Exp Log Applications
Exp Log Applications
Jackie Nicholas
and Christopher Thomas
1998
c University of Sydney
Acknowledgements
Parts of this booklet were previously published as Exponents and Logarithms, written by
Christopher Thomas for the Mathematics Learning Centre in 1991. The remainder is
new.
In his earlier booklet, Chris acknowledged the assistance of Sue Gordon, Trudy Weibel
and Duncan Turpie for their comments on the content and layout and for proof reading. I
would also like to thank Sue Gordon and Paul Clutton for their comments on this version
and for the laborious task of checking the exercises.
Jackie Nicholas
November 1998
Contents
1 Introduction 1
1.1 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
5 Solutions to Exercises 25
5.1 Solutions to Exercises from Section 1 . . . . . . . . . . . . . . . . . . . . . 25
5.2 Solutions to Exercises from Section 2 . . . . . . . . . . . . . . . . . . . . . 27
5.3 Solutions to Exercises from Section 3 . . . . . . . . . . . . . . . . . . . . . 33
5.4 Solutions to Exercises from Section 4 . . . . . . . . . . . . . . . . . . . . . 36
Mathematics Learning Centre, University of Sydney 1
1 Introduction
The exponential and logarithmic functions are important functions in science, engineering
and economics. They are particularly useful in modelling mathematically how populations
grow or decline. You might be surprised to learn that scales used to describe the magnitude
of seismic events (the Richter scale) or noise (decibels) are logarithmic scales of intensity.
In this booklet we will demonstrate how logarithmic functions can be used to linearise
certain functions, discuss the calculus of the exponential and logarithmic functions and
give some useful applications of them.
If you need a detailed discussion of index and log laws, then the Mathematics Learning
Centre booklet: Introduction to Exponents and Logarithms is the place to start. If you
are unsure of the level you need, then do this short quiz. The answers are in section 5.
1.1 Exercises
The following expressions evaluate to quite a ‘simple’ number. If you leave some of your
answers in fractional form you won’t need a calculator.
1 3
1. 9 2 2. 16 4 3. ( 15 )−2 4. (3−1 )2 5. ( 52 )−2
3 2 1 1 2
6. (−8) 2 7. (− 27
8
)3 8. 527 5−24 9. 8 2 2 2 10. (−125) 3
These look a little complicated but are equivalent to simpler ones. ‘Simplify’ them. Again,
you won’t need a calculator.
3n+2 16 1 1
11. n−2 12. 6
13. (a 2 + b 2 )2
3 1
x
x 2 +x
14. (x2 + y 2 ) 2 − x2 (x2 + y 2 )− 2
1 1 1 1 2 1 2
15. 1 16. (u 3 − v 3 )(u 3 + (uv) 3 + v 3 )
x2
17. Sketch the graphs of the functions f (x) = 3x and f (x) = 3−x . On the same diagrams
mark in roughly the graphs of f (x) = 2.9x and 2.9−x .
Using the rules of logarithms, rewrite the following expressions so that just one logarithm
appears in each.
27. 3 log2 x + log2 30 + log2 y − log2 w 28. 2 ln x − ln y + a ln w
29. 12(ln x + ln y) 30. log3 e × ln 81 + log3 5 × log5 w
If you did not get most of those questions correct (in particular, questions 17 to 29), then
you need to go back to the companion book, Introduction to Exponents and Logarithms.
Mathematics Learning Centre, University of Sydney 2
2.1 Introduction
y
y = h(x)
y = g(x)
4
y = f(x)
Do you think that it is possible, by looking at the graphs, to work out what the functions
are? Well, for one of the functions it is relatively easy. The function g(x) has as its graph
a straight line, so it must be of the form g(x) = mx + b where m is the slope of the line
and b is the y-intercept, the point on the y-axis where the line cuts the y-axis. Even if we
can’t work out exactly what m and b are, we still know the general form of the function.
In fact m is equal to 32 , and the y-intercept, b, is equal to 12 . Therefore the straight line
is the graph of the function g(x) = 32 x + 12 .
On the other hand, we can’t really see what the functions f (x) and h(x) are by simply
inspecting the graph. We can’t even really tell the general form of the functions. Actually
f (x) = e0.41x and h(x) = (x + 1)2 , but it would be difficult to guess this by looking at the
graphs.
The basis of this section is the observation that it is easy to recognise a straight line. If
we see a graph which is a straight line then we know that it is the graph of a function
of the form y = mx + b. If we see a graph that is curved then we know that it is not a
straight line but, without more information, we cannot usually say much about what the
form of the function is.
If we have a function of the form aekx (for example y = 3.7e2x ) or axb (for example
y = 3x5 ) then we can transform this function in a simple way to get a function of the
form f (x) = mx + b, the graph of which is a straight line. We can tell from the position
and slope of this straight line what the original function is.
Mathematics Learning Centre, University of Sydney 3
x 1 2 3 4 5
y 2 16 54 128 250
X = ln x 0 0.693 1.099 1.386 1.609
Y = ln y 0.693 2.773 3.989 4.852 5.521
Figure 2: Table of data for function y = 2x3 together with the values of Y = ln y and
X = ln x.
y 5
250
Y = ln 2 + 3X
4
200 y = 2x3
3
150
2
100
1
50
By plotting our data as ln y against ln x we have come up with a graph which is a straight
line, and therefore much easier to understand.
