Statistical mechanics a survival guide 1st Edition A. M. Glazerinstant download
Statistical mechanics a survival guide 1st Edition A. M. Glazerinstant download
or textbooks at https://ebookultra.com
_____ Follow the link below to get your download now _____
https://ebookultra.com/download/statistical-mechanics-a-
survival-guide-1st-edition-a-m-glazer/
https://ebookultra.com/download/thermodynamics-and-statistical-
mechanics-of-macromolecular-systems-bachmann-m/
https://ebookultra.com/download/statistical-mechanics-a-concise-
introduction-for-chemists-1st-edition-b-widom/
https://ebookultra.com/download/statistical-mechanics-of-phases-and-
phase-transitions-1st-edition-steven-a-kivelson/
https://ebookultra.com/download/statistical-mechanics-of-disordered-
systems-a-mathematical-perspective-1st-edition-anton-bovier/
A Survival Guide for Buying a Home Second Edition Sid
Davis
https://ebookultra.com/download/a-survival-guide-for-buying-a-home-
second-edition-sid-davis/
https://ebookultra.com/download/funding-your-retirement-a-survival-
guide-1st-edition-max-newnham/
https://ebookultra.com/download/accident-and-emergency-radiology-a-
survival-guide-nigel-raby/
https://ebookultra.com/download/elements-of-statistical-mechanics-ivo-
sachs/
https://ebookultra.com/download/nonequilibrium-statistical-mechanics-
gene-f-mazenko/
Statistical mechanics a survival guide 1st Edition A. M.
Glazer Digital Instant Download
Author(s): A. M. Glazer, J. S. Wark
ISBN(s): 9780198508168, 0198508166
Edition: 1
File Details: PDF, 5.43 MB
Year: 2001
Language: english
r:
r
Statistical Mechanics
~
A survival guide
~r~ A. M. GLAZER
Department ofPhysics. University ofOxford
r· and
••....
: J. S. WARK
Department ofPhysics. University of Oxford
..',.
~.
.. :
'
'.
~
~
'~.
......
~
....
.......
~
-...
.....
.
....,
~
OXFORD
UNIVBRSITY PRESS
~
.....
~
9
~
~
OXFORD
UNIVBRSITY PRBSS ~
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It fu rthers the University's objective of excellence in research. scholarship.
and education by publishing worldwide in
1
)
Oxford New York To Pam and Tina
Athens Auckland Bangkok Bogota Buenos Aires Cape Town
Chennai Dar es Salaam Delhi Florence Hong Kong Istanbul Karachi J
KollGlta Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi
Paris Sao Paulo Shanghai Singapore Taipei Tokyo Toronto Warsaw
)
with associated companies in Berlin Ibadan
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc.. New York
(!;) Oxford University Press. 2001
The moral rights of the author ha~e been asserted
Database right Oxford University Press (maker)
First published 2001
All rights reseIVed. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted. in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
. Oxford University Press, at the address above
You mqst not drculate this book 'in any other binding or cover
and!you must impose this same condition on any acquirer
I·,
~\i ~~tish Library Cataloguingin Publication Data
,", Data available
rl~Fary of Congress Cataloging in Publication Data
Data available
. ISBN 0 19 850815 8 (hbk)
ISBN 0 19 850816 6 (pbk)
1 3 5 7 9 10 8 6 4 2
Typeset by the author using laTeX
Printed in Great Britain
on acid-free paper by
T.]. International Ltd. Padstow
~
r·
r~
i
i i
(- Preface
r~
r'"
~~ Statistical mechanics is the science of predicting the observable properties
of a many-body system by studying the statistics of the behaviour of its
~. individual constituents/".- be they atoms, molecules, photons etc. It pro-
....
~. vides the link between macroscopic and microscopic states. For example,
,.
most satisfying parts of an undergraduate science course - linking in an
elegant manner the quantum world with everyday observables of systems
. containing large numbers of particles .
~
~.
.. However, in our experie~ce, all too often students fail to appr(~ci(\tr the'
beauty of this bridge between worlds owing to the complexity of the lIlal h-
ematics with which they are presented. As with any subject, statistical III'>·
•••
~
chanics contains its subtleties, nuances and insights that can only bl' fully
understood after a thorough grounding in the underlying theory. I !owcvl'r,
the majority of the important, basic concepts are, in our opinion, re'adily
accessible without resorting to such in-depth analysis. Therefore. right at
'1t the outset, it is important to state that the approach to the subject taken
within this book is going to be slightly handwaving: we make no att (~TlI pt
~
......... to be sophisticated (just as well, some might say). We firmly ]){'li"vI' thaI
it is better to grasp the basic concepts, even if they are put forward in
a non-rigorous way, than be prcsellted with rigorous derivations bill not
understand them. For the bravI~·hearted, t.here arc many morr 1I';,fntill;ht
~
to physics undergraduates at the University of Oxford over Dlany years.
