Vous êtes sur la page 1sur 32

Mathematics for Chemistry

Mathematics catered towards undergraduate chemists with little


maths experience
Contents

0.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
0.1.1 Quantitative methods in chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 Numbers 2
1.1 Number theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Partial fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.3 Polynomial division . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.4 Substitutions and expansions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.1 Functions as tools in chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Units and dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3.1 Units, multipliers and prexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3.2 Conversion factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.3 Greek alphabet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.4 Unit labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.5 An aside on scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.6 Making tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.1 Denition of errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.2 Combination of uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 Plotting Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5.1 The properties of graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.6 Complex numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.6.1 Introduction to complex numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2 Trigonometry, matrices 8
2.1 Trigonometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.1 Free Web Based Material from UK HEFCE . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.2 Trigonometry - the sin and cosine rules . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.3 Trigonometric identities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.4 Identities and equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.5 Some observations on triangles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.6 The interior angles of a polygon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

i
ii CONTENTS

2.2 Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1 Free Web Based Material from HEFCE . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.2 Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.3 A summary of vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Matrices and determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.1 Simultaneous linear equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.2 Practice simultaneous equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3.3 Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.4 Matrix multiply practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.5 Finding the inverse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.6 Determinants and the Eigenvalue problem . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.7 Matrices with complex numbers in them . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

3 Calculus 14
3.1 Dierentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.1 Free web-based material from HEFCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.2 The basic polynomial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.3 The chain rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.4 Dierentiating a product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.5 Dierentiating a quotient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.6 Using dierentiation to check turning points . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2.1 Free Web Based Material from HEFCE . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2.2 Logarithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.2.3 Integrating 1/x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.2.4 Integrating 1/x like things . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.2.5 Some observations on innity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.2.6 Denite integrals (limits) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2.7 Integration by substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2.8 Simple integration of trigonometric and exponential Functions . . . . . . . . . . . . . . . . 17
3.2.9 Integration by parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2.10 Integration Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.2.11 Dierential equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.12 The calculus of trigonometric functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.13 Integration by rearrangement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.14 The Maclaurin series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2.15 Calculus revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3 Some useful aspects of calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.1 Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.2 Numerical dierentiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3.3 Numerical integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
CONTENTS iii

4 Tests and examples 21


4.1 Some mathematical examples applied to chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1.1 Variable names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1.2 van der Waals Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.1.3 A diatomic potential energy surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.1.4 A one-dimensional metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.1.5 Keplers Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.1.6 Newtons law of cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.1.7 Bacterial Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1.8 Partial fractions for the 2nd order rate equation . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2 Tests and exams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2.1 A possible nal test with explanatory notes . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2.2 50 Minute Test III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

5 Further reading 26
5.1 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.1.1 Further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

6 Text and image sources, contributors, and licenses 27


6.1 Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
6.2 Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6.3 Content license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
0.1. INTRODUCTION 1

0.1 Introduction
This book was initially derived from a set of notes used
in a university chemistry course. It is hoped it will evolve
into something useful and develop a set of open access
problems as well as pedagogical material.
For many universities the days when admission to a
Chemistry, Chemical Engineering, Materials Science or
even Physics course could require the equivalent of A-
levels in Chemistry, Physics and Mathematics are prob-
ably over for ever. The broadening out of school curric-
ula has had several eects, including student entry with
a more diverse educational background and has also re-
sulted in the subject areas Chemistry, Physics and Mathe-
matics becoming disjoint so that there is no co-requisite
material between them. This means that, for instance,
physics cannot have any advanced, or even any very sig-
nicant mathematics in it. This is to allow the subject to
be studied without any of the maths which might be rst
studied by the A-level maths group at the ages of 17 and
18. Thus physics at school has become considerably more
descriptive and visual than it was 20 years ago. The same
applies to a lesser extent to chemistry.
This means there must be an essentially remedial com-
ponent of university chemistry to teach just the Mathe-
matics and Physics which is needed and not too much, if
any more, as it is time consuming and perhaps not what
the student of Chemistry is most focused on. There is
therefore also a need for a book Physics for Chemistry.

0.1.1 Quantitative methods in chemistry


There are several reasons why numerical (quantitative)
methods are useful in chemistry:

Chemists need numerical information concerning


reactions, such as how much of a substance is con-
sumed, how long does this take, how likely is the
reaction to take place.
Chemists work with a variety of dierent units, with
wildly dierent ranges, which one must be able to
use and convert with ease.

Measurements taken during experiments are not


perfect, so evaluation and combination of errors is
required.
Predictions are often desired, and relationships rep-
resented as equations are manipulated and evaluated
in order to obtain information.
Chapter 1

Numbers

1.1 Number theory metrical angles, e.g. sin 60 = 23 and so they appear fre-
quently in mathematical expressions regarding 3 dimen-
1.1.1 Numbers sional space.

Notation
Real numbers come in several varieties and forms;
The notation used for recording numbers in chemistry is
the same as for other scientic disciplines, and appropri-
Integers are whole numbers used for counting in- ately called scientic notation, or standard form. It is a
divisible objects, together with negative equivalents way of writing both very large and very small numbers
and zero, e.g. 42, 7, 0 in a shortened form compared to decimal notation. An
example of a number written in scientic notation is
Rational numbers can always be expressed as frac-
tions, e.g. 4.673 = 4673/1000. 4.65 106

Irrational numbers, unlike rational numbers, cannot with 4.65 being a coecient termed the signicand or the
be expressed mantissa, and 6 being an integer exponent. When written
as a fraction or as a denite decimal, in decimal notation, the number becomes
e.g. and 2
4650000 .
Natural numbers are integers that are greater than or
equal to zero. Numbers written in scientic notation are usually nor-
malised, such that only one digit precedes the decimal
point. This is to make order of magnitude comparisons
It is also worth noting that the imaginary unit and there-
easier, by simply comparing the exponents of two num-
fore complex numbers are used in chemistry, especially
bers written in scientic notation, but also to minimise
when dealing with equations concerning waves.
transcription errors, as the decimal point has an assumed
position after the rst digit. In computing and on calcu-
Surds lators, it is common for the 10 (times ten to the power
of) to be replaced with E (capital e). It is important
The origin of surds goes back to the Greek philosophers. not to confuse this E with the mathematical constant e.
It is relatively simple to prove that the square root of 2 Engineering notation is a special restriction of scientic
cannot be a ratio of two integers, no matter how large notation where the exponent must be divisible by three.
the integers may become. In a rather Pythonesque inci- Therefore, engineering notation is not normalised, but
dent the inventor of this proof was put to death for heresy can easily use SI prexes for magnitude.
by the other philosophers because they could not believe
such a pure number as the root of 2 could have this impure Remember that in SI, numbers do not have commas
property. between the thousands, instead there are spaces, e.g.
18 617 132 , (an integer) or 1.861 713 2x107 . Com-
(The original use of quadratic equations is very old, Baby- mas are used as decimal points in many countries.
lon many centuries BC.) This was to allocate land to farm-
ers in the same quantity as traditionally held after the
great oods on the Tigris and Euphrates had reshaped the Exponents
elds. The mathematical technology became used for the
same purpose in the Nile delta. Consider a number xn , where x is the base and n is the
When you do trigonometry later you will see that surds exponent. This is generally read as x to the n or x to
are in the trigonometric functions of the important sym- the power of n . If n = 2 then it is common to say x

2
1.1. NUMBER THEORY 3

squared, and if n = 3 then x cubed. Comparing pow- We can reverse this process by use of a common denom-
ers (exponentiation) to multiplication for positive integer inator.
values of n (n = 1, 2, 3 . . . ) , it can be demonstrated that 3
1 = 3(x1)(x+1)
2(x+1) 2(x1) 2(x+1)(x1)
4x = x + x + x + x , i.e. four lots of x added together
The numerator is 2x 4 , so it becomes
x4 = xxxx , i.e. x multiplied by itself four times. (x2)
(x+1)(x1)
1 0
For x , the result is simply x . For x the result is 1 .
which is what we started from.
So we can generate a single term by multiplying by the
Order of operations denominators to create a common denominator and then
add up the numerator to simplify. A typical application
When an expression contains dierent operations, they might be to convert a term to partial fractions, do some
must be evaluated in a certain order. Exponents are evalu- calculus on the terms, and then regather into one quotient
ated rst. Then, multiplication and division are evaluated for display purposes. In a factorised single quotient it will
from left to right. Last, addition and subtraction are eval- be easier to see where numerators go to zero, giving so-
uated left to right. Parentheses or brackets take prece- lutions to f (x) = 0 , and where denominators go to zero
dence over all operations. Anything within parentheses giving innities.
must be calculated rst. A common acronym used to re- A typical example of a meaningful innity in chemistry
member the order of operations is PEMDAS, for Paren- might be an expression such as
theses, Exponents, Multiplication, Division, Addition,
A
Subtraction. Another way to remember this acronym is (EEa )2
Please Excuse My Dear Aunt Sally. The variable is the energy E, so this function is small ev-
Keep in mind that negation is usually considered multi- erywhere, except near Ea . Near Ea a resonance occurs
plication. So in the case of x2 , the exponent would be and the expression becomes innite when the two ener-
evaluated rst, then negated, resulting in a negative num- gies are precisely the same. A molecule which can be
ber. electronically excited by light has several of these reso-
nances.
Take note of this example:
Here is another example. If we had to integrate the fol-
3 + x 2 , let x=5.
lowing expression we would rst convert to partial frac-
If evaluated incorrectly (left-to-right, with no order of op- tions:
erations), the result would be 16. Three plus ve gives 3x 3x A B
= 2(x+1)(x2) = x+1 + x2
eight, times two is 16. The correct answer should be 13. 2x2 2x4
Five times two gives ten, plus three gives 13. This is be- so
cause multiplication is solved before addition. 3
x = A(x 2) + B(x 1)
2
let x = 2 then 3 = B
1.1.2 Partial fractions let x = 1 then 3
2 = A
therefore the expression becomes
Partial fractions are used in a few derivations in thermo-
dynamics and they are good for practicing algebra and x2 2(x+1)
3 3

factorisation. Later you will learn that these expressions integrate to


It is possible to express quotients in more than one way. give simple ln expressions.
Of practical use is that they can be collected into one term
or generated as several terms by the method of partial
fractions. Integration of a complex single term quotient Problems
is often dicult, whereas by splitting it up into a sum, a
3 4x2
sum of standard integrals is obtained. This is the principal (1) x2 1 (2) x2 +2x
chemical application of partial fractions. 4 7x+6
(3) (2x+1)(x3) (4) x2 +3x+2
An example is
x+1
A(x1)+B(x+1)
(5) 2x2 6x+4
x2 A B
(x+1)(x1) = (x+1)(x1) = (x+1) + (x1)