Mathematics Learning Centre, University of Sydney 4
What usually happens in practise is this: Some data is obtained, usually through exper-
iment or a sampling procedure. The person analysing the data thinks that they behave
according to a function of the form y = axb , but is not sure of this, or of what values the
constants a and b take. The researcher takes the data, calculates Y = ln y and X = ln x
and plots them. If these data lie roughly along a straight line then the researcher knows
that the data are behaving roughly according to the relationship Y = ln a + bX, and by
measuring the slope and intercept of a line of best fit is able to make an estimate of the
values of ln a and b. a is found using the equation a = eln a .
Example: The data in the table in Figure 4 are measurements made by a researcher
studying blood sugar levels. The researcher suspects that the quantities x and y should
It is clear that the data do fall (roughly) along a straight line. A ‘line of best fit’ has
been drawn through the data. You may disagree with where the line has been drawn; its
position is largely a subjective choice though in some contexts there are formulae which
may be applied to tell us precisely where the line should be drawn. Anyway, let us suppose
that we have agreed on the position of the line as in Figure 6. Since the data do fall along
a straight line, we know that they are related by the formulae y = axb .
Y
10.00
8.00
6.00
4.00
2.00
X
0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00
y = axb
ln y = ln(axb ) taking logs of both sides
= ln a + ln(xb )
= ln a + b ln x
So, Y = ln a + bX
To find the slope we take two points on the line and use them to calculate the slope. It is
very important to use two points actually on the line of best fit, not (X, Y ) points (unless
the line of best fit happens to pass through (X, Y ) points). Since, in this case, the points
(0.92, 5.23) and (1.76, 6.42) lie on the line, the slope of the line is therefore
6.42 − 5.23
b ≈
1.76 − 0.92
1.19
≈
0.84
≈ 1.4.
Therefore b ≈ 1.4. The best way to calculate the constant a is to substitute the coordinates
of one of the points on the line into the equation Y = ln a + 1.4X. If we use the point
(1.76,6.42) then we obtain 6.42 ≈ ln a + 1.4(1.76), or ln a ≈ 3.96. Now, since a = eln a , we
get that a ≈ 52.46.
Thus the data conforms approximately to the relationship y = 52.5x1.4 .
Exercises
1. The following data conform approximately to a relationship of the form y = axb .
Mathematics Learning Centre, University of Sydney 6
In the previous subsection we saw how we could use a log transformation to analyse data
that conformed approximately to the relationship y = axb . In the same way we can
use this transformation to analyse data conforming to the relationship y = aekx . Note
carefully the difference between these two cases. In the former case the independent
variable (in this example x) appears as the base of the exponential expression, whereas
now the independent variable appears as part of the exponent.
Taking logarithms of both sides of the equation y = aekx gives
ln y = ln a + kx.
We end up with a logarithm of the dependent variable, y, but not of the independent
variable, x. Calling Y = ln y (but notice that we do not have to make any change to x), if
we were to graph Y against x we would come up with a straight line. This line has slope
k and Y -intercept ln a.
Example: A scientist doing research into binary stars observes the data shown in the
table in Figure 7.
The scientist suspects that the data conform to a relationship of the form y = aekx .
Calculate Y (= ln y), and plot Y against x on a graph. Is the scientist’s suspicion justified,
and if so what are the values of the constants a and k?
8.00
Y
6.00
∗
4.00 ∗
2.00
x
100 200 300 400 500 600 700 800
Exercise
2. The data in the table below are approximately related by the equation y = aekx .
Calcluate Y = ln y and plot (x, Y ). Find the approximate values of a and k.
2.4 Summary
Case 1
Data that conform (approximately) to a relationship of the form y = axb will yield
(approximately) a straight line when Y = ln y is plotted against X = ln x. If two points
(X1 , Y1 ) and (X2 , Y2 ) are chosen on the line of best fit then the constant b can be found
from
Y 2 − Y1
b= .
X2 − X1
The constant ln a can be found by using this value of b and either one of the points
(X1 , Y1 ) or (X2 , Y2 ) in the equation Y = ln a + bX. Then a can be found using the
equation a = eln a .
Case2
Data that conform (approximately) to a relationship of the form y = aekx will yield
(approximately) a straight line when y = ln y is plotted against x. If two points (x1 , Y1 )
and (x2 , Y2 ) are chosen on the line of best fit then the constant k can be found using the
equation
Y 2 − Y1
k= .
x2 − x1
The constant ln a can be found by using the value of k and either one of the points (x1 , Y1 )
or (x2 , Y2 ) in the equation Y = ln a + kx. Again use a = eln a to find a.
2.5 Exercises
Each of the following sets of data conforms approximately to one of the relationships
y = axb or y = aekx . Find out to which of the relationships each data set conforms, and
find the constants a and b or k.
3.
x 1.75 2.8 3.9 6.3 8.76 15.2 25.53 42.1
y 1.93 3.32 5.7 8.67 14.3 24.05 42.95 84.8
4.
x 1.12 2.22 3.08 3.94 5.0 6.04 7.14 8.18
y 18.2 35.3 66.9 103.8 209.1 388.6 694.1 1398
5.
x 1.00 1.98 3.28 4.58 6.0 6.9 7.82 9.0
y 7370 4045 2553 1218 655 466 272 179
6.
x 0.286 0.374 0.575 0.825 1.23 2.09
y 16.5 32.5 69.6 161.2 318.2 1099
Mathematics Learning Centre, University of Sydney 9
3.1 Introduction
This section is an introduction to that part of calculus which involves exponential and
logarithmic functions. You should only read it if you have some knowledge of calculus.
Do not attempt this section if you have not seen any calculus before. The Mathematics
Learning Centre booklet: Introduction to Differential Calculus may be useful if you need
to learn calculus.
