Mathematics and Statistics Department

Math 375 Homework Problems On Fall 2023

Higher Order Linear Differential Equations

Doing homework problems is the best way to master the course’s material. To benefit most, resist the urge of looking up
the solution of a problem, until you have completely solved it or seriously attempted it.
The problems on this worksheet refer to material from chapter 2 of the lecture notes. Please report any typos, omissions
and errors to aiffam@ucalgary.ca

1
MATH 375 Higher Order Linear Differential Equations Fall 2023 2

Basics
2.01. Which of the following is a linear differential equation

0
a. y 00 − 5 t y = t y 0 − 25 b. sin(t) y 0 + 2 t y = 0
0
sin(t) y 00 + cos(t) y + 2 t y = 0 d. y 00 + t y 0 − y −1 = ln(t)


c.
y0 + t y 1
e. 00
= et f. y 00 + (2 t + 1) y 0 + y 2 = ln(t)
1+y t

2.02. Given that C1 cos(2 t) + C

2 sin(2 t) is the general solution of y 00 + 4 y = 0, solve
 y 00 + 4 y = 0
the initial value problem √
 y(π/2) = −1 & y 0 (π/2) = 2 3


Express your answer in the form y = A cos 2 t + φ

Given that C1 cos ω t + C2 sin ω t) is the general solution of y 00 + ω 2 y = 0,

2.03.
where
 ω > 0 is a constant, show that the solution of the initial value problem
00
 y + ω2 y = 0
y

is given by y = y0 cos ω (t−t0 ) + 1 sin ω (t−t0 )
0 ω



 y t =y , y t =y
0 0 0 1

Existence and Uniqueness

2.04. Determine the largest open interval over which the unique solution of the given
initial value problem is guaranteed to exist.

 t y 00 + 1 y 0 + y = t  t y 00 + 1 y 0 + y = t
 
t 2−9 t2 − 9
a. b.
 y(1) = 0 & y 0 (1) = 2  y(4) = 0 & y 0 (4) = 2


 y 00 + t y 0 + ln(1 − t) y = ln(2 + t)

 y 00 + y 0 + y = sec(t) 
c. d.
 y(π/4) = 1 & y 0 (π/4) = −1  y(0) = 1 & y 0 (0) = 2



 y 000 +t 1
y 00 + |t − 5| y 0 − ln t2 − 4 y =


e. t + 5 t − 1
 y(−3) = 1, y 0 (−3) = −4 & y 00 (−3) = 2

 y 000 + t y 00 + |t − 5| y 0 − ln t2 − 4 y = 1


f. t+5 t−1
 y(3) = 1, y 0 (3) = −4 & y 00 (3) = 2

2.05. Consider the differential equation t2 y 00 − 3 t y 0 − 5 y = 0. Verify that y1 = t5 and

1
y2 = are solutions of the equation, then solve the initial value problem
t
( 2 00
t y − 3 t y0 − 5 y = 0
y(1) = 4 & y 0 (1) = 2
MATH 375 Higher Order Linear Differential Equations Fall 2023 3

Second Order Homogeneous With Constant Coefficients

2.06. Solve the following second order linear differential equations.

a. 2 y 00 − y 0 − 3 y = 0 b. 9 y 00 − 12 y 0 + 4 y = 0 c. y 00 − 2 y 0 + 5 y = 0
2.07. Solve each of the following.

a. 3 y 00 − 2 y 0 − 8 y = 0 b. 4 y 00 + 4 y 0 + y = 0 c. 4 y 00 − 12 y 0 + 25 y = 0

2.08. Given k > 0 a constant, find the general solution of each of
the following
a. y 00 − 2 k y 0 + 2 k 2 y = 0 b. y 00 + 2 k y 0 + k 2 + 1 y = 0

2.09. A second order linear homogeneous differential equation with constant coefficients
has the given function as a solution, write down the normal form of the equation.

a. 2 e−5 t + 3 e−2 t b. 5 e−2 t − 2 c. 2 e3 t cos(2 t)

Second Order Homogeneous Cauchy-Euler Equations

2.10. Find the general solution of each of the following Cauchy-Euler equation. Assume
t > 0.

a. 2 t2 y 00 + 5 t y 0 + y = 0 b. 3 t2 y 00 + 13 t y 0 + 3 y = 0
c. t2 y 00 + 5 t y 0 + 4 y = 0 d. 4 t2 y 00 − 8 t y 0 + 9 y = 0
e. 4 t2 y 00 + 16 t y 0 + 13 y = 0 f. t2 y 00 + 7 t y 0 + 25 y = 0

2.11. Solve 
the initial value problems
t2 y 00 − 4 t y 0 + 4 y = 0 t2 y 00 + 7 t y 0 + 5 y = 0

a. b.
y(1) = −2, y 0 (1) = −11 y(1) = −1, y 0 (1) = 13

2.12. Find a second order homogeneous Cauchy-Euler differential equation in each of

the following cases.

t4 − 3
a. is a solution the differential equation.
t2
ln(t)
b. is a solution the differential equation.
t3

sin ln(t)
c. is a solution the differential equation.
t

Higher Order Homogeneous With Constant Coefficients

2.13. Find the general solution of the following differential equations

a. y 000 + 3 y 00 − 16 y 0 − 48 y = 0 b. y 000 − 5 y 00 + 4 y 0 − 20 y = 0
c. y (4) − 4 y 000 + 14 y 00 − 20 y 0 + 25 y = 0 d. y (4) − 8 y 00 + 16 y = 0
MATH 375 Higher Order Linear Differential Equations Fall 2023 4

2.14. Find a fundamental set of solutions for the homogeneous linear differential
equations
a. y 000 + y 00 + y 0 + y = 0 b. y 000 − y 00 − y 0 + y = 0
c. y (4) + 6 y 00 + 9 y = 0 d. y 000 + 6 y 00 + 12 y 0 + 8 y = 0

2.15. Write down, in normal form, the linear homogeneous constant coefficients
differential equation associated with the given fundamental
 t t solutiont set.
2t

a. 1, t, e b. e , e cos(t) , e sin(t)
c. { cos(t) , sin(t) , t cos(t) , t sin(t) } d. { cosh(t) , sinh(t) }

2.16. Solve the following initial value problems 

 000  y (4) + 8 y 00 + 16 y = 0
y + y 00 − 9 y 0 − 9 y = 0
a. b. y(0) = 3, y 0 (0) = 2
y(0) = y 0 (0) = y 00 (0) = −3  00
y (0) = 1, y 000 (0) = 2

2.17. A sixth order linear

homogeneous differential equation with constant coefficients has
t 2 + 3 cos(3 t) as a solution. Find the normal form of the equation.
2.18 Find the general solution of y 000 − y 00 − 4 y 0 − 6 y = 0, given that e−x cos(x) is a
solution of the differential equation.

The Wronskian
2.19. Compute the wronskian W (t) of each of the following set of functions.
n o n o
a. er t , t er t b. er t cos(s t), er t sin(s t)
Assume that r, s 6= 0, are real constants.
2.20. Write Abel’s formula for each of the following equations.
a. 3 y 00 − 2 y 0 + t y = 0 b. t y 00 + (5 t − 2) y 0 + t et y = 0

2.21. Let y1 (t) and y2 (t) be two solutions of y 00 + p(t) y 0 + q(t) y = 0, in (a, b), and
y1 (t) y2 (t)
let W (t) = be their wronskian
y10 (t) y20 (t)
a. Show that W 0 (t) + p(t) WR (t) = 0
b. Show that W (t) = K e− p(t) dt , where K is a real constant.

Consider the linear homogeneous differential equation y 00 + t2 + 1 y 0 − 2 y = 0

2.22.
Suppose that y1 (t) and y2 (t), are solutions of the differential equation, and let
W (t) be their wronskian. Given that W (1) = 2, use Abel’s formula to evaluate
W (2).

2.23. The 2nd order linear differential equation y 00 + p(t) y 0 + q(t) y = 0, has solutions
1
y1 (t) = and y2 (t) = et . Find p(t) and q(t).
t
MATH 375 Higher Order Linear Differential Equations Fall 2023 5

2.24. Let y1 (t) and y2 (t) be a fundamental set of solutions for the 2nd order linear
differential equation y 00 + p(t) y 0 + q(t) y = 0. Express p(t) and q(t) in terms of
y1 (t) and y2 (t).

2.25. Let u(t) and v(t) be differentiable in (a, b). Suppose u(0) = 1, u0 (0) = 2,
v(0) = 2 and v 0 (0) = 5. Show that u(t) and v(t) are linearly independent in
(a, b).

Reduction of order

Consider the 2nd order linear differential equation y 00 − 4 t y 0 + 4 t2 − 2 y = 0

2.26.
2
a. Verify that y1 (t) = et is a solution.
b. Use the reduction of order method to find the general solution.

Consider the 2nd order linear differential equation t2 1 − ln(t) y 00 + t y 0 − y = 0

2.27.
a. Verify that y1 (t) = t is a solution
b. Use the reduction of order method to find the general solution.

2.28. Given a solution y1 (t) of the associated homogeneous differential equation, use
the reduction of order method to find the general solution.
a. y1 (t) = et , t y 00 − (t + 1) y 0 + y = t2 e2 t
b. y1 (t) = t, (1 − t) y 00 + t y 0 − y = (1 − t)2

2.29. Given a solution y1 (t) of the differential equation, use the reduction of order
method to find the general solution.
sin(t) 2
a. y1 (t) = , y 00 + y 0 + y = 0
t t  
sin(t) 2 00 0 2 1
b. y1 (t) = √ , t y + t y + t − y=0
t 4

The Undetermined Coefficients Method

2.30. Suppose the method of undetermined coefficients is used to find a particular
solution yp (t). Write down the form of yp (t) in each of the following cases. Do
not compute yp (t).
a. 4 y 00 − 3 y 0 = t e3 t/4 b. y 00 + y = sin(t) − cos(t)
2
c. y 00 − 3 y 0 + 2 y = t + et d. y 00 + 16 y = sin 4 t + π/3

2.31. Find the general solution of the differential equations.

a. y 00 − 4 y 0 + 4 y = t2 b. y 00 + 4 y 0 + 4 y = 8 e−2 t
c. y 00 + y = 4 t cos(t) d. y 00 − 3 y 0 + 2 y = t2 + t e3 t


MATH 375 Higher Order Linear Differential Equations Fall 2023 6

2.32. Solve 
the following initial value problems.
y 00 + y 0 = e−t
 (4)
y − y = 8 et
a. b.
y(0) = 1, y 0 (0) = −1 y(0) = −1, y 0 (0) = y 000 (0) = 0, y 00 (0) = 1

2.33. Suppose the method of undetermined coefficients is used
2 to find a particular solu-
tion yp (t) for y 000 − 4 y 00 + 6 y 0 − 4 y = et + 2 sin(t) . Write down the form of
yp (t).

2.34. Determine whether or not the undetermined coefficients method can be used to
find a particular solution to the given differential equation.
a. y 00 − 4 y 0 + 4 y = cos2 (3 t) b. y 00 + 4 y 0 + 4 t y = t + e−2 t
00
c. y + y = 4 t cosh(t) d. 2 y 00 + y 0 − 3 y = 4 t sin(3 t) cos(5 t)
t
e. y 000 − 3 y 0 − 2 y = t f. y (4) − 3 y 00 + 2 y = cosh(3 t) sin(t)
e

The Variation of Parameters Method

2.35. Find the general solution of the following equations.
et
a. y 00 + y = sec3 (t) b. y 00 − 2 y 0 + y =
t2 + 1

2.36. Find a particular solution of

1 2t + 1
a. y 00 − y = t
b. t2 y 00 − 2 y =
1+e (t + 1)3
Hint: For question b., to find a fundamental set of solutions for the homogeneous
equation, try solutions in the form tr .

2.37. Use the variation of parameters method to find a particular solution to each of
the following Cauchy-Euler equations.

a. t2 y 00 + t y 0 + 9 y = − tan 3 ln(t) b. t2 y 00 + 3 t y 0 + y = t−1

 
c. t2 y 00 − t y 0 + y = t t + 3/ ln(t)

2.38. Find a particular solution yp (t) of the differential equation

t−1 π
a. y 000 + y 00 = 2 b. y 000 + y 0 = tan(t), 0<t<
t 2

Boundary Value Problems

2.39. Solve each of the following boundary value problems

a. y 00 + y = 0, y(0) = 0, y(π) = 0
b. y 00 + y = 1, y(0) = 0, y(1) = 0
c. y 00 + y = 0, y(0) = 0, y(π) = 1
MATH 375 Higher Order Linear Differential Equations Fall 2023 7

2.40. Solve each of the following boundary value problems or else show that it has no
solution.

a. y 00 + 2 y = 0, y 0 (0) = 1, y 0 (π) = 0
b. y 00 + 2 y = t, y(0) = 0, y(π) = 0
c. y 00 + 4 y = cos(t), y(0) = 0, y(π) = 0
d. t2 y 00 + 3 t y 0 + y = t2 , y(1) = 0, y(e) = 0
e. y 00 − 2 y 0 + y = 0, y(−1) = 0, y(1) + y 0 (1) = 5

2.41. Find all values of the parameter k > 0 for which the following boundary value
problems have a solution that is not identically zero.
a. y 00 + k 2 y = 0, y(0) = 0, y 0 (b) = 0
b. y 00 + k 2 y = 0, y(0) = 0, y(b) = 0
c. y 00 + k 2 y = 0, y 0 (0) = 0, y 0 (b) = 1
d. y 00 + k 2 y = 0, y 0 (0) = 1, y 0 (b) = 0

y 00 + λ2 y = sin(t)

2.42. Show that the boundary value problem has a solution
y(0) = 0, y(π) = 1
if and only if λ 6= ±1, ±2, ±3, · · ·

2.43. Find the solution of y 00 − 4 y = 5 e−3 t , that satisfies the conditions lim y(t) = 0
t→+∞
and y(0) = 3.

2.44. Find the eigenvalues/eigenfunctions of the two-point boundary value problem.

 2 00
t y − t y 0 + λ y = 0, 1 < t < L
a.
y(1) = 0 y(L) = 0
 00
y + λ y = 0, 0 < t < 2 π
b.
y(0) = y(2 π) y 0 (0) = y 0 (2 π)
 00
y + λ y = 0, 0 < t < π
c.
y(0) − y 0 (0) = 0 y(π) = 0
 00
y − 2 y 0 + λ y = 0, 0 < t < π
d.
y(0) = 0 y(π) = 0
 00
y + λ y = 0, −L < t < L
e.
y(−L) = y(L) y 0 (−L) = y 0 (L)
 00
y + λ y = 0, 0 < t < L
f.
y(0) = 0 3 y(L) − y 0 (L) = 0

Solutions
MATH 375 Higher Order Linear Differential Equations Fall 2023 8

Linear Second and Higher Order Differential Equations

The second order differential equation F t, y, y 0 , y 00 = 0, is called linear, if it has or can be

written in the form a2 (t) y 00 + a1 (t) y 0 + a0 (t) y = f (t).