In the above x 2 must equal A(x 1) + B(x + 1) since


the denominators are equal. So we set x rst to +1 giving Answers
1 = 2B . Therefore B = 1/2. If we set x = 1
instead 3 = 2A , therefore A = 3/2 . So (1) 3
2(x1) 3
2(x+1) (2) 5
(x+2) 1
x
x2
(x+1)(x1) = 3
2(x+1) 1
2(x1) (3) 4
7(x3) 8
7(2x+1) (4) 8
x+2 1
x+1
4 CHAPTER 1. NUMBERS

(5) 3
2(x2) 2
2(x1) ax2 + bx + c = 0
there is a formula

b (b2 4ac)
1.1.3 Polynomial division x= 2a
(Notice the line over the square root has the same priority
This is related to partial fractions in that its principal use
as a bracket. Of
course
we all know by now that a+b
is to facilitate integration.
is not equal to a + b but errors of priority are among
Divide out the most common algebra errors in practice).
3x2 4x6
1+x There is a formula for a cubic equation but it is
like this rather complicated and unlikely to be required for
undergraduate-level study of chemistry. Cubic and higher
3x - 7 ----------------- x + 1 ) 3x2 4x 6 3x2 +3x ------ equations occur often in chemistry, but if they do not fac-
--- 0 7x 6 7x 7 -------- 1 torise they are usually solved by computer.
So our equation becomes 3x 7 + 1+x 1
Solve:
This can be easily dierentiated, and integrated. If this 2x2 14x + 9
is dierentiated with the quotient formula it is consider- 2
ably harder to reduce to the the same form. The same 1.56(x + 3.67x + 0.014)
procedure can be applied to partial fractions. Notice the scope or range of the bracket.
2x2 4x + 2
1.1.4 Substitutions and expansions 45.1(1.2[A]2 57.9[A] + 4.193)
Notice here that the variable is a concentration, not the
You can see the value of changing the variable by simpli-
ubiquitous x .
fying
3 2
( J(J+1) ) +2( J(J+1) )+1 ( J(J+1) ) 2( J(J+1) )+1


J(J+1) 2
( ) 9 J(J+1) 2
( ) +9 1.3 Units and dimensions
to
x3 +2x+1 x2 2x+1
x2 9 x2 +9
1.3.1 Units, multipliers and prexes
where
It is usually necessary in chemistry to be familiar with
J(J+1)
x= at least three systems of units, Le Systme International
This is an example of simplication. It would actually be d'Units (SI), atomic units as used in theoretical calcula-
possible to dierentiate this with respect to either J or tions and the unit system used by the experimentalists.
using only the techniques you have been shown. The alge- Thus if we are dealing with the ionization energy, the
braic manipulation involves dierentiation of a quotient units involved will be the Joule (J), the Hartree (E , the
and the chain rule. atomic unit of energy), and the electron volt (eV).

Evaluating dy
gives These units all have their own advantages;
dx

+ 2 (x(x2x+1)x
3 2
3x2 +2
2 (x(x+2x+1)x 2x2 The SI unit should be understood by all scientists
x2 9 2 9)2 x2 +9 2 +9)2
regardless of their eld.
Expanding this out to the J s and s would look ridicu-
lous. The atomic unit system is the natural unit for theory
as most of the fundamental constants are unity and
Substitutions like this are continually made for the pur-
equations can be cast in dimensionless forms.
pose of having new, simpler expressions, to which the
rules of calculus or identities are applied. The electron volt comes from the operation of the
ionization apparatus where individual electrons are
accelerated between plates which have a potential
1.2 Functions dierence in Volts.

An advantage of the SI system is that the dimensionality


1.2.1 Functions as tools in chemistry of each term is made clear as the fundamental constants
have structure. This is a complicated way of saying that
The quadratic formula if you know the dimensionality of all the things you are
working with you know an awful lot about the mathemat-
In order to nd the solutions to the general form of a ics and properties such as scaling with size of your sys-
quadratic equation, tem. Also, the same system of units can describe both
1.3. UNITS AND DIMENSIONS 5

the output of a large power station (gigaJoules), or the in- Magnetic Susceptibility
teraction of two inert gas atoms, (a few kJ per mole or
a very small number of Joules per molecule when it has 1 cm3 mol1 = 16.60540984 JT2 x 1030 (joule tesla2 )
been divided by Avogadros number).
In SI the symbols for units are lower case unless derived
from a persons name, e.g. ampere is A and kelvin is K.
Note the use of capitals and lower case and the increment Old units
on the exponent being factors of 3. Notice also centi and
deci are supposed to disappear with time leaving only the Occasionally, knowledge of older units may be required.
powers of 1000. Imperial units, or convert energies from BTUs in a ther-
modynamics project, etc.
In university laboratory classes you would probably be
1.3.2 Conversion factors
given material on the Quantity Calculus notation and
methodology which is to be highly recommended for sci-
The, (sometimes call caret or hat), sign is another nota-
entic work.
tion for to the power of. E means times 10 to the power
of, and is used a great deal in computer program output. A good reference for units, quantity calculus and SI is:
I. Mills, T. Cuitas, K. Homann, N. Kallay, K. Kuchitsu,
Quantities, units and symbols in physical chemistry, 2nd
Energy edition, (Oxford:Blackwell Scientic Publications,1993).
An approximation of how much of a chemical bond each
energy corresponds to is placed next to each one. This
indicates that light of energy 4 eV can break chemical
bonds and possibly be dangerous to life, whereas infrared 1.3.3 Greek alphabet
radiation of a few cm1 is harmless.
1.3.4 Unit labels
1 eV = 96.48530891 kJ mol1 (Near infrared), ap-
proximately 0.26 chemical bonds
The labelling of tables and axes of graphs should be done
1 1
1 kcal mol = 4.184000000 kJ mol (Near in- so that the numbers are dimensionless, e.g. temperature
frared), approximately 0.01 chemical bonds is T /K ,

1 MHz = 0.399031E-06 kJ mol1 (Radio waves), ln(k/k0 ) and energy mol / kJ etc.
approximately 0.00 chemical bonds This can look a little strange at rst. Examine good text
1 cm = 0.01196265819 kJ mol (Far infrared), books like Atkins Physical Chemistry which follow SI
1 1

approximately 0.00 chemical bonds carefully to see this in action.


The hardest thing with conversion factors is to get them
Wavelength, generally measure in nanometres and used the right way round. A common error is to divide when
in UV spectroscopy is dened as an inverse and so has a you should be multiplying, also another common error is
reciprocal relationship. to fail to raise a conversion factor to a power.
required Units
factor Conversion = given Units
Length
1 eV = 96.48530891 kJ mol1
There is the metre, the Angstrom (1010 m), the micron 1 cm1 = 0.01196265819 kJ mol1
(106 m), the astronomical unit (AU)and many old units To convert eV to cm1 , rst convert to kJ per mole
such as feet, inches and light years. by multiplying by 96.48530891 / 1. Then convert
to cm1 by multiplying by 1 / 0.01196265819 giving
Angles 8065.540901. If we tried to go directly to the conversion
factor in 1 step it is easy to get it upside down. However,
1
The radian to degree conversion is 57.2957795130824, common sense tell us that there are a lot of cm s in an
(i.e. a little bit less than 60, remember your equilateral eV so it should be obviously wrong.
triangle and radian sector). 1 inch = 2.54 centimetres. If there is a surface of nickel
electrode of 2 * 1.5 square inches it must be 2 * 1.5 *
2.542 square centimetres.
Dipole moment
To convert to square metres, the SI unit we must divide
1 Debye = 3.335640035 Cm x 1030 (coulomb metre) this by 100 * 100 not just 100.
6 CHAPTER 1. NUMBERS

Dimensional analysis 1.4.2 Combination of uncertainties

The technique of adding unit labels to numbers is espe- In an experimental situation, values with errors are often
cially useful, in that analysis of the units in an equation combined to give a resultant value. Therefore, it is nec-
can be used to double-check the answer. essary to understand how to combine the errors at each
stage of the calculation.

1.3.5 An aside on scaling


Addition or subtraction
One of the reasons powers of variables are so important
is because they relate to the way quantities scale. Physi- Assuming that x and y are the errors in measuring x
cists in particular are interested in the way variables scale and y , and that the two variables are combined by addi-
in the limit of very large values. Take cooking the turkey tion or subtraction, the uncertainty (absolute error) may
for Christmas dinner. The amount of turkey you can af- be obtained by calculating
ford is linear, (power 1), in your income. The size of an
(x)2 + (y)2
individual serving is quadratic, (power 2), in the radius
of the plates being used. The cooking time will be some- which can the be expressed as a relative or percentage
thing like cubic in the diameter of the turkey as it can be error if necessary.
presumed to be linear in the mass.
(In the limit of a very large turkey, say one the diameter
of the earth being heated up by a nearby star, the internal Multiplication or division
conductivity of the turkey would limit the cooking time
and the time taken would be exponential. No power can Assuming that x and y are the errors in measuring x
go faster / steeper than exponential in the limit. The series
and y , and that the two variables are combined by multi-
expansion of ex goes on forever even though 1/n! gets plication or division, the fractional error may be obtained
very small.) by calculating

Another example of this is why dinosaurs had fatter legs ( x 2 y 2
x ) +( y )
than modern lizards. If dinosaurs had legs in proportion
to small lizards the mass to be supported rises as length
to the power 3 but the strength of the legs only rises as
the area of the cross section, power 2. Therefore the big- 1.5 Plotting Graphs
ger the animal the more enormous the legs must become,
which is why a rhino is a very chunky looking version of
a pig. 1.5.1 The properties of graphs
There is a very good article on this in Cooper, Necia
Grant; West, Georey B., Particle Physics: A Los Alamos The most basic relationship between two variables x and
Primer, ISBN 0521347807 y is a straight line, a linear relationship.
y = mx + c
The variable m is the gradient and c is a constant which
1.3.6 Making tables gives the intercept. The equations can be more complex
than this including higher powers of x , such as
1.4 Statistics y = ax2 + bx + c
This is called a quadratic equation and it follows a shape
1.4.1 Denition of errors called a parabola. High powers of x can occur giving
cubic, quartic and quintic equations. In general, as the
For a quantity x the error is dened as x . Consider a power is increased, the line mapping the variables wiggles
burette which can be read to 0.05 cm3 where the volume more, often cutting the x -axis several times.
is measured at 50 cm3 .

