (303 K)2
= +4.88... J K−1 mol−1
∆‡ G = ∆‡ H − T∆‡ S
= (+73.4... × 103 J mol−1 ) − (303 K) × (+4.88... J K−1 mol−1 )
= +��.� kJ mol−1
580 15 REACTION DYNAMICS
E��C.�(a) �e rate constant for a bimolecular gas phase reaction is given by [��C.��a–���]
kT RT ∆‡ S�R −E a �RT
k r = e2 e e
h p−○
To use this expression the given rate constant needs to be converted to units
of m3 mol−1 s−1 . From the perfect gas law [J] = n�V = p�RT, therefore to
convert the rate constant from units of pressure−1 to units of concentration−1
units requires multiplication by RT
hp−○ k r hp−○ Ea
∆‡ S = R ln �k r e E a �RT
� = R ln � �+
e2 kRT 2 e2 kRT 2 T
= (8.3145 J K−1 mol−1 )
(0.0220... m3 mol−1 s−1 ) × (6.6261 × 10−34 J s) × (105 Pa)
× ln � �
e2 (1.3806 × 10−23 J K−1 ) × (8.3145 J K−1 mol−1 ) × (338 K)2
58.6 × 103 J mol−1
+
338 K
= −91.2 J K−1 mol−1
E��C.�(a) In Example ��C.� on page ��� the following expression for the rate constant for
a reaction between structureless particles is derived
1�2
8kT
kr = NA � � σ ∗ e−∆E 0 �RT
πµ
d ln k r
E a = RT 2
dT
d �� �
� �
1�2 �
� ∆E 0 �
�
�
�ln � ∗�
8kT
= RT 2 � N � � σ � − �
dT �� � A
� RT � �
� � �
πµ
�
1 ∆E
= RT 2 � + � = 12 RT + ∆E 0
0
2T RT 2
�is is rearranged to give ∆‡ S, noting that for a collision between like molecules
µ = 12 m
� hp−○ �
1�2
8kT
∆ S = R ln N A �
‡
� σ∗
� πµ e3�2 kRT 2 �
= (8.3145 J K−1 mol−1 )
�
� 8 × (1.3806 × 10−23 J K−1 ) × (300 K)
1�2
× ln �
�(6.0221 × 10 23
mol−1
) × � �
� π × 12 × 65 × (1.6605 × 10−27 kg)
�
× (0.35 × 10−18 m2 )
�
(6.6261 × 10−34 J s) × (105 Pa) �
× �
�
e3�2 (1.3806 × 10−23 J K ) × (8.3145 J K mol ) × (300 K)2 �
−1 −1 −1
�
= −74 J K−1 mol−1
E��C.�(a) It is convenient to convert the units of the frequency factor and express it as
A = 4.6 × 109 m3 mol−1 s−1 . �e relationship between E a and ∆‡ H for a bi-
molecular gas-phase reaction is given by [��C.��–���], ∆‡ H = E a − 2RT =
(10.0 × 103 J mol−1 ) − 2 × (8.3145 J K−1 mol−1 ) × (298 K) = +5.04... kJ mol−1 =
+5.0 kJ mol−1 . �e relationship between A and ∆‡ S for a bimolecular gas-
phase reaction is given by [��C.��a–���]
kT RT ∆‡ S�R
A = e2 e
h p−○
Ap−○ h
hence ∆‡ S = R ln 2
e kRT 2
= (8.3145 J K−1 mol−1 )
(4.6 × 109 m3 mol−1 s−1 ) × (105 Pa) × (6.6261 × 10−34 J s)
× ln
e2 (1.3806 × 10−23 J K−1 ) × (8.3145 J K−1 mol−1 ) × (298 K)2
= −45.8... J K−1 mol−1 = −46 J K−1 mol−1
∆‡ G is found by combining the values of ∆‡ H and ∆‡ S in the usual way
∆‡ G = ∆‡ H − T∆‡ S
= (+5.04... × 103 J mol−1 ) − (298 K) × (−45.8... J K−1 mol−1 )
= +�� kJ mol−1
582 15 REACTION DYNAMICS
Solutions to problems
P��C.� It is convenient to convert the units of the frequency factor and express it as
A = 4.07 × 102 m3 mol−1 s−1 . �e relationship between E a and ∆‡ H for a
bimolecular gas-phase reaction is given by [��C.��–���], ∆‡ H = E a − 2RT =
(65.4 × 103 J mol−1 ) − 2 × (8.3145 J K−1 mol−1 ) × (300 K) = +60.4... kJ mol−1 =
+60.4 kJ mol−1 . �e relationship between A and ∆‡ S for a bimolecular gas-
phase reaction is given by [��C.��a–���]
kT RT ∆‡ S�R
A = e2 e
h p−○
Ap−○ h
hence ∆‡ S = R ln 2
e kRT 2
= (8.3145 J K−1 mol−1 )
(4.073 × 102 m3 mol−1 s−1 ) × (105 Pa) × (6.6261 × 10−34 J s)
× ln
e2 (1.3806 × 10−23 J K−1 ) × (8.3145 J K−1 mol−1 ) × (300 K)2
= −1.80... × 102 J K−1 mol−1 = −181 J K−1 mol−1
∆‡ G is found by combining the values of ∆‡ H and ∆‡ S in the usual way
∆‡ G = ∆‡ H − T∆‡ S
= (+60.4... × 103 J mol−1 ) − (300 K) × (−1.80... × 102 J K−1 mol−1 )
= +��� kJ mol−1
�e relationship between ∆U and ∆H is given by [�B.�–��], ∆H = ∆U +
∆ν g RT, where ∆ν g is the change in stoichiometric coe�cients for gaseous species.
In this case ∆ν g = 1 − 2 = −1, hence
∆‡ U = ∆‡ H − ∆ν g RT = ∆‡ H + RT
= (+60.4... × 103 J mol−1 ) + (8.3145 J K−1 mol−1 ) × (300 K)
= +��.� kJ mol−1
3�2
kT h2 2µr 2 kT −∆E 0 �RT
=κ NA � � e
h 2πkT µ ħ2
1�2
8kT
kr = NA � � σ ∗ e−∆E 0 �RT
πµ
To simplify the notation, the overline, double dagger and standard symbols will
be omitted. With such a complex calculation it is best to break it down into
parts and then assemble them to give the �nal result. First, the pre-multiplying
constants (assuming κ = 1)
Next consider the ratio of the translational partition functions: each is given
by q −○ = Vm−○ �Λ 3 = RT�p−○ Λ 3 , with Λ = h�(2πmkT)1�2 . For the purpose of this
approximate calculation it is su�cient to use integer masses
p−○
3�2 3�2
qHD2 h2 m HD2
= � � � �
qH qD2 trans RT 2πkT m H m D2
105 Pa
=
(8.3145 J K mol−1 ) × (400 K)
−1
Next consider the ratio of the vibrational partition functions: each is given by
q = (1 − e−hc ν̃�k T )−1 . �e vibrational frequency of D� is so high that q = 1.
584 15 REACTION DYNAMICS
Each normal mode of the activated complex has ν = 1000 cm−1 n thus the
contribution to the partition function is computed as
where n vib is the number of vibrational normal modes of the activated complex.
To �nd the rotational partition function for D� requires a knowledge of the
rotational constant, which is computed from the given bond length of H� . �e
e�ective mass for D� is 1 m u thus
h
B̃ =
8π 2 cµR 2
6.6261 × 10−34 J s
=
8π 2 (2.9979 × 1010 cm s−1 ) × (1 × 1.6605 × 10−27 kg) × (74 × 10−12 m)2
= 30.7... cm−1
(a) �e next step is to compute the moments of intertia, and hence the ro-
tational constants, for the isoceles geometry of the activated complex.
From the data given the D–D distance is ��.� pm and the H–D distance
is ��.� pm. Figure ��.� shows the geometry: the �lled circles are D and
the open circle is H. For this approximate calculation integer masses are
used.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 585
��.� pm y
b
x
��.� pm
Figure 15.1
�e centre of mass lies along the vertical bisector a, and the moment of
inertia about this axis is
h (6.6261 × 10−34 J s)
à = =
8π 2 cI 8π 2 (2.9979 × 1010 cm s−1 ) × (1.30... × 10−47 kg m2 )
= 21.3... cm−1
I b = 2 × 2 × m u × x 2 + 1 × m u (y − x)2
= 4 × (1.6605 × 10−27 kg) × (44.4 × 10−12 m)2
+ 1 × (1.6605 × 10−27 kg) × ([85.3... − 17.0...] × 10−12 m)2
= 9.67... × 10−47 kg m2
axis therefore σ = 2.
1 kT 3�2 π 1�2
q= � � � �
σ hc ÃB̃C̃
(1.3806 × 10−23 J K−1 ) × (400 K)
3�2
= 12 � �
(6.6261 × 10−34 J s) × (2.9979 × 1010 cm s−1 )
1�2
π
� �
(21.3... cm−1 ) × (28.9... cm−1 ) × (12.2... cm−1 )
= 47.1...
For this triangular activated complex there are three normal modes, one
of which corresponds to the reaction coordinate. �us, the rate constant
is
(b) Now consider a linear geometry for the activated complex, H–D–D, with
the H–D distance as ��.� pm and the D–D distance as ��.� pm. �e �rst
task is to locate the centre of mass. Assuming that this is a distance x from
the right-hand D it follows that 2x = 2(88.8 − x) + 1(96.2 + 88.8 − x).
which gives x = 72.52 pm. �e moment of inertia is therefore
(c) �e calculations performed so far both give rate constants in excess of the
target (by a factor of about �). �e only contribution which can feasibly
change by this much is the rotational partition function of the activated
complex. To make this smaller, the rotational constants need to be in-
creased, which is achieved by generally shrinking the size of the complex.
Some trial calculations suggest that shrinking the H–D distance to ��%
of the H� distance achieves the desired result, but this seems a rather
implausible structure of the activated complex.
P��C.� �e variation of the rate constant with ionic strength is given by [��C.��–���],
lg k r = lg k r○ + 2Az A z B I 1�2 ; at ��� K and for aqueous solutions A = 0.509. A
plot of lg(k r �k r○ ) against I 1�2 is used to explore whether or not this relationship
applies.
I I 1�2 k r �k r○ lg(k r �k r○ )
0.010 0 0.100 0 8.10 0.908
0.015 0 0.122 5 13.30 1.124
0.020 0 0.141 4 20.50 1.312
0.025 0 0.158 1 27.80 1.444
0.030 0 0.173 2 38.10 1.581
0.035 0 0.187 1 52.00 1.716
1.8
1.6
1.4
lg(k r �k r○ )
1.2
1.0
0.8
0.08 0.10 0.12 0.14 0.16 0.18 0.20
1�2
I
Figure 15.2
�e plot is shown in Fig ��.�: the data fall on a good straight line with slope
+9.18. Such a value implies
P��C.� Some experimentation with various graphs shows that a plot of lg k r against I
(not I 1�2 ) gives a good straight line, as shown in Fig ��.�.
−0.16
lg[k r �(dm3 mol−1 s−1 )]
−0.17
−0.18
0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
I
Figure 15.3
In the limit of zero ionic strength lg[k r○ �(dm3 mol−1 s−1 )] = −0.1815, hence
k r○ = 0.658 dm3 mol−1 s−1 .
From the text [��C.��b–���] gives the dependence of lg k r on the activity coef-
�cients
γ C‡
lg k r = lg k r○ − lg
γA γB
In this case one reactant (say A) is known to have a charge of −1, and the other
reactant (say B) is neutral. �erefore, the charge on the activated complex is
also −1 and hence, if the Debye–Hückel limiting law applies to these species,
γ C‡ = γ A . It therefore follows that
lg k r = lg k r○ + lg γ B
Given that k r (D)�k r (H) = 1�6.4 it follows that ζ = 1.86.... With this value, and
using integer masses for H and D, the expression for ζ becomes
D��D.� Molecular beams may be used to prepare molecules in speci�c rotational and
vibrational states, and then to examine the results of collisions between such
precisely prepared species. Section ��D.�(a) on page ��� describes how molec-
ular beams are prepared such that the molecules in them have a very nar-
row range of velocities and therefore relatively few collisions to redistribute
their energies. Molecules in such beams can be prepared in speci�c vibrational
states, for example, by using lasers to excite vibrations. Crossing two molecular
beams allows collisions to be staged between two sets of precisely characterized
molecules. Detectors can then be used to study the results of those collisions,
recording the number of molecules in which particular states that are scattered
in a given direction.
590 15 REACTION DYNAMICS
Solutions to exercises
E��D.�(a) �e numerator of [��D.�–���] is integrated as follows with the assumption that
P(E) is independent of E and can therefore be written P
∞ ∞
� P(E) e−E�k T dE = P � e−E�k T dE
0 0
∞
= −PkT e−E�k T �0 = PkT
�e e�ect of this term is to make the rate constant increase with temperature.
E��D.�(a) Refer to Fig. ��D.�� on page ���, which shows an attractive potential energy
surface as well as trajectories of both a successful reaction and an unsuccessful
successful trajectory passes through the transition state (marked as ‡ ○). �is
one. �e trajectories begin in the lower right, representing reactants. �e
trajectory is fairly straight from the lower right through the transition state,
indicating little or no vibrational excitation in the reactant. �erefore most
of its energy is in translation. Since it has enough total energy to reach the
transition state, the reactant can be described as being high in translational
energy and low in vibrational energy.
�is successful trajectory moves from side to side along the valley representing
products, so the product is high in vibrational energy and relatively lower in
translational energy. �e unsuccessful trajectory, by contrast, has a reactant
high in vibrational energy; it moves from side to side in the reactant valley
without reaching the transition state.
Solutions to problems
P��D.� �e change in intensity of the beam, dI, is proportional to the number of scat-
terers per unit volume, N , the intensity of the beam, I, and the path length dL.
�e constant of proportionality is the collision cross-section σ, the ‘target area’
of each scatterer.
1
dI = −σN I dL hence dI = −σN dL hence d ln I = −σN dL
I
If the incident intensity at L = 0 is I 0 , and the intensity a�er scattering through
length L is I, integration gives
I L
� d ln I = � −σN dL
I0 0
hence ln I�I 0 = −σN L
k r○ k○ k○
kr = = r � e−λυ e−υ hν�k T = r � e−υ (λ+hν�k T)
′ ′ ′
−λυ −υhν�k T
� δ υυ′ e e
q υ,υ′ q υ′ q υ′
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 591
Writing out the �rst few terms of the sum shows that it is a geometric progres-
sion with common ratio e−(λ+hν�k T)
k r○
= �1 + e−(λ+hν�k T) + e−2(λ+hν�k T) + . . .�
q
k○ 1
= r
q 1 − e−(λ+hν�k T)
where to go to the �nal line the fact that the sum to in�nity of a geometric series
is a�(1 − r), where r is the common ratio and a is the �rst term, is used. �is
relationship requires r < 1, which is the case provided λ > 0.
Solutions to exercises
E��E.�(a) �e rate constant for electron-transfer is given by [��E.�–���] together with
[��E.�–���]
(∆ r G −○ + ∆E R )2
1�2
1 π3
k et = � � H et (d)2 e−∆ G�RT ∆‡ G =
‡
h RT∆E R 4∆E R
For the two reactions given, ∆E R and ∆‡ G are assumed to be the same. �e
distance dependence of H et (d)2 is given by [��E.�–���], H et (d)2 = H et
○2 −βd
e ,
therefore
k et,2 (H et (d)2 )2
= = e−β(d 2 −d 1 )
k et,1 (H et (d)2 )1
hence ln(k et,2 �k et,1 ) = −β(d 2 − d 1 )
ln(k et,2 �k et,1 )
therefore β = −
(d 2 − d 1 )
ln[(4.51 × 104 s−1 )�(2.02 × 105 s−1 )]
β=−
(1.23 nm) − (1.11 nm)
= 12.5 nm−1
592 15 REACTION DYNAMICS
Increasing the distance from �.� nm to �.� nm reduces H et (d)2 to about �.��%
of its initial value.
E��E.�(a) �e rate constant for electron-transfer is given by [��E.�–���] together with
[��E.�–���]
(∆ r G −○ + ∆E R )2
1�2
1 π3
k et = � � H et (d)2 e−∆ G�RT ∆‡ G =
‡
h RT∆E R 4∆E R
With the given data there is only one unknown quantity, ∆E R , but it is not pos-
sible to �nd an analytical expression for this in terms of the other parameters.
However, mathematical so�ware is able to �nd a solution numerically. Before
embarking on such a calculation it is important to make sure that the units of
the various quantities are consistent.
�e choice is made to express the energies as molar quantities (J mol−1 ). It
therefore follows that the term in parentheses has units (J−1 mol), and given
that the units of 1�h are (J−1 s−1 ), it follows that for the rate constant to have
the expected units of s−1 , H et (d)2 must be in J2 mol−1 . Using the conversion
factors from inside the front cover
1.9864 × 10−23 J
2
H et (d)2 = �(0.04 cm−1 ) × � × (6.0221 × 1023 mol−1 )
1 cm−1
= 3.80... × 10−25 J2 mol−1
96.485 × 103 J mol−1
∆ r G −○ = (−0.185 eV) × = −1.78... × 104 J mol−1
1 eV
�e constant factor evaluates to
1�2 1�2
1 π3 1 π3
� � H et (d)2 = � �
h RT 6.6261 × 10−34 J s (8.3145 J K−1 mol−1 )×(298 K)
× (3.80... × 10−25 J2 mol−1 )
= 6.41... × 107 J1�2 mol−1�2 s−1
�e equation to be solved is therefore
−1�2
(37.5 s−1 ) = (6.41... × 107 J1�2 mol−1�2 s−1 ) × ∆E R × e−∆ G�(2.47 ...×10 3 J mol−1 )
‡
Solutions to problems
P��E.� �is Problem is somewhat ill-posed and some of the references to equations
in the text are incorrect. �e key missing item is that it should be assumed
that the electron-transfer process is rate limiting so that k r = Kk et , where K
is the equilibrium constant for the di�usive encounter; this point is discussed
in Section ��E.� on page ���. In addition, the �rst-order rate constant for the
electron-transfer process is written, using transition-state theory, as
∆ r G −○ 2∆ r G −○ ∆E R,DA ∆E R,DA 2
2
∆‡ G DA = + +
4∆E R,DA 4∆E R,DA 4∆E R,DA
∆ r G −○
2
= + 1 ∆ r G −○ + 14 ∆E R,DA ≈ 12 ∆ r G −○ + 14 ∆E R,DA
4∆E R,DA 2
where the approximation �∆ r G −○ � � ∆E R,DA is used in the last step. As
indicated in the Problem, ∆E R,DA may be written
Using this in the above expression for ∆‡ G DA gives the required expres-
sion
(c) �e overall rate constant for the self-exchange process are written k AA =
K AA k et,AA and likewise for DD. �e �rst-order rate constant for the elec-
tron transfer process is written using eqn ��.� as k et,AA = κν‡ e−∆ G AA �RT
‡
594 15 REACTION DYNAMICS
(e) Next, eqn ��.� is used to substitute for ∆‡ G AD in the expression for k r
= K AD κν‡ e−(∆ r G
−
○
+∆‡ G AA +∆ ‡ G DD )�2RT
k r = K AD e−∆ r G [k AA �K AA ] [k DD �K DD ]
−
○
�2RT 1�2 1�2
K AD
kr = (k AA k DD K)1�2
(K AA K DD )
1�2
which is shown on the plot. �e maximum of this function occurs when the
derivative is zero, that is when 2×−2.828×(−∆ r G −○ �eV)+5.942 = 0; this occurs
at (−∆ r G −○ �eV) = 1.05.... �erefore ∆E R = 1.05 eV .