At present the Oxford undergraduate encounters this subject towards t lIP
~
end of the second year of a 3 - 4 year course, having previously recei"f'd
a course on classical equilibrium thermodynamics and an introduction to
quantum mechanics. The material presented here covers the entire syllabus
as it is taught at this level, and a more detailed approach (not indudrd
~
here) is then given for those working towards a more theoretical course.
'vVe supply a considerable number of questions, many of which have been
traditionally asked at Oxford, together with worked solutions. The latter
~
no doubt will not endear us to our colleagues, since they will now have to
find new questions (without written answers) to set their students!
There are many approaches to studying this subject, and the most
~
~
~
VIII Preface Preface ix
appropriate will clearly depend on your starting point. You will get the omirsky and Svetlana Zhitomirskaya produced the diagrams. Peter Sond-
most out of this book if you have a basic knowledge of quantum mechanics hauss patiently answered our questions concerning I¥IE;X. We extend our
(for example, the ability to solve Schrodinger's equation for simple one- thanks to them all.
dimensional potentials), though this is not essential. Furthermore, there is
clearly a very strong connection between thermodynamics and statistical Clarendon Laboratory, Oxford Mike Glazer
mechanics. We assume some prior study of thermodynamics, and thus at 2001 Justin Wark
least a passing familiarity with the concept of entropy - though perhaps you
may still find it somewhat confusing. It is a concept that often engenders
a feeling of unease. If you have taken a thermodynamics course you will no
doubt recall that there are elegant arguments proving that there is such a
quantity called entropy and that it can be shown to be a function of state.
But it would not be surprising if you remained somewhat confused as to
its nature. The other aspect of thermodynamics that can be troubling,
and which illustrates both its elegance and its apparent obscurity, is its
complete generality. The derivation of all the TdS equations and suchlike
is completely independent of the system in question - we can use them for
ideal gases, magnets, and elastic bands - whatever we like. At no stage
in a classical thermodynamic treatment do we worry about the detailed,
nitty-gritty of the system of interest - its quantum states and the nature
of the particles making up the system.
Now that's where statistical mechanics comes in. It will, we hope, give
you a better feel for the nature of entropy, and provide an understanding of
~
that important link between the microscopic and macroscopic world. By the ~
end of this book you should be able to start from the microscopic world of t,•
SchrOdinger's equation and derive the pressure (or entropy, enthalpy, etc.)
of the gas in a steam engine, the magnetic susceptibility of a paramagnet,
the heat capacity of a solid, and many, many more properties of a system
of particles. It is worth realizing that many of the basic ideas in statistical
.'"..-:
mechanics were worked out in the 19th century by those giants of physics, f'
Ludwig Boltzmann, Willard Gibbs, and James Clerk Maxwell. Boltzmann,
in particular, found himself under serious attack by many of his colleagues ~
for espousing the idea that the properties of gases and other systems could ~
be treated in statistical terms. It is most interesting to read through those
early papers and see the often tortuous and difficult arguments with which ~
he had to wrestle, especially since today we think of many of the ideas in
~
much simpler terms. In the event, Boltzmann was eventually shown to be
correct.. It waS not until the 1920s that Heisenberg and others demonstrated ~
the notion that the properties of quantum particles were indeed statistical
~
in nature, rather than deterministic. Therefore, the work of these 19th
century physicists laid the foundation for the modern science of quantum ~
mechanics.
We are very grateful to Stephen Blundell, Dave Donaldson, aI;id Anton ~
Machacek, who took the time to read the draft manuscript of this book and ~~
made many helpful suggestions. Geoff Brooker made helpful comments on ~~
f the original lecture notes upon which this book were based. Victor Zhit- ~
t
.
·
. ~~
~
~
r:r. '.,.1"
:
. .
". j";~;
,I
/
" I
,...
r.
;
; , i\
Contents
~~
!.
h.
.:: I .:j
:..
:.:.
