MAT213 / MAT206: Mathematical Methods I Lecture

Notes
EHIGIE, Julius Osato*

Department of Mathematics, University of Lagos, Nigeria

First Order Ordinary Differential Equations

Contents
1 Exact Equations 1
1.1 Method of Solution for Exact Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Integrating Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1 Exact Equations
Definition 1.1 A differential equation

M (x, y)dx + N (x, y)dy = 0 (1)

is said to be an exact differential equation in domain D if there exist a function F of two variables
(x, y) ∈ D such that

∂F (x, y)
= M (x, y) (2)
∂x
and
∂F (x, y)
= N (x, y) (3)
∂y
for all (x, y) ∈ D.

Theorem 1.1 If M and N have continuous first partial derivatives at all points (x, y) ∈ D, then the
differential equation
M (x, y)dx + N (x, y)dy = 0 (4)
is exact if and only if

∂M (x, y) ∂N (x, y)
= (5)
∂y ∂x
* Not to be reproduced without permission from Julius Ehigie ([email protected])

1
Ehigie, J. O. Mathematical Methods

proof:
First, assume that the expression M (x, y)dx + N (x, y)dy is an exact differential in D, then by definition
of an exact differential, there exist a function F such that

∂F (x, y) ∂F (x, y)
= M (x, y) and = N (x, y)
∂x ∂y
for all (x, y) ∈ D

∂ 2 F (x, y) ∂M (x, y)
=
∂y∂x ∂y
and
∂ 2 F (x, y) ∂N (x, y)
=
∂x∂y ∂x
for all (x, y) ∈ D. Using the continuity of the first partial derivative of M and N , we have

∂ 2 F (x, y) ∂ 2 F (x, y)
=
∂y∂x ∂x∂y
∂M (x, y) ∂N (x, y)
=⇒ =
∂y ∂x
conversely, if equation (5) holds, then we need to show that M dx + N dy = 0 is exact in D that is we
must first prove that there exist a function F such that (2) and (3) are satisfied, Now, from (2)
Z
F (x, y) = M (x, y)∂x + φ(y) (6)
R
where M (x, y)∂x indicates a partial integration with respect to x, holding y constant; and φ is an
arbitrary function of y only. Differentiate (6) partially with respect to y
Z
∂F (x, y) ∂ dφ(y)
= M (x, y)∂x + (7)
∂y ∂y dy
substituting equation (3) into (7) will yield
Z
dφ(y) ∂
= N (x, y) − M (x, y)∂x (8)
dy ∂y
dφ
since φ is a function of y only then will be a function of y only. Hence, RHS of (8), that is,
dy
∂ R
N (x, y) − M (x, y)∂x must be a function of y only.
∂y
Thus, we shall show that  Z 
∂ ∂
N (x, y) − M (x, y)∂x = 0
∂x ∂y
Now,
∂2
 Z  Z
∂ ∂ ∂N (x, y)
N (x, y) − M (x, y)∂x = − M (x, y)∂x
∂x ∂y ∂x ∂x∂y

2
Ehigie, J. O. Mathematical Methods

from equation (2) we know that Z

F (x, y) = M (x, y)∂x (9)

∂2 ∂ 2 F (x, y) ∂ 2 F (x, y) ∂2
Z Z
∴ M (x, y)∂x = =⇒ = M (x, y)∂x (10)
∂x∂y ∂x∂y ∂x∂y ∂x∂y
thus we obtain,
 Z   Z 
∂ ∂ ∂N (x, y) ∂ ∂
N (x, y) − M (x, y)∂x = − M (x, y)∂x
∂x ∂y ∂x ∂y ∂x

as using (9)   Z
∂N (x, y) ∂ ∂F (x, y) ∂N (x, y) ∂
= − = − M (x, y)∂x (11)
∂x ∂y ∂x ∂x ∂y
by assumption of (4), (6) becomes
 Z 
∂ ∂ ∂N (x, y) ∂N (x, y)
= N (x, y) − M (x, y)∂x = − =0 (12)
∂x ∂y ∂x ∂x

hence the proof.

∂M ∂N
Remark 1.1 The condition given in this theorem i.e = is often the useful test for the exactness
∂y ∂x
of a first order ODE.

1.1 Method of Solution for Exact Equation

Now to solve an exact equation (1) given by

M (x, y)dx + N (x, y)dy = 0 (13)

We must have
∂F ∂F
= M and =N
∂x ∂y
i.e
∂F (x, y) ∂F (x, y)
dx + dy = 0
∂x ∂y
or
dF (x, y) = 0 =⇒ F (x, y) = c, (14)
where (14) is the solution (1). There are two ways of solving an exact equation. The first one is by
comparison of F , M , N while the second approach is by grouping. The grouping method is shorter and
faster. The two method are illustrated in the examples below.