More generally, the nth order differential equation F t, y, y 0 , · · · , y (n) = 0, is called linear, if

it has or can be written in the form an (t) y (n) + an−1 (t) y (n−1) + · · · + a1 (t) y 0 + a0 (t) y = f (t).

2.01. a. Rewriting the equation as y 00 − t y − 5 t y = −25, shows that the differential equation is 2nd
order linear.
b. Rewriting the equation as sin(t) y 00 + cos(t) y 0 + 2 t y = et , shows that the differential is 2nd
order linear.
c. Rewriting the equation as sin(t) y 000 + cos(t) y 00 + cos(t) y 0 + 2 t − sin(t) y = 0, shows that

the differential is 3rd order linear.

1
d. Because of the term y −1 = , the differential equation is not linear.
y
e. Multiplying both side of the equation by 1 + y 00 , moving the y 00 term to the left, we get
− et y 00 + y 0 + t y = et . This shows that the differential equation is 2nd order linear.
1
f. Because of the term y 2 , the differential equation is not linear.
t

2.02. We need to select the constants C1 and C2 so that y(t) = C1 cos(2 t) + C2 sin(2 t), satisfies the
initial conditions. We have y 0 (t) = −2 C1 sin(2 t) + 2 C2 cos(2 t). It follows
  
y(π/2) = −1√ C1 cos(π) + C2 sin(π) = −1 √ C1 = 1 √
0 ⇐⇒ ⇐⇒
y (π/2) = 2 3 −2 C1 sin(π) + 2 C2 cos(π) = 2 3 C2 = − 3
√
√

Hence the solution y(t) = cos(2 t) − 3 sin(2 t) = 1 · cos(2 t) + − 3 · sin(2 t)
q √ √
Multiplying and dividing by the amplitude A = (1)2 + (− 3)2 = 1 + 3 = 2, leads to
1 √
3   

y(t) = 2 cos(2 t) − sin(2 t) = 2 cos(π/3) cos(2 t) − sin(π/3) sin(2 t) = 2 cos 2 t + π/3
2 2
where in the last step, we made use of the identity cos(α) cos(β) − sin(α) sin(β) = cos(α + β)

2.03. Select the constants C1 , C2 so that





y t0
= y0 C1 cos ω t0 +
C2 sin ω t0 = y0

0 ⇐⇒
y t 0 = y1 −C1 ω sin ω t0 + C2 ω cos ω t0 = y1

Using Cramer’s rule to solve the system, leads to

y

y

C1 = y0 cos ω t0 − 1 sin ω t0 C2 = y0 sin ω t0 + 1 cos ω t0
ω ω



Substituting into y = C1 cos ω t + C2 sin ω t and rearranging terms, we get
h



i y h



i
y = y0 cos ω t cos ω t0 + sin ω t sin ω t0 + 1 sin ω t cos ω t0 − cos ω t sin ω t0
ω
or else
y

y = y0 cos ω (t − t0 ) + 1 sin ω (t − t0 )
ω
MATH 375 Higher Order Linear Differential Equations Fall 2023 9

Existence and Uniqueness of the Solution of a Linear Initial Value Problem

If
 p(t), q(t), and r(t) are continuous at and around t = t0 , then the initial value problem
y 00 +
p(t) y 0 + q(t) y
= r(t)
has a unique solution that is defined ( at least ) in the largest
y t 0 = y0 & y 0 t 0 = y1


open interval a , b that contains t0 , and where p(t), q(t), r(t) are all continuous.

Similarly, if p2 (t), p1 (t),

 p0000 (t), and 00 r(t) are 0continuous at and around t = t0 , then the
y +
p2 (t) y + p1 (t)
y + p0 (t) y = r(t)
initial value problem has a unique solution
y t0 = y0 & y 0 t0 = y1 & y 00 t0 = y2




that is defined ( at least ) in the largest open interval a , b that contains t0 , and where
p2 (t), p1 (t), p0 (t), r(t) are all continuous.

1 1
2.04.ab. Rewrite the equation in normal form as y 00 + y0 + y = 1
t (t − 3) (t + 3) t
1




p(t) = is continuous in− ∞ , −3 , − 3 , 0 , 0 , 3 , and 3 , +∞
t (t − 3) (t + 3)
1


q(t) = is continuous in − ∞ , 0 , and 0 , +∞
t

r(t) = 1 is continuous in − ∞ , +∞
a. The largest open interval that
contains
the initial time
t = 1 where

p(t), q(t), and r(t)
are all continuous is 0 , 3 ∩ 0 , +∞ ∩ − ∞ , +∞ = 0 , 3 . By the existence
and
uniqueness theorem, the solution of the I.V.P. is guaranteed to exist in 0 , 3
b. The largest open interval that
contains the
initial time t
= 4 where p(t),

q(t), and r(t)
are all continuous is 3 , +∞ ∩ 0 , +∞ ∩ − ∞ , +∞ = 3 , +∞ . By the existence

and uniqueness theorem, the solution of the I.V.P. is guaranteed to exist in 3 , +∞
2.04.c The equation is already in normal form y 00 + y 0 +
y = sec(t)
p(t) = 1, q(t) = 1 are continuous in − ∞ , , +∞
1



r(t) = sec(t) = is continuous in · · · , −3 π/2 , −π/2 , −π/2 , π/2 , π/2 , 3 π/2 , · · ·
cos(t)
The largest open interval that contains the initial time t = π/4

where p(t),
q(t), and r(t) are
all continuous is − ∞ , +∞) ∩ − ∞ , +∞) ∩ − π/2 , π/2 = − π/2 , π/2 . By the existence

and uniqueness theorem, the solution of the I.V.P. is guaranteed to exist in − π/2 , π/2

2.04.d The equation is already in normal form
y 00 + t y 0 + ln(1 − t) y = ln(2 + t)

p(t) = t, is continuous in − ∞ , +∞

q(t) = ln(1 − t) is continuous in − ∞ , , 1

r(t) = ln(2 + t) is continuous in − 2 , +∞
The largest open interval
that contains

the initial time

t = 0
where p(t), q(t), and r(t) are all
continuous is − ∞ , +∞ ∩ − ∞ , 1 ∩ − 2 , +∞ = − 2 , 1 . By the
existence and uniqueness
theorem, the solution of the I.V.P. is guaranteed to exist in − 2 , 1
MATH 375 Higher Order Linear Differential Equations Fall 2023 10

2.04. The equation is in normal form, with

t


p2 (t) = , is continuous in − ∞ , −5 ∪ − 5 , +∞
t+5

p1 (t) = |t − 5| is continuous in − ∞ , , +∞
p0 (t) = − ln t2 − 4) is continuous in − ∞ , , −2 ∪ 2 , +∞



1


r(t) = is continuous in − ∞ , 1 ∪ 1 , +∞
t−1
e. The largest open interval that contains the
initial time t
= −3 where
p2 (t), p1 (t),
p0 (t),
and r(t)
are all continuous is − 5 , +∞ ∩ − ∞ , +∞ ∩ − ∞ , −2 ∩ − ∞ , 1 = −
5 , −2 . By the existence
and uniqueness theorem, the solution of the I.V.P. is guaranteed
to exist in − 5 , −2 .
f. The largest open interval that contains
the initial time
t = 3 where
p2 (t), p1 (t),

p0 (t), and

r(t) are all continuous is − 5 , +∞ ∩ − ∞ , +∞ ∩ 2 , +∞ ∩ 1 , +∞ = 2 , +∞ .
By the existence
and uniqueness theorem, the solution of the I.V.P. is guaranteed to exist
in 2 , +∞ .

2.05. We have

t2 y100 − 3 t y10 − 5 y1 = t2 20 t3 − 3 t 5 t4 − 5 t5 = 20 t5 − 15 t5 − 5 t5 = 0

t2 y200 − 3 t y20 − 5 y2 = t2 2 t−3 − 3 t − t−2 − 5 t−1 = 2 t−1 + 3 t−1 − 5 t−1 = 0

Hence both y1 = t5 and y2 = t−1 , are solutions of the differential equation t2 y 00 −3 t y 0 −5 y = 0.

y (t) t5
= t6 is


The two solutions are linearly independent in 0 , +∞ , since their ratio 1 =
y2 (t) 1/t
not a constant. It follows that the general solution is y(t) = C1 y1 (t) + C2 y2 (t) = C1 t5 + C2 t−1
To solve the initial value problem, we need to determine the constants C1 and C2 so that the
5 −1 0 4 −2
initial conditions
  From y(t) = C1 t + C2 t , it follows y (t) = 5 C1 t − C2 t .
are satisfied.
y(1) = 4 C1 + C2 = 4
Thus ⇐⇒ Making use of Cramer’s Rule, we have
y 0 (1) = 2 5 C1 − C2 = 2

4 1 1 4
2 −1 −4 − 2 5 2 2 − 20
C1 = = =1 and C2 = = =3
1 1 −1 − 5 1 1 −1 − 5
5 −1 5 −1

3
Hence the solution of the initial value problem is y(t) = C1 t5 + C2 t−1 = t5 + 3 t−1 = t5 +
t

Solving a Second Order Homogeneous Constant Coefficients Differential Equation

To solve the second order linear constant coefficients differential equation a y 00 + b y 0 + c y = 0,

find the roots λ1 and λ2 of the characteristic equation a λ2 + b λ + c = 0.

If λ1 6= λ2 are reals, then the general solution is y = C1 eλ1 t + C2 eλ2 t

If λ1 = λ2 = r, then the general solution is y = C1 er t + C2 t er t

If λ1 = α + β i, λ2 = λ1 , then the general solution is y = C1 eα t cos(β t) + C2 eα t sin(β t)

MATH 375 Higher Order Linear Differential Equations Fall 2023 11

3
2.06.a The characteristic equation is 2 λ2 − λ − 3 = 0. The roots are λ1 = −1 and λ2 = . Hence a
n o 2
fundamental set of solutions is e−t , e3 t/2 , and the general solution is

C1 e−t + C2 e3 t/2

2
2.06.b The characteristic equation is 9 λ2 − 12 λ + 4 = 0. The roots are λ1 = λ2 = . Hence a
n o 3
fundamental set of solutions is e2 t/3 , t e2 t/3 , and the general solution is

C1 e2 t/3 + C2 t e2 t/3

2.06.c The characteristic equation is λ2 − 2 λ +

n 5 = 0. The roots are λo = 1 + 2 i and λ2 = 1 − 2 i.

1
et cos 2 t , et sin 2 t , and the general solution is


Hence a fundamental set of solutions is

C1 et cos 2 t + C2 et sin 2 t

4
2.07.a The characteristic equation is 3 λ2 − 2 λ − 8 = 0. The roots are λ1 = − and λ2 = 2. Hence a
n o 3
fundamental set of solutions is e−4 t/3 , e2 t , and the general solution is C1 e−4 t/3 + C2 e2 t

1
2.07.b The characteristic equation is 4 λ2 + 4 λ + 1 = 0. The roots are λ1 = λ2 = − . Hence a
n o 2
fundamental set of solutions is e−t/2 , t e−t/2 , and the general solution is C1 e−t/2 +C2 t e−t/2

3 3
2.07.c The characteristic equation is 4 λ2 − 12 λ + 25 = 0. The roots are λ1 = + 2 i and λ2 = − 2 i.
n 2 o 2
e3 t/2 cos 2 t , e3 t/2 sin 2 t



Hence a fundamental set of solutions is , and the general
solution is C1 e3 t/2 cos 2 t + C2 e3 t/2 sin 2 t

2.08.a. The characteristic equation is λ2 − 2 k λ + 2 k2 = 0 ⇐⇒ (λ − k)2 + k2 = 0. Hence the roots

λ1 = k + k i, λ2 = k − k i and the general solution is y(t) = C1 ek t cos(k t) + C2 ek t sin(k t).

2.08.b. The characteristic equation is λ2 + 2 k λ + k2 + 1 = 0 ⇐⇒ (λ + k)2 + 1 = 0. Hence the roots

λ1 = −k + i, λ2 = −k − i and the general solution is y(t) = C1 e−k t cos(t) + C2 e−k t sin(t).

2.09.a. The fact that 2 e−5 t + 3 e−2 t is a solution of the second order, linear, homogeneous, constant
coefficients differential equation y 00 + b y 0 + c y = 0, implies that λ1 = −5 and λ2 = −2, are
roots of the characteristic equation. Hence the characteristic equation is given by

λ + 5 λ + 2 = 0 ⇐⇒ λ2 + 7 λ + 10 = 0

It follows that the differential equation, in normal form is y 00 + 7 y 0 + 10 y = 0.

2.09.b. The fact that 5 e−2 t − 2 is a solution of the second order, linear, homogeneous, constant coeffi-
cients differential equation y 00 + b y 0 + c y = 0, implies that λ1 = −2 and λ2 = 0, are roots of
the characteristic equation. Hence the characteristic equation is given by

λ + 2 λ − 0 = 0 ⇐⇒ λ2 + 2 λ = 0

It follows that the differential equation, in normal form is y 00 + 2 y 0 = 0.