10
lg k et
6
Figure 15.4
30
20
lg(k et �s−1 )
10
For a typical value σ = 0.4 nm2 and a mass of 2 × 10−26 kg, at ��� K
N A × qmT × (q R )3
k r = (1.5 × 1011 m3 mol−1 s−1 ) × e−∆E 0 �RT
qmT × (q R )3 × qmT × (q R )3
1
= (1.5 × 1011 m3 mol−1 s−1 ) × R 3 e−∆E 0 �RT
(q ) × qmT �N A
1
= (1.5 × 1011 m3 mol−1 s−1 ) ×
(10) × 107
3
�e steric factor is now 9 × 10−8 – very much smaller than for the reaction
between structureless particles.
16 Magnetic resonance
Solutions to exercises
E��A.�(a) �e energy level separation is ∆E = hν where ν = γ N B0 �2π, [��A.�–���].
Hence, in megahertz, the frequency separation is
γ N B0 (6.73 × 107 T−1 s−1 ) × (15.4 T)
ν = 10−6 × = 10−6 × = 165 MHz
2π 2π
E��A.�(a) �e energy level separation is ∆E = hν where ν = γ N B0 �2π, [��A.�–���].
Hence, for a given magnetic �eld, ∆E ∝ γ N . Using γ N (15 N) = −2.712 ×
107 T−1 s−1 and γ N (31 P) = 10.84 × 107 T−1 s−1 , it follows that �γ N (31 P)� >
�γ N (15 N)� and so the separation of energy levels is larger for 31 P than for 15 N.
E��A.�(a) �e ground state has m I = + 12 (α spin) and population N α , and the upper
state has m I = − 12 (β spin) and population N β . �e total population N is N =
N α + N β , and the population di�erence is N α − N β . �e Boltzmann distribution
gives N β �N α = e−∆E�k T , where ∆E is the energy di�erence between the two
states: ∆E = γ N ħB0 . It follows that N β = N α e−∆E�k T . With these results
N α − N β N α − N β N α (1 − e−∆E�k T ) 1 − e−∆E�k T
= = =
N N α + N β N α (1 + e−∆E�k T ) 1 + e−∆E�k T
� ± 12 ħ � 1
θ = cos−1 √ = cos−1 � √ � = ±0.9553 rad = ±54.74○
� 3ħ� 3
2
E��A.�(a) �e energies of the nuclear spin states in a magnetic �eld are given by [��A.�d–
���], E m I = −g I µ N B0 m I where g I is the nuclear g-factor, µ N is the nuclear
magneton, B0 is the magnitude of the magnetic �eld, and the component of
the angular momentum on a speci�ed axis is m I ħ where m I = I, I − 1, ..., −I.
�erefore, since the possible values of m I are ± 32 , ± 12 , the energies of nuclear
spin states are
E m I = −g I µ N B0 m I
= −(0.4289) × (5.0508 × 10−27 J T−1 ) × (6.800 T) × m I
= (−1.473... × 10−26 J) × m I
Solutions to problems
P��A.� �e resonance condition in NMR is [��A.�–���], hν = γ N ħB0 , where ν is
the resonance frequency, γ N is the nuclear magnetogyric ratio, and B0 is the
magnetic �eld strength. Hence, assuming the same magnetic �eld for both
neutrons and �H,
hν n γ N (n)ħB0 γ N (n)
= =
hν 1H γ N (1 H)ħB0 γ N (1 H)
With the data given, and taking the modulus of the negative magnetogyric ratio
of the neutron,
N β − N α −γ N (n)ħB0
=
N 2kT
�e above expression for B0 is used to give
N β − N α −γ N (n)ħ 2πν 1H −γ N (n)hν 1H
= × =
N 2kT γ N ( 1H) 2kTγ N ( 1H)
602 16 MAGNETIC RESONANCE
D��B.� �is is discussed in detail in Section ��B.�(d) on page ���. In summary, two
nuclei are chemically equivalent if they are related by a symmetry operation of
the molecule. Such nuclei will have the same resonance frequency, that is, the
same chemical shi�. Examples are the protons in benzene, and H-� and H-� in
�-nitrophenol. In benzene the protons are related by a C 6 operation (as well as
others) and in �-nitrophenol the protons are related by re�ection in a mirror
plane perpendicular to the plane of the ring and passing through C-� and C-�.
Two chemically equivalent nuclei A and A′ are magnetically equivalent if the
coupling between A and any other magnetic nucleus in the molecule, say Z,
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 603
Solutions to exercises
E��B.�(a) �e resonance frequency ν is given by [��B.�–���], ν = ν○ + (ν spect �106 )δ,
where ν○ is the resonance frequency of the standard. �e frequency separation
of the two peaks, ∆ν, is
ν spect ν spect ν spect
∆ν = ν 2 − ν 1 = [ν○ + � � δ 2 ] − [ν○ + � 6 � δ 1 ] = � 6 � (δ 2 − δ 1 )
106 10 10
400.130000 × 10 Hz
6
=� � (9.80 − 2.20) = 3040 Hz
106
106 106
∆δ = δ 2 − δ 1 = ∆ν × � � = (550 Hz) × = �.��
ν spec 400.130000 × 106 Hz
νX νA
J AX = 10 Hz J AX = 10 Hz
250 Hz
ν increasing
For a spectrometer operating at 800 MHz the resonance frequencies of the two
spins are
νX νA
J AX = 10 Hz J AX = 10 Hz
800 Hz
ν increasing
��
E��B.�(a) F NMR: the four 19 F nuclei are equivalent and so all have the same shi�. �is
resonance is split by coupling to the single 10 B nucleus into 2nI + 1 lines; here
n = 1 and I = 3 so there are � lines of equal intensity. �e splitting between
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 605
each is the B–F coupling constant, and the multiplet is centred at the shi� of
the �uorine.
��
B NMR: the 10 B nucleus is coupled to four equivalent 19 F nuclei, giving a
multiplet with 2nI + 1 lines; here n = 4 and I = 12 so there are � lines. �ese
lines have intensities in the ratio �:�:�:�:�, each line is separated from the next
by the B–F coupling constant, and the whole multiplet is centred at the shi� of
the boron.
E��B.�(a) �e multiplet resulting from coupling to four equivalent spin- 12 nuclei orig-
inates from the resonance of nucleus A being split into two by coupling with
one X nucleus, and then each of those two lines being split into two by coupling
to the second X nucleus. �ese lines are then each further split into two by
coupling to the third X nucleus, and �nally each of these lines is split into two
by the fourth X nucleus.
As a result of the X nuclei being equivalent, each causes the same splitting and
thus some lines are coincident and give rise to absorption lines of increased
intensity. As shown in Fig. ��.�, the overall result is a �:�:�:�:� quintet .
J AX
ν
Figure 16.1
E��B.�(a) �e multiplet resulting from coupling to two non-equivalent spin- 12 nuclei orig-
inates from the resonance of nucleus A being split into two by coupling with
the X1 nucleus, and then each of these lines being split into two by coupling to
the X2 nucleus.
Because the coupling between A and X1 is not the same as that between A and
X2 , no lines are necessarily coincident. Coupling to two non-equivalent spin- 12
nuclei thus results in a multiplet with four lines of equal intensity, in this context
usually referred to as a doublet of doublets, as shown in Fig. ��.�.
606 16 MAGNETIC RESONANCE
J AX1
J AX2
ν
Figure 16.2
E��B.�(a) Coupling to one I = 52 nucleus results in the original line being split into (2I +
1) = 6 lines. Each of these lines is then split into six in the same way by coupling
to the second I = 52 nucleus.
As a result of each nucleus being equivalent, the splittings are the same and
thus some lines are coincident and give rise to absorption lines of increased
intensity. �e result is a �:�:�:�:�:�:�:�:�:�:� multiplet , as is shown in Fig. ��.�.
J AX
ν
Figure 16.3
E��B.��(a) A group of nuclei are chemically equivalent if they are related by a symmetry
operation of the molecule and therefore have the same chemical shi�. Two
chemically equivalent nuclei A and A′ are magnetically equivalent if the cou-
pling between A and any other magnetic nucleus in the molecule, say Z, is
the same as the coupling between A′ and Z. If Z itself is to be magnetically
equivalent to Z′ , then the couplings A–Z, A–Z′ , A′ –Z, and A′ –Z′ must all be
equal: that is, the couplings between any member of a group of equivalent spins
with any member of another group of equivalent spins must be the same in
order for the groups each to be magnetically equivalent.
In �-chloro-�-bromobenzene, H2 and H6 are chemically equivalent because
they are related by a 180○ degree rotation about the C1 –C4 axis. Similarly, H3
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 607
and H5 are chemically equivalent because they are related by the same opera-
tion.
�e coupling between H2 and H3 is not the same as between H2 and H5 , there-
fore the test for magnetic equivalence is not satis�ed: H2 and H6 are not mag-
netically equivalent, nor are H3 and H5 .
Cl
H6 H2
H5 H3
Br
Fa
Fe
P Fe
Fe
Fa
E��B.��(a) Coalescence of two NMR lines due to rapid nuclei exchange occurs when the
condition given in [��B.��–���], τ = 21�2 �πδν, is satis�ed; in this expression τ
is the lifetime of an environment, and δν is the di�erence between the Larmor
frequencies of the two environments. Using [��B.�–���],
For a proton jumping between two sites with �rst-order rate constant k, k = 1�τ
ν − ν○ 750 Hz
δ= × 106 = × 106 = �.�
ν ○ (500.130000 × 106 Hz)
Solutions to problems
���
P��B.� Xe NMR spectrum: the signal is split by coupling to the ��F atom (���%
natural abundance, spin 12 ) into a doublet with coupling constant J = 7600 Hz.
19
F NMR spectrum: this consists of a superposition of spectra from the dif-
ferent isotopes of Xe, and the intensity of these spectra re�ects the abundance
of each isotope. A doublet, with splitting ���� Hz, arises from those ��% of
molecules containing ���Xe. �e remaining ��% of molecules do not contain
Xe nuclei which give rise to splitting, so a single line appears.
�e observed spectrum is not a triplet, in the sense that it does not arise from
coupling to two equivalent spin- 12 nuclei. Rather it is a superposition of a singlet
and a doublet.
.. Fa ..
Fe Fe
S Fe S : F S F
Fe Fe F F
Fa Fa
A B C
Structure A. �ere are two environments for �uorine atoms: the three Fe are
equivalent as they are related by a threefold rotation about the S–Fa bond; Fa
is unique. Each Fe has the same coupling to Fa , so the Fe are magnetically
equivalent. �e spectrum will therefore consist of a doublet from Fe and a
quartet from Fa .
Structure B. �ere are two environments for �uorine atoms: the two Fe are
equivalent as they are related by a twofold rotation about the S–lone pair axis;
the two Fa atoms are equivalent because they are related by re�ection in the
mirror plane formed by the two Fe and the S. Each Fe has the same coupling to
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 609
P��B.� Figure ��.� shows the three possible staggered conformations of XYCHCHR� R� .
When a staggered conformation is adopted, the angle between adjacent C–H
bonds is � = 60○ . Conformation a is trans (� = 180○ , 3 J HH = 3 J t ) and confor-
mations b and c are gauche (� = 60○ , 3 J HH = 3 J g ).
R� H R�
X H X H X H
H R� R� R� R� H
Y Y Y
a b c
Figure 16.4
For R3 = R4 = H, all conformations have the same energy and each occurs with
equal probability (P = 13 ). �us
3
J HH = 13 (3 J t + 2 × 3 J g ) = 7.3 Hz (��.�)
For R3 = CH3 and R4 = H, the steric bulk of the methyl group makes con-
formation c very unfavourable. �us, only equivalent conformations a and
b are relevant for calculating 3 J HH : these have equal energy and hence equal
probability (P = 12 ). �us
3
J HH = 12 (3 J t + 3 J g ) = 8.0 Hz (��.�)
For R3 = R4 = CH3 , the steric bulk of the methyl group makes conformations
b and c very unfavourable due to close proximity to the bulky X and Y groups.
�erefore only conformation a is relevant for calculating 3 J HH . �us
3
J HH = 3 J t = 11.2 Hz (��.�)
�e latter two results are used to give an average value 3 J g = 5.1 Hz and 3 J t =
11.2 Hz.
Using the original Karplus equation, 3 J HH = A cos2 � + B, the gauche and trans
couplings are calculated as
3
J g = A cos2 (60○ ) + B = 14 A + B 3
J t = A cos2 (180○ ) + B = A + B
�ere is poor agreement between these values and the ones determined from
the experimental data.
∂ 3 J HH ∂
= [A + B cos � + C cos 2�]
∂� ∂�
= −B sin � − 2C sin 2�
∂ 3 J HH
= −B sin � − 4C sin � cos �
∂�
= −(B + 4C cos �) sin �
Hax Hax
Hax
Heq
Heq Heq
Heq Heq
Heq
Hax
Hax Hax
Fa
Fe
P Fe
Fe
Fa
(c) Unless water is excluded rigorously from the sample there will exchange
between the OH of the alcohol and the residual water in the solvent. �is
process leads to a collapse of the splitting due to any coupling to the
OH proton because its spin state is scrambled as it leaves and returns
to the molecule. �e spectrum then consists of a triplet and a quartet,
characteristic of a CH� CH� group.
If water is excluded, the exchange is inhibited and then splittings due to
coupling to the OH proton are seen, giving more complex multiplets.
612 16 MAGNETIC RESONANCE
D��C.� Relaxation results from the interaction between individual spins and local mag-
netic �elds which arise in various ways. As is described in Topic ��A, the inter-
action between a spin and a magnetic �eld depends on the magnetogyric ratio
of the spin: the greater the magnetogyric ratio, the stronger the interaction.
�us, a given local �eld has a greater interaction with a nucleus with a greater
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 613
magnetogyric ratio and, all other things being equal, the relaxation caused by
these �elds will therefore be faster the greater the magnetogyric ratio.
�e magnetogyric ratio for �H is about four times that of ��C, hence the inter-
actions between �H and local �elds is stronger leading to faster relaxation and
shorter relaxation times. A somewhat separate issue is that in typical molecules
H atoms tend to be more accessible than C atoms, and so if the local �eld is from
outside the molecule it may be felt more strongly by the more exposed H atoms.
Solutions to exercises
E��C.�(a) Ethanoic acid has two carbon environments: a methyl carbon and a carboxylic
acid carbon. �e ��C NMR spectrum shows a signal at δ = 21, which is due
to the methyl carbon. �is C atom is adjacent to three protons and couples to
them, giving rise to a quartet. �e splitting of the quartet is the one-bond C–H
coupling: 1 J = 130 Hz.
�e signal at δ = 178 is due to the carboxylic acid carbon. �is C atom also
couples to the three methyl protons, but this time over two bonds which results
in a much smaller coupling. As with the methyl carbon, a quartet is seen; the
splitting is the much smaller two-bond C–H coupling.
When the ��C NMR spectrum is recorded with proton decoupling, the split-
tings due to C–H couplings are collapsed, leaving two singlets.
E��C.�(a) �e maximal NOE enhancement is given by [��C.�–���], η = γ X �2γ A , where
γ X and γ A are the magnetogyric ratios of nuclei X and A, respectively.
γ1 H (26.752 × 107 T−1 s−1 )
η= = = �.���
2γ 31 P 2 × (10.840 × 107 T−1 s−1 )
E��C.�(a) From the discussion in Section ��C.�(a) on page ���, the �ip angle � of a pulse
of duration ∆τ is given by � = γN B1 ∆τ, where B1 is the strength of the applied
�eld. It follows that B1 = ��γN ∆τ. With the data given
� π
B1 = = = 9.40 × 10−4 T
γN ∆τ (26.752 × 107 T−1 s−1 ) × (12.5 × 10−6 s)
�e corresponding 90○ pulse has half the �ip angle so, for the same B1 , ∆τ is
halved: ∆τ = 12 × (12.5 µs) = 6.25 µs .
E��C.�(a) �e e�ective transverse relaxation time, T2∗ , is given by [��C.�–���], T2∗ =
1�π∆ν 1�2 , where ∆ν 1�2 is the width of the (assumed) Lorentzian signal mea-
sured at half the peak height. Hence
1 1
T2∗ = = = 0.21 s
π∆ν 1�2 π × (1.5 Hz)
E��C.�(a) �e amplitude of the free induction decay, S(t), is proportional to e−t�T2 , where
t is time and T2 is the transverse relaxation time . If S(t 2 ) = 12 S(t 1 ) then
−∆t ∆t
ln 12 = −∆t�T2 therefore T2 = =
ln 12 ln 2
(1.0 s)
T2 = = 1.4 s
ln 2
Solutions to problems
P��C.� (a) �e �ip angle � of the pulse is given by � = ∆τγ N B1 (Section ��C.�(a) on
page ���). �erefore, increasing the pulse duration ∆τ increases the �ip
angle. �e strongest signal will be observed a�er application of a ��○ pulse
because such a pulse rotates the magnetization completely to the xy-plane.
A ���○ pulse rotates the magnetization from +z to −z and therefore creates
no transverse magnetization, and hence no observable signal.
From the description of the experiment it is clear that pulses of duration
�.� µs and �.� µs correspond to �ip angles of less than ��○ , and about ��○ ,
respectively. �e �.� µs pulse must correspond to a �ip angle of more than
��○ because the signal is weaker than for a shorter pulse. �e ��.� µs pulse
gives no signal and so must correspond to a ���○ pulse.
(b) �e ��.� µs pulse gives no signal and so corresponds to a �ip angle of ���○ .
A ��○ pulse has half the duration of a ���○ pulse, that is ∆τ 90 = 5.0 µs .