1 Back to Basics 1
1.1 The Role of Statistical Mechanics, 1
~.' 1.2 Introductior(,to Coin Tossing
1.3 Heads I Win, Tails You Lose
"" 2
2
:.'.•
1.4 Stirling's Theorem 5
1.5 More Flipping Coins '" 6
1.6 Distinguishable Particles " ", i'. 8
1.7 Summary ,\;,
11
:.••
. ~;.' ;.- ,,~,:. ;.
12
1.8 Problems
2 The Statistics of D1stinguishabl~P~ii~i~s , ! ' ;', 14
2.1 The Boltzmann Distribution for DistinguishablePar-'"
ticles ' . .. 14
2.2 Lagrange's Method of Undetermined Multipliers: 15
•• 2.3
2.4
2.5
The Single-Particle Partition Function ' ( ' "
Degeneracy
The Partition Function of a System
21
22
23
~
• 3
2.6
2.7
Summary
Problems
Paramagnets and Oscillators
25
26
29
29
~ 3.1 A Spin-1/2 Paramagnet
3.2 Paramagnets with Angular Momentum J 34
~ 3.3 The Simple Harmonic Oscillator 36
~
Gases 43
'S 4.1 Distinguishable and Indistinguishable States 43
4.2 Identical Gas Particles - Counting the States 45
~ 4.3 The Partition Function of a Monatomic Ideal Gas 51
~
4.4 Properties of the Monatomic Ideal Gas 52
4.5 More about Adiabatic Expansions 53
~ •
xii Contents
I
,
l
~ Contents xiii
..,
~~
~
0'1.8 Summary 59 Revisited 110
-t.!) Problems 61 10.5 Fermi-Dirac Distribution Revisited 110
10.6 Bose-Einstein Distribution Revisited 111 .~
!j Diatomic Ideal Gases 62 10.7 Summary 111
:>.1 Other Degrees of Freedom 62 :t
c .} 10.8 Problems 112 ~
.J._ Rotational Heat Capacities for Diatomic Gases 63
:" "
.) ...
Tile Vibrational Partition Function of a Diatomic 11 The End is in Sight 113 I
65 11.1 Phase Space 113
~
l:a'i
j.J PULLing it All Together for an Ideal Gas 66 11.2 Equipartition of Energy 115
~
,l.:)
. Summary 67 11.3 Route Map through Statistical Mechanics 116 ~
.''d3 Pmulems 68 11.4 Summary 118
11.5 Problems 118 ~
6 Quantum Statistics 69
1>.1 Indistinguishable Particles and Quantum Statistics 69 A Worked Answers to Problems 119 ~
6.2 Bose-Einstein Statistics 71 A.l Chapter 1 119 ~
1),;\ Fermi-Dirac Statistics 73 A.2 Chapter 2 120
ri.'! More on t.he Quantulll Dist.ribut.ion Functions 73 A.3 Chapter 3 122 ~
6.5 Summary 76 A.4 Chapter 4 124 ~
G.G Problems 76 A.5 Chapter 5 125 "
,
, A
A.6 Chapter 6 126
7 Eledl'oll!l in Metals 77 A.7 Chapter 7 126 '"
7.1 Fermi-Dirac Statistics: Electrons in Metals 77
A.8 Chapter 8 128 ~
7.2 The Heat Capacity of a Fermi Gas 79 A.9· Chapter 9 . ~ "
132 1t
~;,
7.3 The Quantum-Classical Transition 82 A.lO Chapter 10 133
7.4 Summary 85 A.ll Chapter 11 134 tJ
7.5 Problems 86
8 Photons and Phonons
8.1 The Photon Gas
88
88
B
C
U sefnl Integrals
Physical Constants
137
138
"
\
8.2 Generalized Derivation of the Density of States 89
D Bibliography 139 ~
8.3 Blackbody Radiation 90
8.4 Phonons 92 Index 140 C
8.5 Summary /
93 ~
8.6 Problems 94
~
9 Bose-Einstein Condensation 97 ~
9.1 Introduction 97 }
~
~
9.2 The Phenomenon of Bose-Einstein Condensation 97 ,.
~
~
9.3 The Quantum-Classical Transition Revisited 102 ~ ,:: ...
9.4 Summary 102 ~
9.5 Problems 103 l. ::"l
I '-=
10 Ensembles
.1 i
104 t
~
I
10.1 Introduction
I
104 I ..
10.2 The Chemical Potential /
!
i
106 if ~
,
j
:~
C
A
l
1
Back to Basics
Table 1.1 The 5 macrostates for flipping a coin 4 times with the associated
far·more tightly 'peaked' than in the example involving 4 coins above (as
microstates
an exercise, work out the numbers for tossing 6 coins). It is because the
Macrostate Microstates distribution becomes so tightly peaked that we make the second postulate
above; that the.macrostate observed is the one with the most microstates.