Practice
The absolute error is x, 0.05 cm3
Plot x2 1 between 3 and +2 in units of 1.
The fractional error is x
x , 50 = 0.001
0.05

Plot x2 + 3x between 4 and +1 in units of 1.


The percentage error is 100 x
x = 0.1 % Plot 2x3 5x2 12x between 5 and +4 in units of 1.
1.6. COMPLEX NUMBERS 7

1.6 Complex numbers Eulers equation is obvious from looking at the Maclaurin
expansion of ei .
1.6.1 Introduction to complex numbers To nd the square root of i we use de Moivres theorem.

ei 2 = 0 + 1i
The equation:
2
so de Moivres theorem gives ei 2 = cos 2.2 + i sin 2.2
x + 6x + 13 = 0
= 12 + 12 i
Does not factorise
Check this by squaring up to give i .
x = 3 9 13
The other root comes from:
4 without complex
numbers does not exist. However
the number i = 1 behaves exactly like any other num- 4 + i sin 4 = 2 2 i
5 5 1 1

ber in algebra without any anomalies, allowing us to solve


de Moivres theorem can be used to nd the three cube
this problem.
roots of unity by
The solutions are 3 2i .
1 = cos + i sin
2i is an imaginary number. 32i is a complex number.
where can be 0 2/n .
Two complex numbers (a, b) = a + ib are added by
Put n = 1/3 , cos = 1/2 and sin = 3/2
(a, b) + (c, d) = (a + c, b + d) = a + c + i(b + d) (
)3 ( )( )
. 1 + 3i = 1 3 3i 1 + 3i
2 2 4 4 2 2 2
Subtraction is obvious: (a, b) (c, d) = (a c, b d) . ( )( )
= 12 3
2 i 21 + 3
2 i
(a, b).(c, d) = ac + i(ad + bc) bd
Division can be worked out as an exercise. It requires This is the dierence of two squares so
( 1 )2 3 2
(c + id)(c id) as a common denominator. This is c2 4 i = 14 + 34
2 2 2 2
(id) , (dierence of two squares), and is c + d .
Similarly any collection of n th roots of 1 can be obtained
This means this way.
(a,b) ac+bd+i(cbad)
(c,d) = c2 +d2 Another example is to get the expressions for cos 4 and
In practice complex numbers allow one to simplify the sin 4 without expanding cos(2 + 2) .
mathematics of magnetism and angular momentum as cos 4 + i sin 4 = (cos + i sin )4
well as completing the number system.
Remember Pascals Triangle
There is an apparent one to one correspondence between
the Cartesian xy plane and the complex numbers, x+iy 1 1 2 1 1 3 3 1 1 4 6 4 1 1 5 10 10 5 1 1 6 15 20 15 6 1
. This is called an Argand diagram. The correspon- = cos4 + 4 cos3 i sin + 6 cos2 i2 sin2 +
dence is illusory however, because say for example you 4 cos i3 sin3 + i4 sin4
raise the square root of i to a series of ascending pow-
= cos4 6 cos2 sin2 + sin4 + i(4 cos3 sin
ers. Rather than getting larger it goes round and round in
4 cos sin3 )
circles around the origin. This is not a property of ordi-
nary numbers and is one of the fundamental features of Separating the real and imaginary parts gives the two ex-
behaviour in the complex plane. pressions. This is a lot easier than
Plot on the same Argand diagram 2i, 3i1, 22i cos(2 + 2)
Solve Use the same procedure to get
2
x + 4x + 29 = 0 cos 6 and sin 6 .
4x2 12x + 25 = 0
x2 + 2ix + 1 = 0
(Answers 2 plus or minus 5i, 3/2 plus or minus 2i, i(1
plus or minus root 2)
2 important equations to be familiar with, Eulers equa-
tion:
ei = cos + i sin
and de Moivres theorem:
n
(cos + i sin ) = cos n + i sin n
Chapter 2

Trigonometry, matrices

2.1 Trigonometry 2.1.3 Trigonometric identities

cos2 + sin2 = 1
2.1.1 Free Web Based Material from UK
HEFCE Remember this is a consequence of Pythagoras theorem
where the length of the hypotenuse is 1.
There is a DVD on trigonometry at Math Tutor. sin( + ) = sin cos + cos sin
cos( + ) = cos cos sin sin
The dierence of two angles can easily be generated by
2.1.2 Trigonometry - the sin and cosine putting = and remembering sin = sin
rules and cos = cos . Similarly, the double angle for-
mulae are generated by induction. tan( + ) is a little
more complicated but can be generated if you can handle
B the fractions! The proofs are in many textbooks but as a
chemist it is not necessary to know them, only the results.

2.1.4 Identities and equations

Identities and equations look very similar, two things con-


h a nected by an equals sign. An identity however is a mem-
(hypotenuse) ory aid of a mathematical equivalence and can be proved.
An equation represents new information about a situation
(opposite) and can be solved.
For instance,
cos2 + sin2 = 1
is an identity. It cannot be solved for . It is valid for all
A b C . However,
(adjacent) cos2 = 1
is an equation where = arccos 1 .
Trigonometric triangle
If you try and solve an identity as an equation you will go
round and round in circles getting nowhere, but it would
In the following trigonometric identities, a , b and c are be possible to dress up cos2 + sin2 = 1 into a very
the lengths of sides in a triangle, opposite the correspond- complicated expression which you could mistake for an
ing angles A , B and C . equation.

The sin rule a


= b
= c
sin A sin B sin C 2.1.5 Some observations on triangles
The cosine rule c2 = a2 + b2 2ab cos C Check you are familiar with your elementary geome-
try. Remember from your GCSE maths the properties
sin A
The ratio identity tan A = cos A of equilateral and iscoceles triangles. If you have an is-

8
2.2. VECTORS 9

coceles triangle you can always dispense with the sin and the 3 dimensions:
cosine rules, drop a perpendicular down to the base and i + j + k

use trig directly. Remember that the bisector of a side or
an angle to a vertex cuts the triangle in two by area, an- so that
gle and length. This can be demonstrated by drawing an v = Axi + Ayj + Az k

obtuse triangle and seeing that the areas are 12 12 .R.h .
The hat on the i, j, k signies that it is a unit vector. This
is usually omitted.
2.1.6 The interior angles of a polygon Our geographical analogy allows us to see the meaning of
vector addition and subtraction. Vector products are less
Remember that the interior angles of a n -sided polygon obvious and there are two denitions the scalar product
are n * 180 360, and the vector product. These are dierent kinds of math-
( n 2 ). ematical animal and have very dierent applications. A
scalar product is an area and is therefore an ordinary num-
For benzene there are six equilateral triangles if the centre
ber, a scalar. This has many useful trigonometrical fea-
of the ring is used as a vertex, each of which having an
tures.
interior angle of 120 degrees. Work out the the angles in
azulene, (a hydrocarbon with a ve and a seven membered The vector product seems at rst to be dened rather
ring), assuming all the C-C bond lengths are equal, (this strangely but this denition maps onto nature as a very
is only approximately true). elegant way of describing angular momentum. The struc-
ture of Maxwells Equations is such that this deni-
tion simplies all kinds of mathematical descriptions of
2.2 Vectors atomic / molecular structure and electricity and mag-
netism.

2.2.1 Free Web Based Material from


HEFCE 2.2.3 A summary of vectors
There is a DVD on vectors at Math Tutor. The unit vectors in the 3 Cartesian dimensions:
i + j + k

2.2.2 Vectors a vector v is:
Imagine you make a rail journey from Doncaster to Bris- v = Axi + Ayj + Az k
tol, from where you travel up the West of the country The hat on the i, j, k signies that it is a unit vector.
to Manchester. Here you stay a day, travelling the next
morning to Glasgow, then across to Edinburgh. At the
end of a days work you return to Doncaster. Mathemat- Vector magnitude
ically this journey could be represented by vectors, (in
2 dimensions because we are at earthers on this scale).
2 2 2
At the end of the 2nd journey (D-B) + (B-M) you are Vmag = Ax + Ay + Az
only a short distance from Doncaster, 50 miles at 9.15
on the clockface. Adding two more vectors, (journeys)
takes you to Edinburgh, (about 250 miles at 12.00). Com- A constant times a vector
pleting the journey leaves you at a zero vector away from
Doncaster, i.e. all the vectors in this closed path add to vnew = cAxi + cAyj + cAz k

zero.
Mathematically we usually use 3 dimensional vectors over
the 3 Cartesian axes x , y and z . Vector addition

It is best always to use the conventional right handed axes A + B = (A + B )i + (A + B )j + (A + B )k


x x y y z z
even though the other way round is equally valid if used
consistently. The wrong handed coordinates can occa- Notice B + A = A + B
sionally be found erroneously in published research pa-
pers and text books. The memory trick is to think of a
sheet of graph paper, x is across as usual and y up the Vector subtraction
paper. Positive z then comes out of the paper.

A unit vector is a vector normalised, i.e. multiplied by a A B = (Ax Bx )i + (Ay By )j + (Az Bz )k

constant so that its value is 1. We have the unit vectors in Notice A B = (A + B)
10 CHAPTER 2. TRIGONOMETRY, MATRICES

Scalar Product If we have the equations:


3x + 4y = 4
A.B = Ax .Bx + Ay .By + Az .Bz
3x + 2y = 1
Notice B.A = A.B
If these are simultaneously true we can nd a unique so-
Notice that if A = B this reduces to a square. lution for both x and y .
If A and B have no common non-zero components in x By subtracting the 2 equations a new equation is created
, y and z the value is zero corresponding to orthogonal- where x has disappeared and the system is solved.
ity, i.e. they are at right angles. (This can also occur by
sign combinations making A.B zero. corresponding to 2y = 3
non axis-hugging right angles.) Substituting back y = 1.5 gives us x .
This was especially easy because x had the same coef-
Vector product cient in both equations. We can always multiply one
equation throughout by a constant to make the coe-
AxB = (Ay Bz Az By )i(Ax Bz Az Bx )j+(Ax By cients the same.

