Mathematics for Chemistry/Print version

< Mathematics for Chemistry

Table of contentsEdit

  1. Introduction
  2. Number theory
  3. Functions
  4. Units and dimensions
  5. Statistics
  6. Plotting graphs
  7. Complex numbers
  8. Trigonometry
  9. Vectors
  10. Matrices and determinants
  11. Differentiation
  12. Integration
  13. Some useful aspects of calculus
  14. Enzyme kinetics
  15. Some mathematical examples applied to chemistry
  16. Tests and exams
  17. Further reading

IntroductionEdit

  1. 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 probably over for ever. The broadening out of school curricula has had several effects, including student entry with a more diverse educational background and has also resulted in the subject areas Chemistry, Physics and Mathematics 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 significant mathematics in it. This is to allow the subject to be studied without any of the maths which might be first 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 component of university chemistry to teach just the Mathematics 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.

Quantitative methods in chemistryEdit

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 consumed, how long does this take, how likely is the reaction to take place.
  • Chemists work with a variety of different units, with wildly different 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 represented as equations are manipulated and evaluated in order to obtain information.

NumbersEdit

  1. ==Numbers==
For more details on this topic, see Number Theory.

Real numbers come in several varieties and forms;

  • Natural numbers are integers that are greater than or equal to zero.
  • Integers are whole numbers used for counting indivisible objects, together with negative equivalents and zero, e.g. 42, -7, 0
  • Rational numbers can always be expressed as fractions, e.g. 4.673 = 4673/1000.
  • Irrational numbers, unlike rational numbers, cannot be expressed as a fraction or as a definite decimal, e.g.   and  

It is also worth noting that the imaginary unit and therefore complex numbers are used in chemistry, especially when dealing with equations concerning waves.

SurdsEdit

The origin of surds goes back to the Greek philosophers. It is relatively simple to prove that the square root of 2 cannot be a ratio of two integers, no matter how large the integers may become. In a rather Pythonesque incident the inventor of this proof was put to death for heresy by the other philosophers because they could not believe such a pure number as the root of 2 could have this impure property.

(The original use of quadratic equations is very old, Babylon many centuries BC.) This was to allocate land to farmers in the same quantity as traditionally held after the great floods on the Tigris and Euphrates had reshaped the fields. The mathematical technology became used for the same purpose in the Nile delta.

When you do trigonometry later you will see that surds are in the trigonometric functions of the important symmetrical angles, e.g.   and so they appear frequently in mathematical expressions regarding 3 dimensional space.

NotationEdit

The notation used for recording numbers in chemistry is the same as for other scientific disciplines, and appropriately called scientific notation, or standard form. It is a way of writing both very large and very small numbers in a shortened form compared to decimal notation. An example of a number written in scientific notation is

 

with   being a coefficient termed the significand or the mantissa, and   being an integer exponent. When written in decimal notation, the number becomes

 .

Numbers written in scientific notation are usually normalised, such that only one digit precedes the decimal point. This is to make order of magnitude comparisons easier, by simply comparing the exponents of two numbers written in scientific notation, but also to minimise transcription errors, as the decimal point has an assumed position after the first digit. In computing and on calculators, it is common for the   ("times ten to the power of") to be replaced with "E" (capital e). It is important not to confuse this "E" with the mathematical constant e.

Engineering notation is a special restriction of scientific notation where the exponent must be divisible by three. Therefore, engineering notation is not normalised, but can easily use SI prefixes for magnitude.

Remember that in SI, numbers do not have commas between the thousands, instead there are spaces, e.g.  , (an integer) or  . Commas are used as decimal points in many countries.

ExponentsEdit

Consider a number  , where   is the base and   is the exponent. This is generally read as “  to the  ” or “  to the power of  ”. If   then it is common to say “  squared”, and if   then “  cubed”. Comparing powers (exponentiation) to multiplication for positive integer values of n  , it can be demonstrated that

 , i.e. four lots of   added together

 , i.e.   multiplied by itself four times.

For  , the result is simply  . For   the result is  .

  can be reduced to multiplication like this if   is an integer:  

Order of operationsEdit

When an expression contains different operations, they must be evaluated in a certain order. Exponents are evaluated first. Then, multiplication and division are evaluated from left to right. Last, addition and subtraction are evaluated left to right. Parentheses or brackets take precedence over all operations. Anything within parentheses must be calculated first. A common acronym used to remember the order of operations is PEMDAS, for "Parentheses, Exponents, Multiplication, Division, Addition, Subtraction". Another way to remember this acronym is "Please Excuse My Dear Aunt Sally".

Keep in mind that negation is usually considered multiplication. So in the case of  , the exponent would be evaluated first, then negated, resulting in a negative number.

Take note of this example:

 

If evaluated incorrectly (left-to-right, with no order of operations), the result would be 16. Three plus five gives eight, times two is 16. The correct answer should be 13. Five times two gives ten, plus three gives 13. This is because multiplication is solved before addition.

Partial fractionsEdit

Partial fractions are used in a few derivations in thermodynamics and they are good for practicing algebra and factorisation.

It is possible to express quotients in more than one way. 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 is often difficult, whereas by splitting it up into a sum, a sum of standard integrals is obtained. This is the principal chemical application of partial fractions.