We will use these results to differentiate functions which involve exponentials or loga-
rithms. We will not derive these results here but if you want to see how they are derived
then have a look at any calculus textbook.
We will make extensive use of the chain rule or composite function rule for differentiation
and will give it here to remind you. Of course, we will use the other rules of differentiation
as required.
Have a look at the function h(x) = (x2 + 1)17 . We can think of this function as being
the result of combining two functions. If g(x) = x2 + 1 and f (u) = u17 then the result of
substituting g(x) into the function f is
g f
x −→ x2 + 1 −→ (x2 + 1)17 .
We start off with x. The function g takes x to x2 + 1, and the function f then takes
x2 + 1 to (x2 + 1)17 . Combining two (or more) functions like this is called composing the
functions, and the resulting function is called a composite function. For a more detailed
discussion of composite functions you might wish to refer to the Mathematics Learning
Centre booklet Functions.
Mathematics Learning Centre, University of Sydney 10
Suppose that h is the composite of the differentiable functions y = f (u) and u = g(x).
Then h is a differentiable function of x whose derivative is
Another formulation of the chain rule, which gives us less information but may be easier
to remember, is
dy dy du
= × .
dt du dt
Solution: This function is a composite function so we must use the composite function
rule for differentiation together with result (1).
we have
h (x) = f (g(x))g (x) = e5x 5 = 5e5x .
2
d(e3x )
Example: Find .
dx
Solution: This is an application of the chain rule together with result (1).
we have
h (x) = f (g(x))g (x) = e3x · 6x = 6xe3x .
2 2
Mathematics Learning Centre, University of Sydney 11
Solution: We solve this by using the chain rule and our knowledge of the derivative of ln x.
These rules arise from the chain rule and results (1) and (2), and can speed up the process
of differentiation. It is not necessary that you remember them. If you forget, just use the
chain rule as in the examples above.
Exercises
Differentiate the following functions.
7
1. f (x) = ln 2x3 2. f (x) = ex 3. f (x) = ln(11x7 )
2 3 −2
4. f (x) = ex +x 5. f (x) = e7x 6. f (x) = ln(ex + x3 )
2x
7. f (x) = e x2 +1 8. f (x) = ln( 5x−2
x2 +3
) 9. f (x) = ln(ex x8 )
Recall that if h(x) = f (g(x)) is the composite of y = f (u) and u = g(x). Then
For example, if
1
h(x) = ln(x2 + 1) then h (x) = 2x.
x2 +1
Now, if we think of integration as the reverse of diffentiation we get,
1
2x dx = ln(x2 + 1) + C.
x2 +1
In general, if we integrate both sides of the equation h (x) = f (g(x))g (x) with respect
to x we get,
f (g(x))g (x) dx = h (x) dx = h(x) + C = f (g(x)) + C.
To use this result, you will need to be able to recognise when a function has this form.
That is, the function to be integrated has the form of a product of two functions: One is a
composite function, and the other is the derivative of the ‘inner’ function of the composite.
Note that in some cases, this derivative is a constant.
We can best illustrate this method with some examples.
Example: Find 3e3x dx.
Solution: In this example, e3x is a composite function and the derivative of the ‘inner’
function is 3. So,
3e3x dx = e3x 3 dx = f (g(x))g (x) dx with u = g(x) = 3x and f (u) = eu .
If you have doubts about this, then differentiate the answer to check it out.
2
Example: Find 4xe2x dx
2
Solution: In this example, the composite function is e2x , and the derivative of the ‘inner’
function is 4x. So,
4xe 2x2
= e 2x2
4x dx = f (g(x))g (x) dx with u = g(x) = 2x2 and f (u) = eu .
Solution: Here the composite function is (ln x)2 and the derivative of the ‘inner’ function
is x1 . So,
1
(ln x) · dx = f (g(x))g (x) dx
2
with u = g(x) = ln x and f (u) = u2 .
x
2x − 4
Example: Find dx.
x2 − 4x + 1
1
Solution: Here the composite function is x2 −4x+1
and the derivative of the ‘inner’ function
is 2x − 4. So,
2x − 4 1
dx = f (g(x))g (x) dx with u = g(x) = x2 − 4x + 1 and f (u) = .
x2 − 4x + 1 u
Since if f (u) = 1
u
then f (u) = ln |u| we get,
2x − 4
dx = f (x2 − 4x + 1) + C = ln |x2 − 4x + 1| + C.
x2 − 4x + 1
The examples above suggest the following rules, which can be very useful when integrating
certain functions.
f (x)ef (x) dx = ef (x) + C (7)
f (x)
dx = ln |f (x)| + C. (8)
f (x)
These rules actually follow from results (3) and (4). It is not necessary that you remember
them. If you forget, just use a substitution as in the examples above.
6
Example: Find x5 ex dx.
6
Solution: This one is a little tricky as the composite function is ex but the derivative
of the ‘inner’ function is 6x5 and we have only x5 in the product. We can get around this
difficulty by writing
6 1 6
x5 ex dx = 6x5 ex dx.
6
Mathematics Learning Centre, University of Sydney 14
4x + 2
Example: Find dx.
x2 + x
Solution: Again we need to make a minor change to the integrand first as the derivative
of the ‘inner’ function is 2x + 1.
4x + 2 2x + 1
dx = 2 dx
x2 + x x2 + x
= 2 ln |x2 + x| + C.
Exercises
Evaluate the following indefinite integrals.