Ay Bx )k If the equations were:
Notice BxA = AxB 3x + 4y = 4
The minus sign on the middle term comes from the def- and
inition of the determinant, explained in the lecture. De-
6x + 8y = 8
terminants are dened that way so they correspond to
right handed rotation. (If you remember our picture of things would go horribly wrong when you tried to solve
cos2 + sin2 = 1 going round the circle, as one coordi- them because they are two copies of the same equation
nate goes up, i.e. more positive, another must go down. and therefore not simultaneous. We will come to this
Therefore rotation formulae must have both negative and later, but in the meantime notice that 3 times 8 = 4 times
positive terms.) Determinants are related to rotations and 6. If our equations were:
the solution of simultaneous equations. The solution of 3x + 4y + 9z = 4
n simultaneous equations can be recast in graphical form
as a rotation to a unit vector in n -dimensional space so 3x + 2y 4z = 1
therefore the same mathematical structures apply to both 9x + 2y 2z = 1
space and simultaneous equations.
we can still solve them but would require a lot of algebra
to reduce it to three (2x2) problems which we know we
can solve. This leads on to the subject of matrices and
2.3 Matrices and determinants determinants.
Simultaneous equations have applications throughout the
2.3.1 Simultaneous linear equations physical sciences and range in size from (2x2)s to sets of
equations over 1 million by 1 million.
If we have 2 equations of the form y = mx + c we may
have a set of simultaneous equations. Suppose two rounds
of drinks are bought in a cafe, one round is 4 halves of 2.3.2 Practice simultaneous equations
orange juice and 4 packets of crisps. This comes to 4
pounds 20. The thirstier drinkers at another table buy 4 Solve:
pints of orange juice and only 1 packet of crisps and this
comes to 6 pounds 30. So we have: x 2y = 1

4.20 = 2x + 4y x + 4y = 8
and and

6.30 = 4x + y x + 5y = 10

i.e. y = 4.20
24 x and y = 6.30 4x 7x 4y = 10
4
If you plot these equations they will be simultaneously Notice that you can solve:
true at x = 1.5 and y = 0.30 . 3x + 4y + 9z = 4
Notice that if the two rounds of drinks are 2 pints and 3x + 2y 4z = 1
2 packets of crisps and 3 pints and 3 packets of crisps
we cannot solve for the prices! This corresponds to two 0x + 0y 2z = 1
parallel straight lines which never intersect. because it breaks down into a (2x2) and is not truly a
2.3. MATRICES AND DETERMINANTS 11

(3x3). (In the case of the benzene molecular orbitals, As discussed earlier the solving of simultaneous equa-
which are (6x6), this same scheme applies. It becomes tions is equivalent in some deeper sense to rotation in n
two direct solutions and two (2x2) problems which can -dimensions.
be solved as above.)

2.3.4 Matrix multiply practice


2.3.3 Matrices i) Multiply the following (2x2) matrices.
( )( ) ( )
3 5 1 2 3x1 + 5x3 3x2 + 5x4
The multiplication of matrices has been explained in the =
6 4 3 4 6x1 + 4x3 6x2 + 4x4
lecture.

( ) 3 3 ii) Multiply the following (3x3) matrices.
0 1 2
A= B = 4 4 AB = 38.15 42.17 4.02 0.205 0.665 1
0 1 2
5 5 4.69 0.76 5.35 0.181 0.647 1
( )
14 14 9.94 8.20 2.74 0.207 0.476 1
14 14 You will notice that this gives a unit matrix as its product.

0 6 12 1 0 0
BA = 0 8 16 0 1 0
0 10 20 0 0 1

( ) 4 5 The rst matrix is the inverse of the 2nd. Computers use
ifA = 1 2 3 and B = 6 7 AB = the inverse of a matrix to solve simultaneous equations.
8 9
( ) y1 = a11 x1 + a12 x2 + a13 x3 ..... + a1n xn
40 60 y2 = a21 x2 + a22 x2 + a23 x3 ..... + a2n xn

but BA cannot exist. To be multiplied two matrices must If we have .....

have the 1st matrix with the same number of elements in .....
a row as the 2nd matrix has elements in its columns. yn = an1 xn + an2 xn + an3 x3 ..... + ann xn
aij = k aik bkj In matrix form this is....
where the aij s are the elements of A . Y = AX

a11 a12 a13 A1 Y = X
a21 a22 a23 In terms of work this is equivalent to the elimination
a31 a32 a33 method you have already employed for small equations
Look at our picture of cos and sin as represented by a but can be performed by computers for 10 000 simulta-
unit vector in a circle. The rotation of the unit vector neous equations.
j about the z -axis can be represented by the following
(Examples of large systems of equations are the tting of
mathematical construct.
reference data to 200 references molecules, dimension
cos() sin() 0 200, or the calculation of the quantum mechanical gra-
sin() cos() 0 dient of the energy where there is an equation for every
0 0 1 way of exciting 1 electron from an occupied orbital to an
excited, (called virtual, orbital, (typically 10 000 equa-
0 sin()
1 = cos() tions.)
0 0
In two dimensions we will rotate the vector at 45 degrees 2.3.5 Finding the inverse
between x and y :
( )( r ) (
cos
r rsin
)
How do you nd the inverse... You use Maple or Matlab
cos() sin() 2 2
+ 2
= r sin = on your PC but if the matrix is small you can use the
sin() cos() r + r cos
( ) 2 2 2 formula...
r
0 A1 = 1
DetA AdjA

This is if we rotate by +45 degrees. For = 45o Here Adj A is the adjoint matrix, the transposed matrix
cos(45) = cos 45 and sin(45) = sin 45 . So the of cofactors. These strange objects are best described by
rotation ips over to give (01) . The minus sign is nec- example.....

essary for the correct mathematics of rotation and is in 1 1 2

the lower left element to give a right handed sense to the 2 1 1

rotational sign convention. 3 1 1
12 CHAPTER 2. TRIGONOMETRY, MATRICES

This determinant is equal to: 1 ( 1 x 1 - 1 x (1)) - (1) ( 2.3.6 Determinants and the Eigenvalue
2 x 1 - 1 x 3) + 2 ( 2 x (1) - ( 1 x 3) each of these terms problem
is called a cofactor.

a11 a12 a13 In 2nd year
quantum chemistry you will come across this
E
a21 a22 a23 0 0

a31 a32 a33 E 0
object: =0
0 E
a11 (a22 a33 a23 a32 ) 0 0 E
DetA = 1i+j1 + 1i+j1 a12 (a21 a33 a23 a31 )
+ 1i+j1 a13 (a21 a32 a22 a31 ) You divide by and set ( E)/ to equal x to get:


This 1i+j1 thing gives the sign alternation in a form x 1 0 0
1 x 1 0
mathematicians like even though it is incomprehensible.
0 1 x 1 = 0

Use the determinant 0 0 1 x

1 1 2
Expand this out and factorise it into two quadratic equa-
2 1 1
tions to give:
3 1 1
(x2 + x 1)(x2 x 1) = 0
to solve the simultaneous equations on page 47 by the
matrix inverse method. The matrix corresponding to the which can be solved using x = b etc.
equations on p47.2 is:
1 1 2 6 2 1 1 = 3 3 1 1 6 The cofactors are 2 1 5 Simultaneous equations as linear algebra
1 5 2 3 3 3 You may nd these 9 copies of the
matrix useful for striking out rows and columns to form The above determinant is a special case of simultaneous
this inverse.... 1 1 2 1 1 2 1 1 2 2 1 1 2 1 1 2 1 1 equations which occurs all the time in chemistry, physics
3 1 1 3 1 1 3 1 1 1 1 2 1 1 2 1 1 2 2 1 1 2 1 and engineering which looks like this:
1 2 1 1 3 1 1 3 1 1 3 1 1 1 1 2 1 1 2 1 1 2
2 1 1 2 1 1 2 1 1 3 1 1 3 1 1 3 1 1 These are the (a11 )x1 + a12 x2 + a13 x3 ..... + a1n xn = 0
a21 x2 + (a22 )x1 + a23 x3 ..... + a2n xn = 0
little determinants with the 1 to the (n-1) factors and the
.....
value of the determinant is 9. The transposed matrix of
.....
cofactors is 2 1 3 1 5 3 5 2 3 So the inverse is 2
1 3 1/9 X 1 5 3 5 2 3 Giving a solution 2 1 a1n xn + a2n xn + an3 x3 ..... + (a11 )x1 = 0
3 6 1 1/9 X 1 5 3 X 3 = 1 5 2 3 6 2 This equation in matrix form is (A 1)x = 0 and the
This takes a long time to get all the signs right. Elimi- solution is Det(A 1) = 0 .
nation by subtracting equations is MUCH easier. How- This is a polynomial equation like the quartic above. As
ever as the computer cannot make sign mistakes it is not you know polynomial equations have as many solutions as
a problem when done by computer program. the highest power of x i.e. in this case n . These solutions
The following determinant corresponds to an equation can be degenerate for example the orbitals in benzene
which is repeated three times giving an unsolvable set of are
6
a degenerate pair because of the factorisation of the
simultaneous equations. x polynomial from the 6 Carbon-pz orbitals. In the 2nd
year you may do a lab exercise where you make the ben-
1 2 3
zene determinant and see that the polynomial is
1 2 3

2 4 6 (x2 4)(x2 1)(x2 1) = 0
Matrix multiplication is not necessarily commutative, from which the 6 solutions and the orbital picture are im-
which in English means AB does not equal BA all the mediately obvious.
time. Multiplication may not even be possible in the case The use of matrix equations to solve arbitrarily large
of rectangular rather than square matrices. problems leads to a eld of mathematics called linear al-
I will put a list of the properties and denitions of matrices gebra.
in an appendix for reference through the later years of the
course.
2.3.7 Matrices with complex numbers in
them

Work out the quadratic equation from the 3 determinants



x i x 1 1 + i
x 2
i x 1 x 1 2

2 i2 x
2.3. MATRICES AND DETERMINANTS 13

They are all the same! This exemplies a deeper property


of matrices which we will ignore for now other than to
say that complex numbers allow you to calculate the same
thing in dierent ways as well as being the only neat way
to formulate some problems.
Chapter 3