An example is

 

In the above   must equal   since the denominators are equal. So we set   first to +1 giving  . Therefore B = -1/2. If we set   instead  , therefore  . So

 

We can reverse this process by use of a common denominator.

 

The numerator is  , so it becomes

 

which is what we started from.

So we can generate a single term by multiplying by the denominators to create a common denominator and then add up the numerator to simplify. A typical application might be to convert a term to partial fractions, do some calculus on the terms, and then regather into one quotient for display purposes. In a factorised single quotient it will be easier to see where numerators go to zero, giving solutions to  , and where denominators go to zero giving infinities.

A typical example of a meaningful infinity in chemistry might be an expression such as

 

The variable is the energy E, so this function is small everywhere, except near  . Near   a resonance occurs and the expression becomes infinite when the two energies are precisely the same. A molecule which can be electronically excited by light has several of these resonances.

Here is another example. If we had to integrate the following expression we would first convert to partial fractions:

 

so

 

let   then  

let   then  

therefore the expression becomes

 

Later you will learn that these expressions integrate to give simple expressions in terms of the natural logarithm.

ProblemsEdit

 

 

 

AnswersEdit

 

 

 

Polynomial divisionEdit

This is related to partial fractions in that its principal use is to facilitate integration.

Divide out

 

like this


               3x  - 7
             -----------------
      x + 1  ) 3x2  -4x   -6
               3x2  +3x
               ---------
                0   -7x   -6
                    -7x   -7
                    --------
                           1

So our equation becomes  

This can be easily differentiated, and integrated. If this is differentiated with the quotient formula it is considerably harder to reduce to the same form. The same procedure can be applied to partial fractions.

Substitutions and expansionsEdit

You can see the value of changing the variable by simplifying

 

to

 

where

 

This is an example of simplification. It would actually be possible to differentiate this with respect to either   or   using only the techniques you have been shown. The algebraic manipulation involves differentiation of a quotient and the chain rule.

Evaluating   gives

 

Expanding this out to the  s and  s would look ridiculous.

Substitutions like this are continually made for the purpose of having new, simpler expressions, to which the rules of calculus or identities are applied.

FunctionsEdit

  1. Functions as tools in chemistryEdit

    The quadratic formulaEdit

    In order to find the solutions to the general form of a quadratic equation,

     

    there is a formula

     

    (Notice the line over the square root has the same priority as a bracket. Of course we all know by now that   is not equal to   but errors of priority are among the most common algebra errors in practice).

    There is a formula for a cubic equation but it is rather complicated and unlikely to be required for undergraduate-level study of chemistry. Cubic and higher equations occur often in chemistry, but if they do not factorise they are usually solved by computer.

    Solve:

     

     

    Notice the scope or range of the bracket.

     

     

    Notice here that the variable is a concentration, not the ubiquitous  .

    Units and dimensionsEdit

    1. ==Units, multipliers and prefixes==

    It is usually necessary in chemistry to be familiar with at least three systems of units, Le Système International d'Unités (SI), atomic units as used in theoretical calculations and the unit system used by the experimentalists. Thus if we are dealing with the ionization energy, the units involved will be the Joule (J), the Hartree (Eh, the atomic unit of energy), and the electron volt (eV).

    These units all have their own advantages;

    • The SI unit should be understood by all scientists regardless of their field.
    • The atomic unit system is the natural unit for theory as most of the fundamental constants are unity and equations can be cast in dimensionless forms.
    • The electron volt comes from the operation of the ionization apparatus where individual electrons are accelerated between plates which have a potential difference in Volts.

    An advantage of the SI system is that the dimensionality of each term is made clear as the fundamental constants have structure. This is a complicated way of saying that if you know the dimensionality of all the things you are working with you know an awful lot about the mathematics and properties such as scaling with size of your system. Also, the same system of units can describe both the output of a large power station (gigaJoules), or the interaction of two inert gas atoms, (a few kJ per mole or a very small number of Joules per molecule when it has been divided by Avogadro's number).

    In SI the symbols for units are lower case unless derived from a person's name, e.g. ampere is A and kelvin is K.

    SI base units
    Name Symbol Quantity
    metre m length
    kilogram kg mass
    second s time
    ampere A electric current
    kelvin K thermodynamic temperature
    candela cd luminous intensity
    mole mol amount of substance
    Derived units used in chemistry
    Quantity Unit Name Symbol
    Area m2
    Volume m3
    Velocity m s-1
    Acceleration m s-2
    Density kg m-3
    Entropy J mol-1 K-1
    Force kg m s-2 newton N
    Energy N m joule J
    Pressure N m-2 pascal Pa
    Frequency s-1 hertz Hz
    Approved prefixes for SI units
    Prefix Factor Symbol
    atto 10-18 a
    femto 10-15 f
    pico 10-12 p
    nano 10-9 n
    micro 10-6 μ
    milli 10-3 m
    centi 10-2 c
    deci 10-1 d
    kilo 103 k
    mega 106 M
    giga 109 G
    tera 1012 T
    peta 1015 P
    exa 1018 E

    Note the use of capitals and lower case and the increment on the exponent being factors of 3. Notice also centi and deci are supposed to disappear with time leaving only the powers of 1000.