MATH 375 Higher Order Linear Differential Equations Fall 2023 12

2.09.c. The fact that e3 t cos(2 t), is a solution shows that the roots of the characteristic equation are
λ1 = 3 + 2 i and λ2 = 3 − 2 i. Consequently the characteristic equation is

λ − (3 + 2 i) λ − (3 − 2 i) = 0 ⇐⇒ λ2 − 6 λ + 13 = 0

Hence the differential equation (in normal form) is

y 00 − 6 y 0 + 13 y = 0

Solving a Second Order Homogeneous Cauchy-Euler Differential Equation

To solve the second order homogeneous differential equation a t2 y 00 + b t y 0 + c y = 0, t > 0,

where a 6= 0, b, c are real constants, find the roots λ1 and λ2 of the indicial/characteristic equation

a λ (λ − 1) + b λ + c = 0.

If λ1 6= λ2 are reals, then the general solution is y = C1 tλ1 + C2 tλ2

If λ1 = λ2 = r, then the general solution is y = C1 tr + C2 tr ln(t)

y = C1 tα cos β ln(t) +C2 tα sin β ln(t)

If λ1 = α+β i, λ2 = λ1 , then the general solution is

2.10a. The characteristic equation is 2 λ (λ − 1) + 5 λ + 1 = (2 λ + 1) (λ + 1) = 0. Hence the roots

n1 = −1/2, λ2 = −1. Aofundamental set of solutions is
λ
y1 = t−1/2 , y2 = t−1 and the general solution is

C C
y(t) = C1 t−1/2 + C2 t−1 = √1 + 2
t t

2.10b. The characteristic equation is 3 λ (λ − 1) + 13 λ + 3 = (3 λ + 1) (λ + 3) = 0. Hence the roots

n1 = −1/3, λ2 = −3. Aofundamental set of solutions is
λ
y1 = t−1/3 , y2 = t−3 and the general solution is

C1 C
y(t) = C1 t−1/3 + C2 t−3 = √
3
+ 32
t t

2.10c. The characteristic equation is λ (λ − 1) + 5 λ + 4 = (λ + 2)2 = 0. Hence the roots λ1 = λ2 = −2.

A fundamental set of solutions
n o is
y1 = t−2 , y2 = t−2 ln(t) and the general solution is

C1 + C2 ln(t)
y(t) = C1 t−2 + C2 t−2 ln(t) =
t2

2.10d. The characteristic equation is 4 λ (λ−1)−8

n λ+9 = (2 λ−3)2 = 0. o Hence the roots λ1 = λ2 = 3/2.
3/2 3/2
A fundamental set of solutions is y1 = t , y2 = t ln(t) and the general solution is

y(t) = C1 t3/2 + C2 t3/2 ln(t) = C1 + C2 ln(t) t3/2

MATH 375 Higher Order Linear Differential Equations Fall 2023 13

2.10e. The characteristic equation is 4 λ (λ − 1) + 16 λ + 13 = (2 λ + 3)2 + 4 = 0. Hence the roots

3
λ1 = − + i, λ2 = λ1 . A fundamental set of solutions is
n 2
o
y1 = t−3/2 cos ln(t) , y2 = t−3/2 sin ln(t)


and the general solution is



C1 cos ln(t) + C2 sin ln(t)
y(t) = C1 t−3/2 cos ln(t) + C2 t−3/2 sin ln(t) =



t3/2

2.10f. The characteristic equation is λ (λ − 1) + 7 λ + 25 = (λ + 3)2 + 16 = 0. Hence the roots

n1 = −3 + 4 i, λ2 = λ1
. A fundamental set of
solutions
λ o is
y1 = t−3 cos 4 ln(t) , y2 = t−3 sin 4 ln(t) and the general solution is



C1 cos 4 ln(t) + C2 sin 4 ln(t)
y(t) = C1 t−3 cos 4 ln(t) + C2 t−3 sin 4 ln(t) =



t3

2.11a. The characteristic equation is λ (λ − 1) − 4 λ + 4 = (λ − 1) (λ − 4) = 0. Hence the roots

λ1 = 1, λ2 = 4. The general solution is then y(t) = C1 t + C2 t4 . To solve the initial value
problem, we select the constants C1 , C2 so that
  
y(1) = −2 C1 + C2 = −2 C1 = 1
⇐⇒ ⇐⇒
y 0 (1) = −11 C1 + 4 C2 = −11 C2 = −3

Hence the solution is y(t) = t − 3 t4 .

2.11b. The characteristic equation is λ (λ − 1) + 7 λ + 5 = (λ + 5) (λ + 1) = 0. Hence the roots

λ1 = −5, λ2 = −1. The general solution is then y(t) = C1 t−5 + C2 t−1 . To solve the initial
value problem, we select the constants C1 , C2 so that
  
y(1) = −1 C1 + C2 = −1 C1 = −3
⇐⇒ ⇐⇒
y 0 (1) = 13 −5 C1 − C2 = 13 C2 = 2

Hence the solution is y(t) = −3 t−5 + 2 t−1 .

t4 − 3
= t2 − 3 t−2 is a solution, indicates that y1 = t−2 , y2 = t2

2.12a. The fact that is a
t2
fundamental set of solutions. Hence −2, 2 are the roots of the characteristic equation a λ (λ −
1) + b λ + c = 0. It follows
 
b−a=0 b=a
a λ (λ − 1) + b λ + c = a (λ + 2) (λ − 2) ⇐⇒ ⇐⇒
c = −4 a c = −4 a

Hence the characteristic equation is a λ (λ − 1) + a λ − 4 a = 0 and the differential equation is

a t2 y 00 + a t y 0 − 4 a y = 0 ⇐⇒ t2 y 00 + t y 0 − 4 y = 0

ln(t)
is a solution, indicates that y1 = t−3 , y2 = t−3 ln(t) is a fundamental

2.12b. The fact that
t3
set of solutions. Hence −3 is a double root of the characteristic equation a λ (λ − 1) + b λ + c = 0.
It follows
 
2 b − a = 6a b = 7a
a λ (λ − 1) + b λ + c = a (λ + 3) ⇐⇒ ⇐⇒
c = 9a c = 9a

Hence the characteristic equation is a λ (λ − 1) + 7 a λ + 9 a = 0 and the differential equation is

a t2 y 00 + 7 a t y 0 + 9 a y = 0 ⇐⇒ t2 y 00 + 7 t y 0 + 9 y = 0
MATH 375 Higher Order Linear Differential Equations Fall 2023 14

sin ln(t)
2.12c. The fact that is a solution, indicates that
t

y1 = t sin ln(t) , y2 = t−1 cos ln(t)
−1


is a fundamental set of solutions. Hence −1 + i
and −1 − i are the roots of the characteristic equation a λ (λ − 1) + b λ + c = 0. It follows
 
  b − a = 2a b = 3a
a λ (λ − 1) + b λ + c = a (λ + 1)2 + 1 ⇐⇒ ⇐⇒
c = 2a c = 2a

Hence the characteristic equation is a λ (λ − 1) + 3 a λ + 2 a = 0 and the differential equation is

a t2 y 00 + 3 a t y 0 + 2 a y = 0 ⇐⇒ t2 y 00 + 3 t y 0 + 2 y = 0

Solving an nth Order Homogeneous Constant Coefficients Differential Equation

To solve the nth order linear constant coefficients differential equation

an y (n) + an−1 y (n−1) + · · · + a1 y 0 + a0 y = 0

1. form the characteristic equation an λn + an−1 λn−1 + · · · + a1 λ + a0 = 0, and find all its
roots λ1 , λ2 , · · · , λn
2. build a fundamental set of solution as follows
– to each real, simple root r, associate the solution er t
– to each real root r, with multiplicity m ≥ 2, associate the m-solutions
er t , t er t , · · · , tm−1 et
– to every pair of simple roots α ± β i, associate the two solutions
eα t cos(β t), eα t sin(β t).
– to every pair of complex roots α ± β i, with multiplicity m, associate the 2 m solutions

eα t cos(β t), t eα t cos(β t), · · · , tm−1 eα t cos(β t)

eα t sin(β t), t eα t sin(β t), · · · , tm−1 eα t sin(β t)
3. when
n done with stepo 2., you should get a fundamental set of solutions
y1 (t), y2 (t), · · · , yn (t) . The general solution is then

C1 y1 (t) + C2 y2 (t) + · · · + Cn yn (t)

2.13a. The characteristic equation is λ3 + 3 λ2 − 16 λ − 48 = 0. Rewriting it as

λ2 λ + 3 − 16 λ + 3 = 0 ⇐⇒ λ2 − 16 λ + 3 = 0

we conclude that the roots are λ1 = −4,n λ2 = −3, and λ3o= 4.

Hence a fundamental set of solutions is e−4 t , e−3 t , e4 t , and the general solution is

y(t) = C1 e−4 t + C2 e−3 t + C3 e4 t

MATH 375 Higher Order Linear Differential Equations Fall 2023 15

2.13b. The characteristic equation is λ3 − 5 λ2 + 4 λ − 20 = 0. Rewriting it as

λ2 λ − 5 + 4 λ − 5 = 0 ⇐⇒ λ2 + 4 λ − 5 = 0

we conclude that the roots are λ1 = 5, nλ2 = 2 i, and λ3 = −2 i o

e5 t , cos 2 t , sin 2 t , and the general solution is



Hence a fundamental set of solutions is

y(t) = C1 e5 t + C2 cos 2 t + C3 sin 2 t

2.13c. The characteristic equation is λ4 − 4 λ3 + 14 λ2 − 20 λ + 25 = 0. Rewriting it as

λ2 + a λ + 5 λ2 + b λ + 5 = 0 ⇐⇒ λ4 + (a + b) λ3 + (a b + 10) λ2 + 5 (a + b) λ + 25 = 0

2
leads to a = b = −2. Thus the characteristic equation takes the form λ2 − 2 λ + 5 = 0. It
follows that the roots are λ1 = λ2 = 1 + 2 i, λ3 = λ4 = 1 − 2 i.
Hence a fundamentalnset of solutions is
o
et cos 2 t , et sin 2 t , t et cos 2 t , t et sin 2 t

and the general solution is

y(t) = C1 et cos 2 t + C2 et sin 2 t + C3 t et cos 2 t + C3 t et sin 2 t

2.13d. The characteristic equation is λ4 − 8 λ2 + 16 =
0. Rewriting it as

2 2
2
λ4 − 8 λ2 + 16 = 0 ⇐⇒ λ2 − 4 = 0 ⇐⇒ λ + 2 λ−2 =0
leads to the roots λ1 = λ2 = −2, nλ3 = λ4 = 2. Hence a fundamental
o set of solutions is
e−2 t , t e−2 t , e2 t , t e2 t
and the general solution is y(t) = C1 e−2 t + C2 t e−2 t + C3 e2 t + C4 t e2 t
2.14a. The characteristic equation is λ
3 + λ2 + λ
+ 1 = 0. Rewriting
it as

λ2 λ + 1 + λ + 1 = 0 ⇐⇒ λ2 + 1 λ + 1 = 0
leads to the roots λ1 = −1, λ2 = i, nand λ3 = −i Hence oa fundamental set of solutions is
e−t , cos(t) , sin(t)

3 2
2.14b.
2
λ − λ − λ2+ 1 =
The characteristic
equation is
0. Rewriting

it as

2
λ λ − 1 − λ − 1 = 0 ⇐⇒ λ − 1 λ − 1 = 0 ⇐⇒ λ + 1 λ − 1 = 0
leads to the roots λ1 = −1, λ2 = λ3 = 1n Hence a fundamental
o set of solutions is
e−t , et , t et

2
2.14c. The characteristic equation is λ4 + 6 λ2 + 9 = 0. Rewriting it as λ2 + 3 = 0, leads to the
√ √
roots λ1 = λ2 = 3 i,n and λ3 = λ4 = − 3 i. Hence a fundamental set of solutions is
√
√
√
√
o
cos 3 t , sin 3 t , t cos 3 t , t sin 3 t

3
2.14d. The characteristic equation is λ3 + 6 λ2 + 12 λ + 8 = 0 ⇐⇒ λ + 2 = 0. Hence the roots
λ1 = λ2 = λ3 = −2, and a fundamental
n set of solutions is o
e−2 t , t e−2 t , t2 e−2 t
2.15a. The solutions y1 (t) = 1 = 1 e , y2 (t) = t = t e0 t , y3 (t) = e2 t , show that the roots of the
0t

characteristic equation are λ1 = λ2 = 0, and λ3 = 2. Consequently the characteristic equation

2
λ − 2 = 0 ⇐⇒ λ2 λ − 2 = 0 ⇐⇒ λ3 − 2 λ2 = 0 Hence the differential equation



is λ − 0
(in normal form), is y 000 − 2 y 00 = 0
MATH 375 Higher Order Linear Differential Equations Fall 2023 16

2.15b. The fact that y1 (t) = et , y2 (t) = et cos(t), and y3 (t) = et sin(t), are solutions shows that the
roots of the characteristic equation are λ1 = 1, λ2 = 1 + i, and λ3 = 1 − i. Consequently the
characteristic equation is

λ − 1 λ − (1 + i) λ − (1 − i) = 0 ⇐⇒ λ − 1 λ2 − 2 λ + 2 = 0 ⇐⇒ λ3 − 3 λ2 + 4 λ − 2 = 0

Hence the differential equation (in normal form) is y 000 − 3 y 00 + 4 y 0 − 2 y = 0

2.15c. y1 (t) = cos(t), y2 (t) = sin(t), y3 (t) = t cos(t), and y4 (t) = t sin(t), being solutions implies that
the roots of the characteristic equation are λ1 = λ2 = i, and λ3 = λ4 = −i. Consequently the
characteristic equation is

2
2 
2  2
= 0 ⇐⇒ λ2 + 1 = 0 ⇐⇒ λ4 + 2 λ2 + 1 = 0


λ−i λ + i = 0 ⇐⇒ λ−i λ+i

Hence the differential equation (in normal form) is y (4) + 2 y 00 + y = 0

1 t 1 1 1
2.15d. Recalling that cosh(t) = e + e−t , and sinh(t) = et − e−t , it follows from the su-
2 2 2 2
perposition principle for homogeneous equations, that y1 (t) = e−t = cosh(t) − sinh(t), and
y2 (t) = et = cosh(t) + sinh(t) are solutions as well. This shows that the roots of the character-
istic equation are λ1 = −1, and λ2 = 1. Consequently the characteristic equation is

λ − (−1) λ − 1 = 0 ⇐⇒ λ + 1 λ − 1 = 0 ⇐⇒ λ2 − 1 = 0

Hence the differential equation (in normal form) is y 00 − y = 0.

