Methods in Calculus

Robert Brignall

May 16, 2005

1 Limits & L’Hôpital’s Rule

1.1 Limits
Definition 1.1. We wish to know whether a function, f (x) has a limit as
x approaches a particular value, a. On the number line, we can approach
a from below or above, and the limit of f (x) is written limx→a− f (x) and
limx→a+ f (x) respectively.
If
lim f (x) = lim f (x)
x→a− x→a+

then we say that the limit exists, and we can just write

lim f (x)
x→a

to mean “the limit of f (x) as x tends to a” (both from above and from
below). When the limit exists, we have limx→a f (x) = f (a).

Example 1.2. (1) f (x) = x2 . Then for all values of a,

lim f (x) = f (a) = a2 .

x→a

(2) f (x) = x1 . Then the limit exists for all values of a except a = 0. As
x tends to 0 from below, f (x) tends to −∞, but as x tends to 0 from
above, f (x) tends to +∞. So
1
lim does not exist.
x→0 x
g(x)
Suppose we have a function f (x) = h(x) for some functions g(x) and
h(x). Observe that limx→a f (x) can only exist if

lim g(x)
x→a

exists.

1
1.2 L’Hôpital’s Rule 2

If, on the other hand,

lim h(x) = 0
x→a

then
limx→a g(x)
lim f (x) =
x→a 0
which in general does not exist. Additionally, if limx→a g(x) exists and
6 0, then limx→a f (x) can never exist.
g(a) =
1+x2
Example 1.3. Let f (x) = x , so g(x) = 1 + x2 and h(x) = x. Then

lim g(x) = g(0) = 1 + 02 = 1

x→0

and
lim h(x) = h(0) = 0
x→0

so limx→0 f (x) does not exist.

g(x)
If f (x) = h(x) and

lim g(x) = lim h(x) = 0 or ± ∞

x→a x→a

then limx→a f (x) might exist.

−1 3
Example 1.4. Let f (x) = 8x 2x−1 . Then limx→1/2 f (x) is an indeterminate
of the form 00 . But the limit still exists:

8x3 − 1 (2x − 1)(4x2 + 2x + 1)

lim = lim
x→1/2 2x − 1 x→1/2 2x − 1
= lim 4x2 + 2x + 1
x→1/2
1 1
= 4· + 2 · + 1 = 3.
4 2

1.2 L’Hôpital’s Rule

g(x)
Theorem 1.5 (L’Hôpital’s Rule). Let f (x) = h(x) , and suppose that, for
some a,
lim g(x) = lim h(x) = 0 or ± ∞
x→a x→a

then
g(x) g 0 (x)
lim = lim 0 .
x→a h(x) x→a h (x)

What does this mean? We can use L’Hôpital’s Rule to find the limit of
indeterminates of the form 00 or ∞
∞.
1.2 L’Hôpital’s Rule 3

Example 1.6. (1) f (x) = sin x

2x , so g(x) = sin x, h(x) = 2x. As x tends
to 0, g(x) → 0 and h(x) → 0 (so f (x) is an indeterminate of the form
0
0 ). Thus using L’Hôpital:

sin x cos x cos 0 1

lim = lim = =
x→0 2x x→0 2 2 2
3−1
(2) f (x) = 8x
2x−1 . We’ve already seen the limit as x → 1/2 is 3, but can
we use L’Hôpital to show this too?
Let g(x) = 8x3 and h(x) = 2x − 1, so g(1/2) = h(1/2) = 0. Then

8x3 − 1 24x2
lim = lim
x→1/2 2x − 1 x→1/2 2

= lim 12x2
x→1/2
1
= 12 · = 3.
4

x2
(3) f (x) = x2 e−x . Rewrite as f (x) = ex so g(x) = x2 and h(x) = ex .
What is limx→∞ f (x)?
As x → ∞, so g(x), h(x) → ∞, and thus we have an indeterminate of
the form ∞
∞ . Using L’Hôpital:

x2 2x
lim = lim x .
x→∞ ex x→∞ e

Note now that both 2x and ex tend to ∞ as x tends to ∞, so we

still have an indeterminate of the form ∞
∞ . But this means we can use
L’Hôpital again:
2x 2
lim = lim
x→∞ ex x→∞ ex
= lim 2e−x = 0.
x→∞
2 Hyperbolic Functions 4

2 Hyperbolic Functions
Definition 2.1. The two main hyperbolic functions are defined as follows:

ex − e−x
sinh x =
2
ex + e−x
cosh x =
2
Now draw two pictures!