    Conversion factorsEdit

    The  , (sometimes call caret or hat), sign is another notation for to the power of. E means times 10 to the power of, and is used a great deal in computer program output.

    EnergyEdit

    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 radiation of a few cm-1 is harmless.

    • 1 eV = 96.48530891 kJ mol-1 (Near infrared), approximately 0.26 chemical bonds
    • 1 kcal mol-1 = 4.184000000 kJ mol-1 (Near infrared), approximately 0.01 chemical bonds
    • 1 MHz = 0.399031E-06 kJ mol-1 (Radio waves), approximately 0.00 chemical bonds
    • 1 cm-1 = 0.01196265819 kJ mol-1 (Far infrared), approximately 0.00 chemical bonds

    Wavelength, generally measure in nanometres and used in UV spectroscopy is defined as an inverse and so has a reciprocal relationship.

    LengthEdit

    There is the metre, the Angstrom (10-10 m), the micron (10-6 m), the astronomical unit (AU)and many old units such as feet, inches and light years.

    AnglesEdit

    The radian to degree conversion is 57.2957795130824, (i.e. a little bit less than 60, remember your equilateral triangle and radian sector).

    Dipole momentEdit

    1 Debye = 3.335640035 Cm x 10-30 (coulomb metre)

    Magnetic SusceptibilityEdit

    1 cm3 mol-1 = 16.60540984 JT-2 x 1030 (joule tesla2)

    Old unitsEdit

    Occasionally, knowledge of older units may be required. Imperial units, or convert energies from BTUs in a thermodynamics project, etc.

    In university laboratory classes you would probably be given material on the Quantity Calculus notation and methodology which is to be highly recommended for scientific work.

    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 edition, (Oxford:Blackwell Scientific Publications,1993).

    Greek alphabetEdit

    Unit labelsEdit

    The labelling of tables and axes of graphs should be done so that the numbers are dimensionless, e.g. temperature is  ,

      and energy mol / kJ etc.

    This can look a little strange at first. Examine good text books like Atkins Physical Chemistry which follow SI carefully to see this in action.

    The hardest thing with conversion factors is to get them the right way round. A common error is to divide when you should be multiplying, also another common error is to fail to raise a conversion factor to a power.

     

    1 eV = 96.48530891 kJ mol-1

    1 cm-1 = 0.01196265819 kJ mol-1

    To convert eV to cm-1, first convert to kJ per mole by multiplying by 96.48530891 / 1. Then convert to cm-1 by multiplying by 1 / 0.01196265819 giving 8065.540901. If we tried to go directly to the conversion factor in 1 step it is easy to get it upside down. However, common sense tell us that there are a lot of cm-1s in an eV so it should be obviously wrong.

    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.

    To convert to square metres, the SI unit we must divide this by 100 * 100 not just 100.

    Dimensional analysisEdit

    The technique of adding unit labels to numbers is especially useful, in that analysis of the units in an equation can be used to double-check the answer.

    An aside on scalingEdit

    One of the reasons powers of variables are so important is because they relate to the way quantities scale. Physicists in particular are interested in the way variables scale in the limit of very large values. Take cooking the turkey for Christmas dinner. The amount of turkey you can afford is linear, (power 1), in your income. The size of an individual serving is quadratic, (power 2), in the radius of the plates being used. The cooking time will be something like cubic in the diameter of the turkey as it can be 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 conductivity of the turkey would limit the cooking time and the time taken would be exponential. No power can go faster / steeper than exponential in the limit. The series expansion of   goes on forever even though   gets very small.)

    Another example of this is why dinosaurs had fatter legs 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 bigger the animal the more enormous the legs must become, which is why a rhino is a very chunky looking version of a pig.

    There is a very good article on this in Cooper, Necia Grant; West, Geoffrey B., Particle Physics: A Los Alamos Primer, ISBN 0521347807

    Making tablesEdit

    StatisticsEdit

    1. Definition of errorsEdit

      For a quantity   the error is defined as  . Consider a burette which can be read to ±0.05 cm3 where the volume is measured at 50 cm3.

      • The absolute error is  
      • The fractional error is  ,  
      • The percentage error is  %

      Combination of uncertaintiesEdit

      In an experimental situation, values with errors are often combined to give a resultant value. Therefore, it is necessary to understand how to combine the errors at each stage of the calculation.

      Addition or subtractionEdit

      Assuming that   and   are the errors in measuring   and  , and that the two variables are combined by addition or subtraction, the uncertainty (absolute error) may be obtained by calculating

       

      which can the be expressed as a relative or percentage error if necessary.

      Multiplication or divisionEdit

      Assuming that   and   are the errors in measuring   and  , and that the two variables are combined by multiplication or division, the fractional error may be obtained by calculating

       

      Plotting graphsEdit

      1. The properties of graphsEdit

        The most basic relationship between two variables   and   is a straight line, a linear relationship.

         

        The variable   is the gradient and   is a constant which gives the intercept. The equations can be more complex than this including higher powers of  , such as

         

        This is called a quadratic equation and it follows a shape called a parabola. High powers of   can occur giving cubic, quartic and quintic equations. In general, as the power is increased, the line mapping the variables wiggles more, often cutting the  -axis several times.