2.16a. The characteristic equation is

λ3 + λ2 − 9 λ − 9 = 0 ⇐⇒ λ2 λ + 1 − 9 λ + 1) = 0 ⇐⇒ λ2 − 9




λ+1 =0

Its roots are λ1 = −3, λ2 = −1, and λ3 = 3. Hence the general solution is

y(t) = C1 e−3 t + C2 e−t + C3 e3 t

To solve the initial value problem, the arbitrary constants C1 , C2 , C3 , should be selected so
that the initial conditions are satisfied. From y 0 (t) = −3 C1 e−3 t − C2 e−t + 3 C3 e3 t and
y 00 (t) = 9 C1 e−3 t + C2 e−t + 9 C3 e3 t , it follows
 
 y(0) = −3  C1 + C2 + C3 = −3
0
y (0) = −3 ⇐⇒ −3 C1 − C2 + 3 C3 = −3
 00
y (0) = −3 9 C1 + C2 + 9 C3 = −3


Solving by Gauss elimination or by Cramer’s rule, we find C1 = 1, C2 = −3, and C3 = −1.

Hence the solution of the initial value problem is

y(t) = e−3 t − 3 e−t − e3 t

2
2.16b. The characteristic equation is λ4 + 8 λ2 + 16 = 0 ⇐⇒ λ2 + 4 = 0. Hence the roots
λ1 = λ2 = 2 i, and λ3 = λ4 = −2 i, and the general solution is



y(t) = C1 + C2 t cos(2 t) + C3 + C4 t sin(2 t)

To solve the initial value problem, we select the arbitrary constants C1 , C2 , C3 , C4 , so that the
initial conditions are satisfied. From
y 0 (t) = C2 + 2 C3 + 2 C4 t cos(2



t) + − 2 C1 + C4 − 2 C2 t sin(2
t)
y 00 (t) = − 4 C1 + 4 C4 − 4 C2 t
cos(2 t) + − 4 C2 − 4 C3 − 4 C4 t
sin(2 t)

y 000 (t) = − 12 C2 − 8 C3 − 8 C4 t cos(2 t) + 8 C1 − 12 C4 + 8 C2 t sin(2 t)

it follows
MATH 375 Higher Order Linear Differential Equations Fall 2023 17

  
y(0) = 3 C =3 C =3
 1  1
  
 0
  
y (0) = 2 C2 + 2 C3 = 2 C2 = −5/4
⇐⇒ ⇐⇒
y 00 (0) = 1 −4 C1 + 4 C4 = 1 C = 13/8
 3
  
 000
  
y (0) = 2 −12 C2 − 8 C3 = 2 C4 = 13/4


The solution of the initial value problem is then

   
5 13 13
y(t) = 3 − t cos(2 t) + + t sin(2 t)
4 8 4

The fact that t 2 + 3 cos(3 t) = 2 t e0 t + 3 t e0 t cos(3 t), is a solution of the differential equation,


2.17.
implies that λ = 0 is a double root, and λ = 0 + 3 i = 3 i is a double root as well, which in turn
implies that λ = 0 − 3 i = −3 i is a double root as well. Hence the characteristic equation is

2
2
2  2
λ−0 λ − 3i λ − (−3 i) = 0 ⇐⇒ λ2 (λ − 3 i) (λ + 3 i) = 0

2
⇐⇒ λ2 λ2 + 9 = 0 ⇐⇒ λ6 + 18 λ4 + 81 λ2 = 0

and the differential equation in normal form is y (6) + 18 y (4) + 81 y 00 = 0

2.18. The characteristic equation is λ3 − λ2 − 4 λ − 6 = 0. The fact that e−t cos(t) is a solution of the
equation, implies that −1 + i and −1 − i are roots of the characteristic equation. It follows

that
(λ+1)2 +2 = λ2 +2 λ+2 is a factor into λ3 −λ2 −4 λ−6, i.e., λ3 −λ2 −4 λ−6 = λ2 +2 λ+2 (λ−3).
Hence the roots of the characteristic equation are λ1 = 3, λ2 = −1 + i, λ3 = −1 − i, and the
general solution is y(t) = C1 e3 t + C2 e−t cos(t) + C3 e−t sin(t).

The Wronskian and Abel’s Formula

Given two differentiable functions u(t) and v(t), the determinant

u(t) v(t)
W (t; u, v) =
u0 (t) v 0 (t)

is called the wronskian of the two functions u(t) and v(t).

If y1 (t) and y2 (t) are any two solutions of y 00 + p(t) y 0 + q(t) y = 0, and if

y1 (t) y2 (t)
W t; y1 , y2 = is the wronskian of the two solutions, then
y10 (t) y20 (t)
R
W (t) = K e− p(t) dt
for some constant K

2.19.a We have
er t t er t 1 t  
W (t) = = er t er t = e2 r t 1 + r t − r t = e2 r t
r er t (1 + r t) er t r 1+rt
MATH 375 Higher Order Linear Differential Equations Fall 2023 18

2.19.b We have
er t cos(s t)
er t sin(s t)

W (t) = rt rt
e r cos(s t) − s sin(s t) e s cos(s t) + r sin(s t)
cos(s t) sin(s t)
= er t er t
r cos(s t) − s sin(s t) s cos(s t) + r sin(s t)
R2 − r R1
cos(s t) sin(s t)
======== e2 r t = e2 r t s cos2 (s t) + s sin2 (s t) = s e2 r t


−s sin(s t) s cos(s t)

2 0 t
2.20.a We first rewrite the differential equation in normal form as y 00 − y + y = 0. Then
3 3
R
W (t) = K e− −2/3 dt
= K e2 t/3

 
2
2.20.b We need to rewrite the equation in normal form as y 00 + 5 − y 0 + et y = 0. It follows
t


R 2
W (t) = K e− 5−2/t dt = K e−5 t+2 ln|t| = K e−5 t eln(t ) = K t2 e−5 t

2.21.a. Differentiating W (t) = y1 (t) y20 (t) − y2 (t) y10 (t), we get

W 0 (t) = y10 (t) y20 (t) + y1 (t) y200 (t) − y20 (t) y10 (t) − y2 (t) y100 (t)
= y1 (t) y200 (t) − y2 (t) y100 (t) remember both y1 (t) and y2 (t) are solutions of
y 00 + p(t) y 0 + q(t) y = 0 ⇐⇒ y 00 = −p(t) y 0 − q(t) y
   
= y1 (t) − p(t) y20 (t) − q(t) y2 (t) − y2 (t) − p(t) y10 (t) − q(t) y1 (t)
 
= −p(t) y1 (t) y20 (t) + p(t) y2 (t) y10 (t) = −p(t) y1 (t) y20 (t) − y2 (t) y10 (t)
= −p(t) W (t)

Hence W 0 (t) + p(t) W (t) = 0.

2.21.b. W 0 (t) + p(t) W (t) = 0 is a first order linear differential equation. An integrating factor is
R
p(t) dt
µ(t) = e . Multiplying both sides of the equation by µ(t), we have
 R
0 R R
e p(t) dt W (t) = 0 =⇒ e p(t) dt W (t) = K =⇒ W (t) = K e− p(t) dt

2.22. By Abel’s formula

R R 2 3
W t = K e− p(t) dt = K e− (t +1) dt = K e−t−t /3 ,


for some constant K

To compute the constant, we set t = 1, and equate the result to 2, to get

W (1) = 2 ⇐⇒ K e−1−1/3 = 2 =⇒ K e−4/3 = 2 =⇒ K = 2 e4/3

3
It follows W (t) = 2 e4/3 e−t−t /3 , and

W 2 = 2 e4/3 e−2−8/3 = 2 e4/3 e−14/3 = 2 e−10/3

MATH 375 Higher Order Linear Differential Equations Fall 2023 19

2.23. First y1 and y2 , being solutions of the differential equation, we have

 00 0
 0 00
 y1 + y1 p(t) + y1 q(t) = 0  y1 p(t) + y1 q(t) = −y1
⇐⇒
 00
y2 + y20 p(t) + y2 q(t) = 0
 0
y2 p(t) + y2 q(t) = −y200

1
Substituting for y1 , and et for y2 , the system becomes
t
1 1 2 2
 
 − 2 p(t) + q(t) = − 3
  p(t) − t q(t) =

t t t ⇐⇒ t

 t 
e p(t) + et q(t) = −et p(t) + q(t) = −1


By Cramer’s rule, we get

2/t −t 1 2/t
−1 1 2/t − t 2 − t2 1 −1 −1 − 2/t −t − 2
p(t) = = = q(t) = = =
1 −t 1+t t (t + 1) 1 −t 1+t t (t + 1)
1 1 1 1

The equation y 00 + p(t) y 0 + q(t) y = 0, then becomes

2 − t2 0 −t − 2
y 00 + t (t + 1) y 00 − t2 − 2 y 0 − t + 2 y = 0



y + y=0 or else
t (t + 1) t (t + 1)

2.24. y1 (t) and y2 (t) being solutions of y 00 + p(t) y 0 + q(t) y = 0, we have

 00 0
 0 00
 y1 + y1 p(t) + y1 q(t) = 0  y1 q(t) + y1 p(t) = −y1
⇐⇒
 00
y2 + y20 p(t) + y2 q(t) = 0 y2 q(t) + y20 p(t) = −y200


Using Cramer’s rule to solve for p(t) and q(t), we get

−y100 (t) y10 (t) y10 (t) y20 (t) y10 (t) y20 (t)
−y200 (t) y20 (t) −y100 (t) −y200 (t) y100 (t) y200 (t) W t; y10 , y20


q(t) = = =− =−
y10 (t)


y1 (t) y1 (t) y2 (t) y1 (t) y2 (t) W t; y1 , y2
y2 (t) y20 (t) y10 (t) y20 (t) y10 (t) y20 (t)

and

y1 (t) −y100 (t) y1 (t) y2 (t) y1 (t) y2 (t)

−y200 (t) −y100 (t) −y200 (t) y100 (t) y200 (t) W 0 t; y1 , y2


y2 (t)
p(t) = = =− =−
y10 (t)


y1 (t) y1 (t) y2 (t) y1 (t) y2 (t) W t; y1 , y2
y2 (t) y20 (t) y10 (t) y20 (t) y10 (t) y20 (t)

2.25. We need to show that if K and L are any two constants such that K u(t)+L v(t) = 0, ∀t ∈ (a, b),
then K = 0 and  L = 0.
K u(t) + L v(t) = 0
Setting t = 0 in leads to the system
K u0 (t) + L v 0 (t) = 0
  
K u(0) + L v(0) = 0 K + 2L = 0 K=0
0 0 ⇐⇒ ⇐⇒
K u (0) + L v (0) = 0 2K + 5L = 0 L=0
MATH 375 Higher Order Linear Differential Equations Fall 2023 20

Reduction of Order Method

Assuming that z = z(t) is a given solution of the homogeneous differential equation
a(t) y 00 + b(t) y 0 + c(t) y = 0, the substitution y = z(t) u, converts the differential equation
a(t) y 00 + b(t) y 0 + c(t) y = f (t), whether f (t) ≡ 0 or not, into

a(t) z(t) · u00 + 2 a(t) z 0 (t) + b(t) z(t) · u0 = f (t)

or equivalently
a(t) z(t) · v 0 + 2 a(t) z 0 (t) + b(t) z(t) · v = f (t)


u0 = v
This way of reducing a second order linear differential equation to two first order linear differential
equations is known as the reduction of order method.

