Mathematics-II (MA1201)

Ordinary Differential Equations Lect 10 & 11

Ratikanta Behera

Second Order Linear Differential Equations

Theorem 1. Let functions p1 and p2 in

L(y)(x) = y 00 (x) + p1 (x)y 0 (x) + p2 (x)y(x) = 0, x ∈ I (1)

be defined and continuous on an interval I. Let φ1 and φ2 be two linearly independent

solutions and x0 ∈ I. Then
 Z x 
W (φ1 , φ2 )(x) = exp − p1 (x)dx W (φ1 , φ2 )(x0 ) (2)
x0

Proof. Given y 00 (x) + p1 (x)y 0 (x) + p2 (x)y(x) = 0, x ∈ I

Now the Wronskian of the two solution:
!
φ1 φ2
W (φ1 , φ2 ) = 0 0
= φ1 φ02 − φ01 φ2 (3)
φ1 φ2

Now
W 0 (φ1 , φ2 ) = (φ01 φ02 + φ1 φ002 − φ001 φ2 − φ01 φ02 ) = φ1 φ002 − φ001 φ2 (4)
Since φ1 and φ2 satisfy (1), we have

φ001 + p1 φ01 + p2 φ1 = 0 ⇒ φ001 = −p1 φ01 − p2 φ1

φ002 + p1 φ02 + p2 φ2 = 0 ⇒ φ002 = −p1 φ02 − p2 φ2

Putting the above value of φ001 and φ002 in (4), we get

W 0 (φ1 , φ2 ) = φ1 (−p1 φ02 − p2 φ2 ) − φ2 (−p1 φ01 − p2 φ1 )

W 0 (φ1 , φ2 ) = −p1 (φ1 φ02 − φ2 φ01 ) = −p1 W (φ1 , φ2 )

Thus, W (φ1 , φ2 ) satisfies a first order linear equation W 0 + p1 W = 0, x ∈ I

Z x
dW dW
= −p1 W ⇒ = −p1 dx ⇒ log W − log c = − p1 dx
dx W x0
 Z x 
W (φ1 , φ2 )(x) = c exp − p1 (x)dx (5)
x0

where c is a constant. By putting x = x0 in (5), we get c = W (φ1 , φ2 )(x0 ). Now using the
value of c in equation (5) we get required result.

Please go on to the next page. . .

Mathematics - II Lect. 10 & 11 Page 2 of 6

Find basis if one solution is know.

Consider homogeneous linear differential equation

y 00 + p(x)y 0 + q(x)y = 0 (6)

We assume a solution y1 of (6) on an open interval I, to be know. To find the basis we need
a second linearly independent solution y2 on I.
Consider y2 = y1 z, then y20 = z 0 y1 + zy10 and y200 = z 00 y1 + 2z 0 y10 + zy100 .
Putting the above value in the given equation (6), we get

z 00 y1 + 2z 0 y10 + zy100 + p(z 0 y1 + zy10 ) + qzy1 = 0 (7)

Collecting the term z 00 , z 0 and z we have

z 00 y1 + z 0 (2y10 + py1 ) + z(y100 + py10 + qy1 ) = 0 (8)

Since y1 is a solution then y100 + py10 + qy1 = 0, further, consider U = z 0 and U 0 = z 00

Now the equation (17) becomes
0
 0 
00 0 2y1 + py1 0 2y1
z +z =0 ⇒U + +p U =0 (9)
y1 y1
This is first order ODE of separation of variables. Now integrating gives
 0  Z
dU 2y1
=− + p(x) ⇒ log |U | = −2 log |y1 | − p(x)dx (10)
U y1
R
Hence we get U = y12 e p(x)dx . But U = z 0 , so that z = U dx. Now the second solution is
R
2
Z
y2 = y1 z = y1 U dx (11)

Here the quotient yy12 = z = U dx can not be constant (U > 0), so that y1 and y2 from a
R

basis of solution.
Example-1 If y1 = x is a solution of (x2 − x)y 00 − xy 0 + y = 0, then find the basis the ODE.
Ans. To find the basis we need a second linearly independent solution y2 .
Consider y2 = zx, then y20 = z 0 x + z and y200 = z 00 x + 2z 0 .
Putting the above value in the given equation, we get

(x2 − x)(z 00 x + 2z 0 ) − x2 z 0 = 0 ⇒ (x2 − x)z 00 + (x − 2)z 0 = 0 (12)

This ODE is of first order in v = z 0 , i.e., (x2 − x)v 0 + (x − 2)z = 0. Hence separation of
variables and integration gives
 
dv x−2 1 2 |x − 1|
=− 2 dx = − ⇒ ln |v| = ln |x − 1| − 2 ln |x| = ln
v x −x x−1 x x2

Please go on to the next page. . .

 Mathematics - II Lect. 10 & 11 Page 3 of 6

Since we don’t need constant (we want to obtain particular solution) so v = x1 − x12 . Now
z = vdx = ln |x| + x1 . Hence y2 = zx = x ln |x| + 1
R

Since y1 and y2 are LI (their quotient is not constant), we obtained required basis of solutions.