(c) �e magnetic �eld strength B1 is determined by rearranging � = ∆τγ N B1
to B1 = ��∆τγ N . �e length of the ���○ pulse is normally used in the
calculation of B1 because such a pulse gives a null signal, which is easier
to determine than the maximum seen for a ��○ pulse. If the duration of
the ���○ pulse is ∆τ 180 it follows that B1 = π�∆τ 180 γ N . Substituting this
expression into that for the B1 Larmor frequency gives
γ N B1 γ N π 1
ν′L = = × =
2π 2π ∆τ 180 γ N 2∆τ 180
1
= = 5.00 × 104 Hz
2 × (10.0 × 10−6 s)
An alternative way at looking at this is to say that the magnetization com-
pletes a full rotation in ��.� µs (a ���○ pulse), so the period of the rotation
is ��.� µs. �e corresponding frequency is 1�period = 1�(20.0×10−6 s) =
50.0 kHz. �is is the B1 Larmor frequency, the frequency at which the
magnetization rotates about B1 .
P��C.� To use the discrete Fourier transform the function needs to be generated as an
array of points, regularly sampled in time. For the plots shown here ��� time-
domain points were generated out to a maximum time of �.� s (some Fourier
transform routines will only transform numbers of data points which are a
power of �, 29 = 512); the interval between the points is therefore (3.0 s)�512 =
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 615
5.86 × 10−3 s. �is means that in the spectrum the maximum frequency which
is represented correctly is 1�(2 × 5.86 × 10−3 s) = 85 Hz; this determines the
scale on the spectra.
Figure ��.� shows the free-induction decay and spectrum for the parameters
given in the Problem. �e peak at �� Hz is higher than that at �� Hz both because
its intensity is greater and also because it is narrower on account of having a
longer T2 ; recall that the area under the peak is constant, so if it broadens the
peak height decreases.
In Fig. ��.� the high-frequency peak has been moved from �� Hz to �� Hz,
and its intensity has been increased from � to �; its peak height is still lower
than the peak at �� Hz on account of the higher-frequency resonance having a
shorter T2 and hence a broader line. In Fig. ��.� the T2 of the high-frequency
peak is reduced from �.� s to �.� s, as is evident from the broadening of the
line and the reduction in its peak height. In all three cases it is evident that
the time-domain signal is changing, but only the frequency-domain signal, the
spectrum, is readily interpretable.
0 20 40 60 80 f / Hz
Figure 16.5
0 20 40 60 80 f / Hz
Figure 16.6
0 20 40 60 80 f / Hz
Figure 16.7
616 16 MAGNETIC RESONANCE
∞
I(ω) = Re �� G(t)eiωt dt�
0
Using the suggested substitution, the free-induction decay G(t) = ae−t�τ cos ω 0 t
is rewritten
∞
I(ω) = aRe �� 1 −t�τ
2
e �e−iω 0 t + eiω 0 t � eiωt dt�
0
∞
= 12 aRe �� eit(ω−ω 0 +i�τ) + eit(ω+ω 0 +i�τ) dt�
0
∞
1 1
= 1
aRe � eit(ω−ω 0 +i�τ) + eit(ω+ω 0 +i�τ) �
2
i(ω − ω 0 + i�τ) i(ω + ω 0 + i�τ) 0
1 1
= − 12 aRe � + �
i(ω − ω 0 + i�τ) i(ω + ω 0 + i�τ)
i i
= 12 aRe � + �
(ω − ω 0 + i�τ) (ω + ω 0 + i�τ)
i(ω − ω 0 − i�τ) i(ω + ω 0 − i�τ)
= 12 aRe � + �
(ω − ω 0 + i�τ)(ω − ω 0 − i�τ) (ω + ω 0 + i�τ)(ω + ω 0 − i�τ)
i(ω − ω 0 ) + 1�τ i(ω + ω 0 ) + 1�τ
= 12 aRe � + �
(ω − ω 0 ) + 1�τ
2 2 (ω + ω 0 )2 + 1�τ 2
a 1 1
= � + �
2τ (ω − ω 0 )2 + 1�τ 2 (ω + ω 0 )2 + 1�τ 2
aτ 1 1
= � 2 + 2 �
2 τ (ω − ω 0 ) + 1 τ (ω + ω 0 )2 + 1
2
P��C.� Note that there is an error in the Problem: the initial condition should be M z (0) =
−M 0
M (τ) t τ
[ln (M z (t) − M 0 )]−Mz 0 = �− �
T1 0
M z (τ) − M 0 τ
ln � �=−
−2M 0 T1
hence M z (τ) = M 0 (1 − 2e−τ�T1 )
(b) Figure ��.� shows plots of M z (τ)�M 0 = 1 − 2e−τ�T1 for di�erent values
of T1 . It is evident that increasing T1 increases the time required for the
z-component of magnetisation to relax back to its equilibrium value, M 0 .
1.0 T1 = 0.2 s
T1 = 1.0 s
0.5 T1 = 2.0 s
M z (τ)�M 0
0.0
−0.5
−1.0
0 1 2 3 4 5
τ� s
Figure 16.8
(c) From the result above, M z (τ) = M 0 (1−2e−τ�T1 ), it follows that M z (τ)�M 0
= (1−2e−τ�T1 ), and hence − 12 [M z (τ)�M 0 −1] = e−τ�T1 . Taking logarithms
gives ln {[M 0 − M z (τ)]�2M 0 } = −τ�T1 . It therefore follows that a plot of
ln {[M 0 − M z (τ)]�2M 0 } against τ is expected to be a straight line of slope
= 1�T1 .
(d) �e data are shown in the table below and the plot in Fig ��.�.
τ� s M z (τ)�M 0 ln {[M 0 − M z (τ)]�2M 0 }
0.000 −1.000 0.000
0.100 −0.637 −0.200
0.200 −0.341 −0.400
0.300 −0.098 −0.600
0.400 0.101 −0.800
0.600 0.398 −1.201
0.800 0.596 −1.599
1.000 0.729 −1.999
1.200 0.819 −2.402
618 16 MAGNETIC RESONANCE
ln {[M 0 − M z (τ)]�2M 0 }
−1
−2
1 1
T1 = − =− = �.��� s
(slope) (−2.001 s−1 )
P��C.� (a) Separate the di�erential equation and integrate between t = 0 and t = τ,
and the corresponding limits M xy (0) and M xy (τ).
M xy (τ) 1 τ 1
� dM xy (t) = − � dt
M xy (0) M xy (t) 0 T2
M xy (τ) t τ
�ln M xy (t)� M (0) = �− �
xy T2 0
M xy (τ) τ
ln � �=−
M xy (0) T2
hence M xy (τ) = M xy (0)e−τ�T2
(b) Divide both sides of this expression by M xy (0) and take the natural log-
arithm to give
M xy (τ) τ
ln =−
M xy (0) T2
(c) �e data are shown in the table below and the plot in Fig ��.��.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 619
0
ln [M xy (τ)�M xy (0)]
−1
−2
−3
0 20 40 60 80 100 120 140
τ� ms
Figure 16.10
1 1
T2 = − =− = 50.0 ms
slope −0.0200 ms−1
γ N ħµ 0 m I
Bloc = − (1 − 3 cos2 θ)
4πR 3
where Bloc is the local magnetic �eld, γ N is the nuclear magnetogyric ratio,
µ 0 is the vacuum permeability, m I ħ is the z-component of the spin angular
momentum, R is the internuclear separation, and θ is the angle between the
direction of local �eld vector and the direction of the applied magnetic �eld.
Rearrange this expression to give R, assume that θ = 0 so that (1 − 3 cos2 θ) =
620 16 MAGNETIC RESONANCE
Solutions to exercises
E��D.�(a) �e total local magnetic �eld is given by [��D.�–���], Bloc = B0 + am I , where
m I = ± 12 and a is the hyper�ne coupling constant. �e centre of the spectrum is
at 332.5 mT. Proton � splits the centre line into two components with separation
a 1 = 2.0 mT.
E��D.�(a) (i) �e three protons in the ⋅CH� radical are equivalent, each with spin quan-
tum number I = 12 . If a radical contains N equivalent nuclei with spin I,
then there are 2N I + 1 hyper�ne lines with an intensity distribution given
by Pascal’s triangle. Hence for the CH� radical, there are (2×3× 12 +1) = 4
hyper�ne lines with intensity distribution 1 ∶ 3 ∶ 3 ∶ 1 .
(ii) �e three 2 H atoms in the ⋅CD� radical are equivalent, each with spin
quantum number I = 1. �e number of hyper�ne lines is (2×3×1+1) = 7,
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 621
and their intensity distribution is the same as that shown in Fig. ��D.� on
page ���, 1 ∶ 3 ∶ 6 ∶ 7 ∶ 6 ∶ 3 ∶ 1 .
E��D.�(a) If a radical contains N equivalent nuclei with spin quantum number I, then
there are 2N I + 1 hyper�ne lines. �e radical contains a single magnetic nu-
cleus, N = 1, and there are a total of four hyper�ne lines. �erefore 2N I + 1 =
2 × 1 × I + 1 = 4 hence I = 32 .
E��D.�(a) As described in Section ��D.� on page ���, the splitting between the lines in
the multiplet is the hyper�ne coupling constant. In this case the splitting is
(334.8 − 332.5) mT = 2.3 mT and (332.5 − 330.2) mT = 2.3 mT, therefore
a = 2.3 mT .
�e central line is una�ected by coupling and so is used to compute the g-value
using [��D.�–���], hν = g µB B0
Solutions to problems
P��D.� �e EPR resonance condition is given by [��D.�–���], hν = g e µ B B0 , where ν
is the spectrometer operating frequency, g e is the g-value of the electron, µ B is
the Bohr magneton and B0 is the magnetic �eld. �e expression is rearranged
to make ν the subject.
the separation between the energy levels associated with molecular vibrations .
It is much greater than the separations of rotational levels, but much less than
the separations of electronic energy levels.
a 0.274 mT
carbon � ρ= = = �.���
Q 2.25 mT
a 0.151 mT
carbon � ρ= = = �.���
Q 2.25 mT
a 0.534 mT
carbon � ρ= = = �.���
Q 2.25 mT
P��D.� �e proportion of time the unpaired electron spends in the N �s orbital is given
by the ratio of the hyper�ne couplings: P(N2s) = (5.7 mT)�(55.2 mT) =
0.103..., that is 10% of the time. �e proportion of time the unpaired electron
spends in the N �p orbital is P(N2pz ) = (1.3 mT)�(3.4 mT) = 0.382..., that is
38% of the time.
�e total probability that the electron will be found on the N atom is P(N) =
10% + 38% = 48% . �e total probability that the electron will be found on the
O atoms is P(O) = 100% − 48% = 52% .
In Problem P�A.� it is shown that the angle θ between two sp λ hybrids is related
to the hybridisation ratio λ by λ = ±(−1� cos θ)1�2 . �e hybridization ratio is
given by λ 2 = (p character)�(s character), therefore in the present case λ 2 =
(38%)�(10%) = 3.8, hence λ = 1.95 . �e angle between the hybrids is found
by rearranging the above expression to cos θ = −1�λ 2 , hence cos θ = −1�3.8
giving θ = 105○ . Such a hybrid is suitable for bonding an angular arrangement,
and so supports the view that this is indeed the geometry for NO� .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 623
− 0.082
0.299 0.271
− 0.135 − 0.129
0.369
CH3
− 0.024
�e computed spin densities for the the two protons ortho to the oxygen,
and to the two protons meta to the oxygen are di�erent. �is is undoubt-
edly a computational artefact, a result of the minimum-energy structure
having one methyl proton in the plane of the ring, which makes the right
and le� side of the ring slightly non-equivalent. Fast internal rotation
is expected to make the two meta protons, and the two ortho protons,
equivalent. �e spin density at the ortho carbons is therefore taken as
�.���, and that at the meta carbons as −0.132.
(b) �e hyper�ne coupling constants resulting from coupling to the ring pro-
tons are computed using the McConnell equation, [��D.�–���]: a = Qρ,
with Q = 2.25 mT. Hence a ortho = 0.285 × (2.25 mT) = 0.641 mT,
and a meta = 0.132 × (2.25 mT) = 0.297 mT. �e hyper�ne coupling for
interaction with the methyl protons cannot be calcuated in the same way
but is given by the so�ware as �.��� mT. �e EPR spectrum will show
a rather complex multiplet arising from splitting into a �:�:� triplet as a
result of the coupling to the ortho protons, each line is then split into a
�:�:� triplet as a result of the coupling to the meta protons, and each line
is split once more into a �:�:�:� quartet as a result of the coupling to the
methyl protons.
I��.� Terms such as [I], [EI], [E], νI , νEI and ν need to be eliminated from the ex-
pressions. First, note that total I (free I plus bound I) is the same as initial I;
similarly, total E (free E plus bound E) is the same as initial E.
[I]tot = [I]0 = [I] + [EI] and [E]tot = [E]0 = [E] + [EI]
Using this, the expression for KI is rewritten
[E][I] ([E]0 − [EI])([I]0 − [EI])
KI = =
[EI] [EI]
Because [I]0 >> [E]0 it follows that [I]0 must also be much greater than [EI],
even if all of the E binds I. Hence [I]0 − [EI] ≈ [I]0 . �is is substituted into the
above expression which is then solved for [EI].
([E]0 − [EI])[I]0 [E]0 [I]0
KI = hence [EI] = (��.�)
[EI] KI + [I]0
624 16 MAGNETIC RESONANCE
�is is rearranged to given an expression for [EI], and then [EI] is substituted
using eqn ��.�
where δν and ∆ν as as de�ned in the text. Finally, [I]0 is made the subject to
give
∆ν
[I]0 = [E]0 − KI
δν
It follows from this that a plot of [I]0 against 1�(δν) is a straight line with slope
[E]0 ∆ν and y-intercept −KI .
17 Molecular Interactions
D��A.� When the applied �eld changes direction slowly, the permanent dipole moment
has time to reorientate and so follows the �eld: the whole molecule rotates
into a new direction in response to a change in the �eld. However, when the
frequency of the �eld is high, a molecule cannot change direction fast enough to
follow the change in direction of the applied �eld and the dipole moment then
makes no contribution to the polarization of the sample. Because a molecule
takes about � ps to turn through about � radian in a �uid, the loss of this contri-
bution to the polarization occurs when measurements are made at frequencies
greater than about 1011 Hz (in the microwave region). It is said that orientation
polarization, the polarization arising from the permanent dipole moments, is
lost at such high frequencies
�e next contribution to the polarization to be lost as the frequency is raised
is the distortion polarization, the polarization that arises from the distortion of
the positions of the nuclei by the applied �eld. �e molecule is distorted by
the applied �eld, and the molecular dipole moment changes accordingly. �e
time taken for a molecule to bend is approximately the inverse of the molec-
ular vibrational frequency, so the distortion polarization disappears when the
frequency of the radiation is increased through the infrared.
At even higher frequencies, in the visible region, only the electrons are mobile
enough to respond to the rapidly changing direction of the applied �eld. �e
polarization that remains is now due entirely to the distortion of the electron
distribution, and the surviving contribution to the molecular polarizability is
called the electronic polarizability.
626 17 MOLECULAR INTERACTIONS
Solutions to exercises
E��A.�(a) �e relationship between the induced dipole moment µ∗ and the electric �eld
strength E is given by [��A.�a–���], µ ∗ = αE, where α is the polarizability.
�e polarizability volume α ′ is related to the polarizability α by [��A.�–���],
α ′ = α�4πε 0 . Combining these equations, rearranging for E, and using 1 V =
1 J C−1 gives
NA α NA µ2 1
Pm = +
3ε 0 9ε 0 k T
which implies that a graph of Pm against 1�T should be a straight line with
slope N A µ 2 �9ε 0 k and intercept N A α�3ε 0 . However, as there are only two data
points it is convenient to calculate the required quantities directly from the
data. Writing the molar polarization at the two temperatures as Pm (T1 ) and
Pm (T2 ) and considering Pm (T2 ) − Pm (T1 ) gives
NA µ2 1 1
Pm (T2 ) − Pm (T1 ) = � − �
9ε 0 k T2 T1
which is rearranged to give
9ε 0 k Pm (T2 ) − Pm (T1 )
1�2
µ=� × �
NA 1�T2 − 1�T1
9 × (8.8542 × 10−12 J−1 C2 m−1 ) × (1.3806 × 10−23 J K−1 )
=�
6.0221 × 1023 mol−1
(62.47 × 10−6 m3 mol−1 ) − (70.62 × 10−6 m3 mol−1 )
1�2
× �
1�(423.2 K) − 1�(351.0 K)
= 5.53... × 10−30 C m = 1.659 D
�e value of α is found using this value of µ together with one of the data points;
both give the same answer. Rearranging [��A.��–���], Pm = (N A �3ε 0 )(α +
µ 2 �3kT), for α and using the data for 351.0 K gives
3ε 0 Pm µ2
α= −
NA 3kT
3 × (8.8542 × 10−12 J−1 C2 m−1 ) × (70.62 × 10−6 m3 mol−1 )
=
6.0221 × 1023 mol−1
(5.53... × 10−30 C m)2
− = 1.008 × 10−39 C2 m2 J−1
3 × (1.3806 × 10−23 J K−1 ) × (351.0 K)
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 627
1 + 2ρPm �M
εr =
1 − ρPm �M
1 + 2 × (1.89 g cm−3 ) × (27.18 cm3 mol−1 )�(92.45 g mol−1 )
= = 4.75
1 − (1.89 g cm−3 ) × (27.18 cm3 mol−1 )�(92.45 g mol−1 )
n r2 − 1 ρN A α 3Mε 0 n r2 − 1
= hence α= × 2
n r2 + 2 3Mε 0 ρN A nr + 2
ε r − 1 ρN A α ρN A α
= hence εr − 1 = (ε r + 2)
ε r + 2 3Mε 0 3Mε 0
ρN A α 2ρN A α 1 + 2ρN A α�3Mε 0
hence ε r �1 − �=1+ hence ε r =
3Mε 0 3Mε 0 1 − ρN A α�3Mε 0
where C = 4πρN A α ′ �3M. Taking the mass density of water from the Resource
section as 0.997 g cm−3 and the molar mass as M = 18.0158 g mol−1
C = 4πρN A α ′ �3M
4π × (0.997 g cm−3 ) × (6.0221 × 1023 mol−1 ) × (1.5 × 10−24 cm3 )
=
3 × (18.0158 g mol−1 )
= 0.209...
1 + 2C 1�2 1 + 2 × 0.209... 1�2
nr = � � =� � = �.�
1−C 1 − 0.209...
E��A.�(a) �e Debye equation, [��A.��–���], is (ε r −1)�(ε r +2) = ρPm �M, where Pm is the
molar polarizability. �e latter is de�ned by [��A.��–���], Pm = (N A �3ε 0 )(α +
µ 2 �3kT), where µ is the dipole moment and α is the polarizability, which is
related to the polarizability volume α ′ according to [��A.�–���], α ′ = α�4πε 0 .