Definition 2.2. As with the trigonometric functions, we can then define

four more functions:
sinh x ex − e−x
tanh x = = x
cosh x e + e−x
1 2
cschx = = x
sinh x e − e−x
1 2
sechx = = x
cosh x e + e−x
1 ex + e−x
coth x = = x
tanh x e − e−x
These functions behave in a way that is very similar to the trigonometric
functions sin, cos, tan and related functions.

2.1 Some identities

Proposition 2.3. The following identities hold for all x:

1. cosh2 x − sinh2 x = 1,

2. cosh x + sinh x = ex ,

3. cosh x + sinh x = e−x ,

4. sinh(2x) = 2 sinh x cosh x,

5. cosh(2x) = 2 cosh2 x − 1 = 1 + 2 sinh2 x.

Remark 2.4. Compare items 1, 4 and 5 to the trigonometric functions:

cos2 x + sin2 x = 1
sin(2x) = 2 sin x cos x
cos(2x) = 2 cos2 x − 1 = 1 − 2 sin2 x

Proof. 1. Tutorial sheet exercise.

2.2 Differentiation and integration 5

2. We have
ex + e−x ex − e−x
cosh x + sinh x = +
2 2
ex + e−x + ex − e−x 2ex
= = = ex
2 2
3. Similar to 2.
4. We have
e2x − e−2x
sinh 2x =
2
e + 1 − 1 − e−2x
2x
=
2
e2x + ex e−x − ex e−x − e−2x
=
2
ex (ex + e−x ) − e−x (ex + e−x )
=
2
(ex − e−x )(ex + e−x )
= 2 = 2 sinh x cosh x
4
5. Tutorial sheet exercise.

2.2 Differentiation and integration

Recall that when we differentiate the trigonometric functions, we get:
d
sin x = cos x
dx
d
cos x = − sin x
dx
What happens if we differentiate sinh x and cosh x? First, recall the relation
d f (x)
e = f 0 (x)ef (x) .
dx
Proposition 2.5.
d
sinh x = cosh x
dx
d
cosh x = sinh x
dx
Proof.
ex − e−x
 
d d
sinh x =
dx dx 2
d x d −x
dx e − dx e
=
2
ex − (−e−x )
= = cosh x
2
2.3 Inverses 6

d
dx cosh x is done similarly.
d
Now we should be able to find out what dx tanh x is. Since

sinh x
tanh x =
cosh x
we can use the quotient rule when differentiating:
d d sinh x
tanh x =
dx dx cosh x
d d
dx (sinh x) · cosh x − sinh x · dx (cosh x)
=
cosh2 x
2 2
cosh x − sinh x −1
= 2 = = −sech2 x
cosh x cosh2 x
d
Compare this to dx tan x = sec2 x without the minus sign.
Since integration is the reverse process to differentiation, we now also
know:
Z
sinh xdx = cosh x + c
Z
cosh xdx = sinh x + c

2.3 Inverses
For a function y = f (x), we wish to know whether it has an inverse function
f −1 , so that x = f −1 (y). What are the inverses of cosh x and sinh x?
If y = cosh x then x cosh−1 y is not a function unless we restrict the
values for x to x ≥ 0, hence y ≥ 1. Then we have:

ex + e−x
y =
2
⇒ 2y = ex + e−x
⇒ 2yex = e2x + 1
⇒ (ex )2 − 2yex + 1 = 0

which is a quadratic equation in ex . Therefore we can use the quadratic

formula to find ex :
p
x 2y ± (−2y)2 − 4 · 1 · 1
e =
p 2·1
= y± y −1 2
3 First Order Differential Equations 7

We have two solutions, but notice that for all values of x, ex > 0 so the
negative value is impossible. Therefore
p
ex = y + y 2 − 1
and so p
x = ln(y + y 2 − 1).
p
Similarly, the inverse of sinh x is x = ln(y + y 2 + 1), left as a tutorial
exercise.