        PracticeEdit

        Plot   between -3 and +2 in units of 1.

        Plot   between -4 and +1 in units of 1.

        Plot   between -5 and +4 in units of 1.

        Complex numbersEdit

        1. ==Introduction to complex numbers==

        The equation:

         

        Does not factorise

         

          without complex numbers does not exist. However the number   behaves exactly like any other number in algebra without any anomalies, allowing us to solve this problem.

        The solutions are  .

          is an imaginary number.   is a complex number.

        Two complex numbers   are added by    .

        Subtraction is obvious:  .

         

        Division can be worked out as an exercise. It requires  as a common denominator. This is  , (difference of two squares), and is  .

        This means

         

        In practice complex numbers allow one to simplify the mathematics of magnetism and angular momentum as well as completing the number system.

        There is an apparent one to one correspondence between the Cartesian   plane and the complex numbers,  . This is called an Argand diagram. The correspondence is illusory however, because say for example you raise the square root of   to a series of ascending powers. Rather than getting larger it goes round and round in circles around the origin. This is not a property of ordinary numbers and is one of the fundamental features of behaviour in the complex plane.

        Plot on the same Argand diagram  

        Solve

         

         

         

        (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, Euler's equation:

         

        and de Moivre's theorem:

         

        Euler's equation is obvious from looking at the Maclaurin expansion of  .

        To find the square root of   we use de Moivre's theorem.

         

        so de Moivre's theorem gives  

         

        Check this by squaring up to give  .

        The other root comes from:

         

        de Moivre's theorem can be used to find the three cube roots of unity by

         

        where   can be  .

        Put  ,   and  

         

         

        This is the difference of two squares so

         

        Similarly any collection of  th roots of 1 can be obtained this way.

        Another example is to get the expressions for   and   without expanding  .

         

        Remember Pascal's Triangle


                                  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
        


         

         

        Separating the real and imaginary parts gives the two expressions. This is a lot easier than

         

        Use the same procedure to get

          and  .

        TrigonometryEdit

        1. == Trigonometry - the sin and cosine rules ==
         
        Trigonometric triangle

        In the following trigonometric identities,  ,   and   are the lengths of sides in a triangle, opposite the corresponding angles  ,   and  .

        • The sin rule  
        • The cosine rule  
        • The ratio identity  

        Trigonometric identitiesEdit

         

        Remember this is a consequence of Pythagoras' theorem where the length of the hypotenuse is 1.

         

         

        The difference of two angles can easily be generated by putting   and remembering   and  . Similarly, the double angle formulae are generated by induction.   is a little more complicated but can be generated if you can handle the fractions! The proofs are in many textbooks but as a chemist it is not necessary to know them, only the results.

        Identities and equationsEdit

        Identities and equations look very similar, two things connected by an equals sign. An identity however is a memory aid of a mathematical equivalence and can be proved. An equation represents new information about a situation and can be solved.

        For instance,

         

        is an identity. It cannot be solved for  ; it is valid for all  . However,

         

        is an equation where  .

        If you try and solve an identity as an equation you will go round and round in circles getting nowhere, but it would be possible to dress up   into a very complicated expression which you could mistake for an equation.

        Some observations on trianglesEdit

        Check you are familiar with your elementary geometry. Remember from your GCSE maths the properties of equilateral and iscoceles triangles. If you have an iscoceles triangle you can always dispense with the sin and cosine rules, drop a perpendicular down to the base and use trig directly. Remember that the bisector of a side or an angle to a vertex cuts the triangle in two by area, angle and length. This can be demonstrated by drawing an obtuse triangle and seeing that the areas are  .

        The interior angles of a polygonEdit

        Remember that the interior angles of a  -sided polygon are  (in degrees) or ( ) (in radians)

        For benzene there are six equilateral triangles if the centre of the ring is used as a vertex, each of which having an interior angle of 120 degrees. Work out the angles in azulene, (a hydrocarbon with a five and a seven membered ring), assuming all the C-C bond lengths are equal, (this is only approximately true).

        VectorsEdit

        1. ==Free Web Based Material from HEFCE==

        There is a DVD on vectors at Math Tutor.

        VectorsEdit

        Imagine you make a rail journey from Doncaster to Bristol, from where you travel up the West of the country to Manchester. Here you stay a day, travelling the next morning to Glasgow, then across to Edinburgh. At the end of a day's work you return to Doncaster. Mathematically this journey could be represented by vectors, (in 2 dimensions because we are flat earthers on this scale). At the end of the 2nd journey (D-B) + (B-M) you are 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). Completing the journey leaves you at a zero vector away from Doncaster, i.e. all the vectors in this closed path add to zero.

        Mathematically we usually use 3 dimensional vectors over the 3 Cartesian axes  ,   and  .

        It is best always to use the conventional right handed axes even though the other way round is equally valid if used consistently. The wrong handed coordinates can occasionally be found erroneously in published research papers and text books. The memory trick is to think of a sheet of graph paper,   is across as usual and   up the paper. Positive   then comes out of the paper.