Calculus

3.1 Dierentiation 3.1.4 Dierentiating a product


d(u.v)
3.1.1 Free web-based material from dx = v du dv
dx + u dx

HEFCE Notice when dierentiating a product one generates two


terms. (Terms are mathematical expression connected by
There is a DVD on dierentiation at Math Tutor. a plus or minus.) An important point is that terms which
represent physical quantities must have the same units
and dimensions or must be pure dimensionless numbers.
You cannot add 3 oranges to 2 pears to get 5 orango-
3.1.2 The basic polynomial
pears. Integration by parts also generates an extra term
each time it is applied.
The most basic kind of dierentiation is:
f (x) = xn

f (x) = nxn1 3.1.5 Dierentiating a quotient
There are two simple rules: You use this to dierentiate tan .
d
( u ) v du
dx u dx
dv

dx v = v2
1. The derivative of a function times a constant is just
the same constant times the derivative.
Problems
2. The derivative of a sum of functions is just the sum
of the two derivatives. Dierentiate with respect to x
3x2 x(1 + x)(1 x)
To get higher derivatives such as the second derivative Notice we have (a2 b2 ) . 4x7 3x2
keep applying the same rules.
(3x + 2)2 + ex
One of the big uses of dierentiation is to nd the station-
8x6 12 x
ary points of functions, the maxima and minima. If the
function is smooth, (unlike a saw-tooth), these are easily x(x + 3)2
located by solving equations where the rst derivative is x2 (3x (2 + x)(2 x))
zero.
Evaluate the inner brackets rst.
Evaluate
( )
3.1.3 The chain rule d
e z
z9
3
dz 18 z 2

dy dy dw ( )
dx = dw . dx
d
c5
dc
This is best illustrated by example: nd dy
given y = d ( 1 )
d 2 3
dx 1 1
9
(x4 + 1) ( ( ))
d
Let y = u9 and u = x4 + 1 . d e
1
2 + 3 2
2
d
( ( ))
Now dy
du = 9u8 and du
dx = 4x3 dr e
5r
ra3 + rb6 + rc9

So using the chain rule we have 9(x4 + 1) .4x3 = a, b and c are constants. Dierentiate with respect to r .
8
8
36x3 (x4 + 1) 3re3r

14
3.2. INTEGRATION 15

d
( ) dy d y 2
dx x6 ex y = f (x) dx = f (x) dx2 = f (x)

Plot x3 + x2 6x between 4 and +3, in units of 1. (It


will speed things up if you factorise it rst. Then you will
Answers
see there are 3 places where f (x) = 0 so you only need
calculate 5 points.) By factorising you can see that this
3x2 + 6x 1
equation has 3 roots. Find the 2 turning points. (Dier-
28x6 6x entiate once and nd the roots of the quadratic equation
ex + 18x + 12 using x = b . . . . This gives the position of the 2 turn-
ing points either side of zero. As the equation is only
48x5 6
x in x3 it has 3 roots and 2 maxima / minima at the most
3x2 + 12x + 9 therefore we have solved everything. Dierentiate your
d2 y
quadratic again to get dx 2 . Notice that the turning point
4x3 + 9x2 8x d2 y
to the left of zero is a maximum i.e. dx2 = ve and the
z z8 6
e + + 2
d y

2 z3 other is a minimum i.e. dx2 = +ve .
5 3/2 5
2c or 2 c3 What is the solution and the turning point of y = x3 .
12 + 2
3 + 3
4 Solve x3 x = 0 , by factorisation.

e + 2
3 3 dy
(The 3 roots are 3,0 and +2. dx = f (x) = 3x2 +

5e5r + 3a
+ 6b
+ 9c 2x 6
r4 r7 r 10
e3r (3 9r) Solutions are 1/3( 19 1) and 1/3( 19 + 1) , i.e.
1.7863 and 1.1196.
x5 ex (x + 6) d2 y

dx2 = f (x) = 6x + 2
d2 y
There are 3 coincident solutions at x = 0 , dx2 = 0 , at
Harder dierentiation problems
0 so this is an inection point.
Dierentiate with respect to x : The roots are 0, 1 and 1.
4 5
x (x + 9)
5
5x2 (x2 7x + 9)
Dierentiate with respect to r
3.2 Integration
3
(r2 + 3r 1) e4r
3.2.1 Free Web Based Material from
Dierentiate with respect to HEFCE
( 2 )
4 3 19
1

e There is a DVD on integration at Math Tutor.


4
Evaluate
d
( 4z ( ))
dz e z 12 The basic polynomial
( ( ))
d 2 2
e 1 1
d 2
f (x) = xn
xn+1
f (x)dx = n+1 +c
3.1.6 Using dierentiation to check turn- This works ne for all powers except 1, for instance the
ing points integral of
1
dy dy x7
dx is the tangent or gradient. At a minimum dx is zero.
This is also true at a maximum or an inection point. The is just
d2 y
second gradient gives us the nature of the point. If dx 2 is 6x1 6 + c
positive the turning point is a minimum and if it is neg-
1 is clearly going to be a special case because it involves
ative a maximum. Most of the time we are interested in
an innity when x = 0 and goes to a steep spike as x
minima except in transition state theory.
gets small. As you have learned earlier this integral is the
If the equation of y = x3 is plotted, is is possible to see natural logarithm of x and the innity exists because the
that at x = 0 there is a third kind of point, an inection log of zero is minus innity and that of negative numbers
dy d2 y
point, where both dx and dx 2 are zero. is undened.
16 CHAPTER 3. CALCULUS

The integration and dierentiation of positive and therefore


negative powers dy 1
dx = x
dy
>>>>>>>>>>>>>>>>>>>>>>Dierentiation dx dx = y
1/3 x*x*x x*x 2x 2 0 0 0 0 1/3 therefore
x*x*x x*x 2x 2 ? ? ? ? Integra-
1
tion<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< x dx = ln x
I(x) H(x) G(x) F(x) ln(x) 1/x 1/(x*x)
Here I, H, G and F are more complicated functions in-
volving ln x .
3.2.4 Integrating 1/x like things
You will be able to work them out easily when you have
done more integration. The thing to notice is that the Just as
calculus of negative and positive powers is not symmet-
d 1
rical, essentially caused by the pole or singularity for x1n dx ln x = x
at x = 0 . so
d 1 du
dx ln u = u dx
3.2.2 Logarithms therefore by the chain rule

d f (x)
Logarithms were invented by Napier, a Scottish Laird, dx ln f (x) = f (x)
in the 17th-century. He made many inventions but his therefore
most enduring came from the necessity of doing the many f (x)
long divisions and trigonometric calculations used in as- f (x) dx = ln |f (x)| + c
tronomy. The Royal Navy in later years devoted great Examples of this are:
time and expense to developing logarithm technology for cos
the purposes of navigation, leading to the commissioning 1+sin d = ln(1 + sin ) + c
of the rst mechanical stored program computer, which or
was theoretically sound but could not be made by Charles sin
1+cos d = ln(1 + cos ) + c
Babbage between 1833 and 1871.
and
The heart of the system is the observation of the proper- y
e y
ties of powers: ey +2 dy = ln(e + 2) + c
ea As the integral of 1/x is ln so the dierential of ln is 1/x
eb
= eab
so
This means that if we have the inverse function of ex we
d
can change a long division into a subtraction by looking dx ln 5x3 = ln 5 + ln x3 = ln 5 + 3 ln x
up the exponents in a set of tables. Before the advent of ln 5 is just a constant so d = 0 so
dx
calculators this was the way many calculations were done.
d 3 3
ln 5x =
y = ex dx x
This can also be done by the chain rule
ln y = x
3
d 3 1 d(5x ) 15x2 3
Napier initially used logs to the base e for his calculations, dx ln 5x = 5x3 . dx = 5x3 = x
but after a year or so he was visited by Briggs who sug- What is interesting here is that the 5 has disappeared com-
gested it would be more practical to use base 10. How- pletely. The gradient of the log function is unaected by
ever base e is necessary for the purposes of calculus and a multiplier! This is a fundamental property of logs.
thermodynamics.

3.2.3 Integrating 1/x 3.2.5 Some observations on innity


dx
x = ey = ey also = x
dy Obviously 1
0 is .
y
This is true because e is our function which reduces or 0
0nd are undened but sometimes a large number over

grows at the rate of its own quantity.
a large number can have dened values. An example is
y = ln x the sin of 90 degrees, which you will remember has a
This is our denition of a logarithm. large opposite over a large hypotenuse but in the limit of
an innitesimally thin triangle they become equal. There-
dx 1
dy = dy also = x fore the sin is 1.
dx
3.2. INTEGRATION 17

3.2.6 Denite integrals (limits) 3.2.9 Integration by parts

Remember how we do a denite integral This is done in many textbooks and Wikipedia. Their
3 3 notation might be dierent to the one used here, which
0
(f (x)) = [F (x)]0 = F (3) F (0) hopefully is the most clear. You derive the expression
where F is the indenite integral of f . by taking the product rule and integrating it. You then

make one of the U V into a product itself to produce the
Here is an example where limits are used to calculate the
expression.
3 areas cut out by a quartic equation:
( )
x 2x x + 2x .
4 3 2 U V = U V U V
We see that x = 1 is a solution so we can do a polynomial (all integration with respect to x . Remember by
division:
U V = U [int] [diff][int]
x3 -x2 2x ------------------- x-1 ) x4 2x3 -x2 +2x x4 ( U gets dierentiated.)
-x3 -------- -x3 -x2 -x3 +x2 ------- 2x2 +2x 2x2 +2x
-------- 0 The important thing is that you have to integrate one ex-
pression of the product and dierentiate the other. In
So the equation is x(x x 2)(x 1) which factorises the basic scheme you integrate the most complicated ex-
2

to pression and dierentiate the simplest. Each time you do


x(x 2)(x + 1)(x 1) . this you generate a new term but the function being dif-
b 4 ferentiated goes to zero and the integral is solved. The
x 2x3 x2 + 2x = expression which goes to zero is U .
[ 5
a
]b
x /5 x4 /2 x3 /3 + x2 a
The other common scheme is where the parts formula
generates the expression you want on the right of the
equals and there are no other integral signs. Then you
3.2.7 Integration by substitution can rearrange the equation and the integral is solved.
This is obviously very useful for trig functions where
du sin cos sin cos sin ad innitum.
sin cos d = u d d
ex also generates itself and is susceptible to the same
where u = sin .
treatment.
x
2
sin cos d = sin2 + c
9 e sin x dx = (ex ) sin x (ex ) cos x dx
sin cos d similarly
10
= sin10 + c
= ex sin x + ex cos x dx

= ex (sin x + cos x) ex sin x dx + c
We now have our required integral on both sides of the
3.2.8 Simple integration of trigonometric equation so
and exponential Functions 1 x
= 2e (sin x + cos x) + c

4e2x dx

9e3x dx 3.2.10 Integration Problems

sin 2d
Integrate the following by parts with respect to x .
sin2 + cos2 d
xex
cos d
x2 sin x
x cos x
Answers x2 ex
x2 ln x
2e2x + c
ex sin x
3e3x + c
e2x cos x
21 cos 2 + c 7
x(1 + x)
+c
Actually this one can be done quite elegantly by parts, to
i.e. the integral of 1d . give a two term expression. Work this one out. Expand-
sin + c ing the original integrand by Pascals Triangle gives:
18 CHAPTER 3. CALCULUS

2 3 4 5 6 7 8 x + 7 x + 21 x + 35 x + 35 x + 21 x + 7 x + It has taken many decades of work to produce computa-


x tionally ecient solutions of this equation for polyatomic
The two term integral expands to molecules. Essentially one expands in coecients of the
atomic orbitals. Then integrates to make a dierential
2 3 4 5 6 7 8 9 1/2 x + 7/3 x + 21/4 x + 7 x + 35/6 x + 3 equation a set of numbers, integrals, in a matrix. Matrix
x + 7/8 x + 1/9 x - 1/72 algebra then nishes the job and nds a solution by solv-
So one can see it is correct on a term by term basis. ing the resultant simultaneous equations.