Replacing α in the expression for Pm by 4πε 0 α ′ gives
NA µ2
Pm = �4πε 0 α ′ + �
3ε 0 3kT
6.0221 × 1023 mol−1 �
= × 4π × (8.8542 × 10−12 J−1 C2 m−1 )
3 × (8.8542 × 10 J C m ) �
−12 −1 2 −1
× (1.23 × 10 −29
m )+
3
3 × (1.3806 × 10−23 J K ) × ([25 + 273.15] K) �
−1
ε r − 1 ρPm 1 + 2ρPm �M
= hence εr =
εr + 2 M 1 − ρPm �M
permittivity is
1 + 2ρPm �M
εr =
1 − ρPm �M
1 + 2 × (1.173 × 106 g m−3 ) × (8.13... × 10−5 m3 mol−1 )�(112.5495 g mol−1 )
=
1 − (1.173 × 106 g m−3 ) × (8.13... × 10−5 m3 mol−1 )�(112.5495 g mol−1 )
= ��.�
δ-
F I
δ- δ-
C δ+ δ+ O O
O
3δ+ δ- δ+
F F δ-
H δ+ H
δ- δ- O O
Figure 17.1
In none of the cases do the dipole moments associated with each bond cancel,
so all three molecules are polar . Note that the O–O bonds in ozone are polar
because the central atom is di�erent from the other two. In the case of CIF�
the dipole moment associated with the C–I bond is much less than that for the
C–F bonds, which results in the dipole moments not cancelling.
An alternative approach is use a symmetry argument. As explained in Sec-
tion ��A.�(a) on page ���, only molecules belonging to point groups C n , C nv or
C s may have a permanent electric dipole moment. CIF� , O� and H� O� belong
to point groups C 3v , C 2v and C 2 respectively, so all three are polar.
µres = �(1.5 D)2 + (0.8 D)2 + 2 × (1.5 D) × (0.8 D) × cos 109.5○ � = 1.4 D
1�2
6
y�nm
−2e
u
µx
⇠
.2 ! !
3 ! µ⇠⇠⇠⇠θ µ y
0! 9⇠
⇠ ?
−e
u ! 20 u
3e ! !○
-
�.�� x�nm
Figure 17.2
all of the point charges have a z coordinate of zero. �e components are then
combined using [��A.�b–���], µ = (µ 2x + µ 2y + µ z2 )1�2 , which is represented
graphically on the right half of Fig. ��.�.
µ y = � QJ yJ
J
and, from the diagram on the right of Fig. ��.�, the direction is given by
7.55... D
θ = tan−1 � � = 12○
31.6... D
Solutions to problems
P��A.� �e molecules are shown in Fig. ��.�.
CH3
CH3
CH3 CH3
CH3
CH3 CH3
Figure 17.3
It is possible to be certain about the result for the �,� isomer, because the sym-
metry of the molecule means that the dipoles must cancel and so µres is neces-
sarily zero.
P��A.� �e individual ethanoic acid molecules have a permanent dipole moment but
the dimers do not because of their symmetry which cause the individual dipoles
associated with polar bonds to cancel. As the temperature increases the dimer
� monomer equilibrium shi�s in favour of the monomers because the pro-
cess of breaking the hydrogen bonds between the molecules is endothermic.
Consequently a greater proportion of the species present are monomers and
therefore have a dipole, and consequently the apparent dipole moment per
molecule appears to increase.
P��A.� �e relationship between induced dipole moment µ ∗ and electric �eld strength
E is given by [��A.�a–���], µ ∗ = αE where α is the polarizability. �e polar-
izability volume α ′ is related to α by [��A.�–���], α ′ = α�4πε 0 , so the induced
dipole is
µ ∗ = αE = 4πε 0 α ′ E
= 4π × (8.8542 × 10−12 J−1 C2 m−1 ) × (2.22 × 10−30 m3 )×(15.0 × 103 V m−1 )
= 3.70... × 10−36 C m = 1.11... × 10−6 D = 1.11 µD
NA α NA µ2 1 NA µ2
Pm = + × or Pm = 43 πN A α ′ +
3ε 0 9ε 0 k T 9ε 0 k
where in the second form the de�nition of polarizability volume, which is given
by [��A.�–���], α ′ = α�4πε 0 , is used to replace α by 4πε 0 α ′ . �is equation im-
plies that a graph of Pm against 1�T should be a straight line of slope N A µ 2 �9ε 0 k
and intercept 43 πN A α ′ .
50
Pm �cm3 mol−1
liquid
40
←� increasing T
solid
30
�e �rst thing to note is that the molar polarization Pm at −80 ○ C and −70 ○ C
is much lower than for the other temperatures. �ese are the temperatures for
which trichloromethane (freezing point −64 ○ C) is a solid. �e lower Pm is
consistent with the molecules in the solid being �xed in position so that they
cannot change orientation to respond to the electric �eld and therefore their
permanent dipole moment makes no contribution to the polarization of the
sample; that is, there is no orientation polarization. In the liquid state there is
a contribution from orientation polarization and therefore the values of Pm are
larger.
For the liquid state, the plot of Pm against 1�T should be linear with a slope of
N A µ 2 �9ε 0 k and intercept 43 πN A α ′ as deduced earlier. In fact, Fig. ��.� shows
that the data do not lie very close to a straight line. �e line drawn on Fig. ��.�
uses only the points corresponding to the three highest temperatures and has
equation
(Pm �cm3 mol−1 ) = 3.76 × 103 × 1�(T�K) + 32.8
Identifying the slope with N A µ 2 �9ε 0 k gives
and hence
Rearranging gives
3
α′ = × (32.8 cm3 mol−1 )
4πN A
3 −1
= −1 × (32.8 cm mol ) = 1.3 × 10
3 −23
cm3
4π × (6.0221 × 10 mol )
23
Note that the intercept of Fig. ��.� corresponds approximately to Pm for the
solid. �is is interpreted as arising because the intercept gives Pm in the limit
of high temperature, under which conditions the molecules then have so much
thermal energy that they cannot be oriented by the applied electric �eld. Con-
sequently there is no orientation polarization contribution to Pm , just as is the
case in the solid where the molecules are prevented from rotating.
NA α NA µ2 1 NA µ2 1
Pm = + × or Pm = 43 πN A α ′ + ×
3ε 0 9ε 0 k T 9ε 0 k T
where in the second form the de�nition of polarizability volume, which is given
by [��A.�–���], α ′ = α�4πε 0 , is used to replace α by 4πε 0 α ′ . �is equation im-
plies that a graph of Pm against 1�T should be a straight line of slope N A µ 2 �9ε 0 k
and intercept 43 πN A α ′ . �e data are plotted in Fig. ��.�.
60
Pm �cm3 mol−1
50
40
and hence
1�2
9ε 0 k
µ=� × (1.5236 × 10−2 m3 mol−1 K)�
NA
9 × (8.8542 × 10−12 J−1 C2 m−1 ) × (1.3806 × 10−23 J K−1 )
=�
6.0221 × 1023 mol−1
1�2
×(1.5236 × 10−2 m3 mol−1 K)� = 5.27... × 10−30 C m = �.��� D
3
α′ = × (5.5432 cm3 mol−1 )
4πN A
3
= × (5.5432 cm3 mol−1 ) = 2.197 × 10−24 cm3
4π × (6.0221 × 1023 mol−1 )
1)�(ε r + 2) = ρPm �M, together with the relationship between refractive index
and relative permittivity which is given by [��A.��–���], n r = ε r . Combining
1�2
n 2r − 1 ρPm M n 2r − 1
= hence Pm = � �
n 2r + 2 M ρ n 2r + 2
M n 2r − 1 M n2 − 1 n2 − 1 RT n 2r − 1
Pm = � 2 �= � 2r � = Vm � 2r �= � �
ρ nr + 2 M�Vm n r + 2 nr + 2 p n 2r + 2
NA α NA µ2
Pm = +
3ε 0 9ε 0 kT
because as discussed above the term involving µ does not contribute at high fre-
quency. Assuming the high frequency contribution to be the same at ��� K, the
di�erence between this value and the static polarization Pm = 57.57 cm3 mol−1
at this temperature corresponds to the term N A µ 2 �9ε 0 kT.
NA µ2
= Pm,static − Pm,high frequency
9ε 0 kT
Hence
1�2
9ε 0 kT
µ=� (Pm,static − Pm,high frequency )�
NA
9 × (8.8542 × 10−12 J−1 C2 m−1 ) × (1.3806 × 10−23 J K−1 ) × (292 K)
=� ×
6.0221 × 1023 mol−1
1�2
�(57.57 × 10−6 m3 mol−1 ) − (5.73... × 10−6 m3 mol−1 )� �
to follow the �eld. �erefore the term µ 2 �3kT in the expression for Pm does
not contribute, and the equation reduces to Pm = N A α�3ε 0 .
Taking the polarizability of methanol as α = 3.59 × 10−40 J−1 C2 m2 from Ta-
ble ��A.� on page ��� gives
1 + 2ρPm �M
εr =
1 − ρPm �M
1 + 2 × (0.7914 g cm−3 ) × (8.13... cm3 mol−1 )�(32.0416 g mol−1 )
=
1 − (0.7914 g cm−3 ) × (8.13... cm3 mol−1 )�(32.0416 g mol−1 )
= 1.75... = �.��
O O HO
2
OH OH O
Figure 17.6
ε r − 1 ρPm 1 + 2ρPm �M
= hence εr =
εr + 2 M 1 − ρPm �M
�is equation implies that if the molar polarization Pm increases then ε r will
increase, because the numerator in the expression for ε r will increase while the
denominator will decrease. �e molar polarization in turn is given by [��A.��–
���], Pm = (N A �3ε 0 )(α + µ 2 �3kT).
�e individual molecules are polar but the dimers are not, because their sym-
metry means the individual dipoles associated with their polar bonds cancel
out. At low temperatures, a signi�cant proportion of molecules exist as the
non-polar dimer and therefore do not contribute to the µ 2 �3kT term in the
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 637
charges which form the dipole: these interactions are opposite in sign but dif-
ferent in magnitude because the distances between the point charge and the
two charges which form the dipole are not the same. As the point charge moves
further away, these interactions both decrease in magnitude. However, because
the distance between these charges is becoming less signi�cant compared to
the distance of the point charge, the magnitude of the interaction with each
becomes more similar. As a result, the terms with opposite sign come closer
to cancelling one another out and the overall interaction therefore goes to zero
faster than does the simple interaction between point charges.
E�ectively, as r increases the two charges which form the dipole merge and
begin to cancel one another. �is is why the potential energy of interaction
falls o� more quickly than it does for the interaction between two point charges.
For quadrupoles and high n-poles the e�ect is greater, leading to an even faster
fall o� of the interaction, presumably because the e�ective cancellation of the
charges happens more quickly when there are more charges forming the n-pole.
Solutions to exercises
E��B.�(a) �e shape of a linear quadrupole is given in Fig. ��A.� on page ���; an example
of such an arrangement is a CO� molecule which has negative charges on the
oxygen atoms and a balancing positive charge on the central carbon. Two such
quadrupoles are shown in Fig. ��.�, arranged so that they are collinear.
r
A B C D E F
-Q +2Q -Q -Q +2Q -Q
l l l l
Figure 17.7
Q2 Q2 l 4 6Q 2 l 4
= × 24x 4 = × 24 � � =
4πε 0 r 4πε 0 r r πε 0 r 5
=
3 × [4π × (8.8542 × 10−12 J−1 C2 m−1 )]2 × (1.3806 × 10−23 J K−1 ) × (298 K)
= 1.62... × 10−78 J m6
640 17 MOLECULAR INTERACTIONS
C 1.62... × 10−78 J m6
Hence �V � = − = − = −1.0 × 10−22 J . �is energy
r 6 (0.5 × 10−9 m)6
corresponds, a�er multiplication by N A , to −62 J mol−1 . �is is very much
smaller than the average molar kinetic energy of the molecules which, as ex-
plained in Section �A.�(a) on page ��, is given by
3
2
RT = 3
2
× (8.3145 J K−1 mol−1 ) × (298 K) = 3.7 kJ mol−1
µ 12 α 2′
V =−
4πε 0 r 6
�(1.85 D) × (3.3356 × 10−30 C m)�(1 D)� × (10.4 × 10−30 m3 )
2
=−
4π × (8.8542 × 10−12 J−1 C2 m−1 ) × (1.0 × 10−9 m)6
= −3.55... × 10−24 J
E��B.�(a) �e London formula for the energy of the dispersion interaction is given by
[��B.�–���]
C I1 I2
V = − 6 C = 32 α 1′ α 2′
r I1 + I2
In the case that the two interacting species are the same, with polarizability
volume α ′ and ionisation energy I, this expression becomes
I 2 3 ′2 3α ′2 I
C = − 32 α ′2 = α I hence V =−
2I 4 4r 6
Table ��A.� on page ��� in the Resource section gives the polarizability volume
of helium as α ′ = 0.20 × 10−30 m3 and Table �B.� on page ��� gives the �rst
ionisation energy as I = 2372.3 kJ mol−1 , therefore
E��B.�(a) �e interaction between a point charge and a point dipole orientated directly
away the charge is given by [��B.�–���], V = −µ 1 Q 2 �4πε 0 r 2 . In this case the
lithium ion has a charge of +e = +1.6022 × 10−19 C.
∆Vm = (1.77... × 10−21 kJ) × (6.0221 × 1023 mol−1 ) = 1.07 × 103 kJ mol−1
E��B.�(a) �e potential energy interaction between two parallel point dipoles separated
by distance r at angle Θ is given by [��B.�b–���], V = µ 1 µ 2 (1−3 cos2 Θ)�4πε 0 r 3 .
µ 1 µ 2 (1 − 3 cos2 Θ)
V=
4πε 0 r 3
(2.7 D) × (2.7 D) × �(3.3356 × 10−30 C m)�(1 D)� × (1 − 3 cos2 45○ )
2
=
4π × (8.8542 × 10−12 J−1 C2 m−1 ) × (3.0 × 10−9 m)3
= −1.3 × 10−23 J which corresponds to −8.1 J mol−1
Solutions to problems
P��B.� (a) In a vacuum the interaction energy is
P��B.� �e energy of a dipole µ in an electric �eld E is given in How is that done? ��A.�
on page ��� by E(θ) = −µE cos θ, where θ is the angle between the direction
of the dipole and the �eld. If the dipole is parallel to the �eld, θ = 0, this energy
is simply −µE, while if it is at θ = 90○ the energy is zero. �e dipole–induced
dipole interaction between a polar molecule such as water with dipole µ and a
polarizable species such as an argon atom with polarizability volume α ′ is given
by [��B.�–���], V = −µ 2 α ′ �4πε 0 r 6 . �e task is to �nd the distance r for which
642 17 MOLECULAR INTERACTIONS
this energy is equal to that of the dipole aligned parallel to the �eld, that is, for
which
µ2 α′
−µE = −
4πε 0 r 6
Rearranging, and noting that 1 V = 1 J C−1 , gives
µα ′
1�6
r=� �
4πε 0 E
(1.85 D) × (3.3356 × 10−30 C m)�(1 D) × (1.66 × 10−30 m3 )
1�6
=� �
4π × (8.8542 × 10−12 J−1 C2 m−1 ) × (1.0 × 103 V m−1 )
= 2.12... × 10−9 m = 2.1 nm
For distances closer than this the arrangement with the dipole directed towards
the argon atom is lower in energy than that with the dipole parallel to the �eld.
µ 12 α 2′
V =−
4πε 0 r 6
�(2.7 D) × (3.3356 × 10−30 C m)�(1 D)� × (1.04 × 10−29 m3 )
2
=−
4π × (8.8542 × 10−12 J−1 C2 m−1 ) × (0.4 × 10−9 m)6
= −1.85... × 10−21 J
P��B.� �e London formula for the energy of the dispersion interaction is given by
[��B.�–���]
C I1 I2
V = − 6 C = 32 α 1′ α 2′
r I1 + I2
Di�erentiation with respect to r gives
dV d C 6C I1 I2 1
F=− = − �− 6 � = − 7 = −9α 1 α 2
dr dr r r I1 + I2 r 7
P��B.� One way of rewriting the Lennard-Jones potential with an exponential repul-
sive term (the exp-� potential) is
r 0 12 r0 6 r0 6
4ε �� � −� � � → Aε �Be−r�r 0 − � � �
r r r
where the constants A and B are introduced to scale the overall potential and to
scale the repulsive term relative to the attractive term; their values can be found
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 643
Aε{Be−1 − 1} = 0
r0 6
V (r) = Aε �e1−r�r 0 − � � �
r
�e function in the curly braces is positive for r > r 0 and negative for r <
r 0 . �erefore, for it to represent a realistic potential with the attractive part
dominating at large distances, the constant A must be negative. Assuming this
to be so, the potential shows a minimum at a distance somewhat beyond r =
r 0 ; the minimum is located by solving dV (r)�dr = 0. �ere is no analytical
solution to the resulting equation, but mathematical so�ware is able to locate
the minimum numerically at r = 1.360 r 0 .
If, like the Lennard-Jones potential, the depth of the well is to be ε at this point,
then A = −1.853. A plot of the potential with this value of the constant is shown
in Fig. ��.�. For comparison, the Lennard-Jones potential is also plotted.
3
exp-�
Lennard-Jones
2
V (r)�ε
−1
1.0 1.5 2.0 2.5 3.0
r�r 0
Figure 17.8
17C Liquids
Answers to discussion questions
D��C.� �is is discussed in Section ��C.� on page ���.
644 17 MOLECULAR INTERACTIONS
Solutions to exercises
E��C.�(a) �e height climbed by a liquid in a capillary tube of radius r is given by [��C.�–
���], h = 2γ�ρg acc r, assuming that the contact angle is zero. Rearranging for
γ, taking ρ = 0.9982 g cm−3 = 998.2 kg m−3 , and noting that 1 N = 1 kg m s−2
gives
γ = 12 ρg acc rh
= 12 ×(998.2 kg m−3 ) × (9.807 m s−2 ) × (0.300 × 10−3 m) × (4.96 × 10−2 m)
= 0.0728... kg s−2 = 72.8 mN m−1
E��C.�(a) �e pressure di�erence between the inside and outside of a spherical droplet
is given by the Laplace equation [��C.�–���], p in = p out + 2γ�r. Hence, taking
the surface tension of water as ��.�� mN m−1 from Table ��C.� on page ���, and
noting that 1 Pa = 1 N m−2 ,
2γ 2 × (72.75 × 10−3 N m−1 )
∆p = p in − p out = = = ��� kPa
r (200 × 10−9 m)
Solutions to problems
P��C.� (a) �e function is plotted in Fig. ��.�
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 645
g(r)
1
0
0 2 4 6 8
r�r 0
Figure 17.9
�e plot resembles Fig. ��C.� on page ��� in that the function oscillates
for short values of r, corresponding to short-range order, but approaches
� for large separations.
