1 Differential operator (2.

3)
In operational calculus, an operator is a transformation that transforms a function into another func-
tion. Hence, differential calculus involves an operator, the differential operator D, which transforms a
(differentiable) function into its derivative. In operator notation we write:

dy
Dy (x) = y 0 (x) =
dx
.
Double D allows to obtain the second derivative of the function y (x):

D2 y (x) = D (Dy (x)) = Dy 0 (x) = y 00 (x) .

Similarly, the nth power of D leads to the nth derivative:

Dn y (x) = y (n) (x)

. Here we assume that the function y (x) is n times differentiable.

So, a homogeneous linear ODE of second order of the form y 00 + ay 0 + by = 0 can be represented as

D2 y + aDy + by = 0
⇒(D2 + aD + bI)y = 0, where I is the identity operator defined as Iy = y
⇒P (D)y = 0, where P (D) = D2 + aD + bI

We call P (D) a second-order differential operator.

Example 1 a. (D2 − 3D)(e2x ) = D2 (e2x ) − 3D(e2x ) = 4e2x − 6e2x = −2e2x .

b. (D + 11I)(x sin x) = D(x sin x) + 11I(x sin x) = x cos x + sin x + 11x sin x = x(cos x + 11 sin x) + sin x.
c. (D − I)(2D + I) sinh 3x = (2D2 − D − I) sinh 3x = 2D2 sinh 3x − D sinh 3x − I sinh 3x = 18 sinh 3x −
3 cosh 3x − sinh 3x = 17 sinh 3x − 3 cosh 3x.

Example 2 Solve (4D2 − I)y = 0 by using factorization of P (D) = 4D2 − I.

Solution: The given equation can be written as

(2D − I)(2D + I)y = 0

Now, we have two different equations

(2D + I)y = 0
and (2D − I)y = 0

The first equation is nothing but a first order linear ODE. The solution is y = e−0.5x (verify!!!). Simi-
larly, for the second equation the solution is y = e0.5x (verify!!!)
Since the solutions are linearly independent, the general solution is

y = c1 e−0.5x + c2 e0.5x .

Note1: If the factors of P (D) are repeated then the above method is not that much useful.
Note2: If it is only the matter to solve the problem, then one can use the techniques of Section 2.2.

1
 2 Euler–Cauchy Equations (2.5)
Euler–Cauchy equations are ODEs of the form
x2 y 00 + axy 0 + by = 0 (1)
with given constants a and b and unknown function y(x). To find the general solution of it, we naturally
choose a solution of the form
y = xm . (2)
Substituting (2) in equation (1) we get the auxiliary equation
m(m − 1) + am + b = 0.
or,
m2 + (a − 1)m + b = 0. (3)
Let m1 and m2 be the roots of (3). There are three different cases on which the general solution of (1)
depends and which are as follows
Case I: (m1 and m2 are real and distinct) General solution is
y = c1 xm1 + c2 xm2

Case II: (m1 and m2 are real and equal) General solution is
y = (c1 + c2 ln x)xm1

Case III: (m1 and m2 are complex) Since complex roots occur in conjugate pair, we consider m1 = p + ιq and
m2 = p − ιq, where p and q are real numbers. Then the general solution is
y = xp (c1 cos(q ln x) + c2 sin(q ln x))

Example 3 Solve (x2 D2 − 4xD + 6I)y = 0

Solution: Let y = xm be a trial solution. Then the auxiliary equation is

m(m − 1) − 4m + 6 = 0
⇒m2 − 5m + 6 = 0
⇒(m − 2)(m − 3) = 0
⇒m = 2, 3
Since the roots are real and distinct, by Case I we have the general solution
y = c1 x2 + c2 x3 ,
where c1 , c2 are arbitrary constants.
Example 4 Solve (x2 D2 − 3xD + 4I)y = 0

Solution: Let y = xm be a trial solution. Then the auxiliary equation is

m(m − 1) − 3m + 4 = 0
⇒m2 − 4m + 4 = 0
⇒(m − 2)2 = 0
⇒m = 2, 2
Since the roots are real and equal, by Case II we have the general solution
y = (c1 + c2 ln x)x2 ,
where c1 , c2 are arbitrary constants.

2
3 Existence and Uniqueness of Solutions. Wronskian (2.6)
In this section we shall discuss the general theory of homogeneous linear ODEs

y 00 + p(x)y 0 + q(x)y = 0 (4)

with continuous, but otherwise arbitrary, variable coefficients p(x) and q(x). This will concern the existence
and form of a general solution of (4) as well as the uniqueness of the solution of initial value problems
consisting of such an ODE with two initial conditions

y(x0 ) = K0 , y 0 (x0 ) = K1 (5)

with given x0 , K0 and K1 .

Theorem 1 (Existence and Uniqueness Theorem for Initial Value Problems) If p(x) and q(x) are
continuous functions on some open interval I and x0 is in I, then the initial value problem consisting of (4)
and (5) has a unique solution on the interval I.

3.1 Linear Independence of Solutions

Let y1 and y2 be two solutions of (4) on an open interval I. We call y1 and y2 linearly independent on I
if the equation
c1 y1 + c2 y2 = 0 on I has the only solution c1 = 0 and c2 = 0
We call y1 , y2 linearly dependent on I if this equation also holds for constants c1 , c2 not both 0.
Also, in another way we can say if there exists some constant k 6= 0 for which y2 = ky1 , then y1 , y2 are
linearly dependent.
Note: If solutions y1 , y2 are linearly independent, then the set {y1 , y2 } is called basis of solutions of
(4).

Theorem 2 (Wronskian) Let the ODE (4) have continuous coefficients p(x) and q(x) on an open interval
I. Then two solutions y1 and y2 of (4) on I are linearly dependent on I if and only if their “Wronskian”

y1 y2
W (y1 , y2 ) = = y1 y20 − y10 y2
y10 y20

is 0 at some x0 in I. Furthermore, if W (y1 , y2 ) = 0 at an x = x0 in I, then W (y1 , y2 ) = 0 on I; hence, if

there is an x1 in I at which W (y1 , y2 ) is not 0, then y1 , y2 are linearly independent on I.

3.2 A general solution of (4)

Theorem 3 (Existence of a General Solution) If p(x) and q(x) are continuous on an open interval I,
then (4) has a general solution on I.

Theorem 4 If the ODE (4) has continuous coefficients p(x) and q(x) on some open interval I, then every
solution of (4) on I is of the form
y = c1 y1 + c2 y2
where {y1 , y2 } is any basis of solutions of (4) on I and c1 , c2 are suitable constants.
1
x
Example 5 We see that W (x, x1 ) = x = − x1 − 1
= − x2 6= 0. So, the set {x, x1 } is linearly indepen-
1 − x12 x

dent.
2x 3x
Example 6 We see that W (2x, 3x) = = 0. So, the set {2x, 3x} is linearly dependent.
 2 3
2
Note: One can check that 2x = 3x. It also implies that the set {2x, 3x} is linearly dependent.
3