Euler's Number, e^x and ln x
The number is the one base for which the exponential function is its own gradient: the curve climbs at exactly the rate . That single property makes and its inverse, the natural logarithm , the functions that calculus, growth and decay are built on. The syllabus asks you to understand the definition and properties of and , including their relationship as inverse functions and their graphs; in the exam they appear in almost every question, from solving equations to differentiation, integration and differential equations.
Where e comes from
Gradients of exponential curves
Every exponential curve (with ) has a gradient that is proportional to its height. To see why, look at the gradient of the chord from to :
The factor comes straight out. As , the second factor tends to a fixed number that depends only on . Call it . Then
Evaluating for a tiny (say ) gives the constant:
| Base | |||
|---|---|---|---|
For the gradient is a bit less than the height; for it is a bit more. Somewhere between and there is a base for which exactly, so the gradient equals the height at every point. That base is .
Euler's number is the base for which the gradient of at every point equals the value of at that point:
The function (also written ) is called the exponential function.
A second route: compound growth
The same number appears when growth is compounded more and more often. Put $1 in an account paying interest a year. Compounded once, you have at the end; compounded times, you have .
The values settle on . You do not need this limit for the exam, but it explains why turns up in every model of continuous growth.
You need to know that is a constant (about ), the graphs and properties of and , and that they are inverse functions. The derivation above is for understanding; the derivative result is used constantly from differentiating ln x and e^x onwards.
The natural logarithm
The natural logarithm of , written , is the logarithm to base :
Because is a logarithm, every law from logarithms and the laws of logarithms holds for it unchanged:
Every exponential can be written with base :
The last identity follows from , so . It shows that , and are all of the form with , and it explains the table above: the constant is exactly (indeed and ).
Simplifying with e and ln
The skill is to bring every power of and every into a form where they can cancel. Move coefficients inside the log first, then cancel:
is not . Use the index law first: . Likewise .
Graphs and properties
The graphs of and are reflections of each other in the line , because the two functions are inverses: swapping and in gives , which is .
The curve through is ; the curve through is ; the straight line is the mirror line .
| Domain | all real | |
| Range | all real | |
| Key point | ||
| Asymptote | (as ) | (as ) |
| Gradient at key point | ||
| Behaviour | increasing, grows faster than any power of | increasing, grows more slowly than any power of |
Points to notice:
- for every . An equation such as or has no solution.
- is negative for , zero at and positive for .
- is undefined for . This is why every log equation needs its domain checked.
- Both functions are one-one (each output comes from exactly one input), so each has an inverse, and taking or of both sides of an equation is always reversible.
- Both functions are increasing, so they preserve inequalities: , and for positive , .
The sketching of transformed versions such as and is covered in exponential graphs.
Solving equations with e and ln
Two moves do almost all the work.
- To free from an exponent, isolate the exponential, then take of both sides: (only possible if ).
- To free from a logarithm, combine into a single , then raise to both sides: .
- Rearrange so that a single exponential term stands alone on one side.
- Check the other side is positive; if not, there is no solution.
- Take natural logs: .
- Solve for and give the answer in exact form unless a decimal is asked for.
- If there are two exponential terms, such as and , or and , look for a quadratic in instead (see the examples).
- Note the domain: every argument of must be positive.
- Combine the logs into a single on each side, writing numbers as logs if needed (, ).
- Remove the logs: , or .
- Solve, then reject any solution outside the domain, giving the reason.
Inverse functions involving e and ln
Because and undo each other, a function built from one has an inverse built from the other. Find it exactly as in P1 (one-one and inverse functions): write , make the subject, then swap the letters. The domain of is the range of , so always find the range of first.
Exponential growth and decay
A quantity that changes at a rate proportional to its current size follows
where is the value when . If it is growth; if (often written with ) it is decay. You will derive this from a differential equation in differential equations; here the skill is to use the model.
- Two data values fix the two constants. Divide one equation by the other to eliminate , then take logs to find .