HHHH If we had many, many coins, the probability becomes so narrowly peaked
4 Heads
around half the number of heads and half the number of tails that to say
3 Heads and 1 Tail HHHT this is the observed macrostate does not introduce a large error.
HHTH Also, it should be clear by now that within statistical mechanics we
HTHH will often be interested in the factorial of a number. We can see this by
THHH thinking a little more generally about the coins. Let's flip a coin N times,
and denote by the symbol n the number of ways of getting n heads (and
2 Heads and 2 Tails HHTT therefore (N - n) tails). We find
HTHT
THHT N'
n= (1.1)
" HTTH n.I(N°- n.
)1
" THTH
It is important to understand this equation. If we have N different objects,
TTHH
we can arrange them in N! different ways. However, when we flip a coin
1 Head and 3 Tails N times, and get n heads, then each result of heads looks the same: if you
TTTH
think of the result HTH H, it looks the same if we interchange any of the
TTHT
THTT heads with any other head. Therefore, the number of ways of arranging the
N flips is NI divided by n! (to take into account that all the heads look
HTTT
the same), and obviously divided by (N - n)! to take into account that all
4 Tails TTTT the tails are the same also.
Notice how this fits in with flipping the coins 4 times. In this case, to
=
get 2 heads (and therefore 2 tails) n 4!/(2! x 2!) 6 as expected. When =
we get to real physical problems we will thus be dealing with numbers like
macrostate of 4 heads. In· total, there are 5 possible macrostates, and the 10 23 !, and before proceeding further we are going to make a slight. diversion
microstates associated with each of these macrostates are shown in Table to the topic of Stirling's theorem, which givcs us a handy way to dpal wil h
1.1. n! when n is extremely large. 'rVe will thcn return to the coins.
In this problem, we have the basic elements of statistical mechanics.
We have learnt that a microstate is the detailed arrangement and each mi- 1.4 Stirling's Theorem
crostate is equally probable. However, the macrostate, which we will define Stirling's theorem is an extremely useful formula for nl that we will us(~
to be the situation of macroscopic interest, can contain several microstates, again and again in our subsequent analysis of microstates. TIll' simpl..st
and obviously the most probable macrostate is the one that contains the form of Stirling's theorem states that for large n
most microstates.
Now, in this simple example, it is clear that the most likely outcome Inn! ~ nlnn - 1/
is 2 heads and 2 tails - but not by much. We could have got 3 heads and You may wish to test the accuracy of this formula using a calculator (sPc
1 tail with a 4 in 16 chance, or even all heads (or tails) with a 1 in 16 Problem 5 in Section 1.8). This remarkable forl1lula call bc jusrifi"d by a
chance. However, it is clear that when we come to deal with real physical graphical method. Study Fig. 1.1. The dot.ted line is t.he function !I = III J',
systems, the numbers are going to get much larger than 4. We know that where x is a continuous variable. The area under the curve fWIIl J' = I to
one gram mole of gas at standard temperature and pressure (STP) contains x = n is the integral, J:
6 x 10 23 (Avogadro's number) molecules. When we come to deal with such
systems we will find that the probability for the most likely macrostate is
I = In In xcix = [x In x - x )'1' = 71 In Il - 71 +I
6 Back to Basics
•
More Flipping Coins 7
..f
(i
~ 1.2 C
y=ln (x) <I)
;.
0.
2.5 ....0 1.0
C
"0
<I)
,~ 0.8
:,
: I
I'll'
I'·
I'. C
(;j
2.0 :' , I ",
E
0
c::
0.6 .
..
. I
, I
\
\
.
.
c
:>.. 1.5
---
en
<I) 0.4
,
,
, I
\ e
l.0
r---.