2 2
2
2.26.a We have y1 = et =⇒ y10 = 2 t et =⇒ y100 = 4 t2 + 2 et . It follows

2 2
2
2
y100 −4 t y10 + 4 t2 −2 y1 = 4 t2 +2 et −4 t 2 t et + 4 t2 −2 et = 4 t2 +2−8 t2 +4 t2 −2 et = 0

2.26.b The reduction of order method, as mentioned above, looks for the general solution in the form
2
y(t) = y1 (t) v(t) = et v.
We have 2 2
2 2 2
y 0 = 2 t et v + et v 0 =⇒ y 00 = 4 t2 + 2 et v + 4 t et v 0 + et v 00
Substituting into the differential equation, we have
2 2
2 2 2
2
et v 00 + 4 t et v 0 + 4 t2 + 2 et v − 4 t et v 0 − 8 t2 et v + 4 t2 − 2 et v = 0
or else
2
et v 00 = 0 ⇐⇒ v 00 = 0 ⇐⇒ v = C1 t + C2
Hence the general solution
2 2 2
y(t) = y1 (t) v(t) = et C1 t + C2 = C1 t et + C2 et

2.27.a We have y1
= t =⇒ y10 = 1 =⇒ y100 =
0
It follows

t2 1 − ln(t) y100 + t y10 − y1 = t2 1 − ln(t) 0 + t 1 − t = 0 + t − t = 0

2.27.b Using the reduction of order method, we look for the general solution in the form

y(t) = y1 (t) v(t) = t v

We have
y 0 = v + t v 0 =⇒ y 00 = 2 v 0 + t v 00
Substituting into the differential equation, leads to

t2 1 − ln(t) 2 v 0 + t v 00 + t v + t v 0 − t v = 0 ⇐⇒ t3 1 − ln(t) v 00 + t2 3 − 2 ln(t) v 0 = 0

or else
3 − 2 ln(t) 1 0
v 00 + v =0
1 − ln(t) t
Setting u = v 0 , the equation becomes
3 − 2 ln(t) 1
u0 + u=0
1 − ln(t) t

t2
This is a first order linear differential. An integrating factor is µ = .
1 − ln(t)
MATH 375 Higher Order Linear Differential Equations Fall 2023 21

Z
3 − 2 ln(t) 1
The substitution s = 1 − ln(t) was used to compute dt.
1 − ln(t) t
1 − ln(t)
Multiplying by the integrating factor and solving for u, we obtain u = C1 .
t2
1 − ln(t) ln(t)
It follows v 0 = C1 =⇒ v = C1 + C2 Hence the general solution is
t2 t
 
ln(t)
y = t v = t C1 + C2 = C1 ln(t) + C2 t
t

2.28.a Using the reduction of order method, we look for the general solution in the form
y(t) = y1 (t) v(t) = et v. We have

y 0 = et v + v 0 =⇒ y 00 = et v + 2 v 0 + v 00

Substituting into the differential equation, leads to

 
1
et t v 00 + (t − 1) v 0 = t2 e2 t ⇐⇒ v 00 + 1 − v 0 = t et
 
t
 
1
Setting u = v 0 , the equation becomes u0 + 1 − u = t et . This is a first order linear
t
et
differential. An integrating factor is µ = .
t
1
Multiplying by the integrating factor and solving for u, we obtain u = C1 t e−t + t et .
2
1 1
It follows v 0 = C1 t e−t + t et =⇒ v = −C1 (t + 1) e−t + (t − 1) et + C2 Hence the general
2 2
solution is
1
y = et v = −C1 (t + 1) + C2 et + (t − 1) e2 t
2

2.28.b Using the reduction of order method, we look for the general solution in the form y(t) =
y1 (t) v(t) = t v(t). We have y 0 = v + t v 0 =⇒ y 00 = 2 v 0 + t v 00 . Substituting into the

differential equation, leads to

t2 − 2 t + 2 0 1−t
t (1 − t) v 00 + t2 − 2 t + 2 v 0 = (1 − t)2 ⇐⇒ v 00 +


v =
t(1 − t) t
 
2 1 1−t
Setting u = v 0 , the equation becomes u0 + −1 + − u= . This is a first order
t t−1 t
t2 e−t
linear differential. An integrating factor is µ = .
1−t
(1 − t) et t2 − 1
Multiplying by the integrating factor and solving for u, we obtain u = C1 2
+ .
 t 0 t t2
t
e 1 e 1
It follows v 0 = −C1 + 1 − 2 =⇒ v = −C1 + t + + C2 Hence the general solution
t t t t
is
y = t v = −C1 et + C2 t + t2 + 1

C1 sin(t) + C2 cos(t) C1 sin(t) + C2 cos(t)

2.29. a. y(t) = b. y(t) = √
t t
MATH 375 Higher Order Linear Differential Equations Fall 2023 22

The Method of Undetermined Coefficients

Consider the linear nonhomogeneous nth order differential equation with constant coefficients

an y (n) + an−1 y (n−1) + · · · + a1 y 0 + a0 y = f (t) · · · (∗)

 
1. If f (t) = eα t R(t) cos(β t) + S(t) sin(β t) , where α, β are real constants, and
R(t), S(t) are polynomials of degree r, s, respectively, then a particular solution to
(∗), is given by
 
ypart = tk eα t P (t) cos(β t) + Q(t) sin(β t)

where P (t), Q(t) are polynomials of degree max r , s , and

0 if α + β i is not a root of the characteristic equation
k=
multiplicity of α + β i if α + β i is a root of the characteristic equation

 
2. If f (t) = f1 (t) + f2 (t), where f1 (t) = eα1 t R1 (t) cos(β1 t) + S1 (t) sin(β1 t)
 
and f2 (t) = eα2 t R2 (t) cos(β2 t) + S2 (t) sin(β2 t) , then a particular solution
to (∗), is given by ypart = ypart,1 + ypart,2 , where ypart,1 is the particular solution
associated with f1 (t) and ypart,2 is the particular solution associated with f2 (t).

2.30a. The characteristic equation of the homogeneous equation is 4 λ2 − 3 λ = 0. It has roots λ1 = 0

3
and λ2 = . The right hand side f (t) = t e(3/4) t has the exponential-polynomial-cosine-
4   3
sine form eα t M (t) cos(β t) + N (t) sin(β t) , with α = and β = 0. It follows that the
4
Undetermined Coefficients Method
 (UCM),
 is applicable, and we look for a particular solution in
k (3/4) t
the form yP (t) = t e A t + B , where k is the multiplicity of α + β i as a root of the
3 3
characteristic equation. Here α + β i = + 0 i = , is clearly a root of the characteristic
4 4
equation, with multiplicity 1. Hence k = 1, and
 
yP (t) = t A t + B e(3/4) t

Finding the constants is not part of the problem, but if you substitute and solve, you will get
 
1 2 4
yp (t) = t − t e3 t/4
6 9

2.30b. The characteristic equation of the homogeneous equation is λ2 + 1 = 0. It has roots λ1 = i

and λ2 = −i. The right hand side f (t) = sin(t)
 − cos(t) has the exponential-polynomial-cosine-
sine form eα t M (t) cos(β t) + N (t) sin(β t) , with α = 0 and β = 1. It follows that the
Undetermined Coefficients
 Method (UCM), is applicable, and we look for a particular solution in
the form yP (t) = tk A cos(t) + B sin(t) , where k is the multiplicity of α + β i as a root of
the characteristic equation. Here α + β i = 0 + 1 i = i, is clearly a root of the characteristic
equation, with multiplicity 1. Hence k = 1, and


yp (t) = t A cos(t) + B sin(t)

Finding the constants is not part of the problem, but if you substitute and solve, you will get
MATH 375 Higher Order Linear Differential Equations Fall 2023 23

1 1
yp (t) = − t cos(t) − t sin(t)
2 2

2.30c. The characteristic equation of the homogeneous equation is λ2 −3 λ+2 = 0 ⇐⇒ (λ−1) (λ−2) = 0
Its roots are λ1 = 1 and λ2 = 2. Expanding the right side, the equation becomes

y 00 − 3 y 0 + 2 y = t2 + 2 t et + e2 t

According to the superposition principle for nonhomogeneous equations, the particular solution is

yp (t) = yp,1 (t) + yp,2 (t) + yp,3 (t)

where
yp,1 (t) is a particular solution of y 00 − 3 y 0 + 2 y = t2
yp,2 (t) is a particular solution of y 00 − 3 y 0 + 2 y = 2 t et
yp,3 (t) is a particular solution of y 00 − 3 y 0 + 2 y = e2 t
According to the undetermined coefficients method, the form of each particular solution, is

yp,1 (t) = tk A t2 + B t + C

with k = 0, since α + β i = 0 + 0 i = 0, is not a root of the characteric equation.

yp,2 (t) = tk D t + E et

with k = 1, since α + β i = 1 + 0 i = 1 is a simple root of the characteristic equation.

yp,3 (t) = tk F e2 t

with k = 1, since α + β i = 2 + 0 i = 2, is a simple root of the characteristic equation.

Hence the form of yp (t), is

yp (t) = yp,1 (t) + yp,2 (t) + yp,3 (t) = A t2 + B t + C + D t2 + E t et + F t e2 t

Finding the constants is not part of the problem, but if you substitute and solve, you will get
1 2 3 7
t + t + − t2 + 2 t et + t e2 t


yp (t) =
2 2 4

2.30d. The characteristic equation of the homogeneous equation is λ2 + 16 = 0. It has roots λ1 = 4 i

and λ2 = −4 i. Rewriting the right side as
√
 π π π 1 3
f (t) = sin 4 t + = sin(4 t) cos + cos(4 t) sin = sin(4 t) + cos(4 t)
3 3 3 2 2
 
shows that it has the exponential-polynomial-cosine-sine form eα t M (t) cos(β t) + N (t) sin(β t) ,
with α = 0 and β = 4. It follows that the (UCM) is applicable, and we look for a particular
solution in the form
yp (t) = tk A cos(4 t) + B sin(4 t)

Here α + β i = 0 + 4 i = 4 i, is a root of the characteristic equation, with multiplicity 1. Thus

k = 1, and

yp (t) = t A cos(4 t) + B sin(4 t) = A t cos(4 t) + B t sin(4 t)
Finding the constants is not part of the problem, but if√you substitute and solve, you will get
1 3
yp (t) = − t cos(4 t) + t sin(4 t)
16 16
MATH 375 Higher Order Linear Differential Equations Fall 2023 24

2.31a. The characteristic equation of the associated homogeneous equation

n is λ2o− 4 λ + 4 = 0. Its
roots are λ1 = λ2 = 2, and a fundamental set of solutions is e , t e2 t . Thus the general
2t

solution of the homogeneous equation is yh (t) = C1 e2 t + C2 t e2 t .

A particular solution yp (t) can be found by using the undetermined coefficients method. It should
have the form yp (t) = A t2 +B t+C. Substituting into the equation and solving for the constants,
1 1 3
we get yp (t) = t2 + t + . Hence the general solution is
4 2 8
1 2 1 3
y(t) = yh (t) + yp (t) = C1 e2 t + C2 t e2 t + t + t+
4 2 8

2.31b. Characteristic equation: λ2 + 4 λ + 4 = 0

The roots: λ1 = λ2 = −2 n o
Fundamental set of solutions: e−2 t , t e−2 t
General solution of the homogeneous: yh (t) = C1 e−2 t + C2 t e−2 t
Particular solution: yp (t) = t2 A e−2 t = 4 t2 e−2 t
General solution of the nonhomogeneous equation

y(t) = yh (t) + yp (t) = C1 e−2 t + C2 t e−2 t + 4 t2 e−2 t

2.31c. Characteristic equation: λ2 + 1 = 0

The roots: λ1 = i and λ2 = −i n o
Fundamental set of solutions: cos(t) , sin(t)
General solution of the homogeneous:
 yh (t) = C1 cos(t) + C2 sin(t)
A t + B) cos(t) + C t + D sin(t) = t cos(t) + t2 sin(t)


Particular solution: yp (t) = t
General solution of the nonhomogeneous equation

y(t) = yh (t) + yp (t) = C1 cos(t) + C2 sin(t) + t cos(t) + t2 sin(t)

2.31d. Characteristic equation: λ2 − 3 λ + 2 = 0

The roots: λ1 = 1, λ2 = 2 n o
Fundamental set of solutions: et , e2 t
General solution of the homogeneous: yh (t) = C1 et+ C2 e2 t 
1 2
Particular solution: yp (t) = A t2 + B t + C) e3 t = t − t + 1 e3 t
2
General solution of the nonhomogeneous equation
 
t 2t 1 2
y(t) = yh (t) + yp (t) = C1 e + C2 e + t − t + 1 e3 t
2

2.32a. Characteristic equation: λ2 + λ = 0

The roots: λ1 = 0, λ2 = −1 n o
Fundamental set of solutions: 1 , e−t
General solution of the homogeneous: yh (t) = C1 + C2 e−t
Particular solution: yp (t) = t e−t A = −t e−t
General solution of the nonhomogeneous equation
y(t) = yh (t) + yp (t) = C1 + C2 e−t − t e−t
Solution of the IVP: y(t) = 1 − t e−t
MATH 375 Higher Order Linear Differential Equations Fall 2023 25

2.32b. Characteristic equation: λ4 − 1 = 0

The roots: λ1 = −1, λ2 = 1, λn3 = i, λ4 = −i o
Fundamental set of solutions: e−t , et , cos(t) , sin(t)
General solution of the homogeneous: yh (t) = C1 e−t + C2 et + C3 cos(t) + C4 sin(t)
Particular solution: yp (t) = t et A = 2 t et
General solution of the nonhomogeneous equation

y(t) = yh (t) + yp (t) = C1 e−t + C2 et + C3 cos(t) + C4 sin(t) + 2 t et

Solution of the IVP: y(t) = e−t + (2 t − 3) et + cos(t) + 2 sin(t)

2.33. The characteristic equation of the homogeneous equation is

λ3 − 4 λ2 + 6 λ − 4 = 0 ⇐⇒ (λ − 2) (λ2 − 2 λ + 2) = 0 ⇐⇒ (λ − 2) (λ − 1)2 + 1 = 0

Hence the roots λ1 = 2, λ2 = 1 + n i, and λ3 = 1 − i. A o fundamental set of solutions for

2t t t
the homogeneous equation is then e , e cos(t) , e sin(t) . Expanding the right side, and
1 − cos(2 t)
making use of the identity sin2 (t) = , the equation becomes
2
y 000 − 4 y 00 + 6 y 0 − 4 y = e2 t + 4 et sin(t) + 4 sin2 (t) = e2 t + 4 et sin(t) + 2 − 2 cos(2 t)

According to the undetermined coefficients method, the form of the particular solution, is

yp (t) = t A e2 t + t et B cos(t) + C sin(t) + D + E cos(2 t) + F sin(2 t)

Finding the constants is not part of the problem, but if you substitute and solve, you will get
1 2t
1 3 1
yp (t) = t e + t et cos(t) − sin(t) − − cos(2 t) − sin(2 t)
2 2 20 20

1 1
2.34a. Rewriting cos2 (3 t) as + cos(6 t), shows that the Undetermined Coefficients Method (UCM)
2 2
can be used to find a particular solution.

2.34b. The differential equation is not constant coefficients, as a result the UCM can not be used to find
a particular solution.

et + e−t
2.34c. Notice that 4 t cosh(t) = 4 t = 2 t et + 2 t e−t . As a result, the UCM can be used to find
2
a particular solution.