2 2 +2 3
10. 2x+1
dx 11. 2xex dx 12. 3x2 ex dx
2x+2 x+2 −2x
13. x2 +2x+7
dx 14. x2 +4x+1
dx 15. e dx
1 −1 1
16. e x2
x3
dx 17. 1−x
dx 18. cos xesin x dx
We now differentiate both sides of this equation with respect to x. To differentiate ln f (x)
we make use of (4):
Mathematics Learning Centre, University of Sydney 15
d 1 f (x)
ln f (x) = · f (x) = .
dx f (x) f (x)
Multiplying both sides of this last equation by f (x) completes the solution.
3 8 cos x
f (x) = f (x) − +
x (2x + 1) sin x
−4 3 8 cos x
3
= x (2x + 1) sin x − +
x (2x + 1) sin x
= 3x2 (2x + 1)−4 sin x + x3 (2x + 1)−4 cos x − 8x3 (2x + 1)−5 sin x
3x2 sin x x3 cos x 8x3 sin x
= + −
(2x + 1)4 (2x + 1)4 (2x + 1)5
We can also use logarithmic differentiation to differentiate functions of the form f (x) = 2x .
f (x) = 2x
ln f (x) = ln(2x )
= x ln 2
1
f (x) = ln 2
f (x)
So, f (x) = f (x) ln 2
= (ln 2)2x
Exercises
Use logarithmic differentiation to find the derivatives of the following functions.
19. f (x) = x7 cos x
(x+1)2
20. f (x) = 3−x 21. f (x) = x−3 (x2 + 3x + 1)4 5−x
3.5 Summary
1
dx = ln |x| + C
x
Mathematics Learning Centre, University of Sydney 16
3.6 Exercises
Calculate the derivatives of the following functions.
9
22. f (x) = ln 3x4 23. f (x) = e2x 24. f (x) = ln 3x6
5 −x2
26. f (x) = ln 4x−6
2
25. f (x) = ex 27. f (x) = ln(ex + 2x4 )
4 3x2 + 7 1 3 x4 +1
31. 4x3 ex dx 32. dx 33. xe dx
x3 + 7x + 7 2
√
sin x ln x 1 − 13
34. dx 35. dx 36. e x dx
cos x + 5 x x4
5
xe−x dx
x 2
37. ex ee dx 38. dx 39.
x
Mathematics Learning Centre, University of Sydney 17
4.1 Introduction
To introduce the topic of exponential growth and decay we can consider the growth in
the population of a single-celled organism. Suppose that the animal reproduces by each
single cell splitting into two cells, and that this occurs regularly, say every hour. Let’s
imagine starting off with just one of these organisms and looking at how the population
grows as time goes by (see Figure 10). If none die then after one hour the organism will
have reproduced and there will be two of them. After a further hour has elapsed each of
these two animals will have itself reproduced and there will now be 4 = 22 of them. After
another hour there will be 8 = 23 of the animals, and so on. This type of increase in the
population is called exponential growth.
Now let us try to write down what is going on here in symbols. Instead of assuming that
the animals reproduce every hour, suppose they reproduce every T units of time. The
unit of time could be seconds, minutes, hours or even years, we don’t care as long as we
stick to the same measuring units throughout the problem.
Let’s measure time from some fixed moment, which we will call t = 0, and we will suppose
now that instead of beginning with one organism we started off with an initial population
of P0 . This means that at time t = 0, the population of the animals is P0 .
Well, at time T each of the P0 single cells has split into two cells, so the population is
2 × P0 . We can represent this by the equation P (T ) = P0 × 2.
After a further period of time T , when t = 2T , each of the 2 × P0 cells has split and the
population is now 4 × P0 = P0 × 22 . Thus P (2T ) = P0 × 22 .
Mathematics Learning Centre, University of Sydney 18
Proceeding like this we see that the population at time nT is P0 × 2n . The equation
which represents the growth of the cell culture is thus P (nT ) = P0 × 2n . If we make the
substitution t = nT , then we obtain the equation
t
P (t) = P0 × 2 T .
Since 2 = eln 2 , this equation can be rewritten as
T
P (t) = P0 × (eln 2 ) t
ln 2
×t
= P0 × e T
= P0 ekt , k= ln 2
T
.
The equation
P (t) = P0 ekt
(with k > 0) is the equation for exponential growth.
Let’s look at a second example, a certain radioactive isotope, isotope X, the atoms of
which decay into another isotope, isotope Y. Suppose that in an interval of time T , half
of the atoms of isotope X decay into atoms of isotope Y. If the initial amount of isotope
X was P0 then after time T the amount of isotope Y left is P0 × 12 .
After a further interval of time T (so that t = 2T ), half of the remaining atoms of isotope
X have decayed, and the amount of isotope X remaining is P0 × 12 × 12 . This can be
represented by the equation P (2T ) = P0 × 2−2 .
Proceeding in this way we find that after a time interval nT the amount of isotope X
remaining is equal to P (nT ) = P0 × 2−n . Have a look at Figure 11.
t=0 t=T t = 2T t = 3T t = 4T
Figure 11: Relative amounts of isotope X remaining for various time periods.
P (t) = P0 × e− ×t
ln 2
T
= P0 ekt , k = − lnT2 .
The equation
P (t) = P0 ekt
(with k < 0) is the equation for exponential decay.
You should notice that the equation representing exponential growth is the same as that
representing exponential decay, except that for growth the constant k is greater than zero,
and k < 0 for decay.
The speed of exponential decay is often specified by giving the half life of the quantity
that is decaying. The half life is the time it takes for a quantity to reduce to half of its
initial size. If we are told the half life of a decaying quantity then we are able to calculate
the constant k in the equation P (t) = P0 ekt .