I
E
en
2 0.2
:
I
,
, I
I
I '.
e
u
Au
1---. :.' '8 ./ I ~
0.0
0.5 1
1
1
I'
•
AL
'-
...0
<I)
e
I' ..0
E
::l 0 NI2 N
e
o 2 4 8 10 Z c
x Number of heads (nonnalized to N)
Fig, 1.1 Graphical proof of Stirling's theorem - see the text for an explanation. Fig. 1.2 The width of the distribution of the number of microstates per
e
macrostate narrows as the number of coins increases: (i) dotted - about 20 coins, e
\V(~ call obviously ignore t.he I for very large n. Also shown on the figure
an' I WI) stai rcases, one abovc (the dashed line) and one below (the solid
(ii) dashed - about 75 coins, (iii) solid - about 150 coins.
e
lille) till' smooth curve. The area under the upper one, Au, between x = 1 n = N /2 - that is to say, half heads and half tails. For the case of 4 coins we e
alld r == n is t.he serics: showed this explicitly by working out the number of ways it could occur,
and we satisfied ourselves that this was the most likely outcome. But how ~
.1
,~.
Au = 1112+ 111:1+ In4+ +Inn
(a.-; III I = a we don't need to include it.). On the other hand between the
= Inn! do we prove this for an arbitrary number of coins, especially if the number
of flips is very large? Well, if n is very large, we can treat it as a continuous,
rather than discrete, variable, and we can work out where the maximum
••
same limits for the lower staircase:
In n = N In N - N - n In n +n - (N - n)ln(N - n) + (N - n)
"
~
The importance of this result is that we shall find later that the quan-
~
For a given total number of flips (i.e. keeping N constant) it is trivial to
tity In n plays a very important role in statistical mechanics. This is most ('0
show that
fortunate because using logarithms changes huge numbers into relatively
small, more tractable numbers. Thank goodness for logarithms! d(lnO) =_lnn+ln(N_n)=ln(N-n)
C'
dn n ~
1.5 More Flipping Coins =
which is zero when (N - n) n, and thus n N/2 as expected. =
Armed with Stirling's theorem let us go back to flipping coins. We know We should note that although we have now proved that the most prob-
intllitively, as we have said before, that t.he most probable result is evident.ly able distribution is half heads and half tails, we have still not worked out
o Back to Basics Distinguishable Jl'lfticll'S 9
how sharp the peak is. This point is dealt with in part in Problem 1 in particles. Strictly speaking, for this to happen there must be som.: illterae-
Section 2.7. Here, we simply quote the result: for coins the width of the tion between the particles so that energy can be transferred between them
peak scales as N 1 / 2 . But the total number of particles scales as N, so the - we will assume that the interaction between the particles is just strong
fmctional width scales as N- 1/ 2 • This is best illustrated by Fig. 1.2, where enough for this transfer to take place, but not sufficiently strong to prevent
the distribution (normalized to the peak) is shown for various numbers of us from treating each particle independently. In order to accommodate this
coins. Similarly, we shall assume in the physics that follows that for sys- energy, some of the particles, let's say nl of them, will be in the first cller~y
tems containing large numbers of particles the fluctuations from the most level of their particular potential wells, each with energy [] , SOIlW of t heill.
probable result are limitingly small, and we are justified in taking the most n2, will be in the second energy level [2, and so all. In gCllccal I h.. n' will
probable macrostate (the one with the most microstates) as representing be ni particles each with energy Ei. Clearly, as we have a fixed 1I1111lbcr of
the so-called equilibrium state of the system. particles:
I herp ar.' with a particular energy is analogous to knowing how many heads
and tails we are likely to get when we flip the coin a set number of times. 41:
With t he coins we could only get heads or tails - now there are lots more ~ This is one of 7
31:
~ microstates in the
possibilities - but the basic idea. remains the same. For the particles, a par- 2E Ul same macrostate
t icular Jllicrostate would entail knowing exactly which particular particles ]E
had which particular energies, rather than just knowing the overall num- o
Iwrs - just as for the coins a microstate entailed knowing exactly which
flips gall' the heads, and which ones the tails. For the coins we know that a)
IIII' Illilnost.at.e of half heads a\l(l half tails is the most, likely, as it contains This is one of 105
IIII' 111',,,1 lIlierost.a\.('$. In the same way, we need to find out the number microstates in the
~)f lIlir·ro"t.atcs in a macrostate for the particles. Once we have a formula ~ same macrostate! -
l"lling us thp number of microst.ates in a macrost.at.c. we can find the one g this macrostate is
that has til(' maximullI probability just ;1.S we did for t.he coins. Ul obviously more
likely
For l he particles, the total number of microstates in a given macrostate,
Q, is given by
b)
(1.2) Fig. 1.4 Two particular arrangements (microstates) of 7 simple harmonic os-
cillators with a total energy of 4e (we have neglected the zero point energy). ]n
(a) all of the energy is taken up by one particular particle, whereas in (b) two
Notice how similar this is to the formula for the coins (eqn (1.1)). Instead of
having just heads or tails (which for the coins meant that we divided the N!