        A unit vector is a vector normalised, i.e. multiplied by a constant so that its value is 1. We have the unit vectors in the 3 dimensions:

         

        so that

         

        The hat on the i, j, k signifies that it is a unit vector. This is usually omitted.

        Our geographical analogy allows us to see the meaning of vector addition and subtraction. Vector products are less obvious and there are two definitions the scalar product and the vector product. These are different kinds of mathematical animal and have very different applications. A scalar product is an area and is therefore an ordinary number, a scalar. This has many useful trigonometrical features.

        The vector product seems at first to be defined rather strangely but this definition maps onto nature as a very elegant way of describing angular momentum. The structure of Maxwell's Equations is such that this definition simplifies all kinds of mathematical descriptions of atomic / molecular structure and electricity and magnetism.

        A summary of vectorsEdit

        The unit vectors in the 3 Cartesian dimensions:

         

        a vector   is:

         

        The hat on the i, j, k signifies that it is a unit vector.

        Vector magnitudeEdit

         

        A constant times a vectorEdit

         

        Vector additionEdit

         

        Notice  

        Vector subtractionEdit

         

        Notice  

        Scalar ProductEdit

         

        Notice  

        Notice that if   this reduces to a square.

        If A and B have no common non-zero components in  ,   and   the value is zero corresponding to orthogonality, i.e. they are at right angles. (This can also occur by sign combinations making   zero. corresponding to non axis-hugging right angles.)

        Vector productEdit

         

        Notice  

        The minus sign on the middle term comes from the definition of the determinant, explained in the lecture. Determinants are defined that way so they correspond to right handed rotation. (If you remember our picture of   going round the circle, as one coordinate goes up, i.e. more positive, another must go down. Therefore rotation formulae must have both negative and positive terms.) Determinants are related to rotations and the solution of simultaneous equations. The solution of   simultaneous equations can be recast in graphical form as a rotation to a unit vector in  -dimensional space so therefore the same mathematical structures apply to both space and simultaneous equations.

        Matrices and determinantsEdit

        1. ==Simultaneous linear equations==

        If we have 2 equations of the form   we may have a set of simultaneous equations. Suppose two rounds of drinks are bought in a cafe, one round is 4 halves of orange juice and 4 packets of crisps. This comes to 4 pounds 20. The thirstier drinkers at another table buy 4 pints of orange juice and only 1 packet of crisps and this comes to 6 pounds 30. So we have:

         

        and

         

        i.e.  

        If you plot these equations they will be simultaneously true at   and  .

        Notice that if the two rounds of drinks are 2 pints and 2 packets of crisps and 3 pints and 3 packets of crisps we cannot solve for the prices! This corresponds to two parallel straight lines which never intersect.

        If we have the equations:

         

         

        If these are simultaneously true we can find a unique solution for both   and  .

        By subtracting the 2 equations a new equation is created where   has disappeared and the system is solved.

         

        Substituting back   gives us  .

        This was especially easy because   had the same coefficient in both equations. We can always multiply one equation throughout by a constant to make the coefficients the same.

        If the equations were:

         

        and

         

        things would go horribly wrong when you tried to solve them because they are two copies of the same equation and therefore not simultaneous. We will come to this later, but in the meantime notice that 3 times 8 = 4 times 6. If our equations were:

         

         

         

        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 determinants.

        Simultaneous equations have applications throughout the physical sciences and range in size from (2x2)s to sets of equations over 1 million by 1 million.

        Practice simultaneous equationsEdit

        Solve:

         

         

        and

         

         

        Notice that you can solve:

         

         

         

        because it breaks down into a (2x2) and is not truly a (3x3). (In the case of the benzene molecular orbitals, which are (6x6), this same scheme applies. It becomes two direct solutions and two (2x2) problems which can be solved as above.)

        MatricesEdit

        The multiplication of matrices has been explained in the lecture.

         

         

         

        but   cannot exist. To be multiplied two matrices must have the 1st matrix with the same number of elements in a row as the 2nd matrix has elements in its columns.

         

        where the  s are the elements of  .

         

        Look at our picture of   and   as represented by a unit vector in a circle. The rotation of the unit vector   about the  -axis can be represented by the following mathematical construct.

         

         

        In two dimensions we will rotate the vector at 45 degrees between   and  :

         

        This is if we rotate by +45 degrees. For     and  . So the rotation flips over to give  . The minus sign is necessary for the correct mathematics of rotation and is in the lower left element to give a right handed sense to the rotational sign convention.

        As discussed earlier the solving of simultaneous equations is equivalent in some deeper sense to rotation in  -dimensions.

        Matrix multiply practiceEdit

        i) Multiply the following (2x2) matrices.

            =  

        ii) Multiply the following (3x3) matrices.

           

        You will notice that this gives a unit matrix as its product.

         

        The first matrix is the inverse of the 2nd. Computers use the inverse of a matrix to solve simultaneous equations.


        If we have  


        In matrix form this is....

         

         

        In terms of work this is equivalent to the elimination method you have already employed for small equations but can be performed by computers for   simultaneous equations.

        (Examples of large systems of equations are the fitting of reference data to 200 references molecules, dimension 200, or the calculation of the quantum mechanical gradient of the energy where there is an equation for every way of exciting 1 electron from an occupied orbital to an excited, (called virtual, orbital, (typically   equations.)