Example: If a quantity P is decaying exponentially with a half life of 250 years, find the
equation expressing the size of this quantity at time t.
Solution: In 250 years time the size of the quantity will be half its present size. If P0 is
the initial size of the quantity then, when t = 250, P = 12 P0 . So,
1
P (250) = P0 = P0 ek×250
2
1
= ek×250
2
1
ln = k × 250
2
ln 12
k = .
250
We have found the value of the constant k, and so the equation representing the decay of
the quantity is
ln 1
2
P (t) = P0 e 250 t .
Since ln 12 = ln 2−1 = − ln 2, the equation representing the decay of the quantity can also
be written as
− ln 2
t
P (t) = P0 e 250 .
Example: Under certain laboratory conditions the population of a particular single celled
organism is known to grow exponentially. A researcher begins an experiment with the
organism and observes that after 3 hours the population has increased to a level 1.5 times
that of the initial population. How long after the beginning of the experiment will the
population double?
Mathematics Learning Centre, University of Sydney 20
If necessary this number may be found in decimal form using a calculator (5.1285 to 4
decimal places). However in most circumstances it is preferable to leave it in the form
given here.
Exercises
1. The size of a quantity P at time t is given by P (t) = 1300e2.1t , where t is measured in
seconds.
(a) Is P increasing or decreasing with time?
(b) What is the initial value of P (ie what is P (0))?
(c) What is the value of P after 1.5 seconds?
(d) In how many seconds (measured from t = 0) will the quantity be equal to 2000?
life cycle of the organism, and whilst the organism is alive the level of Carbon-14 in it
remains roughly constant. When the organism dies no more Carbon-14 is ingested by it
and the level decreases. Measurement of the level of Carbon-14 in ancient dead organic
matter, and comparison of the level with that contained in similar living matter, yields
an estimate of the age of the dead matter.
Analysis of some human remains found in the Gibson Desert shows that the level of
Carbon-14 in the bones was 0.358 times that which would be found in the bones of living
humans. What is the approximate age of these remains?
Solution: If C represents the level of Carbon-14 in the remains, and measuring time t
from the moment of death of the human, then
C(t) = C0 ekt .
After 5730 years (the half life of Carbon-14) C is half its original level, so
1
C(5730) = C0 = C0 ek×5730
2
1
= ek×5730
2
1
ln = k × 5730
2
ln 12
k =
5730
and
1
ln 2
C(t) = C0 e 5730 t .
So far all we have done is use our information about the half life of Carbon-14 to find
the formula by which Carbon-14 decays. Our solution so far would be the same for any
problem involving radiocarbon dating. To find the age of these particular bones we need
to use the fact that the level of Carbon-14 has been reduced to 0.358 of the initial level.
So
1
ln 2
0.358C0 = C0 e 5730 t
ln 1
2 t
0.358 = e 5730
ln 12
ln 0.358 = t
5730
ln 0.358
t = ln 1 2
5730
≈ 8600.
The remains are approximately 8600 years old.
Exercise
3. Plutonium-239 is one of the most toxic substances known. The safe level of Plutonium-
239 that can be ingested by one human has been set at 0.64 micrograms, or 6.4 × 10−7
grams. It decays exponentially, and has a half life of 243,000 years. In January 1968
an atomic bomb was lost in Greenland, and it has been estimated that 400 grams of
Plutonium-239 was released into the marine environment. Calculate the length of time
which must pass before the 400 grams of Plutonium-239 has decayed to an amount which
is considered safe for ingestion by one human being.
Mathematics Learning Centre, University of Sydney 22
This subsection is intended for people who have some knowledge of calculus, and should
be omitted by anyone who has not.
If P is a quantity that is growing or decaying exponentially then the size of the quantity
P at time t is described by the equation P (t) = P0 ekt , where a negative constant k means
decay and a positive k means growth. It is natural to ask about the rate of change in the
size of the quantity.
Questions about rates of change in mathematics are usually answered by means of calculus.
The rate of increase or decrease in the quantity P is equal to dP
dt
.
dP
If P (t) = P0 ekt ,then dt
= kP0 ekt = kP (t).
Thus if the size of the quantity at time t, P (t), is given by the equation P (t) = P0 ekt then
the rate of growth (or decay) of the quantity is proportional to the size of the quantity. In
symbols,
dP
∝ P or, introducing the proportionality constant k,
dt
dP
= kP.
dt
On the other hand, suppose we know that a quantity P increases (or decreases) at a
rate proportional to P (this can be expressed symbolically by the equation dP
dt
= kP or
P (t) = kP ). Then
P (t)
= k
P (t)
P (t)
dt = k dt
P (t)
ln P (t) = kt + C
P (t) = ekt+C = eC ekt .
What we have shown here is that exponential growth or decay is characterised by the
property that the rate of change of the quantity under consideration is proportional to
the size of the quantity. If we are studying a quantity that we know increases (or decreases)
at a rate which is proportional to the size of the quantity then this quantity is growing
(or decaying) exponentially.
The examples with which we introduced exponential growth and decay may make more
sense to you now. If a single celled animal reproduces every hour, and none of the animals
in the culture die, then the rate of increase in the population is proportional to the size of
the population—if there are twice as many organisms then they will have twice as many
offspring. This is why the population of such (imaginary) organisms grows exponentially.
Mathematics Learning Centre, University of Sydney 23
Example: A scientist is studying a cell culture and finds that the the population P of
the culture increases at a rate given by the equation
dP
= 1.7P.
dt
How long does it take for the population to increase to a level 2.5 times that of the initial
population?