(b) �e virial v 2 (r) = r(dV �dr) is obtained by di�erentiating the Lennard-
Jones expression for V
d r 0 12 r0 6 d
v 2 (r) = r 4ε �� � − � � � = r 4ε �r 012 r −12 − r 06 r −6 �
dr r r dr
= r × 4ε �−12r 012 r −13 + 6r 06 r −7 � = 4ε �−12r 012 r −12 + 6r 0 r −6 �
r 0 12 r0 6
= −24ε �2 � � −� � �
r r
�e quantity v 2 (r)�ε is plotted in Fig. ��.��; it falls away steeply to the le�
of the maximum.
2
v 2 (r)�ε
−2
0 1 2 3 4
r�r 0
Figure 17.10
646 17 MOLECULAR INTERACTIONS
17D Macromolecules
D��D.� Polymer melting occurs at a speci�c melting temperature, Tm , above which the
crystallinity of polymers can be destroyed by thermal motion. Higher melting
temperatures correspond to increased strength and number of intermolecular
interactions in the material. Polymers undergo a transition from a state of
high to low chain mobility at the glass transition temperature, Tg . �ere is
su�cient energy available to an elastomer at normal temperatures for limited
bond rotation to occur and the �exible chains writhe. At lower temperatures,
the amplitudes of the writhing motion decrease until a speci�c temperature, Tg ,
is reached at which motion is frozen completely and the sample forms a glass.
Solutions to exercises
E��D.�(a) As explained in Section ��D.�(c) on page ���, the root mean square separation
of the ends of a partially rigid chain with persistence length l p is given by the
value for a free chain, N 1�2 l, multiplied by a factor F, where F is given by
[��D.��–���], F = (2l p �l − 1)1�2 . �e contour length is given by [��D.�–���],
R c = N l, so a persistence length of �.�% of the contour length corresponds to
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 647
2 × (0.050N l)
1�2 1�2
2l p
F=� − 1� =� − 1� = (0.100N − 1)
1�2
l l
= (0.100 × 1000 − 1) = 9.94 . . .
1�2
E��D.�(a) By analogy with [��D.��–���], the radius of gyration R g of a partially rigid coil
is related to that of a freely jointed chain according to R g = F × R g,free where
F = (2l p �l −1)1�2 . �e radius of gyration for a three-dimensional freely jointed
chain is given by [��D.�b–���], R g,free = (N�6)1�2 l, so
1�2
2l p N 1�2
R g = F × R g,free = � − 1� ×� � l
l 6
Rearranging gives
2l p N 6R g2 2l p 2
l 6R g
R g2 = � − 1� � � l 2 hence = −1 hence lp = � 2 + 1�
l 6 Nl2 l 2 Nl
6 × (2.1 nm)2
2
l 6R g 0.150 nm
lp = � 2 + 1� = � + 1� = �.�� nm
2 Nl 2 1000 × (0.150 nm)2
E��D.�(a) Modelling the polyethene as a �D random coil perfect elastomer, the restoring
force is given by [��D.��a–���], F = (kT�2l) ln[(1 + λ)�(1 − λ)] where λ =
x�N l.
�e monomer of polyethene –[CH� CH� ]n – is taken to be CH� CH� , so the
number of monomers in the chain is given by
�e length of each CH� CH� unit is estimated as the length of a two C–C bonds:
one C–C bond in the centre and half a bond length either side where the unit
connects to carbons in adjacent units. From Table �C.� on page ��� a C–C
bond length is approximately ��� pm, so the monomer length l is taken as 2 ×
(154 pm) = 308 pm = 0.308 nm. �e value of λ corresponding to a �.� nm
extension is therefore
x 1.0 nm
λ= = = 1.40... × 10−3
N l (2.31... × 103 ) × (0.308 nm)
648 17 MOLECULAR INTERACTIONS
∑ i m i M i ∑ i (n i M i )M i ∑ i n i M i2
MW = = =
∑i m i ∑i n i M i ∑i n i M i
�e two polymers are present in equal amounts n 1 = n 2 = n, therefore
E��D.�(a) �e root mean square separation of the ends of a freely jointed one-dimensional
chain is given by [��D.�–���], R rms = N 1�2 l, where N is the number of monomer
units and l is the length of each unit. In this case
�e length of each CH� CH� unit is estimated as the length of a two C–C bonds:
one C–C bond in the centre and half a bond length either side where the unit
connects to carbons in adjacent units. From Table �C.� on page ��� in the
Resource section a C–C bond length is approximately ��� pm, so the monomer
length l is taken as 2 × (154 pm) = 308 pm. �e contour length and root mean
square separation are then given by
R rms = N 1�2 l = (9.98... × 103 )1�2 × (308 pm) = 3.07... × 104 pm = 30.8 nm
7.3 × 10−9 m
2
Rg 2
N =� � =� � = 2.2 × 103
l 154 × 10−12 m
650 17 MOLECULAR INTERACTIONS
E��D.�(a) �e probability that the ends of a one-dimensional random coil are a distance
nl apart is given by [��D.�–���], P = (2�πN)1�2 e−n �2N where N is the total
2
number of monomers in the chain and l is the length of each monomer unit.
�e monomer of polyethene –[CH� CH� ]n – is taken to be CH� CH� , so the
number of monomers in the chain is given by
M polymer 65 × 103 g mol−1
N= = = 2.31... × 103
M CH2 CH2 28.0516 g mol−1
�e length of each CH� CH� unit is estimated as the length of a two C–C bonds:
one C–C bond in the centre and half a bond length either side where the unit
connects to carbons in adjacent units. From Table �C.� on page ��� a C–C
bond length is approximately ��� pm, so the monomer length l is taken as 2 ×
(154 pm) = 308 pm.
If the end-to-end distance is d, then d = nl and hence n = d�l. In this case
d = 10 nm and l = 308 pm hence
d 10 × 10−9 m
n= = = 32.4...
l 308 × 10−12 m
�e probability of the ends being this distance apart is therefore
2 1�2 −n 2 �2N
P=� � e
πN
1�2
2
=� � × e−(32.4 ...) �2×(2.31 ...×10 3 )
= �.���
2
π × (2.31... × 103 )
a 3
3 1�2
f (r) = 4π � � r 2 e−a a=� �
2 2
r
π 1�2 2N l 2
where N is the number of monomers in the chain, l is the length of each monomer,
and f (r) dr is the probability that the ends of the chain are a distance between
r and r + dr apart.
�e monomer of polyethene –[CH� CH� ]n – is taken to be CH� CH� , so the
number of monomers in the chain is given by
M polymer 65 × 103 g mol−1
N= = = 2.31... × 103
M CH2 CH2 28.0516 g mol−1
�e length of each CH� CH� unit is estimated as the length of a two C–C bonds:
one C–C bond in the centre and half a bond length either side where the unit
connects to carbons in adjacent units. From Table �C.� on page ��� a C–C
bond length is approximately ��� pm, so the monomer length l is taken as 2 ×
(154 pm) = 308 pm = 0.308 nm. �erefore
1�2
3 1�2 3
a=� � =� � = 0.0826... nm−1
2N l 2 2 × (2.31... × 10 ) × (0.308 nm)2
3
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 651
3
a
f (10.0 nm) = 4π � � r 2 e−a
2 2
r
π 1�2
0.0826... nm−1
3
= 4π � � × (10.0 nm)2 × e−(0.0826 nm ) ×(10.0 nm)
−1 2 2
π 1�2
= 0.0642... nm−1
E��D.��(a) As explained in Section ��D.�(b) on page ���, the radius of gyration of a con-
strained chain is given by the value for a free chain multiplied by a factor F,
where F is given by [��D.�–���], F = [(1 − cos θ)�(1 + cos θ)] . For θ =
1�2
○
109 ,
1 − cos θ 1�2 1 − cos 109○ 1�2
F=� � =� � = 1.40...
1 + cos θ 1 + cos 109○
�is corresponds to a percentage increase of [(1.40...) − 1] × 100% = +��.�% .
�e volume is proportional to the cube of the radius, so the volume of the
constrained chain is related to that of a free chain by a factor of F 3 = (1.40...)3 =
2.75.... �is corresponds to a percentage increase of [(2.75...) − 1] × 100% =
+���% .
Solutions to problems
P��D.� �ere is some lack of clarity in the text over the de�nition of the radius of
gyration, R g . For a polymer consisting of N identical monomer units, R g is
de�ned as
N
R g2 = (1�N) � r 2i (��.�)
i=1
where r i is the distance of monomer unit i from the centre of mass. In other
words, the radius of gyration is the root-mean-square of the distance of the
monomer units from the centre of mass.
A related quantity is the moment of inertia I about an axis, which is de�ned in
the following way
N
I = � md i2 (��.�)
i=1
where m is the mass of the monomer unit and d i is the perpendicular distance
from the monomer to the axis. In general, the distance d i is not the same as r i :
the �rst is the perpendicular distance to the axis, the second is the distance to
the centre of mass.
A radius of gyration can be related to a moment of inertia by imagining a rigid
rotor consisting of a mass m tot equal to the total mass of the polymer held at a
distance R g from the origin; the moment of inertia of this rotor is I = m tot R g2 ,
and hence R g2 = I�m tot . However, note that this radius of gyration is associated
by the rotation about a particular axis.
652 17 MOLECULAR INTERACTIONS
(a) For a solid sphere the mass is distributed continuously rather than at
discrete points as for the simple polymer. �e de�nition in eqn ��.� is
adapted by imagining that the monomer unit at r i is a volume element dV
located at distance r from the centre of mass (the origin). �e sum over r 2i
becomes the integral of r 2 dV over the sphere, and division by N becomes
division by the volume of the sphere Vs . Hence R g2 = (1�Vs ) ∫sphere r 2 dV .
It is convenient to complete the calculation using spherical polar coordi-
nates
1 r=a θ=π �=2π
R g2 = � � � r 2 × r 2 sin θ dr dθ d�
Vs r=0 θ=0 �=0
3 r=a θ=π �=2π
= � r 4
dr � sin θ dθ � d�
4πa 3 r=0 θ=0 �=0
3 a5
= × × 2 × 2π
4πa 3 5
= 35 a 2
a4
=ρ × l × × 2π = ρl a 4 π�2 = m tot a 2 �2
4
where on the last line the total mass is given by m tot = ρ × πa 2 × l. A rigid
rotor with the same total mass has moment of inertia I = m tot R g,��
2
, hence
R g,�� = (2)−1�2 a .
To compute the moment of inertia perpendicular to the long axis, say
about the x-axis, it is necessary to know the perpendicular distance d
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 653
between an arbitrary point (x, y, z) and that axis. �is distance is that
between the points (x, y, z) and (x, 0, 0); by Pythagoras’ theorem d 2 =
y 2 + z 2 = r 2 sin2 � + z 2 . �e moment is inertia is therefore found from the
integral
a4
=ρ × l × × π = ρl a 4 π�4 = m tot a 2 �4
4
where the integral over � is found using Integral T.� with k = 1 and a =
2π.
A rigid rotor with the same total mass has moment of inertia I = m tot R g,�
2
,
hence R g,� = (a 2 �4 + l 2 �12)1�2 .
(c) �e speci�c volume is the volume divided by the mass, υ s = V �m, and
the mass of an individual macromolecule is given by M�N A , where M is
the molar mass.
V (4�3)πa 3 3υ s M
υs = = hence a3 =
m M�N A 4πN A
654 17 MOLECULAR INTERACTIONS
3 1�2 3υ s M 1�3
Rg = � � � �
5 4πN A
3 1�2 3 1�3
=� � � � (υ s M)
1�3
5 4πN A
1�3
3 1�2 3
R g �m = � � � �
5 4π × (6.0221 × 1023 mol−1 )
× �(υ s �cm3 g−1 )(M�g mol−1 ) × 10−6 �
1�3
For a rod R g,�� = (1�2)1�2 a = (1�2)1�2 × (0.50 nm) = �.�� nm . To �nd the
radius of gyration about the perpendicular axis requires a knowledge of l and
this is found from the speci�c volume in a similar way to the method used for
a sphere
V πa 2 l υs M
υs = = hence l = 2
m M�N A πa N A
υs M
l=
πa 2 N A
(0.750 × 10−6 m3 g−1 ) × (100 × 103 g mol−1 )
=
π × (0.50 × 10−9 m)2 × (6.0221 × 1023 mol−1 )
= 1.585... × 10−7 m = 158.5... nm
Hence
a 3
3 1�2
f (r) = 4π � � r 2 e−a a=� �
2 2
r
where
π 1�2 2N l 2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 655
π 1�2 π 8(a ) a
2 2
��� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �
0
3 3 2N l 2
= = � � = Nl2
2a 2 2 3
P��D.� �e walk is constructed by starting at the origin and taking steps of unit length
in a direction speci�ed by a randomly generated angle θ between 0 and 360○ .
Each step then involves incrementing the x coordinate by cos θ and the y co-
ordinate by sin�θ. �e �nal distance r reached at the end of the walk is found
calculating r = x 2 + y 2 where x and y are the �nal x and y coordinates.
10 10
5 5
0 0
−5 −5
−10 −10
−4 −2 0 2 4 −4 −2 0 2 4
Figure 17.11
Two such walks, of �� and ��� steps, are shown in Fig. ��.��. �e �nal values of
r in these cases are �.�� and �.��.
To investigate whether the mean and most probable values of r vary as N 1�2 ,
where N is the number of steps, a large number of random walks with vary-
ing numbers of steps are generated and the value of r found for each. �e
table shows the mean and most probable values of r estimated from samples
of ��� random walks carried out with each of ��, ��, ��, ��, ��, ��, ��, ��, ��
and ��� steps; the most probable values have been estimated by constructing a
histogram of the values of r, �tting a curve to the histogram, and �nding the
maximum of the curve.
If r mean and rmost probable vary as N 1�2 then plots of these values against N 1�2
should give a straight line passing through the origin. �e data are plotted in
Fig. ��.��. In both cases the the data fall on a reasonable straight line that almost
passes through the origin, thus indicating that rmean and rmost probable do indeed
vary as N 1�2 .
656 17 MOLECULAR INTERACTIONS
10 15
8
rmost probable
10
6
rmean
4
5
2
0 0
0 2 4 6 8 10 0 2 4 6 8 10
N 1�2 N 1�2
Figure 17.12
P��D.� �e formula given in the text for the radius of gyration of a sphere of radius R is
R g = (3�5)1�2 R. If the molecules given are globular, that is, roughly spherical,
their speci�c volume υ s should be given by υ s = V �m = (4�3)πR 3 �m where R
is the radius of the sphere and m is the mass of one molecule. Replacing m by
M�N A and rearranging gives
4
πR 3 3υ s M 1�3 3 1�2 3υ s M 1�3
υs = 3
hence R=� � hence Rg = � � � �
M�N A 4πN A 5 4πN A
In the last step R g = (3�5)1�2 R is used. Using this expression, the value of R g
expected for each of the molecules if they are spherical is calculated from the
υ s and M data, and compared to the experimental value of R g . If the values are
similar then there is evidence that the molecules are globular.
For serum albumin
For DNA
For serum albumin and bushy stunt virus the experimental radii of gyration
(�.�� nm and ��.� nm) are similar to the values that would be expected if these
macromolecules were spherical, thus suggesting that they are globular. In the
case of DNA the experimental radius of gyration (���.� nm) is much greater
than the value expected if it were spherical, suggesting that DNA is not globular
and therefore more rod-like.
P��D.� As explained in the How is that done? ��D.� on page ��� the restoring force
for an extended elastomer is given by F = −T(∂S�∂x)T . �is restoring force
is equal to the tension t required to keep the sample at a particular length,
hence t = −T(∂S�∂x)T . �e restoring force therefore depends on (∂S�∂x).
Extension of a polymer reduces the disorder and hence entropy of the chains,
so there is a tendency to revert to the more disordered non-extended state.
17E Self-assembly
Answers to discussion questions
D��E.� �e formation of micelles is favoured by the interaction between hydrocarbon
tails and is opposed by charge repulsion of the polar groups which are placed
close together at the micelle surface. As the salt concentration is increased, the
repulsion of head groups is reduced because their charges are partly shielded
by the ions of the salt. �is favours micelle formation causing the micelles
to be formed at a lower concentration and hence reducing the critical micelle
concentration.
Solutions to exercises
E��E.�(a) �e isoelectric point of a protein is the pH at which the protein has no net
charge and therefore is una�ected by an electric �eld. �is is the pH at which
658 17 MOLECULAR INTERACTIONS
Solutions to problems
P��E.� (a) �e data show that π increases by �.� for every additional CH� group.
�e R group in question, (CH� )� CH� , has two more CH� groups than
(CH� )� CH� which has π = 2.5, so the predicted value for (CH� )� CH� is
2.5 + 2 × 0.5 = �.� .
(b) �e data are plotted in Fig. ��.��. �e points fall on a reasonable straight
line, the equation of which is
�e slope and intercept of the line are therefore −1.49 and −1.95 .
−1.5
−2.0
−2.5
π
−3.0
−3.5
−0.2 0.0 0.2 0.4 0.6 0.8 1.0
log K I
Figure 17.13
[M N ]
K=
([M]total − N[M N ]) N
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 659
where the factors of 1�c −○ are omitted for clarity. For the case that K = 1 and
N = 2 this equation becomes
[M2 ]
1= ([M]total − 2[M2 ]) = M2
2
hence
([M]total − 2[M2 ])2
H +0.75 6 –0.29
C –0.79 C
H 4 H
C8 N
–0.58 7 H
9, 10, 11
H H
4
data in the table the components of the dipole moment are easily com-
puted in units of the elementary charge times Å, and then these values
are converted to Debye in the usual way. �e total dipole moment is
µ = (µ 2x + µ 2y + µ z2 )1�2 .