        Finding the inverseEdit

        How do you find the inverse... You use Maple or Matlab on your PC but if the matrix is small you can use the formula...

         

        Here Adj A is the adjoint matrix, the transposed matrix of cofactors. These strange objects are best described by example.....

         

        This determinant is equal to: 1 ( 1 x 1 - 1 x (-1)) - (-1) ( 2 x 1 - 1 x 3) + 2 ( 2 x (-1) - ( 1 x 3) each of these terms is called a cofactor.

         


         

        This   thing gives the sign alternation in a form mathematicians like even though it is incomprehensible.

        Use the determinant

         

        to solve the simultaneous equations on page 47 by the matrix inverse method. The matrix corresponding to the equations on p47.2 is:

        
             1      -1      2                 6
        
             2       1      1          =      3
        
             3      -1      1                 6
        
        
        The cofactors are
        
             2       1     -5
        
            -1      -5     -2
        
            -3       3      3
        
        
        You may find these 9 copies of the matrix useful for
        striking out rows and columns to form this inverse....
        
            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
        
        
        
            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
        
        
        
            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 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     -3         6          1
        
          -1/9  X      1      -5      3    X    3     =   -1
        
                      -5      -2      3         6          2
        
        

        This takes a long time to get all the signs right. Elimination by subtracting equations is MUCH easier. However as the computer cannot make sign mistakes it is not a problem when done by computer program.

        The following determinant corresponds to an equation which is repeated three times giving an unsolvable set of simultaneous equations.

         

        Matrix multiplication is not necessarily commutative, which in English means   does not equal   all the time. Multiplication may not even be possible in the case of rectangular rather than square matrices.

        I will put a list of the properties and definitions of matrices in an appendix for reference through the later years of the course.

        Determinants and the Eigenvalue problemEdit

        In 2nd year quantum chemistry you will come across this object:  

        You divide by   and set   to equal   to get:

         

        Expand this out and factorise it into two quadratic equations to give:

         

        which can be solved using  

        Simultaneous equations as linear algebraEdit

        The above determinant is a special case of simultaneous equations which occurs all the time in chemistry, physics and engineering which looks like this:

         

        This equation in matrix form is   and the solution is  .

        This is a polynomial equation like the quartic above. As you know polynomial equations have as many solutions as the highest power of   i.e. in this case  . These solutions can be degenerate for example the   orbitals in benzene are a degenerate pair because of the factorisation of the   polynomial from the 6 Carbon-pz orbitals. In the 2nd year you may do a lab exercise where you make the benzene determinant and see that the polynomial is

         

        from which the 6 solutions and the orbital picture are immediately obvious.

        The use of matrix equations to solve arbitrarily large problems leads to a field of mathematics called linear algebra.

        Matrices with complex numbers in themEdit

        Work out the quadratic equation from the 3 determinants

         

        They are all the same! This exemplifies 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 different ways as well as being the only neat way to formulate some problems.

        DifferentiationEdit

        1. ==Free web-based material from HEFCE==

        There is a DVD on differentiation at Math Tutor.

        The basic polynomialEdit

        The most basic kind of differentiation is:

         

         

        There are two simple rules:

        1. The derivative of a function times a constant is just the same constant times the derivative.
        2. The derivative of a sum of functions is just the sum of the two derivatives.

        To get higher derivatives such as the second derivative keep applying the same rules.

        One of the big uses of differentiation is to find the stationary points of functions, the maxima and minima. If the function is smooth, (unlike a saw-tooth), these are easily located by solving equations where the first derivative is zero.

        The chain ruleEdit

         

        This is best illustrated by example: find   given  

        Let   and  .

        Now   and  

        So using the chain rule we have  


        Differentiating a productEdit

         

        Notice when differentiating a product one generates two terms. (Terms are mathematical expression connected by 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 orangopears. Integration by parts also generates an extra term each time it is applied.

        Differentiating a quotientEdit

        You use this to differentiate  .

         

        ProblemsEdit

        Differentiate with respect to  

         

        Notice we have  .  

         

         

         

         

        Evaluate the inner brackets first.

        Evaluate

         

         

         

         

         

        a, b and c are constants. Differentiate with respect to  .

         

         

        AnswersEdit

         

         

         

         

         

         

         

         

         

         

         

         

         

        Harder differentiation problemsEdit

        Differentiate with respect to  :

         

         

        Differentiate with respect to  

         

        Differentiate with respect to  

         

         

        Evaluate

         

         

        Using differentiation to check turning pointsEdit

          is the tangent or gradient. At a minimum   is zero. This is also true at a maximum or an inflection point. The second gradient gives us the nature of the point. If   is positive the turning point is a minimum and if it is negative a maximum. Most of the time we are interested in minima except in transition state theory.

        If the equation of   is plotted, is is possible to see that at   there is a third kind of point, an inflection point, where both   and   are zero.