3x sin x
2x2 cos 2x 3.2.12 The calculus of trigonometric func-
tions
If you integrate x7 sin x you will have to apply parts 7
times to get x to become 1 thereby generating 8 terms:
There are many dierent ways of expressing the same
7 6 5 4 3 -x cos(x) + 7 x sin(x) + 42 x cos(x) - 210 x sin(x) thing in trig functions and very often successful integra-
- 840 x cos(x) + 2 2520 x sin(x) + 5040 x cos(x) - 5040 tion depends on recognising a trig identity.
sin(x) + c (Output from Maple.)
sin 2xdx = 12 cos 2x + c
Though it looks nasty there is quite a pattern to this, 7,
but could also be
7x6,7x6x5 -----7! and sin, cos, -sin, -cos, sin, cos etc
so it can easily be done by hand. sin2 x or cos2 x
(each with an integration constant!).
When applying calculus to these functions it is necessary
3.2.11 Dierential equations to spot which is the simplest form for the current manip-
ulation. For integration it often contains a product of a
First order dierential equations are covered in many function with its derivative like 2 sin x cos x where inte-
textbooks. They are solved by integration. (First order gration by substitution is possible.
dy dy
equations have dx , second order equations have dx and
Where a derivative can be spotted on the numerator and
d2 y
dx2 .) its integral below we will get a ln function. This is how
The arbitrary constant means another piece of informa- we integrate tan .
sin x
tion is needed for complete solution, as with the Newtons tan xdx = cos x dx = ln (cos x) + c
Law of Cooling and Half Life examples.
We can see this function goes to innity at /2 as it should
Provided all the x s can be got to one side and the y s to do.
another the equation is separable.
dy
2y dx = 6x2
3.2.13 Integration by rearrangement
2ydy = 6x2 dx
y 2 = 2x3 + c Take for example:

This is the general solution. cos2 xdx

Typical examples are: Here there is no sin function producted in with the cos
dy
dy
dx
powers so we cannot use substitution. However there are
y = x dx y = x the two trig identities
ln y = ln x + ln A (constant) i.e. y = Ax cos2 + sin2 = 1
by denition of logs. and
dy
dx = ky (1) cos2 sin2 = cos 2
dy
kxy = dx Using these we have
dy
y = kxdx (2) cos2 xdx = 21 (1 + cos 2x)dx

ln y = 1 2
(3) so we have two simple terms which we can integrate.
2 kx +A
This corresponds to:
y = Be 2 kx
1 2 3.2.14 The Maclaurin series
The Schrdinger equation is a 2nd order dierential We begin by making the assumption that a function can
equation e.g. for the particle in a box be approximated by an innite power series in x :
2 d2
2m dx2 + E = 0 f (x) = a0 + a1 x + a2 x2 + a3 x3 + a4 x4 + . . .
3.3. SOME USEFUL ASPECTS OF CALCULUS 19

By dierentiating and setting x = 0 one gets n! denominator becomes large quickly.



f (0) 2 f (0) 3
f (x) = f (0) + f (0)x + 2! x + 3! x +

f (0) 4 Trigonometric power series
4! x + ...
Sin, cos and ex can be expressed by this series approxi- Remember that when you use sin x = x and cos x =
mation 1 x2 /2
x3 x5 x7
sin x = x 3! + 5! 7! ... that x must be in radians.....
2 4 6
cos x = 1 x
2! + x
4! x
6! ...
ex = 1 + x + x2
+ x 3
+ x4
...
3.2.15 Calculus revision
2! 3! 4!
Notice ex also works for negative x . Problems
x
When dierentiated or integrated e generates itself!
integrate x to the power of x with respect to x
When dierentiated sin x generates cos x .
By using series we can convert a complex function into a 1. Dierentiate et cos 2t , with respect to t . (Hint -
polynomial, and can use sin x = x x3 /6 for small x . use the chain rule.)
In actual fact the kind of approximation used inside com- 4
2. Dierentiate x2 (3x + 1) . (Chain rule and product
ax2 +bx+c
puter programs is more like: Ax 2 +Bx+C rule here.)
These have greater range but are much harder to develop 3. Dierentiate ln (7x4 ) . (Hint - split it into a sum of
and a bit ddly on the calculator or to estimate by raw logs rst.)
brain power.
We cannot expand ln x this way because ln 0 is 4. Integrate ln x . (Hint - use integration by parts and
take the expression to be dierentiated as 1.)
. However ln(1 + x) can be expanded.
Work out the series for ln(1 + x) . Answers

1. It is just et . 2 sin 2t et cos 2t . Bring a et


Factorials
out of each term to simplify to a(b + c) .
The factorials you have seen in series come from repeated 2. 2x(9x + 1)(3x + 1) .
3
dierentiation. n! also has a statistical meaning as it is the
number of unique ways you can arrange n objects. 3. ln 7 + 4 ln x - therefore it is 4 times the derivative
of ln x .
0! is 1 by denition, i.e. the number of dierent ways you
can arrange 0 objects is 1. 4. You should get x ln x x by 1 application of parts.
In statistical thermodynamics you will come across many
factorials in expressions such as: W = n0 !nN !
1 !n2 !...

Factorials rapidly get unreasonably large: 6! = 720, 8! =


3.3 Some useful aspects of calculus
40320 but 12! = 479001600 so we need to divide them
out into reasonable numbers if possible, so for example 3.3.1 Limits
8!/6! = 7x8 .
Many textbooks go through the proper theory of dier-
entiation and integration by using limits. As chemists it
Stirlings approximation is possible to live without knowing this so we might well
not have it as an examinable topic. However here is how
Also in statistical thermodynamics you will nd Stirlings we dierentiate sin from 1st principles.
approximation: ( )
dy y
=
ln x! x ln x x dx x
limit x0
sin(x+x)sin x
This is proved and discussed in Atkins Physical Chem- = x
istry. sin x cos x+sin x cos xsin x
= x
How can you use series to estimate ln 2 . Notice that the
series for = x x + sinxx cos x
sin x sin x

ln(1 + x) converges extremely slowly. ex is much faster As sin x = x for small x this expression is cos x .
because the Similarly for cos
20 CHAPTER 3. CALCULUS

dy cos(+)cos
d =
cos cos sin sin cos
=
cos sin cos
=
This is equal to sin .

3.3.2 Numerical dierentiation


You may be aware that you can t a quadratic to 3 points,
a cubic to 4 points, a quartic to 5 etc. If you dierentiate a
function numerically by having two values of the function
dy
x apart you get an approximation to dx by constructing a
triangle and the gradient is the tangent. There is a forward
triangle and a backward triangle depending on the sign of
x . These are the forward and backward dierentiation
approximations.
If however you have a central value with a either side
you get the central dierence formula which is equivalent
to tting a quadratic, and so is second order in the small
value of x giving high accuracy compared with drawing
a tangent. It is possible to derive this formula by tting a
quadratic and dierentiating it to give:
dy f + +f 2f 0
dx = 2
HCl r-0.02 sigma (iso) 32.606716 142.905788
110.299071 HCl r-0.01 sigma (iso) 32.427188
142.364814 109.937626 HCl r0 Total shielding:
paramagnetic : diamagnetic sigma (iso) 32.249753
141.827855 109.578102 HCl r+0.01 sigma (iso)
32.074384 141.294870 109.220487 HCl r+0.02 sigma
(iso) 31.901050 140.765819 108.864769
This is calculated data for the shielding in ppm of the
proton in HCl when the bondlength is stretched or com-
pressed by 0.01 of an Angstrom (not the approved unit
pm). The total shielding is the sum of two parts, the para-
magnetic and the diamagnetic. Notice we have retained
a lot of signicant gures in this data, always necessary
when doing numerical dierentiation.
Exercise - use numerical dierentiation to calculated d
(sigma) / dr and d2 (sigma) / dr2 using a step of 0.01 and
also with 0.02. Use 0.01 to calculate d (sigma(para)) / dr
and d (sigma(dia)) / dr.

3.3.3 Numerical integration


Wikipedia has explanations of the Trapezium rule and
Simpsons Rule. Later you will use computer programs
which have more sophisticated versions of these rules
called Gaussian quadratures inside them. You will only
need to know about these if you do a numerical project
later in the course. Chebyshev quadratures are another
version of this procedure which are specially optimised
for integrating noisy data coming from an experimental
source. The mathematical derivation averages rather than
amplies the noise in a clever way.
Chapter 4

Tests and examples

4.1 Some mathematical examples a very natural description mathematically and maps
ideally onto the physical reality. Here we are inte-
applied to chemistry grating with respect to t and , the operating fre-
quency if it is a laser experiment is a constant, there-
4.1.1 Variable names fore it appears on the denominator in the integration.
2
cos(t) sin(t)dt = cos()2 + c In this case
The ubiquitous x is not always the variable as you will all we can see a physical interpretation for the integra-
know by now. One problem dealing with real applications tion constant. It will be a phase factor. If we were
is sorting out which symbols are the variables and which dealing with sunlight we might well be integrating
are constants. (If you look very carefully at professionally a dierent function over in order to calculate all
set equations in text books you should nd that there are of the phenomenon which has dierent strengths at
rules that constants are set in Roman type, i.e. straight the dierent light frequencies. Our integration lim-
letters and variables in italics. Do not rely on this as it is its would either be from zero to innity or perhaps
often ignored.) over the range of energies which correspond to vis-
Here are some examples where the variable is conven- ible light.
tionally something other than x .