µ x = −0.461 e Å
= (−0.461 × 10−10 m) × (1.6022 × 10−19 C)�(3.3356 × 10−30 C m)
= −2.212 D
20
0
V �(J mol−1 )
−20
−40
−60
0 50 100 150 200 250 300 350
Θ�○
Figure 17.14
I��.� Starting from G = U − T S − tl, the di�erential is formed and the basic equation
dU = TdS + tdl is introduced to give
�e Maxwell relations are derived using the method described in Section �E.�(a)
on page ���. Because G is a state function, dG = −SdT − ldt is an exact
di�erential. �erefore according to the criterion in [�E.�–���], (dg�dy)x =
(dh�dx) y if d f = gdx + hdy is an exact di�erential, it follows that
∂S ∂l ∂S ∂l
−� � = −� � hence � � =� �
∂t T ∂T t ∂t T ∂T t
Applying the same approach to dA = −SdT + tdl gives
∂S ∂t ∂S ∂t
−� � =� � hence � � = −� �
∂l T ∂T l ∂l T ∂T l
To deduce the equation of state, the basic equation dU = TdS + tdl is divided
by dl and the condition of constant T is imposed
∂U ∂S ∂t ∂t
� � = T � � + t = −T � � + t = t − T � �
∂l T ∂l T ∂T l ∂T l
In the second step the Maxwell relation (∂S�∂l)T = −(∂T�∂t) l , derived above,
is used.
I��.� �e osmotic pressure Рis expressed in terms of the concentration [J] by the
virial-like expression of [�B.��–���], Π = RT �[J] + B[J]2 �. In this equation
the units of B are the inverse of the units of [J].
�e osmotic virial coe�cient arises largely from the e�ect of excluded volume.
If a solution of a macromolecule is imagined as being built by the successive
addition of macromolecules of e�ective radius a to the solvent, each one being
excluded by the ones that preceded it, then B is the excluded volume per mole
of molecules. �e volume of a molecule is v mol = (4�3)πa 3 , but the excluded
volume is determined by the smallest distance possible between centres of two
molecules, which is 2a. �erefore the excluded volume is (4�3)π(2a)3 = 8v mol
for a pair of molecules. �e volume excluded per molecule is one-half this
volume or 4v mol . �us, for an e�ective radius of a = γR g = 0.85R g the osmotic
virial coe�cient is
16π 16π
B = 4N A v mol = NA a3 = N A γ 3 R g3
3 3
(a) For a freely jointed chain R g = (N�6)1�2 l hence
16π N 3�2
B= NA γ3 � �
3 6
16π 4000 3�2
= (6.0221 × 1023 mol−1 ) × [0.85×(154 × 10−12 m)]3 � �
3 6
= 0.39 m3 mol−1 .
662 17 MOLECULAR INTERACTIONS
16π N 3�2
B= NA γ3 � �
3 3
16π 4000 3�2
= (6.0221 × 1023 mol−1 ) × [0.85×(154 × 10−12 m)]3 � �
3 3
= 1.1 m3 mol−1 .
I��.� (a) �e Lennard-Jones potential is given by [��B.��–���], VLJ (r) = 4ε{(r 0 �r)12 −
(r 0 �r)6 }. As shown in Fig. ��B.� on page ���, the depth of the potential
well is given by ε and the position of the minimum is given by r e = 21�6 r 0 .
̃ e = 1.51 × 10−23 J and
�e results for He� give the depth of the well as hc D
the position of the minimum as R = 297 pm so it follows that
re 297 pm
ε = 1.51 × 10−23 J and r0 = = = ��� pm
2 1�6 21�6
A plot of the potential with these values is shown in Fig. ��.��.
(b) �e Morse potential is given by [��C.�–���], VM (x) = hc D ̃ e (1 − e−ax )2 ,
̃
where x = r − r e and hc D e is the depth of the well. With this poten-
tial VM (0) = 0, in contrast to the Lennard-Jones potential for which
VLJ (∞) = 0. To compare the two, the Morse potential is replotted as
VM (x) = hc D̃ e �1 − e−a(r−r e ) �2 − hc D
̃ e ; this is shown plotted in Fig. ��.��
using the given values of hc D ̃ e = 1.51 × 10−23 J and a = 5.79 × 1010 m−1 .
4
Lennard-Jones
Morse
2
V �(10−23 J)
−2
250 300 350 400 450 500
r�pm
Figure 17.15
18 Solids
Solutions to exercises
E��A.�(a) �e separation of (hkl) planes d hk l of a cubic lattice is give by [��A.�a–���],
d hk l = a�(h 2 + k 2 + l 2 )1�2 .
(562 pm) (562 pm)
d 112 = = ��� pm d 110 = = ��� pm
(12 + 12 + 22 )1�2 (12 + 12 + 02 )1�2
(562 pm)
d 224 = = ��� pm
(22 + 22 + 42 )1�2
d hk l = (h 2 �a 2 + k 2 �b 2 + l 2 �c 2 )−1�2
−1�2
32 22 12
d 321 = � + + �
(812 pm)2 (947 pm)2 (637 pm)2
= (2.05... × 1019 )−1�2 m = ��� pm
664 18 SOLIDS
E��A.�(a) �e volume of an orthorhombic unit cell is given by V = abc, and the mass
of the unit cell m is given by m = ρV , where ρ is the mass density. Using the
estimate of mass density ρ = 3.9 g cm−3
m NM
ρ= =
V NA V
4 × (154.75 g mol−1 )
= = �.�� g cm−3
(6.0221 × 1023 mol−1 ) × [(634 × 784 × 516) × 10−36 m3 ]
E��A.�(a) Miller indices are of the form (hkl) where h, k, and l are the reciprocals of
the intersection distances along the a, b and c axes, respectively. If the recipro-
cal intersection distances are fractions then the Miller indices are achieved by
multiplying through by the lowest common denominator.
Solutions to problems
P��A.� A face-centred cubic unit cell has lattice points at its � corners and also at the
centres of its six faces. �erefore there are (8 × 18 + 6 × 12 ) = 4 lattice points per
unit cell. �e mass density ρ is therefore ρ = 4m�V , where m is the mass per
lattice point and the volume V is a 3 , where a is the unit cell dimension. �e
molar mass M is calculated from M = mN A
M = N A ρa 3 �4
= (6.0221 × 1023 mol−1 ) × (1.287 × 106 g m−3 ) × (12.3 × 10−9 m)3 × 1
4
P��A.� From Fig. ��A.� on page ��� it is seen that the unit cell can be envisaged as a
prism of height c whose base is rhombus with sides a and interior angle 120○ ,
which is depicted below.
60○
@
a@ 120○ @a@ x
@ @
a
�e area of the rhombus is ax = a × a sin(60○ ), hence the volume is V = cax =
√
c × [a × a sin(60○ )] = ( 3�2)a 2 c .
P��A.� For a monoclinic unit cell, V = abc sin β. From the information given, a =
1.377b and c = 1.436b. Because there are two napthalene molecules within
the unit cell it follows that the mass density is ρ = 2m�V , where m is the mass
per molecule given by m = M�N A , where M is the molar mass of napthalene
(128.1... g mol−1 ).
Using ρV = 2m and V = abc sin β, it follows that abc sin β = 2m�ρ and hence
abc = 2m�(ρ sin β) = 2M�(N A ρ sin β). �e product abc = 1.377 × 1.436 × b 3
and so
1�3
2M
b=� �
N A ρ sin β×1.377×1.436
2×(128.1... g mol−1 )
1�3
=� �
(6.0221 × 1023 mol−1 )×(1.152 × 106 g m−3 )×sin (122.82○ )×1.377×1.436
= ���.� pm
P��A.� �e mass of the unit cell m is given by m = N M�N A , where N is the number
of monomer units per unit cell and M is the molar mass of a monomer unit.
�e mass is also written in terms of the mass density ρ and the volume V as
m = ρV . Hence N M�N A = ρV and so N = ρN A V �M. �e molar mass is
V = abc sin β
= (1.0427 nm) × (0.8876 nm) × (1.3777 nm) × sin(93.254○ )
= 1.27... × 10−27 m3
hence
P��A.� Consider the two-dimensional lattice and planes shown in Fig. ��A.�� on page
���. �e (hk0) planes intersect the a, and b axes at distances a�h and b�k from
the origin, respectively. Using trigonometry,
d hk0 d hk0 h d hk0 d hk0 k
sin � = = cos � = =
a�h a b�k b
Because sin2 θ + cos2 θ = 1 it follows that
d hk0 h 2 d hk0 k 2
� � +� � =1
a b
Rearranging gives
1 h2 k2
2
= +
d hk0 a2 b2
Because the third side of the cell is mutually perpendicular to the other two,
the extension to three dimensions simply involves adding an additional term,
as in the derivation for a cubic lattice
1 h2 k2 l 2
2
= + +
d hk l a2 b2 c2
D��B.� �e scattering factor determines how strongly an atom scatters the X-rays, and
hence how strong the contribution from a particular atom is to a re�ection. It is
de�ned and described in Section ��B.�(c) on page ���. For forward scattering,
the scattering factor is equal to the number of electrons in the atom.
Solutions to exercises
E��B.�(a) �e Bragg law [��B.�b–���], λ = 2d sin(θ), is rearranged to give the glancing
angle as 2θ = 2 sin−1 (λ�2d), where d is the plane separation and λ is the
wavelength of the X-rays. For the case where λ = 154.433 pm,
E��B.�(a) In Section ��B.�(c) on page ��� it is shown that the scattering factor in the
forward direction, f (0), is equal to the total number of electrons in the species,
N e . �us for Br – f (0) = 36 .
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 667
�e indices h, k and l are all integers, and einπ = (−1)n for integer n. All the
exponents in the sum are even multiples of iπ, so all the exponential terms are
equal to +1. Hence Fhk l = f .
E��B.�(a) �e orthorhombic C unit cell is shown in Fig. ��A.� on page ���. �e structure
factor is given by [��B.�–���]
Fhk l = � f j ei� hk l ( j)
j
In this case there are a total of ten terms to include, h = 0 to 9. Figure ��.� shows
a plot of V ρ(x) against x; the electron density is at a maximum of 110�V at
x = 0.5, the centre of the unit cell.
V ρ(x) 100
50
1
P(r) = � �Fhk l � e
2 −2πi(hx+k y+l z)
V hk l
In this case the structure factors are only given for the x direction so the sum
is just over the index h. Furthermore, because Fh = F−h the summation can be
taken from from h = 0 to h = +∞. Using a similar line of argument to that in
Exercise E��B.�(a), the Patterson synthesis is
∞
V P(x) = F02 + 2 � Fh2 cos(2πhx)
h=1
In this case there are a total of ten terms to include, h = 0 to 9. Figure ��.� shows
a plot of V P(x) against x. As expected, there strong feature at the origin; this
arises from the separation between each atom and itself. �ere is also a strong
feature at x = 1 which indicate that atoms are separated by 1 × a unit along the
x-axis.
E��B.�(a) To constructor the Patterson map, choose the position of one atom to be the
origin (here, the boron). �en add peaks to the map corresponding to vec-
tors joining each pair of atoms (Fig. ��.�). Heavier atoms give more intense
contributions than light atoms, so peaks arising from F and F separations are
shown with greater diameter than those representing B and F separations. �e
vector between atom A and atom B has the same magnitude as that between
B and A, but points in the opposite direction; the map therefore includes two
symmetry related peaks on either side of the origin. �e vectors between each
atom and itself give a peak at the centre point of the Patterson map, and the
many contributions at this position create an intense peak.
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 669
1 000
500
V P(x)
F
RBF
RBF
B
RFF
F F
RFF
Figure 18.3
E��B.�(a) Using the de Broglie relation [�A.��–���], λ = h�p = h�(mυ), where p is the
momentum, m is the mass of a neutron and υ its speed, it follows that
h 6.6261 × 10−34 J s
υ= = = �.� km s−1
λm (65 × 10−12 m) × (1.6749 × 10−27 kg)
E��B.�(a) From the equipartition principle the kinetic energy is E k = 12 kT. �is en-
ergy can be written in terms of the momentum as p2 �(2m) and hence p =
(mkT)1�2 . �e de Broglie relation [�A.��–���], λ = h�p, is then used to �nd
the wavelength
h 6.6261 × 10−34 J s
λ= =
(mkT)1�2 [(1.6749 × 10−27 kg) × (1.3806 × 10−23 J K−1 ) × (350 K)]1�2
= 233 pm
E��B.��(a) Bragg’s law [��B.�b–���], λ = 2d sin θ, describes the relationship between wave-
length of the X-rays λ, the Bragg angle θ, and the plane separation d. �us
λ = 2 × (99.3 pm) × sin (20.85○ ) = ��.� pm
670 18 SOLIDS
E��B.��(a) As shown in Fig. ��B.�� on page ���, for the cubic I lattice re�ections from
planes with h + k + l = odd are absent from the di�raction pattern. Hence the
�rst three possible re�ections occur for planes (���), (���) and (���). Using the
Bragg law [��B.�b–���], λ = 2d hk l sin θ, and the expression for the spacing of
the planes [��A.�a–���], d hk l = a�(h 2 + k 2 + l 2 )1�2 , the following table is drawn
up
Solutions to problems
P��B.� �e NaCl unit cell is depicted in Fig. ��B.� on page ���. �e unit cell is a cube
with volume V = a 3 where a is the unit cell side length. �ere are eight Na+
ions at vertices, six Na+ ions on faces, one Cl – ion at centre and 12 Cl – ions
at the edges of the unit cell. �us in total there are four NaCl units per unit
cell. �e mass density ρ is ρ = m�V where m is the total mass per unit cell. It
follows that m = 4m NaCl = 4M�N A where M is the molar mass of an NaCl unit
(M = 58.44 g mol−1 ).
4M 4M 32M sin3 θ
ρ= = =
NA V NA a3 NA λ3
known parameters
1�3
32M
λ=� � sin θ
NA ρ
32 × (58.44 g mol−1 )
1�3
=� � sin(6.0○ )
(6.0221 × 1023 mol−1 ) × (2.17 × 106 g m−3 )
= ��� pm
P��B.� Combining Bragg’s law [��B.�b–���], λ = 2d sin θ, with the expression for the
the separation of planes for a cubic lattice [��A.�a–���], d hk l = a�(h 2 + k 2 +
l 2 )1�2 , gives sin θ = (λ�2a) (h 2 + k 2 + l 2 )1�2 .
�e �rst three re�ections for a cubic P lattice are (���), (���) and (���). Con-
sider the ratio of sin θ for the �rst two of these compared to ratio of sin θ for
the �rst two observed lines:
sin θ 110 (12 + 12 )1�2 sin θ 1st sin 22.171○
= = 1.41... = = 1.15...
sin θ 100 (12 )1�2 sin θ 2nd sin 19.076○
�ese do not match up, so the lattice is not cubic P. For cubic I the �rst three
re�ections are (���), (���) and (���); making the same comparison gives
Cubic F has four atoms per unit cell and so the mass density is ρ = 4m�V =
4M�N A V , where M is the molar mass of silver.
4M 4 × (107.87 g mol−1 )
ρ= = = ��.�� g cm−3
N A a 3 (6.0221 × 1023 mol−1 ) × (408.55 × 10−12 m)3
672 18 SOLIDS
3Z R 3Z −r cos kr R 3Z R
f (θ) = � r sin kr dr = � + 2 3 � cos kr dr
kR 0
3 kR 3 k 0 k R 0
−3Z cos kR 3Z sin kr R 3Z
= + � = (sin kR − kR cos kR)
k2 R2 k2 R3 k 0 k3 R3
6
R = 100 pm Z=3
R = 100 pm Z=6
4 R = 400 pm Z=3
f (θ)
Figure 18.4
A plot of f (θ) against (sin θ)�λ is shown in Fig. ��.�. In the forward direction,
θ = 0, the scattering factor is equal to Z, which is the expected result because
Z is equal to the number of electrons. �e scattering oscillates with increasing
angle but superimposed on this is an overall decay that becomes faster as R
increases. �us, the larger the atom, the more concentrated is the scattering in
the forward direction.
P��B.� �e structure factor is given by [��B.�–���], Fhk l = ∑ j f j e2πi(hx j +k y j +l z j ) . Each
atom A is shared between 8 unit cells and therefore has weight 18 , whereas the
B atom contributes to just one unit cell and so has weight �.
(a) For f A = f , f B = 0 then Fhk l = f for all (hkl) planes so the di�raction
pattern will display no systematic absences.
(b) For f B = 21 f A then Fhk l = f A [1 + 12 (−1)(h+k+l ) ]. For all re�ections with
(h + k + l) odd the intensity will be proportional to the square of 12 f A , and
for all re�ections with (h + k + l) even the intensity will be proportional
to the square of 32 f A .
(c) For f A = f B = f then Fhk l = f [1 + (−1)(h+k+l ) ]. For all re�ections with
(h + k + l) odd the structure factor is zero, and for all re�ections with
(h + k + l) even the structure factor is 2 f . �e di�raction pattern will
show systematic absences for lines where (h + k + l) is odd.
Solutions to exercises
E��C.�(a) �e lattice enthalpy ∆H L is the change in standard molar enthalpy for the pro-
cess MX(s) → M+ (g) + X− (g) and its equivalent. �e value of the lattice en-
thalpy is determined indirectly using a Born–Haber cycle, as shown in Fig. ��.�
(all quantities are given in kJ mol−1 ). From the cycle it follows that
−635 kJ mol−1 + ∆H L = (178 + 1735 + 249 − 141 + 844) kJ mol−1
Ca2+ (g)+O(g)
���
��� Ca2+ (g)+ 12 O� (g) −141 Ca2+ (g) + O – (g)
∆H L
����
Ca(g)+ 12 O� (g)
Figure 18.5
Figure 18.6
E��C.�(a) �e packing fraction is f = NVa �Vc where N is the number of spheres per unit
cell, Va = 4πR 3 �3 is the volume of each sphere of radius R, and Vc is the volume
of the unit cell.
(i) For a primitive cubic unit cell the spheres touch along the edges of the
cell, so the edges of the cube have length 2R and hence Vc = (2R)3 . �ere
is one sphere per unit cell, N = 1, and therefore
4πR 3 �3 π
f = = = �.����
8R 3 6
(ii) For a bcc unit cell, the spheres touch along the body diagonal so the
length of this diagonal is 4R. Imagine a right-angle triangle in which the
√ two sides are an edge of the
hypotenuse is the body diagonal, and the other
√
cube, length a, and a face diagonal, length 2a. It follows that (4R)
√ 3=
2
and as N = 2 it follows
√
that side of the cube is 4R� 3 so N = 2, Va = 4πR 3 �3 and
It follows √
Vc = (4R� 3)3 . �us
√
2 × 4πR 3 �3 3π
f = √ = = �.����
(4R� 3)3 8
(iii) For a fcc unit cell, the spheres touch along a face diagonal which therefore
has length 4R. If the edge of the cube has length a it√follows, by consid-
ering a face, that (4R)2 = a 2 + a 2 and hence a = 2 2R. �e volume is
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 675
√
therefore Vc = (2 2R)3 , and because N = 4 the packing fraction is
4 × 4πR 3 �3 π
f = √ = √ = �.����
16 2R 3 3 2
E��C.�(a) �e unit cell volume V is related to the packing density f and the atomic vol-
ume v by f V = v. Assuming the atoms can be approximated as spheres then
v = 4πR 3 �3 where R is the atomic radius. Using the packing densities calculated
in Exercise E��C.�(a) and the given data
�us transformation from hcp to bcc causes cell volume to expand by 1.6%
Solutions to problems
P��C.� �e packing fraction is f = NVa �Vc , where N is the number of atoms per unit
cell, Va = 4πR 3 �3 is the volume of an atom of radius R, and Vc is the unit cell
volume. �e structure of diamond is shown in Fig. ��C.�� on page ���: there
are � atoms at the vertices of the cell (weight 18 ), � atoms at the face-centres
(weight 12 ), and � atoms within the unit cell (weight �), giving a total of � atoms
per unit cell.