         

        Plot   between -4 and +3, in units of 1. (It will speed things up if you factorise it first. Then you will see there are 3 places where   so you only need calculate 5 points.) By factorising you can see that this equation has 3 roots. Find the 2 turning points. (Differentiate once and find the roots of the quadratic equation using  . This gives the position of the 2 turning points either side of zero. As the equation is only in   it has 3 roots and 2 maxima / minima at the most therefore we have solved everything. Differentiate your quadratic again to get  . Notice that the turning point to the left of zero is a maximum i.e.   and the other is a minimum i.e.  .

        What is the solution and the turning point of  .

        Solve  , by factorisation.

        (The 3 roots are -3,0 and +2.  

        Solutions are   and  , i.e. -1.7863 and 1.1196.

         

        There are 3 coincident solutions at  ,  , at 0 so this is an inflection point.

        The roots are 0, 1 and -1.

        IntegrationEdit

        1. ==Free Web Based Material from HEFCE==

        There is a DVD on integration at Math Tutor.

        The basic polynomialEdit

         

         

        This works fine for all powers except -1, for instance the integral of

         

        is just

         

        -1 is clearly going to be a special case because it involves an infinity when   and goes to a steep spike as   gets small. As you have learned earlier this integral is the natural logarithm of   and the infinity exists because the log of zero is minus infinity and that of negative numbers is undefined.

        The integration and differentiation of positive and negative powersEdit

           >>>>>>>>>>>>>>>>>>>>>>Differentiation
        
           1/3 x*x*x  x*x       2x        2         0        0       0       0
        
           1/3 x*x*x  x*x       2x        2         ?        ?       ?       ?
        
           Integration<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
        
           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 involving  .

        You will be able to work them out easily when you have done more integration. The thing to notice is that the calculus of negative and positive powers is not symmetrical, essentially caused by the pole or singularity for   at  .

        LogarithmsEdit

        Logarithms were invented by Napier, a Scottish Laird, in the 17th-century. He made many inventions but his most enduring came from the necessity of doing the many long divisions and trigonometric calculations used in astronomy. The Royal Navy in later years devoted great time and expense to developing logarithm technology for the purposes of navigation, leading to the commissioning of the first mechanical stored program computer, which was theoretically sound but could not be made by Charles Babbage between 1833 and 1871.

        The heart of the system is the observation of the properties of powers:

         

        This means that if we have the inverse function of   we can change a long division into a subtraction by looking up the exponents in a set of tables. Before the advent of calculators this was the way many calculations were done.

         

         

        Napier initially used logs to the base   for his calculations, but after a year or so he was visited by Briggs who suggested it would be more practical to use base 10. However base   is necessary for the purposes of calculus and thermodynamics.

        Integrating 1/xEdit

         

        This is true because   is our function which reduces or grows at the rate of its own quantity.

         

        This is our definition of a logarithm.

         

        therefore

         

         

        therefore

         

        Integrating 1/x like thingsEdit

        Just as

         

        so

         

        therefore by the chain rule

         

        therefore

         

        Examples of this are:

         

        or

         

        and

         

        As the integral of   is   so the differential of   is   so

         

          is just a constant so   so

         

        This can also be done by the chain rule

         

        What is interesting here is that the 5 has disappeared completely. The gradient of the log function is unaffected by a multiplier! This is a fundamental property of logs.

        Some observations on infinityEdit

        Obviously   is  .

         nd   are undefined but sometimes a large number over a large number can have defined values. An example is the   of 90 degrees, which you will remember has a large opposite over a large hypotenuse but in the limit of an infinitesimally thin triangle they become equal. Therefore the   is 1.

        Definite integrals (limits)Edit

        Remember how we do a definite integral

         

        where   is the indefinite integral of  .

        Here is an example where limits are used to calculate the 3 areas cut out by a quartic equation:

         .

        We see that   is a solution so we can do a polynomial division:


                x3  -x2 -2x
              -------------------
         x-1  ) x4 -2x3 -x2 +2x
                x4  -x3
               --------
                    -x3 -x2
                    -x3 +x2
                    -------
                        -2x2 +2x
                        -2x2 +2x
                        --------
                             0
        

        So the equation is   which factorises to

         .

         

        Integration by substitutionEdit

         

        where  .

         

         

        Simple integration of trigonometric and exponential FunctionsEdit

         

         

         

         

         

        AnswersEdit

         

         

         

         

        i.e. the integral of  .

         

        Integration by partsEdit

        This is done in many textbooks and Wikipedia. Their notation might be different to the one used here, which hopefully is the most clear. You derive the expression by taking the product rule and integrating it. You then make one of the   into a product itself to produce the expression.

         

        (all integration with respect to  . Remember by

         

        (  gets differentiated.)

        The important thing is that you have to integrate one expression of the product and differentiate the other. In the basic scheme you integrate the most complicated expression and differentiate the simplest. Each time you do this you generate a new term but the function being differentiated goes to zero and the integral is solved. The expression which goes to zero is  .

        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 can rearrange the equation and the integral is solved. This is obviously very useful for trig functions where   ad infinitum.

          also generates itself and is susceptible to the same treatment.

         

         

         

        We now have our required integral on both sides of the equation so

         

        Integration ProblemsEdit

        Integrate the following by parts with respect to .