Example 1.1 (Comparison Approach) Solve the differential equation

(3x2 + 4xy)dx + (2x2 + 2y)dy = 0 (15)

3
Ehigie, J. O. Mathematical Methods

Solution:
Given the equation
(3x2 + 4xy)dx + (2x2 + 2y)dy = 0 (16)
We obtain the partial derivatives as
∂F (x, y)
= M (x, y) = 3x2 + 4xy (17)
∂x
∂F (x, y)
= N (x, y) = 2x2 + 2y (18)
∂y
Differentiating functions M and N partially w.r.t y and x, respectively yields
∂M ∂N
= 4x =
∂y ∂x
Hence, equation (15) is exact. Therefore from (17)
Z
F (x, y) = M (x, y)∂x + φ(y) (19)
Z
= (3x2 + 4xy)∂x + φ(y) (20)

=⇒ F (x, y) = x3 + 2x2 y + φ(y) (21)

differentiating equation (21) partially with respect to y, we get
∂F (x, y) ∂φ(y)
= 2x2 + = 2x2 + 2y (22)
∂y ∂y

dφ(y)
=⇒ 2x2 + = 2x2 + 2y
dy
dφ(x)
=⇒ = 2y
dx
Integrating both sides yields
φ(y) = y 2 + c0 (23)
Thus, substituting equation (23)in (21) we obtain

F (x, y) = x3 + 2x2 y + y 2 + c0

But the solution of (1) must be in the form

F (x, y) = c1

∴ we can write the solution as

x3 + 2x2 y + y 2 + c0 = c1
=⇒ x3 + 2x2 y + y 2 = c.

Example 1.2 Solve the differential equation

(x − xy 2 )dx + (8y − x2 y)dy = 0 (24)

4
Ehigie, J. O. Mathematical Methods

Solution:
Given the equation
(x − xy 2 )dx + (8y − x2 y)dy = 0 (25)
We obtain the partial derivatives as

∂F (x, y)
= M (x, y) = x − xy 2 (26)
∂x
∂F (x, y)
= N (x, y) = 8y − x2 y (27)
∂y
Differentiating functions M and N partially w.r.t y and x, respectively yields
∂M ∂N
= −2xy =
∂y ∂x
Hence, equation (24) is exact. Therefore from (26)
Z
F (x, y) = M (x, y)∂x + φ(y) (28)
Z
F (x, y) = (x − xy 2 )∂x + φ(y) (29)

x2 x2 y 2
=⇒ F (x, y) = − + φ(y) (30)
2 2
differentiating equation (30) partially with respect to y, we get

∂F (x, y) ∂φ(y)
= −x2 y + = N (x, y) = 8y − x2 y (31)
∂y ∂y

dφ(x)
=⇒ = 8y
dx
Integrating both sides yields
φ(y) = 4y 2 + c0 (32)
Thus, substituting equation (32)in (30) we obtain

x2 x2 y 2
F (x, y) = − + 4y 2 + c0
2 2
But F (x, y) = c1 is the expected solution of the given exact equation, ∴ we can write the solution as

x2 x2 y 2
− + 4y 2 + c0 = c1
2 2
=⇒ x2 − x2 y 2 + 8y 2 = k.

Example 1.3 By grouping method, obtain the solution of the same ordinary differential equation given
in Example 1.1

5
Ehigie, J. O. Mathematical Methods

Solution:
Given
(3x2 + 4xy)dx + (2x2 + 2y)dy = 0, (33)
we shall group (34) into three separate form: the function of x only, the function of y only, and the
function that consists of both x and y together. Thus, we can write equation (34) as follows.

3x2 dx + (4xydy + 2x2 dy) + 2ydy = 0 (34)

But from elementary differentiation, we know that

d(x3 ) = 3x2 dx, d(2x2 y) = 4xydx + 2x2 dy, d(y 2 ) = 2ydy

Thus equation (34) can be written in differential form as

d(x3 ) + d(2x2 y) + d(y 2 ) = d(c)

Integrating both sides yields

x3 + 2x2 y + y 2 = c
which gives the same result as comparison method. Note that equation (34) could have been integrated
directly to get the above result, but the second and the third term in brackets are functions of two variable
and their integrals carry half the value. Hence to integrate equation (34) directly, we can write either of
the two equations below:
Z Z Z Z
2 1 1 2
3x dx + 4xydx + 2x dy + 2ydy = c
2 2
Z Z Z Z
2 2
=⇒ 3x dx + 2xydx + x dy + 2ydy = c

=⇒ x3 + x2 y + x2 y + y 2 = c
Alternately, we can integrate equation (34) also as
Z Z Z
2 2
3x dx + d(2x y) + 2ydy = c

x3 + 2x2 y + y 2 = c
Thus we can obtain the same result in several ways as it has been shown above.