�e two nearest-neighbour atoms which touch along the body diagonal are at
locations (�,�,�) and ( 14 , 14 , 14 ), where the coordinates are expressed as fractions
of the length of the side of the unit cell, a. �ese two atoms are at the√ opposite
corners of a small cube with edge a�4. �e body diagonal of a cube is 3 times
√ edge, so it follows that the length of the body diagonal of this
the length of the
small cube is 3a�4. As the two atoms √ touch along this diagonal, this distance
is also equal to 2R, hence a = 8R� 3. �e packing fraction is therefore given
by
√
8 × 4πR 3 �3 8 × 4πR 3 �3 3π
f = = √ 3 = = �.����
a �8R� 3� 16
3
676 18 SOLIDS
P��C.� (a) Close-packed spheres form a face-centred cubic structure, which is shown
in Exercise E��C.�(b) to have a packing density of f = 0.7405. A sam-
ple of volume V of diamond therefore contains f V �(4πR 3 �3) carbon
atoms, where R is the atomic radius. �e mass of these carbon atoms is
f V �(4πR 3 �3)×(M�N A ), where M is the molar mass of carbon, therefore
the mass density is
P��C.� �e formation of a band in one dimension results in a set of states which spread,
to a �nite extent, above and below the energy of the original atomic orbital from
which the band is created (Fig. ��C.�(e) on page ���). If the system is extended
to two dimensions, each one of these original states itself becomes the starting
point for a band arising from overlap in the second dimension; this is illustrated
in a highly schematic way in Fig. ��.�.
�e original atomic orbital is indicated on the le�, and overlap of these results
in a one-dimensional band, indicated by the dotted lines. �en, each state in
this band itself gives rise to a further band when interactions are allowed in a
second dimension. �is is illustrated for the states at the very top and bottom
of the band (shown by dotted lines), and a selection of levels between. �e band
clearly increases in overall width, but in addition the density of states increases
in the centre of the two-dimensional band as many of one-dimensional bands
overlap here. In contrast, at the extremities of the band, fewer one-dimensional
bands are overlapping. �ese are only qualitative arguments, but they are in-
dicative of the origin for the change in the density of states which is indeed
observed.
P��C.� (a) �e rock salt (NaCl) structure, shown in Fig. ��C.�� on page ���, exhibits
sixfold coordination. Let the radius of the chloride ions be r l and that of
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 677
atomic
�D
�D
Figure 18.7
�z A z B �e 2 d ∗
C ′ e−d�d = A
∗
hence
4πε 0 d 2
Note that in this expression the distance d is now that which gives the minimum
∗
potential energy. Substituting this expression for C ′ e−d�d into eqn ��.� gives
�z A z B �N A e 2 d∗
E p,min = −A �1 − �
4πε 0 d d
Solutions to exercises
E��D.�(a) Poisson’s ratio, ν P , is de�ned in [��D.�–���], ν P = εtrans �εnorm , where εtrans is
the transverse strain and εnorm is the normal (uniaxial) strain. If the normal
strain is �.�%, it follows that the change in length ∆Lnorm is
It is expected that the result of applying the stress will be to decrease the size
of the cube in the transverse dimension (that is ∆Ltrans is negative), and that
the decrease will be the same in each transverse direction. �e volume a�er the
stress has been applied is therefore
E��D.�(a) �e relationship between the applied pressure p, the bulk modulus K, and the
fractional change in volume ∆V �V is given by [��D.�b–���], K = p�(∆V �V ).
For a fractional change of �%, ∆V �V = 0.01, the pressure is p = 0.01 × 3.43 ×
109 Pa = ��.� MPa .
E��D.�(a) �e Young’s modulus E is related to the stress σ and the strain ε by [��D.�a–���],
E = σ�ε. �e stress is given by σ = F�A where F is the force applied and A is
the cross-sectional area. Hence
F 500 N
σ= = = 1.59... × 108 Pa = 1.6 × 102 MPa
A π(1.0 × 10−3 m)2
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 679
∆L σ 1.59... × 108 Pa
ε= = = = 0.036...
L E 4.42 × 109 Pa
Hence the percentage increase in length L is 3.6%
Solutions to problems
P��D.�
µ(3λ + 2µ) 3λ + 2µ λ
E= [i] K= [ii] G = µ [iii] νP = [iv]
λ+µ 3 2(λ + µ)
2ν P µ
λ= [v]
1 − 2ν P
Substituting [v] into [i], and then using [iii] in the �nal step gives
6ν P µ+2µ(1−2ν P )
µ� � 6ν P µ + 2µ(1 − 2ν P )
E= = = 2µ(1 + ν P ) = 2G(1 + ν P )
1−2ν P
2ν µ+µ(1−2ν )
� P 1−2ν P P � 2ν P + (1 − 2ν P )
as required
Solutions to exercises
E��E.�(a) �e Fermi–Dirac distribution is given by [��E.�b–���], f (E) = 1�[e(E−E F )�k T +
1], where f (E) is the probability of occupation of a state with energy E, and
E F is the Fermi energy. In this case E F = 1.00 eV = 1.60... × 10−19 J, using the
conversion factor from inside of the front cover. With some rearrangement of
the expression for f (E) it follows that
E = kT ln[1� f (E) − 1] + E F
= (1.3806 × 10−23 J K−1 ) × (298 K) × ln (1�0.25 − 1) + (1.60... × 10−19 J)
= 1.64... × 10−19 J = �.�� eV
E��E.�(a) Assuming that the temperature, T, is not so high that many electrons are excited
to states above the Fermi energy, E F , the Fermi–Dirac distribution can be writ-
ten as [��E.�b–���], f (E) = 1�[e(E−E F )�k T + 1], where f (E) is the probability
of occupation of a state with energy E.
For E = E F + kT, f (E F + kT) = 1�[e(E F +k T−E F )�k T + 1] = 1�[e1 + 1] = �.���
Solutions to problems
P��E.� �e Fermi–Dirac distribution is given by [��E.�b–���], f (E) = 1�[e(E−E F )�k T +
1], where f (E) is the probability of occupation of a state with energy E, and
E F is the Fermi energy. Let x = (E − E F )�E F and y = E F �kT so that f (E) can
be written as f (x, y) = 1�(ex y + 1). Note that x can be negative for energies
below the Fermi energy, but y must always be positive.
A set of curves for di�erent combinations of x and y are shown in Fig. ��.�.
Note that as T → ∞, y → 0 and f → 12 since all available energy states have the
same probability of 12 of being occupied. Also, as T → 0, y → ∞ and f tends
towards a step distribution for which f = 1 for x < 0 and f = 0 for x > 0.
P��E.� Substituting eqn [��E.�a–���] into eqn [��E.�–���] and integrating over the full
energy range gives
∞ ∞ ∞ ρ(E)
N =� dN(E) = � ρ(E) f (E) dE = � dE
0 0 0 e(E−µ)�k T + 1
f (E) 0.5
0 4
−4 −2 2
0 2 EF
4 0 kT
E−E F
EF
Figure 18.8
will have a net magnetic moment. �is is the origin of the magnetization of the
sample.
�e same idea applies when considering the interaction between a permanent
electric dipole and an applied electric �eld. In the presence of the �eld the
dipoles favour certain directions and so when averaged over the sample they
do not cancel. �e result is a net electric moment of the sample, called the
polarization. In both cases the magnetization or polarization depends on the
competition between the randomizing e�ect of thermal motion and the order-
ing e�ect of the applied �eld.
Solutions to exercises
E��F.�(a) �e molar susceptibility χ m of a substance is given by [��F.�b–���], the Curie
law,
C N A g e2 µ 0 µ B2 S(S + 1)
χm = where C =
T 3k
�is is rearranged to give the spin quantum number as
3kT χ m
S(S + 1) =
N A g e2 µ 0 µ B2
3 × (1.3806 × 10−23 J K−1 ) × (294.53 K)
=
(6.0221 × 1023 mol−1 ) × (2.0023)2
1.463 × 10−7 m3 mol−1
×
(1.2566 × 10−6 J s2 C−2 m−1 ) × (9.2740 × 10−24 J T−1 )2
= 6.839...
Note the conversion of the molar magnetic susceptibility from units of cm3 mol−1
to m3 mol−1 . To sort out the units the relations 1 T = 1 kg s−2 A−1 and 1 A =
1 C s−1 , hence 1 C = 1 A s, are useful. �e value of S is found by solving the
quadratic
√
S 2 + S − 6.839... = 0 S = 12 (−1 ± 1 + 4 × 1 × 6.839...) = −0.500 ± 2.662...
C N A g e2 µ 0 µ B2 S(S + 1)
χm = where C=
T 3k
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 683
Solutions to problems
P��F.� �e spin contribution to the molar magnetic susceptibility is given by the Curie
lawm [��F.�b–���],
C N A g e2 µ 0 µ B2 S(S + 1)
χm = where C=
T 3k
684 18 SOLIDS
Solutions to exercises
E��G.�(a) �e energy gap is given by
hc (6.6261 × 10−34 J s) × (2.9979 × 108 m s−1 )
∆ε = hν = =
λ (350 × 10−9 m)
= 5.67... × 10−19 J
Converting to eV the band gap is
(5.67... × 10−19 J) × [1 eV�(1.6022 × 10−19 J eV−1 )] = �.�� eV
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 685
Solutions to problems
P��G.� (a) To �nd whether or not a vector is an eigenvector of the hamiltonian ma-
trix, the matrix is allowed to act on the vector
ν̃mon β̃ 1 ν̃ + β̃ 1
� � � � = � mon � = (ν̃mon + β̃) � �
β̃ ν̃mon 1 ν̃mon + β̃ 1
Acting on the vector with the hamiltonian matrix regenerates the original
vector times a constant, which is the eigenvalue ν̃+ = ν̃mon + β̃. Similarly,
for the second proposed eigenvector
ν̃mon β̃ 1 ν̃ − β̃ 1
� �� � = � mon � = (ν̃mon − β̃) � �
β̃ ν̃mon −1 −ν̃mon + β̃ −1
Assuming that Ψb (i) is normalised the �rst and second integrals are = 1,
and with the de�nition S = ∫ Ψb∗ (1)Ψb (2) dτ the third term is 2S; overall
I = 2(1+S). Division of the wavefunction by I 1�2 therefore normalizes the
function, so the normalization constant is N+ = [2(1 + S)]−1�2 . A similar
calculation gives the normalization constant for Ψ− as N− = [2(1−S)]−1�2
(c) �e integral µ dim = ∫ Ψ±∗ µ̂Ψ0 dτ is evaluated by substituting in the given
forms of Ψ± and Ψ0 and using the de�nition µ mon = ∫ Ψb∗ (i) µ̂Ψa (i) dτ.
Note that ∫ Ψb∗ (2) µ̂Ψa (1) dτ = ∫ Ψb∗ (1) µ̂Ψa (2) dτ = 0 because these
correspond to transitions from a level of one monomer to a level of the
other monomer.
µ dim = � Ψ±∗ µ̂Ψ0 d τ
1 1
=� [Ψb (1) ± Ψb (2)]∗ µ̂ 1�2 [Ψa (1) + Ψa (2)] dτ
[2(1 ± S)]1�2 2
� ��� � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ��� � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � ��
µ mon µ mon
�
=
1 � � Ψ ∗ (1) µ̂Ψa (1) dτ ± � Ψ ∗ (2) µ̂Ψa (2) dτ
2(1 ± S) �
1�2 �
b b
�
=0 =0
��� � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � �� ��� � � � � � � � � � � � � � � � � � � � � � � � � � � �� � � � � � � � � � � � � � � � � � � � � � � � � � � � �� �
�
± � Ψb∗ (2) µ̂Ψa (1) dτ + � Ψb∗ (1) µ̂Ψa (2) dτ � �
�
�
1
= (µ mon ± µ mon )
2(1 ± S)1�2
�e trigonometric identities
increases the scattering factor falls o�, but the decay is slower for the more
compact orbital with Z = 2.
For a �s hydrogenic orbital
1.0
Z=1
0.8 Z=2
0.6
f
0.4
0.2
0.0
0.000 0.002 0.004 0.006 0.008 0.010
ξ�(1012 m)
Figure 18.9
1.0
Z=1
Z=2
0.5
f
0.0
0.000 0.001 0.002 0.003 0.004 0.005
12
ξ�(10 m)
Figure 18.10
λ = 2d sin θ, to give
1 δV 1 a 3 (300 K) − a 3 (100 K)
α= = 3
V δT a (100 K) δT
1 (356.66... pm)3 − (355.53... pm)3
= = 4.811 × 10−5 K−1
(355.53... pm)3 (300 − 100) K
Solutions to exercises
E��A.�(a) For a process to be spontaneous it must be accompanied by a reduction in the
Gibbs energy, that is ∆G < 0. �e adsorption of a gas on a surface is likely to
be accompanied by a signi�cant reduction in entropy on account of the loss of
translational degrees of freedom, therefore ∆S < 0. Given that ∆G = ∆H−T∆S,
a process with ∆S < 0 can only have ∆G < 0 if ∆H is su�ciently negative, that
is the process must be exothermic.
2.0158 g mol−1 .
p
Zw =
(2πMkT�N A )1�2
(0.10 × 10−6 Torr)×(133.32 Pa Torr−1 )×(6.0221 × 1023 mol−1 )1�2
=
�2π×(2.0158 × 10−3 kg mol−1 )×(1.3806 × 10−23 J K−1 )×(298.15 K)�
1�2
(ii) For propane, the molar mass M = 3×(12.011 g mol−1 )+8×(1.0079 g mol−1 ) =
44.096 g mol−1 .
p
Zw =
(2πMkT�N A )1�2
(0.10 × 10−6 Torr)×(133.32 Pa Torr−1 )×(6.0221 × 1023 mol−1 )1�2
=
�2π×(44.096 × 10−3 kg mol−1 )×(1.3806 × 10−23 J K−1 )×(298.15 K)�
1�2
For an argon atom, the molar mass M = 39.95 g mol−1 . �us, for A = π(d�2)2 ,
where d is the diameter of the circular surface, rearranging the above expres-
sion gives
r(2πMkT�N A )1�2
p=
A
4.5 × 1020 s−1
=
π×(0.5 × 1.5 × 10−3 m)2
2π×(39.95 × 10−3 kg mol−1 )×(1.3806 × 10−23 J K−1 )×(425 K)
1�2
� �
6.0221 × 1023 mol−1
= 1.25... × 104 Pa = 0.13 bar
E��A.�(a) For a perfect gas, and at constant temperature, p ∝ 1�V , where V is the volume
occupied by the gas at pressure p. �erefore
p 2 V1 V1 p 1
= hence V2 =
p 1 V2 p2
a complete monolayer, but where the volume has been corrected to the same
pressure p.
At 5.0 bar, the volume adsorbed is a complete monolayer and thus V∞ = 22 cm3
at 5.0 bar. At 0.1 bar, this same volume is
(22 cm3 ) × (5.0 bar)
V∞,0.1 bar = = 1100 cm3
0.1 bar
Hence the surface coverage is
10 cm3
θ= = 9.1 × 10−3
1100 cm3
Solutions to problems
P��A.� �e Coulombic energy of interaction of the test ion with a section of lattice is
determined by summing the interaction energy of this ion with each of the ions
in the section of the lattice. Interactions between ions of opposite charge make
a negative contribution to the energy of −C�r, where C is a positive constant
and r is the distance between the test ion and an ion in the lattice. Similarly,
interactions between ions of the same charge make a positive contribution to
the energy of +C�r. De�ne a 0 as the distance between nearest neighbours in
the lattice, that is the lattice spacing.
(a) For a Type � section, and considering the nearest �� ions only, the inter-
action energy with the test ion is
C 1 1 1 C 10 (−1)n C
E2 = × �−1 + − + ... + � = � � � = −0.646 � �
a0 2 3 10 a 0 n=1 n a0
(b) For a Type � section of lattice, with �� atoms in each direction, the inter-
action energy with the test ion is
C 10 10 (−1)n+m C
E1 = � �� � 2 = +0.259 � �
a 0 n=1 m=1 (n + m )
2 1�2 a0
(c) To calculate the energy of interaction between the test ion and the lattice
in arrangement (a), observe that there is one Type � interaction and two
Type � interactions. Hence the interaction energy is given by
C C
E(a) = E 2 + 2E 1 = � � × (−0.646 + 2 × 0.259) = −0.128 � �
a0 a0
To calculate the energy of interaction between the test ion and the lattice
in arrangement (b), observe that there are two Type � interactions and
three Type � interactions. Hence the interaction energy is given by
C C
E(b) = 2E 2 + 3E 1 = � � × [2(−0.646) + 3(0.259)] = −0.516 � �
a0 a0
�e energy of interaction of the probe cation is much lower for (b) than
for (a), therefore (b) is the more favourable arrangement .
692 19 PROCESSES AT SOLID SURFACES
P��A.� From inside the front cover, 760 Torr = 1 atm = 1.01325 × 105 Pa, therefore
1 Torr is 133.32 Pa. �e unit cell for a face-centred cubic lattice is shown in
Fig. ��A.� on page ��� (cubic F) and how the planes are identi�ed using Miller
indices is described in Section ��A.�(a) on page ���.
(a) �e (���) plane is the face of the cube and the arrangement of the atoms
in the plane is shown in Fig. ��.�. �ere two atoms in this face, being the
total of one in the centre and a quarter of each of the four atoms at the
corners. Each atom has surface area πr 2 , where r is the atomic radius,
and the area of the face is a 2 = (352 × 10−12 pm)2 = 1.24 × 10−15 cm2 .
(���) surface
Figure 19.1
(���) surface
r
√
2a
Figure 19.2
√ 2
�e area of the face is 2a , therefore the surface number density n is
2 2
n= √ =√
2a 2 2 × (352 × 10−12 pm)2
= 1.14 × 1019 m−2 = 1.14 × 1015 cm−2
Similar calculations to those above give, for H� , f = 9.4 × 105 s−1 and
f = 0.13 s−1 at p = 100 Pa and p = 0.10 µTorr, respectively. For propane
the rates are f = 2.0 × 105 s−1 and f = 2.7 × 10−2 s−1 .