         

         

         

         

         

         

         

         

        Actually this one can be done quite elegantly by parts, to give a two term expression. Work this one out. Expanding the original integrand by Pascal's Triangle gives:

                 2       3       4       5       6      7    8
          x + 7 x  + 21 x  + 35 x  + 35 x  + 21 x  + 7 x  + x
        


        The two term integral expands to

              2        3         4      5         6      7        8        9
         1/2 x  + 7/3 x  + 21/4 x  + 7 x  + 35/6 x  + 3 x  + 7/8 x  + 1/9 x  - 1/72
        

        So one can see it is correct on a term by term basis.

         

         

        If you integrate   you will have to apply parts 7 times to get   to become 1 thereby generating 8 terms:

        
        
            7             6              5               4               3
          -x  cos(x) + 7 x  sin(x) + 42 x  cos(x) - 210 x  sin(x) - 840 x  cos(x) +
        
                2
          2520 x  sin(x) +  5040 x cos(x) - 5040 sin(x)  + c
        
        
        (Output from Maple.) 
        

        Though it looks nasty there is quite a pattern to this, 7, 7x6,7x6x5 ------7! and sin, cos, -sin, -cos, sin, cos etc so it can easily be done by hand.

        Differential equationsEdit

        First order differential equations are covered in many textbooks. They are solved by integration. (First order equations have  , second order equations have   and  .)

        The arbitrary constant means another piece of information is needed for complete solution, as with the Newton's Law of Cooling and Half Life examples.

        Provided all the  s can be got to one side and the  s to another the equation is separable.

         

         

         

        This is the general solution.

        Typical examples are:

         

         

        by definition of logs.

          (1)

         

          (2)

          (3)

        This corresponds to:

         

        The Schrödinger equation is a 2nd order differential equation e.g. for the particle in a box

         

        It has taken many decades of work to produce computationally efficient solutions of this equation for polyatomic molecules. Essentially one expands in coefficients of the atomic orbitals. Then integrates to make a differential equation a set of numbers, integrals, in a matrix. Matrix algebra then finishes the job and finds a solution by solving the resultant simultaneous equations.

        The calculus of trigonometric functionsEdit

        There are many different ways of expressing the same thing in trig functions and very often successful integration depends on recognising a trig identity.

         

        but could also be

         

        (each with an integration constant!).

        When applying calculus to these functions it is necessary to spot which is the simplest form for the current manipulation. For integration it often contains a product of a function with its derivative like   where integration by substitution is possible.

        Where a derivative can be spotted on the numerator and its integral below we will get a   function. This is how we integrate  .

         

        We can see this function goes to infinity at   as it should do.

        Integration by rearrangementEdit

        Take for example:

         

        Here there is no   function producted in with the   powers so we cannot use substitution. However there are the two trig identities

         

        and

         

        Using these we have

         

        so we have two simple terms which we can integrate.

        The Maclaurin seriesEdit

        We begin by making the assumption that a function can be approximated by an infinite power series in  :

         

        By differentiating and setting   one gets

         

        Sin, cos and   can be expressed by this series approximation

         

         

         

        Notice   also works for negative  .

        When differentiated or integrated   generates itself!

        When differentiated   generates  .

        By using series we can convert a complex function into a polynomial, and can use   for small  .

        In actual fact the kind of approximation used inside computer programs is more like:  

        These have greater range but are much harder to develop and a bit fiddly on the calculator or to estimate by raw brain power.

        We cannot expand   this way because   is

         . However   can be expanded.

        Work out the series for  .

        FactorialsEdit

        The factorials you have seen in series come from repeated differentiation.   also has a statistical meaning as it is the number of unique ways you can arrange   objects.

          is 1 by definition, i.e. the number of different ways you can arrange 0 objects is 1.

        In statistical thermodynamics you will come across many factorials in expressions such as:  

        Factorials rapidly get unreasonably large: 6! = 720, 8! = 40320 but 12! = 479001600 so we need to divide them out into reasonable numbers if possible, so for example  .

        Stirling's approximationEdit

        Also in statistical thermodynamics you will find Stirling's approximation:

         

        This is proved and discussed in Atkins' Physical Chemistry.

        How can you use series to estimate  . Notice that the series for

          converges extremely slowly.   is much faster because the

          denominator becomes large quickly.

        Trigonometric power seriesEdit

        Remember that when you use   and  

        that x must be in radians.....

        Calculus revisionEdit

        ProblemsEdit

        integrate x to the power of x with respect to x

        1. Differentiate  , with respect to  . (Hint - use the chain rule.)
        2. Differentiate  . (Chain rule and product rule here.)
        3. Differentiate  . (Hint - split it into a sum of logs first.)
        4. Integrate  . (Hint - use integration by parts and take the expression to be differentiated as 1.)

        AnswersEdit

        1. It is just  . Bring a   out of each term to simplify to  .
        2.  .
        3.   - therefore it is 4 times the derivative of  .
        4. You should get   by 1 application of parts.

        Some useful aspects of calculusEdit

        1. ==Limits==

        Many textbooks go through the proper theory of differentiation and integration by using limits. As chemists it is possible to live without knowing this so we might well not have it as an examinable topic. However here is how we differentiate sin from 1st principles.

         

         

         

         

        As   for small   this expression is  .

        Similarly for