1. The Euler angles which are used in rotation are con-


ventionally , and not the more usual angle
names and . The rotation matrix for the nal
twist in the
commonest Euler denition
therefore 4. This example is a laser experiment called Second
cos() sin() 0
Harmonic Generation. There is an electric eld F
looks like sin() cos() 0
, frequency and a property constant (2) . 0
0 0 1
is a fundamental constant. We have an intense
2. The energy transitions in the hydrogen atom which monochromatic laser eld uctuating at the fre-
give(the Balmer quency , (i.e. a strong light beam from a big
) series are given by the formula = laser). Fv sin(2t) Therefore the F 2 term con-
RH 212 n12 is just a single variable for the en-
ergy the tilde being a convention used by spectro- tributes 0 (2) Fv2 sin2 (2t) to the polarization.
scopists to say it is wavenumbers, (cm1 ). The H We know from trigonometric identities that sin2
subscript on RH has no mathematical meaning. It can be represented as a cosine of the double angle
is the Rydberg constant and is therefore in Roman cos(2) = 1 2 sin2 Therefore the polarization
type. RH is known very accurately as 109,677.581 is 21 0 (2) Fv2 (1 cos 4t) In this forest of sub-
cm1 . It has actually been known for a substantial scripts and Greek letters the important point is that
fraction of the class to make an error putting this there are two terms contributing to the output com-
fraction over a common denominator in examina- ing from (1 cos 4t) which multiplies the rest of
tion conditions. the stu. In summary we have A sin2 (t) is equal to
B(1 cos(2t)) where everything except the trig(t)
3. In the theory of light is used for frequency and and trig(2t) are to some extent unimportant for the
t not surprisingly for time. Light is an oscillating phenomenon of doubling the frequency. sin and cos
electric and magnetic eld therefore the cosine func- dier only in a phase shift so they represent the same
tion is a very good way of describing it. You can physical phenomenon, i.e. light, which has phase.
also see here the use of complex numbers. Using (One of the important properties of laser light is that
the real axis of the Argand diagram for the electric it is coherent, i.e. it all has the same phase. This is
eld and the imaginary axis for the magnetic eld is fundamentally embedded in our mathematics.)

21
22 CHAPTER 4. TESTS AND EXAMPLES

4.1.2 van der Waals Energy to give the shape of the potential and hence the infra-red
spectrum. E is usually modelled by a very complicated
The van der Waals energy between two inert gas atoms function whose dierentiation is not entered into lightly.
can be written simply as a function of r
A B
Evdw =
r 12 r6 4.1.4 A one-dimensional metal
12
Notice that the r term is positive corresponding to re-
pulsion. The r6 term is the attractive term and is negative A one-dimensional metal is modelled by an innite chain
corresponding to a reduction in energy. A and B are con- of atoms 150 picometres apart. If the metal is lithium
stants tted to experimental numbers. each nucleus has charge 3 and its electrons are modelled
This function is very easy to both dierentiate and inte- by the{ function}
grate. Work these out. In a gas simulation you would cos2 2 150 r

use the derivative to calculate the forces on the atoms and


which repeats every 150 pm. What constant must this
would integrate Newtons equations to nd out where the
function be multiplied by to ensure there are 3 electrons
atoms will be next.
on each atom? (Hint... integrate cos2 between either 2
Another potential which is used is: and + 2 or 75pm and +75pm according to your equa-
Evdw = AeBr Cr6 tion. This integral is a dimensionless number equal to
the number of electrons, so we will have to multiply by a
This has 1 more ttable constant. Integrate and dieren- normalisation constant.)
tiate this.
Here we have modelled the density of electrons. Later in
The rA12 rB6 is called a Lennard-Jones potential and is the second year you will see electronic structure more ac-
often expressed using the 2 parameters of an energy and curately described by functions for each independent elec-
a distance . tron called orbitals. These are subject to rigorous math-
(( ) ( )6 )
12 ematical requirements which means they are quite fun to
E(r) = 4 r r
calculate.
is an energy. Set the derivative of this to zero and nd
out where the van der Waals minimum is. Dierentiate
again and show that the derivative is positive, therefore 4.1.5 Keplers Laws
the well is a minimum, not a turning point.
Another physics problem but a good example of a log-log
plot is the radius and time period relations of the planets.
4.1.3 A diatomic potential energy surface
This data is dimensionless because we have divided by the
time / distance of the earth. We can take logs of both.
Try a least squares t on your spreadsheet program. Using
the Earth and Saturn data: (which is extremely bad labo-
ratory practice, to usejust two points from a data set!)
(logT )/(logr) = 1.5000359
so logT = 1.5logr = logr1.5
so T = r3/2 and T 2 = r3
This is Keplers 3rd law. If you use either a least squares
t gradient or the mercury to saturn data you get the same
powers. We have got away with not using a full data set
because the numbers given are unusually accurate and to
some extent tautological, (remember the planets go round
Interaction energy of argon dimer. The long-range part is due to in ellipses not circles!).
London dispersion forces

In a diatomic molecule the energy is expanded as the bond 4.1.6 Newtons law of cooling
stretches in a polynomial. We set x = (r r0 ) . At r0
the function is a minimum so there is no dE/dx term. is the excess temperature of a cooling body over room
2 3 temperature (20o C say). The rate of cooling is propor-
E = 12 kharm. ddxE2 + 16 kanharm. ddxE3 tional to the excess temperature.
Whatever function is chosen to provide the energy setting d = k
dt
the 1st derivative to zero will be required to calculate r0 .
dx 1
The 2nd and 3rd derivatives will then need to be evaluated using dy = dy
dx
4.2. TESTS AND EXAMS 23

dt
d = k
1
4.1.8 Partial fractions for the 2nd order
This is a dierential equation which we integrate with re- rate equation
spect to to get
In chemistry work you will probably be doing the 2nd or-
t = k1 ln + c
der rate equation which requires partial fractions in order
The water is heated to 80o C and room temperature is 20o to do the integrals.
C. At the beginning t = 0 and = 60 , so
If you remember we have something like
0 = k1 ln 60 + c 1 A B
(2x)(3x) = (2x) + (3x)
therefore
Put the right-hand side over a common denominator
ln 60
c= k 1 A(3x)+B(2x)
(2x)(3x) = (2x)(3x)
t= k1 ln + 1
k ln 60
This gives 1 = 3A Ax + 2B Bx
but ln 60 ln = ln 60

By setting x to 3 we get 1 = -B (B=1). Setting x = 0 and

so ln 60 = kt B = 1
After 5 minutes the water has cooled to 70o C. 1 = 3A 2 (A=+1) Check 1 = 3 -x 2 +x true...
so ln = 5k so k =
50
60 ln 15 5
6 = 1
5 ln 6
5 < math > Therefore
.So < math > k = 0.0365
1
(2x)(3x) dx = 1
(2x) 1
(3x) dx = ln(2 x) +
ln
60 = 0.0365t ln(3 x) + c

60 = e0.0365t noting the sign changes on integrating 1/(2-x) not 1/x.


by the denition of logarithms. This gives the plot of an
exponential decay between 80 and 20o C.
So after 10 minutes t = 20 + 60e0.365 = 61.6o C. 4.2 Tests and exams
After 20 minutes t = 20 + 60e0.73 = 49.9o C. After
30 minutes t = 20 + 60e1.10 = 40.1o C.
4.2.1 A possible nal test with explanatory
notes
4.1.7 Bacterial Growth This test was once used to monitor the broad learning of
university chemists at the end of the 1st year and is in-
2 grams of an organism grows by 1/10 gram per day per tended to check, somewhat lightly, a range of skills in
gram. only 50 minutes. It contains a mixture of what are per-
dm m ceived to be both easy and dicult questions so as to
dt = 10
give the marker a good idea of the students algebra skills
This is a dierential equation which is solved by integra- and even whether they can do the infamous integration by
tion thus parts.
10
m dm = dt (1) Solve the following equation for x
10 ln m = t + c x2 + 2x 15 = 0
t
ln m = 10 +c It factorises with 3 and 5 so : (x+5)(x3) = 0 therefore
therefore the roots are 5 and +3, not 5 and 3!
t
m = ec e 10 (2) Solve the following equation for x
When t = 0 we have 2 grams so 2x2 6x 20 = 0
m = 2et/10 Divide by 2 and get x2 3x 10 = 0 .
For the sample to double in mass This factorises with 2 and 5 so : (x 5)(x + 2) = 0
4 = 2e t/10 therefore the roots are 5 and 2.
(3) Simplify
2 = et/10
ln 2 = t ln w6 4 ln w
10
t = 10 ln 2 = 6.9315days Firstly 6 ln w 4 ln w so it becomes 2 ln w .

Half life calculations are similar but the exponent is neg- (4) What is
1
ative. log2 64
24 CHAPTER 4. TESTS AND EXAMPLES

64 = 8 x 8 so it also equals 23 x 23 i.e. 1


64 is 26 , therefore The rst term is b(b2 1) the second 12 ( b2 0) and
the answer is 6. the 3rd term zero. This adds up to b3 3/2b .
(5) Multiply the two complex numbers (16) What is the general solution of the following dier-
3 + 5i and 3 5i ential equation:
d A
These are complex conjugates so they are 32 minus i2 x dr = r
52 i.e. plus 25 so the total is 34. where A is a constant..
(6) Multiply the two complex numbers = A ln r + k .