3 First Order Differential Equations

These are algebraic equations which contain at least one derivative and any
number of functions of x and/or y.
Definition 3.1. The order of a differential equation is the highest “multiple-
differential” present.
Example 3.2. The equation
dy d2 y
x2 + 3 − sin y =
dx dx2
has second order.
Definition 3.3. A linear differential equation is one which contains differ-
entials of power 1 only.
Example 3.4. Using the notation ẏ = dydt ,
...
ẏ 3 − ÿ − y 2 = sin(ty)
is not linear, while
y 0000 + 4y 00 − 7 = tan ex
is linear (and of 4th order).
We will be looking only at linear first order differential equations.
Differential equations occur throughout the sciences as well as in eco-
nomics. Typical examples include radioactive decay, population growth and
compounding interest on a bank account.
Example 3.5. Repaying a mortagage. At time t = 0 suppose you borrow
an amount of money S0 pounds from the bank. At any given time t, your
debt is S(t) pounds, which has interest rate r percent. At the same time,
you pay off the mortgage at rate p. The amount of money in your account
at any given time therefore obeys the differential equation
dS
= (1 + r)S − p.
dt
How do we solve this so we know how much money, S(t), we have at time
t?
3.1 Solving a Differential Equation 8

3.1 Solving a Differential Equation

This involves rewriting the differential equation in a form where no differ-
ential is present.
dy
Example 3.6. dx = 2x is a differential equation. We can solve this by
integrating both sides, and the general solution is y = x2 + c.
If we know some extra pieces of data, for example “y = 2 when x = 1”,
we can evaluate the constant of integration:
Example 3.7. Using our previous equation y = x2 + c, then 2 = 12 + c so
c = 1, and we have a particular solution,
y = x2 + 1.
So, in general, a differential equation of the form
dy
= f (x)
dx
can simply be solved by direct integration,
Z
y = f (x)dx + c.

However, if there is a function of y also present, for example dy/dx =

f (x)g(y) then we cannot integrate immediately, we must first “separate the
variables”.

3.2 Separable Differential Equations

This method can be applied to all differential equations which can be written
in the form
dy
= f (x)g(y)
dx
and it is solved by integrating both sides after “moving” g(y) to the left-
hand-side of the equation:
Z Z
dy
= f (x)dx
g(y)
1 0 2y
Example 3.8. xy = x2 +1
. Move ys to the left, xs to the right to get
1 dy 2x
= 2
y dx x +1
which we can then integrate to get
ln |y| = ln |x2 + 1| + c
and thus
y = A(x2 + 1)
where A = ec .
3.3 Integrating Factors 9

Example 3.9. From the equation in Example 3.5, lets solve the differential
equation
dS
= (1 + r)S − p
dt
subject to the initial conditions that S = S0 when t = 0.
Rearrange the equation to get
Z Z
dS
= dt
(1 + r)S − p
and integrate both sides, giving
ln |(1 + r)S − p|
= t + c.
1+r
p (1+r)c
More usefully, S = 1+r + Ae(1+r)t for constant A = e 1+r .
p
Now apply the initial conditions, S = S0 at t = 0, to get A = S0 − 1+r ,
and so the solution is
 
p p
S= + S0 − e(1+r)t .
1+r 1+r
When have we paid off the whole debt? Say at time t = T we have
S = 0. Then
 
p p
0 = + S0 − e(1+r)T
1+r 1+r
 
1 1
⇒ T = ln
1+r 1 − (1 + r)S0 /p
Note that if you pay back too slowly (i.e. p is small, and in particular
p < (1 + r)S0 ) then 1 − (1 + r)S0 /p < 0 and the debt will never be paid off!