1.2 Integrating Factor

Definition 1.2 If the differential equation

M (x, y)dx + N (x, y)dy = 0 (35)

is not an exact equation in a domain D for all (x, y) ∈ D, but a function µ(x, y) can be found such that
the differential equation
µ(x, y)M (x, y)dx + µ(x, y)N (x, y)dy = 0 (36)
is exact in D, then µ(x, y) is called an Integrating Factor (IF) of the differential equation (35).

6
Ehigie, J. O. Mathematical Methods

On multiplication of µ(x, y) to both sides of (35) makes the resulting equation (36) exact. Thus there
exist a µ(x, y), which can always reduce an inexact differential equation of the form (35) to an exact
differential equation.

In other words, any given first order differential equation which is known to be inexact equation can be
made to be an exact equation by finding a function µ(x, y) as an integrating factor for the differential
equation, yet it is assumed that an integrating factor can be found. Furthermore, since the integrating
factor µ(x, y) is simply used to multiply the original equation in order to make it an exact equation, it
follows that there can be more than one integrating factor that can make a differential equation an exact
equation.

To obtain an integrating factor µ(x, y) for a differential equation, the following statement may be of as-
sistant when the nature of the given differential equation is considered.

Thus for (35), we give the following rules.

∂M ∂N
−
∂y d∂y R
1. If = f (x), the e f (x) dx is an integrating factor of (35).
N
∂M ∂N
−
∂y d∂y R
2. If = −g(y), the e g(y) dy is an integrating factor of (35).
M
3. If (35) is known to be homogeneous and that xM (x, y) + yN (x, y) 6= 0, then, the expression
1
is an integrating factor of the homogeneous equation.
xM (x, y) + yN (x, y)
4. If (35) can be written in the form yf (x, y)dx + xg(x, y)dy = 0, where f (x, y) 6= g(x, y), then the
expression
1 1
=
xy[f (x, y) − g(x, y)] ∂M (x, y) ∂N (x, y)
−
∂x ∂y
is an integrating factor.
Some examples are now given to illustrate the four cases for problem of the form (35) that are inexact.

Example 1.4 Solve the differential equation

(x2 + y 2 + x)dx + xydy = 0 (37)

Solution:
We can check for the exactness of (37)
∂M ∂N
= 2y, =y
∂y ∂x
∂M ∂N
This shows that (37) is not exact since 6= .
∂y ∂x

7
Ehigie, J. O. Mathematical Methods

Group of terms Integrating Factor (µ(x, y)) Exact ODE

1 xdy − ydx
ydx − xdy − 2
x x2
1 ydx − xdy
ydx − xdy
y2 y2

1 xdy − ydx
ydx − xdy −
xy xy

1 xdy − ydx
ydx − xdy −
x2 + y2 x2 + y 2

1 ydx + xdy
ydx + xdy
xy xy

1 ydx + xdy
ydx + xdy
(xy)n (xy)n

1 ydy + xdx
ydy + xdx
x2 + y 2 x2 + y 2

1 ydy + xdx
ydy + xdx , n>1
(x2 + y 2 )n (x2 + y 2 )n

aydx + bxdy, a, b constant xa−1 y b−1 xa−1 y b−1 (aydx + bxdy)

Table 1: Integrating Factor for first order Homogenous equation of degree n = 1

However, since these derivatives are both of y only, we can try the integration factor of the form
∂M ∂N
−
∂y d∂y 2y − y 1
= = = f (x)
N xy x
Hence the integrating factor is
dx
R
R
f (x)dx
IF = e = e x = eln x = x

That is µ(x, y) = x. Hence multiplying (37) by µ(x, y), we get

(x3 + xy 2 + x2 )dx + x2 ydy = 0 (38)

Hence we can check if the new equation (38) is now exact. Thus from (38) we have
∂M ∂N
= 2xy, = 2xy
∂y ∂x

8
Ehigie, J. O. Mathematical Methods

∂M ∂N
which shows that = .
∂y ∂x
Hence, we can solve (38) by grouping method. That is

(x3 + x2 )dx + (xy 2 dx + x2 ydy) = 0

which can be written as differential equations as follows:

 
1 4 1 3 1
d x + x + d( x2 y 2 ) = d(c)
4 3 2
Integrating to get
1 4 1 3 1 2 2
x + x + x y = c.
4 3 2
Example 1.5 Obtain an integrating factor for the given differential equation

2xdx + xydy = 0 (39)

and solve completely.