(5, 2) and (5, 2) (17) Integrate by parts:
x sin xdx
The real part is 25 plus the 4i2 . The cross terms make Make x the factor to be dierentiated and apply the for-
10i and +10i so the imaginary part disappears. mula, taking care with the signs... sin x x cos x .
(7) Dierentiate with respect to x : (18)The Maclaurin series for which function begins with
1
2 3x
2 these terms?
3x
Answer: 2
6x 1 + x + x2 /2! + x3 /3! + x4 /4! + . . .
3x3
(8) 6
+ 3x3 It is ex ....
x4
Answer: 9x2 24 (19)Express
x5
x2
as partial fractions.
(9) 2 +2 x (x3)(x+4)
x
1 6
Answer: 1 1 It is ..... 7(x3) + 7(x+4)
x x3

(10) x3 (x (2x + 3)(2x 3)) (20) What is 2ei4 cos 4 in terms of sin and cos
Expand out the dierence of 2 squares rst.....collect and This is just Eulers equation..... 2e
i4
= 2 cos 4
multiply....then just dierentiate term by term giving: 2i sin 4
20x4 4x3 + 27x2 so one cos 4 disappears to give ... cos 4 2i sin 4 .
3
(11) 3x cos 3x
This needs the product rule.... Factor out the 9x2 .... 50 Minute Test II
9x2 (cos 3x x sin 3x)
(1) Simplify 2 ln(1/x3 ) + 5 ln x
(12) ln(1 x)2
1
(2)What is log10
This could be a chain rule problem....... 10 000
1
(1x) 2 .2.(1).(1 x) (3) Solve the following equation for t
or you could take the power 2 out of the log and go straight t 3t 4 = 0
2

to the same answer with a shorter version of the chain rule (4) Solve the following equation for w
to: (1x)
2
.
w2 + 4w 12 = 0
(13) Perform the following integrations:
( ) (5) Multiply the two complex numbers
2 cos2 + 2 d (4, 3) and (5, 2)
cos2 must be converted to a double angle form as shown (6) Multiply the two complex numbers 3+2i and 3
many times.... then all 3 bits are integrated giving ....... 2i
cos sin + + 2 (7) The Maclaurin series for which function begins with
( 3 4 )
(14) 8x + 3 dx 8 these terms?
x x
, which goes to ln , this is straightforward x x /6 + x /120 + . . .
3 5
Apart from x4
polynomial integration. Also there is a nasty trap in that (8) Dierentiate with respect to x :
two terms can be telescoped to x163 .
x3 (2 3x)2
( x82 + 4 ln x)
(9) 2x 23x
(15) What is the equation corresponding to the determi-
nant: (10) x4 3x2 + k

b 1 0 where k is a constant.
1 2
b 1 = 0
2 (11) 3x2 4 Ax4
0 1 b
where A is a constant.
4.2. TESTS AND EXAMS 25

(12) 3x3 e3x (17) What is the equation corresponding to the determi-
(13) ln(2 x) 3 nant:

x 1 0
(14) Perform the following integrations:
1 x 0 = 0
( 4 )
3w 2w2 + 5w6 2 dw 0 0 x
x1
(15) (3 cos + ) d (18) Express (x+3)(x4) as partial fractions.
(16) What is the equation belonging to the determinant (19)What is the general solution of the following dier-
\begin{vmatrix} x & 0 & 0\\ 0 & x & i \\ 0 & i & x \\ ential equation:
\end{vmatrix} = 0</math> d r
dr = A
(17) What is the general solution of the following dier-
ential equation: (20) What is ei2 2i sin 2 in terms of sin and cos.
dy
dx = ky
(18) Integrate by any appropriate method:
( )
ln x + x4 dx
x+1
(19) Express (x2)(x+2)

as partial fractions.
(20) What is 2ei2 + 2i sin 2 in terms of sin and cos.

4.2.2 50 Minute Test III

(1) Solve the following equation for t


t2 4t 12 = 0
1
(2) What is log4 16
(3) The Maclaurin series for which function begins with
these terms?
1x2 /2+x4 /24+. . . ---- (4) Dierentiate with respect
to x :
5
x2 8x4

(5) 4
x
2x
6
(6) 5 x + x3
(7) 5
x3 5x 3

(8) x (2x2 (5 + 2x)(5 2x))


2

(9) 2x2 sin x


(10) Multiply the two complex numbers
(2, 3) and (2, 3)
(11) Multiply the two complex numbers 3 i and
3+i
(12) Perform the following integrations:
(1 6
)
3x + 3x2 5x
1
dx
(13)
( 2 2 )
6x + x x82 dx
( 2 )
(14) cos + d
( 3 )
(15) sin cos + 2 d

(16) Integrate by parts: 2x cos xdx
Chapter 5

Further reading

5.1 Further reading Maths for Chemistry is an online resource providing in-
teractive context-based resources which explain how var-
ious aspects of maths can be applied to chemistry. There
5.1.1 Further reading
are quizzes and downloadable les to check understand-
ing.
Books

Paul Monk, Maths for Chemistry: A Chemists


Toolkit of Calculations, (Oxford, 2006), ISBN 978-
0199277414.

Bostock and Chandler, Core Maths for A-level,


Third Edition, (Nelson Thornes, 2000), ISBN 978-
0748755097.

M. C. R. Cockett and G. Doggett, Maths for


Chemists: Numbers, Functions and Calculus v. 1,
(Royal Society of Chemistry, London, 2003), ISBN
978-0854046775

M. C. R. Cockett and G. Doggett, Maths for


Chemists: Power Series Complex Numbers and Lin-
ear Algebra v. 2, (Royal Society of Chemistry, Lon-
don, 2003), ISBN 978-0854044955

G. Currell and T. Dowman,Mathematics and Statis-


tics for Science, (Wiley, 2005), ISBN 978-
0470022290.

Stephen K. Scott, Beginning Maths for Chemistry,


(Oxford, 1995), ISBN 978-0198559306

Peter Tebbutt, Basic Mathematics for Chemists, 2nd


edition, (Wiley, 1994), ISBN 978-0471972839

Online resources

There is much useful free material relevant to this book,


including downloadable DVDs, funded by the HEFCE
Fund for the Development of Teaching & Learning and
the Gatsby Technical Education Project in association
with the Higher Education Academy at Math Tutor.
Discover Maths for Chemists from the Royal Society of
Chemistry is a a one-stop site designed by chemists for
chemists. This new free-to-use site brings together all the
best resources to help you combine maths and chemistry.

26
Chapter 6

Text and image sources, contributors, and


licenses

6.1 Text
Mathematics for Chemistry/Introduction Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Introduction?oldid=
1938631 Contributors: Amigadave and Adrignola
Mathematics for Chemistry/Number theory Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Number_theory?
oldid=2097024 Contributors: Whiteknight, Jguk, MartinY, Mikes bot account, Amigadave, NipplesMeCool, Adrignola and Anonymous:
2
Mathematics for Chemistry/Functions Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Functions?oldid=1612124
Contributors: Amigadave and Adrignola
Mathematics for Chemistry/Units and dimensions Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Units_and_
dimensions?oldid=2587155 Contributors: Whiteknight, Jomegat, Jguk, MartinY, Mikes bot account, QuiteUnusual, Amigadave, Adrignola
and Anonymous: 8
Mathematics for Chemistry/Statistics Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Statistics?oldid=1612175
Contributors: Whiteknight, MartinY, Mikes bot account, Amigadave and Adrignola
Mathematics for Chemistry/Plotting graphs Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Plotting_graphs?
oldid=1612136 Contributors: Amigadave and Adrignola
Mathematics for Chemistry/Complex Numbers Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Complex_
Numbers?oldid=2514602 Contributors: Whiteknight, Jguk, MartinY, Mikes bot account, Amigadave, Adrignola and Anonymous: 1
Mathematics for Chemistry/Trigonometry Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Trigonometry?oldid=
2174342 Contributors: Whiteknight, MartinY, Mikes bot account, Amigadave, Adrignola and HethrirBot
Mathematics for Chemistry/Vectors Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Vectors?oldid=2358133 Con-
tributors: Whiteknight, MartinY, Recent Runes, Mikes bot account, Amigadave, Adrignola and Anonymous: 2
Mathematics for Chemistry/Matrices and Determinants Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/
Matrices_and_Determinants?oldid=2563875 Contributors: Whiteknight, Jguk, MartinY, Mikes bot account, Amigadave, Adrignola, Lla-
maAl and Anonymous: 4
Mathematics for Chemistry/Dierentiation Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Differentiation?oldid=
2168593 Contributors: Whiteknight, Jguk, MartinY, Mikes bot account, Amigadave, Adrignola, Frigotoni and Anonymous: 1
Mathematics for Chemistry/Integration Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Integration?oldid=
2680267 Contributors: Whiteknight, Jomegat, Jguk, MartinY, Xania, Mikes bot account, Jake Wasdin~enwikibooks, Amigadave,
Adrignola and Anonymous: 5
Mathematics for Chemistry/Some Useful Aspects of Calculus Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/
Some_Useful_Aspects_of_Calculus?oldid=1612173 Contributors: Whiteknight, Jguk, MartinY, Mikes bot account, Amigadave and
Adrignola
Mathematics for Chemistry/Some Mathematical Examples applied to Chemistry Source: https://en.wikibooks.org/wiki/
Mathematics_for_Chemistry/Some_Mathematical_Examples_applied_to_Chemistry?oldid=1612172 Contributors: Whiteknight, Jguk,
MartinY, Mikes bot account, Amigadave, Adrignola and Anonymous: 1
Mathematics for Chemistry/Tests and Exams Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Tests_and_Exams?
oldid=1612176 Contributors: Whiteknight, MartinY, Mikes bot account, Amigadave and Adrignola
Mathematics for Chemistry/Further reading Source: https://en.wikibooks.org/wiki/Mathematics_for_Chemistry/Further_reading?
oldid=1731273 Contributors: Thenub314, Amigadave, Adrignola, Msuxg~enwikibooks and Gibbs~enwikibooks

27
28 CHAPTER 6. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

6.2 Images
File:Argon_dimer_potential.png Source: https://upload.wikimedia.org/wikipedia/commons/0/04/Argon_dimer_potential.png License:
CC-BY-SA-3.0 Contributors: Self-made with Gnuplot. Converted to PNG with GIMP Original artist: Poszwa
File:Book_important2.svg Source: https://upload.wikimedia.org/wikipedia/commons/9/91/Book_important2.svg License: CC BY-SA
3.0 Contributors: Own work Original artist: darklama
File:Trigonometric_Triangle.svg Source: https://upload.wikimedia.org/wikipedia/commons/4/48/Trigonometric_Triangle.svg License:
CC-BY-SA-3.0 Contributors: en:Image:Trigonometric Triangle.PNG/Image:Trigonometry triangle.svg Original artist: Traced by User:
Stannered
File:Wikipedia-logo.png Source: https://upload.wikimedia.org/wikipedia/commons/6/63/Wikipedia-logo.png License: GFDL Contrib-
utors: based on the rst version of the Wikipedia logo, by Nohat. Original artist: version 1 by Nohat (concept by Paullusmagnus);

6.3 Content license


Creative Commons Attribution-Share Alike 3.0