3.3 Integrating Factors

This method can be applied to all differential equations that can be written
in the form
dy
+ f (x)y = g(x).
dx
The method is derived from the product rule for differentiation,
d dy
(yh(x)) = h(x) + h0 (x)y
dx dx
which here we are trying to “reverse”.
dy
So we want to multiply the equation dx + f (x)y = g(x) by a function
h(x) so that we have
dy
h(x) + f (x)h(x)y = g(x)h(x)
dx
3.3 Integrating Factors 10

and with h(x) satisfying

h0 (x) = f (x)h(x).
d
ef (x) = f 0 (x)ef (x) , so if we pick h(x) so that
R
Recall that dx
R
f (x)dx
h(x) = e

then R
h0 (x) = f (x)e f (x)dx
= f (x)h(x).
Such an h(x) is called the integrating factor.
So our differential equation can be written
d  R  R
f (x)e f (x)dx y = g(x)e f (x)dx
dx
and integrating we get
R Z R
f (x)dx f (x)dx
yf (x)e = g(x)e dx.

Providing we can solve the integral on the right hand side, we have solved
the differential equation.
Example 3.10. Find the general solution of the differential equation
dy
− 2y = e3x .
dx
Here, our function f (x) = −2 (not really a function of x!), so the inte-
grating factor is R
h(x) = e −2dx = e−2x
and so we may write
d
ye−2x = ex .


dx
Integrating, we obtain ye−2x = ex dx = ex + c so the general solution is
R

y = e3x + ce2x .

Example 3.11. Solve

dy
+ y sin x = ecos x
dx
subject to the conditions y = 0 when x = π/2.
Our integrating factor is
R
h(x) = e sin xdx
= e− cos x

so multiply through by e− cos x to get

d
ye− cos x = 1


dx
4 Integration techniques 11

which has general solution y = ecos x (x + c).

We now apply the conditions to find c. Putting in y = 0 and x = π/2:
π 
0 = ecos π/2 +c
2
π
c = −
2
and so the particular solution is y = ecos x (x − π/2).

4 Integration techniques
4.1 Integration by Parts
Suppose we have a function f (x) which is the product of two functions, i.e.
f (x) = u(x)v(x). Then, using the product rule, we may differentiate:
df (x) du(x) dv(x)
= v(x) + u(x)
dx dx dx
or, in shorthand, f (x) = [u(x)v(x)] = u (x)v(x) + u(x)v 0 (x).
0 0 0

If we now integrate both sides of the equation:

Z Z Z
[u(x)v(x)]0 dx = u0 (x)v(x)dx + u(x)v 0 (x)dx

and rearrange to get

Z Z Z
u(x)v (x)dx = [u(x)v(x)] dx + u0 (x)v(x)dx
0 0

Evaluating the first term on the right hand side gives us the standard formula
for integration by parts:
Z Z
u(x)v (x)dx = u(x)v(x) − u0 (x)v(x)dx.
0

Where we are required to evaluate a definite integral, we may use the

form Z b h ib Z b
0
u(x)v (x)dx = u(x)v(x) − u0 (x)v(x)dx.
a a a
Example 4.1. To evaluate xex dx, we let u(x) = x and v 0 (x) = ex . Thus
R

u0 (x) = 1 and v(x) = ex (ignore constant of integration). Using the formula

above, we obtain
Z h i Z
xe dx = xe − ex dx
x x

= xex − ex + C
Note that we only need to include the constant of integration once we
have evaluated all the integrals. It is not strictly necessary to do this, as if
you forget then you will just end up with 2 (or more) constants.
4.2 Reduction Formulas 12

Tricks There are a number of ways Integration by Parts can be extended

to solve more advanced problems.
Example 4.2. 1. Repeated Parts: you may need to apply parts more
than once before the integral can be solved. For example,
Z Z
x2 ex dx = x2 ex − 2xex dx
Z
= x2 ex − 2xex + 2ex dx