P (t) = P0 e1.7t
we will work it out as follows. Rearranging the equation P (t) = 1.7P gives
P (t)
= 1.7.
P (t)
Taking the indefinite integral (with respect to t) of both sides of this equation gives
P (t)
dt = 1.7dt
P (t)
ln P (t) = 1.7t + C
eln P (t) = e1.7t+C
P (t) = eC e1.7t .
P (t) = P0 e1.7t .
We are asked to find how long until the population of the culture grows to a level 2.5
times that of the initial population. In other words, when is P (t) = 2.5P0 ?
Well
Exercise
4. (Attempt only if you know some calculus.) A quantity P being measured in an
experiment increases according to the equation dPdt
= 4.5P . The initial amount of the
quantity was 480 units. Find an expression for the amount P (t) of the quantity at time
t.
Mathematics Learning Centre, University of Sydney 24
4.3 Summary
The exponential growth or exponential decay of a quantity P are both represented by the
equation
P (t) = P0 ekt
where
P (t) = the size of the quantity P at time t,
P0 = the initial size of the quantity (at time t = 0),
k = a constant, k > 0 for growth and k < 0 for decay,
t = time, measured in any convenient units.
The rate of exponential decay is often specified by stating the half life of the quantity.
The half life is the time for taken for a quantity to decay to half its initial size.
If a quantity P is growing (or decaying) in an exponential fashion then the rate of growth
(or decay) of the quantity is proportional to the size of the quantity. In symbols
dP
∝ P or
dt
dP
= kP.
dt
On the other hand, any quantity that grows (or decays) in such a fashion is growing (or
decaying) exponentially.
dP
In short, exponential growth (or decay) is characterised by the property that dt
= kP .
4.4 Exercises
5. A quantity A is known to grow exponentially. After 2 minutes of a laboratory exper-
iment the quantity was measured to be 153, and exactly one minute later the quantity
was measured to be 247.
(a) What was the initial value of A?
(b) Write down an expression for A(t).
(c) What is the size of A after 7 minutes?
(d) When will the value of A(t) be 2000?
7. Radium decays exponentially, with a half life of 1466 years. This means that it takes
1466 years for the amount of radium in a given sample to decrease to half its initial level.
(a) Find the equation governing the decay of radium.
(b) How long does it take for the amount of radium in a given sample to decrease to one
fifth of its initial level?
Mathematics Learning Centre, University of Sydney 25
5 Solutions to Exercises
3. ( 15 )−2 = 1
( 15 )2
= 25
4. (3−1 )2 = 3−2 = 1
32
= 1
9
5. ( 52 )−2 = ( 25 )2 = 4
25
3
6. (−8) 2 is not defined.
7. ( −27 ) 3 = (( −27 ) 3 )2 = ( −3
2 1 9
8 8 2
)2 = 4
1 1 1 1 1 1 1 1
13. (a 2 + b 2 )2 = (a 2 )2 + 2a 2 b 2 + (b 2 )2 = a + 2a 2 b 2 + b
14.
x2
(x2 + y 2 ) 2 − x2 (x2 + y 2 )− 2 = (x2 + y 2 ) 2 −
1 1 1
1
(x2 + y 2 ) 2
1 1
(x2 + y 2 ) 2 (x2 + y 2 ) 2 − x2
= 1
(x2 + y 2 ) 2
x2 + y 2 − x2
= 1
(x2 + y 2 ) 2
y2
= 1
(x2 + y 2 ) 2
1 1
x 2 +x x2 x 1
15. 1 = 1 + 1 = 1 + x2
x2 x2 x2
16.
1 1 2 1 2 1 2 1 1 1 2 1 2 1 1 1 2
(u 3 − v 3 )(u 3 + (uv) 3 + v 3 ) = u 3 u 3 + u 3 (uv) 3 + u 3 v 3 − v 3 u 3 − v 3 (uv) 3 − v 3 v 3
= u−v
Mathematics Learning Centre, University of Sydney 26
17. The graphs are drawn in Figures 12 and 13 below. Notice that the graph of f (x) =
2.9x is very close to the graph of f (x) = 3x , and similarly for the other pair of graphs.
y
40 y = 3x
y = 2.9 x
30
20
10
x
-4 -3 -2 -1 1 2 3 4
y = 3-x 40
y = 2.9 -x 30
20
10
x
-4 -3 -2 -1 1 2 3 4
2
22. ln ee21 = ln e2−21 = −19
ln e7 7 7
23. log11 121
= log11 112
= 2
9 9
25. eln 2 = 2
√
3 √
26. eln 27
= 3
27 = 3
30x3 y
27. 3 log2 x + log2 30 + log2 y − log2 w = log2 w
2 a
28. 2 ln x − ln y + a ln w = ln x2 − ln y + ln wa = ln x yw
The data have been plotted and a line of best fit drawn in Figure 15. The points
Y 8.00
6.00
4.00
*
*
2.00
X
0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50
(1, 3.53) and (2, 4.55), each marked with an asterix, are on the line of best fit. We
calculate the slope of this line, b,
4.55 − 3.53
b =
2−1
1.02
≈
1
≈ 1.02
Mathematics Learning Centre, University of Sydney 28
To find the value of the constant ln a we substitute this value of b and the coordinates
of one of the points on the line, say (1, 3.53) into the equation Y = ln a + bX.
3.53 = ln a + 1.02(1)
ln a = 3.53 − 1.02
≈ 2.51
So, a = e2.51 = 12.3. Thus the relationship between x and y is given approximately
by the equation
y = 12.3 × x1.02 .