2.34d. Making use of the identity 2 sin(a) cos(b) =
sin(a + b) + sin(a − b), we rewrite the right side
as 4 t sin(3 t) cos(5 t) = 2 t sin(8 t) + sin(−2 t) = 2 t sin(8 t) − 2 t sin(2 t), which shows that the
UCM can be used to find a particular solution.

t
2.34e. From = t e−t , we see that the UCM can be used to find a particular solution.
et

e3 t + e−3 t 1 1
2.34f. From cosh(3 t) sin(t) = sin(t) = e3 t sin(t) + e−3 t sin(t), it follows that the UCM
2 2 2
can be used to find a particular solution.
MATH 375 Higher Order Linear Differential Equations Fall 2023 26

The Variation of Parameters Method

Consider the linear nonhomogeneous second order differential equation

a2 (t) y 00 + a1 (t) y 0 + a0 (t) y = f (t) · · · (∗)

n o
Given a fundamental set of solutions y1 (t) , y2 (t) of the associated homogeneous equation, the
variation of parameters method looks for a particular solution of (∗), in the form

yp (t) = u1 (t) y1 (t) + u2 (t) y2 (t)

y1 (t) u01 (t) + y2 (t) u02 (t) = 0

u1 (t), u2 (t) satisfy the system
y10 (t) u01 (t) + y20 (t) u02 (t) = f (t)/a2 (t)
f (t) y1 (t) y2 (t)
Letting g(t) = , W (t) = , and solving leads to
a2 (t) y10 (t) y20 (t)

1 0 y2 (t) y2 (t) g(t) 1 y1 (t) 0 y1 (t) g(t)

u01 (t) = =− u02 (t) = =
W (t) g(t) y20 (t) W (t) W (t) y10 (t) g(t) W (t)

Next integrate without adding the constants of integration to get u1 (t) and u2 (t). Finally substitute
back into yp (t) = u1 (t) y1 (t) + u2 (t) y2 (t) to get the particular solution.

Notice: adding the constants of integration will produce the full general solution.

2.35a. The characteristic equation of the homogeneous equation is λ2 + 1 = 0. It has roots λ1 = i and
λ2 = −i. Hence a fundamental set of solutions for the homogeneous equation is
n o
y1 (t) = cos(t) , y2 (t) = sin(t)

According to the variation of parameters method, we look for a particular solution in the form

yp (t) = y1 (t) u1 (t) + y2 (t) u2 (t)

with u01 (t) and u02 (t) solutions of the system


1 0 y2 (t)
u0 (t) =

0 0
y20 (t)
 
 y1 (t) u1 (t) + y2 (t) u2 (t) = 0  1 W (t) f (t)


⇐⇒
y10 (t) u01 (t) + y20 (t) u02 (t) = f (t) 1 y1 (t) 0
 

 u02 (t) =


W (t) y10 (t) f (t)

y1 (t) y2 (t)
where W (t) = , is the wronskian of the fundamental set of solutions. We have
y10 (t) y20 (t)

cos(t) sin(t)
W (t) = =1
− sin(t) cos(t)

1 0 sin(t) 1
u01 (t) = = − sin(t) sec3 (t) = − tan(t) sec2 (t) =⇒ u1 (t) = − tan2 (t)
1 sec3 (t) cos(t) 2
and
1 cos(t) 0
u02 (t) = = cos(t) sec3 (t) = sec2 (t) =⇒ u1 (t) = tan(t)
1 − sin(t) sec3 (t)
MATH 375 Higher Order Linear Differential Equations Fall 2023 27

Hence the particular solution

 1   
yp (t) = u1 (t) y1 (t) + u2 (t) y2 (t) = − tan2 (t) cos(t) + tan(t) sin(t)
  2  
1 1 1
= − tan(t) cos(t) + sin(t) tan(t) = − sin(t) + sin(t) tan(t) = sin(t) tan(t)
2 2 2
sin2 (t) 1 − cos2 (t) 1 1
= = = sec(t) − cos(t)
2 cos(t) 2 cos(t) 2 2
1
Now because cos(t), is a solution of the associated homogeneous equation ( remember cos(t)
2
1
is part of the fundamental set of solutions ), we can drop off − cos(t) from yp (t) to simply get
2
1
yp (t) = sec(t)
2
1
The general solution of the equation is then y(t) = C1 cos(t) + C2 sin(t) + sec(t) · · · (∗)
2
1
Notice that if we haven’t dropped off − cos(t) from the particular solution, the general solution
2
would have been
 
1 1 1 1
y(t) = C1 cos(t) + C2 sin(t) + sec(t) − cos(t) = C1 − cos(t) + C2 sin(t) + sec(t)
2 2 2 2

Now because C1 is an arbitrary constant, that’s the same solution as (∗)

2.35b. The characteristic equation of the homogeneous equation is λ2 − 2 λ + 1 = 0 ⇐⇒ (λ − 1)2 = 0.

It has roots λ1 = λ2 = 1. Hence a fundamental set of solutions for the associated homogeneous
equation is n o
y1 (t) = et , y2 (t) = t et
According to the variation of parameters method, we look for a particular solution in the form
yp (t) = y1 (t) u1 (t) + y2 (t) u2 (t), with u01 (t) and u02 (t) solutions of the system

0 1 0 y2 (t)
 u1 (t) =

0 0
W (t) f (t) y20 (t)
 
 y1 (t) u1 (t) + y2 (t) u2 (t) = 0 

⇐⇒
 0
y1 (t) u01 (t) + y20 (t) u02 (t) = f (t) 1 y1 (t) 0


 u02 (t) =


W (t) y10 (t) f (t)

y1 (t) y2 (t)
where W (t) = , is the wronskian of the fundamental set of solutions.
y10 (t) y20 (t)
We have
et t et
W (t) = = (t + 1) e2 t − t e2 t = e2 t
et (t + 1) et
1 0 t et −t e2 t t 1
u01 (t) = =⇒ u1 (t) = − ln t2 + 1


t 2 =
=− 2
e2 t e /(t + 1) (t + 1) et e2 t 2
t +1 t + 1 2
and
1 et 0 e2 t 1
u02 (t) = =
= 2 =⇒ u2 (t) = tan−1 (t)
e2 t et t 2
e /(t + 1) e2 t t2 + 1 t +1
 1
  
Hence the particular solution yp (t) = et − ln t2 + 1 + t et tan−1 (t) , and the general
2
solution is
1 t
y(t) = C1 et + C2 t et − e ln t2 + 1 + t et tan−1 (t)


2
MATH 375 Higher Order Linear Differential Equations Fall 2023 28

2.36a. The roots of the characteristic

n equationo λ2 − 1 = 0 are λ1 = −1, λ2 = 1. Hence a fundamental
−t
set of solutions is y1 = e , y2 = et The wronskian of these solutions is

e−t et
W (t) = =2
−e−t et

1 0 et et 1
u01 (t) = ln 1 + et


=− =⇒ u1 (t) = −
2 1/(1 + et ) et 1 + et 2

e−t 1 e−t
 
0 1 0 1 1 1 1 1
u2 (t) = = =
= −
2 −e−t 1/(1 + et ) 2 1 + et 2 et 1 + et 2 et 1 + et
1 −t 1 e−t
= e −
2 2 e−t + 1
Integrating, leads to

1 + et
 
1 −t 1 1 −t
e + ln e−t + 1 = − e−t + ln e + ln 1 + et + t





u2 (t) = − t
=−
2 2 e 2
Hence a particular solution is
1
1
yp (t) = y1 u1 + y2 u2 = − e−t ln 1 + et − et e−t + ln 1 + et + t



2 2
t
1 t

= − cosh(t) ln 1 + e − 1 + te
2

n o
2.36b. A fundamental set of solutions is y1 = t−1 , y2 = t2 . The wronskian of these solutions is

t−1 t2
W (t) = =3
−t−2 2t

1 0 t2 2t + 1 4t + 3
u01 (t) = 2t + 1 =− =⇒ u1 (t) =
3 2t 3 (t + 1)3 6 (t + 1)2
t2 (t + 1)3

1 t−1 0 2t + 1 1 1 1 1
0
u2 (t) = −2 2t + 1 = =− 2 + 3 + +
3 −t 3 t3 (t + 1)3 3t 3t 3 (t + 1)2 3 (t + 1)3
t2 (t + 1)3
Integrating, leads to
1 1 1 1
u2 (t) = − 2 − −
3t 6t 3 (t + 1) 6 (t + 1)2
Hence a particular solution is
 
1 4t + 3 1 1 1 1
yp (t) = y1 u1 + y2 u2 = + t2 − 2 − −
t 6 (t + 1)2 3t 6t 3 (t + 1) 6 (t + 1)2
1
=
2 t (t + 1)
MATH 375 Higher Order Linear Differential Equations Fall 2023 29

2.37a. This is a Cauchy-Euler equation. Its indicial/characteristic equation is

λ (λ − 1) + λ + 9 = λ2 + 9 = 0

It has roots ±3 i. Hence a fundamental set of solutions is

n

o
y1 = cos 3 ln(t) , y2 = sin 3 ln(t)

The wronskian of these solutions is

cos 3 ln(t)
sin 3 ln(t)
3
W (t) = −1 −1 =
−3 t sin 3 ln(t) 3 t cos 3 ln(t) t

0 t 0 sin 3 ln(t)
sin 3 ln(t) tan 3 ln(t)
u1 (t) = =
3 −t−2 tan 3 ln(t) 3 t−1 cos 3 ln(t)


3t
1



= sec 3 ln(t) − cos 3 ln(t)
3t



t cos 3 ln(t) 0 sin 3 ln(t)
u02 (t) = −1

−2

=−
3 −3 t sin 3 ln(t) −t tan 3 ln(t) 3t
Integrating, leads to
1

1
1

u1 (t) = ln sec 3 ln(t) + tan 3 ln(t) − sin 3 ln(t) , u2 (t) = cos 3 ln(t)
9 9 9
Hence a particular solution is
1



yp (t) = y1 u1 + y2 u2 = cos 3 ln(t) ln sec 3 ln(t) + tan 3 ln(t)
9

2.37b. This is a Cauchy-Euler equation. Its indicial/characteristic equation is

2
λ (λ − 1) + 3 λ + 1 = λ + 1 = 0
n o
It has −1 as a double root. Hence a fundamental set of solutions is y1 = t−1 , y2 = t−1 ln(t) .
t−1 t−1 ln(t)

The wronskian of these solutions is W (t) = = t−3 ,
−t−2 −2
t 1 − ln(t)

t−1 ln(t)
t−1

0 0
u01 (t) = t3 −3 −2 = −t−1 ln(t), u02 (t) = t3 = t−1
t 3t 1 − ln(t) −t−2 t−3

1 2
Integrating, leads to u1 (t) = − ln (t), u2 (t) = ln(t). Hence a particular solution is
2
1 −1 2
yp (t) = y1 u1 + y2 u2 = t ln (t)
2

2.37c. This is a Cauchy-Euler equation. Its indicial/characteristic equation is

2
λ (λ − 1) − λ + 1 = λ − 1 = 0
n o
It has 1 as a double root. Hence a fundamental set of solutions is y1 = t , y2 = t ln(t) . The
t t ln(t)
wronskian of these solutions is W (t) = = t,
1 1 + ln(t)
MATH 375 Higher Order Linear Differential Equations Fall 2023 30

0 t ln(t) 3
u01 (t) = t−1 = − ln(t) −
1 + 3/t ln(t) 1 + ln(t) t
t 0 3
u02 (t) = t−1 =1+
1 1 + 3/t ln(t) t ln(t)
Integrating, leads to


u1 (t) = t − (t + 3) ln(t), u2 (t) = t + 3 ln ln(t)

Hence a particular solution is

yp (t) = y1 u1 + y2 u2 = t2 − 3 t ln(t) + 3 t ln(t) ln ln(t)

Because t ln(t) is a solution of the associated homogeneous differential equation, it can be dropped
and we get the simpler particular solution

yp (t) = t2 + 3 t ln(t) ln ln(t)

The Variation of Parameters Method for Third Order Equations

Consider the linear nonhomogeneous third order differential equation

a3 (t) y 000 + a2 (t) y 00 + a1 (t) y 0 + a0 (t) y = f (t) · · · (∗)

n o
Given a fundamental set of solutions y1 (t) , y2 (t) , y3 (t) of the associated homogeneous
equation, the variation of parameters method looks for a particular solution of (∗), in the form

yp (t) = u1 (t) y1 (t) + u2 (t) y2 (t) + u3 (t) y3 (t)

0 0 0

 y1 u1 + y2 u2 + y3 u3 = 0
u1 (t), u2 (t), u3 (t) satisfy the system y10 u01 + y20 u02 + y30 u03 = 0
 00 0
y1 u1 + y200 u02 + y300 u03 = f (t)/a3 (t)
y1 (t) y2 (t) y3 (t)
f (t)
Letting g(t) = , W (t) = y10 (t) y20 (t) y30 (t) , and solving for u01 , u02 , u03 , leads to
a3 (t)
y100 (t) y200 (t) y300 (t)

0 y2 (t) y3 (t) y1 (t) 0 y3 (t)

1 1
u01 (t) = 0 y20 (t) y30 (t) u02 (t) = y10 (t) 0 y30 (t)
W (t) W (t)
g(t) y200 (t) y300 (t) y100 (t) g(t) y300 (t)

y1 (t) y2 (t) 0
1
u03 (t) = y10 (t) y20 (t) 0 with
W (t)
y100 (t) y200 (t) g(t)
Next integrate without adding the constants of integration to get u1 (t), u2 (t) and u3 (t).
Finally substitute back into yp (t) = u1 (t) y1 (t) + u2 (t) y2 (t) + u3 (t) y3 (t) to get the particular
solution.
Notice: adding the constants of integration will produce the full general solution.
MATH 375 Higher Order Linear Differential Equations Fall 2023 31

2.38a. The characteristic equation of the homogeneous equation is λ3 + λ2 = 0 ⇐⇒ λ2 (λ + 1) = 0. It

has roots λ1n= λ2 = 0 and λ3 = −1. Hence a fundamental
o set of solutions of the homogeneous
equation is y1 (t) = 1 , y2 (t) = t , y3 (t) = e−t
We have
1 t e−t
W (t) = 0 1 −e−t = (1) (1) (e−t ) = e−t
0 0 e−t

0 e−t
t
1 t−1 t e−t t−1 1 − t2
u01 (t) = −t
= −t 2 − t e−t − e−t =


0 −e 1 = −t 2 −t
e −t e t 1 −e e t t2
(t − 1)/t2 e−t
0
1 1
= − 1 =⇒ u1 (t) = − − t
t2 t
1 0 e−t
1 1 0 −e−t 1 e−t (t − 1) t−1
u02 (t) = 0 0 −e−t = −t 2 −t = −t = 2
e −t
2 −t e (t − 1)/t e e t2 t
0 (t − 1)/t e
1 1 1
= − 2 =⇒ u2 (t) = ln |t| +
t t t
1 t 0  t 0
1 t−1 et · t − et · 1 e et
u03 (t) = 0 1 0 = e t
(1) (1) = = =⇒ u 3 (t) =
e−t t2 t2 t t
0 0 (t − 1)/t2

Hence the particular solution

     t
1 1 e
yp (t) = − − t 1 + ln |t| + t+ e−t = 1 − t + t ln |t|
t t t

2.38b. The characteristic equation of the homogeneous equation is λ3 + λ = 0 ⇐⇒ λ (λ2 + 1) = 0.