- Halving or doubling time does not depend on the starting value: gives .
- Long-term behaviour: as , for . A model like therefore tends to .
Worked examples
Simplify, giving exact answers: (a) ; (b) ; (c) ; (d) .
Solution
(a) .
(b) .
(c) .
(d) .
Solve the equation , giving your answer correct to 3 significant figures.
Solution
Isolate the exponential: .
Take natural logs: .
So (3 s.f.).
(The exact answer is equivalent, since .)
Solve , giving your answers in exact form.
Solution
Multiply every term by (which is never zero), using :
Let : , so and or .
; .
Solve , giving in exact form and correct to 3 significant figures.
Solution
Domain: (which also makes ).
Collect the terms: , so
Since , this is positive and so valid.
The function is defined by for .
(a) State the range of .
(b) Find an expression for and state the domain of .
Solution
(a) takes every positive value, so and . The range is .
(b) Let . Then
So , with domain (the range of ). The domain is also exactly where the argument of is positive, which is a good check.
The mass, grams, of a radioactive substance days after a measurement began is modelled by , where and are positive constants. When , , and when , .
(a) Show that and find , correct to 3 significant figures.
(b) Find the time at which the mass has fallen to grams.
(c) Find the half-life of the substance.
Solution
(a) The data give and . Divide the first by the second to eliminate :
Numerically . Then (3 s.f.).
(b) .
Using unrounded values, days (3 s.f.).
(c) Half-life : days (3 s.f.).
- Taking logs before isolating. From , writing is wrong. Divide by first, or use .
- Splitting the log of a sum. . If two exponential terms are added, look for a quadratic in .
- Accepting impossible values. In a quadratic in , a root gives no , because . Say so: " has no solution".
- Writing of a negative number. is not an answer. If you reach it, either the equation has no solution or there is an earlier slip.
- Forgetting the domain of an inverse. The domain of is the range of , not "all real numbers".
- Rounding too early. In a model, keep unrounded (store it on the calculator) for later parts; using can change the third significant figure of a later answer.
- "Exact" means leave and in the answer: , . A decimal alone loses the accuracy mark.
- "Show that " means the printed form must appear as the last line, reached by visible steps. Dividing the two equations to eliminate the other constant is the expected method.
- When you reject a root, give the reason in words ("", " is undefined for "). Examiners repeatedly report solutions that list both roots and leave them.
- Inverse-function questions usually carry a mark for the domain of : find the range of before you start the algebra.
- is the base for which : the gradient equals the height.
- ; ; all log laws apply to .
- for and for all ; , .
- : domain , range , through . : domain , range , through . Reflections in .
- : every exponential is an .
- Solve by isolating the exponential and taking , or combining logs and raising ; spot quadratics in .
- is never zero or negative; needs a positive argument.
- Growth and decay: ; divide two data equations to find ; half-life .
Practice
- Simplify exactly: (a) ; (b) ; (c) ; (d) .
- Solve , giving the exact answer.
- Solve .
- Solve , giving your answer in exact form.
- The function is defined by for . Find and state its range.
- Solve , giving exactly and to 3 significant figures.
- The temperature, , of a drink minutes after it is poured is . State the initial temperature and the long-term temperature, and find the time taken for the temperature to fall to .
- Solve , explaining why only one solution exists.
- A population of insects is modelled by , where is in years. The population doubles every years. (a) Find the exact value of . (b) A second population is modelled by . Find the value of at which the two populations are equal, giving your answer correct to 3 significant figures.
Answers
-
(a) . (b) . (c) . (d) .
-
, so .
-
Domain . . Reject (outside the domain), so .
-
Let : . is impossible since . , so .
-
. So for . Its range is the domain of : .
-
Domain . , so , giving and (3 s.f.). It is positive, so valid.
-
At : . As , , so . For : minutes (3 s.f.).
-
Multiply by : . has no solution because for all . So and is the only solution.
-
(a) . (b) years (3 s.f.).