by 7I! and (N - n)!), we now have all the different possible energies f 1, f2, f3
particular particles each have energies of Ie while another particular particle has
I
an energy of 2e. The arrangement shown in (a) is one of 7 microstates in the same
macrostate, whereas the arrangement shown in (b) is one of 105 microstates in
,
~
'(
\
etc., containing nl, n2, n3 particles, so we divide the N! by nl!, n2!, n3! etc.
Defore we wade further into the algebra, let us look at a simple example
that sheds light on equation (1.2). Let us suppose that we have 7 localized
the same macrostate.
just as getting half heads and half tails is the most probable macrostate
•
\.1
,~.
particles, with energy levels that are equally spaced with energies 0, f, 2f, 3f
etc., such as occurs in simple harmonic oscillators, ancl we give our system
of particles a total energy 4f. Fig. 1.4 illustrates two possible microstates
for flipping coins. Obviously, just as with the coins, we would like to find a
general way of finding the macrostate which corresponds to the maximum
value Q for the general case where N is very large. How this is done will
•-
~
(there are many, many more in this system). For instance, (i) is a microstate be revealed in the next chapter.
where there are 6 particles in one level and 1 particular one in another. How f
many different ways in total can this be done? Answer = 7 (we could put 1.7 Summary (
any of the particles in the upper state). That is, there are 7 microstates To summarize the important points of this chapter:
in the corresponding macrostate. As for the microstate shown in (ii), this C
• A microstate describes which particles are in which state. All allowed
particular microstate is one of a total of 105 in the same macrostate! The
macrostate corresponds to having 4 particles in the ground state, 2 in the
microstates are assumed equally probable. ~
• A macrostate describes how many particles are in each energy level.
level with energy €, and 1 in the level with energy 2f, and of course the
We assume that the observed macrostate is the one with the most
C
microstate shown tells you elCactly which of the seven particles has which
microstates, ~
energy. There are 105 microstates in this macrostate, as Q =
7!/{4! x
• Stirling's formula (in the form we shall use) is
2! x I!) = 105. One thing you might notice immediately is that the more <.'
probable distribution (the one with 105 microstates) looks as if it could be Inn! = nlnn - n C'
falling roughly exponentially with energy (4 particles in the ground state,
~
I
2 in the next, and 1 in the following state). This exponentiaL fall-off -
the so-called Boltzmann distribution law - is one of the main results of • If we have N distinguishable non-degenerate particles, such that there
statistical mechanics. This law will be justified in Chapter 2. It so happens are ni particles in the energy level fi, the number of arrangements (i.e. ~
that this macrostate is the most prohil.hl~ one for t.his particlIlar problem, the number of microst.ates per macrostate) is ~
Problems 13
12 Hack to Hasics
N! N! when tossing 6 coins, and the number of ways of getting 2 heads and
n = nl.n2·
I In I n .1
3· .. · ' .... TI n t ! 4 tails, and 1 head and 5 tails, and so forth. Do the same for 10 coins.
i PI~t a graph of your results, with the probability normalized to the
peak of the distribution in each case, and with the number of particles
• For an isolated system of distinguishable particles, the number of par- normalized to the total number. Convince yourself that, in fractional
ticles and the energy are constant: terms, the distribution is getting slightly more sharply peaked as the
number of coins increases. . /
Lnj=N U = Lnjei
1.8 Problems
1. To win the UK National Lottery one must supposedly pick 6 different
numbers from the 49 available. The order in which the numbers are
chosen does not matter. If, being mean, we buy only one ticket, what
are our chances of winning the jackpot?
2. (i) Mr. and Mrs. Fatchance have two children. If they tell you that at
least one of them is a boy, but you do not know the gender of the other
child, what is the probability that they have two boys?
(ii) If they had three children and told you again that at least one was
a boy, what would the probability be that they were all boys?
(iii) Finally, suppose that they had never worked out what was causing
all this procreation, and had N children, and told you again that at
least one was a boy, what would the probability be that they were all
boys?