Its roots are λ1 = 0, λn 2 = i and λ3 = −i. Hence a fundamental
o set of solutions of the
homogeneous equation is y1 (t) = 1 , y2 (t) = cos(t) , y3 (t) = sin(t) . The Wronskian of the
above solutions is
1 cos(t) sin(t)
− sin(t) cos(t)
W (t) = 0 − sin(t) cos(t) = =1
− cos(t) − sin(t)
0 − cos(t) − sin(t)

0 cos(t)
sin(t)
1 cos(t) sin(t)
u01 (t) = 0 − sin(t)
cos(t) = tan(t)
1 − sin(t) cos(t)
tan(t) − cos(t)
− sin(t)
2 2

= tan(t) cos (t) + sin (t) = tan(t) =⇒ u1 (t) = ln |sec(t)|
1 0 sin(t)
1 0 cos(t)
u02 (t) = 0 0 cos(t) = = − cos(t) tan(t) = − sin(t) =⇒
1 tan(t) − sin(t)
0 tan(t) − sin(t)
u2 (t) = cos(t)
1 cos(t) 0
1 − sin(t) 0 sin2 (t)
u03 (t) = 0 − sin(t) 0 = = − sin(t) tan(t) = −
1 − cos(t) tan(t) cos(t)
0 − cos(t) tan(t)
1 − cos2 (t)
=− = − sec(t) + cos(t) =⇒ u3 (t) = − ln |sec(t) + tan(t)| + sin(t)
cos(t)
MATH 375 Higher Order Linear Differential Equations Fall 2023 32

Hence the particular solution

 
yp (t) = ln |sec(t)| (1) + cos(t) cos(t) + − ln |sec(t) + tan(t)| + sin(t) sin(t)

or else
yp (t) = ln |sec(t)| − sin(t) ln |sec(t) + tan(t)|

Second Order Boundary Value Problems

To solve the boundary value problem

 00
y + p(x) y 0 + q(x) y = f (x) a<x<b
αa y(a) + βa y 0 (a) = ya and αb y(b) + βb y 0 (b) = yb

Start by finding the general solution

y(x) = C1 y1 (x) + C2 y2 (x) + yp (x)

of the differential equation. Next determine the constants C1 and C2 , if possible, so that the
boundary conditions αa y(a) + βa y 0 (a) = ya and αb y(b) + βb y 0 (b) = yb , are satisfied.

2.39 The characteristic equation of the differential equation is λ2 + 1 = 0. It has roots ±i.
Hence the general solution is y(x) = C1 cos(x) + C2 sin(x).

2.39a. The general solution is y(x) = C1 cos(x) + C2 sin(x).

We need to select the constants C1 , C2 , such that
 
y(0) = 0 C1 = 0
⇐⇒ ⇐⇒ C1 = 0
y(π) = 0 −C1 = 0

Which leads to the solution y(x) = C sin(x), where C is an arbitrary real constant.
2.39b. A particular solution of y 00 + y = 1 is yp (x) = 1. It follows that the general solution is

y(x) = C1 cos(x) + C2 sin(x) + 1

We need to select the constants C1 , C2 , such that

 
y(0) = 0 C1 + 1 = 0
⇐⇒
y(1) = 0 C1 cos(1) + C2 sin(1) + 1 = 0

  C1 = −1
C1 = −1
⇐⇒ ⇐⇒ −1 + cos(1)
− cos(1) + C2 sin(1) + 1 = 0  C2 =
sin(1)

−1 + cos(1)
Which leads to the solution y(x) = − cos(x) + sin(x) + 1.
sin(1)

2.39c. The general solution is y(x) = C1 cos(x) + C2 sin(x).

We need to select the constants C1 , C2 , such that
 
y(0) = 0 C1 = 0
⇐⇒
y(π) = 1 −C1 = 1
MATH 375 Higher Order Linear Differential Equations Fall 2023 33

That is impossible, and as a result the boundary value problem has no solution.

2.40a. The general solution of the differential equation is

√
√
√ √
√ √

y(t) = C1 cos 2 t + C2 sin 2 t =⇒ y 0 (t) = − 2 C1 sin 2 t + 2 C2 cos 2 t

Now, we select the constants C1 , C2 so that


1 √

 0
 √  C1 = √ cot 2 π

y (0) = 1 2 C2 = 1 √
√
√ √
2
⇐⇒ ⇐⇒
y 0 (π) = 0 − 2 C1 sin 2 π + 2 C2 cos 2 π = 0  C2 = √
 1
2
Hence √
√
√
√



1 1 cos 2 (t − π)
y(t) = √ cot 2 π cos 2 t + √ sin 2 t = √ √

2 2 2 sin 2 π

2.40b. The general solution of the differential equation is

√
√
1
y(t) = C1 cos 2 t + C2 sin 2 t + t
2
Now, we select the constants C1 , C2 so that

   C1 = 0
y(0) = 0 C1 = 0 √
π
⇐⇒ ⇐⇒ √

y(π) = 0 C2 sin 2 π + π/2 = 0  C2 = − 2 sin 2 π

Hence √

π sin 2 t 1
y(t) = − √
+ t
2 sin 2 π 2

2.40c. The general solution of the differential equation is

y(t) = C1 cos(2 t) + C2 sin(2 t) + 3 cos(t)

Now, we select the constants C1 , C2 so that

  
y(0) = 0 C1 + 3 = 0 C1 = −3
⇐⇒ ⇐⇒
y(π) = 0 C1 − 3 = 0 C1 = 3

That is impossible. Therefore the boundary value problem doesn’t have a solution.

2.40d. This is a nonhomogeneous Cauchy-Euler differential equation. Its general solution is

1 2
y(t) = C1 t−1 + C2 t−1 ln(t) + t
9
Now, we select the constants C1 , C2 so that
  
1 1 1
  C1 + = 0
  C1 = −
  C1 = −

y(1) = 0 9 9 9
⇐⇒ 2 ⇐⇒ 2 ⇐⇒ 3
y(e) = 0  C1 e−1 + C2 e−1 + e = 0
  C1 + C2 = − e
  C2 = 1 − e

9 3 9
1 1 − e3 −1 1
Hence y(t) = − t−1 + t ln(t) + t2 .
9 9 9
MATH 375 Higher Order Linear Differential Equations Fall 2023 34

2.40e. The general solution of the differential equation is

y(t) = C1 et + C2 t et = C1 + C2 t et =⇒ y 0 (t) = C1 + C2 + C2 t et

Now, we select the constants C1 , C2 so that

C1 − C2 e−1
 

y(−1) = 0
=0
⇐⇒ ⇐⇒ C1 = C2 = e−1
y(1) + y 0 (1) = 5 2 C1 + 3 C2 e = 5

Hence y(t) = (t + 1) et−1 .

2.41 The characteristic equation of the differential equation is λ2 + k2 = 0. It has roots ±i k.

Hence the general solution is y(x) = C1 cos(k x) + C2 sin(k x) and its derivative is
y 0 (x) = −k C1 sin(k x) + k C2 cos(k x).

2.41a. We need to select the constants C1 , C2 , such that

  
y(0) = 0 C1 = 0 C1 = 0
⇐⇒ ⇐⇒
y 0 (b) = 0 −k C1 sin(k b) + k C2 cos(k b) = 0 k C2 cos(k b) = 0

C2 = 0, leads to the zero solution.

(2 n − 1) π (2 n − 1) π
cos(k b) = 0 =⇒ k b = , n = 1, 2, 3, · · · ⇐⇒ k = , n = 1, 2, 3, · · ·
2 2b
 
(2 n − 1) π (2 n − 1) π
For k = , the nonzero solution is y(x) = sin x or any multiple thereof.
2b 2b

2.41b. We need to select the constants C1 , C2 , such that

  
y(0) = 0 C1 = 0 C1 = 0
⇐⇒ ⇐⇒
y(b) = 0 C1 cos(k b) + C2 sin(k b) = 0 C2 sin(k b) = 0

C2 = 0, leads to the zero solution.

nπ
sin(k b) = 0 =⇒ k b = n π, n = 1, 2, 3, · · · ⇐⇒ k = , n = 1, 2, 3, · · ·
b
nπ nπ 
For k = , the nonzero solution is y(x) = sin x or any multiple thereof.
b b
2.41c. We need to select the constants C1 , C2 , such that
 0  
y (0) = 0 k C2 = 0 C2 = 0
0 ⇐⇒ ⇐⇒
y (b) = 1 −k C1 sin(k b) + k C2 cos(k b) = 1 −k C1 sin(k b) = 1

 C2 = 0
⇐⇒ 1
 C1 = − , kb 6= n π, n = 1, 2, 3, · · ·
k sin(k b)

nπ cos(k x)
For k 6= , n = 1, 2, 3, · · · , the nonzero solution is y(x) = − .
b k sin(k b)

2.41d. We need to select the constants C1 , C2 , such that

 0  
y (0) = 1 k C2 = 1 C2 = 1/k
⇐⇒ ⇐⇒
y 0 (b) = 0 −k C1 sin(k b) + k C2 cos(k b) = 0 −k C1 sin(k b) + cos(k b) = 0

 C2 = 1/k
⇐⇒ cos(k b)
 C1 = , k b 6= n π, n = 1, 2, 3, · · ·
k sin(k b)
MATH 375 Higher Order Linear Differential Equations Fall 2023 35

nπ cos(k b) 1
For k 6= , n = 1, 2, 3, · · · , the nonzero solution is y(x) = cos(k x) + sin(k x).
b k sin(k b) k

2.42. The general solution of the differential equation is

1
y(t) = C1 cos(λ t) + C2 sin(λ t) + sin(t), if λ 6= ±1
λ2 − 1
and
1
y(t) = C1 cos(t) + C2 sin(t) −
t cos(t), if λ = ±1
2
Imposing the boundary conditions y(0) = 0 and y(π) = 1, leads to the system
 
C1 = 0 C1 = 0
⇐⇒
C1 cos(λ π) + C2 sin(λ π) = 1 sin(λ π) C2 = 1

in the first case and to the system


C1 = 0
−C1 + π/2 = 1

in the second case. Hence the boundary value problem has a unique solution if and only if
λ 6= ±1, ±2, ±3, · · ·

2.43. The characteristic equation of the homogeneous equation is λ2 − 4 = 0. It has roots ±2. The
general solution of the homogeneous equation is yh (t) = C1 e−2 t +C2 e2 t . Using the undetermined
coefficients method, we look for a particular solution in the form yp (t) = A e−3 t . Substituting
into y 00 − 4 y = 5 e−3 t , leads to A = 1. Hence yp (t) = e−3 t and the general solution is

y(t) = yh (t) + yp (t) = C1 e−2 t + C2 e2 t + e−3 t

Imposing the limit-at-infinity condition

C1 e−2 t + C2 e2 t + e−3 t = 0 ⇐⇒ C2 = 0


lim y(t) = 0 ⇐⇒ lim
t→+∞ t→+∞

Hence y(t) = C1 e−2 t + e−3 t . The boundary condition y(0) = 3, leads to

C1 + 1 = 3 =⇒ C1 = 2

Thus y(t) = 2 e−2 t + e−3 t .

2.44a. t2 y 00 − t y + λ y = 0 is a one-parameter homogeneous Cauchy-Euler differential equation. Its

indicial/characteristic equation is

r (r − 1) − r + λ = r2 − 2 r + 1 + λ − 1 = (r − 1)2 + λ − 1 = 0

case 1: λ − 1 > 0 ⇐⇒ λ > 1 √

The roots of the indicial equation are 1 ± λ − 1 i and the general solution is
√
√

y(t) = C1 t cos λ − 1 ln(t) + C2 t sin λ − 1 ln(t)

C1 = 0 √
Imposing the boundary conditions y(1) = y(L) = 0, leads to

C2 L sin λ − 1 ln(L) = 0
Taking C1 = C2 = 0, produces the solution y = 0. √

Taking C1 = 0, C2 6= 0, and λ > 1 such that sin λ − 1 ln(L) = 0, produces the nonzero
√

solution y(t) = C2 t sin λ − 1 ln(t) . Now
MATH 375 Higher Order Linear Differential Equations Fall 2023 36

 
λ >√
1 λ
√> 1

⇐⇒
sin λ − 1 ln(L) = 0 λ − 1 ln(L) = n π, n = 1, 2, 3, · · ·
 2
nπ
⇐⇒ λ = 1 + , n = 1, 2, 3, · · ·
ln(L)
Hence  2  
nπ ln(t)
λn = 1 + , yn (t) = t sin n π , n = 1, 2, 3, · · ·
ln(L) ln(L)
are eigenvalue-eigenfunction pairs of our boundary value problem.

case 2: λ − 1 < 0 ⇐⇒ λ < 1 √ √

The roots of the indicial equation are r1 = 1 − 1 − λ, r2 = 1 + 1 − λ and the general solution
is
y(t) = C1 tr1 + C2 tr2
Imposing the boundary conditions y(1) = y(L) = 0, leads to

C1 + C2 = 0
⇐⇒ C1 = C2 = 0
C1 Lr1 + C2 Lr2 = 0

Hence the boundary value problem has only the zero solution, and consequently there is no
eigenvalue λ < 1.

case 3: λ − 1 = 0 ⇐⇒ λ = 1
The roots of the indicial equation are r1 = r2 = 1, and the general solution is

y(t) = C1 t + C2 t ln(t)

Imposing the boundary conditions y(1) = y(L) = 0, leads to


C1 = 0
⇐⇒ C1 = C2 = 0
C1 L + C2 L ln(L) = 0

Hence the boundary value problem has only the zero solution, and consequently λ = 1 is not an
eigenvalue.

