Syllabus coverage
- Understand the relationship between logarithms and indices and use the laws of logarithms, excluding change of base
- Understand the definition and properties of and , including their relationship as inverse functions and their graphs, and know the graph of for positive and negative values of
- Use logarithms to solve equations and inequalities in which the unknown appears in an index
- Use logarithms to transform a relationship to linear form, and determine unknown constants from the gradient and the intercept of the resulting straight line
Part of the AS & A Level Maths syllabus 9709.
A logarithm is an index written the other way round
The definition is the whole topic: means exactly the same thing as . The number is the base, and it must be positive and not equal to 1; must be positive, because no power of a positive base is negative or zero. So because , and because .
Two bases are used throughout Paper 3. Base 10 gives the common logarithm, written with no base shown, and it is the one on a calculator button. Base gives the natural logarithm, written , and it is the one the rest of the course is built on. Nothing else needs a base of its own.
Four consequences of the definition are used constantly and are worth knowing cold: because ; ; because the two operations cancel; and . The last two are the same statement seen from each side, and they are what make logarithms and exponentials inverse functions.
Logarithms turn multiplication into addition, which is why they were invented and why they still solve equations that nothing else can touch. A question that says 'solve', where the unknown is in the air, is almost always a question about taking logarithms of both sides.
- - learn this as one two-way statement
- and for every valid base
- , so without any calculation
- You cannot take the logarithm of 0 or of a negative number
- , not : the answer is the index, not the value of the fraction
The laws of logarithms
Three laws cover everything Paper 3 asks. The product law: . The quotient law: . The power law: . All three come from the index laws, and all three need the same base on every logarithm before they can be used.
They are used in both directions. Combining several logarithms into one is what lets you solve an equation: becomes , and then the definition turns it into , so , giving or . The solution must be rejected, because does not exist - a logarithm question that produces two roots almost always wants one of them thrown away with a reason.
Splitting a single logarithm into several is what the later topics need. The power law is how a logarithm of a power is written, and it is the step that makes an equation with the unknown in the index solvable at all. You will use it again in differentiation and integration, where and are the functions that behave most simply.
Change of base is deliberately excluded at 9709, so you will never need to convert a logarithm from one base to another by formula. If an equation involves base 10 and base together, take logarithms in whichever base you like on both sides and carry the base through to the end.
| Law | Statement | Example |
|---|---|---|
| Product | ||
| Quotient | ||
| Power | ||
| Special values | , | , |
| Cancelling | , | , |
The exponential function e^x and the natural logarithm ln x
is a number, approximately 2.718281828, chosen because is the one exponential curve whose gradient at every point equals its height. is called the exponential function, and means , the natural logarithm.
Because they share a base, and undo each other: for every real , and for every . That second restriction matters. is meaningless, because does not exist, so a solution that requires it must be rejected.
The two functions are therefore inverse functions, and this is the relationship the syllabus asks you to understand rather than merely quote. It is what lets you solve : exponentiate both sides to get , so to 3 significant figures. In the other direction, gives , so .
Two facts about the ranges are worth memorising because they decide how many solutions an equation has. is positive for every , so has no solutions. And is defined only for , so does have a solution, namely .
| Statement | Equivalent statement | Restriction |
|---|---|---|
| none | ||
| none | ||
| in the second form only | ||
| always | is undefined for | the domain of is |
The graphs of y = e^x, y = ln x and y = e^kx
The graph of passes through , climbs steeply to the right, and approaches the x-axis as becomes very negative without ever touching it. Its range is and it is increasing everywhere. The graph of passes through , climbs more and more slowly, and approaches the y-axis as approaches zero from above. Its domain is and its range is every real number.
Because the functions are inverses, one graph is the reflection of the other in the line . The point on reflects to on , and the horizontal asymptote of one becomes the vertical asymptote of the other. If a question asks you to sketch both, the line is the mirror, and drawing it makes the sketch easy to justify.
The syllabus asks specifically for for both positive and negative . For the curve grows as increases; for it decays towards the x-axis. Every one of them passes through , because whatever is, and the bigger is, the faster the curve rises or falls. The case is worth knowing separately: is the reflection of in the y-axis.