= ex (x2 − 2x + 2)

2. Invisible Parts: often we may be able to use parts where we set v 0 (x) =
1, which then integrates to x. For example,
Z Z
1
ln xdx = x ln x − x dx
x
= x ln x − x + C

3. Cyclic Parts: For some integrals (notably those involving hyperbolic

or trigonometric functions) we may carry out repeated parts twice or
more and end up back where we started. This does not always mean
the method is useless, however. For example,
Z Z
e cos xdx = e sin x − ex sin xdx
x x

 Z 
= e sin x − − e cos x + ex cos xdx
x x

Z
= e (sin x + cos x) − ex cos xdx
x

ex cos xdx, observe that this equation says

R
But then, if I(x) =

I(x) = ex (cos x + sin x) − I(x)

and therefore 2I(x) = ex (cos x + sin x), in which case we deduce that

ex (cos x + sin x)
Z
ex cos xdx = +C
2

4.2 Reduction Formulas

We can sometimes use integration by parts to obtain reduction formulas for
integrals. These can be used to write an integral involving powers of an
expression in terms of integrals that involve lower powers of the expression.
4.2 Reduction Formulas 13

Example 4.3. Find a reduction formula for sinn xdx.

R

We use integration by parts, setting v 0 = sin x and u = sinn−1 x so that

v = − cos x and u0 = (n − 1) sinn−2 x cos x. Thus
Z Z
sinn xdx = − cos x sinn−1 x + (n − 1) sinn−2 x cos2 xdx

and since cos2 x = 1 − sin2 x, we may write

Z Z Z
n n−1 n−2
sin xdx = − cos x sin x + (n − 1) sin xdx − (n − 1) sinn xdx

and so
Z Z Z
n n n−1
sin xdx + (n − 1) sin xdx = − cos x sin x + (n − 1) sinn−2 xdx

If we let Z
In = sinn xdx

then we get the reduction formula:

1 n−1
In = − cos x sinn−1 x + In−2 .
n n
We can now use this to evaluate integrals for a given value of n:
Example 4.4. Find sin4 xdx.
R

Write
1 3
I4 = − cos x sin3 x + I2
4 4
1 1 1
I2 = − cos x sin x + I0
Z2 2
I0 = dx = x + C

and so we get the result,

Z
1 3 3
sin4 xdx = − cos x sin3 x − cos x sin x + x + D.
4 8 8
What about if n is odd? Then we can still use the reduction formula.
Example 4.5. Find sin3 xdx.
R

First write out the terms we need:

1 2
I3 = − cos x sin2 x + I1
Z3 3
I1 = sin xdx = − cos x + C

and so Z
1 2
sin3 xdx = − cos x sin2 x − cos x + D.
3 3
4.3 Trigonometric and Hyperbolic Substitutions 14

4.3 Trigonometric and Hyperbolic Substitutions

Recall we can often solve integrals by substitution of x for another variable
(we often use u). We can also use trigonometric and hyperbolic functions
as substitutions.
Example 4.6. Evaluate Z
1
√ dx.
4 + x2
We will solve this by making the substitution x = 2 sinh u, so that dx =
2 cosh du. What happens if we do this?
Z Z
1 1
√ dx = p · 2 cosh udu
2
4+x 4 + 4 sinh2 u
Z
cosh udu
= p
1 + sinh2 u
Z
cosh udu
= √
Z cosh2 u
= du = u + C
x
= sinh−1 +C
2
Example 4.7. Evaluate Z
1
√ dx.
4 − x2
This we solve by substituting x = 2 sin u so dx = 2 cos udu:
Z Z
1 2 cos udu
√ dx = p
4−x 2
4 − 4 sin2 u
Z
cos udu
= √
Z cos2 u
= du = u + C
x
= sin−1 +C
2
There are several integrals we can treat in a similar way. See the sheet
on Inverse Hyperbolic and Trigonometric Functions for more examples.