In summary: the eigenvalue-eigenfunctions pairs are

 2  
nπ nπ
λn = 1 + , yn (t) = t sin ln(t) , n = 1, 2, 3, · · ·
ln(L) ln(L)

2.44b. y 00 + λ y = 0 is a one-parameter second order linear second order differential equation. Its
characteristic equation is r2 + λ = 0.

case 1: λ > 0.
The general solution is
√
√
√ √
√ √

y(t) = C1 cos λ t + C2 sin λ t =⇒ y 0 (t) = − λ C1 sin λ t + λ C2 cos λ t

. Imposing the boundary conditions y(0) = y(2 π), y 0 (0) = y 0 (2 π), leads to the system
  √
 √

 1 − cos 2 π λ C1 − sin 2 π λ C2 = 0
√
 √

 sin 2 π λ C1 + 1 − cos 2 π λ C2 = 0
MATH 375 Higher Order Linear Differential Equations Fall 2023 37

√
whose solutions are: C1 = C2 = 0 if λ 6= 1, 2, 3, · · · , and C1 , C2 arbitrary, if
√
√λ = 1, 2, 3, · · · . It follows that the boundary value problem has nonzero solutions only when
λ = 1, 2, 3, · · · . Hence λ = n2 , n = 1, 2, 3, · · · are eigenvalues. The corresponding eigenfunc-
tions are cos(n t), sin(n t).
case 2: λ < 0.
The general solution is
√ √ √ √ √ √
y(t) = C1 e− −λ t + C2 e −λ t =⇒ y 0 (t) = − −λ C1 e− −λ t + −λ C2 e −λ t

Imposing the boundary conditions y(0) = y(2 π), y 0 (0) = y 0 (2 π), leads to the system
  √   √ 
 1 − e−2 π −λ C1 + 1 − e2 π −λ C2 = 0
 √   √ 
 1 − e−2 π −λ C1 + −1 + e2 π −λ C2 = 0

whose solution is C1 = C2 = 0. Hence the boundary value problem has only the zero solution,
and consequently there is no eigenvalue λ < 0.

case 3: λ = 0
The general solution is y(t) = C1 + C2 t. Imposing the boundary conditions y(0) = y(2 π)
and y 0 (0) = y 0 (2 π), leads to C2 = 0 and C1 arbitrary. Hence λ = 0 is an eigenvalue with
corresponding eigenfunction y(t) = 1.

In summary: the eigenvalues of our boundary value problem are λn = n2 , n = 0, 1, 2, 3, · · ·

The corresponding eigenfunctions are 1 and cos(n t), sin(n t), n = 1, 2, 3, · · · .

2.44c. y 00 + λ y = 0 is a one-parameter second order linear second order differential equation. Its
characteristic equation is r2 + λ = 0.

case 1: λ > 0.
The general solution is
√
√
√ √
√ √

y(t) = C1 cos λ t + C2 sin λ t =⇒ y 0 (t) = − λ C1 sin λ t + λ C2 cos λ t

Imposing the boundary conditions y(0) − y 0 (0) = 0, y(π) = 0, leads to the system
√ ( √
C1 = λ C 2

C1 −√ λ C 2 = 0 √
√ √
√
i

⇐⇒ h
cos λ π C1 + sin λ π C2 = 0 λ cos λ π + sin λ π C2 = 0

Choosing C2 = 0 =⇒ C1 = 0, which leads to the solution y(t) = 0. If we choose C2 6= 0,

then λ must be a root of the equation
√ √
√
√ √

λ cos λ π + sin λ π = 0 ⇐⇒ λ + tan λπ = 0

In which case the solution is

√ √
√

y(t) = C2 λ cos λ t + sin λ t

Graphing u = tan(π t) and v = −t for t > 0, shows that the two graphs intersect at infinitely
√
many√points with abscissas 0 < t1 < t2 < t3 < · · · This shows that the equation λ+


tan λπ = 0 has infinitely many roots. Letting

0 < λ1 < λ2 < λ3 < · · ·

be those roots, we see that they are all eigenvalues

p pof our
boundary
p value

problem. Each λn ,
corresponds to the eigenfunction yn (t) = λn cos λn t + sin λn t .
MATH 375 Higher Order Linear Differential Equations Fall 2023 38

case 2: λ < 0.
The general solution is
√ √ √ √ √ √
y(t) = C1 e− −λ t
+ C2 e −λ t
=⇒ y 0 (t) = − −λ C1 e− −λ t + −λ C2 e −λ t

Imposing the boundary conditions y(0) − y 0 (0) = 0, y(π) = 0, leads to the system
 √
√

1 −√ −λ C1 +√ 1 − −λ C2 = 0
⇐⇒ C1 = C2 = 0
e−π −λ C1 + eπ −λ C2 = 0

Hence the boundary value problem has only the zero solution, and consequently there is no
eigenvalue λ < 0.

case 3: λ = 0
The general solution is y(t) = C1 + C2 t. Imposing the boundary conditions y(0) − y 0 (0) = 0
and y(π) = 0, leads to the system

C1 − C2 = 0
⇐⇒ C1 = C2 = 0
C1 + π C 2 = 0

Hence the solution y(t) = 0, and λ = 0 is not an eigenvalue.

In summary:
√
problem are the roots 0 < λ1 < λ2 < λ3 < · · ·
the eigenvalues of our√boundary value
of the equation λ + tan λπ = 0. The corresponding eigenfunctions are
p p
p

λn cos λn t + sin λn t , n = 1, 2, 3, · · ·

2.44d. y 00 − 2 y + λ y = 0 is a one-parameter second order linear differential equation. Its characteristic

equation is
r2 − 2 r + λ = (r − 1)2 + λ − 1 = 0
case 1: λ − 1 > 0 ⇐⇒ λ > 1 √
The roots of the characteristic equation are 1 ± λ − 1 i and the general solution is
√ √
y(t) = C1 et cos λ − 1 t + C2 et sin λ − 1 t

Imposing the boundary conditions y(0) = y(π) = 0, leads to

 
C1 = 0 √ √ C1 = 0 √
⇐⇒
C1 eπ cos λ − 1 π + C2 eπ sin λ − 1 π = 0




C2 sin λ − 1π = 0

Taking C1 = C2 = 0, produces the solution y = 0. √

Taking C1 = 0, C2 6= 0, and λ > 1 such that sin λ − 1 π = 0, produces the nonzero solution
√
y(t) = C2 et sin λ − 1 t . Now

 
λ>√ 1 λ
√> 1

⇐⇒ ⇐⇒ λ = 1+n2 , n = 1, 2, 3, · · ·
sin λ − 1 π = 0 λ − 1 = n, n = 1, 2, 3, · · ·

Hence
λn = 1 + n2 , yn (t) = et sin(n t), n = 1, 2, 3, · · ·
are eigenvalue-eigenfunction pairs of our boundary value problem.

case 2: λ − 1 < 0 ⇐⇒ λ < 1 √

The roots of the characteristic equation are 1 ± 1 − λ and the general solution is
√
√

y(t) = C1 e 1− 1−λ t + C2 e 1+ 1−λ t

Imposing the boundary conditions y(0) = y(π) = 0, leads to

MATH 375 Higher Order Linear Differential Equations Fall 2023 39

(
C1 + C2 = 0

√ √
⇐⇒ C1 = C2 = 0
C1 e 1− 1−λ π + C2 e 1+ 1−λ π
=0
Hence the boundary value problem has only the zero solution, and consequently there is no
eigenvalue λ < 1.

case 3: λ − 1 = 0 ⇐⇒ λ = 1
The roots of the indicial equation are r1 = r2 = 1, and the general solution is y(t) = C1 + C2 t.
Imposing the boundary conditions y(0) = y(π) = 0, leads to

C1 = 0
⇐⇒ C1 = C2 = 0
C1 + π C2 eπ = 0

Hence the boundary value problem has only the zero solution, and consequently λ = 1 is not an
eigenvalue.

In summary: the eigenvalue-eigenfunctions pairs of our boundary value problems are

λn = 1 + n2 , yn (t) = et sin(n t), n = 1, 2, 3, · · ·

2.44e. y 00 + λ y = 0 is a one-parameter second order linear second order differential equation. Its
characteristic equation is r2 + λ = 0.

case 1: λ > 0.
The general solution is
√
√
√ √
√ √

y(t) = C1 cos λ t + C2 sin λ t =⇒ y 0 (t) = − λ C1 sin λ t + λ C2 cos λ t

. Imposing the boundary conditions y(−L) = y(L), y 0 (−L) = y 0 (L), leads to the system
 √

2 C2 sin √λ L
= 0
2 C1 sin λ L = 0

Selecting C1 = C2 = 0 gives the solution y(t) = 0. Selecting C1 6= 0 and C2 6= 0, leads to

√
√  n π 2
sin λ L = 0 =⇒ λ L = n π ⇐⇒ λ = , n = 1, 2, 3, · · ·
L
nπ  nπ 
and gives the solution y(t) = C1 cos t + C2 sin t
L L  n π 2
It follows that our boundary value problem has nonzero solutions only when λ = . Hence
 n π 2 nπ  nπ  L
λ= , n = 1, 2, 3, · · · are eigenvalues and cos t , sin t are corresponding
L L L
eigenfunctions.

case 2: λ < 0.
The general solution is
√ √ √ √ √ √
y(t) = C1 e− −λ t
+ C2 e −λ t
=⇒ y 0 (t) = − −λ C1 e− −λ t + −λ C2 e −λ t

Imposing the boundary conditions y(−L) = y(L), y 0 (−L) = y 0 (L), leads to the system
  √ √   √ √ 
 e −λ L − e− −λ L C1 − e −λ L − e− −λ L C2 = 0 
C1 − C2 = 0
 √ √   √ √  ⇐⇒
 − e −λ L − e− −λ L C1 − e −λ L − e− −λ L C2 = 0 C1 + C2 = 0

whose solution is C1 = C2 = 0.
MATH 375 Higher Order Linear Differential Equations Fall 2023 40

Hence the boundary value problem has only the zero solution, and consequently there is no
eigenvalue λ < 0.

case 3: λ = 0
The general solution is y(t) = C1 + C2 t. Imposing the boundary conditions y(−L) = y(L)
and y 0 (−L) = y 0 (L), leads to C2 = 0 and C1 arbitrary. Hence λ = 0 is an eigenvalue with
corresponding eigenfunction y(t) = 1.

In summary:
 n π 2
the eigenvalues of our boundary value problem are λ0 = 0 and λn = , n = 1, 2, 3, · · · .
nπ  nπ  L
The corresponding eigenfunctions are 1 and cos t , sin t , n = 1, 2, 3, · · · .
L L

2.44f. y 00 + λ y = 0 is a one-parameter second order linear second order differential equation. Its
characteristic equation is r2 + λ = 0.

case 1: λ > 0.
The general solution is
√
√
√ √
√ √

y(t) = C1 cos λ t + C2 sin λ t =⇒ y 0 (t) = − λ C1 sin λ t + λ C2 cos λ t

Imposing the boundary conditions y(0) = 0, 3 y(L) − y 0 (L) = 0, leads to the system
(
C
1 = 0 √
√ √

3 sin λ L − λ cos λ L C2 = 0

Choosing C2 = 0, leads to the solution y(t) = 0. If we choose C2 6= 0, then λ must be a root

of the equation
√
√ √
√
√
3 sin λ L − λ cos λ L = 0 ⇐⇒ 3 tan λ L = λ
√

In which case the solution is y(t) = C2 sin λ t . The roots of the above equation form an
increasing sequence 0 < λ1 < λ2 < λ3 < · ·p · They

are all eigenvalues or our boundary value
problem. Associated eigenfunctions are sin λn t , n = 1, 2, 3, · · ·

case 2: λ < 0.
The general solution is
√ √ √ √ √ √
y(t) = C1 e− −λ t
+ C2 e −λ t
=⇒ y 0 (t) = − −λ C1 e− −λ t + −λ C2 e −λ t

Imposing the boundary conditions y(0) = 0, 3 y(L) − y 0 (L) = 0, leads to the system

C1 + C2 = 0 √
√
√
√
3 + −λ e− −λ L C1 + 3 − −λ e −λ L C2 = 0

which we can rewrite as


  C2 = −C1 √ 
√ 3 − −λ ⇐⇒ C1 = C2 = 0
 e−2 −λ L − √ C1 = 0
3 + −λ
Hence the boundary value problem has only the zero solution, and consequently there is no
eigenvalue λ < 0.
MATH 375 Higher Order Linear Differential Equations Fall 2023 41

case 3: λ = 0
The general solution is y(t) = C1 + C2 t. Imposing the boundary conditions y(0) = 0 and
3 y(L) − y 0 (L) = 0, leads to the system

C1 = 0
(3 L − 1) C2 = 0

1
If L 6= , then the solution is y(t) = 0 and λ = 0 is not an eigenvalue.
3
1
If L = , the solution is y(t) = C2 t and λ = 0 is an eigenvalue.
3
In summary:
the eigenvalues of
√ our
boundary
√ value problem are the roots 0 < λ1 p< λ2 < λ3 < · · · of the


equation 3 tan λ L = λ. with corresponding eigenfunctions sin λn t .
1
In addition, if L = , then λ = 0 and y(t) = t is an eigenvalue-eigenfunction pair.
3