A remark about this solution: yours may be slightly different from this one. Remem-
ber that the choice of line of best fit is a subjective one, and that we have rounded
all numbers to a couple of decimal places. However, if correct, your answer will not
differ from this one by much.
The data have been plotted and a line of best fit drawn in Figure 17. The point
Y
2.00
*
1.00
x
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
-1.00
(x, Y ) = (3.14, −0.04) is on the line. We have marked by an asterix another point on
the line, (7, 1.61). We calculate the slope of this line, k, using these two points. In
calculating the slope of the line we will take the logarithm of the y-values only, not
of the x-values. Taking logarithms to base e of both sides of the equation y = aekx
gives ln y = ln a + kx, and the slope of the line is k.
1.61 − (−0.04)
k ≈
7 − 3.14
1.65
≈
3.86
≈ 0.43.
Mathematics Learning Centre, University of Sydney 29
To find the value of the constant ln a we substitute k = 0.43 and the coordinates of
one of the points on the line into the equation Y = ln a + kx. Substituting the point
(3.14, −0.04) we get
−0.04 ≈ ln a + (0.43)(3.14)
ln a ≈ −0.04 − 1.35
≈ −1.39.
Y 8.00
6.00
4.00
2.00
X
0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50
On the line of best fit we have chosen the points (2.72, 3.18) and (3.74, 4.44) to
calculate the constants a and b.
4.44 − 3.18
b =
3.74 − 2.72
1.26
≈
1.02
≈ 1.24
Mathematics Learning Centre, University of Sydney 30
To find the constant ln a we substitute the coordinates of one of the points on the
line into the equation Y = ln a + bX. We will use the point (2.72,3.18).
3.18 ≈ ln a + (1.24)(2.72)
ln a ≈ 3.18 − 3.37
≈ −0.19
Y 8.00
6.00
4.00
2.00
On the line of best fit we have chosen the points (1.12, 2.90) and (8.18, 7.24), to
calculate the constants a and b,
7.24 − 2.90
k ≈
8.18 − 1.12
4.34
≈
7.06
≈ 0.61
Mathematics Learning Centre, University of Sydney 31
Substituting the coordinates of the point (1.12, 2.90) into the equation Y = ln a + kx,
we get
2.90 ≈ ln a + (1.12)(0.61)
ln a ≈ 2.90 − 0.68
≈ 2.22
So, a ≈ 9.20
Thus the data conform approximately to the relationship y = 9.20 × e0.61x .
Y
8.00
*
*
6.00
4.00
2.00
7.00 − 7.93
k ≈
5.00 − 3.00
−0.93
≈
2.00
≈ −0.47
Substituting the coordinates of the point (5.00, 7.00) into the equation Y = ln a + kx
gives us
7.00 ≈ ln a + (−0.47)(5.00)
Mathematics Learning Centre, University of Sydney 32
ln a ≈ 7.00 + 2.35
≈ 9.35
So, a ≈ 11499
Thus the data conform approximately to the relationship y = 11499e−0.47x .
6. Plotting Y = ln y against X = ln x results in a straight line whereas plotting Y = ln y
against x will result in a curve. Therefore this data conforms to a relationship of the
form y = axb . A table of the values of Y = ln y and X = ln x is given in Figure 24.
A plot of Y = ln y against X = ln x has been made in Figure 25, and a line of best
fit drawn in.
8.00
Y
6.00
4.00
2.00
X
-2.00 -1.50 -1.00 -0.50 0.50 1.00 1.50 2.00 2.50
On the line of best fit we have chosen the points (−1.25, 2.80) and (0.74, 7.00) to
calculate the constants a and b.
7.00 − 2.80
b =
0.74 − (−1.25)
4.20
≈
1.99
≈ 2.11.
To find the constant ln a we substitute the coordinates of one of the points on the
line into the equation Y = ln a + bX. We will use the point (0.74, 7.00).
7.00 ≈ ln a + (2.11)(0.74)
ln a ≈ 7.00 − 1.56
≈ 5.44
So, a ≈ 230.44.
Mathematics Learning Centre, University of Sydney 33
5.
−2
d(e7x ) d(7x−2 ) −2
= × e7x
dx dx
−2
= (−14x−3 ) × e7x
−2
−14e7x
=
x3
6.
d ln(ex + x3 ) d(ex + x3 ) 1
= × x
dx dx e + x3
1
= (ex + 3x2 ) × x
e + x3
x 2
e + 3x
=
ex + x3
7.
2x
d(e x2 +1 ) d( x22x+1 ) 2x
= × e x2 +1
dx dx
(x2 + 1)(2) − (2x)(2x) 2x
= × e x2 +1
(x2 + 1)2
2 − 2x2 2x
= × e x2 +1
(x2 + 1)2
8.
d(ln( 5x−2
x2 +3
)) d( 5x−2
x2 +3
) 1
= × 5x−2
dx dx x2 +3
(x2 + 3)(5) − (5x − 2)(2x) x2 + 3
= ×
(x2 + 3)2 5x − 2
15 + 4x − 5x 2 2
x +3
= ×
2
(x + 3) 2 5x − 2
15 + 4x − 5x 2
=
(x2 + 3)(5x − 2)
Mathematics Learning Centre, University of Sydney 34
9.