The following properties helps us to find a solution of (1)

Properties 1. Consider the linear homogeneous equation

y 00 (x) + p1 (x)y 0 (x) + p2 (x)y(x) = 0, x ∈ I has

(i) y = y1 (x) = x a solution if and only if p1 + p2 x = 0

(ii) y = y1 (x) = x2 a solution if and only if 2 + 2xp1 + x2 p2 = 0

(iii) y = y1 (x) = eax a solution if and only if a2 + ap1 + p2 = 0

Homogeneous Linear equation with constant coefficients

The equation
dn y dn−1 y
+ p 1 + · · · + pn y = 0 (13)
dxn dxn−1
where p1 , p2 , · · · , pn are constants is called linear nth order homogeneous differential equation
with constant coefficients. If y = emx is a solution of (13), then

mn emx + p1 mn−1 emx + · · · + pn emx = 0

Canceling the non-vanishing factor emx , we get

mn + p1 mn−1 + · · · , pn = 0 (14)

Which is called the characteristic equation or the auxiliary equation of the given
differential equation (13). Hence to solve (13) we write the auxiliary equation (14) and solve
it for m. this gives rise to three different cases

Case-1: Distinct real roots

Suppose the roots of the auxillary equations of (14) are the n distinct real numbers
m1 , m2 , · · · , mn . Then em1 x , em2 x , · · · , emn x are n distinct solutions (13) and there are
LI on any interval [a, b]. Hence the general solution is

y = c1 em1 x + c2 em2 x + · · · + cn emn x

where c1 , c2 , · · · , cn are arbitrary constants.

Please go on to the next page. . .

 Mathematics - II Lect. 10 & 11 Page 4 of 6

Example-1 Solve
d2 y dy
2
− − 6y = 0
dx dx
The auxiliary equation is

m2 − m − 6 = 0,
⇒ (m − 3)(m − 2) = 0,

The roots are 3 and −2 which are real and distinct. Hence the general solution is

y(x) = c1 e3x + c2 e−2x

Here we also verify that the solution e3x and e−2x are LI.
!
3x −2x
e e
W (e3x , e−2x ) = = −5ex 6= 0.
3e3x −2e−2x

Case-2: Repeated real roots

Consider seconder-order homogeneous linear ODEs whose coefficients a and b are
constant,
y 00 + ay 0 + by = 0 (15)
Suppose the roots of the auxiliary equation are equal, m = m1 = m2 = −a/2, Hence
only one solution,
y1 = emx = e−(ax/2) (16)
To obtain a second solution y2 ,
we consider y2 = y1 z, then y20 = z 0 y1 + zy10 and y200 = z 00 y1 + 2z 0 y10 + zy100 .
Putting the above value in the given equation (15), we get

z 00 y1 + 2z 0 y10 + zy100 + a(z 0 y1 + zy10 ) + bzy1 = 0

Collecting the term z 00 , z 0 and z we have

z 00 y1 + z 0 (2y10 + ay1 ) + z(y100 + ay10 + by1 ) = 0 (17)

Since y1 is a solution then y100 +py10 +qy1 = 0 and 2y1 +ay1 = 0, as 2y10 = −aemx = −ay1
Now equation (17) becomes z 00 y1 = 0. Hence z 00 = 0. and z = c1 x + c2 .
Thus the second solution y2 = zy1 , we can simple choose c1 = 1, c2 = 0 and get z = x.
Now our solutions are y1 = x and get y2 = xy1 = xemx = xe−ax/2
Hence the general solution is

y = (c1 + c2 x)emx = (c1 + c2 x)e−ax/2 (18)

Please go on to the next page. . .

 Mathematics - II Lect. 10 & 11 Page 5 of 6

where c1 , c2 , · · · , cn are arbitrary constants.

Example-2 Solve
d3 y d2 y
− 3 + 4y = 0
dx3 dx2
The auxiliary equation is m3 − 3m2 + 4 = 0 and its roots are 2, 2, −1.
Hence the general solution is

y(x) = (c1 + c2 x)e2x + c3 3ex

Case-2: Complex conjugate roots

Theorem 2. If the auxilary equation of the differential equation

y 00 + p1 y 0 + p2 y = 0 (19)

has the complex roots α ± β, then the general solution of the differential equation (19)
is
y = eαx (A cos βx + B sin βx)

Proof. Let us consider m1 = α + iβ and m2 = α − iβ the corresponding part of the

general solution of (19) is

y = c1 em1 x + c2 em2 x
= c1 e(α+iβ)x + c2 e(α−iβ)x
= c1 eαx · eiβx + c2 eαx · e−iβx
= eαx (c1 eiβx + c2 e−iβx )
= eαx [c1 (cos βx + i sin βx) + c2 (cos βx − i sin βx)]
= eαx [(c1 + c2 ) cos βx + i(c! − c2 ) sin βx]

Consider c1 + c2 = A and i(c1 − c2 ) = B then the above equation becomes

y = eαx (A cos βx + B sin βx).

Remark 1. Suppose m1 , m2 are conjugate complex, m3 = m4 and the other roots of

the auxiliary equation are distinct and different, then the general solution is

y = eαx (A cos βx + B sin βx) + (c3 + c4 x)em3 x + c5 em5 x + c6 em6 x + · · · + cn emn x (20)

Remark 2. If a pair of complex conjugate complex roots occurs twice, then the corresponding
solution is
y(x) = eαx [(A + A1 x) cos βx + (B + B1 x) sin βx] (21)

Please go on to the next page. . .

Mathematics - II Lect. 10 & 11 Page 6 of 6

Example-3 Solve the IVP

d2 y dy
2
− 4 + 5y = 0, y(0) = 1, y 0 (0) = 0
dx dx
Ans. The auxiliary equation is m2 + 4m + 5 = 0 and its roots are −2 ± i.
Hence the general solution is

y(x) = e−2x (c1 cos x + c2 sin x)

Now y 0 (x) = e−2x [(c2 − 2c1 ) cos x − (2c2 + c1 ) sin x].

Using the initial conditions, we get c1 = 1 and c2 = 2. Hence the required solution is

y(x) = e−2x (cos x + 2 sin x)