3. In a particularly hideous game show, a contestant is asked to guess
behind which of three closed doors a prize is hidden. He/she chooses
one of the three doors, and tells the compere his/her choice. That door
remains closed, but the compere (who knows where the prize actually
lies) then opens one of the two remaining doors, showing it to be empty
to the contestant. With the door of his/her original guess still closed,
the contestant is then given the chance to guess again. Is there any
advantage at this point to him/her in changing doors?
4. If we assume that the birthdays of the population are equally dis-
tributed throughout the year, how maIlY people, taken at random,
need to gather together in a room to give a greater than 50% chance
of at least two of them sharing a cammon birthday?
5. Use your calculator to work out In 10! Compare your answer with the
simple version of Stirling's theorem (N In N - N) . How big must N
be for the simple version of Stirling's theorem to be correct to within
(i) 5% (ii) 1% ?
6. In Section 1.3 we worked out the number of ways we could get heads
or tails when we flipped 4 coins (there were 6 ways of getting 2 heads
and 2 tails, 4 ways of getting 1 head and 3 tails etc. etc.). Using eqn
(1.1), work out the number of ways of getting 3 heads and 3 tails
q
Lagrange's Method of Undetermined MUltipliers 15
The Statistics of =a
d(ln 12)
In 12 a
In 12 0 In 12
-!::l-dnl + -!::l-dn2 + ... -!::l-dni + ... 0
Unl Un2 Uni
=
Distinguishable Particles But from eqn (2.3)
olnn
- - = -(lnnj + 1)
oni
and so
He uses statistics as a drunken man uses lamp-posts - for support rather than
illumination.
d(ln 12) =- L(ln n; + l)dni = 0 (2.4)
A. Lang
But recall that the total number of particles, and the total internal energy
2.1 The Boltzmann Distribution for Distinguishable Parti- are also constant, and so it follows that
cles
.'.'·'1 In Section 1.G we showed that for a system of N particles the macrostate
Ldn; =0 and LCjdnj =0 (2.5)
which has, in general, ni particles in the non-degenerate energy level Ej
I'
contains 12 microstates: Also, because of eqn (2.5), eqn (2.4) reduces to
But we also know that the total number of particles is constant, and, as we 2.2 Lagrange's Method of Undetermined MUltipliers
~#""'defitwd the system as being isolated from the environment, so too is the It can be seen from the working above that in order to maximize In 12 (and
",~ Clj('rgy. That is to say hence 12) we have three conditions to satisfy simultaneously:
\\,}W[I' (r is t.he total internal energy of the syst.em. We now wish to maximise Physically, this corresponds to finding the macrostate with the maximum
t
\2 10 lind out. the most. probable macrostate. Our approach is going to be number of microstates subject to the constraints of a given number of C
V('ry "illlilar to that for the coins. First. of all, we decide to maximise In 12 particles and a given energy. As we have three quantities that must all be
ratllt'r than 12 it.self, for t.he simple reason that the mathematics is made a
~
zero at the same time, we could certainly multiply two of them by arbitrary
101 ":\.~icr with the help of Stirling's theorem. So, taking logarithms of eqn
(:!.l).
constants, say a and {3: r-
Lin n;dn; = 0, aLdnj=O, and (3LE;dnj =0 . ~
Inn = InN! - LInn;!
~
The two constants, a and {3, are known as Lagrange multipliers. We can
ThcII IIsing Stirling's theorem we obtain now write .' ~
In n = N In N - N - L nj In nj + L nj = N In N - 2: ni In nj . (2.3) L dni(ln ni + a + {3Ei) = 0 (2.7) ~
j j ; 1 ~.....
III ord,'r to find the equilibrium state (the observed macrostate) we must How do we satisfy eqn (2.7)? As we have two unknowns, a and (3, we might ~
maxilJlize t.his quantity. For the coins maximizing 12 was fairly trivial - we argue that we could satisfy two of the terms simultaneously, i.e. we could ~. ,
eliminate the terms in dnl and dn2 by satisfying the two simultaneous
equations
I Total Energy = U
Lagrange's Method of Undetermined Multipliers
and
but even if we did this it would still leave all the other terms in the series
totally independent. Furthermore, there is no particular reason for picking
the first and second terms in the series - we could pick any two we liked, and
still leave all the rest independent. The result of this argument is that the N 2 particles
only way to satisfy eqn (2.7) is to treat all of the terms as being mutually m, in level f:'j
independent, and therefore each term in the series must be exactly zero,
that is to say for all values of i
(In nj + (X + j3e;) = 0
from which it follows that
nj =exp(-a-j3e;) =·Aexp(-j3e;) (2.8)
where for convenience we have written A = exp(-a). Fig. 2.1 Two separate systems of distinguishable non-degenerate particles are
This is starting to look a little like the expected Boltzmann distribution in thermal contact.' Their combined total energy, U = U1 + U2 , is constant.