d(ln(ex x8 )) d(ex x8 ) 1
= × x 8
dx dx e x
1
= ((ex )(8x7 ) + (x8 )(ex )) ×
ex x8
ex (x8 + 8x7 )
=
ex x8
x + 8x7
8
=
x8
10. 2
2x+1
dx = ln |2x + 1| + C
2 +2 2 +2
11. 2xex dx = ex +C
3 3
12. 3x2 ex dx = ex + C
13. 2x+2
x2 +2x+7
dx = ln |x2 + 2x + 7| + C
14. x+2
x2 +4x+1
dx = 1
2
2x+4
x2 +4x+1
dx = 12 ln |x2 + 4x + 1| + C
−2x
15. e dx = − 12 e−2x + C
1 −1
(2x−3 )e− x2 dx = 12 e− x2 + C
1 1
1
16. e x2
x3
dx = 2
17. 1
1−x
dx = − ln(1 − x) + C = ln( 1−x
1
)+C
18. cos xesin x dx = esin x + C
19.
x7 cos x
f (x) =
(x + 1)2
ln f (x) = 7 ln x + ln(cos x) − 2 ln(x + 1)
f (x) 7 sin x 2
= − −
f (x) x cos x x + 1
7 sin x 2
f (x) = f (x) − −
x cos x x + 1
x7 cos x 7 sin x 2
= − −
(x + 1)2 x cos x x + 1
20.
f (x) = 3−x
ln f (x) = (−x) ln 3
f (x)
= (−1) ln 3
f (x)
f (x) = f (x)(− ln 3)
= −(ln 3)3−x
Mathematics Learning Centre, University of Sydney 35
21.
9
d(e2x ) 9
23. dx
= 18x8 e2x
5 2
d(ex −x ) 5 −x2
25. dx
= (5x4 − 2x)ex
26.
d(ln 4x−6 ) −24x−7
=
dx 4x−6
−6
=
x
27.
2 2
d(ln(ex + 2x4 )) d(ex + 2x4 ) 1
= × x2
dx dx e + 2x4
2 1
= (2xex + 8x3 ) × x2
e + 2x4
x2 3
2xe + 8x
=
ex2 + 2x4
28.
x4 (x + x2 )3
f (x) =
(x − 1)2
ln f (x) = 4 ln x + 3 ln(x + x2 ) − 2 ln(x − 1)
f (x) 4 1 + 2x 2
= +3 −
f (x) x x+x 2 x−1
4 1 + 2x 2
f (x) = f (x) +3 −
x x + x2 x − 1
x4 (x + x2 )3 4 1 + 2x 2
= + 3 −
(x − 1)2 x x + x2 x − 1
Mathematics Learning Centre, University of Sydney 36
29.
2
f (x) = 4x ex
ln f (x) = x ln 4 + x2 ln e
f (x)
= ln 4 + 2x
f (x)
f (x) = f (x)(ln 4 + 2x)
2
= (ln 4 + 2x)4x ex
30.
4 4
31. 4x3 ex dx = ex + C
3x2 +7
32. x3 +7x+7
dx = ln |x3 + 7x + 7| + C
1 3 x4 +1 1 4 +1 4 +1
33. xe
2
dx = 8
4x3 ex dx = 18 ex +C
− sin x
34. sin x
cos x+5
dx = − cos x+5
dx = − ln | cos x + 5| + C
√ln x 3
35. x
dx = 23 (ln x) 2 + C
1 −1
3x−4 e− x3 dx = 13 e− x3 + C
1 1
1
36. e x3
x4
dx = 3
x ex x
37. e e dx = ee + C
5
38. x
dx = 5 ln |x| + C
xe−x dx = − 12 (−2x)e−x dx = − 12 e−x + C
2 2 2
39.
1. a. Since 2.1 > 0, the quantity P (t) = 1300e2.1t is increasing with time.
b. P (0) = 1300e2.1×0 = 1300.
c. The value of P after 1.5 seconds is P (1.5) = 1300e2.1×1.5 ≈ 30337.
Mathematics Learning Centre, University of Sydney 37
3. The initial amount of plutonium released was 400g so when t = 0, P (0) = P0 = 400.
The equation for decay is P (t) = 400ekt .
Plutonium has a half life of 243,000 years, so when t = 243000, P = 12 400, so
The time taken for 400g to decay to 6.4 × 10−7 g is given by the equation
1
ln 2
6.4 × 10−7 = 400e 243000 t
1
ln 2
t 6.4 × 10−7
e 243000 =
400
ln 12 6.4 × 10−7
t = ln
243000 400
243000 6.4 × 10−7
t = ln
ln 12 400
≈ 7.1 × 106 years.
2000
e0.479t =
58.7
2000
0.479t = ln
58.7
1 2000
t = ln
0.479 58.7
≈ 7.37 minutes.
6. a. Since the quantity grows exponentially, and because the initial quantity is 47.9,
we know that it behaves according to a relationship of the form B(t) = 47.9ekt .
The other piece of information we are given is
B(6.7) = 47.9ek×6.7 = 102, which means that
102
ek×6.7 =
47.9
102
k × 6.7 = ln
47.9
1 102
k = ln
6.7 47.9
≈ 0.113.
The quantity B behaves according to the equation B(t) = 47.9e0.113t .
b. B(t) has increased to 500 when
B(t) = 47.9e0.113t = 500
500
e0.113t =
47.9
500
0.113t = ln
47.9
1 500
t = ln
0.113 47.9
≈ 20.76.
The quantity will have increased to 500 after approximately 20.76 minutes.
b. The amount of radium decays to one fifth of its original level when
ln 1
2 1
e 1466 t =
5
1
ln 2 1
t = ln
1466 5
1466 1
t = ln
ln 12 5
≈ 3404 years.