One algebraic fact ties the family together: , so a curve of the form is the same kind of thing as with . That is why every exponential model in the paper, whether written with a base of 2 or a base of , has the same shape and the same two possibilities - growth or decay.
| Curve | Point it always passes through | Asymptote | Behaviour |
|---|---|---|---|
| the x-axis, as | increasing, range | ||
| the x-axis, as | decreasing, range | ||
| , | the x-axis on the left | growth; steeper for larger | |
| , | the x-axis on the right | decay; faster for more negative | |
| the y-axis, as | increasing, domain |
Solving equations in which the unknown is in the index
When the unknown is in the air, take logarithms of both sides. That step is valid because is an increasing function, so it preserves equality, and it works for any base - and are both acceptable, and the exact answer may be written with either. For , taking gives , so and to 3 decimal places.
When the unknown appears in two indices, use the power law to bring both down, then collect the terms. For : taking logarithms gives ; expanding gives ; collecting gives , so to 3 significant figures. The whole method is one line of expansion and one line of collecting.
Inequalities work the same way with one extra rule: dividing by a negative number reverses the inequality, and a logarithm can be negative when its base is less than 1 or its argument is less than 1. In , taking logarithms gives , so , and multiplying by reverses the sign to give . Since the division does not reverse anything, and to 3 significant figures.
Two smaller habits save marks. State which logarithm you are taking, because that is a method mark in its own right. And keep the exact fraction until the final line: rounded to 1.32 and then reused in a later part can cost the accuracy mark, while the exact form never can.
- Write 'take of both sides' before you use the power law
- becomes in one step
- With the unknown in two indices, expand both sides and collect the terms before dividing
- Dividing by a negative logarithm reverses an inequality - check the sign of the log before you divide
- gives exactly, because and
Reducing a relationship to linear form
Experimental data often obey a power law or an exponential law , and neither can be checked by plotting against , because both produce curves. Taking logarithms turns each of them into the equation of a straight line, and the constants can then be read off the graph or worked out from two pairs of values.
For , take logarithms of both sides and use the product and power laws: . Plotted with up the vertical axis and along the horizontal axis, this is a straight line of gradient and vertical intercept . The gradient gives the power directly, and the intercept has to be exponentiated to give .
For , the same step gives , which is a straight line when is plotted against itself. Here the gradient is and the vertical intercept is again . A relationship of the form is the same case with : the gradient of the straight line is exactly , and the intercept is .
Given two pairs of values, the two constants come out of two simultaneous equations. If with when , and when , then and . Subtracting removes : , so . Then , so .
| Relationship | Linear form | Plot | Gradient | Intercept |
|---|---|---|---|---|
| against | , so | |||
| against | , so | |||
| against | ||||
| against |
How a Paper 3 logarithms question is built
There are four shapes and each has a routine. Evaluating a logarithm is a recall question - write the index. Solving an equation with the unknown in the index is 'take logarithms, apply the power law, collect'. Solving an equation with logarithms on both sides is 'combine with the laws, use the definition, then reject any invalid root'. Transforming a relationship to linear form is 'take logarithms, state what is plotted, name the gradient and the intercept, then convert the intercept back'.
The marks sit in the statements rather than in the arithmetic. Writing 'take of both sides', naming the gradient as , and saying why a root is rejected are each worth a mark that a calculator cannot supply. A bare decimal with no line of working rarely scores more than the final accuracy mark.
- Rewrite using the definition before doing anything clever - most questions are shorter than they look
- Reject any root that makes an argument of a logarithm zero or negative, and say why in one clause
- Give the exact answer and the rounded one when the question asks for a stated accuracy
- In a linear-form question, state the axes you are plotting on: against and against are different graphs
- Check a solution by substituting it back into the original equation - index equations are easy to check numerically
Key terms you must be able to define
- Logarithm
- The index to which a base must be raised to give a number: means .
- Base (of a logarithm)
- The number being raised to a power, written as a subscript in . It must be positive and not equal to 1.