- the number of particles in a given energy level falls off exponentially with
increasing energy. However, we are still left with the problem of determining We now crank the handle, and put this through the sausage machine
the constants A and 13. Note that if we knew what 13 was, then finding A once more (take logarithms, simplify using Stirling's theorem, and set the
would be trivial. A is just a constant of normalization, which can be seen differential of In n equal to 0 to find the maximum):
by substituting eqn (2.8) into eqn (2.2):
.In n = N 1 In N 1 - 2: nj In nj + N 2 ln N 2 - 2: mj In mj
:,.' N.=.A-2]exp(-j3c;)· '. . '. ,'. j
But how do we show what 13 is? Well, first,we will demonstrate that it d(ln n) = -d(2: nj In nj) - d(2: m;ln ~d ;, 0
must be some function of temperature.. j
..
Exploring the Variety of Random
Documents with Different Content
books about election and the elect, which they applied serenely to
other people. They were taught that all the people—outside, in the
street—in the world were destined to endless torments: all but a very
few, including themselves. They believed it, or said they did; and the
words never caused them a shudder, a gleam of pity, a thought of
remonstrance. That is what they called believing the doctrine.
Photo by H. N. King
LAST PHOTOGRAPH OF THE PRINCE CONSORT
Photo by H. N. King
STATUE OF QUEEN AND PRINCE CONSORT AT WINDSOR
Photo by Valentine
THE ALBERT MEMORIAL, LONDON
We have learned that the old revolutionary cry has quite died away
and is almost forgotten. This also is partly the result of the increased
comfort. At the same time the advance of democratic ideas has been
most marked. Slowly but surely, the whole power in the country has
passed into the hands of the Commons. The dominant idea at the
present moment of our people is that the country must be governed for
them and by them. This would have seemed a most terrible thing sixty
years ago. That we should be governed by working-men! Incredible! It
is, however, the fact; we are governed by the people. Only, what the
prophets did not understand, the governing power is delegated by the
people to representatives, who are not, as a rule, working-men; one or
two working-men are in the House and doing well. The people,
however, are very chary of electing one of themselves; they prefer to
send to the House as their representatives such men as John Morley
and James Bryce, scholars and students, responsible persons, whom
they know and can trust; they will not send demagogues and wind-
bags and political adventurers. You have seen how they treat the
House of Lords; so long as it gives no trouble it may remain, but only
on condition that it is recruited from new families. If it were to obstruct
any really popular movement—which the House will not do—we should
see what would happen. Meantime, the people look abroad and judge
for themselves. They observe that the great colonies are all Republics,
and are doing well under republican institutions. If we were not doing
well under our institutions, it is quite certain that the revolutionary cry
would be heard again.
As regards work and wages, the people are firmly persuaded that
they are entitled to be the dictators. They think that they have a right
to exact what wages they think are fair, and to work for such hours as
they think right. There have been desperate struggles, in which the
employers have lost huge sums of money, while the men have suffered
terrible privations. It is not for me to discuss in this place the right or
the wrong of Trades Unions; it is enough to state that the working-men
hold this belief, and are ready, whenever it is possible, to act upon it.
Copyright Foster & Dickinson
THE CONSERVATIVE BENCHES OF THE HOUSE OF LORDS ON THE NIGHT OF THE
DIVISION ON THE HOME RULE BILL,
8th September 1893
The Marquis of Salisbury speaking
These changes for the lads and working-men I foresee very clearly.
With regard to the position of women I also foresee important changes.
At the present moment there is a wild and insensate game of “grab”
going on. Women admit of no restrictions, they claim everything. They
are not satisfied with the whole intellectual field, they would overrun
the field of physical labour. They take the men’s work at half the pay,
they drive the men out of the country, they remove from themselves
the possibility of marriage, they deny the country that increase of
population which the country has a right to expect. This folly will
presently cease; calmer and more sensible counsels will prevail. It will
be recognised that Nature assigns limitations and prescribes certain
kinds of work for men, and certain other kinds for women. Above all, it
will be remembered that if a man owes himself to his country as a
soldier or a workman, so a woman owes to her country the duties of
maternity.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
ebookultra.com