Solution:
Given the differential equation (39), We can check for the exactness by similarly taking the following
partial derivatives
∂M ∂N
= 2x, = 6x
∂y ∂x
That is (39) is inexact.

Hence, we try the integration factor of the form

∂M ∂N
−
∂y ∂y 2x − 6x 1
= = − = −g(y)
M 2xy y
This suggest that the integrating factor has the form

R 2
dy
= e y = y2
R
g(y)dx
µ(x, y) = e

Hence multiplying (39) by µ(x, y), we get

2xy 3 dx + (4y 3 + 3x2 y 2 )dy = 0 (40)

Thus checking if the new equation (40) is now exact. We take partial derivatives
∂M ∂N
= 6xy 2 , = 6xy 2
∂y ∂x
∂M ∂N
which shows that = and (40) is now exact.
∂y ∂x

9
Ehigie, J. O. Mathematical Methods

Thus we can solve (40) by grouping method. That is

(2xy 3 dx + 3x2 y 2 dy) + 4y 3 dy = 0
which can be written as differential equations as follows:
d(x2 y 3 ) + d(y 4 ) = d(c)
Integrating to get
=⇒ x2 y 3 + y 4 = c.
Example 1.6 We shall similarly solve the differential equation
(x4 + y 4 )dx − xy 3 dy = 0 (41)
Solution:
We observe that the given equation (41) is homogeneous, since it can be written in the form
dy y 3 y  y −3 y y
= 2+ = + =f
dx x x x x x
Thus we can solve this equation using the conventional substitution y = vx as was earlier demonstrated
in Section ??. However, we can also obtain an integrating factor to reduce the equation to exact form. By
the statement above an appropriate integrating factor could be of the form
1 1
µ(x, y) = =
xM (x, y) + yN (x, y) x(x4 + y 4 ) + y(−xy 3 )
1
=⇒ µ(x, y) =
x5
the reader should note that given homogeneous equation, (41) is not exact, since
∂M ∂N
= 4y 3 , = y3
∂y ∂x
Thus (41) is not exact.

Hence, multiplying (41) by the integrating factor µ(x, y), we have a exact equation
1 y4 y3
 
+ 5 dx − 4 dy = 0 (42)
x x x
∂M ∂N 4y 3
such that = = 5.
∂y ∂x x
Thus, we can solve (42) by grouping method as follows
 4
y3

1 y
dx + dx − 4 dy
x x5 x
which can be written as
1
d(ln x) − d(y 4 x−4 ) = d(c)
4
on integrating, we obtain
1
ln x − y 4 x−4 = c
4
or
y 4 = x4 (4 ln +c).

10
Ehigie, J. O. Mathematical Methods

Example 1.7 Solve the differential equation

(2y + x2 y 3 )dx + (2x − 2x3 y 2 )dy = 0 (43)

Solution:
The given differential equation can be written in the form

yf (x, y)dx + xg(x, y)dy = 0

and since f (x, y) 6= g(x, y) then the integrating factor will be of the form
1 1
µ(x, y) = =
xM (x, y) − yN (x, y) 2xy + x y − 2xy + 2x3 y 3
3 3

1
=⇒ µ(x, y) = .
3x2 y 3
Hence, multiplying both sides of (43) by µ(x, y), we obtain
   
2 1 2 2
+ dx + − dy = 0 (44)
3x3 y 2 3x 3x2 y 3 3y

(44) is now exact since

∂M ∂N 1
= =− 3 3
∂y ∂x 3x y
Thus we can solve (44) by grouping
 
1 2 2 2
dx + 3 2
dx + 2 3 dy − dy = 0
3x 3x y 3x y 3y
or  
1 1 1 2
d(ln x) − d − d(ln y) = d(c)
3 3 x y2
2 3
Integrating the implicit function yields
1 1 2
ln x − 2 2 − ln y = c.
3 3x y 3

References
[1] BRONSON, R. (2003). Schaum’s easy outline of differential equations (Crash Course) : [includes
fully solved problems for every topic ; expert tips for mastering differential equations ; all you need
to know to pass the course] . New York, McGraw-Hill.

[2] BRONSON, R., COSTA, G. B. (2006). Schaum’s outline of differential equations: [563 fully solved
problems ; concise explanations of all course concepts ; cover first-order, second-order, and nth-
order equations ...]. New York, McGraw-Hill.

[3] Okunuga S. A. (2003). Elementary Mathematical Methods. WIM Publications.

11