- Argument of a logarithm
- The number the logarithm is taken of. It must be positive: requires .
- Common logarithm
- A logarithm to base 10, written with the base left out. It is the one on the calculator button.
- Natural logarithm
- A logarithm to base , written . The one Paper 3 uses most.
- e
- The irrational number 2.718281828..., the base for which has gradient equal to its own height.
- Exponential function
- , always positive, increasing everywhere, with the x-axis as an asymptote.
- Inverse functions
- Two functions that undo each other. and are inverses, so and for .
- Asymptote
- A line a curve approaches but never reaches. has the x-axis as an asymptote; has the y-axis.
- Product law
- . It turns a product into a sum.
- Quotient law
- . It turns a quotient into a difference.
- Power law
- . It brings an index down to the front, which is how index equations are solved.
- Exponential equation
- An equation with the unknown in an index, such as . Solved by taking logarithms of both sides.
- Exponential growth and decay
- The behaviour of : growth when , decay towards the x-axis when .
- Linear form
- A relationship rewritten so that it plots as a straight line, such as .
- Logarithmic transformation
- Taking logarithms of a relationship to turn a power law or an exponential law into a straight line, so that the gradient and intercept give the constants.
Applied
At AS & A Level the marks sit in the setup: choosing the model, stating the variable and writing the equation before any working. These show the shape of a full-mark answer, then you do the algebra yourself.
asks for the power of 2 that gives 32. Since , the answer is 5. asks for the power of 3 that gives , and , so the answer is . No calculator is needed for either, and the whole question is worth two marks.
Solve . Take logarithms of both sides: , so to 3 decimal places. The exact form is the answer to the question; the decimal is what a stated accuracy asks for. Check roughly: , so an answer near 1.54 is sensible.
Solve . Take logarithms: , so to 3 significant figures. Notice that the 2 stays with the rather than disappearing into the fraction: the power law gives , and dividing by is the last step.
Data fit , with when and when . Taking logarithms gives . Subtracting the two equations: . Then , so . Check: , as given.
The mass grams of a sample after days is modelled by . Find the time at which the mass is 10 grams. Put , so . Take logarithms: , so days to 3 significant figures. The negative sign in the index is handled by the logarithm, not by ignoring it.
Flashcards
Click a card to flip it. Mark honestly — the tracker uses this.
Exam-style questions
8 questions, 28 marks. Write your answer, then reveal the mark scheme and tick each point you actually made. Answering honestly here is what makes the rest of the site useful.
Mark scheme — tick each point you actually made
- B1
- B1, because not checked yet
Mark scheme — tick each point you actually made
- M1
- A1
- A1not checked yet
Mark scheme — tick each point you actually made
- M1
- M1
- A1
- A1not checked yet
Mark scheme — tick each point you actually made
- M1
- M1
- A1
- A1not checked yet
Mark scheme — tick each point you actually made
- M1
- M1
- A1not checked yet
Mark scheme — tick each point you actually made
- M1
- M1
- M1
- A1
- A1, from not checked yet
Mark scheme — tick each point you actually made
- M1
- M1
- A1
- A1rejects because is not defined, so not checked yet
Mark scheme — tick each point you actually made
- B1
- B1
- B1
Exam technique for this topic
- Learn the definition as a two-way statement and use it in the first line of any logarithm question. 'Rewrite as ' is a method mark, and it turns most equations into ordinary algebra.
- When the unknown is in an index, write 'take of both sides' on its own line before applying the power law. The method mark is awarded for the logarithm, not for the eventual answer.
- Reject invalid roots explicitly. In the quadratic gives and , and the answer is only complete when you have said that is rejected because is undefined.
- In a linear-form question, state the axes and both constants in words: 'plotting against gives a straight line of gradient and intercept '. Then convert the intercept with - the graph gives you the logarithm of , never itself.
- Give both the exact form and the rounded value when an accuracy is specified. earns the exactness mark, and 3.06 earns the accuracy mark; writing only 3.06 leaves the first of those on the table.