(c) �e (���) plane has the surface structure shown in Fig. ��.�; there is one
atom on this face.
(���) surface
2r
2r
Figure 19.3
√ √ 2
�e area of this rhombus is 2r × 3r = 2 3r √ . Because the atoms touch
√ 2 √ 2
along the face diagonals it follows that 4r = 2a, hence r 2
= a 2 �8. Using
this the area of the face is 2 3r = 3a �4, and thus the surface number
694 19 PROCESSES AT SOLID SURFACES
density is
1 1
n= √ =√
3a 2 �4 3 × (352 × 10−12 pm)2 �4
= 1.86 × 1019 m−2 = 1.86 × 1015 cm−2
Similar calculations to those above give, for H� , f = 5.8 × 105 s−1 and
f = 7.7 × 10−2 s−1 at p = 100 Pa and p = 0.10 µTorr, respectively. For
propane the rates are f = 1.2 × 105 s−1 and f = 1.6 × 10−2 s−1 .
For the BET isotherm assumption (�) is removed so that multi-layer coverage
is possible. In the derivation of this isotherm a distinction is made between the
energetics involved in forming the �rst and subsequent layers.
Solutions to exercises
E��B.�(a) �e isosteric enthalpy of adsorption is de�ne as [��B.�b–���]
∂ ln(α p−○ ) ∆ ad H −○
� � =−
∂(1�T) θ R
E��B.�(a) �e rate constant for desorption is assumed to follow an Arrhenius law, k des =
Ae−E a,des �RT . Recall that for a �rst order process the half life is simply propor-
tional to the inverse of the rate constant, therefore the time needed for a certain
amount to desorb is also inversely proportional to the rate constant. �us
E��B.�(a) �e half-life for a species on the surface is given by [��B.��–���], t 1�2 = τ 0 eE a,des �RT .
at ��� K t 1�2 = (0.1 ps) e(15×10 J mol−1 )�[(8.3145 J K−1 mol−1 )×(400 K)]
3
= 9.1 ps
at ���� K t 1�2 = (0.1 ps) e(15×10 J mol−1 )�[(8.3145 J K−1 mol−1 )×(1000 K)]
3
= 0.61 ps
at ��� K t 1�2 = (0.1 ps) e(150×10 J mol−1 )�[(8.3145 J K−1 mol−1 )×(400 K)]
3
To eliminate α �rst multiply the le�-hand equation by 1�p 2 and the right-hand
equation by 1�p 1
V∞ 1 1 V∞ 1 1
= + = +
p 2 V1 α p 1 p 2 p 2 p 1 V2 α p 1 p 2 p 1
where for the last step top and bottom are multiplied by p 1 p 2 .
With the data given
p1 − p2
V∞ =
p 1 �V1 − p 2 �V2
(145.4 Torr) − (760 Torr)
= = ��.� cm3
(145.4 Torr)�(0.286 cm3 ) − (760 Torr)�(1.443 cm3 )
t 1�2 = (1.0 × 10−14 s) e(120×10 J mol−1 )�[(8.3145 J K−1 mol−1 )×(400 K)]
3
= 47 s
θ θ
αp = hence p=
1−θ α(1 − θ)
698 19 PROCESSES AT SOLID SURFACES
∂ ln(α p−○ ) ∆ ad H −○
� � =−
∂(1�T) θ R
�e data gives the enthalpy of desorption as +10.2 J for 1.00 mmol of gas,
therefore the molar enthalpy of adsorption is −10.2 kJ mol−1 .
−10.2 kJ mol−1 1 1
ln (p 2 �kPa) = ln(12 kPa) + −1 −1 � − � = 2.68...
8.3145 J K mol 313 K 298 K
Solutions to problems
P��B.� (a) �e Langmuir isotherm is [��B.�–���], θ = α p�(1 + α p), inverting both
sides gives 1�θ = 1�α p + 1. Figure ��.� shows a plot of 1�θ against 1�p for
three di�erent values of α.
(b) �e Langmuir isotherm for adsorption with dissociation is [��B.�–���],
θ = (α p)1�2 �[1 + (α p)1�2 ]. Figure ��.� shows a plot of 1�θ against 1�p for
the same values of α used in Fig. ��.�. In contrast to the straight lines
seen in Fig. ��.�, for the case of adsorption with dissociation the 1�θ
against 1�p plot shows pronounced curvature. Such a plot may there-
fore in principle make it possible to distinguish between dissociative and
non-dissociative adsorption.
(c) In [��B.�–���] the BET isotherm is manipulated into a straight-line plot
z 1 (c − 1) zVmon 1 (c − 1)
= + z or = + z
(1 − z)V cVmon cVmon (1 − z)V c c
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 699
100
α = 2 atm−1
80 α = 1 atm−1
α = 0.5 atm−1
60
1�θ
40
20
0
0 20 40 60 80 100
(1�p)�atm −1
Figure 19.4
20
α = 2 atm−1
α = 1 atm−1
α = 0.5 atm−1
15
10
1�θ
0
0 20 40 60 80 100
(1�p)�atm−1
Figure 19.5
z 1 (c − 1)
= + z z = p�p∗
(1 − z)V cVmon cVmon
�us a plot of z�(1 − z)V against z is expected to be a straight line with slope
(c − 1)�cVmon and intercept 1�cVmon ; note that (slope)/(intercept) = c − 1. For
brevity the term z�(1 − z)V is denoted y.
(a) �e data are shown in the table below and the plot is shown in Fig. ��.�.
700 19 PROCESSES AT SOLID SURFACES
10
c=2
c=4
8
c = 10
zVmon �(1 − z)V
6
0
0 2 4 6 8 10
z
Figure 19.6
0.020
0.015
y�(cm−3 )
0.010
0.005
0.000
0.00 0.05 0.10 0.15 0.20 0.25
z
Figure 19.7
(b) �e data are shown in the table below and the plot is shown in Fig. ��.�.
0.010
y�(cm−3 )
0.005
0.000
0.00 0.02 0.04 0.06 0.08 0.10 0.12
z
Figure 19.8
slope 0.07953
= c−1 hence c =1+ = 263
intercept 3.036 × 10−4
60
40
20
50 100 150 200
p�kPa
Figure 19.9
�e data plainly fall on a curve, rather than the expected straight line. On the
assumption that the isotherm is more likely to hold at low pressure, the �rst six
data points are used to construct the line, the equation of which is
∂ ln(α p−○ ) ∆ ad H −○
� � =−
∂(1�T) θ R
For the data in this Problem p−○ is replaced by c −○ . �is equation implies that a
plot of ln(αc −○ ) against 1�T will have slope −∆ ad H −○ �R, for data at constant θ.
26.5
26.0
ln (αc −○ )
25.5
25.0
3.1 3.2 3.3 3.4 3.5 3.6
(1�T)�(10 −3
K )
−1
Figure 19.10
�e value of ∆ ad H −○ is obtained from the slope (the scatter means that high
precision on the result is not warranted)
∆ ad G −○ = ∆ ad H −○ − T −○ ∆ ad S −○
= (−20 kJ mol−1 ) − (300 K) × (+0.146 kJ K−1 mol−1 ) = −64 kJ mol−1
704 19 PROCESSES AT SOLID SURFACES
−1.5
−2.0
ln(w a �g)
−2.5
−3.0
−3.5
−3 −2 −1 0
ln([A]�c )
−
○
Figure 19.11
�e table of data is shown below and the plot is shown in Fig. ��.��; for brevity
the units of c and s are omitted throughout. It is clear that the data fall on a
curve and so do not conform to the Langmuir isotherm.
c s c�s
15.0 0.60 25.0
23.0 0.75 30.7
42.0 1.05 40.0
84.0 1.50 56.0
165 2.15 76.7
390 3.50 111.4
800 5.10 156.9
150
100
c�s
50
0
0 200 400 600 800
c
Figure 19.12
c s ln c ln s
15.0 0.60 2.71 −0.511
23.0 0.75 3.14 −0.288
42.0 1.05 3.74 0.049
84.0 1.50 4.43 0.405
165 2.15 5.11 0.765
390 3.50 5.97 1.253
800 5.10 6.68 1.629
ln s
3 4 5 6
ln c
Figure 19.13
c s sTemkin sFreundlich
15.0 0.60 0.11 0.60
23.0 0.75 0.57 0.75
42.0 1.05 1.23 1.04
84.0 1.50 1.98 1.51
165.0 2.15 2.71 2.18
390.0 3.50 3.64 3.46
800.0 5.10 4.42 5.09
(α A p A + 1)θ A + α A p A θ B = α A p A α B p B θ A + (α B p B + 1)θ B = α B p B
SOLUTIONS MANUAL TO ACCOMPANY ATKINS’ PHYSICAL CHEMISTRY 707
�e terms in θ A are now the same, and so will disappear when the two equations
are subtracted to give
Solutions to exercises
E��C.�(a) �e amount in moles of N� gas is found using the perfect gas law; close attention
to the units is needed.
pV (760 Torr)×(1.01325 × 105 Pa) 3.86 × 10−6 m3
n= = ×
RT 760 Torr (8.3145 J K−1 mol−1 ) × (273.15 K)
= 1.72... × 10−4 mol
Solutions to problems
P��C.� (a) �e Langmuir–Hinshelwood rate law is given in [��C.�b–���]
kr αA αB pA pB
υ=
(1 + α A p A + α B p B )2
708 19 PROCESSES AT SOLID SURFACES
(b) When the partial pressures of the reactants are low, α A p A � 1 and α B p B �
1, and therefore these terms may be ignored in the denominator to give
υlow = k r α A α B p A p B
In this limit the rate law is −1 order in A, �rst order in B, and therefore
overall zeroth order. It does not appear to be possible to achieve zeroth
order for either A or B alone.
αp 1 + αp − αp
θu = 1 − =
1 + αp 1 + αp
1 1
= ≈
1 + αp αp
[p 0 ln p − p]� p 0 = − 23 k c t
p
p 0 ln(p�p 0 ) − (p − p 0 ) = − 23 k c t
0.0
−0.2
y�kPa
−0.4
−0.6
0 50 100 150 200 250
t�s
Figure 19.14
Solutions to exercises
E��D.�(a) (i) �e Butler–Volmer equation is [��D.�–���], j = j 0 (e(1−α) f η −e−α f η ). For
H+ on Ni j 0 = 6.3×10−6 A cm−2 and α = 0.58; at ���.�� K, f = 38.921 V−1 .
For an overpotential of +0.20 V the current density is
(ii) If the current is entirely anodic, only the �rst term is needed
j = (0.79 mA cm−2 )
× (e(1−0.5)×(38.921 V − e−0.5×(38.921 V )
−1 −1
)×(0.010 V) )×(0.010 V)
= 0.31 mA cm−2
j = (0.79 mA cm−2 )
× (e(1−0.5)×(38.921 V − e−0.5×(38.921 V )
−1 −1
)×(0.100 V) )×(0.100 V)
= 5.4 mA cm−2
j = (0.79 mA cm−2 )
× (e(1−0.5)×(38.921 V − e−0.5×(38.921 V )
−1 −1
)×(−5.0 V) )×(−5.0 V)
E��D.�(a) At equilibrium, only the exchange current �ows, therefore for an electrode with
area A the current is j 0 A, and thus the charge passing in time t is (current ×
time): q = j 0 At. If each species passing through the double layer carries one
fundamental change, the number of charges is N = q�e = j 0 At�e. �us the
number per second through an area of �.� cm2 is, for H+ /Pt,
A similar calculation for Fe�+ /Pt gives 1.6 × 1016 s−1 , and for H+ /Pb the result
is 3.1 × 107 s−1 .
�e number of atoms covering � cm2 of electrode is (10−4 m2 )�(280×10−12 m)2
= 1.27... × 1015 . �erefore for H+ /Pt the number of times per second that each
atom is involved in a electron transfer event is (number of such events)/(number
of atoms) = (4.93... × 1015 s−1 )�(1.27... × 1015 ) = �.� s−1 . Similar calculations
for Fe�+ /Pt and H+ /Pb give �� s−1 and 2.4 × 10−8 s−1 , respectively. For H+ /Pb
the time between events is more than � year.
E��D.�(a) In the linear region the current density and overpotential are related by [��D.�–
���], η = RT j�F j 0 , therefore the current density is j = ηF j 0 �RT. For an
electrode of area A the current is I = jA, and therefore the resistance is
η η RT
r= = =
I ηF j 0 A�RT F j 0 A
For H+ /Pt
�e units are resolved by using (from inside the front cover) 1 V = 1 J C−1 and
1 Ω = 1 V A−1 . A similar calculation for H+ /Hg gives 3.3 × 1010 Ω .
E��D.�(a) Because the standard potential of Zn2+ /Zn is −0.76 V, under standard condi-
tions Zn metal will only be deposited when the applied potential is more nega-
tive than −0.76 V. �e current density is given by [��D.�–���], j = j 0 (e(1−α) f η −
e−α f η ), but under these conditions only the second term (the cathodic current)
is signi�cant. Using the data given for H+ , assuming α = 0.5, and recalling that,
at ���.�� K, f = 38.921 V−1
j H+ = − j 0 e−α f η
= −(50 × 10−12 A cm−2 ) e−0.5×(38.921 V = −1.3 × 10−4 A cm−2
−1
)×(−0.76 V)
It is usually considered that the metal can be deposited if the current density
for discharge of H+ is less than about 1 mA cm−2 , which is satis�ed in this case,
but not by a large margin. �e expectation is that zinc metal will be deposited,
but accompanied by signi�cant evolution of H� due to discharge of H+ .
712 19 PROCESSES AT SOLID SURFACES
E��D.�(a) If the anodic process is dominant, the current density is given by [��D.�a–���],
ln j = ln j 0 + (1 − α) f η, where f = F�RT. At ���.�� K
where the units are resolved by recalling 1 V = 1 J C−1 . Taking the di�erence of
two expressions for ln j for di�erent overpotentials gives
ln( j 2 � j 1 ) = (1 − α) f (η 2 − η 1 )
ln( j 2 � j 1 )
hence η 2 = + η1
(1 − α) f
ln(75�55.0)
= + 0.125 V = �.�� V
(1 − 0.39) × (38.921 V−1 )
E��D.�(a) If the anodic process is dominant, the current density is given by [��D.�a–���],
j = j 0 e(1−α) f η , where f = F�RT. At ���.�� K, f = 38.921 V−1 . Rearranging for
j 0 and then using the data given
j 0 = j e−(1−α) f η
= (55.0 mA cm−2 ) e−(1−0.39)×(38.921 V = �.� mA cm−2
−1
)×(0.125 V)
E��D.�(a) If the anodic process is dominant, the current density is given by [��D.�a–���],
j = j 0 e(1−α) f η , where f = F�RT. At ���.�� K, f = 38.921 V−1 . Taking the ratio
of two expressions for j for di�erent overpotentials gives
j 2 � j 1 = j 0 e(1−α) f η 2 � j 0 e(1−α) f η 1
hence j 2 = j 1 e(1−α) f (η 2 −η 1 )
= (1.0 mA cm−2 ) e(1−0.5)×(38.921 V = �� mA cm−2
−1
)×[(0.60−0.40) V)]
Solutions to problems
P��D.� (a) �e current density is given by [��D.�–���], j = j 0 (e(1−α) f η − e−α f η ),
but for positive η the second term (the anodic current) dominates and
therefore ln j = ln j 0 + (1 − α) f η. A plot of ln j against η will have slope
(1 − α) f and intercept ln j 0 . Such a plot is shown in Fig. ��.��.
0
0.00 0.05 0.10 0.15 0.20 0.25
η�V
Figure 19.15
From the slope it follows that (1 − α) × (38.921 V−1 ) = 24.18 V−1 hence
α = 0.38 . �e exchange current density is computed from the intercept
as j 0 = 0.79 mA cm−2 .
(b) For negative overpotentials the cathodic current dominates and j = − j 0 e−α f η .
�e following table is drawn up using the results from (a).
RT
E(Fe2+ �Fe) = E(Fe2+ �Fe)−○ + ln a Fe2+
2F
�erefore with the given concentration, the potential is
E ′ �V η�V υ�(pmol s−1 ) j�(nA cm−2 ) � j c ��(nA cm−2 ) ln[� j c ��(nA cm−2 )]
−0.702 −0.091 1.47 31.2 32.1 3.47
−0.727 −0.116 2.18 46.2 46.7 3.84
−0.752 −0.141 3.11 65.9 66.2 4.19
−0.812 −0.201 7.26 154 154 5.04
5.0
ln[� j c ��(nA cm−2 )]
4.5
4.0
3.5
−0.22 −0.20 −0.18 −0.16 −0.14 −0.12 −0.10 −0.08
η�V
Figure 19.16
From the slope it follows that −α × (38.921 V−1 ) = −14.20 V−1 hence
α = 0.365 . �e exchange current density is computed from the intercept
as j 0 = 8.91 nA cm−2 .
P��D.� �e data given correspond to positive overpotentials, so the anodic current will
dominate and hence ln j = ln j 0 + (1 − α) f η. A plot of ln j against η will have
slope (1 − α) f and intercept ln j 0 . Such a plot is shown in Fig. ��.��.
8
ln[ j�(mA m−2 )]
j t=0→τ = j 0 e(1−α) f [η− +(η+ −η− )t�τ] = j 0 e(1−α) f η− e(1−α) f (η+ −η− )t�τ
716 19 PROCESSES AT SOLID SURFACES
j t=τ→2τ = j 0 e(1−α) f [(2η+ −η− )+(η− −η+ )t�τ] = j 0 e(1−α) f (2η+ −η− ) e(1−α) f (η− −η+ )t�τ
d Q1 Q2 Q1 Q2
F=− =
dr 4πε 0 r 4πε 0 r 2
(1.6022 × 10−19 C)2
=
4π × (8.8542 × 10−12 J−1 C2 m−1 ) × (2.00 × 10−9 m)2
= 57.7 pN
I��.� �e approach is to compute the standard reaction Gibbs energy of the com-
bustion reaction using tabulated standard Gibbs energies of formation. �e
standard cell potential is then computed using ∆ r G −○ = −νFE −○ ; the number
of electrons involved in the reaction is identi�ed by considering the oxidation
numbers of the products and reactants. For brevity the phases of the species
are omitted from the chemical equations.
�e non-relativistic wavelength is
h
λnon-rel =
(2m e e∆�)1�2
6.6261 × 10−34 J s
=
[2×(9.1094 × 10−31 kg)×(1.6022 × 10−19 C)×(50 × 103 V)]1�2
= 5.48 pm
�ere is a signi�cant di�erence of about �%. Whether or not this will a�ect
any particular measurement depends on whether it is necessary to know the
wavelength to high precision.